MIKE SHE USER MANUAL VOLUME 1: USER GUIDE

MIKE SHE USER MANUAL VOLUME 1: USER GUIDE
MIKE SHE USER MANUAL
VOLUME 1: USER GUIDE
MIKE by DHI 2012
2
Please Note
Copyright
This document refers to proprietary computer software which is protected
by copyright. All rights are reserved. Copying or other reproduction of
this manual or the related programs is prohibited without prior written
consent of DHI. For details please refer to your 'DHI Software Licence
Agreement'.
Limited Liability
The liability of DHI is limited as specified in Section III of your 'DHI
Software Licence Agreement':
'IN NO EVENT SHALL DHI OR ITS REPRESENTATIVES (AGENTS
AND SUPPLIERS) BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING, WITHOUT LIMITATION, SPECIAL, INDIRECT,
INCIDENTAL OR CONSEQUENTIAL DAMAGES OR DAMAGES
FOR LOSS OF BUSINESS PROFITS OR SAVINGS, BUSINESS
INTERRUPTION, LOSS OF BUSINESS INFORMATION OR OTHER
PECUNIARY LOSS ARISING OUT OF THE USE OF OR THE INABILITY TO USE THIS DHI SOFTWARE PRODUCT, EVEN IF DHI
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
THIS LIMITATION SHALL APPLY TO CLAIMS OF PERSONAL
INJURY TO THE EXTENT PERMITTED BY LAW. SOME COUNTRIES OR STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL, SPECIAL, INDIRECT,
INCIDENTAL DAMAGES AND, ACCORDINGLY, SOME PORTIONS
OF THESE LIMITATIONS MAY NOT APPLY TO YOU. BY YOUR
OPENING OF THIS SEALED PACKAGE OR INSTALLING OR
USING THE SOFTWARE, YOU HAVE ACCEPTED THAT THE
ABOVE LIMITATIONS OR THE MAXIMUM LEGALLY APPLICABLE SUBSET OF THESE LIMITATIONS APPLY TO YOUR PURCHASE OF THIS SOFTWARE.'
Printing History
December 2011
July 2011
September 2012
3
4
MIKE SHE
Getting Started • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1
2
INTRODUCTION . . . . . . . . . . . . . . . .
1.1 Process models . . . . . . . . . . . . .
1.2 Requirements . . . . . . . . . . . . . .
1.2.1 Input requirements . . . . . . .
1.2.2 Model limits . . . . . . . . . . .
1.2.3 MIKE SHE Demo model limits
1.2.4 Hardware Requirements . . .
1.3 Getting Help . . . . . . . . . . . . . . .
1.4 Service and Maintenance . . . . . . .
1.4.1 Service Packs . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . 15
. . . . . . . . . . . . . . . . . . . . 18
. . . . . . . . . . . . . . . . . . . . 20
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
23
24
24
. . . . . . . . . . . . . . . . . . . . 26
. . . . . . . . . . . . . . . . . . . . 26
. . . . . . . . . . . . . . . . . . . . 27
BUILDING A MIKE SHE MODEL . . . . . . . . . . . . . . . . . . . .
2.1 MIKE Zero . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.1 MIKE Zero Editors . . . . . . . . . . . . . . . . . . . .
2.2 The MIKE SHE User Interface . . . . . . . . . . . . . . . . . .
2.2.1 The Setup Editor . . . . . . . . . . . . . . . . . . . . .
2.2.2 The Setup Data Tree . . . . . . . . . . . . . . . . . .
2.2.3 Background Maps . . . . . . . . . . . . . . . . . . . .
2.2.4 Initial Model Setup . . . . . . . . . . . . . . . . . . . .
2.2.5 Simulation parameters . . . . . . . . . . . . . . . . .
2.2.6 Hot Starting from a previous simulation . . . . . . . .
2.3 Model domain and grid . . . . . . . . . . . . . . . . . . . . . .
2.4 Topography . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5 Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 Precipitation . . . . . . . . . . . . . . . . . . . . . . .
2.5.2 Snow . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3 Evapotranspiration . . . . . . . . . . . . . . . . . . . .
2.5.4 Snow Melt . . . . . . . . . . . . . . . . . . . . . . . .
2.6 Land Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.1 Vegetation . . . . . . . . . . . . . . . . . . . . . . . .
2.6.2 Paved areas . . . . . . . . . . . . . . . . . . . . . . .
2.6.3 Irrigation . . . . . . . . . . . . . . . . . . . . . . . . .
2.7 Channel Flow . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.8 Overland Flow . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.9 Unsaturated Flow . . . . . . . . . . . . . . . . . . . . . . . . .
2.9.1 Soil Profiles . . . . . . . . . . . . . . . . . . . . . . . .
2.9.2 Initial Conditions . . . . . . . . . . . . . . . . . . . . .
2.9.3 Macropore flow . . . . . . . . . . . . . . . . . . . . .
2.9.4 Green and Ampt infiltration . . . . . . . . . . . . . . .
2.9.5 UZ Column Classification . . . . . . . . . . . . . . . .
2.9.6 Coupling Between Unsaturated and Saturated Zone
. . . . . . . 29
. . . . . . . 29
. . . . . . . 30
. . . . . . . 31
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
32
33
34
34
36
37
. . . . . . . 37
. . . . . . . 39
. . . . . . . 40
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
40
41
41
42
. . . . . . . 43
. . . . . . . 44
. . . . . . . 45
. . . . . . . 45
. . . . . . . 46
. . . . . . . 47
. . . . . . . 52
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
52
53
54
54
55
56
5
2.10 Saturated Groundwater Flow . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.1 Conceptual Geologic Model for the Finite Difference Approach
2.10.2 Specific Yield of upper SZ layer . . . . . . . . . . . . . . . . . . .
2.10.3 Numerical Layers . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.4 Groundwater Drainage . . . . . . . . . . . . . . . . . . . . . . . .
2.10.5 Groundwater wells . . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.6 Linear Reservoir Groundwater Method . . . . . . . . . . . . . . .
2.11 Storing of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.11.1 Detailed Time Series Output . . . . . . . . . . . . . . . . . . . . .
2.11.2 Detailed MIKE 11 Time Series Output . . . . . . . . . . . . . . .
2.11.3 Grid Series Output . . . . . . . . . . . . . . . . . . . . . . . . . .
57
58
59
60
60
61
62
62
64
66
67
Results and Calibration • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3
4
5
6
MIKE SHE RESULTS . . . . . . . . . . . .
3.1 Output Files . . . . . . . . . . . . . .
3.1.1 Log files . . . . . . . . . . .
3.2 Multiple simulations . . . . . . . . . .
3.3 Output Items . . . . . . . . . . . . . .
3.3.1 Overland flow velocity . .
3.3.2 Recharge . . . . . . . . . . .
3.3.3 Summary of all output items
. . . . . . . . . . . . . . . . . . . . .
71
. . . . . . . . . . . . . . . . . . . . .
71
. . . . . . . . . . . . . . . . . . . . .
71
. . . . . . . . . . . . . . . . . . . . .
72
. . . . . . . . . . . . . . . . . . . . .
72
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
73
74
75
THE RESULTS VIEWER . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Toolbars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2
Modifying the plot . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Adding additional result files and overlays . . . . . . . . . .
4.2.2 Adding or modifying vectors . . . . . . . . . . . . . . . . . .
4.2.3 Changing the shading and contour settings of gridded data
4.2.4 Changing the legend and colour scale . . . . . . . . . . . .
4.3 Displaying a time series at a point . . . . . . . . . . . . . . . . . . .
4.4 Saturated Zone Cross-section Plots . . . . . . . . . . . . . . . . . .
4.4.1 Saving and loading profiles . . . . . . . . . . . . . . . . . .
4.5 Displaying a MIKE 11 cross-section . . . . . . . . . . . . . . . . . .
4.6 UZ Specific Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6.1 UZ Scatter and Filled Plots . . . . . . . . . . . . . . . . . . .
4.6.2 UZ Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USING THE WATER BALANCE TOOL . . . . . . . . .
5.1 Creating a water balance . . . . . . . . . . . . .
5.1.1 Create a new water balance document
5.1.2 Extract the water balance data . . . . .
5.1.3 Specify your water balance . . . . . . .
. . .
81
. . .
81
. . .
83
.
.
.
.
.
.
.
.
83
85
87
88
. . .
92
. . .
95
. . .
98
. . .
98
.
.
.
.
. .
100
. .
. .
100
101
. . . . . . . . . . . . .
105
. . . . . . . . . . . . .
106
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
106
107
109
MIKE SHE
5.2
5.3
5.4
5.5
5.1.4 Calculate and View the Water Balance . . . . . . . . . . . . . . 113
Calculating Water Balances in Batch Mode . . . . . . . . . . . . . . . . 113
Available Water Balance Items . . . . . . . . . . . . . . . . . . . . . . . 114
5.3.1 Snow Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.3.2 Canopy interception storage . . . . . . . . . . . . . . . . . . . . 119
5.3.3 Ponded water storage . . . . . . . . . . . . . . . . . . . . . . . . 120
5.3.4 Unsaturated Zone Storage . . . . . . . . . . . . . . . . . . . . . 125
5.3.5 Saturated Zone Storage . . . . . . . . . . . . . . . . . . . . . . 128
5.3.6 Limitations for Linear Reservoir and Sub-catchment OL Water Balance 140
Standard Water Balance Types . . . . . . . . . . . . . . . . . . . . . . . 142
Making Custom Water Balances . . . . . . . . . . . . . . . . . . . . . . 144
5.5.1 Customizing the chart output . . . . . . . . . . . . . . . . . . . . 145
Running MIKE SHE • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
6
RUNNING YOUR MODEL . . . . . . . . . . . .
6.1 Preprocessing your model . . . . . . . .
6.1.1 Viewing the pre-processed data
6.1.2 Editing the pre-processed data
6.2 Pre-processed data items . . . . . . . .
6.2.1 MIKE 11 coupling . . . . . . . .
6.2.2 Land Use . . . . . . . . . . . . .
6.2.3 Unsaturated Flow . . . . . . . .
6.2.4 Saturated Flow . . . . . . . . . .
6.3 The Results Tab . . . . . . . . . . . . . .
6.3.1 Detailed Time Series Results .
6.3.2 Gridded Results . . . . . . . . .
6.3.3 MIKE 11 Detailed Time Series .
6.3.4
Run Statistics . . . . . . . . . .
6.4 Controlling your simulation . . . . . . . .
6.4.1 Model Limits . . . . . . . . . . .
6.4.2 Speeding up your simulation . .
6.4.3 Controlling the Time Steps . . .
6.5 Using Batch Files . . . . . . . . . . . . .
6.6 OpenMI . . . . . . . . . . . . . . . . . . .
6.7 Parallelization of MIKE SHE . . . . . . .
. . . . . . . . . . . . . . . . . .
149
. . . . . . . . . . . . . . . . . .
149
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
150
150
. . . . . . . . . . . . . . . . . .
151
.
.
.
.
.
.
.
.
151
151
152
153
. . . . . . . . . . . . . . . . . .
154
.
.
.
.
.
.
.
.
155
156
158
158
. . . . . . . . . . . . . . . . . .
160
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
160
160
161
. . . . . . . . . . . . . . . . . .
164
. . . . . . . . . . . . . . . . . .
166
. . . . . . . . . . . . . . . . . .
167
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Surface Water • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7
SURFACE WATER IN MIKE SHE . . . . . . . . . . . . . . . . . . . . . . . . . 171
7.1 Overland Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7
7.1.1 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.2 Reduced OL leakage to UZ and to/from SZ . . . . . . . . . . . 173
7.1.3 Separated Flow areas . . . . . . . . . . . . . . . . . . . . . . . 174
7.1.4 Paved Area Drainage . . . . . . . . . . . . . . . . . . . . . . . . 174
7.1.5 Overland Flow Velocities . . . . . . . . . . . . . . . . . . . . . . 177
7.2 Overland Flow Performance . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2.1 Stagnant or slow moving flow . . . . . . . . . . . . . . . . . . . 178
7.2.2 Threshold gradient for overland flow . . . . . . . . . . . . . . . 179
7.3 Multi-cell Overland Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.3.1 Evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.3.2 Infiltration to SZ and UZ with the Multi-Grid OL . . . . . . . . . 184
7.3.3 Reduced Leakage with Multi-cell OL . . . . . . . . . . . . . . . 186
7.3.4 Multi-cell Overland Flow + Saturated Zone drainage . . . . . . 187
7.3.5 Test example for impact on simulation time . . . . . . . . . . . 187
7.3.6 Limitations of the Multi-cell Overland Flow Method . . . . . . . 192
7.3.7 Setting up and evaluating the multi-grid OL . . . . . . . . . . . 192
7.4 Channel Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
7.4.1 MIKE 11 Overview . . . . . . . . . . . . . . . . . . . . . . . . . 194
7.5 Building a MIKE 11 model . . . . . . . . . . . . . . . . . . . . . . . . . . 195
7.5.1 MIKE 11 network limitations . . . . . . . . . . . . . . . . . . . . 196
7.5.2 MIKE 11 Cross-sections . . . . . . . . . . . . . . . . . . . . . . 197
7.6 Coupling of MIKE SHE and MIKE 11 . . . . . . . . . . . . . . . . . . . . 199
7.6.1 MIKE SHE Branches vs. MIKE 11 Branches . . . . . . . . . . 201
7.6.2 The River-Link Cross-section . . . . . . . . . . . . . . . . . . . 202
7.6.3 Connecting MIKE 11 Water Levels and Flows to MIKE SHE . 203
7.6.4 Evaluating your river links . . . . . . . . . . . . . . . . . . . . . 205
7.6.5 Groundwater Exchange with MIKE 11 . . . . . . . . . . . . . . 207
7.6.6 Steady-state groundwater simulations . . . . . . . . . . . . . . 212
7.7 Overland Flow Exchange with MIKE 11 . . . . . . . . . . . . . . . . . . 212
7.7.1 Lateral inflow to MIKE 11 from MIKE SHE overland flow . . . 214
7.7.2 Flooding from MIKE 11 to MIKE SHE using Flood Codes . . . 214
7.7.3 Direct Overbank Spilling to and from MIKE 11 . . . . . . . . . 216
7.7.4 Converting from Flood Codes to Overbank Spilling . . . . . . . 217
7.8 Unsaturated Flow exchange with MIKE 11 . . . . . . . . . . . . . . . . 217
7.9 Water balance with MIKE 11 . . . . . . . . . . . . . . . . . . . . . . . . 217
7.10 Coupling MIKE SHE Water Quality to MIKE 11 . . . . . . . . . . . . . . 218
7.11 MIKE 11 User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
7.11.1 MIKE SHE Coupling Reaches . . . . . . . . . . . . . . . . . . . 219
7.12 Common MIKE 11 Error Messages . . . . . . . . . . . . . . . . . . . . . 225
7.12.1 Error No 25: At the h-point ____ the water depth greater than 4 times
max. depth 225
7.12.2 Warning No 47: At the h-point ____ the water level as fallen below
8
MIKE SHE
the bottom of the slot x times 226
7.12.3 Warning No __: Bed levels not the same . . . . . . . . . . . . . 226
Drainage modelling with MIKE URBAN • . . . . . . . . . . . . . . . . . . . . . . 227
8
USING MIKE SHE WITH MIKE URBAN . . . . . . . . . . . . . . . . . . . . . .
8.1 Coupling MIKE SHE and MIKE URBAN . . . . . . . . . . . . . . . . . .
8.1.1 Telling MIKE SHE to couple to MIKE URBAN . . . . . . . . . .
8.1.2 Telling MIKE URBAN that it is coupled to a MIKE SHE model
8.1.3 Creating a MsheMouse.pfs file . . . . . . . . . . . . . . . . . . .
8.1.4 Output Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.2 Warning messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.3 Water Balance Limitations . . . . . . . . . . . . . . . . . . . . . . . . . .
229
233
234
234
235
238
238
239
Groundwater • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
9
UNSATURATED GROUNDWATER FLOW . . . . . . . . . . . . . . . . . . . . 243
9.0.1 UZ Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
9.0.2 Coupling Between Unsaturated and Saturated Zone . . . . . . 245
10
SATURATED GROUNDWATER FLOW . . . . . . . . . . . . . . . . . .
10.1 Conceptualization of the Saturated Zone Geology . . . . . . . .
10.1.1 Lenses . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10.2 Numerical Layers . . . . . . . . . . . . . . . . . . . . . . . . . . .
10.2.1 Specific Yield of the upper SZ numerical layer . . . . . .
10.2.2 SZ Boundary Conditions . . . . . . . . . . . . . . . . . .
10.3 Groundwater Drainage . . . . . . . . . . . . . . . . . . . . . . . .
10.3.1 Saturated Zone drainage + Multi-cell Overland Flow . .
10.4 MIKE SHE versus MODFLOW . . . . . . . . . . . . . . . . . . .
10.4.1 Importing a MODFLOW 96 or MODFLOW 2000 Model
. . . .
249
. . . .
249
. . . .
251
. . . .
252
. . . .
. . . .
252
253
. . . .
253
. . . .
259
. . . .
262
. . . .
264
Water Quality • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
11
SOLUTE TRANSPORT . . . . . . . . . . . . . . .
11.1 Flow Storing Requirements . . . . . . . . .
11.2 Storing of Results . . . . . . . . . . . . . . .
11.3 Simulation and Time Step Control . . . . . .
11.3.1 Calibrating and Verifying the Model
11.4 Executing MIKE SHE WQ . . . . . . . . . .
11.5 Output . . . . . . . . . . . . . . . . . . . . . .
11.6 Coupling MIKE SHE and MIKE 11 WQ . . .
. . . . . . . . . . . . . . . .
271
. . . . . . . . . . . . . . . .
272
. . . . . . . . . . . . . . . .
272
. . . . . . . . . . . . . . . .
273
. . . . . . . . . . . . . . .
273
. . . . . . . . . . . . . . . .
274
. . . . . . . . . . . . . . . .
275
. . . . . . . . . . . . . . . .
278
9
12
13
MIKE SHE + ECO LAB . . . . . . . . . . . . . . . . . . . . . . . .
12.1 ECO Lab Templates . . . . . . . . . . . . . . . . . . . . . .
12.1.1 Developing a Template . . . . . . . . . . . . . . . .
12.1.2 ECO Lab templates in MIKE SHE . . . . . . . . .
12.1.3 State Variables in MIKE SHE . . . . . . . . . . . .
12.1.4 Specifying Constants and Forcings in MIKE SHE .
12.1.5 Running ECO Lab with MIKE SHE . . . . . . . . .
PARTICLE TRACKING (PT) . . . . . . . . . . . . . . . . .
13.1 Requirements in MIKE SHE WM . . . . . . . . . . .
13.1.1 Flow Storing Requirements . . . . . . . . .
13.1.2 Specification of Well Fields . . . . . . . . .
13.2 Output from the PT simulations . . . . . . . . . . . .
13.3 Extraction of particle registrations . . . . . . . . . . .
13.3.1 Running from a batch file . . . . . . . . . . .
13.3.2 Limitations with the PT registration method
13.4 Extraction of particle pathlines . . . . . . . . . . . .
. . . . . . .
279
. . . . . . .
280
.
.
.
.
.
.
.
.
.
.
280
284
287
288
289
. . . . . . . . . . .
291
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . . . . . . .
291
. . . . . . . . . . .
. . . . . . . . . . .
291
292
. . . . . . . . . . .
293
. . . . . . . . . . .
294
. . . . . . . . . . .
. . . . . . . . . . .
294
295
. . . . . . . . . . .
295
Additional Options • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
14
10
EXTRA PARAMETERS . . . . . . . . . . . . . . . . . . . . . . .
14.1 Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.1.1 Negative Precipitation . . . . . . . . . . . . . . . .
14.1.2 Precipitation Multiplier . . . . . . . . . . . . . . . .
14.2 Surface Water . . . . . . . . . . . . . . . . . . . . . . . . .
14.2.1 Time-varying Overland Flow Boundary Conditions
14.2.2 Time varying surface infiltration (Frozen soils) . .
14.2.3 Simplified Overland Flow Options . . . . . . . . . .
14.2.4 Irrigation River Source Factors . . . . . . . . . . .
14.2.5 Explicit Overland Flow Output . . . . . . . . . . . .
14.2.6 Alternative low gradient damping function . . . . .
14.2.7 Paved routing options . . . . . . . . . . . . . . . .
14.3 Unsaturated Zone . . . . . . . . . . . . . . . . . . . . . . . .
14.3.1 Transpiration during ponding . . . . . . . . . . . .
14.3.2 Threshold depth for infiltration (2-Layer UZ) . . . .
14.3.3 Increase infiltration to dry soils . . . . . . . . . . .
14.4 Saturated Zone . . . . . . . . . . . . . . . . . . . . . . . . .
14.4.1 Sheet Pile Module . . . . . . . . . . . . . . . . . . .
14.4.2 SZ Drainage to Specified MIKE 11 H-points . . . .
14.4.3 SZ Drainage Downstream Water Level Check . .
14.4.4 SZ Drainage to MOUSE . . . . . . . . . . . . . . .
14.4.5 Time varying drainage parameters . . . . . . . . .
14.4.6 SZ Drainage River Link Reference Table . . . . .
. . . . . . .
299
. . . . . . .
300
. . . . . . .
. . . . . . .
300
301
. . . . . . .
302
.
.
.
.
.
.
.
.
.
.
.
.
.
.
302
303
304
306
307
308
309
. . . . . . .
309
. . . . . . .
. . . . . . .
. . . . . . .
309
310
311
. . . . . . .
312
.
.
.
.
.
.
312
316
319
319
320
321
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
MIKE SHE
14.4.7 Canyon exchange option for deep narrow channels
14.5 Water Quality . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.5.1 Disable SZ solute flux to dummy UZ . . . . . . . . .
14.5.2 SZ boundary dispersion . . . . . . . . . . . . . . . . .
14.6 Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.6.1 Including OpenMI . . . . . . . . . . . . . . . . . . . .
14.6.2 Plot control for Detailed Time Series Output . . . . .
14.6.3 Extra Pre-Processing output . . . . . . . . . . . . . .
14.6.4 GeoViewer Output . . . . . . . . . . . . . . . . . . . .
. . . . . .
322
. . . . . .
323
. . . . . .
. . . . . .
323
323
. . . . . .
324
.
.
.
.
324
325
325
325
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
MIKE ZERO Options • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
15
EUM DATA UNITS . . . . . . . . . . . . . . . . . . . . . . . . .
15.1 Changing from SI to Imperial (American) data units. . .
15.2 Restoring the default units . . . . . . . . . . . . . . . . .
15.3 Changing the EUM data type of a Parameter . . . . . .
15.3.1 Changing the EUM Type of a .dfs0 Parameter
15.3.2 Changing the EUM Type of a .dfs2 Parameter
. . . . . . . . .
329
. . . . . . . . .
331
. . . . . . . . .
332
. . . . . . . . .
332
. . . . . . . . .
. . . . . . . . .
334
335
Working with Data • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
16
TIME SERIES DATA . . . . . . . . . . . .
16.1 Creating Time Series in MIKE SHE
16.1.1 Import from ASCII . . . . .
16.1.2 Import from Excel . . . . .
16.1.3 Import from old .t0 file . . .
16.2 Working with Spatial Time Series .
16.3 Time Series Types . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
339
. . . . . . . . . . . . . . . . . . . . .
339
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
340
341
341
. . . . . . . . . . . . . . . . . . . . .
341
. . . . . . . . . . . . . . . . . . . . .
342
17
USING MIKE SHE WITH ARCGIS . . . . . . . . . . . . . . . . . . . . . . . . . 345
18
SPATIAL DATA . . . . . . . . . . . . .
18.1 The Grid Editor . . . . . . . . . .
18.2 Gridded Data Types . . . . . . .
18.3 Integer Grid Codes . . . . . . .
18.4 Gridded (.dfs2) Data . . . . . .
18.4.1 Stationary Real Data .
18.4.2 Time-varying Real Data
18.4.3 Integer Grid Codes . .
18.5 Interpolation Methods . . . . . .
18.5.1 Bilinear Interpolation .
18.5.2 Triangular Interpolation
. . . . . . . . . . . . . . . . . . . . . . .
347
. . . . . . . . . . . . . . . . . . . . . . .
347
. . . . . . . . . . . . . . . . . . . . . . .
347
. . . . . . . . . . . . . . . . . . . . . . .
348
. . . . . . . . . . . . . . . . . . . . . . .
349
. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
351
352
353
. . . . . . . . . . . . . . . . . . . . . . .
355
. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
355
359
11
18.5.3 Inverse Distance . . . . . . . . . . . . . . . . . . . . . . . . . .
18.6 Performing simple math on multiple grids . . . . . . . . . . . . . . . . .
18.7 Performing complex operations on multiple grids . . . . . . . . . . . . .
361
361
363
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
12
MIKE SHE
GETTING STARTED
13
14
MIKE SHE
1
INTRODUCTION
In the hydrological cycle, water evaporates from the oceans, lakes and rivers, from the soil and is transpired by plants. This water vapour is transported in the atmosphere and falls back to the earth as rain and snow. It
infiltrates to the groundwater and discharges to streams and rivers as baseflow. It also runs off directly to streams and rivers that flow back to the
ocean. The hydrologic cycle is a closed loop and our interventions do not
remove water; rather they affect the movement and transfer of water
within the hydrologic cycle.
In 1969, Freeze and Harlan (Freeze and Harlan, 1969) proposed a blueprint for modelling the hydrologic cycle. In this original blueprint, different flow processes were described by their governing partial differential
equations. The equations used in the blueprint were known to represent
the physical processes at the appropriate scales in the different parts of the
hydrological cycle.
From 1977 onwards, a consortium of three European organizations developed, and extensively applied, the Système Hydrologique Européen
(SHE) based on the blueprint of Freeze and Harlan (Abbott et al., 1986a &
b). The integrated hydrological modelling system, MIKE SHE, emerged
from this work (see Figure 1.1)
Since the mid-1980's, MIKE SHE has been further developed and
extended by DHI Water & Environment. Today, MIKE SHE is an
advanced, flexible framework for hydrologic modelling. It includes a full
suite of pre- and post-processing tools, plus a flexible mix of advanced
and simple solution techniques for each of the hydrologic processes.
MIKE SHE covers the major processes in the hydrologic cycle and
includes process models for evapotranspiration, overland flow, unsaturated flow, groundwater flow, and channel flow and their interactions.
Each of these processes can be represented at different levels of spatial
distribution and complexity, according to the goals of the modelling study,
the availability of field data and the modeller’s choices, (Butts et al. 2004).
The MIKE SHE user interface allows the user to intuitively build the
model description based on the user's conceptual model of the watershed.
The model data is specified in a variety of formats independent of the
model domain and grid, including native GIS formats. At run time, the
spatial data is mapped onto the numerical grid, which makes it easy to
change the spatial discretisation.
Getting Started
15
Introduction
Figure 1.1
Hydrologic processes simulated by MIKE SHE
MIKE SHE uses MIKE 11 to simulate channel flow. MIKE 11 includes
comprehensive facilities for modelling complex channel networks, lakes
and reservoirs, and river structures, such as gates, sluices, and weirs. In
many highly managed river systems, accurate representation of the river
structures and their operation rules is essential. In a similar manner, MIKE
SHE is also linked to the MOUSE sewer model, which can be used to simulate the interaction between urban storm water and sanitary sewer networks and groundwater. MIKE SHE is applicable at spatial scales ranging
from a single soil profile, for evaluating crop water requirements, to large
regions including several river catchments, such as the 80,000 km2 Senegal Basin (e.g. Andersen et al., 2001). MIKE SHE has proven valuable in
hundreds of research and consultancy projects covering a wide range of
climatological and hydrological regimes, many of which are referenced in
Graham and Butts (2006).
The need for fully integrated surface and groundwater models, like MIKE
SHE, has been highlighted by several recent studies (e.g. Camp Dresser &
16
MIKE SHE
McKee Inc., 2001; Kaiser-Hill, 2001; West Consultants Inc. et al., 2001;
Kimbley-Horn & Assoc. Inc. et al., 2002; Middlemis, 2004, which can all
be downloaded from the MIKE SHE web site). These studies compare and
contrast available integrated groundwater/surface water codes. They also
show that few codes exist that have been designed and developed to fully
integrate surface water and groundwater. Further, few of these have been
applied outside of the academic community (Kaiser-Hill, 2001).
Applications around the world
MIKE SHE has been used in a broad range of applications. It is being used
operationally in many countries around the world by organizations ranging from universities and research centres to consulting engineers companies (Refsgaard & Storm, 1995). MIKE SHE has been used for the
analysis, planning and management of a wide range of water resources
and environmental and ecological problems related to surface water and
groundwater, such as:
z
River basin management and planning
z
Water supply design, management and optimization
z
Irrigation and drainage
z
Soil and water management
z
Surface water impact from groundwater withdrawal
z
Conjunctive use of groundwater and surface water
z
Wetland management and restoration
z
Ecological evaluations
z
Groundwater management
z
Environmental impact assessments
z
Aquifer vulnerability mapping
z
Contamination from waste disposal
z
Surface water and groundwater quality remediation
z
Floodplain studies
z
Impact of land use and climate change
z
Impact of agriculture (irrigation, drainage, nutrients and pesticides,
etc.)
Graham and Butts (2006) contains a list of some easily accessible references for many of the application areas listed above.
Getting Started
17
Introduction
User interface
MIKE SHE’s user interface can be characterized by the need to
1 Develop a GUI that promotes a logical and intuitive workflow, which
is why it includes
–
A dynamic navigation tree that depends on simple and logical
choices
–
A conceptual model approach that is translated at run-time into the
mathematical model
–
Object oriented “thinking” (geo-objects with attached properties)
–
Full, context-sensitive, on-line help
–
Customized input/output units to support local needs
2 Strengthen the calibration and result analysis processes, which is why
it includes
–
Default HTML outputs (calibration hydrographs, goodness of fit,
water balances, etc.)
– User-defined HTML outputs
– A Result Viewer that integrates 1D, 2D and 3D data for viewing
and animation
– Water balance, auto-calibration and parameter estimation tools.
3 Develop a flexible, unstructured GUI suitable for different modelling
approaches, which is why it includes
– Flexible data format (gridded data, .shp files, etc.) that is easy to
update for new data formats
– Flexible time series module for manipulating time-varying data
– Flexible engine structure that can be easily updated with new
numerical engines
The result is a GUI that is flexible enough for the most complex applications imaginable, yet remains easy-to-use for simple applications.
1.1
Process models
MIKE SHE, in its original formulation, could be characterized as a deterministic, physics-based, distributed model code. It was developed as a
fully integrated alternative to the more traditional lumped, conceptual
rainfall-runoff models. A physics-based code is one that solves the partial
differential equations describing mass flow and momentum transfer. The
18
MIKE SHE
Process models
parameters in these equations can be obtained from measurements and
used in the model. For example, the St. Venant equations (open channel
flow) and the Darcy equation (saturated flow in porous media) are physics-based equations.
There are, however, important limitations to the applicability of such
physics-based models. For example,
z
it is widely recognized that such models require a significant amount of
data and the cost of data acquisition may be high;
z
the relative complexity of the physics-based solution requires substantial execution time;
z
the relative complexity may lead to over-parameterised descriptions
for simple applications; and
z
a physics-based model attempts to represent flow processes at the grid
scale with mathematical descriptions that, at best, are valid for smallscale experimental conditions.
Therefore, it is often practical to use simplified process descriptions. Similarly, in most watershed problems one or two hydrologic processes dominate the watershed behaviour. For example, flood forecasting is dominated
by river flows and surface runoff, while wetland restoration depends
mostly on saturated groundwater flow and overland flow. Thus, a complete, physics-based flow description for all processes in one model is
rarely necessary. A sensible way forward is to use physics-based flow
descriptions for only the processes that are important, and simpler, faster,
less data demanding methods for the less important processes. The downside is that the parameters in the simpler methods are usually no longer
physics meaningful, but must be calibrated-based on experience.
The process-based, modular approach implemented in the original SHE
code has made it possible to implement multiple descriptions for each of
the hydrologic processes. In the simplest case, MIKE SHE can use fully
distributed conceptual approaches to model the watershed processes. For
advanced applications, MIKE SHE can simulate all the processes using
physics-based methods. Alternatively, MIKE SHE can combine conceptual and physics-based methods-based on data availability and project
needs. The flexibility in MIKE SHE's process-based framework allows
each process to be solved at its own relevant spatial and temporal scale.
For example, evapotranspiration varies over the day and surface flows
respond quickly to rainfall events, whereas groundwater reacts much
slower. In contrast, in many non-commercial, research-oriented integrated
hydrologic codes (e.g. MODFLOW HMS, Panday et al., 1998; InHM,
Sudicky et al., 2002), all the hydrologic processes are solved implicitly at
Getting Started
19
Introduction
a uniform time step, which can lead to intensive computational effort for
watershed scale models.
Figure 1.2
1.2
Schematic view of the process in MIKE SHE, including the available
numeric engines for each process. The arrows show the available
exchange pathways for water between the process models. Note:
the SVAT evapotranspiration model is not yet available in the commercial version of MIKE SHE.
Requirements
The requirements to build and run a MIKE SHE model depend on the purpose of the model and the trade-offs that must be made between conceptualization and the practicality of simulation time.
1.2.1
20
Input requirements
The flexibility of MIKE SHE means that there is no predefined list of
required input data. The required data depends on the hydrologic process
included and the process model selected, which, in turn, depend on what
MIKE SHE
Requirements
problem you are trying to solve with MIKE SHE. However, the following
basic model parameters are required for nearly every MIKE SHE model:
z
Model extent - typically as a polygon,
z
Topography - as point or gridded data, and
z
Precipitation - as station data (rain gauge data).
Additional basic data is required depending on the hydrologic processes
included, and their options:
z
Reference evapotranspiration - as station data or calculated from meteorological data,
z
Air Temperature - for calculating snowmelt (station data),
z
Solar Radiation - for calculating snowmelt (station data)
z
Sub-catchment delineation - for runoff distribution
z
River morphology (geometry + cross-sections) - for river flow and
water level calculations
z
Land use distribution - for vegetation and paved runoff calculations
z
Soil distribution - for distributing infiltration and calculating runoff
z
Subsurface geology - for calculating groundwater flow
If you also want to calculate water quality then additional basic information includes:
z
Species to be simulated, and
z
Source locations
The data items listed above are the basic input data that define your problem. They are not usually part of the calibration. If we now look at each of
Getting Started
21
Introduction
the hydrologic processes, and the process models available for each, then
we can separate out the principle calibration parameters.
Table 1.1
Principle parameters for MIKE SHE
Principle calibration
parameters
Other parameters
Overland flow
Surface roughness
(finite difference)
Detention storage
Overland flow
(subcatchmentbased)
Surface roughness
Detention storage
Slope parameters
River flow
River bed roughness
River bed leakage coefficient
Unsaturated flow Saturated hydraulic conduc- Soil water contents at
(finite difference) tivity
saturation, field capacity, and wilting point
Soil pedotransfer function parameters
Unsaturated flow Saturated hydraulic conduc- Soil water contents at
saturation, field capac(2-layer method) tivity
ity, and wilting point
Capillary thickness
22
Actual Evapotran- Leaf Area Index
spiration
Root depth
Canopy Interception
FAO Crop coefficient
Kristensen and Jensen
ET parameters
Groundwater flow Hydraulic conductivity
(finite difference) Specific yield
Specific storage
Drain level
Drain time constants
MIKE SHE
Requirements
Table 1.1
Principle parameters for MIKE SHE
Principle calibration
parameters
Other parameters
Groundwater flow Reservoir time constants
Interbasin transfers
(linear reservoir) Reservoir volumes (specific (dead zone storage)
yield, depths)
Water quality
Porosity
Soil bulk density
Dispersivities
Sorption and degradation
rate constants
Source strength
The parameter list in Table 1.1is not complete. There are many other
parameters that can be modified if you are trying to simulate something
specific, such as paved area discharge, or snowmelt, or if you eliminate
one or more of the processes. If you do not simulate a process, then a place
holder parameter is usually required that will need to be calibrated. For
example, if you do not simulate the unsaturated zone and evapotranspiration, then precipitation must be converted to groundwater recharge using
the Net Rainfall Fraction and Infiltration Fraction parameters to account
for losses to evapotranspiration and runoff.
1.2.2
Model limits
Although, there are no physical limits to the size of your model, there are
practical limits and hardware limits.
The practical limits are generally related to run time. We all want the
model to be a little bit bigger or more detailed. However, that little extra
detail or slightly smaller grid size can quickly lead to long run times.
The physical limits are generally related to memory size. If you model
requires more memory than is physically installed on the computer, then
the computer will start to swap data to the hard disk. This will vastly slow
down your simulation. Also, if you are using a 32-bit computer, then the
operating system will also put constraints on the maximum file size and
memory.
If your model reaches the practical or physical limits of your computer,
then critically evaluate your model to see if you really need such a large,
complex model. For example, maybe you can reduce the number of UZ
nodes or increase the grid size.
Getting Started
23
Introduction
If the model is simply too slow, then you may be able do an initial rough
calibration with a less complex model. For example, during the initial calibration, you could use Gravity flow instead of Richards equation, double
the grid spacing, or shorten the calibration period. Afterwards, you can
switch back to the original configuration for the final calibration. You
might even be surprised that the rougher model is actually good enough.
1.2.3
MIKE SHE Demo model limits
If no dongle is installed, or if a current license file is not available, then
MIKE SHE will run in demo mode. In this case, the model size is
restricted. If you need a full size MIKE SHE to perform your evaluation,
then you are welcome to contact your local DHI office to request a 30-day
evaluation license.
The current demo restrictions are as follows:
– number of cells in x- and y-direction: 70
– number of computational cells per layer (incl. boundary cells): 2000
– number of computational saturated zone layers: 2
– number of river links: 250
– number of computational UZ columns (multi-layer UZ): 155
– number of nodes per UZ column (multi-layer UZ): 100
– simulation time: 4444 hours or 185 days
– number of UZ timesteps: 800
– number of SZ timesteps: 200
– no steady-state SZ
– no overbank spilling
– no ECO Lab linkage
– no irrigation
Further, there are some restrictions in the rest of the MIKE Zero tools in
demo mode. The most critical of these is that the Grid and Time Series
Editors do not allow you to save files in demo mode.
1.2.4
24
Hardware Requirements
The hardware requirements for MIKE SHE depend on the model that you
are trying to simulate. As a rule of thumb, any good quality, new computer
should be sufficient for an average MIKE SHE model. Thus, a typical
MIKE SHE
Requirements
machine for an average MIKE SHE model will have at least a 2GHz CPU,
2GB of RAM, and 100 GB of free disk space.
Note, however, these are minimum requirements. In particular, data storage is often a problem. A large model with a long simulation period and a
short saved time step interval can easily generate very large output data
sets. If you save multiple simulations (e.g. calibration runs or scenarios),
then you can quickly have hundreds of Gigabytes of output data.
Note: MIKE SHE must run in a Windows environment and will not run on
Linux workstations.
64-bit CPU
Most of the DHI numerical engines are compiled for a 64-bit processor,
including MIKE SHE.
However, MIKE 11 is an older code that cannot be compiled for a 64-bit
processor. Therefore, MIKE 11 will run as a 32-bit application in a 64-bit
environment.
Multi-core/processor computers
The numerically intensive operations in the MIKE SHE engine have been
optimized for multi-core computers. However, not all of the hydrologic
processes scale equally well. Thus, the simulation speed improvements on
multi-core computers depends on the model.
The AUTOCAL program for parameter optimization and sensitivity analysis has been updated to automatically spread out the simulation load to
the available cores.
The standard MIKE Zero license is supports up to four cores/processors. If
you want to take advantage of more than four cores, then you will need to
contact your local DHI sales office to obtain additional run-time licenses.
RAM
MIKE SHE does not dynamically allocate RAM. That is, the amount of
RAM required by the model is allocated at the beginning of the simulation
based on the specified number of nodes. If you don’t have enough RAM,
then MIKE SHE will swap to the hard disk, which can drastically slow
down your simulation.
The amount of RAM may also be important when running multiple simulations at the same time, since each simulation will require a full memory
space.
Getting Started
25
Introduction
In a computer with a 32-bit operating system, each application is restricted
to 2GB of RAM. If you have a 32-bit system, each MIKE SHE simulation
can only use a maximum of 2GB of RAM - even if you have installed
more than 2GB of RAM.
If your computer has a 64-bit operating system, then there is effectively no
limit to the amount of RAM. In this case, MIKE SHE will use all available
RAM.
CPU Speed
In general, the higher the CPU clock speed, the faster the calculations.
However, simulation speed also depends on the chip design, which
depends on the manufacturer (e.g. Intel vs AMD), the platform (e.g. laptop
vs. desktop), etc. Given the huge range of chip designs and the rapid pace
of development, it is difficult to give specific guidance on choice of CPU other than “faster is usually better, all other things being equal”.
1.3
Getting Help
If you click F1 in any MIKE SHE dialogue, you will land in one of the
sections of The MIKE SHE Reference Guide. Likewise, if you click F1 in
any MIKE 11 or other MIKE Zero dialogue, you will land in a relevant
section of the on-line help.
This manual is a supplement to the basic on-line F1 help and provides you
with additional information on how to use MIKE SHE to get the results
that you want.
1.4
Service and Maintenance
As with any complex software package, the software is being continually
improved and extended. Some of these improvements are fixes of problems that have slipped though our quality control. Others are fixes of
known minor problems with the software. However, the vast majority of
the changes in new releases and service packs are related to improvements
to the functionality of the software.
Your initial purchase of the software is protected by a one-year subscription to our Service and Maintenance Agreement. Your Service and Maintenance Agreement entitles you free support for software problems via
email or telephone and regular updates to the software.
26
MIKE SHE
Service and Maintenance
We strongly recommend that you subscribe to the Service and Maintenance Agreement after the first year to further protect your investment.
Improvements, extensions and fixes are continually being made, and we
will make every effort to help you with any problems that you encounter,
but we cannot provide fixes for any versions older than the current release.
1.4.1
Service Packs
As part of the Service and Maintenance, there is an auto update program
installed with your software. This program automatically checks our website for Service Packs to the currently installed release and downloads the
Service Pack if it is available. You will be asked before the installation
begins, if you want the installation to proceed. We strongly recommend
that you install the latest Service Pack as soon as they are released.
However, some clients prefer not to install the Service Pack during a
project, or close to the end of a project. Occasionally, a fix in the numerical engine will slightly change your simulation results. This may require
you to re-run previously finished simulations to obtain valid comparisons
between simulations.
The Auto Updater overwrites your existing executable files. Therefore, if
you are concerned about potential changes in your results, they you should
backup all of the files in the MIKE SHE installation directory, before
installing the Service Pack.
If you did not back up your installation directory, and you need to restore a
previous version, DHI maintains an archive of all standard patch versions.
Contact your local support centre and we will send you a copy of your previous executable.
Getting Started
27
Introduction
28
MIKE SHE
MIKE Zero
2
BUILDING A MIKE SHE MODEL
The MIKE SHE user interface is organized around the workflow to build a
model. Basically, your work flow follows the data tree. You typically start
at the top of the data tree and work your way down. As you complete each
of the items in the data tree, the red “x” will be replaced by a green checkmark. Thus, the basic work flow for a fully integrated MIKE SHE model
is built around the following components:
1 The MIKE SHE User Interface (V.1 p. 31)
2 Background Maps (V.1 p. 34)
3 Initial Model Setup (V.1 p. 34)
4 Simulation parameters (V.1 p. 36)
5 Model domain and grid (V.1 p. 37)
6 Topography (V.1 p. 39)
7 Climate (V.1 p. 40)
8 Channel Flow (V.1 p. 46)
9 Overland Flow (V.1 p. 47)
10 Unsaturated Flow (V.1 p. 52)
11 Saturated Groundwater Flow (V.1 p. 57)
12 Storing of results (V.1 p. 62)
13 Preprocessing your model (V.1 p. 149)
14 Running your Model (V.1 p. 149)
2.1
MIKE Zero
MIKE SHE is part of the MIKE Zero suite of modelling tools. However,
MIKE Zero is more than a set of modelling tools. MIKE Zero is a project
management interface, with a full range of tools for helping you with your
modelling project.
In any project, it is a challenge to maintain an overview of all of these
files, not to mention keeping regular backups and archives of all of these
files. As you progress through the calibration and validation phases, and
then on to the scenario analysis and report writing phases, the number of
model artifacts can become overwhelming. The MIKE Zero project structure is designed to include all of your modelling files; that is all of the raw
Getting Started
29
Building a MIKE SHE Model
data files, model input files, and model output files, as well as any reports,
spread sheets, plots, etc.
The MIKE Zero project structure is designed to help you keep control of
your project.
There is a separate introduction manual to help you get started working
with MIKE Zero.
2.1.1
MIKE Zero Editors
The MIKE Zero also includes general tools for data editing, analysis and
manipulation. Some of these have their own file types, or documents. The
MIKE Zero documents include (with the tools commonly used for MIKE
SHE in bold):
z
The Time Series Editor (.dfs0)- for time series data
z
The Profile Series Editor (.dfs1)- for time varying 1D data (profiles are
not used in MIKE SHE)
z
The Grid Editor (.dfs2 and .dfs3) - for time varying 2D and 3D data
z
Data Manager - for finite element data
z
The Plot Composer (.plc) - for creating standard report plots
z
Result Viewer (.rev) - for results presentation
z
Bathometry (.batsf) - for sea bed elevations
z
Animator (.mza) - for 3D visualization of 2D surface water and waves
z
ECOLAB (.ecolab) - for water quality in surface water, which can be
used in MIKE 11 (but not yet in the rest of MIKE SHE)
z
AUTOCAL (.auc) - for autocalibration, sensitivity analysis and scenario management
z
EVA Editor (.eva) - for extreme value analysis of surface water flows
z
Mesh Generator (.mdf) - for creating meshes for the finite element versions of MIKE 21 and MIKE 3
z
Data Extraction FM (.dxfm) - for extracting data from finite element
results files
z
MIKE Zero Toolbox (.mzt) - various tools for data manipulation
The documentation for these tools is found in the printed MIKE Zero
books and under MIKE Zero in the on-line help.
30
MIKE SHE
The MIKE SHE User Interface
In addition to the MIKE Zero document-based tools, there are a number of
other important MIKE Zero utilities that are accessed from the Start\Programs\MIKE by DHI menu, including:
2.2
z
MIKE View - a results evaluation utility for MIKE 11 and MOUSE
(sewers) 1D flow results.
z
Image Rectifer - a simple tool for stretching and georeferencing image
files
z
Launch Simulation Engine - a utility for launching and running
MIKE Zero simulation engines independent of the Graphical User
Interface.
z
GeoViewer - a visualization tool for 3D layer data
The MIKE SHE User Interface
The MIKE SHE user interface is organized by task. In every model application you must
1 Set up the model,
2 Run the model, and
3 Assess the results.
The above three tasks are repeated until you obtain the results that you
want from the model.
When you create or open a MIKE SHE model, you will find your self in
the Setup Tab of the MIKE SHE user interface.
The following sections provide a quick overview of the main hydrologic
processes in MIKE SHE. For more detailed information on the individual
parameters, see Setup Data Tab (V.2 p. 19) chapter in the Reference Manual.
Alternatively, this manual also contains detailed user guidance and information in the sections:
Getting Started
z
Surface Water (V.1 p. 169)
z
Unsaturated Groundwater Flow (V.1 p. 243)
z
Saturated Groundwater Flow (V.1 p. 249)
z
Running MIKE SHE (V.1 p. 147)
z
Results and Calibration (V.1 p. 69)
31
Building a MIKE SHE Model
2.2.1
The Setup Editor
Context Sensitive dialogue
Data tree
Task tabs
Data validation area
Figure 2.1
Graphical overview of the in the MIKE SHE GUI, without the Project
Explorer.
The Setup editor is divided into three sections - the data tree, a context
sensitive dialogue and a validation area.
The data tree is dynamic and changes with how you set up your model. It
provides an overview of all of the relevant data in your model. The data
tree is organized vertically, in the sense that if you work your way down
the tree, by the time you come to the bottom you are ready to run your
model.
The context sensitive dialogue on the right allows you to input the
required data associated with your current location in the data tree. The
dialogues vary with the type of data, which can be any combination of
static and dynamic data, as well as spatial and non-spatial data. In the case
32
MIKE SHE
The MIKE SHE User Interface
of spatial and time varying data, the actual data is not input to the GUI.
Rather, a file name must be specified and the link to the file is stored in the
GUI. Furthermore, the distribution of the data in time and space need not
correspond between the various entries. For example, rainfall data may be
entered as hourly values and pumping rates as weekly values, while the
model may be run with daily timesteps.
The validation area at the bottom of the dialogue provides you with immediate feedback on the validity of the data that you have input.
After you have set up your model, you must switch to the Processed Data
tab and run the pre-processing engine on the model. This step reconciles
all of the various spatial and time series data and creates the actual data set
that will be run by MIKE SHE. Once the data has been pre-processed the
simulation can be started. Using the Pre-processing tab at the bottom, you
can view the pre-processed data.
After the simulation is finished, you can switch to the Results tab, where
you can view the detailed time series output as in a report-ready HTML
view. Alternatively, you can use the Results Viewer, which is one of the
generic MIKE Zero tools, for more customized and detailed analysis of
the gridded output.
2.2.2
The Setup Data Tree
Your MIKE SHE model is organized around the Setup Data Tree. The layout of the tree depends on the model components that are active in the current model, which are selected in the Simulation Specification dialogue.
Opposite the data tree is the corresponding dialogue for the currently
selected tree branch.
The data tree is designed to hide the components that are not needed for
the current simulation. However, no data is ever lost if the branch is hidden. That is, all data is retained, even if the branch is not currently visible.
The design of the data tree is such that when you make selections in the
current dialogue, the tree is automatically updated to reflect the selection.
However, the layout of the data tree and the options available in the current dialogue are such that the data tree will only change along the current
branch. That is, if you make a selection in the current dialogue, additional
options or branches may become available further along the branch. However, no changes will occur in other branches of the data tree. For example, if you make a selection in the Precipitation dialogue, this will affect
the Precipitation data branch. It will not affect the Evapotranspiration
branch.
Getting Started
33
Building a MIKE SHE Model
The only exception to the above rule is selections made in the Simulation
Specification dialogue, which is used to set up the entire data tree. Thus,
for example, if you unselect Evapotranspiration in the Simulation Specification dialogue, the entire Evapotranspiration branch will disappear.
2.2.3
Background Maps
Arguably, the first step in building your model is to define where your
model is located. This generally involves defining a basic background
map for your model area.
The Display item is located at the top of the data tree to make it easy to
add and edit your background maps. In the Display item, you can add any
number of images to your model setup, in a variety of formats. The images
are carried over to the various editors, so you can keep a consistent display
between the set up editor and, for example, the Grid Editor and the Results
Viewer.
In the event that you are using scanned paper maps, if your maps are not
rectilinear, or are not correctly georeferenced, then you can use the Image
Rectifer (see on-line help under MIKE Zero) to align your image to the
coordinate system you are using.
Note The display of the Mike 11 network is not carried over to the Results
Viewer.
2.2.4
34
Initial Model Setup
MIKE SHE allows you to simulate all of the processes in the land phase of
the hydrologic cycle. That is, all of the process involving water movement
after the precipitation leaves the sky. Precipitation falls as rain or snow
depending on air temperature - snow accumulates until the temperature
increases to the melting point, whereas rain immediately enters the
dynamic hydrologic cycle. Initially, rainfall is either intercepted by leaves
(canopy storage) or falls through to the ground surface. Once at the ground
surface, the water can now either evaporate, infiltrate or runoff as overland
flow. If it evaporates, the water leaves the system. However, if it infiltrates
then it will enter the unsaturated zone, where it will be either extracted by
the plant roots and transpired, added to the unsaturated storage, or flow
downwards to the water table. If the upper layer of the unsaturated zone is
saturated, then additional water cannot infiltrate and overland flow will be
formed. This overland flow will follow the topography downhill until it
reaches an area where it can infiltrate or until it reaches a stream where it
will join the other surface water. Groundwater will also add to the baseflow in the streams, or the flow in the stream can infiltrate back into the
groundwater.
MIKE SHE
The MIKE SHE User Interface
In the main simulation specification dialogue, you select the processes that
you would like to include in your model. For the main water movement
processes, you can also select the numerical solution method. In general,
the simpler methods will require less data and run more quickly. Your
choice here will be immediately be reflected in the data tree.
Water Quality
In this dialogue, you can also chose to simulate water quality. If you turn
on the water quality, then several additional items will be added to the data
tree. Also, you will be able to chose to simulate water quality using either
the full advection-dispersion method for multiple species including sorption and decay. Or, you can chose to simulate water quality using the random walk particle tracking method.
You can also do water quality scenario analysis by using a common water
movement simulation and defining only the water quality parameters. The
common water movement simulation is defined by first unchecking the
Use current WM simulation for Water Quality checkbox.
The Technical Reference contains detailed information on the numerical
methods that can be selected from this dialogue:
Getting Started
z
Overland Flow - Reference (V.2 p. 265)
z
Channel Flow - Reference (V.2 p. 287)
35
Building a MIKE SHE Model
2.2.5
z
Evapotranspiration - Reference (V.2 p. 295)
z
Unsaturated Flow - Reference (V.2 p. 319)
z
Saturated Flow - Reference (V.2 p. 355)
z
Particle Tracking-Reference (V.2 p. 435)
z
Advection Dispersion - Reference (V.1 p. 739)
Simulation parameters
Once you have selected your processes, then there are several simulation
parameters that need to be defined. None of these are initially critical and
the default values are generally satisfactory initially. You can come back
to all of these at any time.
However, we recommend that you set up you simulation period when you
first create your model. The simulation period is used to verify all of you
time series data to make sure that your time series cover your simulation
period. You can still add your time series files, but if your simulation
period is not correct, then you will get a warning message in the message
field at the bottom of the page and the time series graphs will not display
the proper portion of the time series.
In MIKE SHE, all of the simulation input and output is in terms of real
dates, which makes it easy to coordinate the input data (e.g. pumping
rates), the simulation results (e.g. calculated heads) and field observations
(e.g. measured water levels).
Solver parameters
The default solver parameters for each of the processes are normally reasonable and there is usually no reason to change these unless you have a
problem with convergence or if the simulation is taking too long to run.
For more information on the solver parameters, you should see the individual help sections for the different solvers:
z
OL Computational Control Parameters (V.2 p. 36)
z
UZ Computational Control Parameters (V.2 p. 41)
z
SZ Computational Control Parameters (V.2 p. 42)
Time step control
Likewise, the time step control is important, but the default values are usually reasonable to get your model up and running. Then, you should go
back to the Time Step Control (V.2 p. 31) dialogue to optimize your simulation time stepping. For more information on time step control, you can
36
MIKE SHE
Model domain and grid
go to the help section for the Time Step Control (V.2 p. 31) dialogue, or
see the Controlling the Time Steps (V.1 p. 161) section.
Note: Although the different hydrologic processes can run on different
time steps, the processes exchange water explicitly. Therefore, there are
restrictions on the relationship between the time steps in the processes. In
particular, the longer time steps must be even multiples of the shorter time
steps. In other words, a 24 hour groundwater time step can included four
6-hour unsaturated flow timesteps, which can each include three 2-hour
overland flow timesteps. See Time Step Control (V.2 p. 31) for more information.
2.2.6
Hot Starting from a previous simulation
Your MIKE SHE simulation can be started from a hot start file. A hot start
file is useful for simulations requiring a long warm up period or for generating initial conditions for scenario analysis. Hot starting can also be an
effective way to change parameters that are normally static (e.g. hydraulic
conductivity) during the model process.
To start a model from a previous model run, you must first save the hot
start data, in the Storing of Results (V.2 p. 183) dialogue. In this dialogue,
you specify the storing interval for hot start data. Then in the Simulation
Period (V.2 p. 29) dialogue, you can specify the hot start file and then
select from the available stored hot start times.
Hot start limitations
There are a few limitations and caveats with the hot start process.
2.3
z
The Water Quality simulations cannot be started from a hot start file.
z
There is no append function for the hot start results, so your simulation
will generate an independent set of results.
z
The pre-processed data does not reflect the hot-start information. The
pre-processed data is based on the specified input data, not the results
file from which the simulation will be started. This primarily affects
initial conditions.
Model domain and grid
Regardless of the components included in your model, the first real step in
your model development is to define the model area. On a catchment
scale, the model boundary is typically a topographic divide, a groundwater
divide or some combination of the two. In general, there are no constraints
on the definition of the model boundaries. However, the model boundaries
Getting Started
37
Building a MIKE SHE Model
should be chosen carefully, keeping in mind the boundary conditions that
will be used for both the surface water and groundwater components.
All other spatial data defined in the data tree, such as topography, is interpolated during pre-processing to the Model Domain and Grid.
You can define your model domain and the grid using either a DHI grid
file (dfs2 format) or a GIS shape file (.shp format).
Using a dfs2 file
If you define your model domain using a dfs2 grid file, then you must
define the cell values as follows:
z
Grid cells outside of the model domain must be assigned a delete value
- by default -1.0e-35.
z
Grid cells inside the model domain must be assigned a value of 1.
z
Grid cells on the model boundary must be assigned a value of 2.
This distinction between interior grid cells and boundary cells is to facilitate the definition of boundary conditions. For example, drainage flow can
be routed to external boundaries but not to internal boundaries.
Since the model domain is defined as part of the dfs2 file format, if you
want to change the extent of your model domain, you must edit the .dfs2
file. However, if you want to change the grid spacing, then it is probably
easier to create a new file.
The Model Domain and Grid does not have to have the same dimensions
(size and spacing) as other specified dfs2 files (e.g. Topography). However, if the other dfs2 input files are coincident, that is if the rows and columns align with one another, then an average of the cell values is used. If
the dfs2 files are not coincident, then the Bilinear Interpolation
(V.1 p. 355) method is used to determine the cell value.
Note: The dfs2 files for integer grid codes must be coincident with the
model grid. For more information on this see Integer Grid Codes
(V.1 p. 353).
Using an polygon shape (shp) file
It is much easier to define your Model Domain and Grid via a GIS polygon shape (.shp) file. In this case, the definition of integer code values is
taken care of internally. Once you have defined the polygon file to use,
then you specify the spatial extent and origin location of the model
domain and grid.
38
MIKE SHE
Topography
An important advantage of using a polygon for the model domain, is that
the number of rows and columns can be easily adjusted. See Using MIKE
SHE with ArcGIS (V.1 p. 345) for more information.
Creating dfs2 or shp files
There is a Create button next to the Browse button that opens a dialogue
where you can define a dfs2 grid file. This utility automatically creates the
grid file with the appropriate Item Type.
In this dialogue, you can specify the overall dfs2 grid dimensions and origin. After you have created the file, then you can open and edit the file in
the Grid Editor using the Edit button.
Geographic projections
MIKE SHE supports all available geographic projections. If you have
defined the domain using a dfs2 file, then the geographic projection is
defined in the dfs2 file. If you use polygon shape file, then you must
defined the projection in the Model Domain and Grid (V.2 p. 70) dialogue.
See Using MIKE SHE with ArcGIS (V.1 p. 345) for more information.
Note: All dfs2 and polygon shape files must use the same geographic
projection. Any inconsistencies in the projections will results in an error
during the pre-processing.
2.4
Topography
In MIKE SHE, the topography defines the upper boundary of the model.
The topography is used as the top elevation of both the UZ model and the
SZ model. The topography also defines the drainage surface for overland
flow.
Many of the elevation parameters can be defined relative to the topography by means of a checkbox in the dialogue, including
z
Lower Level (V.2 p. 194),
z
Upper Level (V.2 p. 194),
z
Initial Potential Head (V.2 p. 165), and
z
Drain Level (V.2 p. 176).
Depth parameters, such as ET Surface Depth (V.2 p. 141), are also measured from the topography.
Getting Started
39
Building a MIKE SHE Model
File Formats
Topography is defined from a digital elevation model (DEM) using either
a dfs2 grid file, a point theme shape (GIS) file, or an ASCII XYZ file.
Non-dfs2 files or dfs2 files that have a different grid definition than the
model grid are all interpolated to the grid defined in the Model Domain
and Grid.
The Bilinear Interpolation (V.1 p. 355) method is useful for interpolating
previously gridded DEM data. Whereas, the Triangular Interpolation
(V.1 p. 359) method is useful for contour data digitized from a DEM.
Inverse Distance (V.1 p. 361) is usually used for sparse or irregularly
spaced data.
ArcGIS Grid Files If you have an ArcGIS Grid DEM, this can be converted to a dfs2 file using the MIKE Zero Toolbox. For more information
see the Using MIKE SHE with ArcGIS (V.1 p. 345) section. Alternatively,
a dfs2 plug-in is available for ArcGIS, that allows you to read and write
dfs2 files directly in ArcGIS.
Surfer Grid Files Surfer Grid files can be saved as an ASCII XYZ file
and then interpolated in MIKE SHE.
Other DEM formats Most other DEM formats can be converted to either
an ArcGIS Grid file or an ASCII XYZ file. If you have special requirements or difficulty, please contact your local support office.
2.5
Climate
Climate is the driving force for the hydrologic cycle. Spatial variation in
solar radiation drives the weather resulting in evaporation, rainfall, and
snow.
2.5.1
Precipitation
Precipitation is the measured rainfall. You can specify the precipitation as
a rate, for example in [mm/hr], or as an amount, for example in [mm]. If
you use the amount method, MIKE SHE will automatically convert this to
a rate during the simulation.
If you use a rate, then the EUM Data Units (V.1 p. 329) must be “Precipitation” and the time series must be Mean Step Accumulated (V.1 p. 343).
If you use an amount, then the EUM Data Units must be “Rainfall” and the
time series must be Step Accumulated (V.1 p. 343).
40
MIKE SHE
Climate
The Precipitation Rate item comprises both a distribution and a value. The
distribution can be either uniform, station-based or fully distributed. If the
data is station-based then for each station a sub-item will appear where
you can enter the time series of values for the station.
2.5.2
Snow
If the Include snow melt (V.2 p. 76) checkbox is checked then rain accumulates as snow if the Air Temperature (V.2 p. 84) is below the Threshold
Melting Temperature (V.2 p. 88) (the temperature at which the snow starts
to melt - usually 0 C). If the air temperature is above the threshold, then
the snow will melt at the rate specified by the Degree-day Melting or
Freezing Coefficient (V.2 p. 89).
Dry snow acts like a sponge and does not immediately release melting
snow. Thus, melting snow is added to wet snow storage. When the amount
of wet snow exceeds the Maximum Wet Snow Fraction in Snow Storage
(V.2 p. 91), the excess is added to ponded water, which is then free to infiltrate or runoff.
More detailed information on the snow melt process can be found in the
on-line help for the individual dialogues and in the Snow Melt - Reference
(V.2 p. 289) section.
2.5.3
Evapotranspiration
The calculation of evapotranspiration uses meteorological and vegetative
data to predict the total evapotranspiration and net rainfall due to
z
Interception of rainfall by the canopy,
z
Drainage from the canopy to the soil surface,
z
Evaporation from the canopy surface,
z
Evaporation from the soil surface, and
z
Uptake of water by plant roots and its transpiration, based on soil moisture in the unsaturated root zone.
The primary ET model is based on empirically derived equations that follow the work of Kristensen and Jensen (1975), which was carried out at
the Royal Veterinary and Agricultural University (KVL) in Denmark. This
model is used whenever the detailed Richards equation or Gravity flow
methods are used in the Unsaturated zone.
In addition to the Kristensen and Jensen model, MIKE SHE also includes
a simplified ET model that is used in the Two-Layer UZ/ET model. The
Two-Layer UZ/ET model divides the unsaturated zone into a root zone,
Getting Started
41
Building a MIKE SHE Model
from which ET can occur and a zone below the root zone, where ET does
not occur. The Two-Layer UZ/ET module is based on a formulation presented in Yan and Smith (1994). Its main purpose is to provide an estimate
of the actual evapotranspiration and the amount of water that recharges the
saturated zone. It is primarily suited for areas where the water table is
shallow, such as in wetland areas.
The reference evapotranspiration (ET) is the rate of ET from a reference
surface with an unlimited amount of water. Based on the FAO guidelines,
the reference surface is a hypothetical grass surface with specific characteristics. The reference ET value is independent of everything but climate
and can be calculated from weather data. The FAO Penman-Monteith
method is recommended for determining the reference ET value.
The reference ET is multiplied by the Crop Coefficient to get the Crop
Reference ET. The Crop Coefficient is found in the Vegetation development table in the Vegetation database. If the vegetation database is not
used, then the Reference ET is the maximum ET rate.
The Reference Evapotranspiration item comprises both a distribution and
a value. The distribution can be either uniform, station-based or fully distributed. If the data is station-based then for each station a sub-item will
appear where you can enter the time series of values for the station.
2.5.4
Snow Melt
MIKE SHE includes a comprehensive snow melt module based on a modified degree-day method. Precipitation that occurs when the air temperature is below the freezing point accumulates as solid snow and does not
infiltrate or contribute to runoff. The accumulated snow has a moisture
content, and when the moisture content reaches a critical level, then additional melting contributes to runoff.
Air Temperature
For snow melt, the air temperature is critical. However, the air temperature changes significantly with elevation. In areas with significant elevation changes, snow will accumulate in upland areas - often were there is
limited weather data available. The elevation correction for air temperature allows you to specify an elevation for the temperature stations and a
temperature change rate with elevation. During the pre-processing, a temperature change factor is calculated for each cell and the actual temperature in the cell is calculated during the simulation using this factor.
In terms of snow melt, the air temperature along with the degree day melting coefficient determine the amount of melting that can occur. If you
42
MIKE SHE
Land Use
have daily temperature data it may be difficult to properly account for the
diurnal melting and freezing cycles.
Air temperature can also be an important parameter during water quality
simulations.
For more information on the snow melt parameters, see the specific snow
melt dialogue information in the Climate (V.2 p. 76) section of the on-line
help and User Interface manual
2.6
Land Use
The land surface plays a very important role in hydrology. In principle, the
land use section is used to define the properties of the land surface. The
most important of these is the distribution of vegetation, which is used by
MIKE SHE to calculate a the spatial and temporal distribution of actual
evapotranspiration.
However, the land surface comes into play in many ways and other sections of the data tree also include properties related to land use. Some of
these properties are related to the vegetation distribution, and may even be
spatially identical. For example:
Topography - The topography is a physical property of the land surface
that defines the hydraulics of both the overland flow and the unsaturated
flow. See Topography (V.1 p. 39). Related to topography is the definition
of Subcatchments (V.2 p. 73), which is needed when you are using the
Linear Reservoir method for groundwater or the simple, catchment based
overland flow method.
Flood zones - In MIKE SHE, flood zones can be defined relative to the
MIKE 11 branches using Flood codes. For details on how use Flood codes
see the chapter on Surface Water (V.1 p. 169).
Hydraulic properties - The properties related directly to overland sheet
flow are found under Overland Flow (V.1 p. 47). This includes the Manning number (V.2 p. 118) or surface roughness and the Detention Storage
(V.2 p. 119), both of which are influenced or even defined by the vegetation.
Hydraulic flow - Areas of the land surface can be hydraulically divided
by man-made structures, such as road ways and embankments, which can
be defined by Separated Flow Areas (V.2 p. 123).
Getting Started
43
Building a MIKE SHE Model
Infiltration properties - The infiltration rate is a property of the soil type,
which may be modified by the land use. Related to the gross infiltration
rate is the presence or absence of macropores and other soil features leading to rapid infiltration. Both of these properties are found in the Unsaturated Flow (V.1 p. 52) section. However, land surface sealing and
compaction can be defined as a reduced contact between ponded water
and the subsurface. This is defined in the Overland flow section as a Surface-Subsurface Leakage Coefficient (V.2 p. 121).
Groundwater drainage - As the groundwater table rises, it intersects low
lying topographic features, such as ditches, or other man-made drainage
features, such as buried farm drains. These features are related to land use,
but are specified as Groundwater Drainage (V.1 p. 60)
Paving - Paved areas are treated as a drainage feature for ponded water not rainfall. The paved drainage function is part of the Land use section
but requires Groundwater drainage to be defined and depends completely
on the land use functions above that affect ponding of water.
2.6.1
Vegetation
By default, the only section under Land use is the vegetation distribution
The vegetation properties are used to calculate the actual evapotranspiration from crop reference evapotranspiration defined under Climate.
The primary vegetation properties are Leaf Area Index (LAI) and Root
Depth (RD). The LAI and Root Depth can be specified directly as a time
series. Or, they can be defined as a crop rotation in the Vegetation Properties Editor (V.2 p. 235).
A good source of local information on LAI and root depth is the agronomy
department at your local university.
Leaf Area Index
The LAI is defined as the area of leaves per area of ground surface. The
LAI values are characteristic of the plant type, season, and plant stress.
LAI values are widely available in the literature for most major plant
types.
The LAI is a lumped parameter for a cell that defines the average leaf area
of the cell. In forests, it includes both the leaf area of the forest canopy and
the understory. In more open areas, it is an average for all vegetation
types, such as grass, brush and trees. In areas of largely open water the
LAI is usually zero. If the LAI is zero, there will be no interception storage and no water will be removed from the unsaturated zone.
44
MIKE SHE
Land Use
Root Depth
Root depth is defined as the depth below ground in millimetres to which
roots extend. The root depth is not necessarily the average root depth. In
some cases it may be the maximum root depth. The root depth defines the
depth at which water can be extracted from the unsaturated zone. If the
root depth is deeper than the depth of the capillary zone, then the roots
will be able to extract water from the saturated zone. The thickness of the
capillary zone is defined by the pedotransfer function in the soil properties
for the Richards and Gravity flow methods. In the 2Layer UZ method, the
thickness of the capillary zone is defined by the ET Surface Depth
(V.2 p. 141). If you are using the Richards or Gravity Flow UZ methods,
then you will also be able to use the Root Shape factor (AROOT) for each
vegetation type. This allows you, for example, to extract more water from
the upper UZ cells than the lower cells, which is typical of grasses in
semi-arid climate zones.
2.6.2
Paved areas
The paved area function allows you to drain rainfall directly to the MIKE
11 network. The paved area function is rather complex and restricted in
several important ways. Most importantly, the paved area function
z
requires that the SZ drainage function be turned on, which means that
it only works when you are using the Finite Difference SZ method, and
z
discharges only to river links - not internal depressions or boundaries.
If you turn on the paved area function, then you can enter the paved runoff
coefficient, which is the fraction of the land surface that is paved.
There are two options for the paving function. There is an optional check
on the water level in the discharge point. If the discharge water level is
higher than or equal to the ponded water level, then no water will be discharged. The second option is that you can limit the discharge rate from
paved areas.
For more details on the paved area function, see Paved Area Drainage
(V.1 p. 174).
2.6.3
Irrigation
The irrigation module allows you to simulate the transfer of irrigation
water from multiple sources to multiple control areas.
The available sources include: shallow wells distributed across the cell,
deep bores with defined screen intervals, river stretches defined by an
upstream and downstream chainage, and external sources. The sources
Getting Started
45
Building a MIKE SHE Model
can be defined in a hierarchy, such that when one is unavailable, the water
will be removed from the next.
Irrigation is applied in control areas. Each control area is defined by an
area and a control function that defines when and how much water will be
applied.
For more information on Irrigation, see
2.7
z
Irrigation Command Areas (V.2 p. 103)
z
Irrigation Demand (V.2 p. 111)
z
Irrigation Priorities (V.2 p. 113)
Channel Flow
In the Rivers and Lakes dialogue (below) you can link MIKE SHE to a
MIKE 11 model.
The River Simulation File (.sim11) is the main MIKE 11 simulation file,
which contains the file references to all the files used in the MIKE 11
model. For MIKE SHE, the primary MIKE 11 files are:
46
z
the simulation control file (.sim11),
z
the river network file (.nwk11),
z
the cross-section database (.xns11),
z
the boundary condition file (.bnd11) and
z
the hydrodynamic setup file (.hd11).
MIKE SHE
Overland Flow
In the Rivers and Lakes dialogue, there are two Inundation Areas options
These options are always available for input, but are only used if you have
selected specific options in the MIKE SHE Links dialogue in the MIKE 11
Network Editor. These options are
z
Flood codes - a map used for the direct inundation of flooded areas in
MIKE SHE based on water levels in MIKE 11, and
z
Bathymetry - a detailed topography file that can be used to modify the
defined topography with a more detailed flood plain topography in
areas where Flood Codes have been defined.
Integrating a MIKE SHE and a MIKE 11 model is not very different from
establishing a stand-alone MIKE 11 HD model and a stand-alone MIKE
SHE model. In principle there are three basic set-up steps:
1 Establish a MIKE 11 HD hydraulic model as a stand-alone model and
make a performance test and, if possible, a rough calibration using prescribed inflow and stage boundaries. You can also specify a default
groundwater table (e.g. MIKE SHE’s initial groundwater level) and
leakage coefficients for any leakage calculations.
2 Establish a MIKE SHE model that includes the overland flow component and (optionally) the saturated zone and unsaturated zone components. An SZ drainage boundary can be used to prevent excessive
surface flows in low lying areas and the river flood plain.
3 Couple MIKE SHE and MIKE 11 by defining branches (reaches)
where MIKE 11 HD should interact with MIKE SHE. Modify your
MIKE SHE and MIKE 11 models so that they work together properly.
For example, by removing the specified groundwater table in MIKE 11
and adjusting your SZ drainage elevations if you used these in Step 2.
Detailed information on developing your surface water model, specifying
flow on flood plains, and coupling to MIKE 11 is in the chapter Surface
Water (V.1 p. 169).
Additional documentation on MIKE 11 can be found in the MIKE 11 User
Guide.
2.8
Overland Flow
Overland flow simulates the movement of ponded surface water across the
topography. It can be used for calculating flow on a flood plain or runoff
to streams.
Getting Started
47
Building a MIKE SHE Model
You can run the Overland flow module separately, or you can combine it
with any of the other modules. However, overland flow is required when
you are using MIKE 11 in MIKE SHE, as the overland flow module provides lateral runoff to the rivers.
The Simplified Overland Flow Routing (V.2 p. 279) method can be used
for regional applications when detailed flow is not required. This method
assumes that ponded water in the upland areas of a subcatchment flows
into the flood plain areas of the subcatchment, which in turn discharges
uniformly into the stream network located in the subcatchment.
The Finite Difference Method (V.2 p. 265) uses the diffusive wave approximation and should be used when you are interested in calculating local
overland flow and runoff. There are two solution methods available.
– Successive Over-Relaxation (SOR) Numerical Solution (V.2 p. 270)
– Explict Numerical Solution (V.2 p. 271)
The choice of method is a tradeoff between accuracy and solution time.
The SOR solver is generally faster because it can run with larger time
steps. The Explict method is generally more accurate than the SOR
method, but is often constrained to smaller time steps. The time step constraint prevents flow from crossing a cell in a single time step. The time
step constraint is determined by the cell with the highest velocity and
applied to the entire model in the current time step.
The Explicit method is generally used when the river is allowed to spill
from MIKE 11 onto the flood plain. Alternatively, you can use Flood
codes (V.2 p. 116) to inundated flood plain areas based on the water level
in MIKE 11.
The Multi-grid overland flow option allows you to take advantage of
detailed DEM information if it is available. The multi-grid method, subdivides the overland flow cell into an even number of sub-cells. The gradients between the cells and the flow area between cells water surface elevation in the cell is then calculated based on the volume of water and the
detailed topography information.
In MIKE SHE, the calculation of 2D overland flow can become a very
time consuming part of the simulation. So, you need to be very careful
when setting up your model to minimize the calculation of overland flow
between cells when it is unnecessary.
Detailed information on Overland Flow, the coupling between MIKE 11
and MIKE SHE and the overbank spilling options, and ways to optimize
48
MIKE SHE
Overland Flow
the calculation of overland flow can be found in the chapter Surface Water
(V.1 p. 169).
Mannings M
The Manning M is equivalent to the Stickler roughness coefficient, the use
of which is described in Overland Flow - Reference (V.2 p. 265).
The Manning M is the inverse of the more conventional Mannings n. The
value of n is typically in the range of 0.01 (smooth channels) to 0.10
(thickly vegetated channels). This corresponds to values of M between
100 and 10, respectively. Generally, lower values of Mannings M are used
for overland flow compared to channel flow.
If you don’t want to simulate overland flow in an area, a Mannings M of 0
will disable overland flow. However, this will also prevent overland flow
from entering into the cell.
Detention Storage
Detention Storage is used to limit the amount of water that can flow over
the ground surface. The depth of ponded water must exceed the detention
storage before water will flow as sheet flow to the adjacent cell. For example, if the detention storage is set equal to 2mm, then the depth of water on
the surface must exceed 2mm before it will be able to flow as overland
flow. This is equivalent to the trapping of surface water in small ponds or
depressions within a grid cell.
If you have static ponded water in an area and you do not want to calculate
overland flow between adjacent cells (can be slow), then you can set the
detention storage to a value greater than the depth of ponding.
Water trapped in detention storage continues to be available for infiltration
to the unsaturated zone and to evapotranspiration.
Initial and Boundary Conditions
In most cases it is best to start your simulation with a dry surface and let
the depressions fill up during a run in period. However, if you have significant wetlands or lakes this may not be feasible. However, be aware that
stagnant ponded water in wetlands may be a significant source of numerical instabilities or long run times.
The outer boundary condition for overland flow is a specified head, based
on the initial water depth in the outer cells of the model domain. Normally,
the initial depth of water in a model is zero. During the simulation, the
water depth on the boundary can increase and the flow will discharge
Getting Started
49
Building a MIKE SHE Model
across the boundary. However, if a non-zero value is used on the boundary, then water will flow into the model as long as the internal water level
is lower than the boundary water depth. The boundary will act as an infinite source of water.
If you need to specify time varying overland flow boundary conditions,
you can use the Extra Parameter option Time-varying Overland Flow
Boundary Conditions (V.1 p. 302).
Separated flow areas
The Separated Flow Areas (V.2 p. 123) are typically used to prevent overland flow from flowing between cells that are separated by topographic
features, such as dikes, that cannot be resolved within a the grid cell.
If you define the separated flow areas along the intersection of the inner
and outer boundary areas, MIKE SHE will keep all overland flow inside
of the model - making the boundary a no-flow boundary for overland flow.
Multi-cell overland flow
The main idea behind the 2D, multi-cell solver is to make the choice of
calculation grid independent of the topographical data resolution. The
approach uses two grids:
z
One describing the rectangular calculation grid, and
z
The other representing the fine bathymetry.
The standard methods used for 2D grid based solvers do not make a distinction between the two. Thus, only one grid is applied and this is typically chosen based on a manageable calculation grid. The available
topography is interpolated to the calculation grid, which typically does not
do justice to the resolution of the available data. The 2D multi-grid solver
in MIKE SHE can, in effect, use the two grids more or less independently.
In the Multi-cell overland flow method, high resolution topography data is
used to modify the flow area used in the St Venant equation and the courant criteria. The method utilizes two grids - a fine-scale topography grid
and a coarser scale overland flow calculation grid. However, both grids
are calculated from the same reference data - that is the detailed topography digital elevation model.
In the Multi-cell method, the principle assumption is that the volume of
water in the fine grid and the coarse grid is the same. Thus, given a volume of water, a depth and flooded area can be calculated for both the fine
grid and the coarse grid.
50
MIKE SHE
Overland Flow
In the case of detention storage, the volume of detention storage is calculated based on the user specified depth and OL cell area.
During the simulation, the cross-sectional area available for flow between
the grid cells is an average of the available flow area in each direction
across the cell. This adjusted cross-sectional area is factored into the diffusive wave approximation used in the 2D OL solver. For numerical details
see Multi-cell Overland Flow Method (V.2 p. 275) in the Reference manual.
The multi-grid overland flow solver is typically used where an accurate
bathymetric description is more important than the detailed flow patterns.
This is typically the case for most inland flood studies. In other words, the
distribution of flooding and the area of flooding in an area is more important than the rate and direction of ingress.
The multi-grid option is described in more detail in the chapter Multi-cell
Overland Flow (V.1 p. 181).
Overland Flow Performance
Calculation of overland flow can be a significant source of numerical
instabilities in MIKE SHE. Depending on the model setup, the overland
flow time step can become very short - making the simulation time very
long.
The chapter Surface Water in MIKE SHE (V.1 p. 171) contains many more
details on simulating overland flow and the coupling to MIKE 11. In particular the section Overland Flow Performance (V.1 p. 178) contains
detailed information on improving the performance of the overland flow
in your model.
MIKE FLOOD
MIKE SHE provides a useful means to simulate 2D flooding on a flood
plain that includes the influence of infiltration and evapotranspiration.
However, the detailed simulation of surface water flow paths and velocities on a flood plain can be very difficult. If you need to simulate more
complex flood plain flow, for example the impact of flood plain structures
and embankments, you may need to use MIKE FLOOD instead of MIKE
SHE.
MIKE FLOOD is combination of the 2D MIKE 21 surface water model
for detailed, accurate flow on the flood plain, and MIKE 11 for channel
flow. MIKE FLOOD allows you to define flood plain structures such as
embankments and culverts that can have very significant impacts on flow
velocity and direction. MIKE FLOOD can also more accurately simulate
Getting Started
51
Building a MIKE SHE Model
flood wave propagation on a surface simply because of the higher order
numerical method used.
2.9
Unsaturated Flow
Unsaturated flow is one of the central processes in most model applications. The unsaturated zone is usually heterogeneous and characterized by
cyclic fluctuations in the soil moisture as water is replenished by rainfall
and removed by evapotranspiration and exchange to the groundwater
table.
Unsaturated flow is primarily vertical since gravity plays the major role
during infiltration. Therefore, unsaturated flow in MIKE SHE is calculated only vertically in one-dimension, which is sufficient for most applications. However, this assumption may not be valid, for example, on steep
hill slopes.
There are three options in MIKE SHE for calculating vertical flow in the
unsaturated zone:
z
the full Richards equation, which is the most computationally intensive, but also the most accurate when the unsaturated flow is dynamic;
z
a simplified gravity flow procedure, which ignores capillary forces,
and is suitable when you are primarily interested in the time varying
recharge and not the dynamics in the unsaturated zone; and
z
a simple two-layer water balance that is suitable when the water table
is shallow and groundwater recharge is primarily influenced by evapotranspiration in the root zone.
More detailed information on the setup and calculation of unsaturated
flow is found in the chapter Unsaturated Groundwater Flow (V.1 p. 243).
The Technical Reference manual includes detailed information on the calculation methods - Unsaturated Flow - Reference (V.2 p. 319).
2.9.1
52
Soil Profiles
The unsaturated zone usually includes several different soil types. For
example, the soil profile could include a compacted upper zone or a loamy
active layer with lots of humus and other organic matter. The lower layers
could be alluvial zones with interbedded clay lenses, or less weathered
bedrock layers.
MIKE SHE
Unsaturated Flow
The soil profile that you define can be as detailed as the available information. There is no restriction on the amount of detail that you can input.
However, from a practical point of view, you are probably better off
grouping similar soil types together and simplifying the soil profiles as
much as possible.
The specified soil profile depth must be deeper than the vertical discretization.
In the 2-Layer UZ method, the soil profile is uniform with depth.
Soil properties database
The soil properties database is used to define the unsaturated flow properties and relationships for the different soil types, if you are using one of
the finite difference UZ methods (i.e. the Richards Equation and Gravity
methods). In the database, each soil type has a set of properties, and the
profile is composed of different soil types.
Vertical Grid Discretisation
The vertical discretisation of the soil profile typically contains small cells
near the ground surface and increasing cell thickness with depth. However, the soil properties are averaged if the cell boundaries and the soil
property definitions do not align.
The discretisation should be tailored to the profile description and the
required accuracy of the simulation. If the full Richards equation is used
the vertical discretisation may vary from 1-5 cm in the uppermost grid
points to 10-50 cm in the bottom of the profile. For the Gravity Flow module, a coarser discretisation may be used. For example, 10-25 cm in the
upper part of the soil profile and up to 50-100 cm in the lower part of the
profile. Note that at the boundary between two blocks with different cell
heights, the two adjacent boundary cells are adjusted to give a smoother
change in cell heights.
2.9.2
Initial Conditions
The default initial conditions for unsaturated flow are usually good, which
means that initially there is no flow in the soil column. This means that the
initial soil moisture content is based on the defined pressure-saturation
relationship.
If the 2-Layer UZ method is chosen, then the initial conditions are automatically defined by the method.
Getting Started
53
Building a MIKE SHE Model
2.9.3
Macropore flow
Macropores include vertical cracks, as well as worm and root holes in the
soil profile. Macropores increase the rate of infiltration through the soil
column.
Simple bypass flow - A simple empirical function is used to describe simple bypass flow in macropores. The infiltration water is divided into one
part that flows through the soil matrix and another part, which is routed
directly to the groundwater table, as bypass flow.
The bypass flow is calculated as a fraction of the net rainfall for each UZ
time step. Typically, macropore flow is highest in wet conditions when
water is flowing freely in the soil (e.g. moisture content above the field
capacity, θFC) and zero when the soil is very dry (e.g. moisture content at
the wilting point, θWP).
Simple bypass flow is commonly used to provide some rapid recharge to
the groundwater table. In many applications, if all the rainfall is infiltrated
normally, the actual evapotranspiration is too high and very little infiltration reaches the groundwater table. In reality some infiltration recharges
the groundwater system due to macropores and sub-grid variability of the
soil profile. In other words, there is usually sub-areas in a grid cell with
much higher infiltration rates or where the unsaturated zone thickness is
much less than that defined by the average topography in the cell.
Simple bypass flow is described in the Reference section under Simplified
Macropore Flow (bypass flow) (V.2 p. 334).
Full Macropore Flow - Macropores are defined as a secondary, additional
continuous pore domain in the unsaturated zone. Full macropore flow is
generally reserved for very detailed unsaturated root-zone models, especially in water quality models where solute transformations are occurring
in the macropores. Full bypass flow is described in the Reference section
under Full Macropore Flow (V.2 p. 336).
2.9.4
Green and Ampt infiltration
The Green and Ampt algorithm is an analytical method to increase infiltration in dry soils due to capillarity. It is not applicable when using the
Richards Equation method because capillarity is already included. However, when capillarity is not included (i.e. in the Gravity flow and 2-Layer
methods), dry soils will absorb rainfall at a much higher rate than the
defined infiltration rate (saturated hydraulic conductivity).
For more information on the Green and Ampt method, see the section
Green and Ampt Infiltration (V.2 p. 340) in the Reference Guide.
54
MIKE SHE
Unsaturated Flow
2.9.5
UZ Column Classification
Calculating unsaturated flow in all grid squares for large-scale applications can be time consuming. To reduce the computational burden MIKE
SHE enables you to compute the UZ flow in a reduced subset of grid
squares. The subset classification is done automatically by the preprocessing program according to soil and, vegetation distribution, climatic
zones, and depth to the groundwater table.
Column classification can decrease the computational burden considerably. However, the conditions when it can be used are limited. Column
classification is either not recommended or not allowed when
z
the water table is very dynamic and spatially variable because the classification is not dynamic,
z
if the 2 layer UZ method is used because the method is fast and the
benefit would be limited,
z
if irrigation is used in the model because irrigation zones are not a classification parameter, and
z
if flooding and flood codes are used, since the depth of ponded water is
not a classification parameter
Thus, the column classification should probably be avoided today
because the models have become more complex, MIKE SHE has become
more efficient and computers have become faster.
If the classification method is used, then there are three options for the
classification:
Getting Started
z
Automatic classification With automatic option, the UZ columns are
divided up based on the internal classification rules. The depth to the
water table, Groundwater Depths used for UZ Classification
(V.2 p. 136), is the lower UZ boundary condition.
z
Specified classification With the specified option, you must supply a
list of grid codes, Specified classification (V.2 p. 138), that defines the
computational column and the columns to which the results will be
applied.
z
Calculated in all Grid points (default) In many models the classification system is not feasible or recommended. In this case, the UZ flow
will be calculated in all soil columns.
55
Building a MIKE SHE Model
z
2.9.6
Partial Automatic Finally a combination of the Automatic classification and the Specified classification is available, where an. Integer Grid
Code file must be provide (see Partial automatic classification
(V.2 p. 137)) to define the different areas.
Coupling Between Unsaturated and Saturated Zone
A correct description of the recharge process is rather complicated
because the water table rises as water enters the saturated zone and affects
flow conditions in the unsaturated zone. The actual rise of the groundwater table depends on the moisture profile above the water table, which is a
function of the available unsaturated storage and soil properties, plus the
amount of net groundwater flow (horizontal and vertical flow and
source/sink terms).
The main difficulty in describing the linkage between the two the saturated (SZ) and unsaturated (UZ) zones arises from the fact that the two
components (UZ and SZ) are explicitly coupled (i.e. they run in parallel
and exchange water only at specific times). Explicit coupling of the UZ
and SZ modules is used in MIKE SHE to allow separate time steps that are
representative of the UZ (minutes to hours) and the SZ (hours to days)
domains.
Error in the mass balance originates from two sources:
z
keeping the water table constant during a UZ time step, and
z
using an incorrect estimate of the specific yield, Sy, in the SZ-calculations.
In the first case above, mass balance and convergence problems can be
addressed by making the maximum UZ time step closer to the SZ time
step.
In the second case above, the MIKE SHE forces the specific yield of the
top SZ layer to be equal to the “specific yield” of the UZ zone as defined
by the difference between the specified moisture contents at saturation, θs,
and field capacity, θfc.This correction is calculated from the UZ values in
the UZ cell in which the initial SZ water table is located. For more information see Specific Yield of the upper SZ numerical layer (V.1 p. 252).
UZ - SZ limitations
The coupling between UZ and SZ is limited to the top calculation layer of
the saturated zone. This implies that:
56
MIKE SHE
Saturated Groundwater Flow
z
As a rule of thumb, the UZ soil profiles should extend to just below the
bottom of the top SZ layer.
z
However, if you have a very thick top SZ layer, then the UZ profiles
must extend at least to below the deepest depth of the water table.
z
If the top layer of the SZ model dries out, then the UZ model usually
assumes a lower pressure head boundary equal to the bottom of the
uppermost SZ layer.
z
All outflow from the UZ column is always added to the top node of the
SZ model.
z
UZ nodes below the water table and the bottom of the top SZ layer are
ignored.
For more detailed information on the UZ-SZ coupling see Unsaturated
Flow - Reference (V.2 p. 319). The chapter, Unsaturated Groundwater
Flow (V.1 p. 243), also contains more detailed information on the setup
and evaluation of the unsaturated model.
2.10
Saturated Groundwater Flow
The Saturated Zone (SZ) component of MIKE SHE calculates the saturated subsurface flow in the catchment. In MIKE SHE, the saturated zone
is only one component of an integrated groundwater/surface water model.
The saturated zone interacts with all of the other components - overland
flow, unsaturated flow, channel flow, and evapotranspiration.
By comparison, MODFLOW only simulates saturated groundwater flow.
All of the other components are either ignored (e.g. overland flow) or are
simple boundary conditions for the saturated zone (e.g. evapotranspiration). On the other hand, there are very few difference between the MIKE
SHE numerical engine and MODFLOW. The differences are limited to the
discretisation and to some differences in the way some of the boundary
conditions are defined.
Finite Difference Method
When the Finite Difference method has been selected, MIKE SHE allows
for a fully three-dimensional flow in a heterogeneous aquifer with shifting
conditions between unconfined and confined conditions. The spatial and
temporal variations of the dependent variable (the hydraulic head) is
described mathematically by the 3-dimensional Darcy equation and
solved numerically by an iterative implicit finite difference technique.
MIKE SHE includes two groundwater solvers - the SOR groundwater
solver based on a successive over-relaxation solution technique and the
Getting Started
57
Building a MIKE SHE Model
PCG groundwater solver based on a preconditioned conjugate gradient
solution technique.
Linear Reservoir Method
The linear reservoir module for the saturated zone in MIKE SHE was
developed to provide an alternative to the physically based, fully distributed model approach. In many cases, the complexity of a natural catchment area poses a problem with respect to data availability, parameter
estimation and computational requirements. In developing countries, in
particular, very limited information on catchment characteristics is available. Satellite data may increasingly provide surface data estimates for vegetation cover, soil moisture, snow cover and evaporation in a catchment.
However, subsurface information is generally very sparse.
The linear reservoir method for the saturated zone may be viewed as a
compromise between limitations on data availability, the complexity of
hydrological response at the catchment scale, and the advantages of model
simplicity.
For example, combining lumped parameter groundwater with physically
distributed surface parameters and surface water often provides reliable,
efficient
2.10.1
z
Assessments of water balance and runoff for ungauged catchments,
z
Predictions of hydrological effects of land use changes, and
z
Flood prediction
Conceptual Geologic Model for the Finite Difference Approach
Before starting to develop a groundwater model, you should have developed a conceptual model of your system and have at your disposal digital
maps of all of the important hydrologic parameters, such as layer elevations and hydraulic conductivities.
In MIKE SHE you can specify your subsurface geologic model independent of the numerical model. The parameters for the numerical grid are
interpolated from the grid independent values during the preprocessing.
The geologic model can include both geologic layers and geologic lenses.
The former cover the entire model domain and the later may exist in only
parts of your model area.
You also have the option to set up your conceptual model
z
58
by layers, where you specify the property distribution in the layer, or
MIKE SHE
Saturated Groundwater Flow
z
by units, where you specify the unit distribution in the layer.
Lenses
In building a geologic model, it is typical to find discontinuous layers and
lenses within the geologic units. The MIKE SHE setup editor allows you
to specify such units - again independent of the numerical model grid.
Lenses are often useful when building up a geologic model where the
units are discontinuous. For example, a coarse alluvial flood plain aquifer
can be defined as a lense inside of a regional bedrock aquifer.
Lenses are specified by defining either a .dfs grid file or a polygon .shp
file for the extents of the lenses. The .shp file can contain any number of
polygons, but the user interface does not use the polygon names to distinguish the polygons. If you need to specify several lenses, you can use a
single file with many polygons and specify distributed property values, or
you can specify multiple individual polygon files, each with unique property values.
There are a number of special considerations when working with lenses in
the geologic model.
2.10.2
z
Lenses override layers - That is, if a lense has been specified then the
lense properties take precedence over the layer properties and a new
geologic layer is added in the vertical column.
z
Vertically overlapping lenses share the overlap - If the bottom of
lense is below the top of the lense beneath, then the lenses are assumed
to meet in the middle of the overlapping area.
z
Small lenses override larger lenses - If a small lense is completely
contained within a larger lense the smaller lense dominates in the location where the small lense is present.
z
Negative or zero thicknesses are ignored - If the bottom of the lense
intersects the top of the lense, the thickness is zero or negative and the
lense is assumed not to exist in this area.
Specific Yield of upper SZ layer
MIKE SHE forces the specific yield of the top SZ layer to be equal to the
“specific yield” of the UZ zone as defined by the difference between the
specified moisture contents at saturation, θs, and field capacity, θfc.This
correction is calculated from the UZ values in the UZ cell in which the initial SZ water table is located. This is reflected in the pre-processed data.
For more information on the SZ-UZ specific yield see Specific Yield of
the upper SZ numerical layer (V.1 p. 252).
Getting Started
59
Building a MIKE SHE Model
2.10.3
Numerical Layers
There is no restriction in MIKE SHE on the number of numerical layers in
the SZ model. However, there may be practical limitations depending on
your computer resources. As a rule of thumb, each additional SZ layer will
significantly slow down your simulation.
The upper boundary of the top layer is always either the infiltration/exfiltration boundary, which in MIKE SHE is calculated by the unsaturated
zone component or a specified fraction of the precipitation if the unsaturated zone component is excluded from the simulation.
The lower boundary of the bottom layer is always considered impermeable.
In MIKE SHE, the rest of the boundary conditions can be divided into two
types: Internal and Outer. If the boundary is an outer boundary then it is
defined on the boundary of the model domain. Internal boundaries, on the
other hand, must be inside the model domain.
The UZ model only interacts and exchanges water with the top SZ layer.
Therefore, the bottom of the top SZ layer is usually specified below the
lowest water table level, so that the top SZ layer always includes the water
table.
2.10.4
Groundwater Drainage
Saturated zone drainage is a special boundary condition in MIKE SHE
used to defined natural and artificial drainage systems that cannot be
defined in MIKE 11. It can also be used to simulate simple, lumped conceptual surface water drainage of groundwater.
Saturated zone drainage is removed from the layer of the SZ layer containing the drain level. Water that is removed from the saturated zone by
drains is routed to local surface water bodies, local topographic depressions, or out of the model. The amount of drainage is calculated based on
the groundwater head and the drain level using a linear reservoir formulation.
When water is removed from a drain, it is immediately moved to the recipient. In other words, the drain module assumes that the time step is longer
than the time required for the drainage water to move to the recipient.
Conceptually, you can use a “full pipe” analogy. The drain is a pipe full of
water. As groundwater is added to the pipe, an equivalent amount of water
must be discharged immediately out of the opposite end of the pipe
because the water is incompressible and there is no additional storage in
the pipe.
60
MIKE SHE
Saturated Groundwater Flow
Each cell requires a drain level and a time constant (which is the same as a
leakage factor). Both drain levels and time constants can be spatially
defined. A typical drainage level might be 1m below the ground surface
and a typical time constant may be between 1e-6 and 1 e-7 1/s.
Drainage reference system
MIKE SHE requires a reference system for linking the drainage to a recipient node or cell. There are four different options for setting up the drainage source-recipient reference system
2.10.5
z
Drain Levels The drainage recipient is calculated based on the drain
levels in all the down gradient cells. That is, the location of the recipient cell is calculated as if the drain water was flowing downhill (based
on the drain levels). This is the most common method of specifying
drainage routing and the default setting.
z
Drain Codes The drainage recipient is specified by the user based on a
distribution map of integer code values.
z
Distributed option With this option there are several different drainage possibilities, including a combination of Codes and Levels. The
Distributed option can also be used to define a specific MIKE 11 Hpoint or MOUSE manhole as a recipient.
z
Removed The fourth option is simply a head dependent boundary that
removes the drainage water from the model. This method does not
involve routing and is exactly the same as the MODFLOW Drain
boundary.
Groundwater wells
Groundwater wells can be included in your SZ simulation. The groundwater well locations, filter depth, pumping rates etc. are stored in a .wel file
that is edited using the Well editor (V.2 p. 229).
Getting Started
61
Building a MIKE SHE Model
2.10.6
Linear Reservoir Groundwater Method
In the linear reservoir method, the entire catchment is subdivided into a
number of subcatchments and within each subcatchment the saturated
zone is represented by a series of interdependent, shallow interflow reservoirs, plus a number of separate, deep groundwater reservoirs that contribute to stream baseflow.
The lateral flows to the river (i.e. interflow and baseflow) are by default
routed to the river links that neighbour the model cells in the lowest topographical zone in each subcatchment.
Interflow will be added as lateral flow to river links located in the lowest
interflow storage in each catchment. Similarly, baseflow is added to river
links located within the baseflow storage area
Three Integer Grid Code maps are required for setting up the framework
for the reservoirs,
z
a map with the division of the model area into Subcatchments,
z
a map of Interflow Reservoirs, and
z
a map of Baseflow Reservoirs.
The division of the model area into subcatchments can be made arbitrarily.
However, the Interflow Reservoirs must be numbered in a more restricted
manner. Within each subcatchment, all water flows from the reservoir
with the highest grid code number to the reservoir with the next lower grid
code number, until the reservoir with the lowest grid code number within
the subcatchment is reached. The reservoir with the lowest grid code
number will then drain to the river links located in the reservoir.
For baseflow, the model area is subdivided into one or more Baseflow
Reservoirs, which are not interconnected. However, each Baseflow Reservoir is further subdivided into two parallel reservoirs. The parallel reservoirs can be used to differentiate between fast and slow components of
baseflow discharge and storage.
For more detailed information on the Linear Reservoir method, see the
section Linear Reservoir Method (V.2 p. 373) in the Reference manual.
2.11
Storing of results
The integrated nature of MIKE SHE means that very large amounts of
output can be generated during a simulation. Thus, the output specifica-
62
MIKE SHE
Storing of results
tion is designed to allow you to save only the necessary information. However, the downside is that if you failed to save a specific output during the
simulation run, then you will have to re-run the simulation to obtain this
information.
The output in MIKE SHE can be divided into two types: Time series and
Grid Series. From a practical viewpoint, time series output generated during the simulation is saved at every simulation time step, whereas grid
series output is saved at a specified time interval. You can easily obtain
missing time series from a grid series output file, but the time resolution
will be the same as the specified saving interval.
Thus, at the locations where you want detailed results of a particular
value, you define a point in the Detailed Time Series dialogue. If you are
interested in the spatial and general temporal trends of a parameter, then it
is usually sufficient to save only the Grid Series output.
Water balance output
The water balance is often a vital part of assessing the results of a MIKE
SHE simulation. The water balance describes the flow of water within
your catchment.
If the water balance checkbox is turned on, then all of the data necessary
for calculating the water balance will be automatically saved. If you do not
check on this box, then you will not be able to calculate a water balance
for your simulation and you will have to re-run your simulation to generate the needed output data.
Water balances are calculated using a separate water balance utility, which
is described in detail in the chapter Using the Water Balance Tool
(V.1 p. 105).
Hot start output
It is often very useful to be able to start a simulation from a consisten predefined starting point. For example, you may want to simulate the first
five years and then start all of your scenarios from this starting point. This
could save you considerable calculation time.
You can append individual simulation output files together using the Concatination tool in the MIKE Zero Toolbox. However, you will not be able
to create a water balance of the entire period including the first five years.
Using the hot start involves:
z
Getting Started
Turning on the hot start by checking the hot start checkbox,
63
Building a MIKE SHE Model
z
Then either storing the hot start data at the end of the simulation only
(which will create only one possible hot start point), or
z
Storing the hot start information at regular storing intervals. Frequent
hot start storage can create very large files and may slow down the simulation as all of this data must be written to the hard disk.
Water quality output
If you want to run a water quality simulation after the water movement
simulation, then you must turn on the storing of the water quality output.
If the water quality is turned on the main Simulation Specification dialogue, then the water quality output is automatically stored during the
water movement simulation. Manual activation is only required if the
water movement simulation is being run separately.
Storing intervals
Storing intervals for both the water movement and the mass balance
define the frequency at which grid data is stored. Grid data is the most
space consuming output.
The grid output data is viewed in the Results Viewer and is used for calculating the water balance. Thus, you cannot calculate a water balance or
spatial output maps at a finer temporal resolution than the storing intervals. If you want detailed output of a specific parameter at more frequent
intervals, then you should use the Detailed Time Series Output function.
2.11.1
Detailed Time Series Output
The detailed time series output allows you to save any output parameter at
every time step of the particular process. Since the different processes run
at different time steps, you may get, for example, much more detailed output for the unsaturated zone than for the saturated zone.
Each item in the Detailed time series is displayed automatically in an
HTML format graph on the Run tab while the simulation is running.
You can also add observation data to each of the detailed time series items
A full list of available output items, as well as more detail on the individual items is found in the section Output Items (V.1 p. 72).
Importing ASCII data
Detailed MIKE SHE Time Series data can be imported directly into the
Detailed MIKE SHE Time Series dialogue using the Import button. The
data file must be a tab- delimited ASCII file without a header line. The file
must contain the following fields and be in the format specified below.
64
MIKE SHE
Storing of results
Name>data typeCode>NewPlot>X >Y >Depth>UseObsdata>dfs0Filename>dfs0ItemNumber
where the > symbol denotes the Tab character and
Name - is the user specified name of the observation point. This is the
name that will be used for the time series item in the Dfs0 file created
during the simulation.
data typeCode - This is a numeric code used to identify the output data
type. See the list of available Data Type Codes in Table 3.1 and
Table 3.2 under Output Items (V.1 p. 72).
NewPlot - This is a flag to specify whether a new detailed time series
HTML-plot will be created on the Results Tab:
0 = the output will be added to the previous plot.
1 = Create a new plot
X, Y - This is the (X, Y) map coordinates of the point in the same EUM
units (ft, m, etc.) as specified in the EUM Database for Item geometry
2-dimensional. (see EUM Data Units)
Depth -This is the depth of the observation point below land surface for
subsurface observation points. The value is in same EUM units (ft, m,
etc.) as specified in the EUM Database for Depth Below Ground (see
EUM Data Units). A depth value must always be included, even if not
needed.
UseObsData - This is a flag to specify whether or not an observation file
needs to be input: 0 = No; 1 = Yes
dfs0FileName - This is the file name of the dfs0 time series file with
observation data. The path to the dfs0 file must be relative to the directory containing the MIKE SHE *.she document. The .dfs0 extension is
added to the file name automatically and should be not be included in
the file name. For example, the following input line
.\Time\Calibration\GroundwaterObs
refers to the file GroundwaterObs.dfs0 located in the subdirectory Time\Calibration, which is found in the same directory as
the .she model document.
dfs0ItemNumber - This is the Item number of the observation data in the
specified DFS0 file.
Getting Started
65
Building a MIKE SHE Model
Import Example
The following is a simple example of a tab delimited ASCII file with two
MIKE SHE observation points, where the file containing the observations
is called obsdata.dfs0:
Obs_1
Obs_2
Obs_3
2.11.2
20
15
16
1234500. 456740.
1239700. 458900.
0241500. 459310.
0. 0
10. 1
20. 1
.\time\obsdata
.\time\obsdata
.\time\obsdata
1
2
3
Detailed MIKE 11 Time Series Output
MIKE 11 output is normally analysed using the MIKE View program.
However, the default MIKE 11 output is only at specified time intervals.
Every item in the Detailed MIKE 11 Time Series table is output at every
MIKE 11 time step.
Like the Detailed Time Series Output (above), each item in this table is
output automatically to an HTML graph in the Run Tab. You can also
specify an observation file for each item, which is more convenient that
using MIKE VIEW.
Importing ASCII data
Detailed MIKE 11 Time Series data can be imported directly into the
Detailed MIKE 11 Time Series dialogue using the Import button. The data
file must be a tab- delimited ASCII file without a header line. The file
must contain the following seven fields:
Name - is the user specified name of the observation point. This is the
name that will be used for the time series item in the Dfs0 file created
during the simulation.
data typeCode - This is a numeric code used to identify the output data
type (1=water level, 2=discharge).
Branch_name - The name of the MIKE 11 branch
Chainage - The location of the MIKE 11 h-point or q-point (the nearest
one will be taken within a tolerance).
UseObsData - This is a flag to specify whether or not an observation file
needs to be input: 0 = No; 1 = Yes
dfs0FileName - This is the file name of the dfs0 time series file with
observation data. The path to the dfs0 file must be relative to the directory containing the MIKE SHE *.she document. The .dfs0 extension is
66
MIKE SHE
Storing of results
added to the file name automatically and should be not be included in
the file name. For example, the following input line
.\Time\Calibration\GroundwaterObs
refers to the file GroundwaterObs.dfs0 located in the subdirectory Time\Calibration, which is found in the same directory as
the .she model document.
dfs0ItemNumber - This is the Item number of the observation data in the
specified DFS0 file.
2.11.3
Grid Series Output
The grid time series output allows you to save spatial output data at every
saved time step of the particular process.
Each item in the Grid time series table is listed on the Run tab. You can
open and plot each of these items while the simulation is running.
A full list of available output items, as well as more detail on the individual items is found in the section Output Items (V.1 p. 72).
Getting Started
67
Building a MIKE SHE Model
68
MIKE SHE
RESULTS AND CALIBRATION
69
70
MIKE SHE
Output Files
3
MIKE SHE RESULTS
The available output from MIKE SHE depends on the processes selected
in the Simulation Specification dialogue. Thus, for example, results for
Overland Flow only appear when Overland flow is being calculated.
3.1
Output Files
The output from MIKE SHE is stored in a combination of files.
.sheres - this is an ASCII file that is a catalogue of all the output files associated with a simulation.
.frf - this is a binary output file containing all of the static information on
the simulation, as well as all of the time series results that cannot be easily
stored in a dfs format.
dfs files - The rest of the output is stored in a series of dfs0, dfs2 and dfs3
files.
The dfs file format is a binary time series format. Each file can contain
multiple output items, but each of the items must be stored at the same
time step interval. Thus, the output for each of the processes that has an
independent storing time step is stored in separate output file (e.g. OL
water depth is stored separately from SZ Recharge, even though each is a
2D output item).
Viewing Output Files
The primary means of viewing the dfs2 and dfs3 output is the Results
Viewer. The gridded output files can be also viewed in the Grid Editor.
The Grid Editor includes icons in the icon bar to step between layers and
time steps, as well as to switch between output items.
Dfs0 output is viewed most easily in the Time Series Editor.
All three of these are MIKE Zero tools and are described in the MIKE
Zero documentation. See the section on The Results viewer (V.1 p. 81) for
more details on the Results Viewer.
3.1.1
Log files
There are three main log files (where the xx refers to your document file
name). All three of these files are found in the default results directory
along with the other result files.
Results and Calibration
71
MIKE SHE Results
xx_PP_Print.log - This is the main output file from the pre-processor.
xx_WM_Print.log - This is the main output from the water movement
engine.
xx_WQ_Print.log - This is the main output from the water quality engine.
3.2
Multiple simulations
There are several things to consider when running multiple MIKE SHE
simulations.
3.3
z
If you run simulations one after another, the results files will be overwritten unless you move or copy them first.
z
If you set up multiple simulations using the same MIKE 11 model, the
MIKE 11 results files will be overwritten. To prevent this, you must
create different .sim11 files and change the results file name in the
.sim11 file.
z
If you are starting from a Hot Start file, then you need to be careful that
the Hot start file you are using is the one you want. The easiest way to
ensure this is to change the name of the hotstart file.
z
You can run a chain of models - hot starting from the end of the last
simulation. This can be done using a batch command, for example.
You can concatenate the results files using the Concatenate Tool in the
MIKE Zero toolbox. This will allow you to build up a set of continuous
results files that includes the entire simulation. However, you will not
be able to create a continuous water balance because the .sheres file
and the .frf files will not be correct.
Output Items
Some of the available output items are calculated as part of another process. For example, the depth of overland water is calculated based on seepage to and from the groundwater and as part of the MIKE 11 surface water
calculations, even if the overland flow is not directly simulated.
Furthermore, some of the output items require that more than one process
be simulated. For example, the leaf area index is only available if both
evapotranspiration and unsaturated flow are calculated.
In the absence of an explicit remark, the sign convention for MIKE SHE’s
output is positive in the positive direction. In other words, all flows in the
72
MIKE SHE
Output Items
direction of increasing X, Y and Z coordinates are positive. Thus, vertical
downward flows, such as infiltration are negative.
Flows that do not have a direction are positive if storage or outflow is
increasing. Thus, all flows leaving the model are positive, and water balance errors are positive if the model is generating water.
Also important to remember is that the output items related to flow are
accumulated over the storing time step. In many cases, these values are
required for the Water Balance program described in the section Using the
Water Balance Tool (V.1 p. 105). The values that are part of the water balance are automatically turned on when the water balance option is
selected.
However, the output items that are not flows, such as temperature, water
depth and Courant number represent the current value at the end of the
storing time step.
Finally, some of the output items are actually input items. For example,
precipitation is usually input as a time series for several polygons or grid
code areas. The output file is a fully distributed dfs2 version of the input
time series files.
The available output items for gridded data and time series data are listed
in Table 3.1 and Table 3.2. Table 3.2 lists a number of additional output
items, such as the number of solver iterations, that can only be displayed
as a time series.
Code - In Table 3.1 and Table 3.2, a Data Type Code is needed when
importing time series items into the Detailed time series output
(V.2 p. 186) dialogue.
The following are some additional notes on the gridded output items
3.3.1
Overland flow velocity
The overland flow velocity in the list of available output items is used for
the water balance calculations. It is not the cell velocity.
The cell velocity cannot be directly calculated because the overland water
depth is an instantaneous value output at the end of storing time step.
The overland flow in the x- and y- directions are mean-step accumulated
over the storing time step. Thus, the overland flow is not the flow in the
cell, but rather the accumulated flow across the cell face on the positive
side of the cell.
Results and Calibration
73
MIKE SHE Results
You may be tempted to calculate a flow velocity from these values. But,
you can easily have the situation where the accumulated flow across the
boundary is non-zero, but at the end of the storing time step, the water
depth is zero. Or, you could have a positive inflow and a zero outflow,
which may be misleading when looking at a map of flow velocities.
The overland flow velocities are discussed in more detail in the section,
Overland Flow Velocities (V.1 p. 177).
3.3.2
Recharge
The data item Total recharge to SZ (positive for downwards flow) contains
the following items:
z
Exchange between UZ and SZ, calculated by the UZ solver
z
Recharge from Bypass or Macropores if included
z
Direct flow between SZ and overland (when groundwater table is
above ground)
z
Transpiration from SZ (when the roots reach the groundwater)
So neither baseflow (SZ-M11) nor drain flow is included. These items can
be found in the two data items:
z
- SZ exchange flow with river (positive when flow from SZ to M11,
negative the other way)
z
- SZ drainage flow from point
The Total recharge to UZ should correspond with the water balance items,
but note the sign. The easiest way to check this is to look at a Saturated
zone water balance, table type:
z
Recharge: exchange between UZ and SZ + Bypass flow or Macropore
recharge if included + direct flow between SZ and Overland + transpiration from SZ, all POSITIVE UPWARDS
z
Drain: Drainage flow from point
z
SZ->River: SZ exchange flow with river, positive for flow to the river
Note the various units. The total recharge result type is a flux (i.e. mm/d,
mm/h, m/s, etc) depending on the chosen user unit for Recharge. Whereas,
the SZ river exchange and Drainage are flows (i.e. m3/s or similar). The
Water balance output is in units of Storage depth (mm). That is, it is normalized with the catchment area (using the area inside the outer bound-
74
MIKE SHE
Output Items
ary), or the subcatchment area if a sub-catchment water balance has been
extracted.
3.3.3
Summary of all output items
The following table includes a summary of all output items for both the
gridded data and the time series data.
Table 3.1
Available output items for gridded data and time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion (Water Quality)
–PT - Particle Tracking
–SM - Snow melt
Code Output Item
Appears with these processes
10
precipitation rate
Always
This is the distributed actual precipitation in the model, accumulated per
storing time step.
128
average water content in the root zone
UZ+ET
2LUZ+ET
11
rooting depth
UZ+ET
2LUZ+ET
12
leaf area index
UZ+ET
2LUZ+ET
182
crop coefficient
UZ+ET
2LUZ+ET
15
actual evapotranspiration
UZ+ET
2LUZ+ET
16
actual transpiration
UZ+ET
2LUZ+ET
13
actual soil evaporation
UZ+ET
17
actual evaporation from interception
UZ+ET
2LUZ+ET
18
actual evaporation from ponded water
UZ+ET
2LUZ+ET
19
canopy interception storage
UZ+ET
2LUZ+ET
14
evapotranspiration from SZ
SZ+UZ+ET
SZ+2LUZ+ET
total snow storage
ET+SM
Dry snow storage
ET+SM
Wet snow storage
ET+SM
Wet snow storage fraction
ET+SM
Fraction of cell area covered by snow
ET+SM
Precipitation and Irrigation added to snow
ET+SM
Total snow converted to overland flow
ET+SM
Freezing due to air temperature
ET+SM
Melting due to air temperature
ET+SM
Results and Calibration
75
MIKE SHE Results
Table 3.1
Available output items for gridded data and time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion (Water Quality)
–PT - Particle Tracking
–SM - Snow melt
Code Output Item
Melting due to SW solar radiation
Appears with these processes
ET+SM
Melting due to energy in rain
ET+SM
99
Snow evaporation
ET+SM
61
depth of overland water
OL
This is the instantaneous depth of water at the end of the storing time step.
58
overland flow in x-direction
This is the flow across the boundary from celli to celli+1 in volume/time
e.g. m3/s.
OL
59
overland flow in y-direction
(this is the flow across the boundary from celli to celli+1 in volume/time
e.g. m3/s)
OL
SZ
SubOL
flow from flooded areas to river
Overland flow to MOUSE
External sources to Overland (for OpenMI)
62
paved area drainage to river or MOUSE
OL+M11/MOUSE +Drainage+
Paved
Overland water elevation
Mean OL wave courant number (explicit OL)
Max OL wave courant number (explicit OL)
Max output OL-OL per cell volume (explicit OL)
141
Water content in root zone (2-layer UZ)
2LUZ
142
Water content below root zone (2-layer UZ)
2LUZ
143
Maximum water content (2-layer UZ)
2LUZ
144
Minimum water content (2-layer UZ)
2LUZ
121
infiltration to UZ (negative)
UZ
2LUZ
SZ
LR
122
exchange from UZ to SZ (negative)
UZ
2LUZ
SZ
LR
UZ
2LUZ
bypass flow UZ (negative)
57
UZ deficit
infiltration to macropores (negative)
macropore recharge to SZ (negative)
76
MIKE SHE
Output Items
Table 3.1
Available output items for gridded data and time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion (Water Quality)
–PT - Particle Tracking
–SM - Snow melt
Code Output Item
37
average soil moisture content in top 5 compartments
Appears with these processes
LR+UZ
Total recharge to SZ (positive down)
This is a sum of:
+ exchange from UZ to SZ
+ bypass flow to SZ (if active)
+ macropore flow to SZ (if active)
+ flow between OL and SZ (if water table at/above topography)
+ transpiration from SZ
SZ drainage and exchange to MIKE 11 is NOT included in this term.
The Recharge is positive downwards and has the EUM units of precipitation rate, so that it can be used directly as input to another MIKE SHE
model in place of precipitation.
119
rate of change in UZ storage
UZ
123
epsilon calculated in UZ
UZ
120
accumulated error in UZ
(water balance in the UZ cells only)
UZ
2LUZ
column mean macropore water content above GW
column total net exchange matrix to macropores
column total exchange matrix to macropores
column total exchange macropores to matrix
45
groundwater feedback to the unsaturated zone
LR+UZ
117
unsaturated zone flow
UZ
118
water content in unsaturated zone
UZ
159
pressure head in unsaturated zone
UZ
129
root water uptake
UZ+ET
LR+2LUZ
macropore water content
root water uptake
20
irrigation: actual water content in root zone
135
irrigation: soil moisture deficit in root zone
UZ+ET+Irrigation
21
total irrigation
UZ+ET+Irrigation
26
irrigation from river
M11+UZ+ET+Irrigation
Results and Calibration
UZ+ET+Irrigation
77
MIKE SHE Results
Table 3.1
Available output items for gridded data and time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion (Water Quality)
–PT - Particle Tracking
–SM - Snow melt
Code Output Item
Appears with these processes
28
irrigation from wells
SZ+UZ+ET+Irrigation
22
irrigation from external source
UZ+ET+Irrigation
23
irrigation index
UZ+ET+Irrigation
24
irrigation shortage
UZ+ET+Irrigation
25
irrigation total demand
UZ+ET+Irrigation
153
sprinkler irrigation
UZ+ET+Irrigation
154
drip and sheet irrigation
UZ+ET+Irrigation
27
ground water extraction for irrigation
SZ+UZ+ET+Irrigation
106
depth to phreatic surface (negative)
SZ
101
head elevation in saturated zone
SZ
107
seepage flow SZ -overland
(the flow up from SZ onto the topography)
SZ
108
seepage flow overland - SZ (negative)
(the flow down into the saturated zone)
SZ
External inflow to SZ drain (for OpenMI)
113
3D UZ recharge to SZ (negative)
SZ+NegPrec
102
groundwater flow in x-direction
(a flow rate, e.g. in [m3/s]
SZ
103
groundwater flow in y-direction
(a flow rate, e.g. in [m3/s]
SZ
104
groundwater flow in z-direction
(a vertical darcy flow rate, e.g. in [mm/day]
SZ
SZ head elevation stored with SZ flows
Use this if you wan to display heads in SZ cross-sections in the Results
Viewer
SZ
109
groundwater extraction
SZ+Extraction
115
SZ exchange flow with river
SZ+River
112
SZ drainage flow from point
SZ+Drainage
SZ flow to general head boundary
SZ+GHB
105
SZ flow to MOUSE
78
MIKE SHE
Output Items
Table 3.1
Available output items for gridded data and time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion (Water Quality)
–PT - Particle Tracking
–SM - Snow melt
Code Output Item
Appears with these processes
External sources to SZ (for OpenMI)
216
Overland concentration
OC
217
Overland sorbed concentration
OC
218
Overland mass/area
OC
219
Air temperature
OC
220
UZ concentration (matrix phase)
UZ
221
UZ sorbed concentration (matrix phase)
UZ
222
UZ concentration (macropore phase)
UZ
223
UZ sorbed concentration (macropore phase)
UZ
224
UZ mass flux (matrix phase)
UZ
225
UZ mass flux (macropore phase)
UZ
226
UZ soil temperature
UZ
227
SZ concentration (mobile phase)
SZ
228
SZ sorbed concentration (mobile phase)
SZ
229
SZ concentration (immobile phase)
SZ
230
SZ sorbed concentration (immobile phase)
SZ
231
SZ soil temperature
SZ
232
SZ porosity
SZ
233
Number of particles
SZ
234
Number of registered particles
SZ
235
Most recent registration zone code
SZ
236
Average age
SZ
237
Average transport time to nearest registration cell
SZ
Results and Calibration
79
MIKE SHE Results
.
Table 3.2
Additional output items for time series.
–Key to symbols
–ET - Evapotranspiration
–OL - Finite Difference Overland Flow
–SubOL - Sub-catchment based Overland Flow
–UZ - Richards or Gravity Unsaturated flow,
–2LUZ - 2-Layer Unsaturated Water Balance
–SZ - Finite Difference Saturated Zone flow,
– LR - Linear Reservoir groundwater
–AD - Advection Dispersion
–PT - Particle Tracking
Code Output Item
Appears with these processes
145
SimStatus: Basic time step length
UZ
146
SimStatus: SZ time step length
SZ
147
SimStatus: No. of SZ iterations / time step
SZ
148
SimStatus: Avg. no. UZ iterations / column / time step
UZ
149
SimStatus: No. of Overland iterations per time step
OL
29
recharge to interflow reservoirs
LR
30
interflow from interflow reservoirs
LR
31
percolation from interflow reservoirs
LR
32
interflow reservoir storage
LR
33
change in interflow reservoir storage
LR
34
inflow to baseflow reservoir
LR
211
dead zone inflow to baseflow reservoir
LR
35
baseflow from baseflow reservoir
LR
36
groundwater feedback from baseflow reservoir
LR+UZ
44
pumping from baseflow reservoir
LR
46
storage in baseflow reservoir
LR
212
dead zone storage in baseflow reservoir
LR
38
change in subcatchment storage in baseflow reservoir
LR
213
change in dead zone storage in baseflow reservoir
LR
155
simple overland water depth
SubOL
156
simple overland exchange to lower zone or river
SubOL
157
simple overland recharge
SubOL
80
OL
LR+2LUZ
MIKE SHE
Toolbars
4
THE RESULTS VIEWER
4.1
Toolbars
Many of the functions in the Results Viewer are the same as those available in other DHI software tools (e.g., 2D Grid Editor). Additional tools
available in the result viewer are summarized in Table 4.1.
Table 4.1
Button
Description of Result Viewer tools
Name
Description
Rewind
Rewinds result files to first time step
Previous Step Rewinds result files to the previous time
step.
Video
Reverse
Generates an avi file from the current time
step to the first time step
Play Reverse Plays result files from the current time step
to the first time step. Identical to Video
Reverse except an avi file is not generated.
Stop Animation
Stops forward and reverse playing of result
files and creation of avi files
Play Forward Plays result files from the current time step
to the last time step. Identical to Video Forward except an avi file is not generated.
‘
Video Forward
Generates an avi file from the current time
step to the last time step
Next Step
Advances result files from the current time
step to the next time step
Wind
Advances results files to the last time step
Go to time
step
Rewinds or advances result files to the
specified time step
81
The Results viewer
Table 4.1
Button
Description of Result Viewer tools
Name
Description
Time step
Change the time step used by the result
viewer. The time step can be less than or
greater than the result file time step
Default
Default extraction tool.
Time Series
extractor
Tool to extract time series data from result
files. Multiple time series can be extracted
by holding down the Ctrl key while left-clicking. A single extraction or the last multiple
extraction is selected using a double leftclick. See Displaying a time series at a point
(V.1 p. 92)
Profile extrac- Tool to extract vertical profiles (cross-sector
tions) from 3D result files. Vertices of a profile line are specified with a single left click
and the profile line is closed with a double
left-click. See Saturated Zone Cross-section Plots (V.1 p. 95)
Cross-secTool to extract cross-sections of MIKE 11
tion extractor results at H-points. Additional information is
given below
UZ Plot
extractor
Tool to extract a UZ plot of the water content in the unsaturated zone. This tool generates a plot of water content versus depth
with time. This tool can only be used on one
cell at a time. A cell is selected by double
left-click. Additional information is given
below.
UZ Scatter
plot
Limits displays of results to unsaturated
zone calculation cells. This button is only
activated if unsaturated zone data is displayed in the result viewer. Additional information is given below.
UZ Filled Plot Displays interpolated unsaturated zone
results in non-calculation cells and unsaturated zone results in calculation cells. This
button is only activated if unsaturated zone
data is displayed in the result viewer. Additional information is given below.
82
MIKE SHE
Modifying the plot
4.2
Modifying the plot
When the Results viewer is opened from MIKE SHE, a default plot is created. However, in many cases, you will want to edit these plot settings.
Typical changes fall into four broad categories:
4.2.1
z
Adding additional result files and overlays (V.1 p. 83)
z
Adding or modifying vectors (V.1 p. 85)
z
Changing the shading and contour settings of gridded data (V.1 p. 87)
z
Changing the legend and colour scale (V.1 p. 88)
Adding additional result files and overlays
A project (or view or plot) in the Results Viewer is a collection of results
files and overlays. You can add additional results files or overlays to your
current plot by following these steps
1 Select Projects/Add Files to Project... from the top pull-down menu.
This will open the dialogue below
:
2 Click on the Add item button in this dialogue to add a line to the list of
files attached to the current project.
3 In the left hand column select the type of file to add, including image
files, additional results files, and MIKE 11 files.
83
The Results viewer
4 Click on the browse button, to find the file that you want to add. All
project files will be displayed that are the correct data type.
5 If you are adding shape files, you must remember to specify the coordinate axes or the file will not be displayed properly. To do this, you must
scroll the dialogue to the right and change the units in the Units combobox.
6 After adding the additional file or files, you can modify the drawing
order from the Overlay Manager tab
The up arrow and down arrow buttons are used to move an item up or
down in the drawing order. The Overlay Manager uses the convention
that items are drawn from the lowest to highest item number (i.e., items
on the bottom of the overlay list are drawn last and are on top of all
other items). The Overlay Manger can also be used to turn overlays on
and off by selecting or unselecting overlay items using the check box.
The Overlay Manager can also be accessed from the menu bar by
selecting Project / Active View Setting / Overlay Manager.
84
MIKE SHE
Modifying the plot
7 After adding the file, open the Property dialogue by right-clicking in
the results map and selecting Properties from the pop-up menu or by
using the top menu Projects / Active View Settings / Horizontal.
8 For an image file (above), you will need to specify the coordinates just
like in the Display items in the Setup Tab.
4.2.2
Adding or modifying vectors
Vectors can be added by adding a MIKE SHE results files that contains
flow data, which are project_overland.dfs2 and the
project_3DSZFlow.dfs3 files. To add vectors follow these steps:
85
The Results viewer
1 Add the a flow data file to your results view by following the directions
in the section Adding additional result files and overlays (V.1 p. 83).
2 After adding the flow file, open the Property dialogue by right-clicking
in the results map and selecting Properties from the pop-up menu or by
using the top menu Projects / Active View Settings / Horizontal.
3 When you added the flow results file, the grid data is by default displayed - hiding the original grid data in your results view. To turn off
this grid date find the grid item for the data file that you just added and
click off the Draw Grid checkbox
4 Then you need to find the Vector item for the flow file that you just
added and check on the Draw Vectors checkbox
5 From the comboboxes for X and Y Items, select the flow data for the x
and y directions.
6 Finally, select a Vector Scale. A suitable scale can only be found by
trial and error. Normally, a large number is good to start with. For
example, 10000. If the vectors are too large, then reduce the scale. If
they are too small, then increase the scale.
Note. Their is no vector data for the initial time step in the MIKE SHE
results files.
7 If your cell size is small or your flows are high you can plot a reduced
number of vectors by modifying the Draw every __ vector option.
86
MIKE SHE
Modifying the plot
4.2.3
Changing the shading and contour settings of gridded data
Gridded data is visually interpolated to a colour scale. The display of the
nodal values can be smoothed to make a more aesthetic plot. Finally, the
plot can customized to contain isolines with labels, a colour legend, etc.
The display properties of the gridded results can be modified by rightclicking in the graphical view and selecting Properties from the menu or
by selecting Projects/Active View Settings/Horizontal from the top menu.
These steps will open the dialogue below
The right hand side of the dialogue lists the available display items. Typically, this is a set of overlays from the MIKE SHE setup tab, plus the grid
file that is being displayed. If you have added other output files to the plot,
then these will also be listed.
Draw Grid - The Draw grid check box turns the gridded data display on
and off.
Contour Type - The interpolation of the gridded values to a colour scale
is controlled by the Contour Type. The available interpolation methods
include Box contours or Shaded contours. Box contours present a uniform
colour for every cell and Shaded contours smooth the colour gradation
between and across the cells. The transparency, copy colours and blend
colours options control how the display of the other overlays interacts
87
The Results viewer
with the gridded data. You should experiment with these settings to get a
feeling for how they interact.
Isolines - Isolines can be with or without labels. In the current version, the
format of the labels cannot be modified. However, the colour scale settings can be used to change the contour intervals.
Miscellaneous - Additional options are available to display the model
mesh and legend. The legend scale is controlled by the Colour tab.
Item/Layer - The Item/Layer section allows you to switch to a different
model layer for the gridded data. Note. You must remember to manually
change the layer number of any other displayed data, such as vectors, in
the other display items.
4.2.4
Changing the legend and colour scale
In some cases the default colour scheme may not be appropriate for the
intended purpose. The colour scheme and/or contour intervals can be
modified by right-clicking in the graphical view and selecting Properties
from the pop-up menu or using the Projects / Active View Settings / Horizontal keystrokes and navigating to the grid file entry that you want to
modify and the Colour tab for the grid entry (Figure 4.1). Options for
modifying the colour scheme and/or contour intervals include making a
New scheme/contour interval, Editing the existing scheme/contour interval, Opening an existing scheme/contour interval, Saving the current
scheme/contour interval, or Resetting (not implemented yet) the current
scheme/contour interval to default values. When making a new
scheme/contour interval it is possible to modify the Max and/or Min
value(s) used to generate the ranges used in the contour intervals
(Figure 4.1).
Figure 4.1
88
Colour modification property tab.
MIKE SHE
Modifying the plot
An example of making a new scheme/contour interval using the maximum
range is summarized in Figure 4.2 to Figure 4.5. Modification of the
number of contour intervals from the default value of 16 to 6 is shown in
Figure 4.2. Figure 4.3 shows the available colour schemes that are available and shows use of the Seismic colour scheme. The legend title (Palette
title) and palette type (Linear auto Scaled, Fixed, Land/Water Auto
Scaled, Land/Water Fixed, and Angle Fixed (Circular)) can also be modified on the first Palette Wizard window (Figure 4.2). Press Next after
making the desired changes to move to the next Palette Wizard window.
The colours used for each contour interval (Colour) and the ranges used
for each contour interval (Value) can be modified on the second Palette
Wizard window (Figure 4.4). Press Next after making the desired changes
to move to the next Palette Wizard window.
The third and final Palette Wizard window allows you to review the modified colour scheme and contour intervals before accepting the changes
(Figure 4.5). Press the Finish button if all of the modifications are acceptable. Otherwise, press the < Back button to make additional modifications
or Cancel button to cancel all changes.
Figure 4.2
Step 1 of 3 - modification of the number of colours used in the colour
scheme.
89
The Results viewer
90
Figure 4.3
Step 1 of 3 - modification of the number of colours model used in the
colour scheme.
Figure 4.4
Step 2 of 3 - modification of the colours used in the colour scheme
and the values colours are applied to.
MIKE SHE
Modifying the plot
Figure 4.5
Step 3 of 3 - acceptance of the colour scheme modification.
After Accepting the colour scheme/contour interval modifications the
Apply button should be pressed on the Result Data Properties window to
modify the look of the Result Viewer plot (Figure 4.6). The resulting modified Result Viewer plot is shown in Figure 4.7.
Figure 4.6
Applying the modified colour scheme to the current result viewer file.
91
The Results viewer
Figure 4.7
Result Viewer file after modification of the default colour scheme.
Users should experiment with the Palette Wizard to develop a better
understanding of available functionality than presented in this simple discussion.
4.3
Displaying a time series at a point
The Time Series tool allows you to plot a time series of all the data available in the current view.
Time series data can be selected from multiple locations in the active
model area using this tool. A single time series can be selected by doubleclicking in the desired location. Time series can be extracted from multiple locations by holding down the Ctrl-key and left clicking on each
desired location. When selecting multiple locations the Ctrl-key should be
held down while double clicking on the last location.
After selecting the locations of the time series files to extract you have the
option to deselect some of the selected points and to accumulate the data
over the simulation period (Figure 4.8). After making the appropriate
selections/deselections press the OK button to generate the time series
plot. The entire extraction process can be stopped by pressing the Cancel
button. An example of a time series plot generated in the Results Viewer is
shown in Figure 4.9.
92
MIKE SHE
Displaying a time series at a point
Figure 4.8
Selection of time series items to extract.
Figure 4.9
Time series plot generated using the time series extraction tool.
Addition graphical functions can be accessed by right-clicking in the
graphical view including zooming, exporting images, exporting time
series data as dfs0 files, and modification of the time series plot properties
(Figure 4.10). Most of the functionality can also be accessed via the menu
bar. For example, modification of the time series plot properties can be
accessed using Projects/Active View Settings/Timeseries.
93
The Results viewer
Figure 4.10
Modifying the properties of time series plots in the result viewer.
Because of the rich functionality available in the Result Viewer with
respect to time series output, users should experiment with the available
options. An example of the available functionality for modification of the
time series plot properties is shown in Figure 4.11. For example, as shown
in the upper left of Figure 4.11, time series items can be added or deleted
from a plot on the items tab.
94
MIKE SHE
Saturated Zone Cross-section Plots
Figure 4.11
4.4
Modification of A) items displayed on the time series plot, B) x-axis
properties, C) y-axis properties, and D) time series plot title.
Saturated Zone Cross-section Plots
To display a cross section plot of a set of 3D gridded data, you must click
on the Profile icon,
. Clicking on this icon will allow you to interactively define a cross-section by left-clicking at each vertex of the profile
line and double-clicking to close the profile.
After closing the profile, the following dialogue will be displayed listing
the available output items.
95
The Results viewer
.
Only one of these items can be selected. After selecting your item, click
OK and the profile will be displayed.
The profiles exactor tool can be used to extract a cross-section through
simulated MIKE SHE and MIKE 11 results. The type of cross-section created is dependent on the simulated data displayed in the result viewer. For
example, if the result viewer contains simulated 3D heads and MIKE 11
results then the cross-section will have simulated water levels and simulate MIKE 11 canal stages.
After defining the profile, the items to be displayed on the profile should
be selected. The resulting profile is shown in Figure 4.12. As with the
other tools, extracted profiles can be animated on the screen and/or
exported as avi and image files.
96
MIKE SHE
Saturated Zone Cross-section Plots
Figure 4.12
Resultant profile generated with the profile extractor tool.
You can modify the plot by right clicking on the plot and selecting Properties form the pop-up menu. In this dialogue, only the Graphical Items Tab
is relevant for MIKE SHE results (below).
In this dialogue, if you click on the Details... button for the item, you will
get the following dialogue, where you can change the colour scale and
plotting characteristics for the cross-section.
97
The Results viewer
As with the other tools available in the result viewer, users should experiment with the available options to learn how to fully use the result viewer
profile extractor.
4.4.1
Saving and loading profiles
If you have a profile open, under the View/Profile item in the top menu
bar, you can save the current profile location. This allows you to create
standard profiles for comparing scenarios.
To load a saved profile, make the plan view plot active, by either minimising or closing open profile plots. The View/Profile/Load option becomes
active and you can load a saved profile and select the profile item normally.
4.5
Displaying a MIKE 11 cross-section
MIKE 11 results can also be added to the result viewer and simulated
canal water levels can be displayed using the cross-section extractor. The
cross-section extractor shows simulated stages and the geometry of the
cross-section being viewed. The process of adding MIKE 11 results to the
result viewer are given in the section Adding additional result files and
overlays.
After selecting the cross-section extractor tool, move the cursor over the
location you want to extract the MIKE 11 results from (Figure 4.13). The
simulated results are displayed along with the cross-section geometry
98
MIKE SHE
Displaying a MIKE 11 cross-section
(Figure 4.14). As with the other tools, extracted profiles can be animated
on the screen and/or exported as .avi and image files.
Figure 4.13
Selection of a MIKE 11 cross-section location in the result viewer.
Figure 4.14
Resultant MIKE 11 cross-section plot.
Addition graphical functions can be accessed by right-clicking in the
graphical view (Figure 4.15). Modification of the profile properties is one
functionality available using the right-click. Since the cross-section plots
are relatively simple, modifications are limited to changing line and
marker properties, cross-section markers, etc. (Figure 4.16).
99
The Results viewer
Figure 4.15
Accessing addition functionality in the extracted MIKE 11 cross-section plot.
Figure 4.16
MIKE 11 cross-section properties that can be edited.
4.6
UZ Specific Plots
4.6.1
UZ Scatter and Filled Plots
For unsaturated zone results, scatter or filled plots can be generated. UZ
Scatter and Filled Plots are only different for simulations that do not use
the “calculation in all cells” UZ module option.
Scatter plots only show simulated results for UZ calculation cells. The
number of UZ calculation cells may be less than the total number of active
model domain used by the overland and saturated zone modules if the UZ
module for the simulation is not using the “calculation in all cells” option.
An example of when use of the UZ scatter plot is useful is shown in
Figure 4.17.
100
MIKE SHE
UZ Specific Plots
Figure 4.17
A) UZ Scatter Plot and its relationship to B) the UZ calculation cells.
An example of a UZ Filled Plot is shown in Figure 4.18. In cases where
the UZ module for the simulation is not using the “calculation in all cells”
option the Result Viewer interpolates values from the calculation cells to
adjacent inactive UZ cells.
Figure 4.18
4.6.2
Filled UZ plot.
UZ Plot
UZ Plots can only be extracted from simulated unsaturated zone water
contents and flow. This is because UZ plots display results for a single column for all of the UZ calculation nodes in the column. Other simulated
UZ results show net values for the entire UZ (i.e., infiltration, recharge to
the SZ, etc.).
After selecting the UZ Plot extractor tool move the cursor over the column
you want to extract the results from and double-click (Figure 4.19).
Results from multiple UZ columns cannot be displayed on the same UZ
Plot.
101
The Results viewer
Figure 4.19
Extracting a UZ Plot from simulated unsaturated water contents and
flow.
The simulated water content results for the selected column are displayed
in Figure 4.20. The UZ Plots show either water content or unsaturated
zone flow for each node in the column (y-axis) for the entire simulation
(x-axis).
Figure 4.20
Example UZ plot of unsaturated zone water content.
Addition graphical functions can be accessed by right-clicking in the
graphical view. Modification of the UZ Plot properties is one functionality
available using the right-click (Figure 4.21). Modifications that can be
made include changing the interpolation methods, adding the mesh, adding isolines, changing the colour schemes, etc. (Figure 4.22).
102
MIKE SHE
UZ Specific Plots
Figure 4.21
Modification of the UZ plot properties.
Figure 4.22
Available UZ plot properties that can be modified.
An example of a modified UZ plot with the mesh displayed and only
showing the upper five meters of the soil column is shown in Figure 4.22.
Additional information on modifying the interpolation and colour scheme
are given in the sections Changing the shading and contour settings of
gridded data and Changing the legend and colour scale.
Figure 4.23
Figure 25Close up of upper 5 meters of soil column with the calculation grid displayed.
103
The Results viewer
104
MIKE SHE
5
USING THE WATER BALANCE TOOL
The water balance utility is a post-processing tool for generating water
balance summaries from MIKE SHE simulations. Water balance output
can include area normalized flows (storage depths), storage changes, and
model errors for individual model components (e.g., unsaturated zone,
evapotranspiration, etc.).
A water balance can be generated at a variety of spatial and temporal
scales and in a number of different formats, including dfs0 time series
files, dfs2 grid series files, and ASCII text output suitable for importing to
Microsoft Excel. You can also automatically create a picture that visualizes the interrelationships between the various water balance components
(see Figure 5.1).
The water balance utility can be run from within the MIKE Zero interface
or from a MSDOS batch file. The batch functionality allows you to calculate water balances automatically after a MIKE SHE simulation that is
also run in batch mode. Alternatively, you can also calculate water balances as part of an AUTOCAL simulation and use the results as part of an
objective function
Figure 5.1
Graphical water balance output example
105
Using the Water Balance Tool
5.1
Creating a water balance
Before you can create a water balance for a MIKE SHE WM simulation,
you must have saved the water balance data during the simulation. Saving
of the water balance data is specified in the Storing of Results (V.2 p. 183)
dialogue. If you have forgotten to save the water balance data, then you
will need to re-run your simulation.
After you have run your WM simulation, creating and running a water balance in MIKE SHE is quite simple, following these steps
1 Create a new water balance document (V.1 p. 106),
2 Extract the water balance data (V.1 p. 107)
3 Specify your water balance (V.1 p. 109), and
4 Calculate and View the Water Balance (V.1 p. 113).
5.1.1
106
Create a new water balance document
The new water balance document is created by selecting the File/New
item in the top menu, or clicking on the New icon in the top menu bar. In
the dialogue that appears, select MIKE SHE and Water Balance Calculations in the right hand box, as shown below.
MIKE SHE
Creating a water balance
5.1.2
Extract the water balance data
To extract the water balance data, specify the MIKE SHE simulation by
selecting the simulation catalogue file (.sheres file), then specify the area
of your model that you want the water balance for, and, finally, extract the
MIKE SHE water balance data from the results files.
Once you have created a new water balance document, the first tab is as
shown below.
107
Using the Water Balance Tool
Flow result catalogue file
A MIKE SHE simulation generates various output files depending on the
options and engines selected for the MIKE SHE simulation. The .sheres
file is a catalogue of all the various output files generated by the current
MIKE SHE run. When you select the .sheres file, you are not specifying
the particular output, but actually just a set of pointers to all the output
files.
The extraction process reads all of the output files and makes itself ready
to produce specific water balances. In the extraction dialogue, you specify
the .sheres file for the simulation that you wish to calculate the water balance for. The .sheres file is located in the same directory as your results.
Note Although, this is an ASCII file, you should be careful not to make
any changes in the file, or you may have to re-run your simulation.
Type of Extraction
You can choose to calculate the water balance on the entire model domain
or in just a part of the domain. By default the calculation is for the entire
domain, or catchment. If you choose the subcatchment area type, they you
will be able to use a dfs2 integer grid code file to define the areas that you
want individual water balances for.
108
MIKE SHE
Creating a water balance
If you use an area resolution, then the water balance will be a summary
water balance for either the entire catchment or the sub-areas that you
define.
If you use a single-cell resolution, you will be able to generate dfs2 maps
of the water balance.
Sub-catchment grid codes
The subcatchment integer grid code file is only used if you have selected
the sub-catchment water balance type. You can specify a delete value to
exclude areas from the water balance. The grid spacing and dimensions in
this dfs2 file do not have to match the model grid exactly. However, the
sub-catchment grid must be both coarser than and aligned with the original grid.
You can also specify a polygon shape file to define the sub-catchment
areas. The shape file may contain multiple polygon, with the same or different codes. Further, the shape file length units do not have to be the same
as the model length units (e.g. feet vs. meters).
Gross files
The pre-processor extracts the water balance data from the standard MIKE
SHE output files and saves the data in a set of “gross” files. The file names
of the gross files is built up from the project name and prefix specified
here. The default value is normally fine.
Run the extraction
To run the extraction, you simply have to click on the Run Extraction icon,
, or the Run/Extraction top menu item.
5.1.3
Specify your water balance
After you have extracted the water balance data from the MIKE SHE
results files, then you can switch to the post-processing tab. Here you can
create any number of individual water balances by simply clicking on the
Add item icon and specifying the water balance parameters in the parameter dialogue.
109
Using the Water Balance Tool
A single Postprocessing item is created by default when the water balance
file is created. The default Postprocessing name can be change to a more
appropriate name. Postprocessing items that are no longer needed can be
deleted using the Delete button.
Use default Config file
Unchecking the Use default Config file checkbox, allows you to specify
the location of a custom water balance Config file. Development of custom water balance configuration files is described in detail in Making
Custom Water Balances (V.1 p. 144).
For each item in the Postprocessing list above, a new item will be added to
the data tree. If you expand the data tree, each will have the following dialogue.
110
MIKE SHE
Creating a water balance
Water Balance
Multiple postprocessings can be run on each water balance extraction.
More detail on the types of available water balances data are discussed in
the Available Water Balance Items (V.1 p. 114) section. In brief, the available types include
z
The total water balance of the entire model catchment or sub-catchments in an ASCII table, a dfs0 file, a dfs2 map file, or a graphical
chart (also by layer),
z
Model errors for each hydrologic component (overland, unsaturated
zone, etc.) in an ASCII table, a dfs0 file, or a dfs2 map file (also by
layer),
z
The snow melt and canopy/interception water balance in an ASCII
table, or a dfs0 file,
z
An abbreviated or detailed water balance for overland or unsaturated
flow in an ASCII table, or a dfs0 file, and
z
An abbreviated or detailed water balance by layer for saturated flow in
an ASCII table, or a dfs0 file.
111
Using the Water Balance Tool
Output Period
An output period different from the total simulation period can be specified by unchecking Use default period and setting the Start date and
End date to the period of interest
Output Time Series Specification
Incremental or Accumulated water balances can be calculated. An incremental water balance is calculated (summed) for each output time step in
the Output period. An accumulated water balance each output time step is
accumulated over the Output period
Layer Output Specifications
If you are using water balance types that calculate data on a layer basis,
you can specify whether you want All layers or just the Specified layer,
where you also must specify a layer number.
Sub-catchment Selection
If you extracted sub-catchment data from the WM results, then you must
specify a subcatchment number or the name of the polygon for which you
want the water balance for. The combobox contains a list of valid ID numbers or polygon names.
Single Cell Location
If you extracted the WM data by cell and you are not creating a map output, you have to specify a cell location for which you want a water balance.
Output File
If you are creating a table or time series water balance, then you can write
the output to either a dfs0 file or to a tab-delimited ASCII file for import to
MSExcel, or other post-processing tool. If you are creating a map, then the
output will be to a dfs2 file, with the same grid dimensions and spacing as
the model grid. If you are creating a chart, then the output will be written
to an ASCII file, with a special format for creating the chart graphic.
Run the Post Processing
To run the post processing, you have two choices. You can click on the
Run Selected Post-Processing icon,
, which runs only the current postprocessing item. Or, you can click on the Run All Post-Processing icon,
, which runs all of the post-processing items in the list. These two
options are also available in the Run top menu.
112
MIKE SHE
Calculating Water Balances in Batch Mode
5.1.4
Calculate and View the Water Balance
The data tree for the results tab lists all of the calculated water balances.
The dialogue for each item, includes the file name and an Open button that
will open an editor for the file. For ASCII output, this will be your default
ASCII editor - usually Notepad. For dfs0 and dfs2 files, the DHI Time
Series Editor or Grid Editor will be opened. For the chart output, the
graphic will be displayed by the program WblChart.
Units for the water balance
The values in the water balance are in the EUM unit type Storage Depth.
This normalization allows water balances for different models or model
areas to be more easily compared. The Storage Depth values can be converted to volume by multiplying by the internal model area. The number
of internal model cells can be found in the _WM_PRINT.LOG file. Thus,
the internal area is the number of cells times the area per cell. If you have
calcuated a water balance on a sub-area, the volumes must be calculated
based on the number of internal cells in the sub-area.
The default units are [mm], but this can be changed to any length unit (e.g.
inches) by changing the EUM unit of the variable Storage Depth.
5.2
Calculating Water Balances in Batch Mode
Like most DHI software, the water balance utility can be run in batch
mode. Some possible ways to run the water balance utility in batch mode
are:
z
Running the water balance utility immediately after completion of a
MIKE SHE simulation run in batch mode.
z
Running the water balance utility for a MIKE SHE simulation without
using the water balance utility graphical users interface.
z
Running multiple water balance Postprocessing stages automatically
without using the water balance utility graphical users interface.
The water balance graphical utility stores all of its information in a .wbl
file. The .wbl file is an ASCII file that can be edited with Notepad or other
text editor, but the format of the water balance file must be preserved. For
more information on editing the .wbl file and creating custom water balances, see Making Custom Water Balances (V.1 p. 144).
To run the water balance utility in batch mode, the .wbl file must be created prior to executing it and all file names in the .wbl file need to be
valid. If during calibration the same MIKE SHE file name is used for each
113
Using the Water Balance Tool
simulation then the same water balance file can be used for all calibration
runs. If the MIKE SHE simulation to be evaluated is different from the
MIKE SHE simulation used to set up the water balance file, you will have
to edit the water balance file.
To run the Extraction and Postprocessing steps in batch mode, the your
PATH statement needs to include directory where MIKE SHE was
installed. The default directory is
C:\Program Files\DHI\MIKEZero\bin
An example is shown below of a batch file that generates water balance
data for three postprocessing steps, using a water balance utility file
named WaterConservationAreas.WBL.
rem ------------------------------------------MSHE_Wbl_Ex.exe WaterConservationAreas.WBL
MSHE_Wbl_Post.exe WaterConservationAreas.WBL 1
MSHE_Wbl_Post.exe WaterConservationAreas.WBL 2
MSHE_Wbl_Post.exe WaterConservationAreas.WBL 3
The MSHE_Wbl_Ex.exe command runs the Extraction phase of the
water balance utility. The command
MSHE_Wbl_Post.exe WaterConservationAreas.WBL 1
runs the first Postprocessing item in the water balance file, WaterConservationAreas.WBL. The number after the water balance file
name in the Postprocessing command indicates which Postprocessing
item to run and this number must consistent with the water balance utility
file (i.e., the number cannot be greater than the number of Postprocessing
items in the file). Otherwise, the program will terminate with an error. The
Postprocessing step cannot be executed before an Extraction step but only
one Extraction step needs to be run for a single water balance utility file.
The water balance batch file can contain Extraction and Postprocessing
steps from multiple water balance utility files.
5.3
Available Water Balance Items
The .shres file contains a list of all of the simulation output files generated
during the WM or WQ simulation. When you use the water balance
extraction utility, all of these files are processed and a special set of water
114
MIKE SHE
Available Water Balance Items
balance files are created - the .wblgross files. One file is created for each
of the water balance components:
z
Snowmelt and precipitation - projectname_sm.wblgross
z
Canopy interception - projectname_ci.wblgross
z
Ponded surface water - projectname_ol.wblgross
z
Unsaturated zone - projectname_uz.wblgross
z
Saturated zone - projectname_sz.wblgross
The contents of each of these files can be output using the “Detailed”
water balances. All of the items in these files are listed and described in
the following tables:
z
Table 5.1 SM - Precipitation and snowmelt items (p. 117)
z
Table 5.2 CI - Canopy interception water balance items (p. 119)
z
Table 5.3 OL - Overland flow items (p. 122)
z
Table 5.4 UZ - Unsaturated Zone items (p. 126)
z
Table 5.5 SZ - Saturated Zone - all layers (p. 130)
z
Table 5.6 SZ - Saturated Zone - specified by layers (p. 136)
z
Table 5.7 SZ - Saturated Zone - Linear Reservoir all layers (p. 137)
The water balance utility is a very flexible tool that allows you to modify
existing Water balance types or create custom Water balance types to suit
your needs. The water balance calculations use a water balance Config(uration) file to define Water balance types using the available water
balance items and a macro language to control program execution.
To modify existing or custom Water balance types you must understand
the available items and what data they contain.
Sign Conventions
MIKE SHE uses a sign convention that is positive in the positive coordinate direction. In other words, water flowing upward in the model is a positive flow in MIKE SHE. Likewise, flow in the direction of increasing x or
y is also positive. Boundary flows and other flows that do not have a
direction are positive outwards.
However, the water balance utility uses a control volume sign convention,
such that all inflows are negative and all outflows are positive. This can
cause confusion when calculating a water balance. For example, a vertical
115
Using the Water Balance Tool
downward flow through the unsaturated zone will always be a negative
result in MIKE SHE. In the water balance control volume, a downward
flow into the unsaturated zone will be a positive outflow in the water balance for ponded water, but a negative inflow into the unsaturated zone
water balance.
The sign convention for the water balance error of each storage is such
that an increasing storage is positive. Thus, a positive water balance error
means that the change in storage plus the total outflows is greater than the
total inflows. In other words, the error is positive if your model is creating
water.
5.3.1
Snow Storage
The snow storage items include all of the water balance items related to
rainfall and the conversion to and from snow.
The items listed in Table 5.1 are those found in the “Snow Melt component
- detailed” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'SM_DETAIL'
DisplayName = 'Snow Melt component - detailed'
Description = 'Detailed Snow Melt component water balance'
NoGroups = 11
Group = 'Precip and Irr -> Snow(sm.qprecandirrtosnow)'
Group = 'AirTemp Freezing(sm.qfreezing)'
Group = 'AirTemp Melting(sm.qthawing)'
Group = 'Radiation Melting(sm.qradmelting)'
Group = 'Rain Melting(sm.qrainmelting)'
Group = 'Snow -> OL(sm.qsnowtool)'
Group = 'Snow Evap(sm.qesnow)'
Group = 'Dry Snow Stor.Change(sm.dsnowsto-sm.dwetsnowsto)'
Group = 'Wet Snow Stor.Change(sm.dwetsnowsto)'
Group = 'Total Snow Stor.Change(sm.dsnowsto)'
Group = 'Error(sm.smwblerr)'
EndSect // WblTypeDefinition
The sign convention is such that precipitation is negative (inflow) and
melting is positive (outflow). All of the noted tems together should add to
zero. The freezing and thawing items are not included in the error term
because they are internal transfers of water between dry snow and wet
snow storages.
116
MIKE SHE
Available Water Balance Items
The snow storage items are found in the projectname_sm.wblgross file.
This file also contains the terms sm.qP, sm.qPad, and sm.PIrrSprinkler,
which are not included in the detailed water balance output because they
are included in the term sm.qPrecAndIrrToSnow.
Table 5.1
SM - Precipitation and snowmelt items
Item
Description
Sign Convention in the
Water balance
sm.qPrecAndIrrToSnow
Precipitation plus irrigation added Inflow - negative
to snow storage when the air temperature is below the freezing
temperature
yes
sm.qFreezing
Amount of wet snow converted to Negative
dry snow due to freezing
no
sm.qThawing
Amount of water removed from Positive when melting
dry snow storage due to tempera- occurs
ture melting
no
sm.qRadMelting
Amount of water removed from Positive when melting
dry snow storage due to radiation occurs
melting
no
sm.qRainMelting
Amount of water removed from
dry snow storage due to melting
from rain
no
sm.qSnowToOL
Outflow - positive when
Amount of wet snow storage
transfered to interception storage. water is added to canopy
Actually, this amount is added to interception
qPad, which is the input to canopy
interception. Then the water is
added to ponded water via interception throughfall.
Note: Freezing of ponded water to
snow storage is not accounted for
in MIKE SHE
Positive when melting
occurs
Included
in Wbl
Error
yes
117
Using the Water Balance Tool
Table 5.1
SM - Precipitation and snowmelt items
Item
Description
sm.qESnow
Outflow - positive when
Amount of evaporation from
evaporation/sublimation
snow. This is a combination of
occurs
sublimation from dry snow and
evaporation from wet snow.
Evaporation is removed first from
wet snow storage. When wet snow
storage is zero, then sublimation
from dry snow is removed
because of the higher energy
required for sublimation.
yes
sm.dWetSnowSto
Change in wet snow storage
no
sm.dSnowSto
Change in total snow storage
Positive when total snow
Note: Change in dry snow storage storage increases
is (dSnowSto - dWetSnowSto)
sm.smWblErr
Snow storage water balance error. Positive if water generated
Sum of marked items.
(∆storage + Outflow >
Inflow)
sm.qP
Total precipitation
Inflow - negative
(not used in detailed SM WB output)
no
sm.qIrrSpinkler
Total Irrigation
Inflow - negative
(not used in detailed SM WB output)
no
sm.qPad
Outflow - positive
Total precipitation reaching the
canopy (Precipitation + sprinkler
irrigation + snowmelt to ponded
water). Same as ci.qPad.
(not used in detailed SM WB output)
no
118
Sign Convention in the
Water balance
Positive when wet snow
storage increases
Included
in Wbl
Error
yes
MIKE SHE
Available Water Balance Items
5.3.2
Canopy interception storage
The canopy interception is a seperate storage on the leaves of the vegetation. If the LAI is zero, then the canopy interception will be zero, as will
all of the items in this storage.
The items listed in Table 5.2 are those found in the “Canopy Interception
component” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'CI'
DisplayName = 'Canopy Interception component'
Description = 'Canopy Interception component waterbalance items'
NoGroups = 5
Group = 'Precip(ci.qpad)'
Group = 'Can. ThroughFall(-ci.qpnet)'
Group = 'Evaporation(ci.qeint)'
Group = 'Can.Stor.Change(ci.dintsto)'
Group = 'Error(ci.ciwblerr)'
EndSect // WblTypeDefinition
The sign convention in the water balance is such that precipitation is negative (inflow) and evaporation is positive (outflow). All of the items
together should add to zero.
Note, however, the negative sign in front of the ci.qpnet term in the water
balance definition above. This is because the canopy throughfall is a vertical downward flow in MIKE SHE - making it a negative value in the
MIKE SHE results files. Whereas, it must be a positive outflow in the
water balance calculation.
Table 5.2
CI - Canopy interception water balance items
Item
Description
Sign Convention in the
Water balance
ci.qPad
Total precipitation reaching the
Inflow - negative
canopy (Precipitation + sprinkler
irrigation + snowmelt to OL - precipitation converted to snow)
-ci.qPnet
Canopy throughfall to ponded
water
Included
in Wbl
Error
yes
Outflow - positive
yes
(Note sign change in water
balance definition)
119
Using the Water Balance Tool
Table 5.2
CI - Canopy interception water balance items
Item
Description
ci.qEInt
Evaporation from intercepted stor- Outflow - positive
age
ci.dIntSto
Change in interception storage
Positive when interception yes
storage increases
ci.ciWblErr
Interception storage water balance error
Positive if water generated
(∆storage + Outflow >
Inflow)
5.3.3
Sign Convention in the
Water balance
Included
in Wbl
Error
yes
Ponded water storage
Water on the ground surface belongs to the ponded water storage. Rainfall
is added to ponded storage. Ponded storage evaporates, infiltrates or flows
to MIKE 11.
The items listed in Table 5.3 are those found in the “Overland flow detailed” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'OL_DETAIL'
DisplayName = 'Overland flow - detailed'
Description = 'Detailed Overland component water balance'
NoGroups = 23
Group = 'qpnet(ol.qpnet)'
Group = 'qirrdrip(ol.qirrdrip)'
Group = 'qeol(ol.qeol)'
Group = 'qh(ol.qh+ol.qhmp)'
Group = 'qolszpos(-ol.qolszpos)'
Group = 'qolszneg(-ol.qolszneg)'
Group = 'qsztofloodpos(-ol.qsztofloodpos)'
Group = 'qsztofloodneg(-ol.qsztofloodneg)'
Group = 'qolin(ol.qolin)'
Group = 'qolout(ol.qolout)'
Group = 'qolrivpos(ol.qolrivpos)'
Group = 'qolrivneg(ol.qolrivneg)'
Group = 'qocdr(ol.qocdr)'
Group = 'qocdrtoM11HPoint(ol.qocdrtoM11HPoint)'
Group = 'qfloodtorivin(ol.qfloodtorivin)'
Group = 'qfloodtorivex(ol.qfloodtorivex)'
Group = 'qoldrtoMouse(ol.qoldrtoMouse)'
120
MIKE SHE
Available Water Balance Items
Group = 'qolMousepos(ol.qolMousepos)'
Group = 'qolMouseneg(ol.qolMouseneg)'
Group = 'qOlExtSink(ol.qOlExtSink)'
Group = 'qOlExtSource(ol.qOlExtSource)'
Group = 'dolsto(ol.dolsto)'
Group = 'olwblerr(ol.olwblerr)'
EndSect // WblTypeDefinition[WblTypeDefinition]
The sign convention for a ponded water control volume is such that precipitation is negative (inflow), and boundary outflow, infiltration and
evaporation are all positive (outflow). All of the Wbl Error items together
should add to zero.
Note, however, the negative sign in front of some of the terms in the water
balance definition above. This is because the SZ exchange to ponded storage is an upwards positive flow in MIKE SHE - making it a positive value
in the MIKE SHE results files when flowing to ponded water and a negative value when infiltrating to SZ. Whereas, these flows must be the opposite sign in the water balance calculation.
Special considerations for water balances in Flood Code cells
Water on the ground surface belongs to ponded storage - except in active
flood code cells. Active flood code cells are those where the cell is flooded
and the water level is controlled by the water level in MIKE 11.
There are four terms in the water balance related to flood codes: qSZToFloodPos/Neg, and qFloodToRivIn/Ex.
When the groundwater table is at or above the land surface, water can
exchange directly between ponded water and the saturated zone. The
unsaturated zone does not exist. If the land surface is an active flood code
cell, then then the water is added to or removed from the storage available
for exchange with MIKE 11 and the two terms qSZFloodPos and qSZFloodNeg may be non-zero.
The exchange between ponded water and MIKE 11 in active flood code
cells is calculated based on the change of storage due to the various
source/sink terms over the MIKE SHE overland time step. This includes
overland flow between flooded and non-flooded cells, rainfall, evaporation, infiltration to UZ, direct flow between SZ and flooded cells when the
groundwater table is above ground. Thus, in a flood code cell
121
Using the Water Balance Tool
1 At the beginning of the overland time step, the ponded water level is
set equal to the corresponding water level in MIKE 11 (if this is above
the MIKE SHE ground level) and the status of the cell is set to active.
2 At the end of the overland flow time step, MIKE SHE calculates the
change in ponded water level and adds or subtracts this as lateral
inflow to MIKE 11 over the next MIKE 11 time step(s), covering the
period of the MIKE SHE Overland time step.
Thus, qsztoflood is not directly added as lateral inflow to MIKE 11. But
it’s one of the source/sink terms that contribute to the change in storage in
flooded cells – which is then added to MIKE 11 as qfloodtoriv
The two terms, qFloodToRivIn and qFloodToRivEx, together are the net
lateral inflow to MIKE 11 from active flood code cells. In other words,
summed together, they are the actual exchange between flood code areas
and MIKE 11.
Table 5.3
OL - Overland flow items
Item
Description
ol.qPnet
Inflow - negative
Canopy throughfall to ponded
water.
This is the same value as ci.qPnet,
but with the opposite sign.
yes
ol.qIrrDrip
Irrigation added to ponded water. Inflow - negative
This includes both drip irrigation
and sheet irrigation, since both are
added directly to ponded storage.
yes
ol.qEOL
Direct evaporation from ponded
water
Outflow - positive
yes
ol.qH
Infiltration from ponded water
into the UZ
Outflow - positive
yes
ol.qHmp
Infiltration from ponded water
into the UZ macropores
Outflow - positive
yes
-ol.qOLSZpos
Direct flow up from SZ to OL.
Inflow - negative
yes
This is a positive upwards flow in (Note sign change in water
the MIKE SHE results files.
balance definition)
122
Sign Convention in the
Water balance
Included
in Wbl
Error
MIKE SHE
Available Water Balance Items
Table 5.3
OL - Overland flow items
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
-ol.qOLSZneg
yes
Direct flow down from OL to SZ. Outflow - positive
(Note sign change in water
This is a negative downwards
balance definition)
flow in the MIKE SHE results
files.
-ol.qSZToFloodPos
yes
Direct flow upwards from SZ to Inflow - negative
an active flood code cell
(Note sign change in water
balance definition)
(active means that it is actually
flooded and the water level is controlled by the water level in MIKE
11).
This is a positive upwards flow in
the MIKE SHE results files.
Only non-zero when the groundwater table is at or above the
ground surface.
yes
-ol.qSZToFloodNeg Direct flow downwards from an Outflow - positive
an active flood code cell to SZ.
(Note sign change in water
balance definition)
(active means that it is actually
flooded and the water level is controlled by the water level in MIKE
11).
This is a negative downwards
flow in the MIKE SHE results
files.
Only non-zero when the groundwater table is at or above the
ground surface.
ol.qOLin
Inflow to overland storage across Inflow - negative
the boundary of the model, or
inflow across the boundary of the
water balance sub-area
yes
ol.qOLout
Outflow from overland storage
Outflow - positive
across the boundary of the model,
or outflow across the boundary of
the water balance sub-area
yes
123
Using the Water Balance Tool
Table 5.3
OL - Overland flow items
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
ol.qOLRivPos
Overland outflow to MIKE 11
Outflow - positive
yes
olqOLRivNeg
Inflow from MIKE 11 to overland Inflow - negative
storage
yes
ol.qOCDr
Overland flow in paved areas that Outflow - positive
is added to the SZ drainage network and thus directly to MIKE
11
yes
ol.qOCDrToM11HPo Overland flow in paved areas that Outflow - positive
int
is added to the SZ drainage network and then directly to a specified MIKE 11 h-point
yes
ol.qOLDrToMouse
Overland flow in paved areas that Outflow - positive
is added to the SZ drainage network and then directly to a specified Mouse/MIKE Urban manhole
yes
ol.qFloodToRivIn
Net lateral inflow exchange
Inflow (negative) or Outbetween active flood code cells
flow (positive)
and MIKE 11 nodes that are inside
the current water balance area
yes
ol.qFloodToRivEx
Inflow (negative) or OutNet lateral inflow exchange
flow (positive)
between active flood code cells
and MIKE 11 nodes that are outside the current water balance area
This is always zero unless the
water balance is being calculated
on a sub-area.
yes
ol.qOLMousePos
Outflow from overland storage to Outflow - positive
Mouse/MIKE Urban
yes
ol.qOLMouseNeg
Inflow from Mouse/MIKE Urban Inflow - negative
to overland storage
yes
ol.qOLExtSink
Outflow to OpenMI sink
Outflow - positive
yes
ol.qOLExtSource
Inflow from OpenMI sink
Inflow - negative
yes
124
MIKE SHE
Available Water Balance Items
Table 5.3
OL - Overland flow items
Item
Description
Sign Convention in the
Water balance
ol.dOLSto
Change in overland storage
Positive if storage increases yes
ol.OLWblErr
OL water balance error
Positive if water generated
(∆storage + Outflow >
Inflow)
5.3.4
Included
in Wbl
Error
Unsaturated Zone Storage
Unsaturated zone storage includes all the water between the ground surface and the water table. Thus, all water stored in the root zone is also
included here.
The items listed in Table 5.4 are those found in the “Unsaturated Zone detailed” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'UZ_DETAIL'
DisplayName = 'Unsaturated Zone - detailed'
Description = 'Detailed Unsaturated zone component water balance'
NoGroups = 10
Group = 'qh(uz.qh)'
Group = 'qhmp(uz.qhmp)'
Group = 'qeuz(uz.qeuz)'
Group = 'qtuz(uz.qtuz)'
Group = 'qrech(-uz.qrech)'
Group = 'qrechmp(-uz.qrechmp)'
Group = 'qgwfeedbackuz(-uz.qgwfeedbackuz)'
Group = 'duzdef(-uz.duzdef)'
Group = 'uzszstocorr(uz.uzszstocorr)'
Group = 'uzwblerr(uz.uzwblerr)'
EndSect // WblTypeDefinition
The sign convention in the UZ water balance is such that infiltration from
the surface is negative (inflow) and recharge to SZ is positive (outflow).
All of the items together should add to zero.
Note, however, the negative sign in front of some of the terms (e.g.
uz.qRech) in the water balance definition above. This is because the
recharge to SZ is a vertical downward flow in MIKE SHE - making it a
125
Using the Water Balance Tool
negative value in the MIKE SHE results files. The negative sign in the
water balance conforms the sign to the water balance sign convention of
positive outflows.
Table 5.4
UZ - Unsaturated Zone items
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
uz.qH
Infiltration from ponded water
into the UZ matrix
Inflow - negative
yes
uz.qHmp
Infiltration from ponded water
into the UZ macropores
Inflow - negative
yes
uz.qEuz
Direct evaporation from the top
Outflow - positive
UZ node when using the Richards
or Gravity flow finite-difference
method
yes
uz.qTuz
Transpiration from the root zone
yes
-uz.qRech
yes
Recharge out of the bottom of the Outflow - positive
soil column to SZ via the UZ soil (Note sign change in water
matrix.
balance definition)
In the MIKE SHE results,
recharge is a vertical downward
flow (in the negative direction).
In the UZ water balance it is an
outflow and must be a positive
value.
-uz.qRechMp
yes
Recharge out of the bottom of the Outflow - positive
soil column to SZ via the UZ
(Note sign change in water
macropores.
balance definition)
In the MIKE SHE results,
recharge is a vertical downward
flow (in the negative direction).
In the UZ water balance it is an
outflow and must be a positive
value.
126
Outflow - positive
MIKE SHE
Available Water Balance Items
Table 5.4
UZ - Unsaturated Zone items
Item
Description
Sign Convention in the
Water balance
-uz.qGWFeedBackUZ
yes
Inflow - negative
Feedback from LR to UZ
This value is only non-zero if the (Note sign change in water
balance definition)
Linear Reservoir groundwater
option is used. In this case, the
baseflow reservoirs will add water
to the UZ as a fraction of the discharge to MIKE 11.
In the MIKE SHE results, the
feedback to UZ is a positive value.
But, in the water balance it is an
inflow and must have a negative
sign.
-uz.dUzDef
Change in UZ deficit.
The UZ deficit is essentially the
amount of air in the profile. It is
the opposite of the UZ storage.
A decreasing deficit means that
the soil is getting wetter, which
equals increasing UZ storage.
An increasing deficit means that
the soil is getting drier, which
equals decreasing UZ storage.
Internally in MSHE, the value of
dUzDef is calculated as a change
in storage.
The negative sign is added to convert the change in storage to a
change in deficit.
Negative for increasing UZ
deficit
(Note sign change in water
balance definition)
Included
in Wbl
Error
yes,
but in
the error
term
calculation the
negative sign
is not
used
127
Using the Water Balance Tool
Table 5.4
UZ - Unsaturated Zone items
Item
Description
Sign Convention in the
Water balance
uz.UzSzStorCorr
Water balance correction to
account for changing thickness of
the UZ zone as the groundwater
table rises and falls.
yes
Positive for a falling
groundwater table, because
the amount of UZ storage is
increasing.
Negative for a rising
groundwater table, because
the amount of UZ storage is
decreasing.
uz.uzWblErr
UZ Water balance error
Positive if water generated
(∆storage + Outflow >
Inflow)
5.3.5
Included
in Wbl
Error
Saturated Zone Storage
The saturated zone storage includes all water below the water table. All
groundwater pumping is from the saturated zone, including irrigation
extraction from groundwater.
The items listed in Table 5.5 are those found in the “Saturated Zone detailed” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'SZ_DETAIL'
DisplayName = 'Saturated Zone - detailed'
Description = 'Detailed Saturated ... balance (depth-integrated)'
NoGroups = 28
Group = 'qszprecip(sz.qszprecip)'
Group = 'qrech(uz.qrech)'
Group = 'qrechmp(uz.qrechmp)'
Group = 'qolszpos(sz.qolszpos)'
Group = 'qolszneg(sz.qolszneg)'
Group = 'qetsz(sz.qetsz)'
Group = 'qszin(sz.qszin)'
Group = 'qszout(sz.qszout)'
Group = 'dszsto(sz.dszsto)'
Group = 'qszabsex(sz.qszabsex)'
Group = 'qszdrin(sz.qszdrin)'
Group = 'qszdrout(sz.qszdrout)'
Group = 'qszdrtorivin(sz.qszdrtorivin)'
128
MIKE SHE
Available Water Balance Items
Group = 'qszdrtorivex(sz.qszdrtorivex)'
Group = 'qszdrtoM11HPoint(sz.qszdrtoM11HPoint)'
Group = 'qszrivneg(sz.qszrivneg)'
Group = 'qszrivpos(sz.qszrivpos)'
Group = 'qszfloodneg(ol.qsztofloodneg)'
Group = 'qszfloodpos(ol.qsztofloodpos)'
Group = 'qgihbpos(sz.qgihbpos)'
Group = 'qgihbneg(sz.qgihbneg)'
Group = 'qirrwell(sz.qirrwell)'
Group = 'qszdrtoMouse(sz.qszdrtoMouse)'
Group = 'qszMousepos(sz.qszMousepos)'
Group = 'qszMouseneg(sz.qszMouseneg)'
Group = 'qSzExtSink(sz.qSzExtSink)'
Group = 'qSzExtSource(sz.qSzExtSource)'
Group = 'Error(sz.szwblerrtot)'
EndSect // WblTypeDefinition
The sign convention in the SZ water balance is such that infiltration from
the unsaturated zone is negative (inflow) and discharge to overland flow is
positive (outflow). All of the items together should add to zero.
The use of negative signs in the SZ water balance is avoided by explicitly
including both inflow (negative) and outflow (positive) terms. For example, sz.qOlSzPos is the flow from the saturated zone directly to ponded
water when the groundwater table is at or above the ground surface. In the
MIKE SHE results, this is a positive upwards flow, and in the water balance it is a positive outflow. Similarly, sz.qOlSzNeg is the downward flow
129
Using the Water Balance Tool
from ponded water directly to the saturated zone, which is a negative
downward flow and a negative water balance inflow to SZ.
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
Sign Convention in the
Water balance
sz.qSzPrecip
Precipitation added directly to the
SZ layer.
This can only be non-zero when
the simulation does not include
UZ. If UZ is included, but the
groundwater table is at the ground
surface (no UZ cells), the precipitation to SZ is included in the term
sz.qOlSzNeg.
Can be an outflow if the negative
precipitation option specified in
the Extra Parameters (Negative
Precipitation (V.1 p. 300)).
In this case, negative precipitation
can be removed from multiple SZ
layers.
Inflow - negative
yes
Can be positive (outflow) if
negative precipitation
option specified.
uz.rech
Recharge out of the bottom of the Inflow - negative
UZ soil column to SZ via the UZ
soil matrix.
In the MIKE SHE results,
recharge is a vertical downward
flow, thus in the negative direction. This is the same sign as the
water balance convention of negative inflow.
130
Included
in Wbl
Error
yes
MIKE SHE
Available Water Balance Items
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
uz.rechmp
Recharge out of the bottom of the Inflow - negative
UZ soil column to SZ via the UZ
macropores or by-pass flow.
In the MIKE SHE results,
recharge is a vertical downward
flow, thus in the negative direction. This is the same sign as the
water balance convention of negative inflow.
yes
sz.qOlSzPos
Upward flow directly from SZ to Outflow - positive
ponded water.
This is non-zero only when the
groundwater table is at or above
the ground surface.
The sign is positive upwards
which is the same as the positive
outflow water balance sign convention.
yes
sz.qOlSzNeg
Inflow - negative
Downward flow directly from
ponded water to SZ.
This is non-zero only when the
groundwater table is at or above
the ground surface.
The sign is positive upwards
which is the same as the negative
inflow water balance sign convention.
yes
sz.EtSz
Evapotranspiration directly from
SZ.
Positive - outflow
yes
sz.qSzIn
Inflow - negative
Inflow to SZ storage across the
boundary of the model, or inflow
across the boundary of the water
balance sub-area.
Inflow from internal fixed head
cells is also included in this term.
yes
131
Using the Water Balance Tool
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
sz.qSzOut
Outflow from SZ storage across Outflow - positive
the boundary of the model, or outflow across the boundary of the
water balance sub-area.
Outflow to internal fixed head
cells is also included in this term,
as well as drainage to local
depressions that contain a fixed
head boundary condition.
yes
sz.dSzSto
Change in SZ storage
Positive when storage
increases
yes
sz.qSzAbsEx
Groundwater pumping from SZ.
This does not include irrigation
wells and shallow irrigation wells,
but includes outflow to fixed head
drain internal boundary conditions.
yes
Outflow - positive
Can be negative (Inflow) if
injection specified for
wells.
sz.qSzDrIn
SZ drainage to local depressions Inflow - negative
in the current water balance area
from areas outside of the current
water balance sub-area.
This term also includes inflow to
the SZ drainage system added via
OpenMI.
yes
sz.qSzDrOut
SZ drainage to the model bound- Outflow - positive
ary, SZ drainage removed directly
from the model.
This term also includes SZ drainage to local depressions located
outside of the current water balance sub-area.
yes
sz.qSzDrToRivIn
SZ drainage to MIKE SHE River Outflow - positive
Links inside of the water balance
sub-area.
yes
132
Sign Convention in the
Water balance
Included
in Wbl
Error
MIKE SHE
Available Water Balance Items
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
sz.qSzDrToRivEx
SZ drainage to MIKE SHE River Outflow - positive
Links outside of the water balance
sub-area.
This can only be non-zero if the
water balance is calculated for a
sub-area.
yes
sz.qSzDrToM11HPoi SZ drainage to specified MIKE 11 Outflow - positive
nt
h-points.
These are specified in the Extra
Parameter option in SZ Drainage
to Specified MIKE 11 H-points
(V.1 p. 316)
yes
sz.qSzRivPos
Baseflow from SZ to MIKE River Outflow - positive
Links
yes
sz.qSzRivNeg
Infiltration from MIKE SHE
River Links to SZ
Inflow - negative
yes
ol.qSZToFloodPos
Direct flow upwards from SZ to
an active flood code cell
(active means that it is actually
flooded and the water level is controlled by the water level in MIKE
11).
This is a positive upwards flow in
the MIKE SHE results files.
Only non-zero when the groundwater table is at or above the
ground surface.
Outflow - positive
(Note sign change compared to detailed Ponded
Storage water balance)
yes
133
Using the Water Balance Tool
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
ol.qSZToFloodNeg
Direct flow downwards from an
an active flood code cell to SZ.
(active means that it is actually
flooded and the water level is controlled by the water level in MIKE
11).
This is a negative downwards
flow in the MIKE SHE results
files.
Only non-zero when the groundwater table is at or above the
ground surface.
Inflow - negative
(Note sign change compared to detailed Ponded
Storage water balance)
yes
sz.qGihbPos
Outflow from SZ storage to inter- Outflow - positive
nal general head boundaries (GHB
cells)
yes
sz.qGihbNeg
Inflow from internal general head Inflow - negative
boundaries (GHB cells) to SZ
storage
yes
sz.qIrrWell
Groundwater pumping from irrigation wells.
This includes both specified irrigation wells and shallow wells.
Outflow - positive
yes
sz.qSzDrToMouse
Outflow - positive
SZ drainage to specified
MOUSE/MIKE Urban manholes.
These are specified in the Extra
Parameter option in SZ Drainage
to MOUSE (V.1 p. 319)
yes
sz.qSzMousePos
Outflow from SZ storage to
Mouse/MIKE Urban pipes
yes
sz.qSzMouseNeg
Inflow from Mouse/MIKE Urban Inflow - negative
pipes to SZ storage
yes
sz.qSzExtSink
Outflow to external sinks specified via OpenMI
yes
134
Outflow - positive
Outflow - positive
MIKE SHE
Available Water Balance Items
Table 5.5
SZ - Saturated Zone - all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
sz.qSzExtSource
Inflow from external sources
specified via OpenMI
Inflow - negative
yes
sz.szWblErrTot
Aggregated SZ water balance
error for all layers
Positive if water generated
(∆storage + Outflow >
Inflow)
Saturated zone layers
The saturated zone water balance can also be calculated by numerical
layer. This means that all of the items in Table 5.5 are repeated for each
numerical layer. However, in this case the water balance error term,
sz.szWblErrTot is replace by a water balance error for each layer.
The layer water balance is slightly more complicated. It includes terms for
the exchange between layers, and the upper layer includes the terms for
the exchange with UZ and ponded water.
In particular, the output for each SZ layer water balance only includes the
exchange with the layer above. This is found in the two additional layer
water balance terms qSzZpos and qSzZneg.
The first term, qSzZpos, is the flow from the current layer upwards to the
layer above. In the results files, this term is in the positive (upwards)
direction. In the water balance, the term is also a positive outflow.
The second term, qSzZneg, is the flow from the layer above downwards
into the current layer. In the results files, this term is in the negative
(downwards) direction. In the water balance, the term is also a negative
inflow to the current layer.
Note: The layer water balance error includes the flows to and from the
layers above and below. However, when summing up the flows, the sign
135
Using the Water Balance Tool
must be changed for the qSzZpos and qSzZneg terms that originate from
the layer below.
Table 5.6
SZ - Saturated Zone - specified by layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
sz.qSzZpos
Upward SZ flow from the current Outflow - positive
layer to the layer above
Only available for LAYER water
balances
yes
sz.qSzZNeg
Downward SZ flow from the layer Inflow - negative
above to the current layer.
Only available for LAYER water
balances
yes
sz.szWblErr
SZ water balance error for the cur- Positive if water generated
(∆storage + Outflow >
rent layer
only available for LAYER water Inflow)
balances
Saturated Zone Linear Reservoir water balance
If the linear reservoir method is used for the saturated zone, the water balance terms are basically the same but are slightly less transparent.
The layer output for the linear reservoir method divides the SZ into two
layers - the interflow reservoirs and the baseflow reservoirs. For the linear
reservoir layers, there is no distinction between the two parallel baseflow
reservoirs, or the cascading interflow reservoirs.
The items listed in Table 5.7 are those found in the “Saturated Zone - layers(Linear Reservoir)” water balance output in the water balance configuration file:
[WblTypeDefinition]
Name = 'SZ_LAYER_LR'
DisplayName = 'Saturated Zone - layers(Linear Reservoir)'
Description = 'Saturated zone water balance for linear reservoir'
NoGroups = 13
Group = 'recharge(uz.qrech+uz.qrechmp)'
136
MIKE SHE
Available Water Balance Items
Group = 'evapotranspirationSZ(sz.qetsz)'
Group = 'lateral IN(sz.qszin)'
Group = 'lateral OUT(sz.qszout)'
Group = 'percolation(sz.qszzneg)'
Group = 'To river(sz.qszrivpos)'
Group = 'From river(sz.qszrivneg)'
Group = 'storagechange(sz.dszsto)'
Group = 'deadzonestoragechange(sz.dszsto_dead)'
Group = 'pumping(sz.qszabsex)'
Group = 'Irr.pumping(sz.qirrwell)'
Group = 'feedbackUZ(sz.qUZfeedback)'
Group = 'Error(sz.szwblerr)'
EndSect // WblTypeDefinition
Table 5.7
Item
SZ - Saturated Zone - Linear Reservoir all layers
Description
recharge
This is the total recharge into the
(uz.qrech+uz.qrechm interflow reservoirs.
p)
If UZ is not simulated, then
uz.qrech is still calculated based
on the infiltration from OL.
evapotranspirationSZ (sz.qetsz)
Sign Convention in the
Water balance
Included
in Wbl
Error
Inflow - negative
yes
Outflow - postitive
This is the direct ET from the
water table. In the LR SZ method,
the water table is constant and
fixed at the beginning of the simulation. If the root zone reaches the
water table, then ET will be taken
from the water table as an infinite
sink when the reference ET is not
satisfied by the other sources.
yes
137
Using the Water Balance Tool
Table 5.7
SZ - Saturated Zone - Linear Reservoir all layers
Item
Description
lateral IN
(sz.qszin)
In the LR SZ model, infiltration to Inflow - negative
the interflow reservoirs and percolation to the baseflow reservoirs is
distributed equally to the entire
reservoir.
When you calculate the water balance in a sub-area, sz.qszin is the
amount of recharge/percolation
that is distributed into the subarea.
For example, if all your recharge
occurs outside of your sub-area,
this is the increase in groundwater
storage that occurs inside your
sub-area.
This can only be non-zero for subarea water balances.
yes
lateral OUT
(sz.qszout)
In the LR SZ model, infiltration to Outflow - positive
the interflow reservoirs and percolation to the baseflow reservoirs is
distributed equally to the entire
reservoir.
When you calculate the water balance in a sub-area, sz.qszout is the
amount of recharge/percolation
that is distributed to areas outside
of the sub-area.
For example, if all your recharge
occurs inside your sub-area, this is
the increase in groundwater storage that occurs outside your subarea.
This can only be non-zero for subarea water balances.
yes
138
Sign Convention in the
Water balance
Included
in Wbl
Error
MIKE SHE
Available Water Balance Items
Table 5.7
SZ - Saturated Zone - Linear Reservoir all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
percolation
(sz.qszzneg)
Infiltration from interflow reser- Inflow - negative
voirs to baseflow reservoirs.
(to the baseflow reservoir)
This is defined only for the lower
(baseflow) layer in the water balance output, but is used in the
water balance error calculation of
the interflow reservoirs with the
opposite sign.
The term sz.qszzpos is not
included here because the LR
method does not allow any transfer of water from the baseflow reservoir upwards to the interflow
reservoir.
yes
To river
(sz.qszrivpos)
Outflow from interflow and base- Outflow - positive
flow reservoirs to MIKE SHE
River Links.
yes
From river
(sz.qszrivneg)
Inflow - negative
Inflow from MIKE SHE River
Links to the baseflow reservoir.
(to the baseflow reservoir)
For the interflow reservoirs, this is
always zero because MIKE 11
only discharges to the baseflow
reservoirs.
yes
storagechange
(sz.dszsto)
Change in storage in the interflow Positive if storage increases yes
and baseflow reservoirs.
deadzonestoragechange
(sz.dszsto_dead)
Change in storage in the deadzone Outflow - Positive
storage. This is calculated as a
change in storage, but it is equal to
the outflow to dead zone storage
because there is no option in
MIKE SHE to reduce the dead
zone storage.
yes
139
Using the Water Balance Tool
Table 5.7
SZ - Saturated Zone - Linear Reservoir all layers
Item
Description
Sign Convention in the
Water balance
Included
in Wbl
Error
pumping
(sz.qszabsex)
Groundwater pumping from the
baseflow reservoirs.
This is always zero for the interflow reservoirs.
Outflow - positive
But, can be negative if
injection rates specified in
wells
yes
Irr.pumping
(sz.qirrwell)
This is the sum of groundwater
Outflow - positive
pumping for irrigation - irrigation
wells + shallow irrigation wells.
yes
feedbackUZ
(sz.qUZfeedback)
This is a fraction of the discharge Outflow - positive
from the baseflow reservoirs to
(from baseflow reservoirs
MIKE 11 to account for discharge only)
to riparian zones that is lost to ET.
yes
Error
(sz.szwblerr)
SZ water balance error for the cur- Positive if water generated
(∆storage + Outflow >
rent layer
only available for LAYER water Inflow)
balances
Error
(sz.szwblerrtot)
Positive if water generated
SZ water balance error for the
both the interflow and baseflow (∆storage + Outflow >
Inflow)
reservoirs combined.
This is only available for the total
water balance option.
5.3.6
Limitations for Linear Reservoir and Sub-catchment OL Water Balance
The water balance calculations have the following restrictions on singlecell, sub-catchment water balances, with the SZ Linear Reservoir and
Simple OL:
z
140
single-cell : won't be correct for TOTAL, OL, SZ water balances. But
can be used for UZ and others.
MIKE SHE
Available Water Balance Items
z
sub-catchment: For TOTAL and OL water balances the smallest valid
water balance sub-catchment is one Overland flow zone (i.e. topographical zone) within one hydrological sub-catchment. If a water balance sub-catchment excludes part of an Overland flow zone within one
hydrological sub-catchment, the water balances will be wrong in many
cases because the OL storage is not necessarily uniformly distributed
over one Overland flow zone, while there is only one value for flows
between OL flow zones, source/sink terms, etc.
z
For TOTAL and SZ water balances: Same restrictions apply, but here
with the interflow reservoirs.
There are no restrictions with respect of the baseflow reservoir distributions.
The pre-processor warns in case the above restrictions are violated. It
can't give an error, because this program doesn't know which type
(TOTAL/OL/SZ/...) the user will specify in the water balance Post-processor.
Basically, sub-catchment water balances can be misleading when using
the linear reservoir method. For example, a baseflow reservoir receiving
percolation from several subcatchments only "sees" the total amount of
percolation. If you make a sub-catchment water balance for one of the
sub-catchments, then the water balance program will return the amount of
percolation for the subcatchment. However, the baseflow reservoir only
received the "average" over the area (total percolation/baseflow res. area).
The difference between these two values will be reflected in the water balance as a "boundary flow" for the sub-catchment, which is obviously not
really correct. The same situation applies for river link infiltration to baseflow reservoirs.
141
Using the Water Balance Tool
5.4
Standard Water Balance Types
Table 5.8 summarizes the 31 standard water balance types defined in the
water balance configuration file. Some of the water balances cannot be
used in certain conditions and these restrictions are listed in the table.
Table 5.8
Water balance types available in the default configuration files.
Water balance type
Description
Total waterbalance
General water balance of the entire model
setup
Error of each component
The water balance error of each model component
Snow Melt component
Snow Melt component water balance items
Canopy Interception component
Canopy Interception component water balance items
Overland flow
Overland component water balance
Overland flow - detailed
Detailed Overland component water balance
Unsaturated Zone
Unsaturated zone component water balance
Unsaturated Zone - detailed Unsaturated zone component water balance
Saturated Zone
Saturated zone component water balance
(depth-integrated)
Saturated Zone - layer(s)
Saturated zone component water balance
(each or specified layer)
Saturated Zone - detailed
Detailed Saturated zone component water
balance (depth-integrated)
Saturated Zone - detailed - Detailed Saturated zone component water
layer(s)
balance (each or specified layer)
142
Saturated Zone (Linear
Reservoir)
Saturated Zone component water balance
for the linear reservoir
Saturated Zone -layers
(Linear Reservoir)
Saturated Zone component water balance
for the linear reservoir
Irrigation component
Irrigation component water balance
MIKE SHE
Standard Water Balance Types
Table 5.8
Water balance types available in the default configuration files.
Water balance type
Description
MOUSE-coupling terms
MIKE SHE - MOUSE exchange (depthintegrated)
MOUSE-coupling terms,
Saturated zone - layer(s)
MIKE SHE sat.zone - MOUSE exchange
(each or specified layer)
Map output: Total error
Distributed output: Total water balance
error
Map output: Overland flow Distributed output: Overland water balance
error
error
Map output: Unsat. Zone
error
Distributed output: Unsat.zone water balance error
Map output: Sat. Zone error Distributed output: Saturated zone water
balance error (depth-integrated)
Map output: Sat. Zone error Distributed output: Saturated zone water
- layer(s)
balance error (each or specified layer)
Map output: Total irrigation Distributed output: Total irrigation
Chart output: Total water
balance
Chart output: General water balance of the
entire model (depth-integrated)
Chart output: Total + each
SZ layer
Chart output: General water balance of the
entire model (each SZ layer)
Chart output: Total water
Chart output: Generel vandbalance for hele
balance TEXT IN DANISH modellen (dybde-integreret)
Chart output: Total + each
SZ layer TEXT IN DANISH
Chart output: Generel vandbalance for hele
modellen (hvert SZ-lag)
Saturated Zone
StorageSaturated zone Storage (depth-integrated)
Saturated Zone
Storage - layer(s)
Saturated zone Storage (each or specified
layer)
Map output: Saturated Zone Distributed output: Saturated zone Storage
Storage
(depth-integrated)
Map output: Saturated Zone Distributed output: Saturated zone Storage
Storage - layer(s)
(each or specified layer)
143
Using the Water Balance Tool
5.5
Making Custom Water Balances
The first combobox in the Post-processing dialogue contains a list of all
the available water balance types. This list is read from the water balance
configuration file, MSHE_Wbl_Config.pfs, which is found in the
MIKE SHE installation \bin directory. The default location of this directory depends on the operating system of your computer.
You can add extra items to the list of available water balance types by
defining additional water balances at the end of the configuration file.
To illustrate how you could add an additional water balance type, the table
below describes the format for each line of the water balance type definition. The example is for an extra water balance type to calculate the net
vertical flow in a specified SZ layer. This water balance type can only be
used with the single-cell resolution and specified output layers options.
Line item
Comment
// Created: 2004-06-2 16:28:48
// DLL id : C:\WINOWS\System32\pfs2000.dll
// PFS version: Mar 3 2004 21:35:12
File header
[MIKESHE_WaterBalance_ConfigFile]
FileVersion = 3
NoWblTypes = 31
NoWblTypes = the number of water balance types in the configuration file.
Remember to change this number if you
add a water balance item to the file
[WblTypeDefinition]
Name = 'TOTAL'
...
Group = 'SZ Storage(sz.szsto)'
EndSect // WblTypeDefinition
Existing water balance definitions
[WblTypeDefinition]
First line of the water balance definition
Name = 'SZ_LAYER_NET_VERT_FLOW_MAP'
Internal name. No spaces allowed
DisplayName = 'Map output: Net Vertical Name displayed in the combobox
Saturated Zone Flow - layer(s)'
144
Description = 'Distributed output:
Saturated zone Storage (specified
layer)'
Description displayed under the combobox
NoGroups = 1
Number of calculation groups in the output file
MIKE SHE
Making Custom Water Balances
Line item
Comment
Group = 'SZ Vertical Flow(sz.qszzpos+
sz.qszzneg)'
EndSect
EndSect
Definition of the calculation group, consisting of a name and a sum of the particular water balance items (no spaces) from
Table 5.1 to Table 5.7. Map items can
only have one group (NoGroups = 1)
// WblTypeDefinition
// MIKESHE_WaterBalance_ConfigFile
last line in the file
When making custom water balance types the format of the default water
balance configuration file must be maintained. Variable names, including
names in square brackets, are case sensitive and the number of spaces in
variable names must be consistent with the default configuration file.
5.5.1
Customizing the chart output
The chart water balance is a special water balance function that creates an
ASCII file that is read by another program to generate the graphic in
Figure 5.1.
The default setup of the items in the chart output do not follow the typical
sign convention of the water balance. The sign convention has been
adjusted to make the chart output more logical. Thus, in the chart output
both precipitation and evapotranspiration are positive values. Whereas, in
the standard water balance, precipitation is negative.
The items included in the graphic are in the water balance configuration
file. The Group sections include a range of options for displaying the output on the graphic, including arrow directions and locations.
145
Using the Water Balance Tool
.
Line item
Comment
// Created: 2004-06-2 16:28:48
// DLL id : C:\WINOWS\System32\pfs2000.dll
// PFS version: Mar 3 2004 21:35:12
File header
[MIKESHE_WaterBalance_ConfigFile]
FileVersion = 3
NoWblTypes = 31
[WblTypeDefinition]
Name = 'TOTAL'
...
Group = 'SZ Storage(sz.szsto)'
EndSect // WblTypeDefinition
Existing water balance definitions
[WblTypeDefinition]
First line of the water balance definition
Name = 'TOTAL_CHART'
Internal name. No spaces allowed
DisplayName = 'Chart output: Total
Water balance'
Name displayed in the combobox
Description = 'Chart output: General water
balance of the entire model (depthintegrated)'
NoGroups = 23
Group = 'SKY TV 45 40 Precipitation(sm.qp)'
EndSect
EndSect
146
NoWblTypes = the number of water balance types in the configuration file.
Remember to change this number if you
add a water balance item to the file
Description displayed under the combobox
Number of calculation groups in the output file
Various display items for the arrows and
items
// WblTypeDefinition
// MIKESHE_WaterBalance_ConfigFile
last line in the file
MIKE SHE
RUNNING MIKE SHE
147
148
MIKE SHE
Preprocessing your model
6
RUNNING YOUR MODEL
In the top icon bar, there is a three-button set of icons for running your
model.
.
PP - The PP button starts the preprocessing. You must first PreProcess
your model data to create the numerical model from your grid independent
data. See Preprocessing your model (V.1 p. 149).
WM - The WM button starts the Water Movement simulation. You can
only run your water movement simulation after you have preprocessed
your data. See Running your Model (V.1 p. 149).
WQ - The WQ button starts the Water Quality simulation. After you have
successfully run a water movement simulation to completion, you can run
a water quality simulation.
In addition to the three icon buttons, there is a Run menu. In this menu,
you can check on and off all three of the above options. Finally, there is an
Execute... menu sub-item that runs only the checked items above it. The
Execute option can also be launched using the Alt - R - E hot-key
sequence.
6.1
Preprocessing your model
In the Setup Tab, you specify the input data required by the model including the size of the model and the numerical grid. However, most of
the setup data is independent of the model extent and grid. When you preprocess you model set up, MIKE SHE’s pre-processor program scans
through your model set up and interpolates all spatial data to the specified
model grid. This interpolated set up data is stored in a .fif file, which is
read during the simulation by the MIKE SHE engine. However, the preprocessed data does not include any time information. All time series
information must be interpolated dynamically during the run because
MIKE SHE dynamically changes the time step during the simulation in
response to stresses on the system.
The Preprocessed Data Tab is used to display the pre-processed data.
Before you run your simulation, you should carefully check the preprocessed data for errors. Errors found in the preprocessed data are typically
related to incorrectly specified parameters, file names, etc. in the Setup
Tab.
Running MIKE SHE
149
Running your Model
On the main pre-processed dialogue, there is a uneditable text box containing the file and location of the pre-processed data. This is a .pfs ASCII
file containing the file references for all of the data. The actual data is
stored in a .fif file, as well as a number of dfs2 and dfs3 files.
After you have successfully preprocessed your model, the pre-processed
data will be automatically loaded when you expand the data tree. The data
tree reflects all the spatial data defined in the model set up tab. In other
words, if the overland flow is not included in the Simulation Specification
(V.2 p. 27) dialogue, then the Overland item will not be included in the
pre-processed data tree.
Note If you change your model setup data, the pre-processed data will not
reflect the changes until you pre-process your model again.
6.1.1
Viewing the pre-processed data
In all map and time series views, there is a View button. This view button
will open the dfs0, dfs2 or dfs3 file that was generated by the pre-processor in either the Grid Editor or the Time Series Editor. However, each of
these files usually contains a large number of data items. The Grid or the
Time Series Editor opens at the first item, so you must use the scrolling
function in the editor to find the data item that you want.
6.1.2
Editing the pre-processed data
MIKE SHE reads only the .fif file during the simulation. The .dfs2 and
dfs3 files are created to make it easier to view and plot the preprocessed
data. If you edit the dfs2 or dfs3 files, the changes will not be used in the
simulation.
If you want to change the pre-processed data and use the changed data in
the simulation, you have a couple of options.
Option 1
1 Right click on the map view and save the data to a new dfs2 file,
2 open the new dfs2 file in the Grid Editor, and
3 make the changes in the new dfs2 file and save the file.
Option 2
1 Use the View button to open the dfs2 or dfs3 pre-processed file in the
Grid Editor,
2 make your changes in the file, and
3 save the file with a new name.
150
MIKE SHE
Pre-processed data items
In both options above, you then use the new dfs2 or dfs3 file as input in
the Setup tab.
6.2
Pre-processed data items
The following sections describe in more detail some of the pre-processed
data items.
6.2.1
MIKE 11 coupling
The coupling between MIKE 11 and MIKE SHE is made via river links,
which are located on the edges that separate adjacent grid cells. The river
link network is created by the pre-processor, based on the MIKE 11 coupling reaches. The entire river system is always included in the hydraulic
model, but MIKE SHE will only exchange water with the coupling
reaches.
The location of each of MIKE SHE river link is determined from the coordinates of the MIKE 11 river points, where the river points include both
digitised points and H-points on the specified coupling reaches. Since the
MIKE SHE river links are located on the edges between grid cells, the
details of the MIKE 11 river geometry can be only partly included in
MIKE SHE, depending on the MIKE SHE grid size. The more refined the
MIKE SHE grid, the more accurately the river network can be reproduced.
This also leads to the restriction that each MIKE SHE grid cell can only
couple to one coupling reach per river link. Thus, if, for example, the distance between coupling reaches is smaller than half a grid cell, you will
probably receive an error, as MIKE SHE tries to couple both coupling
reaches to the same river link.
The river links are shown on Rivers and Lakes data tree pages, as well as
the SZ Drainage to River page.
Related Items:
z Surface Water in MIKE SHE (V.1 p. 171)
z
6.2.2
Coupling of MIKE SHE and MIKE 11 (V.1 p. 199)
Land Use
The vegetation distribution is displayed on a map, but if you use the vegetation database for specifying the crop rotation, this information will not
be displayed in the pre-processor.
Running MIKE SHE
151
Running your Model
Shape files
If you have used shape files for the Land Use distribution, then the PP output order may not reflect the input order if the polygons are labeled with
text strings. In this case, the PP program reads the polygons and orders
them in the order that they are encountered during the pre-processing. The
6.2.3
Unsaturated Flow
The Unsaturated Flow data tree in the pre-processed data contains a two
noteworthy data items.
Soil profiles
Under the unsaturated zone, you will find a map with the grid codes for
each of the soil profiles used. Accompanying this map is a text page containing the details of all the soil profiles. At the top of this page is the path
and file name of the generated text file, which you can open in any text
editor.
Note If you are using one of the finite difference methods, the pre-processor modifies the vertical discretisation wherever the vertical cell size
changes. Thus, if you have 10 cells of 20cm thickness, followed by 10
cells of 40cm thickness, the location of the transition will be moved such
that the two cells on either side will be have an equal thickness. In this
case, cells 10 and 11 will both be 15cm.
UZ Classification Codes
If certain conditions are met, then the flow results for a 1D unsaturated
zone column can be applied to columns with similar properties. If you
chose to use this option, then a map will be generated that shows the calculation cells and the corresponding cells to which the results will be copied.
The cell with a calculation is given an integer grid code with a negative
value. The flows calculated during the simulation in the cells with the negative code, will be transferred to all the cells with the same positive grid
code value. For example, if an UZ recharge to SZ of 0.5 m3/day is calculated for UZ grid code -51, then all the SZ cells below the UZ cells with a
grid code of +51 will also be given the same recharge.
Tip This map can be difficult to interpret without using the Grid Editor.
Related Items:
z Unsaturated Zone (V.2 p. 127)
z
152
Soil Profile Definitions (V.2 p. 132)
MIKE SHE
Pre-processed data items
6.2.4
z
Partial automatic classification (V.2 p. 137)
z
Specified classification (V.2 p. 138)
Saturated Flow
The saturated zone data is generally written to a dfs3 file. In the map view,
there is a combo box where you can specify the layer that you want to
view.
Specific Yield of upper SZ layer
MIKE SHE forces the specific yield of the top SZ layer to be equal to the
“specific yield” of the UZ zone as defined by the difference between the
specified moisture contents at saturation, θs, and field capacity, θfc.This
correction is calculated from the UZ values in the UZ cell in which the initial SZ water table is located. This is reflected in the pre-processed data.
For more information on the SZ-UZ specific yield see Specific Yield of
the upper SZ numerical layer (V.1 p. 252).
Saturated Zone Drainage
The rate of saturated zone drainage is controlled by the drain elevation and
the drain time constant. However, the destination of the drainage water is
controlled by the drain levels and the drain codes, which determine if the
water flows to a river, a boundary, or a local depression. The algorithm for
determining the drainage source-recipient reference system is described in
Groundwater Drainage (V.1 p. 60).
During the preprocessing, each active drain cell is mapped to a recipient
cell. Then, whenever drainage is generated in a cell, the drain water will
always be moved to the same recipient cell. The drainage source-recipient
reference system is displayed in the following two grids
Drainage to local depressions and boundary - This grid displays all the
cells that drain to local depressions or to the outer boundaries. All drainage from cells with the same negative value are drained to the cell with the
corresponding positive code. If there is no corresponding positive code,
then that cell drains to the outer boundary, and the water is simply
removed from the model. Cells with a value of zero either do not generate
drainage, or they drain to a river link.
Drainage to river - This grid displays all of the cells that drain to river
links. All drainage from cells with the same negative value are drained to
the cell with the corresponding positive code. Cells with a value of zero
either do not generate drainage, or they drain to a the outer boundary or a
local depression.
Running MIKE SHE
153
Running your Model
Related Items:
z Groundwater Drainage (V.1 p. 60)
6.3
z
Drainage (V.2 p. 173)
z
Drain Level (V.2 p. 176)
z
Drain Time Constant (V.2 p. 177)
z
Drain Codes (V.2 p. 178)
z
Option Distribution (V.2 p. 179)
The Results Tab
All the simulation results are collected in the Results tab. This includes
Detailed time series output for both MIKE SHE and MIKE 11, as well as
Grid series output for MIKE SHE.
A Run Statistics tool is available for helping you assimilate the calibration
statistics for each of the detailed time series plots.
A link to the GeoScene3D program is also included, where you can visualize your results in a dynamic 3D environment. This program is widely
used in Denmark and can be independently purchased from www.i-gis.dk.
The Results post-processing section contains options for post-processing
the random walk particle tracking results.
154
MIKE SHE
The Results Tab
6.3.1
Detailed Time Series Results
The MIKE SHE Detailed time series tab includes an HTML plot of each
point selected in the Setup Editor. The HTML plots are updated during the
simulation whenever you enter the view. Alternatively, you can select the
Refresh button to refresh the plot.
Note: The HTML plot is regenerated every time you enter the Detailed
Time Series page. So, if you have a lot of plots and a long simulation, then
the regeneration can take a long time.
For information on the statistics see Statistic Calculations (V.2 p. 217).
Running MIKE SHE
155
Running your Model
6.3.2
Gridded Results
Gridded data results for MIKE SHE can be viewed by selecting the Gridded Data Results Viewer item on the Results tab. The table is a list of all
gridded data saved during a MIKE SHE simulation. The items in this list
originate from the list of items selected in the Grid series output
(V.2 p. 192) dialogue from the Setup tab.
Clicking on the View result button will open the Results Viewer to the current item. All overlays from MIKE SHE (e.g. shape files, images, and grid
files) will be transferred as overlays to the result view. However, the
MIKE 11 river network is not transferred as an overlay.
Layer number - For 3D SZ data files, the layer number can be specified
at the top of the table. However, the layer number can be changed from
within the Results Viewer (see Adding additional result files and overlays
(V.1 p. 83)) By default the top layer is displayed.
Vectors - Vectors can be added to the SZ plots of results, by checking the
Add X-Y flow vectors checkbox. These vectors are calculated based on the
Groundwater flow in X-direction and Groundwater flow in Y-direction
data types if they were saved during the simulation.
In the current version, velocity vectors cannot be added for overland flow
output.
156
MIKE SHE
The Results Tab
The “Overwrite existing file” warning
When the Result Viewer opens one of the items in the table, it creates a
setup file for the particular view with the extension .rev. The name of the
current .rev file is displayed in the title bar of the Results Viewer.
Initially, the .rev file includes the default view settings and the overlay
information from MIKE SHE. However, if you make changes to the view,
such as changes in the way contours are displayed, when you close the
view, you will be asked if you want to save your changes. The .rev file can
be opened directly at any time and your results will be displayed using the
saved settings.
However, the next time you open the item in the table, you will be asked if
you want to overwrite the existing .rev file. If you click on “Yes”, then a
new .rev file will be created. If you click on “No”, then your previous settings will be re-loaded, and your results will open with the settings from
the previous time you opened these results.
Running MIKE SHE
157
Running your Model
6.3.3
MIKE 11 Detailed Time Series
The MIKE 11 Detailed time series tab includes an HTML plot of each
point selected in the Setup Editor. The HTML plots are updated during the
simulation whenever you enter the view. Alternatively, you can select the
Refresh button to refresh the plot.
Note: The HTML plot is regenerated every time you enter the Detailed
Time Series page. So, if you have a lot of plots and a long simulation, then
the regeneration can take a long time.
For information on the statistics see Statistic Calculations (V.2 p. 217).
6.3.4
158
Run Statistics
Run statistics can be generated in HTML format for a MIKE SHE simulation. The run statistics table information can be copied and pasted directly
into any word processing program, such as Microsoft Word, or spreadsheet, such as Microsoft Excel. The Run Statistics HTML document
MIKE SHE
The Results Tab
includes MIKE SHE and MIKE 11 results for all items included in the
MIKE SHE and MIKE 11 detailed time series sections that also include
observation data.
To calculate Run Statistics for a simulation, navigate to the Results Tab
and the Run Statistics item on the menu tree. Press the Generate Statistics
button on the Run Statistics window to perform the statistical calculations.
For some simulations with long simulation periods and/or a lot of calibration data it can take a while to generate the run statistics.
After successful completion of the Generate Statistics phase, the Run Statistics HTML document will be displayed in the window on the Run Statistics page (see below).
Similar to the detailed time series output, the Run Statistics can be viewed
during a simulation. Press the Refresh button on the Run Statistics page to
update the Run Statistics using the most recent model results during a simulation
For information on the statistics see Statistic Calculations (V.2 p. 217).
Shape file output for run statistics
A shape file of statistics is also generated when the html document is generated. The shape file contains all of the information contained in the
HTML document and can be used to generate maps of model errors that
can be used to evaluate spatial bias. The shape file is created in the simulation directory and is named ProjectName_Stat.shp where ProjectName is
the name of the *.she file for the simulation. Note: the Run Statistics shape
file does not have a projection file associated with it and this file should be
created using standard ArcGIS methods.
The statistics contained in the HTML document and the shape file are calculated using the same methods used to calculate statistics for the detailed
time series output. The reader is referred to the Detailed Time Series Output section for more information on how the statistics are calculated.
Running MIKE SHE
159
Running your Model
6.4
Controlling your simulation
6.4.1
Model Limits
Although, there are no physical limits to the size of your model, there are
practical limits and hardware limits.
The practical limits are generally related to run time. We all want the
model to be a little bit bigger or more detailed. However, that little extra
detail or slightly smaller grid size can quickly lead to long run times.
The physical limits are generally related to memory size. If you model
requires more memory than is physically installed on the computer, then
the computer will start to swap data to the hard disk. This will vastly slow
down your simulation. The section, Hardware Requirements (V.1 p. 24),
outlines some hardware considerations when using MIKE SHE.
If your model reaches the practical or physical limits of your computer,
then may we suggest the following:
1 Critically evaluate your model to see if you really need such a large,
complex model. For example, you may be able to reduce the number of
UZ elements or the slightly increase the grid size.
2 Do a rough calibration with a smaller model first. The model independent structure of MIKE SHE makes it reasonable to refine your model
later with a minimum of effort. For example, you can use Gravity flow
instead of Richards equation, double the grid spacing, or shorten the
calibration period, during the initial calibration and switch back to the
original during the final calibration. You might even be surprised that
the rougher model is actually good enough.
6.4.2
Speeding up your simulation
In most cases, the best way to speed up your model is to make it simpler.
You should look very carefully at your model and ask yourself the following questions, for example:
z
160
Do you really need a fine discretisation during calibration? A
coarser grid may allow you to do many more calibration runs. Then
when the model is calibrated, you can refine the grid for the final simulations - but remember to check you calibration first.
MIKE SHE
Controlling your simulation
z
Do you really need the Richards equation for unsaturated flow?
For regional models, the two layer water balance method is usually
sufficient, which is very fast. The gravity flow method is also, typically
2-5 times faster than the Richards equation method. Again during the
calibration it can be a good idea to use one of the simpler methods and
the more detailed method for the final simulations.
z
Is your MIKE 11 simulation too detailed? If your MIKE 11 crosssections are too close together, MIKE 11 will run with a very short time
step. Regional models can often be run with the simple routing methods in MIKE 11, which are very fast.
If your simulation is still too slow, then several sections in the manual
might be of help. In particular,
6.4.3
z
Hardware Requirements (V.1 p. 24) contains information on different
hardware configurations,
z
Controlling the Time Steps (V.1 p. 161) contains information on how
the dynamic time step control works,
z
Overland Flow Performance (V.1 p. 178) contains information on how
to improve the efficiency of the overland flow solution, which can be
very time consuming if you have permanently ponded water,
z
Parallelization of MIKE SHE (V.1 p. 167) contains information on the
using MIKE SHE with multi-core PCs and 64-bit operating systems.
Controlling the Time Steps
Each of the main hydrologic components in MIKE SHE run with independent time steps. Although, the time step control is automatically controlled, whenever possible, MIKE SHE will run with the maximum
allowed time steps.
The component time steps are independent, but they must meet to
exchange flows, which leads to some restrictions on the specification of
the maximum allowed time steps.
Running MIKE SHE
z
If MIKE 11 is running with a constant time step, then the Max allowed
Overland (OL) time step must be a multiple of the MIKE 11 constant
time step. If MIKE 11 is running with a variable time step, then the
actual OL time step will be truncated to match up with the nearest
MIKE 11 time step.
z
The Max allowed UZ time step must be an even multiple of the Max
allowed OL time step, and
161
Running your Model
z
The Max allowed SZ time step must be an even multiple of the Max
allowed UZ time step.
Thus, the overland time step is always less than or equal to the UZ time
step and the UZ time step is always less than or equal to the SZ time step.
If you are using the implicit solver for overland flow, then a maximum OL
time step equal to the UZ time step often works. However, if you are using
the explicit solver for overland flow, then a much smaller maximum time
step is necessary, such as the default value of 0.5 hours.
If the unsaturated zone is included in your simulation and you are using
the Richards equation or Gravity Flow methods, then the maximum UZ
time step is typically around 2 hours. Otherwise, a maximum time step
equal to the SZ time step often works.
Groundwater levels react much slower than the other flow components.
So, a maximum SZ time step of 24 or 48 hours is typical, unless your
model is a local-scale model with rapid groundwater-surface water reactions.
Precipitation-dependent time step control
Periods of heavy rainfall can lead to numerical instabilities if the time step
is too long. To reduce the numerical instabilities, the a time step control
has been introduced on the precipitation and infiltration components. You
will notice the effect of these factor during the simulation by suddenly
seeing very small time steps during storm events.
The parameters controlling the time step adjustment are in the Time Step
Control (V.2 p. 31) dialogue. In particular, the following three parameters
control the time step during rainfall events:
Max precipitation depth per time step If the total amount of precipitation [mm] in the current time step exceeds this amount, the time step
will be reduced by the increment rate. Then the precipitation time
series will be resampled to see if the max precipitation depth criteria
has been met. If it has not been met, the process will be repeated with
progressively smaller time steps until the precipitation criteria is satisfied. Multiple sampling is important in the case where the precipitation
time series is more detailed than the time step length. However, the criteria can lead to very short time steps during short term high intensity
events. For example, if your model is running with maximum time
steps of say 6 hours, but your precipitation time series is one hour, a
high intensity one hour event could lead to time steps of a few minutes
during that one hour event.
162
MIKE SHE
Controlling your simulation
Max infiltration amount per time step If the total amount of infiltration
due to ponded water [mm] in the current time step exceeds this
amount, the time step will be reduced by the increment rate. Then the
infiltration will be recalculated. If the infiltration criteria is still not
met, the infiltration will be recalculated with progressively smaller
time steps until the infiltration criteria is satisfied.
If your model does not include the unsaturated zone, or if you are using
the 2-Layer water balance method, then you can set these conditions up by
a factor of 10 or more. However, if you are using the Richards equation
method, then you may have to reduce these factors to achieve a stable
solution.
Input precipitation rate requiring its own time step If the precipitation
rate [mm/hr] in the precipitation time series is greater than this amount,
then the simulation will break at the precipitation time series measurement times. This option is added so that measured short term rainfall
events are captured in the model.
For example, assume you have hourly rainfall data and 6-hour time
steps. If an intense rainfall event lasting for only one hour was
observed 3 hours after the start of the time step, then MIKE SHE
would automatically break its time stepping into hourly time steps during this event. Thus, instead of a 6-hour time step, your time steps during this period would be: 3 hours, 1 hour, and 2 hours. This can also
have an impact on your time stepping, if you have intense rainfall and
your precipitation measurements do not coincide with your storing
time steps. In this case, you may see occasional small time steps when
MIKE SHE catches up with the storing time step.
Actual time step for the different components
As outlined above the overland time step is always less than or equal to
the UZ time step and the UZ time step is always less than or equal to the
SZ time step. However, the exchanges are only made at a common time
step boundary. This means that if one of the time steps is changed, then all
of the time steps must change accordingly. To ensure that the time steps
always meet, the initial ratios in the maximum time steps specified in this
dialogue are maintained.
After a reduction in time step, the subsequent time step will be increased
by
timestep = timestep × ( 1 + IncrementRate )
Running MIKE SHE
(6.1)
163
Running your Model
until the maximum allowed time step is reached.
Relationship to Storing Time Steps
The Storing Time Step specified in the Detailed time series output
(V.2 p. 186) dialogue, must also match up with maximum time steps.
Thus,
– The OL storing time step must be an integer multiple of the Max
UZ time step,
– The UZ storing time step must be an integer multiple of the Max
UZ time step,
– The SZ storing time step must be an integer multiple of the Max SZ
time step,
– The SZ Flow storing time step must be an integer multiple of the
Max SZ time step, and
– The Hot start storing time step must be an integer multiple of the
maximum of all the storing time steps (usually the SZ Flow storing
time step)
For example, if the Maximum allowed SZ time step is 24 hrs, then the SZ
Storing Time Step can only be a multiple of 24 hours (i.e. 24, 48, 72 hours,
etc.)
6.5
Using Batch Files
A ’batch’ file contains native DOS commands in a programming structure.
When executed each of the DOS commands in the batch file is executed
sequentially. Since, most MIKE Zero and MIKE SHE programs can be
executed in this way, a properly constructed batch file allows you to run
multiple models sequentially when you are not at the computer, such as
over night.
Basically, to run MIKE SHE in batch mode, you must
1
Setup the different models with different names using the Setup Editor
2 Create a .BAT file containing the DOS commands to run the models
3 Run the .BAT file and analyse the results using the standard MIKE
Zero analysis tools (e.g. the Results Viewer)
164
MIKE SHE
Using Batch Files
Setup the different models
Your original model can be saved to a new name and the necessary
changes made in the new set up. We highly recommended that you create
and set up the different models in the MIKE SHE Setup Editor. In principle, you could edit the .SHE file, which is a text file containing all of the
information on the model set up, but the file is typically very large and
confusing, and the format of this file must be preserved exactly.
Create the batch file
To create a batch file, you must create a text file with the extension .BAT.
Then add the DOS commands in the order that you would like them executed. But, before you can run the MIKE SHE executables, you must add
the MIKE SHE installation directory to your PATH variable. The default
installation directory depends on your operating system. For example, for
MS Vista (64-bit) the default directory is:
C:\Program Files(x86)\DHI\2011\bin\
The DOS command to add the default path to the PATH variable is:
Set PATH=%PATH%,C:\Program Files(x86)\DHI\2011\bin\
To run MIKE SHE from the batch file you must add the following two
DOS command lines after the PATH statement above:
MSHE_PreProcessor MyModel.she
MSHE_watermovement MyModel.she
The above two lines will run both the preprocessor and the water movement engine separately. If you want to run them together, then you can
replace the two lines with
MSHE_Simulation MyModel.she
The examples above will run silently. That is, no progress information will
be displayed. If you want to display progress information, then you should
use the MzLaunch utility. Using
MzLaunch.exe MyModel.she -e MSHE_Simulation
will leave the MzLaunch utility open when the simulation finishes,
whereas
MzLaunch.exe MyModel.she -e MSHE_Simulation -exit
Running MIKE SHE
165
Running your Model
will close the MzLaunch utility when the simulation finishes.
Analyse the Results
The MSHE_watermovement.exe program automatically generates all of
the output asked for in the Setup Editor. Thus, to look at your output, you
only need to open the model at look at your results in the normal way.
If you want to run the water balance program, which is described in the
Using the Water Balance Tool chapter, you can add the following lines to
you batch file:
MSHE_Wbl_Ex.exe //apv My_WB_areas.WBL
MSHE_Wbl_Post.exe //apv My_WB_areas.WBL 1
MSHE_Wbl_Post.exe //apv My_WB_areas.WBL 2
In the above, the first command runs the Extraction phase of the water balance utility, while the subsequent commands run the Post-processing
items in the water balance file. The number after the water balance file
name indicates which Post-processing item to run. Post-processing steps
cannot be executed before an Extraction step but only one Extraction step
needs to be run for a each water balance utility file.
6.6
OpenMI
OpenMI stands for Open Modelling Interface. OpenMI is a standard,
which facilitates the linking of simulation models and model components
of environmental and socio-economic processes. It thus enables managers
to more fully understand and predict the likely impacts of their policies
and programmes.
The OpenMI Association is the organisation responsible for the development, maintenance, and promotion of OpenMI. DHI active in the OpenMI
Association and was one of the original founding members. On the
OpenMI Association web site at www.openmi.org, you can learn which
models are already OpenMI compliant, get help on OpenMI model migration, request new features, exchange opinions and provide feedback
related to OpenMI implementations.
MIKE SHE is OpenMI compliant. That is, MIKE SHE can be linked to
other OpenMI compliant programs. If you have specific questions on
using MIKE SHE with OpenMI, please contact your local support centre.
166
MIKE SHE
Parallelization of MIKE SHE
Linking MIKE SHE with OpenMI
If you want to link MIKE SHE to another program using OpenMI, then
you will need to initialize MIKE SHE to produce the required OpenMI
linkages. This is done using the Extra Parameter option: Including
OpenMI (V.1 p. 324).
OpenMI limitations of MIKE SHE
The OpenMI GUI has been compiled for "any CPU". So, if you are using
a 64-bit CPU, the OpenMI GUI will act like a 64-bit application - and
expect the OpenMI components to also be 64-bit applications.
When using MIKE SHE on a 64-bit machine and adding a MIKE SHE
model in the OpenMI GUI, an error will be generated. This is a limitation
of the current version of OpenMI.
The workaround is to download the source code of the OpenMI editor,
change the setting from "any CPU" to "x86", recompile and use the new
.exe file instead.
6.7
Parallelization of MIKE SHE
For Release 2011, the MIKE SHE solvers have been parallelized as much
as possible and have been updated for 64-bit operating systems. Also, significant improvements in the memory and calculation efficiency have
been made. However, the scalability of the parallelization is dependent on
the individual modules. Thus, every model will scale differently with
respect to the running time.
The unsaturated module is highly scalable because each UZ column is
completely independent. The saturated zone and overland flow modules,
on-the-other-hand, are not nearly as scalable because of the connections
between the cells. As an approximation, a typical model with a mix of
modules will probably run between 1.8 and 2.5 times faster on a four-core
computer.
The AUTOCAL program has also been updated to take advantage of
multi-core computers. In this case, multiple simulations are sent automatically to each of the cores.
In all cases, the use of additional cores is restricted by the available
licenses. The default number of run-time licenses is limited to four, which
means that the parallelization and AUTOCAL will support up to four
cores. If you want to use more than four cores, then you must contact your
local DHI office for additional run-time licenses.
Running MIKE SHE
167
Running your Model
168
MIKE SHE
SURFACE WATER
169
170
MIKE SHE
Overland Flow
7
SURFACE WATER IN MIKE SHE
Hydrologically, surface water can occur in defined channels or distributed
as ponded water in lakes or on the flood plain. Surface water interacts with
the rest of the hydrologic cycle through evaporation and exchange to/from
groundwater.
In MIKE SHE, ponded surface water can be simulated directly using the
2D overland flow module. The water flow on the ground surface is calculated by MIKE SHE’s Overland Flow Module, using the diffusive wave
approximation of the Saint Venant equations, or using a semi-distributed
approach based on the Mannings equation. This chapter concentrates on
the diffusive wave, finite-difference method.
Historically, MIKE SHE also included its own 1D channel flow module,
but this was replace several years ago by MIKE 11.
This chapter describes the interaction and coupling between MIKE 11 and
MIKE SHE, as well as guidance on the modelling of both channel flow
and flooding in MIKE SHE.
For detailed technical information on 2D surface water flow, see the Overland Flow - Reference (V.2 p. 265) chapter.
For detailed technical information on MIKE 11, refer to either the .pdf
versions of the MIKE 11 Reference Manuals that are installed with MIKE
11, or the MIKE 11 documentation in the on-line help. These can be easily
accessed from your Windows Start menu under MIKE by DHI.
7.1
Overland Flow
When the net rainfall rate exceeds the infiltration capacity of the soil,
water is ponded on the ground surface. This water is available as surface
runoff, to be routed downhill towards the river system. The rate of overland flow is controlled by the surface roughness and the gradient between
cells. The direction of flow is controlled by the gradient of the land surface
- as defined by the topography. The quantity of water available for overland flow is the available ponded depth minus the detention storage, as
well as the losses due to evaporation and infiltration along the flow path.
Overland flow can be a very time consuming part of the simulation. There
are many ways to reduce this burden - often without significantly impacting the accuracy of the results.
Surface Water
171
Surface Water in MIKE SHE
7.1.1
Parameters
The main overland flow parameters are the surface roughness which controls the rate of flow, the depth of detention storage which controls the
amount of water available for flow, plus the intial and boundary conditions.
Surface Roughness
The Stickler roughness coefficient is equivalent to the Manning M. The
Manning M is the inverse of the commonly used Mannings n. The value
of n is typically in the range of 0.01 (smooth channels) to 0.10 (thickly
vegetated channels), which correspond to values of M between 100 and
10, respectively.
If you don’t want to simulate overland flow in an area, a Mannings M of 0
will disable overland flow. However, this will also prevent overland flow
from entering into the cell.
Detention Storage
Detention Storage is used to limit the amount of water that can flow over
the ground surface. The depth of ponded water must exceed the detention
storage before water will flow as sheet flow to the adjacent cell. For example, if the detention storage is set equal to 2mm, then the depth of water on
the surface must exceed 2mm before it will be able to flow as overland
flow. This is equivalent to the trapping of surface water in small ponds or
depressions within a grid cell.
Water trapped in detention storage continues to be available for infiltration
to the unsaturated zone and to evapotranspiration.
Detention storage also affects the exchange with MIKE 11. Only ponded
water in excess of the detention storage will flow to MIKE 11. Also,
flooding from MIKE 11 will only happen when the water level in the river
link is above the topography plus detention storage.
The paved area drainage is also linked to the detention storage. Only the
available ponded water will be routed to the SZ drainage network - that is
the ponded depth above the detention storage. If you want to route all of
the ponded water to the SZ drainage network, then you can use the Extra
Parameter: Paved routing options (V.1 p. 309).
Initial and Boundary Conditions
In most cases it is best to start your simulation with a dry surface and let
the depressions fill up during a run in period. However, if you have significant wetlands or lakes this may not be feasible. However, be aware that
172
MIKE SHE
Overland Flow
stagnant ponded water in wetlands may be a significant source of numerical instabilities or long run times.
The outer boundary condition for overland flow is a specified head, based
on the initial water depth in the outer cells of the model domain. Normally,
the initial depth of water in a model is zero. During the simulation, the
water depth on the boundary can increase and the flow will discharge
across the boundary. However, if a non-zero initial condition is used on
the boundary, then water will flow into the model as long as the internal
water level is lower than the boundary water depth. The boundary will act
as an infinite source of water.
Time varying OL boundary conditions
If you need to specify time varying overland flow boundary conditions,
you can use the Extra Parameter option Time-varying Overland Flow
Boundary Conditions (V.1 p. 302).
7.1.2
Reduced OL leakage to UZ and to/from SZ
The Surface-Subsurface Leakage Coefficient (V.2 p. 121) reduces the
infiltration rate at the ground surface. It works in both directions. That is,
it reduces both the infiltration rate and the seepage outflow rate across the
ground surface.
Conceptually, the leakage coefficient is used to account for soil compaction and fine sediment deposits on flood plains in areas that otherwise
have similar soil profiles.
If the groundwater level is at the ground surface, then the exchange of
water between the surface water and ground water is based on the specified leakage coefficient and the hydraulic head between surface water and
ground water. In other words the UZ model is automatically replaced by a
simple Darcy flow description when the profile becomes completely saturated.
If the groundwater level is below the ground surface, then the vertical
infiltration is determined by the minimum of the moisture dependent
hydraulic conductivity (from the soils database) and the leakage coefficient.
This option is often useful under lakes or on flood plains, which may be
permanently or temporarily flooded, and where fine sediment may have
accumulated creating a low permeable layer (lining) with considerable
flow resistance.
Surface Water
173
Surface Water in MIKE SHE
The value of the leakage coefficient may be found by model calibration,
but a rough estimate can be made based on the saturated hydraulic conductivities of the unsaturated zone or in the low permeable sediment layer,
if such data is available.
The specified leakage coefficient is used wherever it is specified. In areas
where a delete value is specified, the vertical hydraulic conductivity of the
top SZ layer is used.
In the processed data, the item, Surface-subsurface Exchange Grid Code,
is added, where areas with full contact are defined with a 0, and areas with
reduced contact are defined with a 1.
7.1.3
Separated Flow areas
The Separated Flow Areas (V.2 p. 123) are typically used to prevent overland flow from flowing between cells that are separated by topographic
features, such as dikes, that cannot be resolved within a the grid cell.
In many detailed models, surface drainage on flood plains and irrigation
areas is highly controlled. The Seperated Flow Areas option allows you to
define these drainage control land features in the model.
If you define the separated flow areas along the intersection of the inner
and outer boundary areas, MIKE SHE will keep all overland flow inside
of the model - making the boundary a no-flow boundary for overland flow.
However, seperated flow areas are not respected by the other hydrologic
processes, such as the SZ drainage function. Thus, lateral flow out of the
model may still occur via SZ drainage, SZ boundary conditions, MIKE 11,
irrigation control areas, etc. even when the seperated flow areas are
defined. Therefore, if you use seperated flow areas, you should carefully
evaluate your results, for example, by using the water balance tool, to
make sure that the water flow is behaving as you expect.
Also, you should note that Overland flow cannot cross a river link. So, the
cell faces with river links always define a seperated flow boundary.
7.1.4
Paved Area Drainage
In MIKE SHE, there is a paved area function to account for the increased
runoff in urban areas. However, the paved area function is rather complex
and currently two limitations.
The paved area function in MIKE SHE relies on the saturated zone drainage reference system to move drainage to the river links. Thus, an impor-
174
MIKE SHE
Overland Flow
tant limitation is that paved area drainage can only be simulated if you are
using the finite difference SZ model.
The second limitation is that the SZ drainage reference is used for routing
paved drainage to river links, specified MIKE 11 h-points and MOUSE
manholes. Paved drainage is not routed to SZ boundaries or internal SZ
depressions.
The paved area drainage is calculated from ponded water. It is not calculated from rainfall directly. However, the order of operations in the three
UZ solvers (Richards, Gravity, and 2Layer Water Balance) is such that
rainfall is added to existing ponded storage before ET and infiltration are
calculated, and, thus, before the paved drainage is calculated. Therefore,
the paved drainage effectively acts on rainfall.
The amount of paved drainage is calculated based on the available ponded
water. That is, a specified fraction of the amount of available ponded
water is routed directly to the SZ drainage system. The SZ drainage system immediately discharges this water into the MIKE 11 river network.
This is analagous to a full-pipe of water. That is, for any inflow an equal
amount of outflow is generated instantaneously.
Rainfall is added to the ponded depth and then the drainage fraction is
removed. However, if at the end of the time step, there is still ponded
water in the cell, the paved fraction will be applied to the remainder again
in the next time step. Thus, if your cell is half paved and half permanently
ponded, the permanently ponded half will eventually drain away.
The paved area drainage is also linked to the detention storage. Only the
available ponded water will be routed to the SZ drainage network - that is
the ponded depth above the detention storage. If you want to route all of
the ponded water to the SZ drainage network, then you can use an Extra
Parameter found in the Paved routing options (V.1 p. 309).
The SZ drainage parameters, drain level and drain time constant, do not
impact the paved area drainage function. However, the paved area drainage does use the SZ drainage reference system, which means that the SZ
drainage levels may play a role in defining the receiving river link.
The paved area drainage does not check to make sure that you do not create any physically impossible feedback loops. So, flood code cells, and
overbank spilling from MIKE 11 should be directed to cells that where
paved area drainage is active. If this happens, you may encounter excessive feedback between MIKE SHE overland flow and MIKE 11.
Surface Water
175
Surface Water in MIKE SHE
Finally, the paved area function has a time step dependency on the UZ
time step length. If the UZ time step is less than the maximum UZ
timestep length, then the paved area drainage will be reduced. To ensure
that the paved area drainage is as expected, you should adjust your preciptation controls such that the UZ time step is not reduced.
Maximum discharge rate
If a cell is 100% paved, then the paved drainage function will remove all
of the ponded water in the time time step. That is, by default the rate of
drainage is not controlled. This does not reflect the conveyance restrictions of the drainage network. This is addressed by an option for specifying a maximum discharge rate from paved areas. The maximum discharge
rate can be specified as a constant value or a distributed value.
The maximum discharge rate can be used to control the inflow to the SZ
drainage system. At every time step, the available drainage volume will be
checked against the maximum drainage rate. Thus, ponded water will be
retained on a grid cell and drained at a controlled rate into the river system. While the water is ponded it will be subject to infiltration and ET.
The rate of leakage below the cell can be controlled by the Surface-Subsurface Leakage Coefficient (V.2 p. 121).
If the maximum discharge rate is set high, then the paved area fraction can
be used to route a fraction of rainfall directly to the river network. This is
reasonable when the travel time in the drainage network is similar to the
time step length, and losses in the drainage network are minimal.
If the maximum discharge rate is set low, then the paved area fraction can
be used to control inflow to and outflow from, for example, small scale
surface impoundments.
The combination of maximum discharge rate and the OL leakage coefficient, along with the multi-cell OL, allows you to simulate distributed on
grid surface water storages. You can use the combination to define, for
example, distributed farm dams that release water to streams at a fixed
rate. The volume of the storage is defined using the multi-grid OL. The
ponded water is subject to evaporation, and you can use the OL leakage
coefficient to control leakage to groundwater.
Note: When the Multi-cell Overland Flow (V.1 p. 181) option is used, a
uniform value for the maximum discharge rate will be used within each
coarse cell. Further, the depth of ponded water is calculated on the subscale, and used to calculate the paved area flow for each sub-scale cell
176
MIKE SHE
Overland Flow
7.1.5
Overland Flow Velocities
MIKE SHE does not calculate overland flow velocities. The OL velocity
can be calculated based on the water depth and the OL flow in the X- and
Y- directions.
However, the calculation is not straight forward. The x- and y- flow is the
flow across the cell boundary in the positive x and y directions. That is,
the flow across the right and top cell boundaries. The average flow in a
cell is, in fact, the mean of the inflow and the outflow in the x and y directions.
Further complicating the calculation is the fact that the x- and y- flow is
saved for calculating the water balance - not for velocity calculation.
Thus, the flow is the mean-step accumulated flow over the storing time
step. If this were not true then it could not be used for the water balance.
However, the ponded depth is an instantaneous value at the time it is
saved. Thus, the depths and flows are not consistent in time, which has
serious implications for the velocity calculation. For example, if your storing time step is a month, your flows will be a monthly average. The velocity will be saved at midnight on the last day of the month. Depending on
the timing of your events, you could easily have a high average flow and a
zero depth.
In principle, you could compensate for the above limations by using a very
short time storing time step, and averaging the flows across the cell. However, then you have to ask yourself, why you want these flows.
MIKE SHE calculates overland flow based on the diffusive wave approximation, which neglects the momentum. Further, the depth and flow rates
are averages for a cell, which does not take into account the actual distribution of velocities and water depths in a natural topography. Finally, if
the area of interest is next to a river, then the physical exchange with the
river depends on the calculation method used. Even in the best case,
exchange between the river and the flood plain is conceptual. There is no
velocity calculated for the river-OL exchange. Water is simply taken from
the river and put on the flood plain cell, or vice versa. The rate of
exchange depends the water level difference and the weir coefficients
used.
Thus, the calculated velocity is probably not very useful for things like
damage assessment. If velocities are important, then MIKE FLOOD is a
much better tool. MIKE SHE, on the other hand, is good at calculating
overland water depths, general flow directions and the exchange of ponded water with the subsurface and rivers.
Surface Water
177
Surface Water in MIKE SHE
7.2
Overland Flow Performance
The overland flow can be a significant source of numerical instabilities in
MIKE SHE. Depending on the setup, the overland flow time step can
become very short - leading to very long.
Overland flow has both an Implicit and Explicit solver. Your choice of
solver affects both the accuracy of your results and the simulation run
time.
The Implicit solver is faster than the Explicit solver because it can run
with longer time steps. However, the it must iterate to converge on a solution. Thus, if each time step takes several iterations because of the dynamics of the overland flow, then the implicit solver can become slow. The
most obvious sign of poor convergence is the presence of warning messages in the projectname_WM_Print.log file about the overland flow
solver not converging. You may be able to live with a few warning messages, but the if the Implicit solver frequently fails to converge then this
will significantly slow down your simulation. If this happens, then you
have a few options.
The first option is to reduce your OL time step. This make increase the
stability of the solver and actually reduce your run times. You can also
increase the convergence criteria. This will decrease the accuracy, but if
there is a troublesome area outside of your area of interest, then this may
be acceptable.
If you switch to the Explicit solver, then the time step becomes dynamic
depending on the Courant Criteria. This will likely reduce your numerical
instabilities because the courant criteria is very restrictive, but the simulation is likely to be slower. However, the difference may not be that great if
you are having a lot of convergence problems.
7.2.1
Stagnant or slow moving flow
The solution of overland flow is sensitive to the surface water gradient. If
the surface water gradient is very small (or zero), then a numerically stable
solution will generally require a very short time step.
Slow moving flow is a problem when you have long term ponded water,
for example in wetlands. If you are only interested in the water levels in
the wetland areas, but not the flow velocity and flow directions, then solving the overland flow equations is not necessary for decision making.
If you want to turn off the overland flow solver in slow moving or stagnant areas, then you can convert these areas to flood codes and allow
178
MIKE SHE
Overland Flow Performance
MIKE 11 to control the water levels. Lateral overland runoff to these areas
will still be calculated, as will evapotranspiration and infiltration. For
more detail on using Flood Codes, see Overland Flow Exchange with
MIKE 11 (V.1 p. 212).
An alternative is to turn off the overland flow calculation in these cells.
You can turn off the overland flow in a cell by setting the Mannings M
number to zero. However, this also turns off the lateral overland inflow
into the cell, as well.
Another option is to use the detention storage parameter to restrict the
amount of available water. In this case, overland flow is allowed into and
out of the cell, but overland flow is not actually calculated until the depth
of water in the cell exceeds the detention storage.
The Threshold gradient for overland flow (see next Section) is also a way
to reduce the influence of stagnant water on the time step. However, you
cannot specify a spatially varying threshold. So, the appropriate value may
be difficult to select if you want to restrict flow in one area, yet keep surface flow in other less stagnant areas.
If you are using the Explicit OL solver, there are several dfs2 output
options that make it easier to find the model areas that are contributing to
reducing the time step. These include:
7.2.2
z
Mean OL Wave Courant number,
z
Max OL Wave Courant number, and
z
Max Outflow OL-OL per Cell Volume.
Threshold gradient for overland flow
In flat areas with ponded water, the head gradient between grid cells will
be zero or nearly zero. As the head gradient goes to zero, ∆t must also
become very small to maintain accuracy. To allow the simulation to run
with longer time steps and dampen any numerical instabilities in areas
with low lateral gradients, the calculated intercell flows are multiplied by
a damping factor when the gradients are close to zero.
Essentially, the damping factor reduces the flow between cells. You can
think of the damping function as an increased resistance to flow as the gradient goes to zero. In other words, the flow goes to zero faster than the
time steps goes to zero. This makes the solution more stable and allows
for larger time steps. However, the resulting gradients will be artificially
high in the affected cells and the solution will begin to diverge from the
Mannings solution. At very low gradients this is normally insignificant,
Surface Water
179
Surface Water in MIKE SHE
but as the gradient increases the differences may become noticeable.
Therefore, the damping function is only applied when the gradient
between cells is below a user-defined threshold.
The details of the available functions can be found in the section Low gradient damping function (V.2 p. 272) found in the Reference manual.
For both functions and both the explicit and implicit solution methods,
each calculated intercell flow in the current timestep is multiplied by the
local damping factor, FD, to obtain the actual intercell flow. In the explicit
method, the flow used to calculate the courant criteria are also corrected
by FD.
The damping function is controlled by the user-specified threshold gradient (see Common stability parameters (V.2 p. 38) for the Overland Flow),
below which the damping function becomes active.
The choice of appropriate threshold value depends on the slope of the flow
surface. Based on both actual model tests in Florida and synthetic setups,
the following conclusions can be reached:
z
A Threshold gradient greater than the surface slope can lead to excessive OL storage on the surface that takes a long time to drain away.
z
A Threshold gradient equal to the surface slope is often reasonable, but
there may still be some excess storage on the surface.
z
Threshold values less than the surface slope typically cause rapid
drainage and give nearly the same answers.
z
Threshold values below 1e-7 do not significantly improve the results
even if the topography is perfectly flat.
z
In general, you should used the highest value possible. Lower values
may increase accuracy but at the expense of run time.
Therefore, we can safely recommend a Threshold gradient in the range of
1e-4 to 1e-5, with a default value of 1e-4. For many floodplains, 1e-4 or
1e-5 should be sufficient. In flood plains with very flat relief, 1e-6 may be
used. Lower values are probably never necessary.
Since most discharge happens during and immediately after an event, the
Threshold gradient is likely to be most important when there is significant
ponding that lasts over several time steps and drains to a boundary or
MIKE 11. Ponded water that infiltrates or evaporates and experiences limited lateral flow will not be affected by the Threshold value.
180
MIKE SHE
Multi-cell Overland Flow
If the topography slope requires a low Threshold, but the solution is unstable at low threshold values, solution stability may be improved with the
Explicit solver by reducing the Maximum Courant number until the solution becomes stable. With the Implicit solver, you may need to change the
solver parameters.
Performance Impact
A low Threshold gradient will increase your simulation time. So, the final
value that you use, may be a compromise between simulation length and
accuracy of the flow in low gradient conditions.
If you have stagnant ponded water in your model, then the intercell gradient in these areas will be nearly zero. If you lower your Threshold gradient, your simulation performance may be adversely impacted, simply
because the OL solver will begin to calculate flow sloshing back and forth
in these areas. Not only will the OL solver have to work harder, the OL
time step will likely also decrease because of the very low gradients. Thus,
the Threshold gradient effectively reduces intercell flow in stagnant areas
to zero allowing the Courant criteria to be satisfied at much higher time
step lengths. See Figure 11.4 in Threshold gradient for overland flow
(V.1 p. 179) in the Reference manual.
7.3
Multi-cell Overland Flow
The main idea behind the 2D, multi-grid solver is to make the choice of
calculation grid independent of the topographical data resolution. The
approach uses two grids:
z
One describing the rectangular calculation grid, and
z
The other representing the fine bathymetry.
The standard methods used for 2D grid based solvers do not make a distinction between the two. Thus, only one grid is applied and this is typically chosen based on a manageable calculation grid. The available
topography is interpolated to the calculation grid, which typically does not
do justice to the resolution of the available data. The 2D multi-grid solver
in MIKE SHE can, in effect, use the two grids more or less independently.
In the Multi-cell overland flow method, high resolution topography data is
used to modify the flow area used in the St Venant equation and the courant criteria. The method utilizes two grids - a fine-scale topography grid
and a coarser scale overland flow calculation grid. However, both grids
Surface Water
181
Surface Water in MIKE SHE
are calculated from the same reference data - that is the detailed topography digital elevation model.
In the Multi-cell method, the principle assumption is that the volume of
water in the fine grid and the coarse grid is the same. Thus, given a volume of water, a depth and flooded area can be calculated for both the fine
grid and the coarse grid. See Figure 7.1.
In the case of detention storage, the volume of detention storage is calculated based on the user specified depth and OL cell area.
During the simulation, the cross-sectional area available for flow between
the grid cells is an average of the available flow area in each direction
across the cell. This adjusted cross-sectional area is factored into the diffusive wave approximation used in the 2D OL solver. For numerical details
see Multi-cell Overland Flow Method (V.2 p. 275) in the Reference manual.
The multi-grid overland flow solver is typically used where an accurate
bathymetric description is more important than the detailed flow patterns.
This is typically the case for most inland flood studies. In other words, the
distribution of flooding and the area of flooding in an area is more important than the rate and direction of ingress.
Figure 7.1
182
The constant volume from the coarse grid is transfered to the fine
scale grid.
MIKE SHE
Multi-cell Overland Flow
Figure 7.2
Flooded area is a function of the surface water level in the grid cell.
Elevations
The elevation of the coarse grid nodes and the fine grid nodes are calculated based on the input data and the selected interpolation method. However, the coarse grid elevation is adjusted such that it equals the average of
the fine grid nodal elevations. This provides consistency between the
coarse grid and fine grid elevations and storage volumes. Therefore, there
may be slight differences between the cell topography elevations if the
multi-cell method is turned on or off. This could affect your model inputs
and results that depend on the topography. For example, if you initial
water table is defined as a depth to the water table from the topography.
7.3.1
Evaporation
Evaporation is adjusted for the area of ponded water in the coarse grid
cell. That is, evaporation from ponded water is reduced by a ponded area
fraction, calculated by dividing the area of ponding in the fine grid cell by
the total cell area.
The ET from soil evaporation is also reduced to the areas where there is
not ponding.
Total transpiration does not need to be adjusted for the non-ponded area
because evaporation from the ponded area is calculated prior to transpiration, and this is already adjusted for the ponded area in the cell. Thus, if
the cell is fully ponded, then the Reference ET will be satisfied from ponded storage and there will be no transpiration. If the cell is only partially
ponded, then the area fraction of the RefET will be first extracted from the
ponded water and the remainder from the root zone. Since, there is only
one UZ column to extract from, the entire root zone will be available for
transpiration.
The Extra Parameter option,Transpiration during ponding (V.1 p. 309),
allows transpiration from the root zone beneath ponded areas. In this case,
Surface Water
183
Surface Water in MIKE SHE
transpiration is calculated before evaporation from ponding. This option
includes a reduction factor to account for the reduced ET under saturated
conditions. The application of this factor will be changed so that it only
applies to the ponded fraction of the cell.
7.3.2
Infiltration to SZ and UZ with the Multi-Grid OL
If ponded water is flowing between cells, the multi-scale topography will
ensure that only the lowest part of each cell will be flooded, and the rate of
flow between the cells will be adjusted for the flooded depth. However,
the infiltration also needs to be adjusted to account for the fact that there is
a driving pressure head in some parts of the cell.
Infiltration to UZ
The UZ infiltration is calculated based on three UZ calculations: one for
the ponded fraction of the cell, one for the non-ponded fraction of the cell
and finally a calculation is made using the area weighted infiltration of the
two first UZ calculations. The last step is needed as there is only one UZ
column below the multi-cells.
In a MIKE SHE simulation without multi-cell infiltration, the engine calculates an average storage depth, which is available for infiltration. This
storage depth is then used for the infiltration calculations. The storage
depth is calculated as follows:
1 Assuming that the OL depth from the previous OL time step is known,
2 the OL depth is updated using the current net precipitation and any sink
and source terms (irrigation, by pass flow, paved area drainage etc.).
Noting that:
z
Bypass flow is extracted from the net precipitation before the infiltration calculation, and
z
Paved area drainage is also extracted before the infiltration calculation. If you want to calculate the infiltration before paved area
drainage this is available as an Extra parameter:
Parameter
Name
Type
infiltration before Boolean
paved routing
z
184
Value
On
The updated OL depth is used for the infiltration calculation;
MIKE SHE
Multi-cell Overland Flow
z
If the reduced contact option is used, the leakage coefficient is used
to calculate the maximum infiltration rate.
When using the multi-cell infiltration, the infiltration is calculated based
on three cases which depend on the ponded area fraction from the latest
OL time step:
1 Non ponded (Ponded Area Fraction = 0). Only one infiltration calculation based on the available storage depth. This is done in the same way
as a situation without the multi-cell option.
2 Fully ponded (Ponded Area Fraction = 1). Only one infiltration calculation based on the available storage depth. This is done in the same
way as a situation without the multi-cell option.
3 Partly ponded (0 < Ponded Area Fraction < 1). Three infiltration calculations are made; ponded area, non-ponded area and a final calculation using the area weighted storage depth.
In the partly ponded case, it is assumed that the net-precipitation is equally
distributed across the whole cell, while ponding from the previous OL
time step only occurs in the ponded part of the cell. For the infiltration calculation in the non-ponded are, the available water depth is calculated as
DepPrec = precipitation x dt
(7.1)
The remaining part of the available water (ponding + precipitation on the
ponded part) is scaled to an equivalent water depth in the ponded area:
DPonded = (OLDepth + DepPrec) x (1-PAreaFrac) / PAreaFrac
(7.2)
where OLDepth is the depth of ponded water from the previous time step,
and PAreaFrac is the ponded area from the previous time step.
Disabling multi-cell infiltration
Multi-cell infiltration is automatically activated when the multi-cell option
is invoked. However, an Extra Parameter option is available if you want to
disable this function - perhaps for backwards compatability with older
models.
Parameter
Name
Type
disable multi-cell Boolean
infiltration
Surface Water
Value
On
185
Surface Water in MIKE SHE
When this is specified, the infiltration will be calculated based on the values of the course grid, and any ponding occurring in any sub-grid cells
will not be included.
7.3.3
Reduced Leakage with Multi-cell OL
If reduced contact is specified, Surface-Subsurface Leakage Coefficient
(V.2 p. 121), the OL leakage coefficient is used, meaning that:
z
Reduced contact only in ponded areas (activated). Leakage coefficient
is only used in the ponded areas and not in the non-ponded areas.
z
Reduced contact only in ponded areas (not activated). Leakage coefficient is used in both the ponded and the non-ponded case.
The two infiltration calculations (for the ponded and the non-ponded case)
result in two infiltration rates from which an area weighted infiltration
rate, QinfAWghtd, is calculated:
QinfAWghtd = Qinfntpd x (1 - PAreaFrac) + Qinfpd x (PAreaFrac) (7.3)
where Qinfntpd is the infiltration rate from the non-ponded area, and Qinfpd is the infiltration rate for the ponded area. The area weighted infiltration rate is then used in the final UZ calculation (when reduced contact is
not used).
Reduced leakage in ponded areas only
In many cases, the ponded areas will have a lower infiltration rate than the
surrounding dry areas. The land surface in the dry areas will tend to be
broken up macropores etc. Whereas, surface sealing will occur beneath
ponded areas. To yield a more realistic flow surface drainage for flooded
areas, an option for reduced contact (Ol leakage coefficient) in only the
ponded part of the cell is available.
Note This is only used in the UZ infiltration and NOT in the exchange
between SZ and OL.
Activating this option will allow you to include a distributed dfs2 integer
grid code file. The reduced leakage will be applied in all areas with a positive integer value. In all other areas (with a negative, zero or delete
value), the reduced leakage condition will be applied to the whole cell
with the following constraints:
z
186
The option will be applied to the ponded area from the previous time
step. This will ensure that rainfall infiltrates normally in the non-ponded areas and currently ponded water will be retained.
MIKE SHE
Multi-cell Overland Flow
z
After the rainfall in non-ponded areas is infiltrated, then intercell lateral flow will be calculated and a new ponded area determined.
This method ensures that ponded water is able to flow laterally between
cells with limited losses. By adjusting the leakage rate, you can decrease
the losses along the OL flow path. This will essentially lead to a sub-grid
scale drainage network that will ensure that runoff will eventually reach
the river.
However, this option only applies to cells that are ponded, and thereby
ensuring that ponded water remains on the surface. During high intensity
rainfall in the current time step, this option will not encourage the creation
of flooded areas, as the reduced leakage coefficient will first be applied in
the following time step if ponded water is present at the end of the time
step.
On flood plains, where the ponding occurs from overbank spilling from
rivers or streams, the option will likely result in a more realistic description of the flow paths on the flood plain, as it prevents the flooded water
from infiltrating.
7.3.4
Multi-cell Overland Flow + Saturated Zone drainage
The topography is often used to define the SZ drainage network. Thus, a
refined topography more accurately reflects the SZ drainage network.
The SZ drainage function uses a drain level and drain time constant. The
drain level defines the depth at which the water starts to drain. Typically,
this is set to some value below the topography to represent the depth of
surface drainage features below the average topography. This depth
should probably be much smaller if the topography is more finely defined
in the sub-grid model. The drain time constant reflects the density of the
drainage network. If there are a lot of drainage features in a cell then the
time constant is higher and vice versa.
Details related to the use of the Multi-cell OL with the SZ Drainage function are found along with the rest of the user guidance on SZ Drainage in
the section: Saturated Zone drainage + Multi-cell Overland Flow
(V.1 p. 259).
7.3.5
Test example for impact on simulation time
The increased accuracy of the multi-cell overland flow method does not
come for free. There is a performance penalty when you turn on the multicell option. However, the penalty relative to the increased accuracy of
water depths is small.
Surface Water
187
Surface Water in MIKE SHE
A test done on a large, complex model in Florida, USA illustrates the performance penalty of the multi-cell method.
In the model the grid cell size is 457.2 m (1500 ft). However, a high resolution (5-ft) DEM is available for the whole model domain based on
LIDAR data. This makes it attractive to use of higher resolution map with
the Multi-cell option to account more accurately for the OL flow between
1500-ft grid cells.
Figure 7.3
188
Aerial photo of part of the model area.
MIKE SHE
Multi-cell Overland Flow
Surface Water
Figure 7.4
5-foot LIDAR data for part of the model area
Figure 7.5
Interpolation of the 5-foot LIDAR data to the 125ft model grid
189
Surface Water in MIKE SHE
Figure 7.6
Interpolation of the 5-foot LIDAR data to the 1500ft model grid
Impact on simulation time
In this test, we tested multi-cell factors of 1, 2, 3, 4, 6, and 12. The smallest grid size was 125-ft, which is 12 times smaller than the coarse 1500-ft
grid.
The following graphs illustrate the impact of the multi-grid option on the
running times for the test model. Figure 7.7 shows that the OL run time
increases linearly with higher multi-cell factors. In the test model, a multicell factor of 12 caused the OL portion of the simulation time took 30
times longer. Figure 7.8 shows that the multi-cell factor also impacts the
run time for MIKE 11. However, this impact is not linear, with the impact
on MIKE 11 leveling off after a multi-cell factor of four.
The test model run time is dominated by MIKE 11. In this case, the original run time for the OL is not very long and the multi-cell factor increases
the OL run time considerably. However, as a fraction of the total run time,
the OL is still small. When the OL cells are subdivided, there is probably
some significant changes in the lateral inflow to MIKE 11. However, as
the multi-cell factor increases, the increased resolution of the inflows is
not signficant above a factor of about four.
190
MIKE SHE
Multi-cell Overland Flow
Figure 7.7
Increase in OL run time as a function of multi-cell factor
Figure 7.8
Increase in MIKE 11 run time as a function of multi-cell factor.
Impact on model results
The model contains a mix of natural, urban, and agricultural areas. The
model also includes a complex river network with the relevant man-made
canals and structures, and most of the natural flow ways. However, there
is a natural flow way in the southeast part of the model that is not conceptualized in MIKE 11. Since, the surface water flow in this area is relying
only on overland flow, the multi-cell option should significant changes in
the OL flow prediction in this natural flow way.
In urban and agricultural areas, the drainage and the OL flow components
in MIKE SHE route the water into the canal network. The drainage component would keep the water table level below the ground most of the time
in those areas. However, it is of interest to test the OL flows predicted with
the multi-cell option in those areas during storms events.
Surface Water
191
Surface Water in MIKE SHE
7.3.6
Limitations of the Multi-cell Overland Flow Method
In principle, all of the exchange terms in MIKE SHE could be adjusted to
reflect the fine scale water levels and flooded areas. However, some of
these are easier to implement than others and of greater importance. Thus,
in current release, the exchange with MIKE 11, as well as UZ and SZ,
depend only on the coarse scale grid elevations.
Overland flow exchange with MIKE 11
Overland flow exchange with MIKE 11 does not consider the multi-cell
method. That is, flow into and out of the River Links is controlled by the
water level calculated from the elevation defined in the coarse grid cell.
Likewise the flow area for exchange with MIKE 11 is calculated as the
coarse water depth times the overall grid size. Also, the elevation used
when calculating flood inundation with flood codes only considers the
average cell depth of the coarse grid. See Overland Flow Exchange with
MIKE 11 (V.1 p. 212).
However, if you choose to modify the topography based on a bathymetry
file, or the MIKE 11cross-sections, then this information will be used
when calculating the multi-cell elevations. See Inundation options by
Flood Code (V.1 p. 222).
7.3.7
Setting up and evaluating the multi-grid OL
The multi-cell overland flow method is activated in the OL Computational
Control Parameters (V.2 p. 36) dialogue. In this dialogue, you can check
on the option and then specify a sub-division factor. The coarse grid will
be divided into this number of cells in both directions. That is, for a factor
of two, the coarse grid will be divided into four cells. Likewise a factor of
five will lead to 25 fine cells per coarse grid cell.
In addition
Pre-processed data
When you enable the Multi-grid OL option the following new items will
be available in the preprocessed data:
z
192
Multi-Cell ground levels (subscale topography) In Figure 7.9, two
interpolations of the same DEM are shown. The top figure is a plan
view of the interpolation to a 100m grid resolution and a 25m grid resolution respectively. The bottom figure is a cross-section across the
middle of the top figure, where you can clearly see the more accurate
resolution of the drainage features.
MIKE SHE
Multi-cell Overland Flow
Figure 7.9
Example of preprocessed data - topography using a 100 meter resolution and a sub-scale factor of 4 (25 m sub-scale resolution).
When SZ Drainage is also active, then the following items are also available:
z
Surface Water
Max. MC Drain Level - displays the maximum drain level of the subcells within each of the model cells
193
Surface Water in MIKE SHE
z
Min MC Drain level - displays the minimum drain level of the subcells within each of the model cells
z
Min MC Drain Depth - displays the minimum drain depth of the subcells within each of the model cells
z
Max MC Drain Depth - displays the maximum drain depth of the subcells within each of the model cells
Additional results options
When the Multi-grid OL option is active, the following additional items
will be available in the result items:
z
Depth of Multi-Cell overland water - displays the depth of overland
water using the sub-scale resolution
z
Multi-Cell overland water elevation - displays the overland water elevation using the sub-scale resolution
7.4
Channel Flow
7.4.1
MIKE 11 Overview
MIKE 11 is a comprehensive 1D channel flow model for simulating rivers
and surface water bodies that can be approximated as 1-dimensional flow
(as strict 1-Dimensional flow does not occur in nature). Basically, MIKE
11 can be applied anywhere average values of levels, velocities, concentrations etc. at a point are acceptable, including:
194
z
River hydrodynamics
z
Structure/reservoir operational control
z
Water quality (e.g. wetlands, salinity)
z
Sediment transport & morphology
z
Flood studies (e.g. mapping, hazard assessment)
z
Flood forecasting (on-line, real-time)
z
Dam break
z
Sediment transport (e.g. Long term morphology)
z
River restoration
z
Integrated with groundwater and flooding
MIKE SHE
Building a MIKE 11 model
MIKE 11 plays a critical role in MIKE SHE Both the overland flow and
groundwater flow modules are linked directly to MIKE 11. The MIKE
SHE-MIKE 11 coupling enables:
7.5
z
the one-dimensional simulation of river flows and water levels using
the fully dynamic Saint Venant equations.
z
the simulation of a wide range of hydraulic control structures, such as
weirs, gates and culverts.
z
area-inundation modelling, using a simple flood-mapping procedure
that is based on simulated river water levels and a digital terrain model.
z
dynamic overland flooding flow to and from the MIKE 11 river network.
z
the full, dynamic coupling of surface and sub-surface flow processes in
MIKE 11 and MIKE SHE.
Building a MIKE 11 model
Integrating a MIKE SHE and a MIKE 11 model is not very different from
establishing a stand-alone MIKE 11 HD model and a stand-alone MIKE
SHE model. In principle, there are three basic set-up steps:
1 Build a stand-alone MIKE 11 HD hydraulic model and make a performance test and, if possible, a rough calibration using prescribed
inflow and stage boundaries. If needed, you can specify a default
groundwater table (e.g. MIKE SHE’s initial groundwater level) and
leakage coefficients for any leakage calculations.
2 Build a stand-alone MIKE SHE model that includes the overland flow
component and (optionally) the saturated zone and unsaturated zone
components. An SZ drainage boundary can be used to prevent excessive surface flows in low lying areas and the river flood plain.
3 Couple MIKE SHE and MIKE 11 by defining branches (reaches)
where MIKE 11 HD should interact with MIKE SHE. Modify your
MIKE SHE and MIKE 11 models so that they work together properly.
For example, by removing the specified groundwater table in MIKE 11
and adjusting your SZ drainage elevations if you used these in Step 2.
In the above scheme, the first step in coupling MIKE 11 to MIKE SHE is
to create a normal MIKE 11 HD model without coupling it with MIKE
SHE. In this regard, a few things should be emphasised:
Surface Water
195
Surface Water in MIKE SHE
7.5.1
z
In a normal MIKE 11 river model only the river chainage (dx) is
important for the results. Geographic positioning of river branches and
cross-sections are only important for the graphical presentation. When
interfacing MIKE 11 to MIKE SHE geographic positioning is critical,
as MIKE SHE needs information on the river location.
z
A reasonably high number of river cross-sections should be included to
ensure that the river elevations are reasonably consistent with the surface topographic features.
MIKE 11 network limitations
There are a few features of MIKE 11 that do not relate well to MIKE SHE.
Short branches
In MIKE 11 there is no restriction on how short your branches are. If you
are trying to simulate discontuous lakes or structures on the flood plain,
for example. you may have very short branches. However, MIKE SHE
does not allow MIKE 11 branches to be shorter than the cell size. Generally, though, short branches are a sign that you should probably reconsider
your model conceptualization - or switch to MIKE FLOOD, which allows
flood plain structures.
Parallel branches
Like short branches, MIKE SHE does not like it when your branches are
too close together. If you have parallel branches that are too close together,
then the branches may be mapped to the same river link. However, each
river link must be mapped to a unique branch. As a rule of thumb, parallel
branches should be greater than a cell width apart. However, this is not
uniformly true, since the two close parallel branches may map onto opposite sides of a cell, if they are located on either side of a cell mid-point.
Thus, you may have unexpected problems, if you change the cell size in a
model that was working and you have branches that are closer together
than one cell size.
Long coupling links
MIKE SHE links to MIKE 11 branches. However, when two branches are
connected, water is passed between the branches directly. The link has not
physical length or storage itself. If your links are too long, there will be an
error in the timing of the flows between the two branches. So, the links
should be kept short. MIKE 11 does not have any restrictions on how long
the links can be, but MIKE SHE will issue a warning if the links are
longer than a cell size. The warning is simply to informing you that there
is no possibility for groundwater-surface water exchange in the link.
196
MIKE SHE
Building a MIKE 11 model
Long distances between cross sections
MIKE 11 controls the distance between the calculation nodes. The properties at the calculation nodes are linearly interpolated from the available
cross-sections. This includes geometric properties such as bank and bottom elevations, marker locations, etc. However, linear interpolation can
easily result in inconsistences between elevations in MIKE SHE.and
marker elevations in MIKE 11. If the bank elevation is higher than the
topography, then overland flow into the river will be restricted. If the
downstream river bottom elevation is higher than the side branch bottom
elevation, then MIKE 11 will likely be unstable.
Long distances between calculation nodes
This is not the same as long distances between cross-sections. MIKE 11
manages the water at the q-points directly linked to the river links. MIKE
SHE and the river link system automatically interpolates the nearest river
link. However, if the calculation nodes are very far apart or very close
together, then the linear interpolation of water volumes between the calculation points may lead to discrepancies in the available water volumes
especially if the river links are being used for irrigation or the river is losing water. In this sense, the distance between the calculation nodes, should
be similar to the MIKE SHE grid spacing.
7.5.2
MIKE 11 Cross-sections
Whenever there is a significant change in the bed slope there should, in
principle, be a cross-section defined in MIKE 11. If only a few cross-sections are available, it may be sufficient to estimate the cross-section shape
based on neighbouring cross-sections and estimate the bank/bed elevation
based on the surface topographic information in MIKE SHE or other topographic maps.
Cross-sections vs. time step
However, every cross-section in MIKE 11 is a calculation node. The time
step in MIKE 11 is sensitive to the Courant number, which is proportional
to the distance between calculation nodes. So, if the cross-sections are
close together, then you may experience very short time steps in MIKE 11.
Thus, if you are have very short MIKE 11 time steps, then you might want
to check your river network to make sure you do not have cross-sections
that are too close together. This frequently occurs when the cross-sections
have been imported. If you do have cross-sections that are too close
together, then you can easily eliminate one or more of them, as long as the
conveyance of the different cross-sections is roughly the same. In other
words, you can eliminate duplicate cross-sections if their Q/H relationships are roughly the same, even though the physical shape of the two
Surface Water
197
Surface Water in MIKE SHE
cross-sections may appear quite different. This is often the case in braided
stream networks, where the location of the main channels may move left
or right, but the overall conveyance of the river bed is relatively constant.
Cross-sections versus MIKE SHE topography
In the absence of flooding, ponded water discharges to the MIKE 11 river
as overland flow. As a general rule, the topography must be higher than or
equal to the bank elevation. If the bank elevation is higher than the topography, water will not be able to flow into the river in that cell, but will run
laterally along the river until it reaches a place for it to flow into the river.
An easy trick to see where this is happening is to run a simulation with no
infiltration, ET, or detention storage and set the initial water depth at 1m.
Then look at the results to find places were the water is piling up against
the river links.
In the pre-processor log file, a table is create that lists all the river links
where the bank elevation is different than the topography.of the adjacent
cell. The critical river links with bank elevations above the topography are
highlighted with the ==> symbol. This list can be surprisingly long
because the river link bank elevations are interpolated from the neighbouring cross-sections. Whereas the topography is already defined. So, frequently the interpolated bank elevations do not line up precisely with the
topography.
If overland flow on the flood plain is essentially absent, for example, due
to infiltration or evapotranspiration, then these differences are not relevant
and there is no need to modify the topography. However, if the overland to
river exchange is important then you may have to carefully modify your
topography file or your bank elevations so that they are consistent.
Hint In many cases, your topography is from a DEM that is different from
your model grid - either because it is a .shp or xyz file, or if it is a different
resolution than your model grid. In this case, it may be easier to save the
pre-processed topography to a dfs2 file (right click on the topography map
in the pre-processed tab). Then modify and use the new dfs2 file as the
topography in your model setup. The disadvantage of this, is that if you
change your model domain or grid, then you will have to redo your topography modifications.
Hint You can also use one of the Flood code options to automatically
modify your topography, if you have wide cross-sections or a detailed
DEM of the floodplain. In this case, after you have set up your MIKE 11
model, you can specify a constant grid code for the whole model and let
MIKE SHE calculate a modified topography based on the cross-sections
198
MIKE SHE
Coupling of MIKE SHE and MIKE 11
Main Branch
Connection
Tributary
MIKE SHE River Links
Figure 7.10
MIKE 11 H-Points
MIKE 11 Branches and H-points in a MIKE SHE Grid with River
Links
or bathymetry. Then save the topography file as above and then use it as
the model topography.
7.6
Coupling of MIKE SHE and MIKE 11
The coupling between MIKE 11 and MIKE SHE is made via river links,
which are located on the edges that separate adjacent grid cells. The river
link network is created by MIKE SHE’s set-up program, based on a userspecified sub-set of the MIKE 11 river model, called the coupling
reaches. The entire river system is always included in the hydraulic
model, but MIKE SHE will only exchange water with the coupling
reaches. Figure 7.10 shows part of a MIKE SHE model grid with the
MIKE SHE river links, the corresponding MIKE 11 coupling reaches, and
the MIKE 11 H-points (points where MIKE 11 calculates the water levels).
Surface Water
199
Surface Water in MIKE SHE
The location of each of MIKE SHE river link is determined from the coordinates of the MIKE 11 river points, where the river points include both
digitised points and H-points on the specified coupling reaches. Since the
MIKE SHE river links are located on the edges between grid cells, the
details of the MIKE 11 river geometry can be only partly included in
MIKE SHE, depending on the MIKE SHE grid size. The more refined the
MIKE SHE grid, the more accurately the river network can be reproduced.
If flooding is not allowed, the MIKE 11 river levels at the H-points are
interpolated to the MIKE SHE river links, where the exchange flows from
overland flow and the saturated zone are calculated.
If flooding is allowed, via Flood Codes, then the water levels at the MIKE
11 H-points are interpolated to specified MIKE SHE grid cells to determine if ponded water exists on the cell surface. If ponded water exists,
then the unsaturated or saturated exchange flows are calculated based on
the ponded water level above the cell.
If flooding is allowed via overbank spilling, then the river water is allowed
to spill onto the MIKE SHE model as overland flow.
In each case, the calculated exchange flows are fed back to MIKE 11 as
lateral inflow or outflow.
Each MIKE SHE river link can only be associated with one coupling
reach, which restricts the coupling reaches from being too close together.
This can lead to problems when you have a detailed drainage or river network with branches less than one half a cell width apart. It will also lead to
problems if your MIKE 11 branches are shorter than your MIKE SHE cell
size.
If you have coupling reaches that are too short or too close together, you
will receive an error message. If this happens, you can
z
decide not to include one of the branches as a coupling reach (it is still
included in the MIKE 11 HD model), or
z
remove some of the branches (this error often occurs when you have a
detailed looped drainage network), or
z
refine your MIKE SHE grid until all coupling reaches are assigned to
unique river links.
If you have a regional model with large cells (say 1-2km wide), then you
cannot expect the river-aquifer interaction to be accurate at the individual
cell level (e.g. all your cell properties – topography, conductivity, Man-
200
MIKE SHE
Coupling of MIKE SHE and MIKE 11
ningsM, etc. – are all average values over 1-4 km2). Rather, most often
you will be interested in having a correct overall water balance along the
stream. Typically, this is achieved by calibrating a uniform average river
bed leakage coefficient against a measured outflow hydrograph. In such a
model, you may also be tolerant of higher groundwater residuals.
On the other hand, if you need more detailed site specific results (and you
have data and measurements to calibrate against), then you will use a local
scale model, with a smaller grid (say 50-200m) and discrepancies between
topography and river bank elevation will largely disappear. In this case,
you will be more likely to be able to make accurate local scale predictions
of groundwater-surface water exchange.
7.6.1
MIKE SHE Branches vs. MIKE 11 Branches
A MIKE 11 branch is a continuous river segment defined in MIKE 11. A
MIKE 11 branch can be sub-divided into several coupling reaches.
A MIKE SHE branch is an unbroken series of coupling reaches of one
MIKE 11 branch.
One reason for dividing a MIKE 11 branch into several coupling reaches
could be to define different riverbed leakage coefficients for different sections of the river.
If there are gaps between the specified coupling reaches, the sub-division
will result in more than one MIKE SHE branch. Gaps of this type are not
important to the calculation of the exchange flows between the hydrologic
components (e.g. overland to river, or SZ to river). The exchange flows
depend on the water level in the MIKE 11 river, which is unaffected by
gaps in the coupling reaches.
However, MIKE SHE can calculate how much of the water in the river is
from the various hydrologic sources (e.g. fraction from overland flow and
SZ exfiltration). However, this sort of calculation is only possible if the
MIKE SHE branch is continuous. If there is a gap in a MIKE SHE branch,
then the calculated contributions from the different hydrologic sources
downstream of the gap will be incorrect. If there are gaps in the MIKE
SHE branch network, then the correct contributions from the different
sources must be determined from the MIKE 11 output directly.
Furthermore, the MIKE 11/MIKE SHE coupling for the water quality
(AD) module will not work correctly if there are gaps in the MIKE SHE
branch network.
Surface Water
201
Surface Water in MIKE SHE
There is one further limitation in MIKE SHE. That is, no coupling branch
can be located entirely within one grid cell. This limitation is to prevent
multiple coupling branches being located within a single grid cell.
Connections Between Tributaries and the Main Branch
Likewise, the connections between the tributaries and the main branch are
only important for correctly calculating the downstream hydrologic contributions to the river flow and in the advection-dispersion (AD) simulations. The connections are not important to the calculation of the exchange
flows between the hydrologic components (e.g. overland to river, or SZ to
river).
In the example shown in Figure 7.10, the river links of the tributary are
correctly connected to the main branch. This will happen automatically
when
z
the hydraulic connection is defined in the MIKE 11 network, AND
z
the connection point (the chainage) on the main branch is included in a
coupling reach, AND
z
the connection point (the chainage) on the tributary is included in a
coupling reach.
If the connection does not satisfy the above criteria, then there may be a
gap in the MIKE SHE branch network and the limitations outlined above
will apply.
7.6.2
The River-Link Cross-section
The MIKE 11(HD) hydraulic model uses the precise cross-sections, as
defined in the MIKE 11 .xns11 (cross-section) file, for calculating the river
water levels and the river volumes. However, the exchange of water
between MIKE 11 and MIKE SHE is calculated based the river-link crosssection.
The river-link uses a simplified, triangular cross-section interpolated (distance weighted) from the two nearest MIKE 11 cross-sections. The top
width is equal to the distance between the cross-section’s left and right
bank markers. The elevation of the bottom of the triangle equals the lowest depth of the MIKE 11 cross-section (the elevation of Marker 2 in the
cross-section). The left and right bank elevations in MIKE 11 (cross-section markers 1 and 3 in MIKE 11) are used to define the left and right bank
elevations of the river link (See Figure 7.11).
202
MIKE SHE
Coupling of MIKE SHE and MIKE 11
Figure 7.11
A typical simplified MIKE SHE river link cross-section compared to
the equivalent MIKE 11 cross-section.
If the MIKE 11 cross-section is wider than the MIKE SHE cell size, then
the river-link cross-section is reduced to the cell width.This is a very
important limitation, as it embodies the assumption that the river is narrower than the MIKE SHE cell width. If your river is wider than a cell
width, and you want to simulate water on the flood plain, then you will
need to use either the Flooding from MIKE 11 to MIKE SHE using Flood
Codes (V.1 p. 214) option or the Direct Overbank Spilling to and from
MIKE 11 (V.1 p. 216) option.
If you don’t want to simulate flooding, then the reduction of the river link
width to the cell width will not likely cause a problem, as MIKE SHE
assumes that the primary exchange between the river and the aquifer takes
place through the river banks. For more detail on the river aquifer
exchange see Groundwater Exchange with MIKE 11 (V.1 p. 207).
For more detail on flooding and overland exchange with MIKE 11 see
Overland Flow Exchange with MIKE 11 (V.1 p. 212)
7.6.3
Connecting MIKE 11 Water Levels and Flows to MIKE SHE
In MIKE 11, every node in the river network requires information on the
river hydraulics, such as cross-section and roughness factors. These nodes
are known as H-points, and MIKE 11 calculates the water level at every
H-point (node) in the river network. Halfway between each H-point is a
Surface Water
203
Surface Water in MIKE SHE
Storing Q-point, where MIKE 11 calculates the flow, which must be constant between the H-points.
The water levels at the MIKE 11 H-points are transferred to the MIKE
SHE river links using a 2-point interpolation scheme. That is, the water
level in each river link is interpolated from the two nearest H-points
(upstream and downstream), calculated from the centre of the link. The
interpolation is proportionally distance-weighted.
The volume of water stored in a river link is based on a sharing of the
water in the nearest H-points. In Figure 7.12, River Link A includes all the
water volume from H.points 1 and 2, plus part of the volume associated
with H-point 3. The volume in River Link B is only related to the volume
in H-point 3. While the volume in River Link C includes water from Hpoints 3 and 4. This is done to ensure consistency between the river volumes in MIKE 11 and MIKE SHE, as the amount of water that can infiltrate or be transfered to overland flow is limited by the amount of water
stored in the river link.
Figure 7.12
Sharing of MIKE 11 H.point volumes with MIKE SHE river links.
The water levels and flows at all MIKE 11 H-points located within the
coupling reaches can be retrieved from the MIKE SHE result file.
However, since the MIKE 11 flows are not used by MIKE SHE, the river
flows stored in the MIKE SHE result file are not the flows calculated at
the MIKE 11 Storing Q-points. Rather, the flows stored in the MIKE SHE
result file are the estimated flows at the MIKE 11 H-points. That is, the
flows in the MIKE SHE result file have been linearly interpolated from
the calculated flows at the Storing Q-point locations to the H-point locations on either side of the Storing Q-point. If the exact Q-point discharges
are needed, they must be retrieved or plotted directly from the MIKE 11
result file.
204
MIKE SHE
Coupling of MIKE SHE and MIKE 11
7.6.4
Evaluating your river links
The river links are evaluated during the pre-processing. In the pre-processor log file (yourprojectnamePP_print.log), there is a table that contains
all of the river link details:
In this table, the locations where the river links are higher than the topography are marked in the outside left column.
The reference system used in the table is illustrated below:
The explanation of the columns is:
Link: River Link ID number. ID starts at 1 and increases by 1.
IX,IY: coordinate of one end of the link segment. They are referred to the
preprocessed grid such that (IX,IY)=(1,1) at the left-bottom corner of the
model grid. The link segment can be drawn starting from (IX,IY) coordinate, and then following east direction if Side="S" or following the north
direction if Side="W".
Surface Water
205
Surface Water in MIKE SHE
Side: relative position of the (IX1,IY1) cell with respect to the link segment. "S" stands for south and "W" for west.
IX1,IY1: coordinate of the cell on the south side of the link if Side="S", or
the cell on the west side of the link if Side= "W". The left-bottom corner
cell of the model grid has coordinates (IX1,IY1)= (1,1).
Topo1: Pre-processed Topo elevation (in meters) of the cell (IX1,IY1).
Bank1: Interpolated cross section bank elevation (in meters) at marker 1
or 3 at the link chainage (last column). The marker (1 or 3) corresponding
to Bank1 depends on the position of the cell (IX1,IY1) with respect to the
direction of increasing chainage. Marker 1 is the left marker in the increasing chainage direction.
IX2,IY2: Coordinate of the cell on the opposite side to (IX1,IY1). In other
words, it is the cell on the north side of the link if Side="S", or the cell on
the east side of the link if Side= "W". The left-bottom corner cell of the
model grid has coordinates (IX2,IY2)= (1,1).
Topo2: Pre-processed Topo elevation (in meters) of the cell (IX2,IY2).
Bank2: Interpolated cross section bank elevation (in meters) at marker 1
or 3 at the link chainage (last column). The marker (1 or 3) corresponding
to Bank2 depends on the position of the cell (IX2,IY2) with respect to the
direction of increasing chainage. Marker 1 is the left marker in the increasing chainage direction.
Bed: Interpolated cross section elevation (in meters) at marker 2 at the
link chainage (last column). In other words, it is the river bed bottom elevation interpolated at that chainage.
Width: Interpolated cross section width (in meters) at the link chainage
(last column). The cross section width is the distance between markers 1
and 3 in the cross section profile.
Leak-opt: The Conductance option used in the coupling reach in which
this river link is contained. The value is from in the MIKE SHE links table
of the MIKE 11 Coupling Reaches dialogue. The three possible options
are "Aq+Bed", "Aq only", and "Bed only". See Groundwater Exchange
with MIKE 11 (V.1 p. 207) and Figure 7.13.
Leak-coeff: The Leakage Coef. value used in the coupling reach in which
this river link is contained found in the MIKE SHE links table of "Coupling Reaches". See Groundwater Exchange with MIKE 11 (V.1 p. 207)
and Figure 7.13.
206
MIKE SHE
Coupling of MIKE SHE and MIKE 11
Spill: Indicates whether the Allow overbank spilling option is checked for
the coupling reach in which the river link is contained. The two possible
values are "On" and "Off". See Figure 7.13.
WeirCoeff: The Weir coefficient value used in the coupling reach in
which the river link is contained. See Figure 7.13.
HExpo: The Head exponent value used in the coupling reach in which the
river link is contained. See Figure 7.13.
FullWdepth: The Minimum upstream height above bank for full weir
width value used in the coupling reach in which the river link is contained.
See Figure 7.13.
ThrVolSpill: Threshold volume value in cubic meters, which is the product between the Minimum flow are for overbank spilling value (for the
coupling reach in which this river link is contained. See Figure 7.13) and
the MIKE SHE cell size.
Chainage: Chainage of the MIKE 11 network that corresponds to the
center of the link segment. They are sorted from highest to lowest chainage values for the same branch.
Branch: Name of the MIKE 11 Branch. Branches are sorted alphabetically.
7.6.5
Groundwater Exchange with MIKE 11
The exchange flow, Q, between a saturated zone grid cell and the river
link is calculated as a conductance, C, multiplied by the head difference
between the river and the grid cell.
Q = C ⋅ ∆h
(7.4)
Note that Eq. (7.4) is calculated twice - once for each cell on either side of
the river link. This allows for different flow to either side of the river if
there is a groundwater head gradient across the river, or if the aquifer
properties are different.
Referring to Figure 7.11, the head difference between a grid cell and the
river is calculated as
∆h = h grid – h riv
Surface Water
(7.5)
207
Surface Water in MIKE SHE
where hgrid is the head in the grid cell and hriv is the head in the river link,
as interpolated from the MIKE 11 H-points.
If the ground water level drops below the river bed elevation, the head difference is calculated as
∆h = z bot – h riv
(7.6)
where zbot is the bottom of the simplified river link cross section, which is
equal to the lowest point in the MIKE 11 cross-section.
In Eq. (7.4), the conductance, C, between the cell and the river link can
depend on
z
the conductivity of the aquifer material only. See Aquifer Only Conductance (V.1 p. 208), or
z
the conductivity of the river bed material only. See River bed only conductance (V.1 p. 209), or
z
the conductivity of both the river bed and the aquifer material. See
Both aquifer and river bed conductance (V.1 p. 210).
Aquifer Only Conductance
When the river is in full contact with the aquifer material, it is assumed
that there is no low permeable lining of the river bed. The only head loss
between the river and the grid node is that created by the flow from the
grid node to the river itself. This is typical of gaining streams, or streams
that are fast moving.
Thus, referring to Figure 7.11, the conductance, C, between the grid node
and the river link is given by
K ⋅ da ⋅ dx
C = ------------------------ds
(7.7)
where K is the horizontal hydraulic conductivity in the grid cell, da is the
vertical surface available for exchange flow, dx is the grid size used in the
SZ component, and ds is the average flow length. The average flow
length, ds, is the distance from the grid node to the middle of the river
bank in the triangular, river-link cross-section. ds is limited to between 1/2
and 1/4 of a cell width, since the maximum river-link width is one cell
width (half cell width per side).
There are three variations for calculating da:
208
MIKE SHE
Coupling of MIKE SHE and MIKE 11
z
If the water table is higher than the river water level, da is the saturated
aquifer thickness above the bottom of the river bed. Note, however,
that da is not limited by the bank elevation of the river cross-section,
which means that if the water table in the cell is above the bank of the
river, da accounts for overland seepage above the bank of the river.
z
If the water table is below the river level, then da is the depth of water
in the river.
z
If the river cross-section crosses multiple model layers, then da (and
therefore C) is limited by the available saturated thickness in each
layer. The exchange with each layer is calculated independently, based
on the da calculated for each layer. This makes the total exchange independent of the number of layers the river intersects.
This formulation for da assumes that the river-aquifer exchange is primarily via the river banks, which is consistent with the limitation that there is
no unsaturated flow calculated beneath the river.
River bed only conductance
If there is a river bed lining, then there will be a head loss across the lining. In this case, the conductance is a function of both the aquifer conductivity and the conductivity of the river bed. However, when the head loss
across the river bed is much greater than the head loss in the aquifer material, then the head loss in the aquifer can be ignored (e.g. if the bed material is thick and very fine and the aquifer material is coarse). This is the
assumption used in many groundwater models, such as MODFLOW.
In this case, referring to Figure 7.11, the conductance, C, between the grid
node and the river link is given by
C = L c ⋅ w ⋅ dx
(7.8)
where dx is the grid size used in the SZ component, Lc is the leakage coefficient [1/T] of the bed material, and w is the wetted perimeter of the
cross-section.
In Eq. (7.8), the wetted perimeter, w, is assumed to be equal to the sum of
the vertical and horizontal areas available for exchange flow. From
Figure 7.11, this is equal to da + lh, respectively. The horizontal infiltration
length, lh, is calculated based on the depth of water in the river and the
geometry of the triangular river-link cross-section.
The infiltration area of the river link closely approximates the infiltration
area of natural channels when the river is well connected to the aquifer. In
Surface Water
209
Surface Water in MIKE SHE
this case, the majority of the groundwater-surface water exchange occurs
through the banks of the river and decreases to zero towards the centre of
the river. However, for losing streams separated from the groundwater
table by an unsaturated zone, the majority of the infiltration occurs vertically and not through the river banks. In this case, the triangular shape of
the river link does not really approximate wide losing streams.and the calculated infiltration area may be too small - especially if the MIKE 11 bank
elevations are much higher than the river level. This can be compensated
for by either choosing a lower bank elevation or by increasing the leakage
coefficient.
There are three variations for calculating da:
z
If the water table is higher than the river water level, da is the saturated
aquifer thickness above the bottom of the river bed. Note, however,
that da is not limited by the bank elevation of the river cross-section,
which means that if the water table in the cell is above the bank of the
river, da accounts for overland seepage above the bank of the river.
z
If the water table is below the river level, then da is the depth of water
in the river.
z
If the river cross-section crosses multiple model layers, then da (and
therefore C) is limited by the available saturated thickness in each
layer. The exchange with each layer is calculated independently, based
on the da calculated for each layer. This makes the total exchange independent of the number of layers the river intersects.
This formulation for da assumes that the river-aquifer exchange is primarily via the river banks, which is consistent with the limitation that there is
no unsaturated flow calculated beneath the river.
Both aquifer and river bed conductance
If there is a river bed lining, then there will be a head loss across the lining. In this case, the conductance is a function of both the aquifer conductivity and the conductivity of the river bed and can be calculated as a serial
connection of the individual conductances. Thus, referring to Figure 7.11,
the conductance, C, between the grid node and the river link is given by
1
C = -------------------------------------------------------ds
1
------------------------- + -----------------------K ⋅ da ⋅ dx L c ⋅ w ⋅ dx
(7.9)
where K is the horizontal hydraulic conductivity in the grid cell, da is the
vertical surface available for exchange flow, dx is the grid size used in the
210
MIKE SHE
Coupling of MIKE SHE and MIKE 11
SZ component, ds is the average flow length, Lc is the leakage coefficient
[1/T] of the bed material, and w is the wetted perimeter of the cross-section. The average flow length, ds, is the distance from the grid node to the
middle of the river bank in the triangular, river-link cross-section. ds is
limited to between 1/2 and 1/4 of a cell width, since the maximum riverlink width is one cell width (half cell width per side).
In Eq. (7.8), the wetted perimeter, w, is assumed to be equal to the sum of
the vertical and horizontal areas available for exchange flow. From
Figure 7.11, this is equal to da + lh, respectively. The horizontal infiltration
length, lh, is calculated based on the depth of water in the river and the
geometry of the triangular river-link cross-section.
The infiltration area of the river link closely approximates the infiltration
area of natural channels when the river is well connected to the aquifer. In
this case, the majority of the groundwater-surface water exchange occurs
through the banks of the river and decreases to zero towards the centre of
the river. However, in the case of losing streams separated from the
groundwater table by an unsaturated zone, the majority of the infiltration
occurs vertically and not through the river banks. In this case, the horizontal infiltration area may be too small, if the MIKE 11 bank elevations are
much higher than the river level. This can be compensated for by either
choosing a lower bank elevation or by increasing the leakage coefficient.
There are three variations for calculating da:
z
If the water table is higher than the river water level, da is the saturated
aquifer thickness above the bottom of the river bed. Note, however,
that da is not limited by the bank elevation of the river cross-section,
which means that if the water table in the cell is above the bank of the
river, da accounts for overland seepage above the bank of the river.
z
If the water table is below the river level, then da is the depth of water
in the river.
z
If the river cross-section crosses multiple model layers, then da (and
therefore C) is limited by the available saturated thickness in each
layer. The exchange with each layer is calculated independently, based
on the da calculated for each layer. This makes the total exchange independent of the number of layers the river intersects.
This formulation for da assumes that the river-aquifer exchange is primarily via the river banks, which is consistent with the limitation that there is
no unsaturated flow calculated beneath the river.
Surface Water
211
Surface Water in MIKE SHE
7.6.6
Steady-state groundwater simulations
For steady-state groundwater models, MIKE 11 is not actually run. Rather
the initial water level in MIKE 11 is used for calculating da in the conductance formulas and hriv for the head gradient.
To improve numerical stability during steady-state groundwater simulations, the actual conductance used in the current iteration is an average of
the currently calculated conductance and the conductance used in the previous iteration.
Canyon option for steady-state groundwater simulations
In the case of a deep, narrow channel crossing multiple model layers, the
head difference used in Equations (7.4) and (7.5) can optionally be limited
by the bottom elevation of the layer. Thus,
∆h = h grid – max ( h riv, z )
(7.10)
where z is the bottom of the current layer.
The above formulation reduces the infiltration from upper layers by reducing the available gradient. Without the ‘Canyon’ option, MIKE SHE
effectively assumes that the river is hydraulically connected to the upper
most model layer, since MIKE SHE calculates the exchange flow with all
layers that intersect the river based on the difference between the river
level and the water table.
Currently, this option is only available for steady-state models. It is activated by means of the boolean Extra Parameter, Enable Canyon
Exchange. For more information on the use of extra parameters, see Extra
Parameters (V.1 p. 299).
7.7
Overland Flow Exchange with MIKE 11
The exchange between overland flow and MIKE 11 rivers can be calculated in three different ways. If the flooding from MIKE 11 to MIKE SHE
cells is ignored (the "no flooding" option) then the exchange from overland flow is one way - that is overland flow only discharges to MIKE 11
rivers. If the you want to simulate flooding from MIKE 11 to MIKE SHE
then the water can be transfered from MIKE 11 to MIKE SHE using
"Flood Codes" or via direct overbank spilling using a wier formula. In
principle, the flood code option does not impact the solution time signficantly, is relatively easy to set up for simple cases and is sufficient when
detailed flood plain flow is not required. Direct overbank spilling com-
212
MIKE SHE
Overland Flow Exchange with MIKE 11
bined with the explicit solution method requires more detailed topopgraphy data and is useful when detailed flood plain flow is required, but can
be significantly slower from a numerical perspective.
Flooding with Overbank Spilling
If you are simulating flooding on the flood plain using the overbank spilling option, then the MIKE 11 cross-sections are normally restricted to the
main channel. The flood plain is defined as part of the MIKE SHE topography. Since, the bank elevation is used to define when a cell floods, it is
more critical that the cross-sections are consistent with your topography,
in the areas where you want to simulate flooding. The table in the simulation log file mentioned above is useful to locate these inconsistencies. It is
usually necessary to have a very fine grid and a detailed DEM for such
simulations, which tends to reduce the inconsistencies because it reduces
the amount of interpolation and averaging when creating the model topography.
Flooding with Flood Codes
If you are simulating flooding on the flood plain using the flood code
option, then flood plain elevation should be consistent with the cross-sections. Otherwise, the flood plain storage will be inconsistent with the river
storage based on the cross-sections.
When you are using Flood Codes, you typically specify wide cross-sections for your rivers. The wide cross-sections can then account for the
increased flood plain storage during flood events. MIKE 11 then places
water on the MIKE SHE cells that are defined by flood codes - if the water
level in the river is above the cell topography. The flood water is then free
to infiltrate or evaporate as determined by MIKE SHE.
In such flooded cells, overland flow is no longer calculated, so there is no
longer any overland exchange to MIKE 11 in flooded cells. Thus, the bank
elevation is not so critical, as long as the cell is flooded. However, when
the flood recedes, the cells revert back to normal overland flow cells and
the same considerations apply as if the cells were not flooded - namely the
bank elevation should be below the topography to ensure that overland
flow can discharge to the river link.
Flood codes are also commonly used for lakes and reservoirs. In this case,
you specify the lake bed bathymetry as the topography (or using the
Bathymetry option). The lake area is defined using flood codes and the
MIKE 11 cross-sections stretch across the lake. MIKE 11 calculates the
lake level and floods the lake. Overland flow adjacent to the lake intersects the flooded cells and the overland water is added to the lake cell (and
to MIKE 11 as lateral inflow). Groundwater exchange to the lake is
Surface Water
213
Surface Water in MIKE SHE
through the lake bed as saturated zone discharge. In principle, the saturated zone could discharge to the river link, but the local groundwater gradients would probably make this exchange very small.
Combining Flood Codes and Overbank Spilling
Flooding using Overbank spilling and Flood Codes is possible in the same
model and even in the same coupling reach. The only restriction is that
there is no overland flow calculated in cells flooded by means of Flood
Codes. So, in a long coupling reach, you could allow overbank spilling
and calculate overland flow using the explicit solver, but define flood
codes in the wide downstream flood plain were the surface water gradients
are very low during flooding and in the wide shallow reservoir half way
down the system.
7.7.1
Lateral inflow to MIKE 11 from MIKE SHE overland flow
MIKE SHE’s overland flow solver calculates the overland flow across the
boundary of the MIKE SHE cells. If a river link is located on the cell
boundary, any overland flow is intercepted by the river link and added to
the water balance of the river link. However, two checks are first made to
ensure exchange to the river is physically possible. The level of ponded
water in the cell must be above the
1 water level in the river link, and
2 bank elevation of the river link.
In the second case, the level of ponded water is checked against the appropriate left and right bank elevations of the river link.
However, there is no mechanism for exchange from MIKE 11 to overland
flow. If the water level in the river rises above the bank elevation, then the
bank elevation is simply extended vertically upwards.
7.7.2
Flooding from MIKE 11 to MIKE SHE using Flood Codes
The MIKE SHE/MIKE 11 coupling allows you to simulate large water
bodies such as lakes and reservoirs, as well as flooded areas. If this option
is used, MIKE SHE/MIKE 11 applies a simple flood-mapping procedure
where MIKE SHE grid points (e.g. grid points in a lake or on a flood
plain) are linked to the nearest H-point in MIKE 11 (where the water levels are calculated). Surface water stages are then calculated in MIKE SHE
by comparing the water levels in the H-points with the surface topographic elevations.
Conceptually, you can think of the flooded cells as “side storages”, where
MIKE 11 continues to route water downstream as 1D flow. But, at the
214
MIKE SHE
Overland Flow Exchange with MIKE 11
same time, the water is available to the rest of MIKE SHE for evaporation
and infiltration.
Determination of the Flooded Area and Water Levels
The flooded area in MIKE SHE must be delineated by means of integer
flood codes, where each coupling reach is assigned a flood code.
During the simulation, the flood-mapping procedure calculates the surface
water level on top of each MIKE SHE cell with a flood code by comparing
the MIKE 11 surface water level to the surface topography in the model
grid. A grid cell is flooded when the MIKE 11 surface water level is above
the topography. The MIKE 11 water level is then used as the level of ponded surface water.
The actual water level in the grid cell is calculated as a distance weighted
average of the upstream and downstream MIKE 11 H-points.
Calculation of the Exchange Flows
After the MIKE SHE overland water levels have been updated, MIKE
SHE calculates the infiltration to the unsaturated and saturated zones and
evapotranspiration. Thus, MIKE SHE simply considers any water on the
surface, including MIKE 11 flood water as ‘ponded water’, disregarding
the water source. In other words, ponded rainfall and ponded flood water
are indistinguishable.
MIKE SHE does not calculate overland flow between cells that are
flooded by MIKE 11. Nor, does MIKE SHE calculate overland exchange
to MIKE 11, if the cell is flooded by MIKE 11. However, lateral overland
flow to neighbouring non-flooded cells is allowed. Thus, if there is a
neighbouring, non-flooded cell with a topography lower than a flooded
cell’s water level , then MIKE SHE will calculate overland flow to the
non-flooded cell as normal.
The calculated exchange flow between the flooded grid cells and the overland, saturated, unsaturated zone or other source/sink terms is fed back to
MIKE 11 as lateral inflow or outflow to the corresponding H-point in the
next MIKE 11 time step.
In terms of the water balance, the surface water in the inundated areas
belongs to the MIKE 11 water balance. In other words, if there is ponded
water on the surface when the grid cell floods, the existing ponded water is
added to the MIKE 11 water flow in the river. As long as the element is
flooded, any exchange to or from the surface water is managed by MIKE
11 as lateral inflow and regular overland flow is not calculated.
Surface Water
215
Surface Water in MIKE SHE
If the element reverts back to a non-flooded state, then any subsequent
ponded water is again treated as regular overland flow and the water balance is accounted for within the overland flow component.
7.7.3
Direct Overbank Spilling to and from MIKE 11
If you want to calculate 2D overland flow on the flood plain during a
storm event, then you cannot use the Flooding from MIKE 11 to MIKE
SHE using Flood Codes (V.1 p. 214) method. The Area Inundation method
is primarily used as a way to spread river water onto the flood plain and
make it available for interaction with the subsurface via infiltration and
evapotranspiration.
The Overbank spilling option treats the river bank as a weir. When the
overland flow water level or the river water level is above the left or right
bank elevation, then water will spill across the bank based on the standard
weir formula
H ds – H w k
Q = ∆x ⋅ C ⋅ ( H us – H w ) ⋅ 1 –  ----------------------
H us – H w
k
·
0 385
(7.11)
where Q is the flow across the weir, ∆x is the cell width, C is the weir
coefficient, Hus and Hds refer to the height of water on the upstream side
and downstream side of the weir respectively, Hw is the height of the weir,
and k is a head exponent.
The units of the weir coefficient depend on the exponent. In MIKE SHE,
the default exponent is 1.5, which means that the weir coefficient has units
of m1/2/s.
If the water levels are such that water is flowing to the river, then the overland flow to the river is added to MIKE 11 as lateral inflow. If the water
level in the river is higher than the level of ponded water, then the river
water will spill onto the MIKE SHE cell and become part of the overland
flow.
If the upstream water depth over the weir approaches zero, the flow over
the weir becomes undefined. Therefore, the calculated flow is reduced to
zero linearly when the upstream height goes below a threshold.
If you use the overbank spilling option, then you should also use the
Explict Numerical Solution (V.2 p. 271) for overland flow.
216
MIKE SHE
Unsaturated Flow exchange with MIKE 11
7.7.4
Converting from Flood Codes to Overbank Spilling
The explicit solver and overbank spilling from MIKE 11 to overland flow
are new in the 2007 Release. In principle, if you were careful setting up
your flood codes, then the conversion to overbank spilling should result in
the same flooded area, with similar depths. The only difference will be
that the water on the flooded area is flowing.
However, in practice the conversion is not likely to be this smooth. Flood
code setups are typically done manually and the topography is typically
not very closely controlled - as long as it was inundated when it was supposed to be. Furthermore, the need for detailed surface roughness (ManningsM) will require additional data. Finally, the complication of fully
dynamic (diffusive wave) 2D flow can lead to complicated water flows
across the flood plain. So, there is likely to be substantial adjustment and
re-calibration to get the flooding right.
Fortunately, you can mix Flood codes and Overbank spilling in the same
model and even in the same coupling reach. This allows you to update
only the parts of your model where the overbank spilling is important and
leave the Flood code option intact elsewhere.
7.8
Unsaturated Flow exchange with MIKE 11
Direct exchange between MIKE 11 and the unsaturated zone is not currently supported. Groundwater exchange is assumed to be a line source
and sink at the boundary between cells and the exchange mechanism
assumes that the primary exchange takes place along the river banks. This
is a suitable assumption when the river is well connected to the aquifer.
However, when MIKE 11 can exchange water with overland flow via
overbank spilling or flood codes, then river water is added to the ponded
water on a MIKE SHE cell, which can then infiltrate to the unsaturated
zone.
7.9
Water balance with MIKE 11
The water balance tool in MIKE SHE (Using the Water Balance Tool
(V.1 p. 105)) includes the exchange with MIKE 11, but it does not include
the water balance within MIKE 11. In other words, once water enters
MIKE 11 it is no longer part of the MIKE SHE water balance. Thus, there
are numerous water balance items that detail the different exchanges to
and from MIKE 11.
Surface Water
217
Surface Water in MIKE SHE
Water exchanges within MIKE 11 can be evaluated using the MIKE View
tool. In some cases, this may require you to include the additional output
for MIKE 11, which is selected in the Additional Output tab in MIKE 11’s
HD editor.
Note: output in MIKE 11 is instantaneous, whereas the output in MIKE
SHE is generally accumulated within a time step. Therefore, a flow at a
rate at a point in MIKE 11 (e.g. a weir) will be the instantaneous flow at
the end of the time step. In MIKE SHE, however, the flow into a cell will
be the average flow over the time step.
7.10
Coupling MIKE SHE Water Quality to MIKE 11
Detailed information on the MIKE 11 Water Quality modules are found in
the MIKE 11 documentation.
The coupling between MIKE 11 and the rest of MIKE SHE’s hydrologic
processes is relatively automatic. You must set up a MIKE 11WQ model
independent of MIKE SHE and specify this .sim11 file in the Rivers and
Lakes dialog. This .sim11 file must only have the same network geometry
as the WM .sim11 file. It does not have to be the same .sim11 file.
The MIKE 11 WQ model can also include EcoLab, which will allow you
simulate eutrophication, etc. in the surface water.
There are a few caveats/limitations that you need to be aware of:
218
z
Species names must be identical in MIKE SHE and MIKE 11. If they
are not identical, then the solutes will be transfered to the river as an
infinite sink, but will not be transported in MIKE 11.
z
The overland WQ must be included if you want to simulate water quality coupled to MIKE 11.
z
Recycling of WM results is not supported in MIKE 11. This means that
if you want to simulate the coupling between MIKE 11 and the rest of
MIKE SHE, your WQ simulation must be continuous.
z
There is no solute transfer from MIKE 11 to MIKE SHE via overbank
spilling or flood codes. Only the water is transfered to flood codes and
overbank spilling. Any solutes will remain in MIKE 11. Thus, solute
transfer from MIKE 11 to MIKE SHE’s SZ is the only transfer supported. Solute transfer from MIKE SHE to MIKE 11 is supported for
both overland and saturated flow.
MIKE SHE
MIKE 11 User Interface
7.11
MIKE 11 User Interface
The following section provides additional information for the MIKE 11
dialogues that are commonly used with MIKE SHE.
7.11.1
MIKE SHE Coupling Reaches
Each MIKE 11 branch that exchanges water with MIKE SHE is called a
coupling reach. A MIKE 11 branch can be sub-divided into several coupling reaches. A reason for doing so could be to allow different riverbed
leakage coefficients for different parts of the river.
The upper half of the dialogue displays the properties of the current coupling reach. While, the bottom half of the dialogue is a table listing all of
the coupling reaches defined.
Figure 7.13
MIKE SHE River Links dialogue in the tabular view of the MIKE 11
Network Editor
Include all branches button
If the Include all branches button is pressed all the branches in the MIKE
11 setup will be copied to the MIKE SHE Links table. Branches that
should not be in the coupling can subsequently be deleted manually and
the specifications for the remaining branches completed. Thus, you may
have a large and complex hydraulic model, but only couple certain reaches
to MIKE SHE. All branches will still be in the hydraulic MIKE 11 model
Surface Water
219
Surface Water in MIKE SHE
but MIKE SHE will only exchange water with branch reaches that are
listed in the MIKE SHE links table.
Note The Include all branches button will erase all existing links that have
been specified.
Location
The branch name, upstream chainage and downstream chainage
define the stretch of river that can exchange water with MIKE SHE. A
MIKE 11 branch can be sub-divided into several coupling reaches, to
allow, for example, different riverbed leakage coefficients for different
parts of the river.
River Aquifer Exchange
Conductance
The river bed conductance can be calculated in three ways.
Aquifer only - When the river is in full contact with the aquifer material,
it is assumed that there is no low permeable lining of the river bed. The
only head loss between the river and the grid node is that created by the
flow from the grid node to the river itself. This is typical of gaining
streams, or streams that are fast moving. More detailed information on this
option can be found in Aquifer Only Conductance (V.1 p. 208).
River bed only - If there is a low conductivity river bed lining, then there
will be a head loss across the lining. In this case, the conductance is a
function of both the aquifer conductivity and the conductivity of the river
bed. However, when the head loss across the river bed is much greater
than the head loss in the aquifer material, then the head loss in the aquifer
can be ignored (e.g. if the bed material is thick and very fine and the aquifer material is coarse). This is the assumption used in many groundwater
models, such as MODFLOW. More detailed information on this option
can be found in River bed only conductance (V.1 p. 209).
Aquifer + Bed - If there is a low conductivity river bed lining, then there
will be a head loss across the lining. In this case, the conductance is a
function of both the aquifer conductivity and the total conductivity of the
between the river and the adjacent groundwater can be calculated as a
serial connection of the individual conductances. This is commonly the
case, when the aquifer material presents a significant head loss. For example, when the aquifer is relatively fine and the groundwater cells are quite
large.More detailed information on this option can be found in Both aquifer and river bed conductance (V.1 p. 210).
220
MIKE SHE
MIKE 11 User Interface
Leakage Coefficient - [1/sec]
This is the leakage coefficient for the riverbed lining in units of [1/seconds]. The leakage coefficient is active only if the conductance calculation
method includes the river bed leakage coefficient.
Linear Reservoir Exchange
If you are using the Linear Reservoir method for groundwater in MIKE
SHE, then by default the Interflow and Baseflow reservoirs discharge uniformly to all the river links within the reservoir. This is generally true in
the lower reaches. However, in the upper reaches many rivers discharge to
the groundwater system.
In this dialogue, you can define whether or not a branch is a Gaining
branch (default) or a Losing branch. If the branch is a:
z
Gaining branch, then the leakage coefficient and wetted area are
ignored and the rate is discharge from the Baseflow reservoir to the
river is calculated based on the Linear Reservoir method.
z
Losing branch, then the rate of discharge from the river to the Baseflow
reservoir is calculated using:
Q=water depth*bank width* branch length* leakage coefficient.
The gaining and losing calculations are done in MIKE SHE for every river
link within the Baseflow reservoir. For the losing river links, the water
level is interpolated from the nearest H-points, the bottom elevation and
bank width is interpolated from the nearest cross-sections. The length is
simply the cell size. MIKE SHE keeps track of the inflow volumes to
ensure that sufficient water is available in the river link.
Weir Data for overland-river exchange
The choice of using the weir formula for overland-river exchange is a global choice made in the MIKE SHE OL Computational Control Parameters
(V.2 p. 36) dialogue. If the weir option is chosen in MIKE SHE, then all
MIKE 11 coupling reaches will use the weir formula for moving water
across the river bank.The weir option is typically used when you want to
simulate overbank spilling and detailed 2D surface flow in the flood
plains. The following parameters and options are available when you
specify the weir option in MIKE SHE. If you chose the Manning equation
option in MIKE SHE, then these parameters are ignored.
Surface Water
221
Surface Water in MIKE SHE
Weir coefficient and Head exponent
The Weir coefficient and head exponent refer to the C and k terms respectively in Equation (7.11). The default values are generally reasonable.
Both the weir coefficient and the head exponent are dimensionless.
Minimum upstream height above bank for full weir width
In Equation (7.11), when the upstream water depth above the weir
approaches zero, the flow over the weir becomes undefined. To prevent
numerical problems, the flow is reduced linearly to zero when the water
depth is below the minimum upstream height threshold. The EUM data
type is Water Depth.
Allow overbank spilling
This checkbox lets you define which branches are allowed to flood over
their banks. Thus, you can allow flooding from MIKE 11 only in branches
with defined flood plains, or only in areas of particular interest.
If overbank spilling is not allowed for a particular branch, then the overland-river exchange is still calculated using the weir formula, but the
exchange is only one way - that is from overland flow to the river.
Minimum flow area for overbank spilling
The minimum flow area threshold prevents overbank spilling when the
river is nearly dry. The flow area is calculated by dividing the volume of
water in the coupling reach by the length of the reach. The EUM data type
is Flow Area, which by default is m2.
The default value is 1 m3/m length of river. This is quite a small amount of
water for most reasonable rivers and should be adjusted based on the river
width. For example, if your river is 10m wide, then spilling will occur
when the water level is 10cm above the bank elevation. However, if your
river is 200m wide, then spilling would start when the water level is only
5mm above the bank elevation.
The cell size also plays a role here. When a cell is flooded, the entire cell
is covered by water. If the cell size is 1000m x 1000m, then a flood of 1
m3/m of river will be only 1mm deep across the cell.
Inundation options by Flood Code
The Inundation method allows specified model grid cells to be flooded if
the MIKE 11 water level goes above the topography of the cell. In this
case, water from MIKE 11 is “deposited” onto the flooded cell. The flood
water can then infiltrate, or evaporate. However, overland flow between
flooded cells and to the river is not calculated. Also, the flooded water
222
MIKE SHE
MIKE 11 User Interface
remains as part of the MIKE 11 water balance and is only transferred to
MIKE SHE when it infiltrates.
Inundation areas and their associated Flood codes are specified on a coupling reach basis.
Flood Area Option
The following three options are available for the Flood Area Option:
z
No Flooding (default) With the No flooding option, the MIKE 11 river
is confined between the left and right banks. If the water level goes
above the bank elevation, then the river is assumed to have vertical
banks above the defined left and right bank locations. No flooding via
flood codes will be calculated.
Note If neither inundation nor overbank spilling is allowed, then the
overland flow exchange to the river is one way only. The only mechanism for river water to flow back into MIKE SHE is through baseflow
infiltration to the groundwater. If overland flow does spill into the
river, there is first a check to make sure that the water level in the river
is not higher than the ponded water.
Surface Water
z
Manual If the Manual option is selected, then you must supply a Flood
code map in MIKE SHE. This Flood code map is used to established
the relationship between MIKE 11 h-points and individual model grids
in MIKE SHE. MIKE SHE then calculates a simple flood-mapping
during the pre-processing that is used during the simulation to assign
river water stages to the MIKE SHE cells if the river level is above the
topography.
z
Automatic The automatic flood mapping option is useful if the river
network geometry is not very complex or for setting up the initial flood
mapping, for later refinement. The automatic method, maps out a polygon for each coupling reach based on the left and right bank locations
of all the cross-sections along the coupling reach. All cells within this
polygon are assigned an integer flood code, unique to the coupling
reach. The automatic method works reasonably well along individual
branches with cross-sections that represent the flood plain. At branch
intersections the assigned flood code may not be correct. However, this
is often not serious because at river confluences the water levels in the
different branches are roughly the same anyway. In any case, the flood
code map is available in MIKE SHE’s preprocessed tab, where you can
check its reasonableness. Right clicking on the map will give you the
option of saving the map to a dfs2 file, which you can then correct and
use with the Manual option.
223
Surface Water in MIKE SHE
Flood Code
If the Manual option is selected, then you must specify a Flood code for
the coupling reach. The flood code is used for mapping MIKE SHE grids
to MIKE 11 h-points. You must click on the Flood Code checkbox in
Figure 7.13, and then specify an integer flood code file in MIKE SHE.
The specified flood code for the coupling reach must exist in the dfs2
Flood Code file. It is important to use unique flood codes to ensure correct
flood-mapping.
Bed Topography
Since the flood mapping procedure will only flood a cell when the river
water level is above the cell’s topography, accurate flood inundation mapping requires accurate elevation data. If one of the flood options are
selected, then you have the option to refine the topography of the flood
plain cells based on the actual cross-section elevations or on a more
detailed local-scale DEM, if it exists.
z
Use Grid Data (default) If Grid Data option is selected, the MIKE
SHE topography value is used to determine whether or not the cell is
flooded. However, the program first checks to see if a Bathymetry file
has been specified.
If a Bathymetry file is available, the topography values of the cells
with flood codes are re-interpolated based on the bathymetry data. The
bathymetry option is useful when a more detailed DEM exists for the
flood plain area compared to the regional terrain model.
z
Use Cross-section If the Cross-section option is specified the topography values of the cells with flood codes are re-interpolated based on
the cross-section data.
When the cross-section option is selected, the pre-processor maps out a
flood-plain polygon for the coupling reach, based on the left and right
bank locations of all the cross-sections along the coupling reach. Interpolated cross-sections are created between the available actual crosssections, if the cross-section spacing is greater than ½ ∆x (grid size).
All the cross-sections (real and interpolated) are sampled to obtain a set
of point values for elevation in the flood plain. The topography values
of all cells with the current flood code that are within the flood-plain
polygon are re-interpolated using the bilinear interpolation method to
obtain a new topography value.
In principle, the Cross-section option ensures a good consistency
between MIKE SHE grid elevations and MIKE 11 cross-sections.
There will, however, often be interpolation problems related to river
meandering, tributary connections, etc., where wide cross-sections of
224
MIKE SHE
Common MIKE 11 Error Messages
separate coupling reaches overlap. Thus, you can make the initial
MIKE SHE set-up using the Cross-section option and then subsequently retrieve and check the resulting ground surface topography,
from the pre-processed data. If needed, the pre-processed topography
can be saved to a .dfs2 file (right click on the map), modified and then
used as input for a new set-up, now using the Use Grid Data option.
Bed Leakage
If one of the flood options are selected, then you must also specify if and
how the leakage coefficient will be applied on the flooded cells. The infiltration/seepage of MIKE SHE flood grids is calculated as ordinary overland exchange with the saturated or unsaturated zone. That is, the leakage
coefficient, if it exists, is applied to both saturated exchange to and from
the flooded cell and unsaturated leakage from the flooded cell. In the case
of the unsaturated leakage, the actual leakage is controlled by either the
leakage coefficient or the unsaturated zone hydraulic conductivity relationship - which yields the lowest infiltration rate.
7.12
z
Use grid data In this case, the leakage coefficient specified in Surface-Subsurface Leakage Coefficient is used. If this item has not been
specified, then the leakage coefficient will be calculated based on the
aquifer material only.
z
Use river data (default) In this case, the Leakage Coefficient - [1/sec]
for the coupling reach is actually copied to the flooded cell and used
for all flood grid points of the coupling reach. This makes sense if the
flood plain is frequently flooded and covered with the same sediments
as the river bed. However, in many cases the flood plain material is not
the same as the river bed and the infiltration rate can be substantially
different.
Common MIKE 11 Error Messages
There are a number of common MIKE 11 error messages that you are
likely to encounter when using MIKE 11 with MIKE SHE.
7.12.1
Error No 25: At the h-point ____ the water depth greater than 4 times max.
depth
This error message essentially says that your MIKE 11 model is unstable.
It frequently occurs when there is an inconsistency in your bed elevations
at the branch junctions. For example, if the bed elevation of the main
branch is much greater than the side branch, then the water piles up and
causes this error.
Surface Water
225
Surface Water in MIKE SHE
7.12.2
Warning No 47: At the h-point ____ the water level as fallen below the
bottom of the slot x times
This warning message essentially says that your MIKE 11 model is unstable. The slot is a numerical trick that keeps a very small amount of water
in the MIKE 11 cross-section when the river is dry. So, when the water
level falls below the slot, it implies that your river has dried out. This
warning frequently occurs when there is either an inconsistency in your
bed elevations or there is an error in your boundary conditions that is
keeping water from entering the system.
7.12.3
Warning No __: Bed levels not the same
This warning message is issued when the bed elevation of a side branch is
not the same as the main branch. If the difference is small (say a few cm) it
can usually be ignored. However, if the side branch is much lower than the
main branch then this warning will often be accompanied by Error No 25:
At the h-point ____ the water depth greater than 4 times max. depth, as the
water will pile up and not be able to flow into the main branch. If the side
branch is only slightly lower than the main branch or even if they are the
same, then backward flows can occur in the side branch when the water
level in the main branch rises. If this is realistic fine, but often it is not.
More typically, the side branch is slightly higher than the main branch.
226
MIKE SHE
DRAINAGE MODELLING WITH
MIKE URBAN
227
228
MIKE SHE
8
USING MIKE SHE WITH MIKE URBAN
Coupling MIKE URBAN and MIKE SHE allows you to simulate the
effect of urban drainage and sewer systems on the surface/subsurface
hydrology.
The use of the integrated MIKE SHE/MIKE URBAN system is not very
different from establishing a stand-alone MIKE URBAN model and a
stand-alone MIKE SHE model. In principle there are three basic set-up
steps to have a coupled MIKE SHE-MIKE URBAN model:
1 Establish a MIKE URBANMIKE URBAN hydraulic model as a standalone model, make a performance test and, if possible, a rough calibration using prescribed inflow and boundaries.
2 Establish a MIKE SHE model that includes the overland flow component and (optionally) the saturated zone and unsaturated zone components.
3 Couple MIKE SHE and MIKE URBAN by defining the locations
where MIKE URBAN should interact with MIKE SHE.
When MIKE SHE runs, it will call MIKE URBAN and ask it to perform a
MIKE URBAN time step. If the end of the MIKE SHE time step has not
yet been reached, MIKE SHE will ask MIKE URBAN to calculate the
next MIKE URBAN time step. The MIKE URBAN model will run normally if it is launched directly from MIKE URBAN.
Note: The MIKE URBAN coupling was originally developed for the
stand-alone sewer modelling product called MOUSE, which was later
incorporated into MIKE URBAN. Thus, references in this chapter to
MIKE URBAN can largely be substituted by “MOUSE”. Further, older
MOUSE models can be coupled to MIKE SHE using the same method
described here.
Important: In the command lines in the input files, the word “mouse”
must still be used. For example, the Extra Parameters option to activate
the MIKE URBAN coupling must be “mouse coupling”.
Drainage modelling with MIKE URBAN
229
Using MIKE SHE with MIKE URBAN
Figure 8.1
MIKE SHE to MIKE URBAN coupling linkages
The exchange between MIKE URBAN and MIKE SHE is calculated
based on the following equation
Q = C ⋅ ( HSHE – H MOUSE )
k
(8.1)
where Q is the exchange between MIKE URBAN and MIKE SHE, C is
the exchange coefficient, k is a head difference exponent and
H SHE = Max ( H cell, Z T, Z M )
(8.2)
H MOUSE = Max ( H pipe, Z T, Z M )
(8.3)
where Hcell is the head in the MIKE SHE cell, Hpipe is the head in the
MIKE URBAN pipe, ZT is the topographic elevation in the cell and ZM is
the elevation of the manhole.
There are five variations on how to calculate the exchange based on above
equations:
230
MIKE SHE
MIKE SHE SZ to MIKE URBAN LINKS
This is a leakage-based solution in which the head difference exponent is
1 and the exchange coefficient in Equation (8.1) for the flow to or from the
pipe is calculated by
C = CL ⋅ RH ⋅ L
(8.4)
where CL is the leakage coefficient (see below), RH is the hydraulic radius
for the flow (see below), and L is the length of the MIKE URBAN pipe
(link) in the MIKE SHE cell.
Leakage Coefficient - The leakage coefficient can be defined in two
ways.
Option 1 is the simple method, which is to use the pipe leakage coefficient specified in the MIKE URBAN .ADP file. See Telling MIKE
URBAN that it is coupled to a MIKE SHE model (V.1 p. 234).
Option 2 uses a combination of the pipe leakage coefficient and the aquifer hydraulic conductivity. In this case, the leakage coefficient is calculated as a series connection of the pipe leakage coefficient (Cp) and the
“average” leakage coefficient of the aquifer grid cell (Caq). The average
leakage coefficient of the grid cell is calculated assuming that the
exchange of water between the pipe and the grid cell is both vertical and
horizontal. The leakage coefficient calculation does not calculate a
detailed flow path based on a geometric calculation, since a MIKE
URBAN pipe can be located anywhere in a grid cell. Instead, an average
vertical and horizontal flow distance is used based on 1/4 of the vertical
and horizontal cell dimensions. Thus,
Kx
Kz
C aq = C aqH + C aqV = ------------------ + ----------------( ∆x ) ⁄ 4 ( ∆z ) ⁄ 4
(8.5)
where Kx and Kz are the horizontal and vertical hydraulic conductivities
respectively and ∆x and ∆z are the horizontal and vertical cell dimensions.
The final leakage coefficient is then calculated as the harmonic mean of
both the aquifer leakage coefficient and the pipe leakage coefficient:
11
1
----= --------- + -----CL
C aq C p
Drainage modelling with MIKE URBAN
(8.6)
231
Using MIKE SHE with MIKE URBAN
Hydraulic Radius - MIKE SHE uses the inner hydraulic radius if the
flow is from MIKE URBAN to MIKE SHE. Whereas, it uses the outer
hydraulic radius if the flow is from MIKE SHE to MIKE URBAN. The
hydraulic radii are calculated by MIKE URBAN.
MIKE SHE Overland flow to MIKE URBAN LINKS
If a MIKE URBAN link is defined as link type CRS or Natural Channel
and has a cross section which is "open", then MIKE SHE can exchange
overland flow with it in both directions. In this case, the exchange coefficient in Equation (8.1) is defined as
C = CL ⋅ L
(8.7)
where CL is the conductance and L is the length of the MIKE URBAN
pipe (link) in the MIKE SHE cell.
If the exponent Equation (8.1) is 1.0, then this is a simple drain formulation and the conductance is per length with units of [m/s]. If the exponent
is 1.5, then this is a weir formulation and the units of the conductance term
are [m1/2/s].
MIKE SHE Overland flow to MIKE URBAN Manholes
If the MIKE URBAN manholes are not sealed, then MIKE SHE can discharge overland flow into the MIKE URBAN manholes. In this case, the
exchange coefficient in Equation (8.1) is defined as
C = CL
(8.8)
where CL is the conductance.
If the exponent Equation (8.1) is 1.0, then this is a simple drain formulation and the conductance, CL, is per length with units of [m/s]. If the exponent is 1.5, then this is a weir formulation and the units of the conductance
term are [m1/2/s].
MIKE SHE SZ drain flow to MIKE URBAN Manholes
If drain flow is specified in MIKE SHE, then the drainage can be discharged to a MIKE URBAN manhole. The flow in the drain is calculated
by MIKE SHE based on the groundwater height above the drain level. In
MIKE SHE the distributed drainage option must be chosen (see Drainage
(V.2 p. 173)) and the cells that drain to a manhole must have an option
value of 4 (see Option Distribution (V.2 p. 179)). The references between
232
MIKE SHE
Coupling MIKE SHE and MIKE URBAN
the MIKE SHE drain codes and the MIKE URBAN manholes are defined
in the MsheMouse.pfs file (see Creating a MsheMouse.pfs file
(V.1 p. 235)).
MIKE SHE Paved Areas to MIKE URBAN Manholes
If the paved area option (see Land Use (V.2 p. 93)) is used in MIKE SHE,
then the flow generated on the paved areas can be discharged to a MIKE
URBAN manhole. MIKE SHE’s paved area flow module uses the same
reference system as the drain component. This option is automatically
activated when the MIKE SHE drains in the paved areas point to a MIKE
URBAN manhole.
MIKE URBAN Outlets to MIKE SHE
MIKE URBAN outlets cannot directly discharge to MIKE SHE’s overland
flow. To work around this, you can add a dummy manhole to your MIKE
URBAN pipe and then couple the pipe to the outlet via a small diameter
dummy pipe (See Figure 8.2). This will force most of the water out of the
manhole and into MIKE SHE’s overland flow. Downside of this method,
is that the head loss at the outlet is over estimated, because the discharge
velocity is zero at a manhole.
Figure 8.2
8.1
Work around for discharging MIKE URBAN outlets to MIKE SHE
Coupling MIKE SHE and MIKE URBAN
The MIKE URBAN coupling in MIKE SHE has not yet been added to the
MIKE SHE user interface. Thus, to couple the models together, you must:
1 tell MIKE SHE to look for a MIKE URBAN model,
2 tell MIKE URBAN that it is coupled to a MIKE SHE model
3 create an MsheMouse.pfs file to define where and how the two models
are coupled.
Drainage modelling with MIKE URBAN
233
Using MIKE SHE with MIKE URBAN
8.1.1
Telling MIKE SHE to couple to MIKE URBAN
To tell MIKE SHE that it needs to couple to a MIKE URBAN model, you
must add the following two items in the Extra Parameters (V.2 p. 193) section of the MIKE SHE Setup Editor.
Parameter
Name
Type
Value
mouse coupling
Boolean
On
mouse coupling
file
file name
the file name of the MIKE URBAN coupling .pfs input file
Note, that the parameter names must be spelled exactly as shown. For
more information on the use of extra parameters see Extra Parameters
(V.1 p. 299).
8.1.2
Telling MIKE URBAN that it is coupled to a MIKE SHE model
To couple a MIKE URBAN model to MIKE SHE, MIKE URBAN must
be supplied with some extra information. This information is found in
MIKE URBAN’s .ADP file.
Line item
Comment
[MOUSE_COUPLING]
SYNTAX_VERSION = 1
UNIT_TYPE = 1
CALLER = 'MSHE'
// LineHeader = 'ID', 'LinkType', 'C ','OLExp',
'SzLeakageCoef'
Comment line for headers
COUPLINGMMSHE= 'NODE1', 1, 0.001, 2, ,
One line for each coupling item:
COUPLINGMMSHE= 'LINK1', 2, 0.001, 2, 0.2
ID = Link name
LinkType = 1 for node; 2 for link
C = conductance for Overland flow to
MIKE URBAN, units depend on
OLExp and whether it is a pipe or a
manhole
SzLeakageCoeff = leakage coefficient; needed only when the saturated
zone is coupled to a link
[Endsect]
234
MIKE SHE
Coupling MIKE SHE and MIKE URBAN
8.1.3
Creating a MsheMouse.pfs file
The MsheMouse.pfs file is an ASCII file that includes all of the specifications for the coupling. The following table defines the structure of the file,
along with some information on the parameters. When the MIKE URBAN
coupling has been added to the user interface, the creation of this file will
be automatic.
Note: The pfs format must be adhered to exactly. There is a small utility
(pfsEditor.exe) in the installation \bin directly that you can use for editing
and testing pfs files that you create.
Table 8.1
MsheMouse.pfs file format and description
Line item
Comment
[MIKESHE_MOUSE_Specifications]
FileVersion = 2
Link_SZ_Exchange_Option = 2
1 = Leakage coefficient based only on
MIKE URBAN pipe leakage coefficient
2 = Leakage coefficient based on a
series connection of the MIKE
URBAN pipe leakage coefficient and
the MIKE SHE aquifer properties
Mouse_MPR_file name =
|.\MOUSE_NASSJO\handskeryd.mpr|
(See Note below this table)
Name of the MOUSE .mpr file or the
MIKE URBAN .mex file.
The MIKEZero file name format ( | | )
indicates that the file name is relative
to the location of this document.
SZ_Coupling = 1
1 or 0 to include/exclude SZ<->MIKE
URBAN coupling
OL_Coupling = 1
1 or 0 to include/exclude Overland<>MIKE URBAN coupling
Dynamic_Coupling = 1
1 for dynamic coupling. Otherwise the
initial MIKE URBAN conditions will
be used.
Drainage modelling with MIKE URBAN
235
Using MIKE SHE with MIKE URBAN
Table 8.1
MsheMouse.pfs file format and description
Line item
Comment
Drainage_To_Manholes = 1
1 to include SZ (and paved area) drain
to manholes. In this case the SZ drain
option must be Levels and Codes
(should rather be named Distributed
Option). In the areas with drain to
MIKE URBAN the Distributed option
code must be 4. For each drain code
value found in areas with Distributed
code 4 a reference from the code to a
MIKE URBAN manhole must be
defined in the Drainage_Manholes
section (see below).
Smooth_SZ_Inflow = 1
Ensures a more smooth calculation of
flows to MIKE URBAN when the
MIKE SHE time steps are large compared to the MIKE URBAN time step.
The MIKE URBAN coupling is only
made at every integer multiple of the
MIKE SHE time step. If the Smooth
option is not activated, the flows to
MIKE URBAN can stop after a
number of MIKE URBAN time steps
because the calculated flow volume
exceeds the volume of the MIKE SHE
SZ/Overland grid cells. The Smooth
option tries to use a reduced flow rate
which equals the available volume /
coupling time.
Smooth_OL_Inflow = 1
[Dynamic_Coupling_Specifications]
236
Limit_Inflow = 0:
Specify 1 if the inflow to MIKE
URBAN should be limited so the
MIKE URBAN volume + inflow does
not exceed a specified fraction of the
maximum MIKE URBAN volume.
This is used to avoid instabilities due
to high pressure.
Limit_Outflow = 0:
Specify 1 if the outflow from MIKE
URBAN should be limited so the
MIKE URBAN volume - outflow
doesn't come below a specified fraction of the maximum MIKE URBAN
volume. This is used to avoid instabilities due to drying / negative volume.
MIKE SHE
Coupling MIKE SHE and MIKE URBAN
Table 8.1
MsheMouse.pfs file format and description
Line item
Comment
[Inflow_Limitations]
MaxVolFac_Links = 0.99
MaxVolFac_Manholes = 0.99
EndSect // Inflow_Limitations
[Outflow_Limitations]
MinVolFac_Links = 0.05
MinVolFac_Manholes = 0.05
EndSect // Outflow_Limitations
The inflow and outflow fractions are
specified here:
EndSect // Dynamic_Coupling_Specifications
No_Of_Storing_reaches = 2
[Storing_Reaches]
[Storing_Reach_1]
No_Of_Links = 2
LinkName_1 = 'Dike_01l1'
LinkName_2 = 'Dike_03l1'
EndSect // Storing_Reach_1
[Storing_Reach_2]
No_Of_Links = 1
LinkName_1 = 'Dike_04l1'
EndSect // Storing_Reach_2
EndSect // Storing_Reaches
When No_Of_Storing_reaches is
greater than 0, the [Storing_Reaches]
section must be specified, and inside
this the [Storing_Reach_1],
[Storing_Reach_2], ... defining the no.
of links and link names for each reach.
[Drainage_Manholes]
No_Of_DrainCodes = 8
[Draincode_1]
Draincode= 12
ManholeName='DNB3182'
Endsect // Draincode_1
.
.
Endsect // Draincode_8
EndSect // Drainage_Manholes
When No_Of_Storing_reaches is
greater than 0, the [Storing_Reaches]
section must be specified, and inside
this the [Storing_Reach_1],
[Storing_Reach_2], ... defining the no.
of links and link names for each reach.
EndSect // MIKESHE_MOUSE_Specifications
Note on file names:
The pfs file line item is always “Mouse_MPR_file name =”
When coupling MIKE SHE to an old MOUSE model, the MOUSE file
name has the extension “.mpr”.
When coupling MIKE SHE to MIKE URBAN, the equivalent file is the
“.mex” file. This file contains all the necessary information for the coupling and is generated automatically by MIKE URBAN.
Drainage modelling with MIKE URBAN
237
Using MIKE SHE with MIKE URBAN
To create the .mex file, you must start a sewer simulation from MIKE
URBAN. However, since the .mex file is only created when the simulation
is launched, if you make changes to the sewer network, then you must recreate the .mex file by first restarting the sewer simulation in MIKE
URBAN. Otherwise, your changes to the sewer network will not be
reflected in the coupled models.
8.1.4
Output Files
Output from the coupled run is written to a number of .dfs0 results filesall located in the standard results directory. In the case of storing reaches,
there is one item in the .dfs0 file for each storing reach.
Table 8.2
8.2
File names and conditions for output for the MIKE SHE-MIKE
URBAN coupling. ’setupname’ refers to the name of the model
setup file.
file name
The file is created when...
.\setupname\setupname_SZ2MouseReaches.dfs0
...the MIKE SHE SZ coupling is
included.
.\setupname\setupname_OL2MouseReaches.dfs0
...the MIKE SHE Overland coupling is included.
.\setupname\setupname_OL2MouseManholes.dfs0
...the MIKE SHE Overland flow
coupling to manholes is included.
.\setupname\setupname_SZDrain2MouseManholes.dfs0
...the MIKE SHE SZ drain coupling to manholes is included.
.\setupname\setupname_PavedDrain2MouseManholes.dfs0
...the MIKE SHE SZ paved areas
to manholes is included.
Warning messages
Exchange inflows reduced
Warning: Exchange inflows from Overland to MOUSE
reduced by Overland house-keeping in order to avoid
instabilities
No. of time steps: 27000 of 27000
Total a priori inflows: 1332286 m3
Total reduced inflows: 920643.0 m3 (69.10%)
MIKE SHE calculates tine in/out flows after an overland time step and
feeds them to MIKE URBAN for one or more MIKE URBAN time steps.
The calculations of these flows are not included in the implicit overland
flow solver. Thus, the “Total a priori flows” are the rough inflows calculated using Equation (8.1). However, to prevent water balance errors,
238
MIKE SHE
Water Balance Limitations
MIKE SHE checks the volume of water available in the grid cell. If the
volume is insufficient, then the inflow is reduced to the available amount.
The final value of inflows is the “Total reduced inflows”. Note though that
the total NET inflow to MIKE URBAN will be less than this value if the
flow goes from MIKE URBAN to MIKE SHE in other grid cells or other
time steps.
Ideally, the Total reduced inflow should be 100%, but in practice this is
rarely achieved.
8.3
Water Balance Limitations
The interaction with MIKE SHE is not included in the MIKE URBAN
Summary HTM file. Thus, the water added from MIKE SHE appears as
an error (i.e. 6: Continuity balance in MIKE Urban).
Drainage modelling with MIKE URBAN
239
Using MIKE SHE with MIKE URBAN
240
MIKE SHE
GROUNDWATER
241
242
MIKE SHE
9
UNSATURATED GROUNDWATER FLOW
Unsaturated flow is one of the central processes in MIKE SHE and in
most model applications. The unsaturated zone is usually heterogeneous
and characterized by cyclic fluctuations in the soil moisture as water is
replenished by rainfall and removed by evapotranspiration and recharge to
the groundwater table. Unsaturated flow is primarily vertical since gravity
plays the major role during infiltration. Therefore, unsaturated flow in
MIKE SHE is calculated only vertically in one-dimension.
Vertical, 1D unsaturated flow is sufficient for most applications. However,
this assumption may not be valid in some situations, such as on very steep
hill slopes with contrasting soil properties in the soil profile.
MIKE SHE includes an iterative coupling procedure between the unsaturated zone and the saturated zone to compute the correct soil moisture and
the water table dynamics in the lower part of the soil profile.
There are three options in MIKE SHE for calculating vertical flow in the
unsaturated zone:
z
the full Richards equation, which requires a tabular or functional relationship for both the moisture-retention curve and the effective conductivity,
z
a simplified gravity flow procedure, which assumes a uniform vertical
gradient and ignores capillary forces, and
z
a simple two-layer water balance method for shallow water tables.
The full Richards equation is the most computationally intensive, but also
the most accurate when the unsaturated flow is dynamic. The simplified
gravity flow procedure provides a suitable solution when you are primarily interested in the time varying recharge to the groundwater table based
on actual precipitation and evapotranspiration and not the dynamics in the
unsaturated zone. The simple two-layer water balance method is suitable
when the water table is shallow and groundwater recharge is primarily
influenced by evapotranspiration in the root zone.
9.0.1
UZ Classification
Calculating unsaturated flow in all grid squares for large-scale applications can be time consuming. To reduce the computational burden MIKE
SHE allows you to optionally compute the UZ flow in a reduced subset of
grid squares. The subset classification is done automatically by the preprocessing program according to soil and, vegetation distribution, climatic
zones, and depth to the groundwater table.
Groundwater
243
Unsaturated Groundwater Flow
Column classification can decrease the computational burden considerably. However, the conditions when it can be used are limited. Column
classification is either not recommended or not allowed when
z
the water table is very dynamic and spatially variable because the classification is not dynamic,
z
if the 2 layer UZ method is used because the method is fast and the
benefit would be limited,
z
if irrigation is used in the model because irrigation zones are not a classification parameter, and
z
if flooding and flood codes are used, since the depth of ponded water is
not a classification parameter
Thus, the column classification should probably be avoided today
because the models have become more complex, MIKE SHE has become
more efficient and computers have become faster.
If the classification method is used, then there are three options for the
classification:
z
Automatic classification The automatic classification requires a distribution of groundwater elevations (see Groundwater Depths used for
UZ Classification). This can be either the initial depth to the groundwater based on the initial heads, or you can supply a .dfs2 map of the
groundwater elevations. In both cases, you must supply a table of intervals upon which the classification will be based. The number of computational columns depends on how narrow the intervals are specified.
If, for example, two depths are specified, say 1 m and 2 m, then the
classification with respect to the depth to groundwater will be based on
three intervals: Groundwater between 0 m and 1 m, between 1 m and 2
m, and deeper than 2 m.
One tip is to extract a map of the calculated potential head in the very
upper saturated zone layer from a previous simulation. The map should
represent the time of the year when the largest variations of the groundwater table are expected (deep groundwater in the hills and shallow
groundwater close to the rivers). Repeat the procedure as calibration
improves.
If the Linear Reservoir method is used for the groundwater, then the
Interflow reservoirs are also used in the classification. However, since
feedback to the UZ only occurs in the lowest Interflow reservoir of
each subcatchment, the Interflow reservoirs are added to the Automatic
244
MIKE SHE
Classification in two zones - those that receive feedback and those that
don’t.
9.0.2
z
Specified classification Alternatively a data file specifying Integer
Grid Codes, where UZ computations are carried out can be specified,
with grid codes range from 2 up to the number of UZ columns (see
Specified classification). The location of the computational column is
specified by a negative code and the simulation results are then transferred to all grids with the an equivalent positive code. For example, if
a grid code holds the value -2 a UZ computation will be carried out for
the profile located in that grid. Simulation results will subsequently be
transferred to all grid codes with code value 2. An easy way to generate
a .dfs2 file to be used for specification of UZ computational columns is
to let the MIKE SHE setup program generate an automatic classification first, and subsequently extract the UZ classification grid codes.
The extracted .dfs2 file can be edited in the 2D editor as desired and
used to specify UZ computational grids.
z
Calculated in all Grid points (default) For most applications you
should specify that computations are to be carried out in all soil columns.
z
Partial Automatic Finally a combination of the Automatic classification and the Specified classification is available. If this option is chosen an Integer Grid Code file must be provide (see Partial automatic
classification) with the following grid codes: In grid points where automatic classification should be used the grid code 1 must be given. In
grid points where computation should be performed for all cells the
grid code 2 must be given.
Coupling Between Unsaturated and Saturated Zone
The following procedure should be used to ensure that the unsaturated
zone does not drop below the bottom of the first calculation layer of the
saturated zone:
Groundwater
z
After a simulation, create a map of grid statistics of the potential head
in the first calculation layer of the saturated zone
z
Subtract the map of the minimum potential head from the map of the
bottom level of the first calculation layer of the saturated zone.
z
View the difference map. If the difference is very small in some areas
of the map (e.g. <0.5 m), it is strongly advised to move the bottom
level of the first calculation layer of the saturated zone downwards.
z
Repeat this procedure until there are no small differences.
245
Unsaturated Groundwater Flow
The water balance program can be used to get an overview of errors due to
a bad setup of the unsaturated zone. The follow procedure can be used to
make a map of UZ-errors:
z
Create a sub catchment map by retrieving UZ-classification codes from
the input file.
z
Replace negative values of the classification code map by positive values in the 2D graphical editor.
z
Use the sub catchment map in the water balance setup file to make a
UZ map of the water balance, which will create your map of UZ-errors.
Vertical discretisation - The vertical discretisation of the soil profile typically contains small cells near the ground surface and increasing cell
thickness with depth. However, the soil properties are averaged if the
cell boundaries and the soil boundaries do not align. .
The discretisation should be tailored to the profile description and the
required accuracy of the simulation. If the full Richards equation is
used the vertical discretisation may vary from 1-5 cm in the uppermost
grid points to 10-50 cm in the bottom of the profile. For the Gravity
Flow module, a coarser discretisation may be used. For example, 10-25
cm in the upper part of the soil profile and up to 50-100 cm in the lower
part of the profile. Note that at the boundary between two blocks with
different cell heights, the two adjacent boundary cells are adjusted to
give a smoother change in cell heights.
Specific Yield of upper SZ layer
MIKE SHE forces the specific yield of the top SZ layer to be equal to the
“specific yield” of the UZ zone as defined by the difference between the
specified moisture contents at saturation, θs, and field capacity, θfc.This
correction is calculated from the UZ values in the UZ cell in which the initial SZ water table is located. For more information on the SZ-UZ specific
yield see Specific Yield of the upper SZ numerical layer (V.1 p. 252).
Limitations of the UZ - SZ coupling
The coupling between UZ and SZ is limited to the top calculation layer of
the saturated zone. This implies that:
246
z
As a rule of thumb, the UZ soil profiles should extend to just below the
bottom of the top SZ layer.
z
However, if you have a very thick top SZ layer, then the UZ profiles
must extend at least to below the deepest depth of the water table.
MIKE SHE
Groundwater
z
If the top layer of the SZ model dries out, then the UZ model usually
assumes a lower pressure head boundary equal to the bottom of the
uppermost SZ layer.
z
All outflow from the UZ column is always added to the top node of the
SZ model.
z
UZ nodes below the water table and the bottom of the top SZ layer are
ignored.
247
Unsaturated Groundwater Flow
248
MIKE SHE
Conceptualization of the Saturated Zone Geology
10
SATURATED GROUNDWATER FLOW
The saturated groundwater component of MIKE SHE includes all of the
water below the water table. If the water table is at or above the ground
surface then the unsaturated zone is turned off this this cell.
The unsaturated zone geology is not related to the saturated zone geology.
Instead the unsaturated zone geology is essentially independent of the saturated zone geology.
10.1
Conceptualization of the Saturated Zone Geology
The development of the geological model is probably the most time consuming part of the initial model development. Before starting this task,
you should have developed a conceptual model of your system and have at
your disposal digital maps of all of the important hydrologic parameters,
such as layer elevations and hydraulic conductivities.
In MIKE SHE you can specify your subsurface geologic model independent of the numerical model. The parameters for the numerical grid are
interpolated from the grid independent values during the preprocessing.
The geologic model can include both geologic layers and geologic lenses.
The former cover the entire model domain and the later may exist in only
parts of your model area. Both geologic layers and lenses are assigned
geologic parameters as either distributed values or as constant values.
The alternative is to define the hydrogeology based on geologic units. In
this case, you define the distribution of the geologic units and the geologic
properties are assigned to the unit.
Each geologic layer can be specified using a dfs2 file, a .shp file or a distribution of point values. However, you should be aware of the way these
different types of files are interpolated to the numerical grid.
The simplest case is that of distributed point values. In this case, the point
values are simply interpolated to the numerical grid cells based on the
available interpolation methods.
In the case of shp files, at present, only point and line theme .shp files are
supported. Since lines are simply a set of connected points, the .shp file is
essentially identical to the case of distributed point values. Thus, it is
interpolated in exactly the same manner.
Groundwater
249
Saturated Groundwater Flow
The case of .dfs2 files is in fact two separate cases. If the .dfs2 file is
aligned with the model grid then the cell value that is assigned is calculated using the bilinear method with the 4 nearest points to the centre of
the cell. If the .dfs2 file is not aligned with the model grid then the file is
treated exactly the same as if it were a .shp file or a set of distributed point
values.
The geologic model is interpolated to the model grid during preprocessing, by a 2 step process.
1 The horizontal geologic distribution is interpolated to the horizontal model grid. If Geologic Units are specified then the integer grid
codes are used to interpret the geologic distribution of the model grid.
If distributed parameters are specified then the individual parameters
are interpolated to the horizontal model grid as outlined above.
2 The vertical geologic distribution is interpolated to the vertical
model grid. In each horizontal model grid cell, the vertical geologic
model is scanned downwards and the soil properties are assigned to the
cell based on the average of the values found in the cell weighted by
the thickness of each of the zones present. Thus, for example, if there
were 3 different geologic layers in a model cell each with a different
Specific Yield, then the Specific Yield of the model cell would be
S y1 ⋅ z 1 + S y2 ⋅ z 2 + S y3 ⋅ z 3
S y = ------------------------------------------------------------z 1 + z 2 + z3
(10.1)
where z is the thickness of the geologic layer within the numerical cell.
Conductivity values
Hydraulic conductivity is a special parameter because it can vary by many
orders of magnitude over a space of a only few meters or even centimeters. This necessitates some special interpolation strategies.
Horizontal Interpolation - The horizontal interpolation of hydraulic conductivity interpolates the raw data values. Thus, in Step 1 above, when
interpolating point values that range over several orders of magnitude,
such as hydraulic conductivity, the interpolation methods will strongly
weight the larger values. That is, small values will be completely overshadowed by the large values.
In fact, the interpolation in this case should be done on the logarithm of
the value and then the cell values recalculated. Until this option is
available in the user interface, you should interpolate conductivities
outside of MIKE SHE using, for example, Surfer. Alternatively, the
250
MIKE SHE
Conceptualization of the Saturated Zone Geology
point values could be input as logarithmic values and the Grid Calculator Tool in the MIKE SHE Toolbox can be used to convert the logarithmic values in the .dfs2 file to conductivity values.
Vertical Interpolation - In Step 2 above, the geologic model is scanned
down and interpreted to the model cell. Although, horizontal conductivity can vary by several orders of magnitude in the different geologic
layers that are found in a model cell, the water will flow horizontally
based on the highest transmissivity. Thus, the averaging of horizontal
conductivity can be down the same as in the example for Specific Yield
above. Vertical flow, however, depends mostly on the lowest hydraulic
conductivity in the geologic layers present in the model cell. In this
case a harmonic weighted mean is used instead. For a 3 layer geologic
model in one model cell, the vertical conductivity would be calculated
by
z 1 + z 2 + z3
K z = ------------------------------------z1
z2
z3
-------- + -------- + -------K z1 K z2 K z3
(10.2)
where z is the thickness of the geologic layer within the numerical cell.
10.1.1
Lenses
In building a geologic model, it is typical to find discontinuous layers and
lenses within the geologic units. The MIKE SHE setup editor allows you
to specify such units - again independent of the numerical model grid.
Lenses are also a very useful way to define a complex geology. In this
case, the lenses are used to define the subsurface geologic units within a
larger regional geologic unit.
Lenses are specified by defining either a .dfs grid file or a polygon .shp
file for the extents of the lenses. The .shp file can contain any number of
polygons, but the user interface does not use the polygon names to distinguish the polygons. If you need to specify several lenses, you can use a
single file with many polygons and specify distributed property values, or
you can specify multiple individual polygon files, each with unique property values.
In the case of lenses, an extra step is added to the beginning of the 2-step
process outlined in the previous section. The location of the lenses is first
interpolated to the horizontal numerical grid. Then the lenses become
essentially extra geologic layers in the columns that contain lenses. How-
Groundwater
251
Saturated Groundwater Flow
ever, there are a number of special considerations when working with
lenses in the geologic model.
z
Lenses override layers - That is, if a lense has been specified then the
lense properties take precedence over the layer properties and a new
geologic layer is added in the vertical column.
z
Vertically overlapping lenses share the overlap - If the bottom of
lense is below the top of the lense beneath, then the lenses are assumed
to meet in the middle of the overlapping area.
z
Small lenses override larger lenses - If a small lense is completely
contained within a larger lense the smaller lense dominates in the location where the small lense is present.
z
Negative or zero thicknesses are ignored - If the bottom of the lense
intersects the top of the lense, the thickness is zero or negative and the
lense is assumed not to exist in this area.
10.2
Numerical Layers
10.2.1
Specific Yield of the upper SZ numerical layer
The specified value for specific yield is not used for the specific yield of
the upper most SZ numerical layer if UZ is included in the simulation.
By definition, the specific yield is the amount of water release from storage when the water table falls. The field capacity of a soil is the remaining
water content after a period of free drainage. Thus, specific yield is equal
to the saturated water content minus the field capacity.
To avoid water balance errors at the interface between the SZ and UZ
models, the specific yield of the top SZ layer is set equal to the he saturated water content minus the field capacity. The value is determined once
at the beginning of the simulation. The water content parameters are taken
from the UZ layer in which the initial SZ water table is located.
In principle, having different values between the SZ and UZ models does
not directly cause a water balance error, but it may cause numerical problems that could lead to water balance errors. By definition, the steady-state
water table location will be identical in both the SZ and UZ models.
Pumping from the SZ will lower the SZ water table by an amount equal to
the specific yield divided by the cell area times the pumping rate. However, if the field capacity is not correlated to the specific yield, then the
amount of water released from storage in the UZ will be more or less than
the amount extracted from the SZ cell. This will result in different water
252
MIKE SHE
Groundwater Drainage
tables in the SZ and UZ models. If pumping stops, the system will again
reach an equilibrium with the same water table in both the SZ and UZ simply because of the pressure head redistribution.
As mentioned, the upper Sy value is calculated only at the beginning of
the simulation based on the UZ layer in which the initial SZ water table is
located. If the soil profile has multiple soil types with different field
capacities and saturated water contents, then the specific yield in the SZ
and UZ model may diverge during the simulation. With slowly moving
water tables, the differences may not be that large and the errors generated
will likely be tolerable. If the water table drops into a lower SZ layer, then
the specified Sy will be used. ,
The actual value used in the model is displayed in the pre-processed tab
under Specific Yield.
10.2.2
SZ Boundary Conditions
The upper boundary of the top layer is always either the infiltration/exfiltration boundary, which in MIKE SHE is calculated by the unsaturated
zone component or a specified fraction of the precipitation if the unsaturated zone component is excluded from the simulation.
The lower boundary of the bottom layer is always considered as impermeable.
In MIKE SHE, the rest of the boundary conditions can be divided into two
types: Internal and Outer. If the boundary is an outer boundary then it is
defined on the boundary of the model domain. Internal boundaries, on the
10.3
Groundwater Drainage
Saturated zone drainage is a special boundary condition in MIKE SHE
used to defined natural and artificial drainage systems that cannot be
defined in MIKE 11. It can also be used to simulate simple overland flow,
if the overland flow system can be conceptualized as a shallow drainage
network connected to the groundwater table - for example, on a flood
plain.
Saturated zone drainage is removed from the layer of the Saturated Zone
model containing the drain level. Water that is removed from the saturated
zone by drains is routed to local surface water bodies, local topographic
depressions, or out of the model.
Groundwater
253
Saturated Groundwater Flow
When water is removed from a drain, it is immediately moved to the recipient. In other words, the drain module assumes that the time step is longer
than the time required for the drainage water to move to the recipient. This
is the same as a “full pipe”. That is, water added to the end of a full pipe of
water causes an equal amount of water to immediately flow out the opposite end - regardless of the length of the pipe.
Drain flow is simulated using an simple linear reservoir formula. Each cell
requires a drain level and a time constant (leakage factor). Both drain levels and time constants can be spatially defined. A typical drainage level is
1m below the ground surface and a typical time constant is between 1e-6
and 1 e-7 1/s.
Drainage reference system
MIKE SHE also requires a reference system for linking the drainage to a
recipient node or cell. The recipient can be a MIKE 11 river node, another
SZ grid cell, or a model boundary.
There are four different options for setting up the drainage source-recipient reference system
Drainage routed downhill based on adjacent drain levels
This option was originally the only option in MIKE SHE. The reference
system is created automatically by the pre-processor using the slope of the
drains calculated from the drainage levels in each cell.
Thus, the pre-processer calculates the drainage source-recipient reference
system by
1 looking at each cell in turn and then
2 look for the neighbouring cell with the lowest drain level.
3 If this cell is an outer boundary cell or contains a river link, the search
stops.
254
MIKE SHE
Groundwater Drainage
4 If this cell does not contain a boundary or river link, then the search is
repeated with the next downstream neighbour until either a local minimum is found or a boundary cell or river link is found.
The result of the above search for each cell is used to build the sourcerecipient reference system.
If local depressions in the drainage levels exist, the SZ nodes in these
depressions may become the recipients for a number of drain flow producing nodes. This often results in the creation of a small lake at such local
depressions. If overland flow is simulated, then the ponded drainage water
will become part of the local overland flow system.
If the drain level equals the topography, drainage will be turn off in
that cell. Likewise, drain levels above the topography are not allowed. In
this case a warning will be written to the PP_Print.log and the drain level
will be automatically adjusted to a value just below the topography.
The drain level method is not allowed when using Time varying drainage
parameters (V.1 p. 320) because the source-recipient reference system is
only calculated once at the beginning of the simulation.
The drain-slope based reference system has been used in MIKE SHE for
many years and works well in most situations. However, when MIKE
SHE is applied where there is very little surface topographic relief, it is
often difficult to establish a suitable reference system based on the surface
topography/drain slopes. For example, often it is assumed that the drains
are located 50 to 100 cm below the terrain. In flat areas, this may generate
many undesired local depressions, which may receive drainage water from
a large area, thus generating lakes in places where there should not be a
lake.
If the drain level is perfectly flat, drainage is turned off. In other
words, if the drain-slope method cannot find a downhill neighbour
because all the neighbours have the same elevation as the cell, the drain
slope method assumes that the cell is a local depression. However, the
depression has no sources of drainage except itself. Thus, the drainage
function is effectively turned off.
Tip: MIKE SHE considers a grid point to be a local depression even if the
drainage level in the four surrounding model grids is only 1 mm higher.
The only way to avoid such problems is to create a drain level map that
does not contain “wrong” local depressions. For large models this may be
difficult and time consuming. In this case, one of the other drainage
options may be better.
Groundwater
255
Saturated Groundwater Flow
Remember, the drainage is routed to a destination. It does not phyisically
flow downhill. The drain levels are only used to build the drainage sourcerecipient reference system, and to calculate the amount of drainage.
Drainage routing based on grid codes
This method is often used when the topography is very flat, which can
result in artificial depressions, or when the drainage system is very well
defined, such as in agricultural applications.
In this method, the drainage levels and the time constants are defined as in
the previous method and the amount of drainage is calculated based on the
drain levels and the time constant.
If the drainage routing is specified by Drain Codes, a grid code map is
required that is used to restrict the search area for the source-recipient reference system. In this case, the pre-processer calculates the reference system within each grid code zone, such that all drainage generated within
one zone is routed to recipient nodes with the same drain code value.
When building the reference system, the pre-processor looks at each cell
and then
1 looks for the nearest cell with a river link with the same grid code
value,
2 if there is no cells with river links, then it looks for the nearest outer
boundary cell with the same grid code,
3 if there are no cells with outer boundary conditions, then it looks for
the cell with the same grid code value that has the lowest drain level. In
this case, the reference system is calculated as if it was based on Drain
Levels (see previous section).
The result of the above search for each cell is used to build the sourcerecipient reference system.
The above search algorithm is valid for all positive Drain Code values.
However, all cells where
Drain Code = 0 - will not produce any drain flow and will not receive any
drain flow, and
Drain Code < 0 (negative) - will not drain to river links, but will start at
Step 2 above and only drain to either a outer boundary or the lowest
drain level.
256
MIKE SHE
Groundwater Drainage
Tip: One method that is often used is to specify only one Drain Code
value for the entire model area (e.g. Drain Code = 1). Thus, all nodes can
drain and any drain flow is routed to the nearest river link. If there are no
rivers, the drain flow will be routed to the nearest boundary. If you want to
route all drain flow to the boundaries instead of the rivers, a negative drain
code can be specified for the entire area (e.g. Drain Code = -1).
Distributed drainage options
Choosing this method, adds the Option Distribution item to the data tree.
With the Option Distribution, you can specify an integer grid code distribution that can be used to specify different drainage options in different
areas of your model.
Code = 1 - In grid cells with a value of 1, the drainage reference system is
calculated based on the Drain Levels.
Code =2 - In grid cells with a value of 2, the drainage reference system is
calculated based the Drain Codes.
Code = 3 - Drainage in grid cells with a value of 3 is routed to a specified
MIKE 11 branch and chainage. At the moment, this options requires
the use of Extra Parameters (V.2 p. 193) and is described in SZ Drainage to Specified MIKE 11 H-points (V.1 p. 316).
Code = 4 - Drainage in grid cells with a value of 4 is routed to a specified
MOUSE man hole. At the moment, this options requires the use of
Extra Parameters (V.2 p. 193) and is described in the section Using
MIKE SHE with MIKE URBAN (V.1 p. 229).
Drain flow not routed, by removed from model
The fourth option is simply a head dependent boundary that removes the
drainage water from the model. This method does not involve routing and
is exactly the same as the MODFLOW Drain boundary.
Groundwater
257
Saturated Groundwater Flow
Drain Code Example
z
The grid cells with Drain Code 3 drain to a local depression since no
boundary or river link is found adjacent to a grid with the same drain
code.
z
The grid cells with Drain Code 1 or 2 drain to nearest river link located
adjacent to a grid with the same drain code.
z
The grid cells with drain code 0 do not contain drains and thus no
drainage is produced.
z
The grid cells with Drain Code -1 drains to local depression since no
boundary is found adjacent to a grid with the same drain code.
z
The grid cells with Drain Code -2 drains to nearest boundary grid with
the same drain code.
The Pre-processed Drainage Reference System
During the preprocessing, each active drain cell is mapped to a recipient
cell. Then, whenever drainage is generated in a cell, the drain water will
always be moved to the same recipient cell. The drainage source-recipient
reference system is displayed in the following two grids in the Pre-processed tab, under the Saturated Zone:
Drain Codes - The value in the pre-processed Drain Codes map reflects
the Option Distribution specified. For example, those cells with an Option
Distribution equal to 1 (Drainage routed based on Drain Levels) will have
258
MIKE SHE
Groundwater Drainage
a pre-processed Drain Code equal to 0, because the Drain Codes are not
being used for those cells.
Drainage to local depressions and boundary - This grid displays all the
cells that drain to local depressions or to the outer boundaries. All drainage from cells with the same negative value are drained to the cell with the
corresponding positive code. If there is no corresponding positive code,
then that cell drains to the outer boundary, and the water is simply
removed from the model. Cells with a delete value either do not generate
drainage, or they drain to a river link.
Drainage to river - This grid displays the river link number that the cell
drains to. Adjacent to the river links, the cells are labeled with negative
numbers to facilitate the interpretation of flow from cells to river links.
Thus, in principle, all drainage from cells with the same positive code are
drained to the cell with the corresponding negative code.
However, this is slightly too simple because the cells actually drain
directly to the river links. In complex river systems, when the river
branches are close together, you can easily have cells connected to multiple branches on different sides. In this case, the river link numbers along
the river may not reflect the drainage-river link reference used in the
model.
If you want to see the actual river links used in all cells, you can use the
Extra Parameter, SZ Drainage River Link Reference Table (V.1 p. 321), to
generate a table of all the river link-cell references in the PP_Print.log file.
Cells with a value of zero either do not generate drainage, or they drain to
a the outer boundary or a local depression.
10.3.1
Saturated Zone drainage + Multi-cell Overland Flow
The topography is often used to define the SZ drainage network. Thus, a
refined topography more accurately reflects the SZ drainage network.
The SZ drainage function uses a drain level and drain time constant. The
drain level defines the depth at which the water starts to drain. Typically,
this is set to some value below the topography to represent the depth of
surface drainage features below the average topography. This depth
should probably be much smaller if the topography is more finely defined
in the sub-grid model. The drain time constant reflects the density of the
drainage network. If there are a lot of drainage features in a cell then the
time constant is higher and vice versa.
Groundwater
259
Saturated Groundwater Flow
When using the multi-cell OL, the drainage system is updated in the sense
that the drain level will be defined using the sub-scale topography information. The SZ drainage will include the following when using sub-scale:
z
Multi-scale SZ drainage supported only in the PCG transient SZ solver
z
Each sub-grid cell will have the same drain time constant defined by
the value in the coarse grid.
z
If the drain level is defined as an elevation, then all sub-grids will have
the same drain level.
z
If the drain level is defined by depth below the surface, then each subgrid may have a unique drain level, since each sub-grid can have a different "Each coarse grid cell has a water table that is common for all
fine scale grids within the coarse grid.
z
If the coarse cell water table is above the fine scale drain level, then
drainage is calculated based on the drain time constant and the depth of
water above the fine scale drain level.
z
Total drainage in a coarse cell is the sum of all the fine scale drainage
volumes.
z
Drainage routing by levels will be determined by the coarse grid. However to make it more realistic with respect to the fine scale hydrology,
the drainage routing by levels will be based on the lowest drain level in
a coarse cell.
z
Drainage to local depressions will be added to the SZ cell, and resultant ponding will then follow the multi-scale OL flow.
Internal validation of the drainage scheme
MIKE SHE performs an internal validation of the SZ drainage scheme.
The following are used in connection with the sub-scale feature:
z
Drainage depths of zero are allowed and drainage depths above the
topography are set to the topography. This allows drain levels at the
ground surface. This check will be done on the coarse grid. That is, if
the coarse grid drain level is above the coarse grid topography, a warning will be issued and all the sub-grid drain depths will be set to zero.
Note for Release 2011 In Release 2011 and prior releases, a drain level
of zero turned off SZ drainage, and drain levels above topography were
260
MIKE SHE
Groundwater Drainage
set to the topography (and turned off drainage). For backwards compatibility an Extra Parameter is available.
Parameter
Name
Type
Value
disable drains at
or above ground
Boolean
On
Drain levels vs River link elevations There is an optional Extra
Parameter check in the drainage routing by levels that checks on the
river link bottom elevation.
Parameter
Name
Type
check drain level Boolean
against bed level
Value
On
If the river link bottom elevation is higher than the drain level, the cell
becomes a local depression. However, this will likely create a lot of
local depressions beside the rivers.
When using the multi-grid OL option, the drainage in a coarse cell is
controlled by the minimum drainage level in the cell. If one sub-grid
cell has a drainage level below the bed level then the drainage in the
entire cell is transferred to an internal depression.
Note for Release 2011 The check was originally added to prevent the
"lifting" of drainage water up to a river link. However, in most cases,
such lifting is probably unintentional. That is, the river bed has been
poorly interpolated. Prior to Release 2012, this was the default behaviour and the check above has been added for backwards compatability.
z
Groundwater
There is a check on the drain levels below the bottom of the model. If
the coarse grid drain level is below the coarse grid bottom of the
model, then a warning will be printed and the drain level will be
adjusted to the bottom of the model. In the sub-grids, you may have the
situation where the sub-grid drain level is below the bottom of the
model, but the average drain level is above. In this case, the sub-grid
drain level will be the maximum elevation of the bottom of the model
and the drain level. Meaning if the drain level of a sub-grid is below
the bottom of the model, the drain level is adjusted to the maximum
value of i) the bottom of the model and ii) the drainage elevation.
261
Saturated Groundwater Flow
Disabling Multi-Cell Drainage
By default, when if the multi-cell OL option is invoked, multi-cell drainage will be active. If you want to disable multi-cell drainage, perhaps for
backwards compatability with older models, an Extra parameter option is
availble to switch off multi-cell drainage: .
Parameter
Name
Type
disable multi-cell Boolean
drainage
Value
On
If this option is used, then the multi-cell drainage is switched off and the
drainage will function using the groundwater level and drain level based
on the course cells.
10.4
MIKE SHE versus MODFLOW
The MIKE SHE can be used to simulate all of the processes in the land
phase of the hydrologic cycle, including overland flow, channel flow,
groundwater flow in the unsaturated zone and saturated groundwater flow.
MODFLOW, on the other hand, is restricted to simulating flow only in the
saturated groundwater zone. Although many of the processes simulated in
MIKE SHE are used in a similar way when simulating groundwater flow
with MODFLOW, they are not actually “simulated” by MODFLOW.
Let’s take groundwater recharge as an example. MODFLOW allows you
to include recharge as an upper boundary condition to the groundwater
model, where recharge is defined as the amount of water reaching the
groundwater table after accounting for evapotranspiration, surface runoff
and changing storage in the unsaturated zone. In MODFLOW, the modeller has to account for these processes herself - usually by applying a constant rule-of-thumb fraction to the measured precipitation data. In most
cases, the model results are very sensitive to this fraction and since the
modeller has little data on this fraction, she will assume an initial value
and use this parameter as a calibration parameter. Thus, she will adjust the
amount of recharge during the calibration process until the measured
groundwater levels match the calculated values.
However, the fraction of precipitation reaching the groundwater table is
constant in neither space nor time. The actual amount of precipitation
reaching the groundwater table depends strongly on the maximum rate of
infiltration, which is a characteristic of the soil and will vary spatially over
262
MIKE SHE
MIKE SHE versus MODFLOW
the model domain. Further, since the maximum rate occurs when the soil
is saturated, different amounts of water will infiltrate during wet periods
compared to dry periods. To complicate matters further, the length of the
preceding dry period will determine the amount of available storage in the
unsaturated zone. For example, if there has been a long dry summer
period, then evapotranspiration may have created a large deficit of water
in the unsaturated zone that must be satisfied before any water reaches the
water table.
This example shows that infiltration of precipitation is a very dynamic
process. It depends on a complex interaction between precipitation,
unsaturated zone soil properties and the current soil moisture content, as
well as vegetation properties.
In MIKE SHE, the saturated zone is only one component of an integrated
groundwater/surface water model. The saturated zone interacts with all of
the other components - overland flow, unsaturated flow, channel flow, and
evapotranspiration.
In comparison, MODFLOW only simulates the saturated flow. All of the
other components are either ignored (e.g. overland flow) or are simple
boundary conditions for the saturated zone (e.g. evapotranspiration).
On the other hand, there are very few difference between the MIKE SHE
Saturated Zone numerical engine and MODFLOW. In fact, they share the
same PCG solver. The differences that are present are limited to differences in the discretisation and to some differences in the way boundary
conditions are defined.
Setting up the saturated zone hydraulic model involves defining the:
z
the geological model,
z
the vertical numerical discretisation,
z
the initial conditions, and
z
the boundary conditions.
In the MIKE SHE GUI, the geological model and the vertical discretisation are essentially independent, while the initial conditions are defined as
a property of the numerical layer. Similarly, subsurface boundary conditions are defined based on the numerical layers, while surface boundary
conditions such as wells, drains and rivers (using MIKE 11) are defined
independently of the subsurface numerical layers.
Groundwater
263
Saturated Groundwater Flow
The use of grid independent geology and boundary conditions provides a
great deal of flexibility in the development of the saturated zone model.
Thus the same geological model and many of the boundary conditions can
be re-used for different model discretisation and different model areas.
10.4.1
Importing a MODFLOW 96 or MODFLOW 2000 Model
A FORTRAN executable is automatically installed with MIKE SHE and
located in the MIKE SHE bin directory. The program can be used to read a
MODFLOW file set and extract the stationary distributed data to a set of
point theme shape files. The shp files can then be used directly in MIKE
SHE.
To extract data from a MODFLOW model, open a command prompt in the
directory containing the input files. On the command prompt line, type
MShe_ModflowExtraction.exe file_name.pfs
The extraction will proceed silently - that is without any messages. To run
the extraction with the messages, you need to use
MZLaunch file_name.pfs -e
MShe_ModflowExtraction.exe
which will start the MZLaunch utility. The file_name.pfs variable is
the input file for the MODFLOW extractor. The input file has the standard
MIKEZero Pfs format. The input fields of the file are explained below.
Lines starting with '//' are not read, but rather can be used as comment
lines.
Table 10.1 is an example .pfs file for the MODFLOW data extractor program:
Table 10.1
264
MODFLOW Extraction.pfs file format and description
Line item
Comment
[MIKESHE_ModflowExtraction]
FileVersion = 3
File version 3 is for Release 2009 and
up
ModflowModel = 'MODFLOW-96'
\\ModflowModel = = 'MODFLOW-2000'
The ModflowModel variable should
be changed to MODLFOW-2000, if
the MODFLOW model is a MODFLOW 2000 model.
MIKE SHE
MIKE SHE versus MODFLOW
Table 10.1
MODFLOW Extraction.pfs file format and description
Line item
Comment
NameFileName = |.\Airport5.nam|
The NameFileName is the name of
the MODFLOW name file that contains all of the references to the other
input files.
The '|' around the name-file name and
the path of the specified file name
must be relative to the location of the
pfs file.
XMin = 300.
YMin = 400.
XMax = 3032.
YMax = 1132
The minimum and maximum (X,Y)
coordinates are used to determine the
exact spatial coordinates of the nodal
points.
XMin and YMin are the UTM coordinates of lower left MODFLOW corner.
Xmax and Ymax are the UTM coordinates of the upper right MODFLOW
corner.
See figure next page.
TimeUnit = 'DAYS'
The TimeUnit is not currently used,
but must be input.
Valid values for TimeUnit are DAYS,
HOURS, MINUTES and SECONDS.
LengthUnit = 'METER'
The LengthUnit is not currently used,
but must be input.
Valid values for LengthUnit are
METER and FEET.
StartDate = 2005,1,1,0,0
The start date and time of the MODFLOW simulation.
Format: YYYY, MM, DD, HH, MM
WellExtraction = 1
Extract well data to a dfs0 file.
On: Flag = 1
Off: Flag = 0
RechargeExtraction = 1
Extract recharge input to a dfs2 file.
On: Flag = 1
Off: Flag = 0
Note: only works with uniform MODFLOW grids.
Groundwater
265
Saturated Groundwater Flow
Table 10.1
MODFLOW Extraction.pfs file format and description
Line item
Comment
HeadExtraction = 1
Extract head results to a dfs2 file.
On: Flag = 1
Off: Flag = 0
Note: only works with uniform MODFLOW grids
EndSect // MIKESHE_ModflowExtraction
Note MODFLOW does not have any internal unit checking. The units
written in the MODFLOW file are only for display purposes. Also, the
units that you define in your MODFLOW user interface may not be the
same as those written to the MODFLOW files. So, you need to be careful
of units and know what units the MODFLOW files are written in.
The MODFLOW name file has the usual MODFLOW format. However,
you should
z
Specify a new name for the LIST file in order not to overwrite the
LIST file of an existing simulation, and
z
Make copies of, or rename, all output files (lines starting with DATA).
Existing result files might otherwise be overwritten during the execution of the extraction routine.
The coordinate information is the UTM coordinates of the lower left and
upper right MODFLOW model corners - not the MODFLOW block-centered nodal coordinates.
Xmax, Ymax
MODFLOW
Model
Grid
UTM Map
266
Xmin, Ymin
MIKE SHE
MIKE SHE versus MODFLOW
These coordinates plus the DELR, DELC vectors from the MODFLOW
files are used to defined the spatial location of the shape file and dfs2 output.
For a MODFLOW model, the extraction routine reads and outputs the following MODFLOW static parameters to a point theme shape file:
Top, Bot, Shead, Tran, Hy, Vcont, Sf1, and Sf2
Plus, it outputs the Specific storage, which is calculated as Sf1 divided by
the layer thickness.
If the well output option is selected, a dfs0 file will be created. In this file,
every cell in the MODFLOW file containing a well will have a seperate
item in the dfs0 file.
If the recharge data and head results is selected, a dfs2 file will be created
for each of these. However, the dfs2 format does not allow for variable
grid spacing, which means that variable grid spacing will be ignored. The
DELR and DELC for the first column and row will be used as the grid
spacing in the dfs2 file. Thus, the recharge and head results output option
is really only useful for MODFLOW models with a uniform grid spacing.
The extraction routine outputs point theme shape files -one file per data
type - with one item for each extracted layer. The shape file names reflect
the MODFLOW manual naming convention (Top.shp, Vcont.shp, etc.).
The points represent the centre of each grid square. The model orientation
is calculated from the user-specified coordinates of lower left (origin) and
upper right corner of the model.
To use the MODFLOW data in MIKE SHE, select the Point/Line .shp
option for the static variable. Then browse to the appropriate .shp file. The
.shp file will contain one item for each model layer in the MODFLOW
model. The appropriate item is selected in the file browse dialogue. Once
the file has been assigned, MIKE SHE will automatically interpolate the
data to the model grid.
Internal inactive zones
Currently, it is not possible to extract the inactive zones from the MODFLOW model and convert these to inactive cells in MIKE SHE. MODFLOW and MIKE SHE treat internal inactive zones quite differently. In
MIKE SHE, the internal inactive zones are simply treated as cells with a
very low hydraulic conductivity, whereas, MODFLOW ignores them in
the solution. Furthermore, the extraction program only writes points to the
.shp file for the active nodes. Thus, when it comes to the interpolation in
Groundwater
267
Saturated Groundwater Flow
MIKE SHE, the interpolation does not know about the inactive zone and
interpolates through the inactive zone - there are simply no data points in
the inactive zones.
268
MIKE SHE
WATER QUALITY
269
270
MIKE SHE
11
SOLUTE TRANSPORT
The complete MIKE SHE advection-dispersion (AD) module is comprised of four independent components, each describing the transport
processes in one of the parts of the hydrological cycle. Used in combination they describe solute transport in the entire hydrological cycle. The
four components are:
z
Overland Transport
z
Channel Transport (MIKE 11)
z
Unsaturated Zone Transport
z
Groundwater Transport
A number of processes relevant for simulating reactive solute transport are
included in MIKE SHE including
z
Water and solute transport in macro pores,
z
Sorption of solutes described by either equilibrium sorption isotherms
(Linear, Freundlich or Langmuir) or kinetic sorption isotherms, which
include effects of hysteresis in the sorption process,
z
Attenuation of solutes described by an exponential decay, and
z
Plant uptake of solutes.
Current Limitations
The solute transport module in MIKE SHE currently does not support
z
exchange from MIKE 11 to Overland flow,
z
any solute transfer via irrigation,
z
any solute transfer via flood codes, and
z
solute migration from UZ to OL.
In the first three cases, the solutes will remain in the source location and
only the water will be transfered. This will lead to increasing concentrations at the source.
In the last case, there is no mechanism in MIKE SHE to transfer water
from UZ to OL, so there is also no means to move solutes from the UZ
cells onto the ground surface. This has implications for salinity modelling,
as there is no way for runoff to remove surface salts that migrate upwards
due to capillarity and concentrate on the ground surface due to evaporation.
271
Solute Transport
11.1
Flow Storing Requirements
Solute transport calculations in MIKE SHE AD are based on the water
fluxes from a MIKE SHE Water Movement (WM) simulation. To ensure
that all the needed WM result data types are stored, you have to specify
that results should be stored for an AD simulation. See Storing of Results
(V.2 p. 183).
The WM data should be stored frequently enough to describe the dynamics of the flow. The selected storing frequencies of flow results will usually be a compromise between limitations in disk space and resolution of
the flow dynamics. The maximum computational time steps in a transport
simulation are often restricted by advective and dispersive stability criterions. However, the transport time step cannot be greater than the flow storing time step in each component.
11.2
Storing of Results
The simulated concentration distribution in each component as well as the
mass balances and fluxes will be stored in dfs2 and dfs3 files with different time steps. Besides these result files, the program also writes output to
the error log, which describes errors encountered during execution and a
print log which contains execution step information, statistics on the run
and a mass balance (if requested).
Normally, the results from the saturated zone (species concentration in
each grid) is by far the most disk consuming parameter. So, be careful
with the storing time step. Mass balances, which includes time series of
mass storage and fluxes between components (and sources, drains, boundaries etc.) can be stored at smaller time steps.
When you select the time step you should also be aware of the time scale
of the process. The time scale for transport processes in groundwater is
usually much larger than the time scale for transport in a river.
Enter the desired time steps - notice that the unit is hours - in each of the
edit fields. There are no limitations on this time step but if you select a
time step less than the simulation time step, the storing time step will be
the new simulation time step.
272
MIKE SHE
Simulation and Time Step Control
11.3
Simulation and Time Step Control
Simulation time steps are to some extent controlled by the user. Several
possibilities for time step control exist to make the execution as fast as
possible with no numerical dispersion and instabilities.
The first possibility for controlling the simulation time steps in the different components is simply to define the maximum time step in each component. Note that time steps should be given in increasing order i.e.
dtRIVER ≤ dtOVERLAND ≤ dtUZ ≤ dtSZ. Also note that this is the MAXIMUM
time step. That is, the actual simulation time step is controlled by the stability criterions with respect to advective and dispersive transport given
below. Furthermore, time steps for transport cannot exceed the storing
time step for the relevant data in the flow result file from a MIKE SHE
flow simulation.
Enter the maximum allowable Courant number for each component. The
Courant number is defined by V x dt/dx (velocity times time step divided
by “grid size”). This number should normally not exceed 1.0 for one- and
two-dimensional transport (UZ, Overland and Channel Flow) and 0.8 for
three-dimensional transport (SZ). The maximum time step will be calculated accordingly.
Enter the maximum allowable dispersive Courant number for each component. The dispersive Courant number is defined by D x dt/dx2 (Dispersion coefficient times time step divided by “grid size” squared). This
number should normally not exceed 0.5. The maximum time step will be
calculated accordingly.
The transport limits are used to avoid negative concentrations in cases
with extreme gradients (e.g. close to sources) or in areas with highly irregular velocity fields. Enter the maximum allowable transport from a node
or grid as a fraction of the storage in the node or grid. A recommended
value for all components is 0.9, which ensures that this option is in use
(the value 0 determines that this option is not in use).
11.3.1
Calibrating and Verifying the Model
The advection-dispersion of solutes depends largely on the simulated
flows and fluxes calculated by the MIKE SHE flow model. After your first
AD simulations, you will usually have to go back and improve the calibration of your flow model. Rarely, can the simulated concentrations and
mass fluxes be calibrated to the measured concentrations by tuning only
the solute transport model.
273
Solute Transport
It is important to recognise that a transport model must be calibrated. This
is true for all applications larger than the laboratory scale since model output cannot necessarily be compared directly to measured values. Measurements are mostly point measurements at a certain time whereas results
often are mean values over larger volumes and longer times.
The purpose of the calibration is to tune the model so that it is able to
reproduce measured conditions for a particular period in a satisfactory
way. This period - known as the calibration period - should be chosen long
enough to include events of similar kind as the ones you are going to
investigate.
A satisfactory calibration is reached when the model is able to reproduce
the measured values taking the following conditions into account:
z
uncertainty in the measurements (time, space, equipment)
z
representativeness of measurements (point/average grid values)
z
differences between your conceptual model and nature
z
uncertainty in other model parameters and data (source description
etc.)
In general, it is impossible to specify an exact level of divergence between
measured data and computed results before the model is satisfactorily calibrated. In each application you have to consider all factors influencing
your result.
After the calibration, you should verify your model by running one or
more simulations for which measurements are available without changing
your model parameters. If the model is able to reproduce the validation
measurements you can consider your calibration to be successful. This
ensures that simulations can be made for any period similar to the calibration and the verification period with satisfactory results.
11.4
Executing MIKE SHE WQ
In the top icon bar, there is a three-button set of icons for running your
model.
.
WQ - The WQ button starts the Water Quality simulation. After you have
successfully run a water movement (WM) simulation to completion, you
can run a water quality simulation.
274
MIKE SHE
Output
In addition to the three icon buttons, there is a Run menu. In this menu,
you can check on and off all three of the above options. Finally, there is an
Execute... menu sub-item that runs only the checked items above it. The
Execute option can also be launched using the Alt - R - E hot-key
sequence.
MIKE SHE WQ can also be launched from a batch file with or without the
MZLaunch function. For more information on this, see Using Batch Files
(V.1 p. 164)
11.5
Output
The output of the MIKE SHE AD is stored to several dfs0, dfs2 and dfs3
files which can be viewed and processed with the different tools available
for these files in MIKE ZERO.
For each species, a concentration file is created for each hydrologic process - a dfs2 file for OL, and a dfs3 file each for UZ and SZ.
For each species, the total WQ mass balance is stored in two dfs0 files.
The xx_species_AllItems.dfs0 includes all of the possible water quality
mass balance items. The xx_species.dfs0 is a copy of the _AllItems.dfs0
file with all of the non-zero items removed.
The first 20 items in the _AllItems.dfs0 file define the global mass balance
(see Table 11.2). There are four items: Storage, Input, Output and Error.
Each of these is calculated for each of the five storage items: SZ,
Immob(SZ), UZ, MP(UZ), and OL.
The item Immob(SZ) is for solutes stored in the SZ matrix porosity when
the dual porosity option in SZ is turned on. In this case, water and solutes
move in the fractures and solutes diffuse into the rock matrix. The fractures are then the primary porosity.
The item MP(UZ) is for solutes stored in the UZ macropores when the
macropore option in UZ is turned on. In this case, water and solutes move
in both the matrix and the macropores, but the volume of water in the
macropores is generally much less than the volume of water in the matrix.
So, the primary porosity is the UZ matrix.
The Error is calculated for each of the five items as:
Error = Output – Input + ∆Storage
275
Solute Transport
However, the mass in the SZ and UZ items includes the mass in both the
primary and secondary porosities.
The Output, Input and Storage are all displayed as positive values in the
dfs0 file and the WQ log file. A positive change in storage denotes an
increase in mass.
Table 11.1
WQ mass balance items in the _AllItems.dfs0 file.
Mass balance item
Component
Storage
SZ, Immob(SZ), UZ, MP(UZ), OL
Input
SZ, Immob(SZ), UZ, MP(UZ), OL
Output
SZ, Immob(SZ), UZ, MP(UZ), OL
Error
SZ, Immob(SZ), UZ, MP(UZ), OL
The rest of the items in the _AllItems.dfs0 file are only non-zero if the
item is relevant for the current WQ simulation. Table 11.2 lists all of the
rest of the items in the _AllItems.dfs0 organized by the source of the solute.
Table 11.2
Available WQ mass balance items in the _AllItems.dfs0 file
From
To
Comment
Sources
SZ, UZ, OL, River
Note that sources can be specified as positive or
negative.
Ext. Sources (OpenMI)
SZ, UZ, OL
This is non-zero only if a model is linked to MIKE
SHE by OpenMI that adds mass to the component.
Ext. Input (OpenMI)
SZ Drain
This is non-zero only if a model is linked to MIKE
SHE by OpenMI that adds mass to the component.
However, this is a special case because you can
add mass directly to the SZ drain without it actually becoming part of the SZ model. The mass is
then added to the model at the location were the
drain discharges (i.e. river link, SZ boundary, or
local SZ depression)
Boundary
SZ, UZ, OL, River
The (Boundary to River) item is typically zero, but
can be non-zero in a couple of rare cases:
If you have
1) a fixed head SZ boundary, or
2) a fixed concentration boundary,
next to a river link, then mass from this cell to the
river will be added to (Boundary to River).
276
MIKE SHE
Output
Table 11.2
Available WQ mass balance items in the _AllItems.dfs0 file
From
To
Comment
Precip
UZ, OL
Mass from precipitation is always added to either
OL or UZ, even though precipitation can be added
to SZ in the WM module.
If you have a SZ-only simulation, then mass from
precipitation is included in the (Precip to OL) and
then (OL to SZ)
Decay
SZ, Immob(SZ), UZ, MP(UZ), OL
This is the mass that has decayed in each of the
processes
Sorp/DeSorp
SZ, Immob(SZ), UZ, MP(UZ), OL
This is the net Sorption and Desorption to the soil
matrix in each of the processes. If mass is sorbed
to the soil matrix it is removed from solution and
this value will be negative. If mass desorbs from
the soil matrix it is added to solution and this value
will be positive.
Colloid-Sorp/DeSorp
SZ, Immob(SZ), UZ, MP(UZ), OL
This is the net Sorption and Desorption to colloids
in each of the processes. If mass is sorbed to the
colloids it is removed from solution and this value
will be negative. If mass desorbs from the colloids
it is added to solution and this value will be positive.
However, this is normally zero, because colloid
transport is not available in the commercial version of MIKE SHE, only a research version.
EcoLab
SZ, Immob(SZ), UZ, MP(UZ), OL
This is the mass change resulting from passing solutes to and from EcoLab - positive if EcoLab
causes the mass to increase and negative if mass
decreases.
SZ
UZ, MP(UZ), OL, Sinks, Sources,
Ext.Sinks(OpenMI), Decay,
Sorp/DeSorp, Colloid-Sorp/DeSorp,
EcoLab, Plant Uptake
Note that there is no mass transfer to SZ drains
because the SZ drains have not storage. Mass that
discharges to SZ drains passes straight through the
drain and is added to the end recipient (i.e. a river,
boundary, or local depression.
Immob(SZ)
UZ, Decay, Sorp/DeSorp, ColloidSorp/DeSorp, EcoLab
SZ Baseflow
River
This is a special item that includes only the mass
from SZ to rivers.
SZ Drain
River, SZ (Local Dep), Boundary
This is a special item that divides up the SZ mass
discharge to drains by the end recipient.
SZ Flow
Boundary
This is a special item to distinguish between SZ
mass discharge to boundaries via drains and direct
discharge to the boundary.
SZ(Fract)
Immob(SZ)
This is a special item that includes only the mass
exchange between fractures and the matrix when
the dual porosity option is selected in the SZ.
277
Solute Transport
Table 11.2
Available WQ mass balance items in the _AllItems.dfs0 file
From
To
Comment
UZ
SZ, Immob(SZ), OL, Sinks, Sources,
Ext.Sinks(OpenMI), Boundary, Decay,
Sorp/DeSorp, Colloid-Sorp/DeSorp,
EcoLab, Plant Uptake
Note that the (UZ to Boundary) item refers to mass
discharge from UZ to SZ boundaries. This can
arise, for example, when a UZ column discharges
into an SZ cell that contains an internal boundary
condition, such as a constant head.
Note that the (UZ to Immob(SZ)) item will only be
non-zero when the dual porosity option in SZ is
turned on. In this case, as the water table increases,
there will be a transfer of UZ mass to water in both
the SZ fractures and SZ immobile matrix based on
the ratios of their porosities.
Related to the above, if macropores are active,
then mass in the UZ macropores will be distributed to both the SZ matrix and the SZ fractures.
MP(UZ)
SZ, Immob(SZ), UZ(Matr), Decay,
Sorp/DeSorp, Colloid-Sorp/DeSorp,
EcoLab
UZ(Matr)
MP(UZ)
OL
SZ, UZ, MP(UZ), River, Sinks,
Sources, Ext.Sinks(OpenMI), Boundary, Decay, Sorp/DeSorp, ColloidSorp/DeSorp, EcoLab
River
SZ, OL, Boundary
11.6
This is a special item that includes only the mass
exchange between macropores and the matrix
when the macropore option is selected in the UZ.
(River to OL) is always zero because this
exchange has not yet been implemented.
Coupling MIKE SHE and MIKE 11 WQ
Coupling the WQ modules between MIKE 11 and MIKE SHE is
described in the section: Coupling MIKE SHE Water Quality to MIKE 11
(V.1 p. 218).
278
MIKE SHE
Coupling MIKE SHE and MIKE 11 WQ
12
MIKE SHE + ECO LAB
ECO Lab is an open and generic equation solver. ECO Lab is mostly used
for water quality and ecosystem modeling, such as modelling eutrophication of lakes, calculating the fate and transport of heavy metals and determining ecology indicators (e.g. distribution of sea grass). Originally, ECO
Lab was developed as a tool to simulate water quality reactions in surface
water, but has been expanded to include agent based modeling and other
more complex reactions and components.
In MIKE SHE, ECO Lab can be used from basic water quality kinetic
reactions in surface and groundwater, to complex coupled feedback interactions between nutrients, plant growth and hydrology.
The initial implementation of ECO Lab in MIKE SHE depends on MIKE
SHE’s WQ module. That is, you must run a WQ simulation after the WM
simulation. When running MIKE SHE + ECO Lab, ECO Lab reads concentrations (state variables) from MIKE SHE’s WQ module, reads other
necessary input data files, generates additional output and passes modified
concentrations (state variables) back to MIKE SHE.
ECO Lab acts on a cell basis. That is, it is called for each cell in the model.
By default it is called at every time step in the MIKE SHE WQ simulation,
but optionally can be called less frequently.
Thus, using ECO Lab is a multi-step process, whereby you:
1 Create an ECO Lab template file that specifies the equations to be
solved, including all the forcing (spatially and time varying input),
constants (spatially varying constants), state variables (parameters calculated in MIKE SHE, i.e species concentrations), and derived outputs
(results).
2 Specify the name of the template file in the MIKE SHE WQ model.
3 Define the links between the template variables and the MIKE SHE
parameters.
4 Run the MIKE SHE Water Quality model. During the simulation,
MIKE SHE passes the State Variables to ECO Lab. ECO Lab updates
the State Variable values and passes them back to MIKE SHE. At the
same time, ECO Lab will write to any specified output files.
The output files are standard dfs2 or dfs3 output files. These files can be
used as input in subsequent WM or other WQ simulations, or viewed in
the Results Viewer, etc.
279
MIKE SHE + ECO Lab
12.1
ECO Lab Templates
An ECO Lab Template contains the mathematical description of the ordinary differential equations to be solved. These could, for example,
describe an ecosystem including the processes affecting the ecosystem.
An ECO Lab template contains six components:
z
State Variables - State variables represent the state that the user wants
to predict (at least one state variable must be specified).
z
Constants - Constants are used as arguments in the mathematical
expressions of processes in an ECO Lab model. They are constant in
time, but can vary in space.
z
Forcings - Forcings are used as arguments in the mathematical expression of processes in an ECO Lab model. They can vary in time and
space. They basically represent variables of an external nature that can
affect the ecosystem.
z
Auxiliary Variables - Auxiliary Variables are arguments in the mathematical expression in an ECO Lab model. They can be used as intermediate calculations that can include any state variable, constant or
forcing. They can also be used to specify results directly.
z
Processes - Processes describe the transformations that affect the state
variables. That means processes are used as arguments in the differential equations that ECO Lab solves to determine the state of the State
Variables.
z
Derived Outputs - Derived Outputs are output files that are derived
based on the model results.
Additional details on developing ECO Lab templates is available in the
MIKE ZERO ECO Lab manual.
Important Note: Units
All concentrations passed from MIKE SHE to ECO Lab are in units of
[g/m3], which is equivalent to [mg/L].
Thus all parameters and equations defined in the ECO Lab template must
reflect these units - either directly or via an appropriate scaling factor. For
example, the correct units for a decay rate constant might be [g/m3/day] or
[mg/L/day].
12.1.1
280
Developing a Template
Creating and developing an ECO Lab template involves several steps.
MIKE SHE
ECO Lab Templates
Create an ECO Lab template
First you must create an ECO Lab template from the File/New menu or
the New File icon. In both cases, you will chose MIKE Zero and then
ECO Lab (.ecolab) in the New File dialogue. This will create a new blank
ECO Lab template file.
Alternatively, you can copy and edit an existing ECO Lab template.
A few tips will be useful before you start.
z
You should try to keep the names of the Constants, Forcings, etc. as
short as practical. The names are used when defining Processes, Auxiliary Variables, and Derived Outputs.
z
The names used in the definitions are case-sensitive.
z
The names must be unique within the list of Constants, Forcings etc.
z
To add a new Constant, Forcing, etc. right click on the item and chose
the appropriate option.
Add State Variables
In the current coupling to MIKE SHE, the only available State Variables
are species concentrations. Thus, you must add one State Variable item for
each species in MIKE SHE that you want ECO Lab to modify during the
WQ simulation.
An ECO Lab template must include at least one State Variable.
The State Variable name must be exactly the same as the Species name in
MIKE SHE.
Dual domain mass transfer
The exception to the exact naming rule is when simulating dual domain
mass transfer. In this case, the State Variable name must use the reserved
suffix "_2" for the solute in the secondary porosity. For example, OXYGEN and OXYGEN_2 would be the State Variable names for the species
OXYGEN in MIKE SHE.
Add one or more Constants
Constants are spatially distributed values that are constant in time. Each
constant may or may not be User Defined.
User Defined = NO
If the Constant is a parameter that is pre-defined then the Constant is not
User Defined (i.e. User Defined = NO). If this is selected, a list of Con-
281
MIKE SHE + ECO Lab
stants is available in the combo box. However, the only Constant on this
list that is relevant for MIKE SHE is
“MIKE_SHE_SUPPLIED_CONSTANT”. All others will be ignored.
There are a limited number of Constants that can be passed directly from
MIKE SHE. These are mostly geometry related parameters, including the
grid area, the cell volume, the topography, and the depth to the top and
bottom of the cell. However, some specific Constants are also available
for the different domains, such as the Detention Storage in the Overland
flow, and the Bulk Density and the Porosity in the saturated zone.
User Defined = YES
If your Constant is not in the list of available pre-defined Constants from
MIKE SHE, then you must chose User Defined = YES. In this case, you
can define any spatially distributed, static parameter.
The actual values and spatial distribution of the Constant will be defined
in the MIKE SHE Setup Editor. In the template, the only thing you need to
specify is the name of the Constant. The name is then used when defining
the Processes, Derived Outputs, etc.
In some cases, the Constant may already be defined in the Setup Editor,
but only those in the list will be automatically passed to ECO Lab. So, if
your Constant is not available, then you will need to define it again in the
list of Constants in the Setup Editor.
Add one or more Forcings
A Forcing is any spatially distributed value that is time varying. You can
think of it as a value that is affecting the State Variable during the simula-
282
MIKE SHE
ECO Lab Templates
tion. For example, air temperature will affect the degradation rate of a solute.
Similar to the Constants, the Forcings can be User Defined = YES, or User
Defined = NO.
User Defined = YES
If the forcing is not user defined (“User Defined = NO”), then a long list
of available forcings are listed in a combo box. However, the only Forcing
on this list that is relevant for MIKE SHE is
“MIKE_SHE_SUPPLIED_FORCING”. All others will be ignored.
Note: The various other “MIKE_SHE_” forcings are used by MIKE 11 to
define concentrations and mass of solute entering the river.
For all forcings that are not user-defined, there will be a list of pre-defined
available forcings in the MIKE SHE Setup editor depending on the
domain.
User Defined = NO
If your Forcing is not in the list of available pre-defined Forcings from
MIKE SHE, then you must chose User Defined = YES. In this case, you
can define any spatially distributed, time-varying parameter.
The actual values and spatial distribution of the Forcing will be defined in
the MIKE SHE Setup Editor. In the template, the only thing you need to
specify is the name of the Forcing. The name is then used when defining
the Processes, Derived Outputs, etc.
In some cases, the Forcing may already be defined in the Setup Editor, but
only those in the list will be automatically passed to ECO Lab. So, if your
Forcing is not available, then you will need to define it again in the list of
Forcings in the Setup Editor.
Create Auxiliary Variables, Processes and Derived Outputs
Auxiliary variables are used for intermediate calculations and must be
defined as 3D (UZ and SZ) or 2D variables.
Processes transform a State Variable or calculate another result. Spatial
variation and type must be defined for each process. Each process can be
included in the results file by choosing "YES" in the "Output" box.
Derived Outputs allow the user to define output files based on the process
results.
283
MIKE SHE + ECO Lab
12.1.2
ECO Lab templates in MIKE SHE
ECO Lab is only available when Water Quality is selected. Thus, to be
able to use ECO Lab, the Water Quality option in the main Simulation
Specification dialogue must be selected.
Note also that ECO Lab will only work with the Finite Difference AD
method.
284
MIKE SHE
ECO Lab Templates
After activating the Water Quality module, the ECO Lab option must be
selected in the WQ Simulation Specification dialogue.
A new data tree branch will appear, where ECO Lab templates can be
specified for each of the hydrologic processes - Overland Flow, the
Unsaturated Zone, and the Saturated Zone. Separate templates are
required for each of these zones because the processes in each of these
domains are very different.
In the ECO Lab Template Specification dialogue, there is a checkbox for
each of the processes. These checkboxes are active if the Water Quality is
activated for the process in the WQ Simulation Specification dialogue
285
MIKE SHE + ECO Lab
.
When you enable ECO Lab for the specific process, you will be able to
browse to the required ECO Lab template. Specified templates can be
directly modified by clicking on the Edit button.
When you browse to a template, the template file is read by the Setup Editor and the number of components (i.e. State Variables, Forcings, Processes, etc.) that have been specified in the template are displayed in the
Template summary.
When calculating the concentrations (State Variables) for the next time
step, an explicit time-integration of the transport equations is made.
Depending on the desired accuracy of this numerical integration, you can
chose one of three different integration methods. The methods are in
increasing level of accuracy (and numerical effort), starting with the Euler
method and finishing with the Runge Kutta 5th Order method).
286
MIKE SHE
ECO Lab Templates
Finally, for each template you can specify an update frequency (see
above). The update frequency tells MIKE SHE how frequently to call
ECO Lab. If the water quality processes are slow relative to the simulation
time step, you can save considerable simulation time by calling ECO Lab
less frequently. For example, to call ECO Lab every second or third simulation time step, you would specify an Update frequency of 2 or 3.
12.1.3
State Variables in MIKE SHE
The State Variables defined in the ECO Lab template are passed to MIKE
SHE as Species. This is the only way to pass information from ECO Lab
to MIKE SHE.
The Species Name in MIKE SHE must be exactly the same as the State
Variable Name used in the ECO Lab Template (except in the case of dual
domain mass transfer, which uses the reserved suffix “_2”).
There are four species types in MIKE SHE. Species can be either:
z
Dissolved - Dissolved species are mobile in the water. They are
active in the subsurface and surface water. Disolved species
have a default concentration of [µg/m3].
z
Sorbed - Sorbed species are only available in the subsurface.
They are fixed to the soil matrix and do not move with the water.
Sorbed species have a default concentration of [g/g].
z
Suspended - Suspended species are only available in ponded
water. They do not infiltrate to the UZ or SZ, and they cannot
become Sorbed species. If the ponded water infiltrates, the species is left behind. Suspended species have a default concentration of [µg/m3].
z
Fixed - A fixed species is neither disolved or nor sorbed. It is
used as an immobile state variable by ECOLab. This allows
ECOLab to read and write arbitrary values to MIKE SHE during
the simulation. Fixed species have an undefined unit.
In particular, the Fixed species is especially interesting in MIKE SHE. It is
a species type without pre-defined units of concentration. The non-dimensional species cannot be transported with the flow and can be used as a
287
MIKE SHE + ECO Lab
book-keeping mechanism for resulting processes. ECO Lab itself can read
a value from any file, update the value and write it back to the file. However, ECO Lab cannot append a new value to a file. That is, ECO Lab cannot read a value, update the value and add the new value as a new timestep
in file. The Fixed species type gives you a mechanism for handling various auxiliary user defined quantities during the simulation. In other words,
ECO Lab can read the current Fixed species value, and return a new value
to MIKE SHE. MIKE SHE moves forward in time, and ECO Lab starts
over again. In the mean time, MIKE SHE has saved the value from the
previous time step. This mechanism greatly increases the flexibility of the
ECO Lab coupling in MIKE SHE. It allows you for example, to create
things like ecological indexes that map changing ecohydrologic conditions over time.
12.1.4
Specifying Constants and Forcings in MIKE SHE
When you browse to the ECO Lab template, the template is read by MIKE
SHE’s Setup Editor to find the MIKE_SHE_SUPPLIED_FORCINGS and
MIKE_SHE_SUPPLIED_CONTANTS used in the template. These are
separated into Built-in and User Specified Forcings and Constants. In both
cases, the Forcing or Constant is added to the appropriate list of Forcings
and Constants in the MIKE SHE data tree - separated into the different
domains (OL, UZ, and SZ).
Built-in Forcings and Constants
If the Forcing or Constant is not user defined (User Defined = “NO”), then
the Forcing or Constant is an internal value in MIKE SHE and will be
passed automatically to ECO Lab. The list of available parameters is quite
short - primarily geometry related (e.g. cell volume), plus a few domain
specific constants (e.g. porosity).
After selecting the parameter from the list of available parameters in the
combo box, select the units that are being used in the template. The list of
units is taken from the standard available units in the EUM database for
the particular item. The Constant or Forcing can be used in various equations in the template. However, there is no check on the units being used.
So, it is expected that the Forcing or Constant will used one of the options
288
MIKE SHE
ECO Lab Templates
from the list of units in the EUM database. Otherwise, and appropriate
conversion must be done in the template equation.
User Specified Forcings and Constants
If the Forcing or Constant is user defined (User Defined = “YES”), then
the Forcing or Constant must be specified explicitly in MIKE SHE.
In this case, there is nothing to specify on the main list of Forcings and
Constants, but a Forcing or Constant item is added to the data tree down
under the appropriate branch in the data tree. In this branch you will find a
table of user specified Forcings and Constants, plus individual sections for
each item. Each item can be spacially defined similarly to other constants
or time varying values in MIKE SHE.
Note: The Forcings and Constants are defined by Water Quality layer in
the Saturated Zone. Thus, you have to define at least one Water Quality
Layer in the Saturated Zone.
The list of user defined Forcings and Constants is initially taken from the
template. However, the list is not fully dynamic. Thus, if you add items to
the template, these will be added to the list. However, if you remove items
from the template, or change the name in the template, the item will not be
removed from the list. This allows you maintain your data inputs while
you are developing your template. Any data that is not used in the template will be pre-processed like any other data, but will not be used in
ECO Lab. If you don’t need the items any more, you can delete them from
the list
Tip: The fact the list of user specified Forcings and Constants is not permanently linked to the template, allows you to pre-process any static or
time-varying data and map it to the numerical grid. You can then use this
pre-processed data in other grid operations in MIKE SHE.
12.1.5
Running ECO Lab with MIKE SHE
To run ECO Lab with MIKE SHE, simply pre-process and run the model
normally. You can view the user specified Constants and Forcing in the
289
MIKE SHE + ECO Lab
Pre-processed tab. The Derived Outputs will be written to the default output directly along with all of the regular output from MIE SHE. The Auxiliary Variables and Processes will also be written to this directory, if the
Output option is on.
All output can be viewed or processed normally with all of the regular
MIKE Zero tools.
290
MIKE SHE
Requirements in MIKE SHE WM
13
PARTICLE TRACKING (PT)
The PT module calculates the flow path of a hundreds, thousands, or even
millions, of hypothetical particles, which are moved in the three-dimensional, saturated groundwater zone (SZ). The particles are displaced individually in a number of time steps. The movement of each particle is
composed of a deterministic part, in which the particle is moved according
to the local groundwater velocity calculated by the MIKE SHE water
movement (WM) module, and a stochastic part where the particle is
moved randomly based on the local dispersion coefficients.
Particle tracking is only calculated for the saturated zone (SZ) and particles that leave SZ are not traced any further. Initially, the user assigns a
number of particles to each model grid cell. Particles are added during the
simulation from sources, for example solutes in precipitation or from
boundary or internal defined concentration cells. Particles leave SZ when
they arrive at a boundary or an internal constant concentration cell or
when they go to a sink. Possible sinks in the Particle Tracking are wells,
rivers, drains, and exchange with the unsaturated zone or overland flow.
All particles are assigned a mass, which means that a number of particles
within a specific volume correspond to a solute concentration. The Particle Tracking module can therefore be used for solute transport simulations
and is in some cases superior to the conventional numerical solution of the
advection-dispersion equation since numerical dispersion is negligible.
However, the module is mostly used for delineation of abstraction well
capture zones and upstream zones and for determination of groundwater
age and conservative solute transport times.
The PT module uses the concept of 'registration' cells. This records particle data when particles enter certain model cells. Registration can be used
to delineate capture zones or to observe particles passing through some
region of interest, such as a redox layer.
13.1
Requirements in MIKE SHE WM
Prior to running a PT simulation, the MIKE SHE Water Movement (WM)
simulation must be run. This section describes what needs to be specified
in the WM simulation to run the PT simulation afterwards.
13.1.1
Flow Storing Requirements
Particle transport calculations in MIKE SHE PT are based on the groundwater flows from a MIKE SHE WM simulation. In principle, only groundwater fluxes are needed but to ensure that all the needed WM result data
291
Particle Tracking (PT)
types are stored the user has to specify that results should be stored for an
AD run in the WM input. Thus, you must check the appropriate checkbox
under “Storing of Results” in the MIKE SHE WM GUI.
The user can choose between “SZ only” and “All hydraulic components”,
however, for PT-simulations “SZ only” will be sufficient, since particle
tracking is only calculated for the saturated zone.
The simulated particle distribution is stored with a desired frequency in
the MIKE SHE WM GUI under “Storing of results” => “Grid series output”. It is important that the SZ and SZ-flow use the same storing frequency in order to run the following PT simulation. The WM result files to
be used in the PT-simulations will be located in a folder with the same
name as the *.SHE file.
13.1.2
292
Specification of Well Fields
To be able to retrieve particle locations based on well fields, it is necessary
to specify the well fields in the MIKE SHE well database file .
MIKE SHE
Output from the PT simulations
13.2
Output from the PT simulations
The result files will be located in standard Results directory for your
project. The PT result files are:
z
projectname.PTRES: An ASCII file in pfs-format listing the abstraction wells and the computational cells, where abstraction occurs. Used
for retrieval of particle location - see PT Registration Extraction
(V.2 p. 220).
z
projectname.PTREG and projectname.trf: Two binary files that cannot be opened directly.
z
projectname.PTBIN: An optional binary file containing all of the particle locations at every saved time step. Individual path lines can be
extracted using the Extraction of particle pathlines (V.1 p. 295).
z
projectname.PTGross.shp: An optional point theme shape file containing the path line information of every particle at every saved time
step. As part of the shape file, a .shx and a .dbf file are also created.
The .dbf file can be opened in Excel if it is less than 65536 lines.
z
projectname_AD_3DSZ.dfs3: Temporal and spatially varying SZ concentrations in the mobile phase based on the mass of the particles.
z
projectname_PT_3DSZ.dfs3: Temporal and spatially varying PT
results including:
– Number of particles - this is the actual number of particles in each
cell
– Accumulated particle count – this is the number of particles that
have entered the cell during the simulation
– Number of registered particles – this is the number of particles
that started in this cell that have become registered
– Most recent registration zone – this is the registration zone
attached to the last particle to be registered that originated in the cell
– Average age – this is the average age of all the particles in the cell
– Average transport time – this is the average length of time from
when the particle was born in this cell until it was registered somewhere
Besides these result files, the program also writes output to two log files.
The error log list errors encountered during execution and the print log file
contains execution step information, statistics on the run and a mass balance (if requested).
293
Particle Tracking (PT)
13.3
Extraction of particle registrations
After the particle tracking has been run, the registration information needs
to be read from the output files and processed in a useful way. The Results
Tab includes an utility to sift through all the particles and their registrations, to find the ones that you want. This is output from this utility is an
ArcGIS point-theme, shape file with the starting locations of the all the
particles that meet your registration criteria. The extraction utility allows
your to filter the results for:
z
Destination type:
– Specific sink types (drain, river, unsaturated zone, well, constant
concentration boundary or constant concentration sink)
– Registration codes specified by the user
– Wells found in the flow results
– Well fields found in the flow results*)
z
Layer from which the particles originated
z
Release (birth) time
z
Transport time
Note: To extract particle locations based on well fields requires that different well fields have been defined, see section Specification of Well Fields
(V.1 p. 292).
The results can be written to either
z
a single shape file where the point attributes allow further selection of
the particles in ArcView, or
z
separate files for each destination type and optionally for each layer
e.g. one file for each sink type/layer combination.
More detailed information on the actual extraction mechanics can be
found in the PT Registration Extraction (V.2 p. 220) section.
13.3.1
Running from a batch file
The registration extraction can be run from a command line. To execute
the program open a command line and type:
Ptoutputretrieval.exe projectname.she extraction_num
294
MIKE SHE
Extraction of particle pathlines
The extraction_num is the item number in the table of extraction items in
the PT Registration Extraction (V.2 p. 220) dialog. The extraction will proceed silently - that is without any messages. To run the extraction with the
messages, you need to use
MZLaunch project_name.she -e Ptoutputretrieval.exe
which will start the MZLaunch utility
projectname_ptoutputretrieval.err: If errors occur during execution of
the program these are written to this log file.
13.3.2
Limitations with the PT registration method
When using registration zones to identify particles that move through certain parts of the model it should be noted that particles can appear more
than once in the output. As they move from one zone to the next they are
repeatedly registered and are finally also registered when they are
removed from the model by a sink. An example would be a particle moving into a registration zone with code 1. The particle is then registered as
being in an 'active cell' and the registration zone code and travel time to
this zone is memorised. If the particle is at a later time removed by a well
it will again be registered but now it will be registered as being removed
by the 'Well' sink.
If there are multiple wells within one cell and output for wells is requested
then the output can contain the same particle more than once. As the
model does not know which of the wells the particle should be assigned to
(the program looks at the total well sink for the cell and cannot distinguish
individual wells) the particle will be repeated for each of the wells within
the cell.
13.4
Extraction of particle pathlines
It is too cumbersome to extract and plot the pathlines for all of the thousands of particles that can be generated during a PT simulation. The PT
Pathline Extraction utility allows you to extact the pathlines for specific
particles if you have saved the intermediate locations in the Storing of
Results (V.2 p. 183) dialog.
To extract a particle pathline you need a Particle ID. These can only be
found after the simulation by evaluating the PT output. For example, you
can find the particle ID by extracting the particle start locations that end in
a specific well and then finding the ID numbers of the particles that you
want in the shape file that was created.
295
Particle Tracking (PT)
In the Results tab, the PT Pathline Extraction (V.2 p. 223) utility is available to make this extraction.
Running from a batch file
The pathline output retrieval program can be run from a command line. To
execute the program, open a command line and type:
PtBinRetrieval.exe file_name.she extraction_num
The extraction_num is the item number in the table of extraction items in
the PT Pathline Extraction (V.2 p. 223) dialog. The extraction will proceed
silently - that is without any messages. To run the extraction with the messages, you need to use
MZLaunch file_name.she -e PtBinRetrieval.exe
which will start the MZLaunch utility. Particle IDs can be found by using
the PT Output Retrieval utility.
296
MIKE SHE
ADDITIONAL OPTIONS
297
298
MIKE SHE
14
EXTRA PARAMETERS
The Extra Parameters section is a special section of the Setup data tree that
allows you to input parameters for options that have not yet been included
in the MIKE SHE user interface.
The Extra Parameters are only recognized if the Name (e.g. “sheet piling
module”) are spelled exactly correct. After the initial run, you should
check in the Preprocessor_print.log file to ensure that the module has actually been activated.
Available Extra Parameters include:
Climate
z Negative Precipitation (V.1 p. 300)
z
Precipitation Multiplier (V.1 p. 301)
Surface water
z Time-varying Overland Flow Boundary Conditions (V.1 p. 302)
z
Time varying surface infiltration (Frozen soils) (V.1 p. 303)
z
Simplified Overland Flow Options (V.1 p. 304)
z
Irrigation River Source Factors (V.1 p. 306)
z
Explicit Overland Flow Output (V.1 p. 307)
z
Alternative low gradient damping function (V.1 p. 308)
z
Paved routing options (V.1 p. 309)
z
Transpiration during ponding (V.1 p. 309)
Unsaturated Zone
z Threshold depth for infiltration (2-Layer UZ) (V.1 p. 310)
z
Increase infiltration to dry soils (V.1 p. 311)
Saturated Zone
z Sheet Pile Module (V.1 p. 312)
Additional Options
z
SZ Drainage to Specified MIKE 11 H-points (V.1 p. 316)
z
SZ Drainage Downstream Water Level Check (V.1 p. 319)
z
SZ Drainage to MOUSE (V.1 p. 319)
z
Time varying drainage parameters (V.1 p. 320)
299
Extra Parameters
z
SZ Drainage River Link Reference Table (V.1 p. 321)
z
Canyon exchange option for deep narrow channels (V.1 p. 322)
Water Quality
z Disable SZ solute flux to dummy UZ (V.1 p. 323)
z
SZ boundary dispersion (V.1 p. 323)
Miscellaneous
z Including OpenMI (V.1 p. 324)
z
Plot control for Detailed Time Series Output (V.1 p. 325)
z
Extra Pre-Processing output (V.1 p. 325)
z
GeoViewer Output (V.1 p. 325)
14.1
Climate
14.1.1
Negative Precipitation
Negative precipitation is sometimes required when net groundwater
recharge has been calculated using an external program, such as DAISY
GIS. In this case, the evapotranspiration may exceed infiltration leading to
a net upward flux of water from the groundwater table. However, the
standard precipitation module in MIKE SHE does not recognize negative
rainfall. In this case, you must specify the negative rainfall using the following Extra Parameters options:
Parameter Name
Type
Value
use negative precipitation
Boolean
On
If the negative precipitation is uniformly distributed:
300
negative precipitation max
depth
float
greater than zero
negative precipitation max
layer
integer
greater than zero
MIKE SHE
Climate
Parameter Name
Type
Value
If the negative precipitation is spatially distributed:
negative precipitation max
depth dfs2 file
file name
.dfs2 file
negative precipitation max
depth dfs2 item
integer
item number in dfs2 file,
greater than zero
negative precipitation max
layer dfs2 file
file name
dfs2 file
negative precipitation max
layer dfs2 item
integer
item number in dfs2 file,
greater than zero
Max depth - This represents the depth of the root zone plus the thickness
of the capillary fringe and is the maximum depth from which negative precipitation can be extracted.
Max layer - This is the maximum layer depth from which negative precipitation can be extracted.
Note: the negative precipitation option will only work if there is no UZ
model active.
14.1.2
Precipitation Multiplier
To facilitate calibration and sensitivity analysis of recharge, in models
where measured precipitation is not being used, a multiplication factor has
been implemented.
Parameter
Name
Type
precipitation fac- float
tor
Value
greater than zero
If this extra parameter is used, then all precipitation values are multiplied
by the factor prior to being used in MIKE SHE.
Additional Options
301
Extra Parameters
14.2
Surface Water
14.2.1
Time-varying Overland Flow Boundary Conditions
The default boundary condition for overland flow in MIKE SHE is a constant water level on the outer boundary. The value of this boundary condition is determined by the initial water depth on the boundary. In most
models the recommended value is a water depth of zero. In this case, if the
water level adjacent to the boundary increases, water will discharge across
the boundary and out of the model. If you want to prevent overland outflow then you can use the Seperated Flow Areas option to restrict lateral
flow out of the model.
If you specify a non-zero value for initial water depth on the boundary,
then this value becames a constant for the entire simulation. If the water
level inside the model decreases below this value, the boundary will act as
an infinite source of inflow to the model.
However, in many models - especially those with significant wetland
areas - the constant water level condition on the boundary is too restrictive.
The following extra parameter options allow you to specify a time varying
condition for the outer boundary of the overland flow. If you initialize this
option, then you must supply a dfs2 integer grid code file that defines the
locations at which you want a time varying boundary. The input requirements have been set up such that you can re-use the model domain dfs2
output file from the pre-processor. In the model domain pre-processed output, the outer cells are defined by a value of 2 and the inner cells are
defined by a value of 1.
If the grid code value on the boundary is:
z
2 - the cell is a time varying boundary node, or
z
1 - the cell will have a constant water depth equal to the initial water
depth.
The second required file is the actual time-varying water level values.
These can be obtained from any MIKE SHE simulation, where the overland water elevation has been stored as a grid series output. There is no
requirement that they be stored on the same grid. Internally, the actual
boundary condition values will be interpolated from the nearest input values. Thus, the OL boundary conditions can be taken from a coarse
regional model and applied to a local scale model.
302
MIKE SHE
Surface Water
Finally, each filename must be accompanied by an integer item number
that defines which item in the dfs2 file should be used.
Parameter
Name
Type
Value
time varying ol
boundary
Boolean
On
ol boundary code filename
file name
.dfs2 file
ol boundary code integer
item number
item number in dfs2 file, greater than
zero
ol boundary head filename
file name
.dfs2 file
ol boundary head integer
item number
item number in dfs2 file, greater than
zero
The Hot Start function is not impacted by the time varying OL boundary.
If the continuing simulation includes the time varying OL function then it
will be used. If the continuing simulation does not include the time varying OL function the head from the hot start time point.
14.2.2
Time varying surface infiltration (Frozen soils)
A common characteristic in cold climates is that infiltration is reduced
during the winter months. When the air temperature is cold enough to
maintain precipitation as snow, then infiltration will be limited in any
case. However, in the spring, when snow storage is melting, then infiltration may still be limited for some period of time.
Although this function was conceived as a way to support reduced infiltration in winter, it can be used any time a time varying leakage is required.
The time varying infiltration function is a modification of the SurfaceSubsurface Leakage Coefficient (V.2 p. 121) to allow it to be time varying.
Additional Options
303
Extra Parameters
Parameter
Name
Type
Value
time varying ol
leakage coefficient
Boolean
On
leakage coefficient dfs2 file
name
filename
.dfs2 file
leakage coefficient item
number
integer
item number in dfs2 file, greater than
zero
mean step accumulated leakage
coefficient
Boolean
On
The time varying leakage coefficient dfs2 file contains a uniform time
series of leakage values. By default the leakage values are instantaneous
values. However, the last option above allows you to specify mean step
accumulated values.
Note that the areas in which these values will be applied has not changed.
The areas are defined in the original Surface-Subsurface Leakage Coefficient (V.2 p. 121) dialogue. That is, the leakage coefficient is active if a
non-delete value is specified in this file.
14.2.3
Simplified Overland Flow Options
Avoiding the redistribution of ponded water
In the standard version of the Simplified Overland Flow solver, the solver
calculates a mean water depth for the entire flow zone using the available
overland water from all of the cells in the flow zone. During the Overland
flow time step, ET and infiltration are calculated for each cell and lateral
flows to and from the zone are calculated. At the end of the time step, a
new average water depth is calculated, which is assigned to all cells in the
flow zone.
In practice, this results in a redistribution of water from cells with ponded
water (e.g. due to high rainfall or low infiltration) to the rest of the flow
zone where cells potentially have a higher infiltration capacity. To avoid
304
MIKE SHE
Surface Water
this redistribution, an option has been added where the solver only calculates overland flow for the cells that can potentially produce runoff, that is,
only in the cells for which the water depth exceeds the detention storage
depth.
Parameter
Name
Type
Value
only simple OL
from ponded
Boolean
On
Routing simple overland flow directly to the river
In the standard version of the Simplified Overland Flow solver, the water
is routed from 'higher' zones to 'lower' zones within a subcatchment. Thus,
overland flow generated in the upper zone is routed to the next lowest
flow zone based on the integer code values of the two zones. In other
words, at the beginning of the time step the overland flow leaving the
upper zone (calculated in the previous time step) is distributed evenly
across all of the cells in the receiving zone. In practice, this results in a distribution of water from cells in the upstream zone with ponded water (e.g.
due to high rainfall or low infiltration) to all of the cells in the downstream
zone with potentially a large number of those cells having a higher infiltration capacity. In this case, then, overland flow generated in the upper
flow zone may never reach the stream network because it is distributed
thinly across the entire downstream zone.
To avoid excess infiltration or evaporation in the downstream zone, an
option was added that allows you to route overland flow directly to the
stream network. In this case, overland flow generated in any of the overland flow zones is not distributed across the downstream zone, but rather it
is added directly to the MIKE 11 stream network as lateral inflow.
Additional Options
Parameter
Name
Type
Value
no simple OL
routing
Boolean
On
305
Extra Parameters
14.2.4
Irrigation River Source Factors
A global “river source volume factor” and “river source discharge factor”
are available as extra parameters for increased control of river sources
during irrigation.
Parameter
Name
Type
Value
river source volume factor
float
positive
river source discharge factor
float
0 or positive
None, one, or both can be specified. If the factor is not specified, then a
Volume factor of 0.99 and a Discharge factor of 0.0 will be used.
The factors are used in the calculation of the available water (depth) of a
river source:
C s ⋅ ∆t F V ⋅ V L F D ⋅ D L ⋅ ∆t
Depth = MIN  ---------------, ----------------- + ----------------------------
A
A
A
(14.1)
where Depth is the available water depth in the river link, Cs is the source
capacity, ∆t is the time step length, Fv is the specified volume factor, VL is
the volume of water in the link, FD is the specified volume discharge, DL
is the river link discharge, and A is the cell area.
The river link discharge is the same as used when checking with the
threshold discharge for switching on/off the source. It is the absolute discharge in the middle of the MIKE SHE river link, interpolated between
two MIKE 11 H-points.
MIKE SHE prints the following message in the xxx_WM_Print.log file
when the parameters are specified:
Extra-parameter specified:
river source volume factor
value = 1.500000
Extra-parameter specified:
river source discharge factor
value = 1.000000
306
MIKE SHE
Surface Water
MIKE SHE also prints the following warnings in the
xxx_WM_Init_Messages.log file if one or both of the factors may result in
water balance errors or numerical instabilities
WARNING: Specified value for river source volume factor
is greater than 1 : 1.500000.
There is a risk of water balance errors and/or instabilities in the coupling between MIKE SHE and MIKE 11.
WARNING: Specified value
factor is greater than 0
There is a risk of water
bilities in the coupling
for river source discharge
: 1.000000.
balance errors and/or instabetween MIKE SHE and MIKE 11.
Note: This option is less useful now that River Sources are defined by
both an Upstream and Downstream chainage. The option is maintained
for backward compatability.
14.2.5
Explicit Overland Flow Output
If you are using the explicit overland flow solver, the time step depends on
the location in the model with the critical courant criteria. The grid series
output allows you to save the courant criteria, so that you can see where
the critical locations are. However, the grid series output is an average
courant number over the storing time step, where there can be hundreds of
OL timesteps in a storing time step. If you are experiencing very short
time steps due to short duration rainfall events, for example, the critical
information can be difficult to distill from the dfs2 grid series output.
To make it easier to find the critical locations, an extra parameter option
was added that writes out the critical locations at every time step, if the
time step is reduced below a user-defined fraction of the storing time step.
Parameter
Name
Type
adaptive OL time Float
step info threshold fraction
Value
between 0.0 and 1.0
Default = 0.01
The default value is 0.01. This means that if the reduced time step is less
than 0.1 times the Max OL time step, then a message will be printed in the
_WM.log file. Such as:
Adaptive time step info from Explictit OL solver:
OL step no: 59: ... Final time step = 1.8108 seconds
Additional Options
307
Extra Parameters
with the following four reasons:
Critical:
Critical:
Critical:
Critical:
OL Wave Courant number. Cell (8,21)...
Net outflow from OL cell to River. Cell (8,21) ...
Net OL outflow from cell. Cell: (17,19) ...
Net outflow from River to OL. River link between ...
If you experience frequent severe reductions in the OL time step when
using the explicit OL solver, then this threshold can cause very large log
files to be created. If you are not interested in this information, then you
can reduce this threshold to reduce the frequency of the output.
14.2.6
Alternative low gradient damping function
In flat areas with ponded water, the head gradient between grid cells will
be zero or nearly zero, which means that as the gradient goes to zero ∆t
also goes to zero. To allow the simulation to run with longer time steps
and dampen any numerical instabilities in areas with low lateral gradients,
the calculated intercell flows are multiplied by a damping factor when the
gradients are close to zero.
Compared to the default damping function, an alternative damping function is available as an Extra Parameter that goes to zero more quickly and
is consistent with the function used in MIKE FLOOD.
The alternative function is a single parabolic function (see Figure 11.5 in
the Reference Manual)
To activate the alternate function, you must specify the following boolean
parameter in the Extra Parameters (V.2 p. 193) dialog:.
Parameter
Name
Type
Value
Enable Alternative Damping
Function
Boolean
On
For more detail, see the section Low gradient damping function
(V.2 p. 272) in the Reference manual.
308
MIKE SHE
Unsaturated Zone
14.2.7
Paved routing options
By default, the paved area function routes the available ponded water to
the SZ drainage network. However, the available ponded depth does not
include the detention storage.
If you want to route all of the ponded water in a cell - including the water
in detention storage.- to the SZ drainage network, then you can define the
following Extra Parameter:
Parameter
Name
Type
allow paved rout- Boolean
ing of detention
storage
Value
On
There is an option to restrict the maximum drainage rate for paved drainage. If this is specified, then the actual drainage rate will not exceed this
value. In Release 2012, this has been added to the user interface. .
Parameter
Name
Type
Value
max paved flow
rate mm/d
Float
greater than zero
14.3
Unsaturated Zone
14.3.1
Transpiration during ponding
In general, plants are not very tolerant of saturated soil in their root zone.
Saturated soil is quickly depleted of oxygen and the roots will soon die.
MIKE SHE normally takes care of this automatically by removing ET
from ponded water before calculating transpiration from the unsaturated
or saturated zones. If there is sufficient ponded water then the entire ET
will be satisfied from the ponded water.
However, some plants, such as rice, are more tolerant of saturated soils
and still extract ET from saturated soils, although normally at a reduced
rate. If ET from the soil zone is ignored, then the distribution of water supplied to ET will be incorrect.
The transpiration during ponding option changes the order in which the
ET is calculated. In this case, the ET rate is multiplied by an anaerobic
Additional Options
309
Extra Parameters
tolerance factor and ET is removed from the soil before being removed
from the ponded water.
Parameter
Name
Type
Value
allow transpira- Boolean
tion during ponding
On
global anaerobic
tolerance factor
Greater than or equal to 0
Less than or equal to 1
Float
optional (instead of global value)
14.3.2
anaerobic tolerance factor dfs2
file name
file name
.dfs2 file
anaerobic tolerance factor item
number
integer
item number in dfs2 file, greater than
zero
Threshold depth for infiltration (2-Layer UZ)
The 2-Layer water balance method for the unsaturated zone does not
include evapotranspiration from the soil surface. Thus, even a small
amount of water on the ground surface will infiltrate. If you use this extra
parameter, then you can define a depth of overland water that must be
exceeded before infiltration will occur. This keeps small amounts of precipitation from infiltrating and allows them to evaporate instead.
The calculated infiltration is simply reduced if the remaining overland
water depth will be smaller than the specified threshold value.
Parameter
Name
310
Type
Value
use threshold
Boolean
depth for infiltration
On
threshold depth
for infiltration
meter
Greater than zero
Float
MIKE SHE
Unsaturated Zone
Note: This option is less useful with the ET Deficit Factor introduced in
the 2008 Release, which maintains ET at the full rate until the specified
deficit is reached. The option is maintained for backward compatability.
14.3.3
Increase infiltration to dry soils
In dry soils the rate of infiltration can be higher than the saturated hydraulic conductivity because capillarity will draw water into the soil and
increase the rate of infiltration. The Increase Infiltration to Dry Soils extra
parameter is available to account for this process, when Richards equation
is not being used.
If the actual water content in the root zone is below the field capacity, θfc ,
then the infiltration capacity is calculated as
K infiltration = K infiltration ⋅ InfiltrationFactor
(14.2)
if
θ fc – θ wp
θ actual > ------------------------------------------------ + θ wp
InfiltrationFactor
(14.3)
where θwp is the wilting point water content.
Otherwise, the infiltration capacity is calculated as
θ fc – θ wp 
K infiltration = K infiltration ⋅  ---------------------------- θ actual – θ wp
Additional Options
Parameter
Name
Type
Value
increase infiltration to dry soils
Boolean
On
max infiltration
rate factor
Float
Greater than 1.0
(14.4)
311
Extra Parameters
14.4
Saturated Zone
14.4.1
Sheet Pile Module
The Sheet Piling module is not yet included in the MIKE SHE GUI. However, the input for the module is fairly simple and is handled via the Extra
Parameters options
The Sheet Piling module is activated by including the following two
parameters in the Extra Parameters section of the data tree, and creating
the required module input file:
Parameter
Name
Type
Value
sheet piling mod- Boolean
ule
On
sheet piling file
the file name of the Sheet Pile input file
file name
Sheet Pile Location
The location of the sheet piles is defined using a dfs2 file with integer grid
codes. One file (or item) is required for each computational layer with
sheet piling. Each file must have the same grid size as the MIKE SHE
model. The grid codes are “composed” of simple sums of 100, 10, 1, 0
where:
100 = a N-S sheet piling “link” between the actual cell and the next
cell in positive x-direction,
10 = a E-W sheet piling “link” between the actual cell and the next
cell in the positive y-direction,
1 = a Horizontal sheet-piling “surface” between the actual layer and
the layer above (ground surface if actual layer is 1), and
0 = no sheet piling.
Thus, for example, a cell containing the code “110” defines the existence
of sheet piling along the Eastern and Northern cell boundaries. A cell containing the code “11” defines a sheet piling along the Northern cell boundary and at the top of the layer.
312
MIKE SHE
Saturated Zone
Leakage Coefficient
The Leakage Coefficient is required for flow in the x-, y-, and z-direction
for each layer containing sheet piling. The Leakage Coefficient is required
in the x-direction if any cell contains a “100” value, in the y-direction if
any cell contains a “10” value, and in the z-direction if any cell contains a
“1” value.
The leakage coefficients can be specified as a global value (per layer) or
as a distribution in a dfs2 file. In the case of a dfs2 file, the values must be
specified in the cells where the grid codes are specified. The EUM type
(unit) of the dfs2 files must be “Leakage coefficient/Drain time constant”
with the unit 1/Time.
Top and bottom levels (optional)
This option can be used when the vertical sheet piling only extends across
part of a layer. The levels are specified in the same cells as the leakage
coefficients in the x- and y-direction, one set of top and bottom levels for
each direction.
The levels can be specified as global values (per layer) or as a distribution
in a dfs2 file. Both can be absolute levels or relative to ground. The EUM
type of the dfs2 files must be “elevation” for absolute levels, and “depth
below ground” (positive values) or “height above ground” (negative values) when specified relative to the ground surface. The type and unit of the
global value is “elevation” (m) when absolute, and “height above ground”
(m) (negative value) when relative.
In cells where the sheet pile extends across the entire layer, the top and
bottom levels should simply be set to large positive and negative values
respectively (e.g. 1.0E+30 and -1.0E+30).
Input File for the Sheet Pile Module
The name of the input file is specified in the Extra Parameters section
described above. The file has the general MIKEZero parameter file (pfs)
format. The exact format of the file is given below, along with a description of the different data items.
Note: The pfs format must be adhered to exactly. There is a small utility
(pfsEditor.exe) in the installation \bin directly that you can use for editing
and testing pfs files that you create.
Additional Options
313
Extra Parameters
.
Line item
Comment
[MIKESHE_SheetPiling_File]
FileVersion = 2
[SheetPiling]
FileVersion can be 1 or 2, but
must be 2, if you want to check
for the SpecifiedXYLevels option
NrOfLayers = 1
SpecifiedXYLevels = 1
Total number of SZ layers with
sheet piling
0: not specified.
1: top and bottom levels specified
for each layer
Note: only checked when
FileVersion > 1
[Layer_1]
314
This section must be repeated for
each -NrOfLayers- sheet piling
layer. The sections must be
named Layer_1, Layer_2, etc.
LayerNumber = 1
The MIKE SHE SZ layer number
of the actual sheet piling layer (1
= top layer).
[GridCodes]
Type = 1
FixedValue = 0
[DFS_2D_DATA_FILE]
FILE_NAME = |.\SPGrid_1.dfs2|
ITEM_COUNT = 1
ITEM_NUMBERS = 1
EndSect // DFS_2D_DATA_FILE
EndSect // GridCodes
[GridCodes] section Specification of grid codes for the current
layer.
Type Normally 1 because a dfs2
file is required. 0 means global
value.
FILE_NAME Name of the dfs2
file with grid codes. The file
name is enclosed in "|" which
tells the system that the name is
relative to the location of this
module input file.
ITEM_NUMBERS : One
number (because ITEM_COUNT
must be 1) defining the item of
the dfs2 file to be used.
MIKE SHE
Saturated Zone
Line item
Additional Options
Comment
[X_Leakage]
Type = 0
FixedValue = 1.0E-7
[DFS_2D_DATA_FILE]
FILE_NAME = |.\maps\SPLeakX_1.dfs2|
ITEM_COUNT = 1
ITEM_NUMBERS = 1
EndSect // DFS_2D_DATA_FILE
EndSect // X_Leakage
[X_Leakage] section Required
if there are any cells with N-S
sheet piling affecting the flow in
the x-direction (codes containing
100).
Type Set to 0 if a global value is
specified and 1 if using a dfs2
file.
FixedValue The global value
(1/s) which is read if Type = 0.
FILE_NAME and
ITEM_NUMBERS Dfs2 file
name and item number if Type =
1 (relative file name as explained
under Grid Codes).
[Y_Leakage]
Type = 0 //(0:Fixed value,1:DFS2 file)
FixedValue = 2.0E-7
[DFS_2D_DATA_FILE]
FILE_NAME = |.\maps\SPLeakY_1.dfs2|
ITEM_COUNT = 1 //(must be 1)
ITEM_NUMBERS = 1 1
EndSect // DFS_2D_DATA_FILE
EndSect // Y_Leakage
Y_Leakage] section : Required if
there are any cells with E-W
sheet piling affecting the flow in
the y-direction (codes containing
10).
[Z_Leakage]
Type = 0 //(0:Fixed value,1:DFS2 file)
FixedValue = 3.0E-7
[DFS_2D_DATA_FILE]
FILE_NAME = |.\maps\SPLeakZ_1.dfs2|
ITEM_COUNT = 1
ITEM_NUMBERS = 1
EndSect // DFS_2D_DATA_FILE
EndSect // Z_Leakage
[Z_Leakage] section : Required if
there are any cells with horizontal
sheet piling affecting the vertical
flow (codes containing 1).
[X_TopLevel]
RelativeToGround = 0 // 0: no, 1: yes
Type = 1 //(0:Fixed value,1:DFS2 file)
FixedValue = 0.0
[DFS_2D_DATA_FILE]
FILE_NAME = |.\YLevels_1.dfs2|
ITEM_COUNT = 1 //(must be 1)
ITEM_NUMBERS = 1
EndSect // DFS_2D_DATA_FILE
EndSect // Y_TopLevel
[X_TopLevel] section : Required
if SpecifiedXYLevels=1 and
there are any codes containing
100.
315
Extra Parameters
Line item
Comment
[X_BottomLevel]
RelativeToGround = 0 // 0: no, 1: yes
Type = 1 //(0:Fixed value,1:DFS2 file)
FixedValue = 0.0
[DFS_2D_DATA_FILE]
FILE_NAME = |.\YLevels_1.dfs2|
ITEM_COUNT = 1 //(must be 1)
ITEM_NUMBERS = 2
EndSect // DFS_2D_DATA_FILE
EndSect // Y_BottomLevel
[X_BottomLevel] section :
Required if SpecifiedXYLevels=1 and there are any codes
containing 100.
[Y_TopLevel]
RelativeToGround = 0 // 0: no, 1: yes
Type = 1 //(0:Fixed value,1:DFS2 file)
FixedValue = 0.0
[DFS_2D_DATA_FILE]
FILE_NAME = |.\YLevels_1.dfs2|
ITEM_COUNT = 1 //(must be 1)
ITEM_NUMBERS = 1
EndSect // DFS_2D_DATA_FILE
EndSect // Y_TopLevel
[Y_TopLevel] section : Required
if SpecifiedXYLevels=1 and
there are any codes containing
10.
[Y_BottomLevel]
RelativeToGround = 0 // 0: no, 1: yes
Type = 1 //(0:Fixed value,1:DFS2 file)
FixedValue = 0.0
[DFS_2D_DATA_FILE]
FILE_NAME = |.\YLevels_1.dfs2|
ITEM_COUNT = 1 //(must be 1)
ITEM_NUMBERS = 2
EndSect // DFS_2D_DATA_FILE
EndSect // Y_BottomLevel
[Y_BottomLevel] section :
Required if SpecifiedXYLevels=1 and there are any codes
containing 10.
EndSect // Layer_1
EndSect // SheetPiling
EndSect // MIKESHE_SheetPiling_File
14.4.2
SZ Drainage to Specified MIKE 11 H-points
The Reference Drainage (RFD) option allows you to route drainage from
the saturated zone drains and paved area runoff directly to MIKE 11 Hpoints. This is different from the normal drainage function, which routes
drainage and paved area discharges to river links rather than directly to Hpoints. Further, this option can route drainage to MIKE 11 branches that
are not defined in the MIKE SHE coupling section of the MIKE 11 network file.
The following steps are required to activate the RFD option:
316
MIKE SHE
Saturated Zone
1 Create a pfs file containing information for each specified drainage
area to be routed to the specific MIKE 11 H-points.
Line item
Comment
[MIKESHE_MIKE11DrainageReach_File]
[SpecifiedMIKE11ReachesForDrainage]
NrOfReaches = 1
RiverChainageUnit = 'meter'
[Reach_1]
DrainCode = 1
BranchName = 'Lammehavebækken'
Upstream_Chainage = 6000.
Downstream_Chainage = 8459
NrOfReaches is the number of
items specified in the section
below
For each specified reach, you
must include a section specifying
the MIKE SHE drain code, and
the MIKE 11 branch name and
the upstream and downstream
chainage.
EndSect // Reach_1
EndSect // SpecifiedMIKE11ReachesForDrainage
EndSect // MIKESHE_MIKE11DrainageReach_File
The drain code references the area that drainage and/or paved area discharge is routed to the specified MIKE 11 branch and chainage. The
drain code must be greater than or equal to zero. Drain code values
equal to zero (0) are not included in the reference drainage system.
Furthermore, an error condition will occur if the specified drain code
does not exist in the drainage code file used in MIKE SHE
The branch name must be spelled correctly and include all spaces contained in the name, if any. The branch name should not be enclosed in
quotes. An error condition will occur if the specified branch is not
present in the MIKE 11 network.
The chainages refer to the starting and ending chainage of the specified
branch which drainage and/or paved area discharge is routed to. The
interval does not have to correspond exactly to specific MIKE 11 Hpoints because the MIKE SHE pre-processor finds the closest H-points
to the specified interval. If the upstream and downstream chainages are
the same, the drainage and/or paved area discharge is routed to the
closest H-point.
Additional Options
317
Extra Parameters
2 Add the following items to the Extra Parameters list
Parameter
Name
Type
Value
use specified
Boolean
reaches for drainage
On
specified reaches file name
for drainage
the pfs file name, including the path
3 In the Drainage item under the Saturated Zone, select distributed
drainage options. See Drainage (V.2 p. 173).
4 Specify drain codes is the same manner as usual. Remember that all
drain codes in the RFD option pfs file must exist in the active domain
of the model or you will get an error.
5 Specify where the RFD option should be used in Drainage Distribution
item in the data tree under the Saturated Zone. The RFD option will be
used in all cells with a value of 3. If a combination of the original
drainage method and the RFD option is going to be used, 2 should be
used for areas using the original drainage option and 3 should be used
where you want the RFD option to be used.
6 Pre-process and run your MIKE SHE model normally.
If the MIKE SHE setup does not successfully preprocess you should
review the above steps to see if you have any error in the setup. The
projectname_PreProcessor_Messages.log file (where projectname is the
name of your *.she file) in your simulation subdirectory should help you
identify why the MIKE SHE setup failed to preprocess.
If the MIKE SHE setup successfully preprocesses you should also look at
the preprocessed data (on the Processed data tab) and the
YourSetup_PreProcessor_Print.log file in your simulation subdirectory to
make sure you are comfortable with how the preprocessor has set up the
drainage reference system. You can search for Making setup of Specified
MIKE 11 Reaches For Drainage in the YourSetup_PreProcessor_Print.log
file to find the start of the section that details the drainage reference system.
Water balance
The water balance utility (e.g., Saturated zone - detailed) can be used to
look at differences between drainage discharges from areas using the original drainage option and the RFD option. The MIKE SHE water balance
318
MIKE SHE
Saturated Zone
configuration file (MSHE_Wbl_Config.pfs in the installation directory)
should be reviewed to see which water balance types segregate standard
drainage flow (data type sz.qszdrtorivin) and RFD drainage flow (data
type sz.qszdrtoM11Hpoint) (see Using the Water Balance Tool
(V.1 p. 105))
14.4.3
SZ Drainage Downstream Water Level Check
In Release 2011, you can optionally check the downstream water level
before calculating SZ drainage. This prevents drainage from being added
to rivers during a flood, for example. It also prevents recirculation of SZ
drainage water when using Flood Codes.
Testing has shown that the test on drainage to local depression can negatively impact runtimes because the number of outer iterations in the PCG
solver may increase. Thus, the downstream check has been seperated into
two Extra Parameters.
14.4.4
Parameter Name
Type
Value
check gradient for drainage to river or mouse
Boolean
On
check gradient for drainage to local depression Boolean
On
SZ Drainage to MOUSE
The MOUSE coupling in MIKE SHE has not yet been added to the MIKE
SHE user interface. Thus, to couple the models together, use the Extra
Parameters options, along with creating a MsheMouse.pfs file to define
where and how the two models are coupled.
To tell MIKE SHE that it needs to couple to a MOUSE model, you must
add the following two items:.
Parameter Name
Type
Value
mouse coupling
Boolean
On
mouse coupling file
file name
the file name of the MOUSE coupling .pfs input file
The MIKE SHE - MOUSE coupling is described fully in Using MIKE
SHE with MIKE URBAN (V.1 p. 229).
Additional Options
319
Extra Parameters
14.4.5
Time varying drainage parameters
In projects where you want to simulate the build out of a drainage network
over time, or changes in the drainage time constants over time, then you
can use this set of extra parameters. Without this set of extra parameters
you would have to hot start your simulation at regular time intervals with
the new drainage parameters.
The time varying drains are also allowed to shift between layers. However, if the drainage level goes above or below the model, the level will be
adjusted and a warning is issued to the log file.
Note: The SOR solver does not allow drainage in any layer except the top
layer and the drain level will be adjusted accordingly.
Note: If you specify time varying drainage parameters, you will not be
able to use any of the drainage routing methods that depend on the drain
level. The preprocessor checks this and gives an error if you have specified
– option 1 (routing based on levels), or
– option 3 (distributed options) AND any of the distributed option
codes are 1 (routing based on levels in these cells).
To activate time varying drainage parameter options, you must specify the
following extra parameters
320
Parameter Name
Type
Value
time varying drainage levels
Boolean
On
time varying drainage constants
Boolean
On
time varying drainage level
dfs2 file name
file name
.dfs2 file
time varying drainage level
item number
integer
item number in dfs2 file,
greater than zero
MIKE SHE
Saturated Zone
Parameter Name
Type
Value
time varying drainage time
constant dfs2 file name
file name
.dfs2 file
time varying drainage time
constant item number
integer
item number in dfs2 file,
greater than zero
Optional if mean step accumulated values instead of instantaneous values:
mean step accumulated drain- Boolean
age levels
On
mean step accumulated drain- Boolean
age time constants
On
The dfs2 Drain Level is an elevation that can be specified using the following three EUM Data Units (V.1 p. 329):
z
Elevation
z
Depth Below Ground (i.e. positive values)
z
Height Above Ground (i.e. negative values)
By default, the Time Series Types (V.1 p. 342) is Instantaneous, but their is
an extra option that allows you to used Mean Step Accumulated values if
you want.
Note The code does not check for the time series type.
All specifications are printed to the projectname_PreProcessor_Print.log
and projectname_WM_Print.log files.
14.4.6
SZ Drainage River Link Reference Table
In the pre-processing tab, the Drain to River grid displays the river link
number that the cell drains to. Adjacent to the river links, the cells are
labeled with negative numbers to facilitate the interpretation of flow from
cells to river links. Thus, in principle, all drainage from cells with the
same positive code are drained to the cell with the corresponding negative
code.
However, this is slightly too simple because the cells actually drain
directly to the river links. In complex river systems, when the river
branches are close together, you can easily have cells connected to multiple branches on different sides. In this case, the river link numbers along
Additional Options
321
Extra Parameters
the river may not reflect the drainage-river link reference used in the
model.
If you want to see the actual river links used in all cells, you can use the
following Extra Parameter to generate a table of all the river link-cell references in the PP_Print.log file. This table can easily be several thousand
lines long. .
14.4.7
Parameter
Name
Type
Value
drainage setup
test print value
Boolean
On
Canyon exchange option for deep narrow channels
In the case of a deep, narrow channel crossing multiple model layers, the
head difference used in Equations (7.5) and (7.6) can optionally be limited
by the bottom elevation of the layer. Thus,
∆h = h grid – max ( h riv, z )
(14.5)
where z is the bottom of the current layer.
The above formulation reduces the infiltration from upper layers by reducing the available gradient. Without the ‘Canyon’ option, MIKE SHE
effectively assumes that the river is hydraulically connected to the upper
most model layer, since MIKE SHE calculates the exchange flow with all
layers that intersect the river based on the difference between the river
level and the water table.
Currently, this option is only available for steady-state models.
322
Parameter
Name
Type
Value
enable canyon
exchange
Boolean
On
MIKE SHE
Water Quality
14.5
Water Quality
14.5.1
Disable SZ solute flux to dummy UZ
The following Extra Parameter is useful, if you are using an alternative
UZ model, such as DAISY, in MIKE SHE and you are trying to couple it
to the WQ.
In this case, you will be typically using the Negative Precipitation
(V.1 p. 300) option. If you use this option, then you will not use a MIKE
SHE UZ, and the UZ-SZ exchange will pass through a “dummy UZ”
layer. When this is coupled to the water quality, solutes will also be passed
to this dummy UZ layer and removed from the SZ domain and the model.
To prevent the upflow of solutes from SZ to the dummy UZ, you must
specify the following Extra Parameter..
Parameter
Name
Type
disable sz trans- Boolean
port to dummy uz
14.5.2
Value
On
SZ boundary dispersion
A detailed test of the MIKE SHE WQ engine comparing an SZ model
with fixed concentration at an inflow boundary with an analytical solution
for a fixed concentration source, showed that MIKE SHE under-estimates
the mass flux into the model when the model includes longitudinal dispersion.
The problem is that the SZ transport scheeme (QUICKEST) doesn't
include dispersive transport to/from open boundary cells. This is as
designed, but apparently not correct. After including the boundary dispersion, the mass input to the model is within 2 % of the analytical solution.
From Release 2011 and onwards, the boundary dispersion has been made
optional for backwards compatibility and is activated with the extraparameter: .
Parameter
Name
Type
enable sz bound- Boolean
ary dispersion
Additional Options
Value
On
323
Extra Parameters
However, the SZ boundary dispersion option (above) does not calculate
dispersive transport to an inflow boundary correctly. Again, this problem
was identified in the tests of MShe_WQ with ECOLab vs analytical solution. For example:
z
Species 1 enters the model via an inflow (flux) boundary with fixed
concentration - including dispersive transport due to the new sz boundary dispersion option.
z
Species 1 decays to Species 2 which again decays to Species 3.
z
The concentrations of Sp2 & Sp3 are too high, especially close to the
inflow boundary.
The analytical solution includes dispersive transport of Sp2 & Sp3 against
the flow direction because the concentration of these species are 0 at the
boundary. However, this dispersive mass flux to the boundary is not
included in the SZ solver due to an old check in the code. When mass flux
to/from a boundary point is reversed compared to the flow direction, the
mass flux is simply reset to 0.
This made sense before the boundary dispersion was implemented
because advective transport against the flow direction would be wrong.
But, now, when the boundary dispersion is active, this situation is allowed.
14.6
Miscellaneous
14.6.1
Including OpenMI
If you want to link a program to MIKE SHE using OpenMI then you must
specify the following Extra Parameter..
Parameter
Name
Type
Value
make omi file
Boolean
On
When enabled, an *.omi file for WM is created called
MIKESHE_WM_SetupName.omi.
If Water Quality is included, a second *.omi file is created called
MIKESHE_WQ_SetupName.omi.
These omi files are to be used in the OpenMI configuration editor.
324
MIKE SHE
Miscellaneous
14.6.2
Plot control for Detailed Time Series Output
On the Results Tab, the Detailed Time Series plots are created in a set of
.html files. The default file length is 5 plots per file. However, you can
control the number of plots per html file by using the following Extra
Parameter. .
Parameter
Name
Type
Value
max number of
detailed ts plots
per html file
Integer
Greater than or equal to 1
Note If the loading of the html file can become very slow if the simulation
is long and there are many plots in the file.
14.6.3
Extra Pre-Processing output
The pre-processing log file, *_PP_Print.log, can be very long. To improve
the readability of the file, some long tables have been removed, including
the tables for drainage references.
To include these tables in the log file, use the following extra parameter: .
Parameter
Name
Type
detailed setup test Boolean
print
14.6.4
Value
On
GeoViewer Output
The GeoViewer is a MIKE Zero tool that is used in the MIKE GeoModel
product for viewing geologic cross-sections in your conceptual model.
The GeoViewer Output extra parameters will create a set of dfs2 output
files during the pre-processing that will allow you to look at your preprocessed model in the GeoViewer.
Additional Options
325
Extra Parameters
The GeoViewer Output is activated by
Parameter Name
Type
Value
make SZ level dfs2 files
Boolean
On
Boolean
On
Optional
adjust dfs2 levels
If this option is active, then the following files will be created:
z
setupname\setupname_GeoLayers.dfs2 - containing the top and bottom
of each geologic layer
If there are lenses:
z
setupname\setupname_GeoLenses.dfs2 - containing the top and bottom of each geologic lense and delete values where there are no lenses
If the computational layers are not defined by geologic layers:
z
setupname\setupname_CompLayers.dfs2 - containing the top and bottom of each computational layer
If the optional second parameter is used, then the top and bottom elevations that are written to the files will be adjusted to be confined between
the topography and the lowest computational layer.
326
MIKE SHE
MIKE ZERO OPTIONS
327
328
MIKE SHE
15
EUM DATA UNITS
All MIKE Zero products use a standard library of data units, called the
Engineering Unit Management (EUM) library. This allows you to change
the displayed units for any value that is included in the library.
Every parameter in MIKE SHE has been added to the EUM library and to
change the displayed unit, you must know the EUM Data Type. In most
cases, the EUM Data Type is displayed in the fly-over text when you put
your mouse cursor in the text field. Alternatively, all items in the on-line
help (F1) list the EUM Data Type in the table at the beginning of the section.
To change the display units of any EUM Data Type, you must close all
open documents and then select ’Options/Edit Unit Base Groups...’ from
the File pull down menu.
When you select this menu item, the Unit Base Group Editing dialogue
appears. By default all of the data units for each active module are displayed. For a clearer overview of the data types, close all of the model
engines that are not relevant.
Next select the data item that you want to change the units of. Then select
the new units from the combobox list of available units.
MIKE ZERO Options
329
EUM Data Units
After you have changed the data units, click ’Save and Close’. This saves
your changes to the default Unit Base Groups (.ubg) file:
C:\Program Files\Common Files\DHI\MIKEZero\MIKEZero.ubg
which is read every time you open a model.
Note! If you have already added data to your model, changing the Unit
Base Group will not convert any of your data. This process simply
changes the displayed units in the user interface and the conversion factors
used to make the input files internally consistent.
In some cases the relevant data item name is not clear, as there may be
several data items with similar names. This is more likely to occur if several modules are selected at the same time. To find out which data item is
correct, close the dialogue and re-open your model. Then either move the
mouse to the relevant textbox, where a fly-over text box should appear
telling you what is the relevant data type for this field. Alternatively, for
gridded data, you can use the Create button to create a data file and then
notice the data type that is displayed in the dialogue.
Finally, occasionally, you may find that the data unit that you are looking
for is not available. In this case, contact your local Technical Support Cen-
330
MIKE SHE
Changing from SI to Imperial (American) data units.
tre, who should forward your request to the developer for inclusion in the
next release.
15.1
Changing from SI to Imperial (American) data units.
The default Unit Base Groups (.ubg) file,
C:\Program Files\Common Files\DHI\MIKEZero\MIKEZero.ubg
is read every time you open a model.
In the same directory there are two standard Unit Base Group files:
MIKEZero_Default_Units.ubg
MIKEZero_US_Units.ubg
MIKE ZERO Options
331
EUM Data Units
The first is the default file and contains standard SI units for all data items
in all of the MIKE Zero products. The second contains standard Imperial
(US) units for most data items in all of the MIKE Zero products.
To change the display units for all of your data items to Imperial units,
load the MIKEZero_US_Units.ubg file, Save and Close the dialogue and
then reopen your model.
If you want to change individual data items to SI or Imperial, you can
change the items individually. Then use the Save and Close button to save
your changes back to the MIKEZero.ubg file. If you want to create special
unit versions, then you can copy the MIKEZero.ubg to a different file
name and reload it.
15.2
Restoring the default units
You can return to your default unit specification at any time, by Loading
either of the default .ubg files:
MIKEZero_Default_Units.ubg
MIKEZero_US_Units.ubg
which are found in the
C:\Program Files\Common Files\DHI\MIKEZero\
directory.
Note! If you want to save any of your model specific changes, then you
should first save the MIKEZero.ubg to a new name.
15.3
Changing the EUM data type of a Parameter
When you create a .dfs0 or .dfs2 parameter file, you must also define the
EUM data type for each parameter in the file. When you assign a .dfs0 or
a .dfs2 file to a parameter value, then MIKE SHE automatically verifies
that the correct EUM data type is being used. If the wrong data type is
present then you will not be able to select OK in the file browser dialogue.
For example, in the following set of dialogues, an Evapotranspiration time
series was selected instead of the correct Precipitation time series file
332
MIKE SHE
Changing the EUM data type of a Parameter
The first error is in the Select Item tab, where there is a message that no
Valid Items are found.
The find out why there is no valid items, you should look in the Constraints Info tab
Here you can see that the Item type is supposed to be Precipitation Rate,
but this constraint has failed.
To find out what the Item Type of the selected file is, look at the Item Info
tab,
where you can see that the current Item Type is Evapotranspiration Rate.
MIKE ZERO Options
333
EUM Data Units
The next two sections outline how to change the EUM Type of an existing
file.
15.3.1
Changing the EUM Type of a .dfs0 Parameter
To change the EUM Data Type of a parameter in a .dfs0 file, open the time
series in the Time Series Editor and then select the Properties... item from
the Edit drop down menu
This opens the item properties dialogue
334
MIKE SHE
Changing the EUM data type of a Parameter
where you can change the EUM Type and the EUM Unit that is assigned
for each time series in the file.
15.3.2
Changing the EUM Type of a .dfs2 Parameter
To change the EUM Data Type of a parameter in a .dfs2 file, open the grid
file in the Grid Editor and then select the Items... item from the Edit drop
down menu
This will open the Edit Properties dialogue for the Grid Editor
where you can change the EUM Type and the associated data EUM Unit
of the item.
MIKE ZERO Options
335
EUM Data Units
336
MIKE SHE
WORKING WITH DATA
337
338
MIKE SHE
Creating Time Series in MIKE SHE
16
TIME SERIES DATA
MIKE SHE uses the dfs0 file format for time series data. Various tools are
available for converting ASCII and EXEL time series to the dfs0 file format. Time series data is required as input for most transient simulations,
for example, daily records of precipitation. Transient simulations can also
generate numerous dfs0 output files.
16.1
Creating Time Series in MIKE SHE
In most cases, you will create dfs0 files using the Create buttons in the
MIKE SHE Setup dialogues. In this way, you can avoid the confusing task
of assigning the Type of time series (e.g. precipitation) and EUM Unit
type (e.g. millimetres) and the TS Type (e.g. reverse step accumulated).
Each of these items are specified automatically.
If you create time a time series using a Create button, the following dialogue will appear:
Working with Data
339
Time Series Data
The principle choice in this dialogue is whether to create an initially uniform time series file or to import a time series from an Excel file or from a
file with the older .t0 file format.
Uniform time series
In a uniform time series, every time step will have the same value. You
should use the uniform time series option if you want to create a time
series file where you do not have any data to import.
Time Series Period
The time series period is the extent of the time series. In a MIKE SHE
simulation, all the time series files must cover the Simulation Period
(V.2 p. 29). The default time series period for a new time series file is the
Simulation Period. However, if you change the time series period so that it
does not cover the simulation period, you will receive an error message
when MIKE SHE tries to run. If you try to add a time series file that does
not cover the simulation period, then the OK button will remain greyed
out and you will not be able to select the file. The constraints tab in the file
selector dialogue gives you the reason that you cannot select the file.
Time Series Interval
The time series interval is the length of the individual time periods. The
number of time periods is the length of the time series period divided by
the period interval. The last period is shortened if necessary.
Time Series File
Every time series has an Item Type which is defined by the valid EUM
Data Unit (see EUM Data Units (V.1 p. 329)) for the particular variable
from which the Create dialogue was launched. In most cases, there is only
one valid Type. In some cases you may have a choice. For example, in
Precipitation, you can chose between Precipitation Rate, which is the
average amount of precipitation per time (e.g. mm/hour) in the time interval, and Rainfall.which is the measured amount of precipitation in the
time interval (e.g. mm).
The Name is simply the name of the data item in the resulting .dfs0 file.
The file name has a default value, that you should change if you will be
creating several files of the same type, such as multiple rain gauge time
series files. Otherwise you may accidentally overwrite the previous file.
16.1.1
340
Import from ASCII
The easiest way to import ASCII data into a dfs0 file is via the Windows
clipboard. In this case, create a uniform time series file with the correct
MIKE SHE
Working with Spatial Time Series
number of time steps and then highlight all of the data values. Then copy
and paste the data from the ASCII file into the table.
However, if you want to import the data from an ASCII file, then you need
to create the file from the File/New menu and choose ASCII file. This is
part of the Time Series Editor itself.
16.1.2
Import from Excel
Only the first Excel Worksheet will be read when reading the Excel file.
However, the worksheet can contain any number of columns of time series
data. If there are multiple columns of data, each will be assumed to be the
of the same type. If the Excel file columns are of different types, then you
can change the data type in the Time Series Editor.
The time series is assumed to have a non-equidistant time axis and the
time series period is read from the first column of the Worksheet.
Worksheet Format
The first row is a header containing the names of each of the columns.
Each subsequent row contains the data. The first column is the date and
time (with DATE or TIME cell format), followed by the data values.
01/01/1981 00:00:00
02/01/1981 00:00:00
03/01/1981 00:00:00
04/01/1981 00:00:00
name1
0.1
0.304
0.025
0.604
namel2
0.2
0.304
0.025
0.604
name3
0.3
0.304
0.025
0.604
16.1.3
Import from old .t0 file
The old .t0 file format is from the X-Motif version of MIKE SHE that
existed before the Windows version was introduced in 2001. The .t0 file
format contains all of the relevant time information. For more information
on the .t0 file format, please refer to your original MIKE SHE documentation.
16.2
Working with Spatial Time Series
In the MIKE SHE Toolbox, there is a Tool in the File Converter section
called dfs2+dfs0 to dfs2. In this utility you specify a dfs2 grid file with
integer grid codes and a dfs0 file with time series data, where the dfs2 file
grid codes are the item numbers in the dfs0 file.
Working with Data
341
Time Series Data
The utility will read the dfs2 file and for each time step in the dfs0 file, it
will substitute the grid code with the time series value.
The result is a dfs2 file with one grid for each time step and the grid values
are the time series values.
16.3
Time Series Types
Specifies how the time step is being defined and how the measured value
is being assigned to the time step. There are five different value types
available:
Instantaneous
The values are measured at a precise instant. For example, the air temperature at a particular time is an instantaneous value.
Accumulated
The values are summed over successive intervals of time and always relative to the same starting time. For example, rainfall accumulated over a
year with monthly rainfall values.
342
MIKE SHE
Time Series Types
Step Accumulated
The values are accumulated over a time interval, relative to the beginning
of the interval. For example, a tipping bucket rain gauge measures stepaccumulated rainfall. In this case, the rain gauge accumulates rainfall until
the gauge is full, then it empties and starts accumulating again. Thus, the
time series consists of the total amount of rainfall accumulated in each
time period - say in mm of rainfall.
Mean Step Accumulated
The values are accumulated over the time interval as in the Step Accumulated, but the value is divided by the length of the accumulation period.
Thus, based on the previous example, the time series consists of the rate of
Working with Data
343
Time Series Data
rainfall accumulated in each time period - say in mm of rainfall per hour
(mm/hr).
Reverse Mean Step Accumulated
In this case, the values are the same as the Mean Step Accumulated, but
the values represent the time interval from now to the start of the next time
interval. The Reverse Mean Step Accumulated time series are primarily
used for forecasting purposes.
344
MIKE SHE
17
USING MIKE SHE WITH ARCGIS
MIKE SHE has been designed to work smoothly with ArcGIS files. In
most cases, distributed data can be linked directly to shape files created by
ArcGIS or any other application. The type of shape file depends on the
type of data. Distributed data, such as initial water levels can be input as
point and line themes, whereas spatial data that is referenced to a time
series, such as precipitation, can be added as a polygon theme. In this case,
each polygon can be assigned a time series of values.
In the reverse direction, all gridded data in the MIKE SHE Setup Editor
can be easily saved as a point theme shape file from the pop-menu when
you right click on a colour shaded map. This includes both interpolated
data in the Setup tab and pre-processed data in the Pre-processed tab.
ArcGIS grids yet cannot be added directly in the MIKE SHE Setup Editor,
but they can be converted to the dfs2 file format. Select New, then the
MIKE Zero Tool box and choose GIS in the list. The Grid2Mike tool will
convert your ArcGIS grid files to the dfs2 file format. Support for native
ArcGIS grid files will be available in a service pack later this year.
The MIKE Zero Tool box also contains tools for converting dfs2 files to
ArcGIS shape files (Mike2Shp) and Grid files (Mike2Grd). These tools
can be useful if you have manipulated your grid files in the MIKE Zero
Grid Editor, since it does not directly support shape file export. Alternatively, you can open any dfs2 file in the MIKE SHE Setup Editor, as along
as the unit type is the same) and then use the right mouse function to
export to a shape file. If you want to convert a dfs3 file to a shape file or a
grid file then you will need to extract a dfs2 file from the dfs3 first using
the 2D Grid from 3D file tool that is found under the Extraction item in the
MIKE Zero Toolbox.
Some items in the MIKE SHE Setup Editor do not support shape files.
Mostly these are related to integer grid codes, such as Drain codes. In this
case, it is difficult to assign integer values based on grid independent polygons. In a complex setup, it would be very difficult to control which cells
are being assigned to which code when the polygons do not coincide with
the cell boundaries. In some areas, the model results could be very sensitive to the code assigned.
345
Using MIKE SHE with ArcGIS
346
MIKE SHE
The Grid Editor
18
SPATIAL DATA
Spatial data includes all model data that can be location dependent, for
example precipitation rates and soil parameters.
18.1
The Grid Editor
The Grid Editor is a generic MIKE Zero grid tool for all MIKE by DHI
software. It is the primary means to edit and manipulate gridded data in
MIKE SHE.
The Grid Editor was originally developed for the Marine programs MIKE
21 and MIKE 3. However, this often leads to confusion in the node and
layer numbering because MIKE 21 and MIKE 3 use a different nodal system because they are based on a node-centered finite difference scheme.
Whereas, MIKE SHE is based on a block-centered finite difference
scheme.
Node numbering in the Grid Editor
In the Grid Editor (and in MIKE 21 and MIKE 3) the nodes are numbered
starting in the lower left from (0,0), whereas in MIKE SHE the nodes are
numbered starting in the lower left from (1,1).
Layer numbering in the Grid Editor
In the Grid Editor (and in MIKE 21 and MIKE 3) the layers are numbered
starting at the bottom from 0, whereas in MIKE SHE the layers are numbered starting at the top from 1.
18.2
Gridded Data Types
There are two basic types of spatial data in MIKE SHE - Real and Integer.
Real data is generally used to define model parameters, such as hydraulic
conductivity. Integer data is generally used to define parameter zones.
Thus, model cells with the same integer value can be associated with a
time series or other characteristic.
Furthermore, real spatial parameters can be distinguished by whether or
not they vary in time. At the moment Integer zones cannot vary with time.
Thus, spatial parameters can be divided into the following:
z
Stationary Real Parameters
347
Spatial Data
z
Time Varying Real Parameters, and
z
Integer Grid Codes
Stationary Real Parameters
Stationary Real Parameters can vary spatially but do not usually vary during the simulation, such as hydraulic conductivity. If such parameters do
vary in time, then you must divide the simulation into time periods and
run the each time period as a separate simulation, starting each simulation
from the end of the previous simulation. This is most easily accomplished
using the Hot Start facility, which is found in the Simulation Period dialogue.
The spatial distribution of stationary real parameters are entered using the
Stationary Real Data dialogue
Time Varying Real Parameters
Many spatial parameters are time dependent, such as precipitation rate. In
this case, both a spatial distribution , as well as a time series for each cell
in the model, must be defined. Spatially distributed parameters that also
vary in time are entered using the Time-varying Real Data dialogues
18.3
Integer Grid Codes
Integer Grid Codes are required when Real data varies in time or when
model functions, such as soil profiles and paved areas, are assigned to particular zones. Integer Grid Codes are always integer values and do not
vary with time.
For information on entering Integer Codes see the Integer Grid Codes section.
The following is an outline of the parameters that require Integer Grid
Codes.
Model Domain
Integer Grid Codes are used to define the inactive areas both inside and
outside the model domain. Inactive areas outside of the model and the
edge of the model are defined in the Model Domain and Grid section,
while inactive, subsurface areas inside of the model are defined as Internal
boundary conditions.
348
MIKE SHE
Gridded (.dfs2) Data
Component Calculations
Integer Grid Codes are used to delineate such things as paved areas. In this
case, the integer code acts like a flag and the calculations that are done are
different depending on how the flag is set.
Model Properties
Integer Grid Codes are used to delineate areas with similar properties. In
this case, the integer value defines the zone to which the cell belongs.
Thus, it defines which set of model properties is to be assigned to the particular cell.
For example, a model may be divided into a five zones each with a different soil profile for the unsaturated zone. In this case, the data tree will
expand under the model property to include five separate sub-branches,
where the soil profile can be defined.
Time Series
Integer Grid Codes are used to define zones for which Real data varies in
time. Thus, a time series for a parameter, such as precipitation rate, can be
assigned to a model zone. Similarly to the Model Properties above, the
model tree will expand under the parameter to include a separate subbranch for each zone, where the time series file can be defined.
Time Varying Integers
Grid Codes and Integer values do not normally vary with time. If such
parameters do vary in time, then you must divide the simulation into time
periods and run each time period as a separate simulation, starting each
simulation from the end of the previous simulation using the Hot Start
options (see Simulation Period).
18.4
Gridded (.dfs2) Data
If the parameter is defined using gridded data, then the data must be in
DHI’s .dfs2 file format.
The easiest way to create the .dfs2 file is to use the
button, which
creates a new grid with the proper default values and attribute type. You
can then edit this grid in the MIKE Zero Grid Editor, which can be
accessed using the
button.
Alternatively, a .dfs2 file can be created using the Grid Series editor,
which can be accessed by clicking on File|New in the pull-down menu, or
349
Spatial Data
using the New File icon,
Series.
, in the toolbar, and then selecting Grid
If you create the file from these tools you must be careful to ensure that
the EUM Data Type matches the parameter that you are creating the file
for. For more information on the EUM data types, see EUM Data Units.
The grid for the .dfs2 file does not have to be the same as the numerical
model grid. However, if the grids are not subsets of one another then the
grids will be interpolated using the bilinear interpolation during the preprocessing stage.
The parameter grid and the model grid are aligned with one another if the
parameter grid or the model grid contain an even multiple of the other
grid’s cells. For example, if the parameter grid was two times finer, then
every model grid cell must contain exactly four parameter grid cells.
If the grids are aligned then the parameter grid will be averaged to the
model grid during the pre-processing stage. However, in some cases it
does not make sense to average parameter values. For example, Van
Genuchten soil parameters cannot really be averaged, since they are a
characteristic of the soil. In such cases, you should ensure that the model
grid and the parameter grid file are identical.
350
MIKE SHE
Gridded (.dfs2) Data
18.4.1
Stationary Real Data
Spatially distributed Real parameters, such as conductivity or topography,
can be defined in three ways, namely they can be defined as a uniform
(global) value or they may be distributed and defined using either gridded
data (.dfs2 file), GIS points and polygons (ArcView .shp file), or irregularly distributed point data (x, y, value coordinate file).
It does not make sense to interpolate some parameters to the model grid.
In such cases, the use of line and point data should be avoided.
Uniform
A uniform, global value means that all the grid cells in the model will
have the same value.
GIS point and line data or Distributed point data
If the parameter is defined using irregularly distributed point data or an
ArcView shape (.shp) file, then the data will be interpolated to the model
grid during the pre-processing stage, using the interpolation method
selected.
The following interpolation methods are included:
z
Bilinear Interpolation (V.1 p. 355), or
z
Triangular Interpolation (V.1 p. 359)
z
Inverse Distance (V.1 p. 361).
It does not make sense to interpolate some parameters to the model grid.
In such cases, the use of line and point data should be avoided.
Elevation Data
Elevation data, such as Layer elevations, is handled exactly the same as all
other Stationary Real Parameters, except that the value may be optionally
specified as a depth below the ground surface rather than absolute elevation above the datum.
Note: The value must be negative if it is below the ground level.
351
Spatial Data
Tip: The current tools do not allow you to specify a polygon shape file
with real values. However, this would be desirable in some cases, such as
when implementing Mannings M values based on vegetation distributions.
A trick to get around this limitation is the following:
1 Temporarily assign an integer grid code to each of the polygons.
2 Specify this file as an input file for one of the data items that needs
integer grid codes, such as drain codes.
3 Right click on the map that will be displayed and save the map view to
a dfs2 file
4 Open this dfs2 file in the grid editor and use the grid editor tools to
replace the integer values with real values
5 In the Grid Editor, change the EUM unit to the appropriate value
6 Save the file and then load it into the Data item for which you wanted
it.
18.4.2
Time-varying Real Data
If the time-varying Real parameter does not vary spatially then the parameter must be defined as Global with either a Fixed or Time-varying value
(see Uniform + Constant and Uniform + Time Varying).
Often, time-varying data, such as precipitation rate, are spatially distributed using measurement stations, which in the model are translated into
model zones using, for example, Thiessen polygons. In this case, each station is associated with a .dfs0 time series file that contains the time series
of precipitation rate. Station-based zones are defined using Integer Grid
Codes in either a .dfs2 file as Grid Codes, or in a Shape (.shp) file as polygons with an Integer Code (see Station-based + Grid Codes or Polygons).
Uniform + Constant
The parameter Value will be assigned to every cell in the model or layer as
appropriate and will remain constant throughout the simulation.
352
MIKE SHE
Gridded (.dfs2) Data
Uniform + Time Varying
The time series in the .dfs0 file will be assigned to every cell in the model
or layer as appropriate.
Station-based + Grid Codes or Polygons
Station-based time varying data means that the model domain is divided
into zones that are defined by an Integer Grid Code.
If a .dfs2 file is used, then the Integer Grid Codes are defined on a regular
grid, which is interpreted to the model grid during the Pre-processing
stage.
If the Integer Grid Codes are defined using polygons then you must supply
an ArcView .shp file containing polygons each with an Integer Grid Code.
The item Fill Gaps with: allows you to define the Integer Grid Code to
use in the event that a cell is not included within one of the polygons.
Once the file containing Integer Grid Codes has been defined, a new level
in the data tree will appear below the current level, containing one entry
for every unique Integer Grid Code in the file.
On this level, you must then supply a time series values for every Integer
Grid Code. However, the time series can also be fixed, in the sense that a
constant value over time is used. This makes it easy to use detailed time
series for some zones and constant values for zones where little information exists.
The time series dialogue itself includes two graphical views. The upper
graphic displays the time series that is being applied and the lower graphic
shows where the time series will be applied.
18.4.3
Integer Grid Codes
The dialogues for Integer Codes function essentially same as those for
Stationary Real Data, except that interpolation does not make sense for
integer grid codes.
If Integer Grid Codes are being used to assign Model Properties, such as
soil profiles or time series, then new sub-branches will appear in the data
353
Spatial Data
tree corresponding to the number of unique Integer Grid Codes in the .dfs2
file.
Uniform Value
A Uniform, global value means that all the grid cells in the model will
have the same value. Thus, all cells would belong to the same zone.
Grid File (.dfs2)
If the Integer Code is defined using a grid file, then the Integer Code is
defined on a grid. This grid may be different than the numerical model
grid. However, the grids must be subsets of one another. That is, the Integer Code grid and the model grid must be aligned with one another and the
Integer Code grid or the model grid must contain an even multiple of the
other grid’s cells. For example, if the Integer Code grid was two times
finer, then every model grid cell must contain exactly four Integer Codes.
Normally, the Integer Code will be assigned to the model grid based on the
most prevalent Integer Code in the cell. However, this can lead to problems when the a particular code is both infrequent and widely dispersed.
For example, if a model area contained many small wetland areas that
were much smaller than a grid cell.
For this reason, a bookkeeping count is kept of the assignments to reduce
any bias in the assignment of Integer Codes and ensure that less frequently
occurring Integer Codes will be represented in the resulting model grid.
For example, if their were two different Integer Codes, A and B, used in
the model and A always occurred more frequently in each model cell, the
bookkeeping count would ensure that B would actually be assigned to
some of the model cells. The final frequency of occurrence of the Integer
Codes in the model cells would reflect the underlying frequency of occurrence of the Integer Codes. That is, if A occurred twice as often as B, the
model grid would also contain twice as many A’s as B’s.
Thus, in our widely dispersed wetland example, if every model grid cell
contained 9 Integer Codes for Land Use, and 1/9 of the Land Use grid
codes were for wetlands, then every ninth Model Cell would be assigned a
Land Use grid code for wetlands.
Polygons
In the current version, only some of the parameters are set up to accept
.shp file polygons. Currently, .shp file polygons are only allowed in:
354
z
Model Domain and Grid (V.2 p. 70)
z
Precipitation Rate (V.2 p. 77)
MIKE SHE
Interpolation Methods
z
Vegetation (V.2 p. 94),
z
Reference Evapotranspiration (V.2 p. 81)
z
UZ Soil Profile Definitions (V.2 p. 132),
z
SZ Internal boundary conditions (V.2 p. 169), and
z
Horizontal Extent (V.2 p. 195) of SZ Lenses.
Note The Horizontal Extent (V.2 p. 195) of SZ Lenses accepts polygons,
but the dialogue is still set up for point/line .shp files and an error is given
in the Data Verification window.
Model grid codes are assigned based in which polygon the centre of the
cell is located in.
18.5
Interpolation Methods
The gap filling is based on the concept that we have to calculate the depth
in the point (xc, yc). We define this as the function zc = f(xc, yc). If we
place our self in this point, we can divide the world up into four quadrants
Q1 - Q4. From here it’s a matter of finding some points from the raw data
set relatively close to this point. The search radius for all possible techniques can be entered - in grid cell distance. Points outside this distance
will never be taken into account.
Figure 18.1
18.5.1
Definition of quadrants
Bilinear Interpolation
This technique finds four points from the raw data set - one in each quadrant. The search is done in the following way. A mask of relative indices is
created. The cells in this mask are sorted according to the distance. For the
quadrant Q1 the cells are sorted in the following way, the grid point it self
being excluded.
355
Spatial Data
Figure 18.2
Illustration of the neighbouring grid cells being sorted
Note that the grid cells with a crosshatch pattern contain raw data points.
When the closest raw data point in each quadrant is found, we have four
points that form a quadrangle. This quadrangle contains the centre point,
where we want to calculate the z-value. This is illustrated on Figure 18.3.
Figure 18.3
Illustration of the closest raw data points in each quadrant.
Note that each grid cell might contain more raw data points. If this is the
case, the closest of these is chosen. We now have an irregular quadrangle,
where the elevation is defined in each vertex. We need to compute the elevation in (xc, yc). If we transform our quadrangle into a square, we can
perform bilinear interpolation. This is illustrated on Figure 18.4.
356
MIKE SHE
Interpolation Methods
Figure 18.4
Illustration of bilinear interpolation.
First the interpolation requires the transformation from quadrangle to a
normalized square. This is done in the by computing 8 coefficients in the
following way:
A1 = x0
A2 = y0
B1 = x1 – x0
B2 = y1 – y0
C1 = x3 – x0
(18.1)
C2 = y3 – y0
D1 = x2 – x1 + x0 – x3
D2 = y2 – y1 + y0 – y3
Mapping the coordinates (xc, yc) to the normalized square (dx, dy) is done
by solving equation (18.2).
2
ax + bx + c = 0
(18.2)
357
Spatial Data
where the coefficients are
a = D 1 B2 – D 2 B 1
b = D 2 xc – D 1 yc – D 2 A 1 + D 1 A 2 + C1 B 2 – C2 B1
(18.3)
b = C 2 xc – C1 yc + C 1 A2 – C 2 A1
Solving equation (18.2) gives us dx.
2
– b ± b – 4ac
dx = ----------------------------------2a
(18.4)
where 0 ≤ dx ≤ 1 is used to choose the correct root. dy can now be computed in two ways:
x c – A 1 – B 1 dx
dy = ---------------------------------C 1 – D 1 dx
(18.5)
or
x c – A 2 – B 2 dx
dy = ----------------------------------C 2 – D 2 dx
(18.6)
Choosing between (18.5) and (18.6) is done in such a way, that division by
zero is avoided. (xc, yc) has been mapped to (dx, dy). The task was to compute the elevation in the point (xc, yc) and this is done in the following way
using regular bilinear interpolation:
z c = ( 1 – dx ) ( 1 – dy ) z 2 + dx ( 1 – dy ) z 3 + ( 1 – dx )dy z 1 + dxdy z 0
358
(18.7)
MIKE SHE
Interpolation Methods
If less than four points are found (if one or more quadrants are empty), the
double linear interpolation is replaced with reverse distance interpolation
(RDI). This is done according to the following scheme:
1
w i = ------------------------------------------------------( xi – xc ) 2 + ( yi – yc ) 2
(18.8)
N
ws =
∑ wi
(18.9)
i=1
N
1
z c = ----- ∑ w i z i
ws
(18.10)
i=1
The method works fairly efficiently, but it has one drawback. The quadrant search is heavily dependent on the orientation of the bathymetry. If
the bathymetry is rotated 45 degrees 4 completely different points might
be used for the interpolation. For this reason there is also a Triangular
interpolation method, which can be used, and this method should be direction independent.
18.5.2
Triangular Interpolation
As mentioned previously the ‘Bilinear Interpolation’ is dependent on the
orientation of the bathymetry. The ‘Triangular Interpolation’ is made as an
answer to this problem. First the closest point to (xc, yc) is found. The following figure shows this:
359
Spatial Data
Figure 18.5
Illustration of triangular interpolation
In this example the point (x0, y0, z0) is the closest point. When this point is
identified, two quadrants are identified – indicated by the light grey and
the dark grey areas. The closest point in these two quadrants are then
found. They can be seen on the figure as (x1, y1, z1) and (x2, y2, z2). The
interpolation is then done in two steps. First the coefficients describing the
plane defined by the 3 found points are computed:
–( y1 – y0 ) ( z 2 – z 0 ) + ( y2 – y0 ) ( z1 – z0 )
A = --------------------------------------------------------------------------------------------( x1 – x0 ) ( y2 – y0 ) – ( x2 – x0 ) ( y1 – y0 )
( x1 – x0 ) ( z 2 – z0 ) + ( x2 – x0 ) ( z1 – z0 )
B = ------------------------------------------------------------------------------------------( x1 – x0 ) ( y2 – y0 ) – ( x2 – x0 ) ( y1 – y0 )
(18.11)
C = z 0 – Ax 0 – By 0
And secondly the actual interpolation is done:
z c = Ax c + By c + C
(18.12)
If less than 3 points are found, reverse distance interpolation (RDI) is
used. The triangular interpolation is more time consuming due to the more
complex direction independent search, but better end results should be
achieved with this method.
360
MIKE SHE
Performing simple math on multiple grids
18.5.3
Inverse Distance
The two inverse distance methods both use the nearest point in each quandrant to calculate the interpolated value, weighted by the distance or the
distance squared respectively.
18.6
Performing simple math on multiple grids
In the upper menu of the Grid Editor, under tools, there is an item called
Copy File into Data.
If you select this item then a dialogue appears where you can insert an
existing dfs2 or dfs3 file into the current dfs2 or dfs3 file that you are editing in the Grid editor.
Alternatively, you can define an operation that you want to do with the
file. For example, if you were editing a topography file, you could subtract
all of the values in a lower elevation file, to obtain a thickness distribution
for a layer.
361
Spatial Data
The principle advantage of this tool, is that time varying dfs2 and dfs3
files can be manipulated. However, if the operations are complex, but not
time varying then
Target file
The target file is the current file you are editing in the Grid editor. The
operations that you do are performed on the target file. So, if you don’t
want to edit the target file, copy it to a new name first and edit the copy.
File to Copy
The top section of the dialogue is the name of the source file that you want
to insert into, subtract from, add to, etc. the target file.
Item mapping
If the target file or the source file has more than one item in it, then all of
the items will be listed here and you will be able to choose whether or not
to map the various items to one another.
2D to 3D Layer Mapping
If you are mapping a 2D dfs2 file into a 3D dfs3 file, then you can choose
to map all of the layers or only a single layer.
Sub-area position
You select to map the source file onto the target file starting at a different
location than the origin. In this case, you must specify the coordinates in
the target grid where the origin of the source grid should be positioned.
For example, if you have a 20x20 grid and we wish to copy data into the
4x4 rectangle given by the four nodes (10,14), (13,14), (13,17) and
(10,17), then you should select a 4x4 grid file and specify j-origin=10 and
k-origin=17. Note: the Grid editor starts its nodal numbering at 0,0.
Time Position
The source grid and target grid do not have to have equal time steps or the
same time origin. In this section of the dialogue, you can specify the time
at which the source grid should be added to the target grid. In this way,
you can add additional time steps to the end of a time varying dfs2 file, or
insert hourly information into a monthly time series, for example.
Operation
Finally, you can specify how the source grid file should interact with the
target file.
362
MIKE SHE
Performing complex operations on multiple grids
Copy - all values are copied such that they replace the existing data in the
data set
Copy if target differs from delete value - values in the source file will be
copied into the target file, only if the target value is a delete value
Copy if source differs from delete value - values in the source file will
be copied into the target file, only if both the source value and the target
values are not delete values
Copy if source AND target differs from delete value - values in the source
file will be copied into the target file, only if the source value is not a
delete value
+ - the source values will be added to the target values
- - the source values will be subtracted from the target values
* - the source values will be multiplied by the target values
/ - the source values will be divided by the target values
18.7
Performing complex operations on multiple grids
In the Toolbox, under MIKE SHE/Util, there is a Grid calculator tool,
which allows you to perform complex operations on .dfs2 grid files. However, the grid files must have the same grid dimensions and they may not
included multiple time steps or multiple items. Thus, this tool is much
more restrictive than the grid operations available in the Grid Editor. However, you can make complex chains of operations and save the setup,
which can save you a lot of time if you are doing the same operation many
times.or after each simulation
The Grid Calculator works like a wizard, with Next and Back buttons to
move between dialogues.
363
Spatial Data
364
MIKE SHE
INDEX
365
Index
Symbols
.fif file
. . . . . . . . . . . . . . . . . 149
Numerics
64-bit CPU . . . . . . . . . . . . . . . 25
64-bit support . . . . . . . . . . . . . 167
A
Actual time step
ET . . . . . . . . . . . . . . . .
OL . . . . . . . . . . . . . . .
UZ . . . . . . . . . . . . . . .
Air temperature . . . . . . . . . .
ArcGIS . . . . . . . . . . . . . . .
Area Inundation Exchange . . .
Calculation - Exchange Flows
Auto Updater . . . . . . . . . . .
Auxiliary Variables
ECO Lab . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
163
163
163
. 42
345
214
215
. 27
. . 280
B
Batch Files . . . . .
Create . . . . . .
Results . . . . . .
Setup . . . . . . .
Bilinear Interpolation
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
164
165
166
165
355
C
Calculation
Exchange Flows . . . . . . .
Calibration
Water Quality . . . . . . . . .
Climate . . . . . . . . . . . . . .
Concatenation . . . . . . . . . .
Concentration units . . . . . . .
Consecutive simulations . . . . .
Constants
ECO Lab . . . . . . . . . . . .
Coupling
MIKE SHE and MIKE 11 . . .
UZ and SZ . . . . . . . . . . .
Coupling MIKE 11 to MIKE SHE
All branches . . . . . . . . . .
Bed Leakage . . . . . . . . .
Bed Topography . . . . . . .
366
Exchange Type . . . . . . . . . . 220
Flood Area . . . . . . . . . . . . . 223
Leakage Coefficient . . . . . . . . 221
Coupling MIKE 11 to MIKE SHEFlood
Code . . . . . . . . . . . . . . . . . 224
Coupling MIKE SHE to MOUSE . . 234
Output Files . . . . . . . . . . . . 238
CPU Speed . . . . . . . . . . . . . . . 26
Cross-section
River Link . . . . . . . . . . . . . . 202
D
Define
Grid . . . . . . . . . . .
Model domain . . . . .
DEM
formats . . . . . . . . .
Demo Limits . . . . . . . .
Derived Outputs
ECO Lab . . . . . . . .
Detention Storage . . . .
dfs2 file . . . . . . . . . . .
Drainage . . . . . . . . . .
Dual Core . . . . . . . . .
dual domain mass transfer
Dual porosity . . . . . . .
. . . . . . . 37
. . . . . . . 37
. . . . . . . 40
. . . . . . . 24
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
. .
. .
60,
. .
. .
. .
280
. 49
. 38
253
. 25
281
275
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
280
284
289
280
280
280
329
332
332
331
331
. 41
. 42
215
299
E
. . 215
.
.
.
.
.
.
.
.
.
.
273
. 40
. 72
280
. 72
. . 280
. . 199
. 56, 60
. . 219
. . 225
. . 224
ECO Lab . . . . . . . . . . .
in MIKE SHE . . . . . . .
Running with MIKE SHE
Templates . . . . . . . .
templates . . . . . . . . .
units . . . . . . . . . . . .
EUM Data Units . . . . . . .
Change . . . . . . . . . .
Default units . . . . . . .
Imperial . . . . . . . . . .
SI . . . . . . . . . . . . .
Evapotranspiration . . . . .
Crop Reference ET . . .
Exchange Flows . . . . . .
Extra Parameters . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
MIKE SHE
Index
F
Fixed species . . . . . . . .
Flooded Area . . . . . . . .
Flow Storing Requirements
Flows to MIKE SHE . . . .
Forcings
ECO Lab . . . . . . . . .
frf file . . . . . . . . . . . . .
Frozen soils . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
287
215
291
203
. . . . . 280
. . . . . . 71
. . . . . 303
G
Geologic Model . . . . . . .
Conductivity values . . .
Lenses . . . . . . . . . .
Green and Ampt . . . . . .
Grid Editor . . . . . . . . . .
2D to 3D Layer Mapping
File to Copy . . . . . . .
Item mapping . . . . . .
MIKE SHE vs MIKE 3 . .
node numbering . . . . .
Operation . . . . . . . . .
Sub-area position . . . .
Target file . . . . . . . . .
Time Position . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
58,
. .
59,
. .
. .
. .
. .
. .
. .
. .
. .
. .
. .
. .
249
250
251
. 54
347
362
362
362
347
347
362
362
362
362
H
Hardware . . . . . . . .
Hardware Requirements
Horizontal Interpolation
Hot start . . . . . . . . .
limitations . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 24
. 24
250
. 37
. 37
I
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
288
353
354
354
354
361
. 45
L
. . . . . . . . . . . . . . . . . . . . 44
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 43
. 44
. 59
. 255
. 71
M
Maintenance . . . . . . .
Mannings M . . . . . . .
Maps
adding . . . . . . . . .
Maximum discharge rate
MIKE 11
Unsaturated flow . . .
water balance . . . .
water quality . . . . .
MIKE 11 Water Levels .
MIKE FLOOD . . . . . .
MIKE Zerio . . . . . . . .
Model chains . . . . . . .
Model domain
specifying . . . . . . .
Model Limits
Demo version . . . .
Model limits
Licensed version . . .
MODFLOW . . . . . . .
MOUSE . . . . . . . . . .
MsheMouse.pfs file . . .
Multi-cell overland flow .
Multiple simulations . . .
. . . . . . . 26
. . . . . . . 49
. . . . . . . 34
. . . . . . . 176
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 217
. 217
. 218
. 203
. 51
. 29
. 72
. . . . . . . 37
. . . . . . . 24
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . 23
. 57, 262
. . . 229
. . . 235
. . . 181
. . . 72
N
Numerical layers
Index
Ecological . . .
Integer Grid Codes
Grid File (.dfs2)
Polygons . . . .
Uniform Value .
Inverse distance .
Irrigation . . . . . .
LAI
Land use . . . . . . .
Leaf Area Index (LAI)
Lenses . . . . . . . .
Local depressions . .
Log files . . . . . . .
. . . . . . . . . . . 60
O
OpenMI . . . . . . . . . .
Output
file types . . . . . . .
items . . . . . . . . . .
multiple simulations .
results concatentation
Overland Flow
Boundary Conditions
Overland flow
. . . . 166, 324
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
71
72
72
72
. . . . . . . 49
367
Index
alternative damping function
boundary conditions . . . .
catchment based options .
detailed output option . . .
detention storage . . . . . .
embankments . . . . . . . .
evaluating . . . . . . . . . .
extra parameter options . .
Mannings M . . . . . . . . .
multi-cell . . . . . . . . . . .
no flow boundaries . . . . .
no flow boundary . . . . . .
performance . . . . . . . . .
seperated flow areas . . . .
stagnant water . . . . . . .
Stickler coefficient . . . . .
Threshold gradient . . . . .
time varying boundaries . .
turning off . . . . . . . . . .
velocities . . . . . . . . . . .
Overland Flow Routing
Directly to the river . . . . .
Overland flow velocity . . . . .
. . 308
. . . . 49
. . . 304
. . . 307
. . . . 49
. 50, 174
. . . 179
. . . 302
. . . . 49
. . . 181
. . . . 50
. 50, 174
.178, 181
. . . 174
. . . 178
. . . 172
.179, 180
. . . 302
. . . 179
. . . 177
. . . 305
. . . . 73
P
Parallelization . . . .
run-time licenses
Particle Tracking . .
Paved areas . . . .
Penman-Monteith .
Precipitation . . . . .
Pre-processed data
.fif file . . . . . . .
editing . . . . . .
land use . . . . .
M11 . . . . . . . .
SZ drainage . . .
unsaturated zone
viewing . . . . . .
Processes
ECO Lab . . . . .
PT Simulations
Output . . . . . .
368
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . 167
. . 167
. . 291
45, 174
. . . 42
. . . 40
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
149
150
151
151
153
152
150
. . . . . . . . . 280
. . . . . . . . . 293
R
RAM . . . . . . . . . . . . . . .
Recharge . . . . . . . . . . . .
Redistribution of ponded water
Release 2011 . . . . . . . . . .
Results
detailed time series output .
River links . . . . . . . . . . . .
River-Aquifer Exchange . . . .
Full Contact . . . . . . . . .
Reduced Contact . . . . . .
Root depth . . . . . . . . . . . .
. . .
. . .
. . .
260,
. 25
. 74
304
261
. . .
. . .
. . .
. . .
209,
. . .
. 64
199
207
208
210
. 45
S
Saturated flow . . . . . . . . . . . . . . 57
Saturated zone
drainage . . . . . . . . . . . . . . 253
Service Packs . . . . . . . . . . . . . . 27
Service packs . . . . . . . . . . . . . . 27
Setup
Data Tree . . . . . . . . . . . . . . . 33
Sheet Pile Module . . . . . . . . . . 312
Input File . . . . . . . . . . . . . . 313
Leakage Coefficient . . . . . . . . 313
Location . . . . . . . . . . . . . . 312
Sheet piling . . . . . . . . . . . . . . 312
sheres file . . . . . . . . . . . . . . . . 71
shp file . . . . . . . . . . . . . . . . . . 38
SMA . . . . . . . . . . . . . . . . . . . 26
Snow . . . . . . . . . . . . . . . . . . . 41
Snow melt . . . . . . . . . . . . . . . . 42
Soil profiles . . . . . . . . . . . . . . . 52
Soils database . . . . . . . . . . . . . 53
Solute Transport
Requirements . . . . . . . . . . . 272
Results . . . . . . . . . . . . . . . 272
Solver parameters . . . . . . . . . . . 36
Spatial Data
Stationary Real Parameters . . . 348
Time Varying Real Parameters . 348
Species
Type . . . . . . . . . . . . . . . . . 287
Specific yield . . . . . . 56, 59, 153, 246
State Variables
ECO Lab . . . . . . . . . . . . . . 280
MIKE SHE
Index
Stationary Real Parameters . . . . . 348
Storing Time Steps . . . . . . . . . . 164
Surface water storage . . . . . . . . 176
Surfer
support for . . . . . . . . . . . . . . 40
Suspended species . . . . . . . . . 287
SZ
conceptual model . . . . . . . . . . 58
drainage . . . . . . . . . . . . . . . 60
finite difference . . . . . . . . . . . 57
lenses . . . . . . . . . . . . . . . . . 59
linear reservoir . . . . . . . . . . . . 58
specific yield . . . . . 56, 59, 153, 246
T
Temperature
air . . . . . . . . . . . . . . .
Templates . . . . . . . . . . . .
Threshold gradient . . . . . . .
Time series
detailed output control . . .
Time step control . . . . . . . .
Time Varying Real Parameters
Triangular Interpolation . . . .
Type
Species . . . . . . . . . . . .
. . . . 42
. . . 280
179, 180
.
.
.
.
.
.
.
.
.
.
.
.
325
. 36
348
359
. . . 287
U
Unsaturated and Saturated Zone
Coupling . . . . . . . . . . . . 56, 245
Unsaturated flow . . . . . . . . . . . . 52
UZ
bypass flow . . . . . . . . . . . . . 54
discretization . . . . . . . . . . . . . 53
dry soils . . . . . . . . . . . . . . . . 54
Green and Ampt . . . . . . . . . . . 54
initial conditions . . . . . . . . . . . 53
macropore flow . . . . . . . . . . . 54
soil profiles . . . . . . . . . . . . . . 52
soils database . . . . . . . . . . . . 53
specific yield . . . . . 56, 59, 153, 246
sub-grid variability . . . . . . . . . . 54
UZ Classification . . . . . . . . . 55, 243
UZ mass balance
error sources . . . . . . . . . . . . . 56
UZ-SZ
limitations . . . . . . . . . . . . 56, 246
V
Vegetation . . . . . .
Velocities . . . . . . .
Velocity
overland flow . . .
Veritical Interpolation
. . . . . . . . . 44
. . . . . . . . . 177
. . . . . . . . . 73
. . . . . . . . . 251
W
Water Balance
Units . . . . . . .
Water balance
limitations . . . .
Water Balances
Batch Mode . . .
Custom . . . . .
Standard Types
Types . . . . . .
Water Quality
Calibration . . .
Limitations . . .
Requirements .
Results . . . . .
Time step control
Water quality . . . .
mass balance .
MIKE 11 . . . . .
Well Fields . . . . .
WM Requirements
Work flow . . . . . .
. . . . . . . . . . 113
. . . . . . . . . . 140
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 113
. 144
. 142
. 114
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 273
. 271
. 272
. 272
. 273
. 35
. 275
. 218
. 292
. 291
. 29
.
.
.
.
.
.
369
Index
370
MIKE SHE
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement