Download version 0.1 of EM 1110-2-1420 Hydrologic Engineering Requirements for Reservoirs.pdf

Download version 0.1 of EM 1110-2-1420 Hydrologic Engineering Requirements for Reservoirs.pdf
Department of the Army
EM 1110-2-1420
U.S. Army Corps of Engineers
Washington, DC 20314-1000
Engineering and Design
Distribution Restriction Statement
Approved for public release; distribution is
31 October 1997
U.S. Army Corps of Engineers
Washington, DC 20314-1000
No. 1110-2-1420
EM 1110-2-1420
31 October 1997
Engineering and Design
1. Purpose. This manual provides guidance to field office personnel for hydrologic engineering
investigations for planning and design of reservoir projects. The manual presents typical study
methods; however, the details of procedures are only presented if there are no convenient references
describing the methods. Also, publications that contain the theoretical basis for the methods are
referenced. Many of the computational procedures have been automated, and appropriate references
are provided.
2. Applicability. This manual applies to all HQUSACE elements and USACE commands having civil
works responsibilities.
Colonel, Corps of Engineers
Chief of Staff
U.S. Army Corps of Engineers
Washington, DC 20314-1000
EM 1110-2-1420
No. 1110-2-1420
31 October 1997
Engineering and Design
Table of Contents
Chapter 1
Purpose . . . . . . . . . . . . . . . . . . . . . . . . . .
Applicability . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . .
Related H&H Guidance . . . . . . . . . . . . . .
Part 1 Hydrologic Engineering
Concepts for Reservoirs
Chapter 2
Reservoir Purposes
Congressional Authorizations . . . . . . . . .
Reservoir Purposes . . . . . . . . . . . . . . . . .
Flood Control . . . . . . . . . . . . . . . . . . . . .
Navigation . . . . . . . . . . . . . . . . . . . . . . . .
Hydroelectric Power . . . . . . . . . . . . . . . .
Irrigation . . . . . . . . . . . . . . . . . . . . . . . . .
Municipal and Industrial
Water Supply . . . . . . . . . . . . . . . . . . . . .
Water Quality . . . . . . . . . . . . . . . . . . . . .
Fish and Wildlife . . . . . . . . . . . . . . . . . . .
Recreation . . . . . . . . . . . . . . . . . . . . . . . .
Water Management Goals
and Objectives . . . . . . . . . . . . . . . . . . . .
Chapter 3
Multiple-Purpose Reservoirs
Hydrologic Studies for
Multipurpose Projects . . . . . . . . . . . . . . . 3-1
Relative Priorities of
Project Functions . . . . . . . . . . . . . . . . . . 3-2
Managing Competitive and
Complementary Functions . . . . . . . .
Operating Concepts . . . . . . . . . . . . . .
Construction and Physical
Operation . . . . . . . . . . . . . . . . . . . .
General Study Procedure . . . . . . . . . .
Chapter 4
Reservoir Systems
Introduction . . . . . . . . . . . . . . . . . . . .
System Description . . . . . . . . . . . . . .
Operating Objectives
and Criteria . . . . . . . . . . . . . . . . . . .
System Simulation . . . . . . . . . . . . . . .
Flood-Control Simulation . . . . . . . . .
Conservation Simulation . . . . . . . . . .
System Power Simulation . . . . . . . . .
Determination of Firm Yield . . . . . . .
Derivation of Operating
Criteria . . . . . . . . . . . . . . . . . . . . . .
System Formulation Strategies . . . . .
General Study Procedure . . . . . . . . . .
Part 2 Hydrologic
Chapter 5
Hydrologic Engineering
Meteorological Data . . . . . . . . . . . . . 5-1
Topographic Data . . . . . . . . . . . . . . . 5-2
Streamflow Data . . . . . . . . . . . . . . . . 5-3
EM 1110-2-1420
31 Oct 97
Adjustment of Streamflow
Data . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
Simulation of
Streamflow Data . . . . . . . . . . . . . . . . . . 5-5
Chapter 6
Hydrologic Frequency
Introduction . . . . . . . . . . . . . . . . . . . . . . .
Duration Curves . . . . . . . . . . . . . . . . . . .
Flood-Frequency Determinations . . . . . . .
Estimating Frequency Curves . . . . . . . . .
Effect of Basin Developments
on Frequency Relations . . . . . . . . . . . . .
Selection of Frequency Data . . . . . . . . . .
Climatic Variations . . . . . . . . . . . . . . . . .
Frequency Reliability Analyses . . . . . . . .
Presentation of Frequency
Analysis Results . . . . . . . . . . . . . . . . . .
Chapter 7
Flood-Runoff Analysis
Introduction . . . . . . . . . . . . . . . . . . . . . . . 7-1
Flood Hydrograph Analysis . . . . . . . . . . . 7-2
Hypothetical Floods . . . . . . . . . . . . . . . . . 7-3
Chapter 8
Water Surface Profiles
Introduction . . . . . . . . . . . . . . . . . . . . . . .
Steady Flow Analysis . . . . . . . . . . . . . . .
Unsteady Flow Analysis . . . . . . . . . . . . .
Multidimensional Analysis . . . . . . . . . . .
Movable-Boundary Profile Analysis . . . . .
Chapter 9
Reservoir Sediment Analysis
Introduction . . . . . . . . . . . . . . . . . . . . . . .
Sediment Yield Studies . . . . . . . . . . . . . .
Reservoir Sedimentation Problems . . . . .
Downstream Sediment Problems . . . . . . .
Sediment Water Quality . . . . . . . . . . . . .
Sediment Investigations . . . . . . . . . . . . . .
Part 3 Reservoir Storage
Chapter 10
Flood-Control Storage
General Considerations . . . . . . . . . . . . . . 10-1
Regulated Release Rates . . . . . . . . . . . . . 10-2
Flood Volume Frequencies . . . . . . . . .
Hypothetical Floods . . . . . . . . . . . . . . .
Operation Constraints
and Criteria . . . . . . . . . . . . . . . . . . . .
Storage Capacity
Determinations . . . . . . . . . . . . . . . . .
Spillways . . . . . . . . . . . . . . . . . . . . . . .
Flood-Control System
Formulation . . . . . . . . . . . . . . . . . . . .
General Study Procedure . . . . . . . . . . .
Chapter 11
Conservation Storage
General Considerations . . . . . . . . . . . .
Water Supply . . . . . . . . . . . . . . . . . . . .
Navigation and Low-Flow
Augmentation . . . . . . . . . . . . . . . . . .
Fish and Wildlife . . . . . . . . . . . . . . . . .
Hydroelectric Power . . . . . . . . . . . . . .
Water Quality Considerations . . . . . . .
Water Quality Requirements . . . . . . . .
Reservoir Water Quality
Management . . . . . . . . . . . . . . . . . . .
Chapter 12
Conservation Storage Yield
Introduction . . . . . . . . . . . . . . . . . . . . .
Problem Description . . . . . . . . . . . . . .
Study Objectives . . . . . . . . . . . . . . . . .
Types of Procedures . . . . . . . . . . . . . . .
Factors Affecting Selection . . . . . . . . .
Time Interval . . . . . . . . . . . . . . . . . . .
Physical Constraints . . . . . . . . . . . . . .
Priorities . . . . . . . . . . . . . . . . . . . . . . .
Storage Limitations . . . . . . . . . . . . . . .
Effects of Conservation
and Other Purposes . . . . . . . . . . . . . .
Simplified Methods . . . . . . . . . . . . . . .
Detailed Sequential Analysis . . . . . . . .
Effects of Water Deficiencies . . . . . . . .
Shortage Index . . . . . . . . . . . . . . . . . .
General Study Procedures . . . . . . . . . .
Chapter 13
Reservoir Sedimentation
Introduction . . . . . . . . . . . . . . . . . . . . . 13-1
Reservoir Deposition . . . . . . . . . . . . . . 13-2
Distribution of Sediment
Deposits in the Reservoir . . . . . . . . . . 13-3
EM 1110-2-1420
31 Oct 97
Chapter 15
Dam Freeboard Requirements
Basic Considerations . . . . . . . . . . . . . . . .
Wind Characteristics
over Reservoirs . . . . . . . . . . . . . . . . . . .
Computation of Wave and Wind
Tide Characteristics . . . . . . . . . . . . . . . .
Wave Runup on
Sloping Embankment . . . . . . . . . . . . . .
Freeboard Allowances
for Wave Action . . . . . . . . . . . . . . . . . .
Chapter 16
Dam Break Analysis
Introduction . . . . . . . . . . . . . . . . . . . . . . .
Dam Breach Analysis . . . . . . . . . . . . . . .
Dam Failure Hydrograph . . . . . . . . . . . . .
Dam Break Routing . . . . . . . . . . . . . . . . .
Inundation Mapping . . . . . . . . . . . . . . . .
Chapter 17
Channel Capacity Studies
Introduction . . . . . . . . . . . . . . . . . . . .
Downstream Channel
Capacity . . . . . . . . . . . . . . . . . . . . .
Stream Rating Curve . . . . . . . . . . . . .
Water Surface Profiles . . . . . . . . . . . .
Part 4 Hydrologic
Engineering Studies
for Reservoirs
Chapter 14
Spillways and Outlet Works
Function of Spillways
and Outlet Works . . . . . . . . . . . . . . . . .
Spillway Design Flood . . . . . . . . . . . . . . .
Area and Capacity of the Reservoir . . . . .
Routing the Spillway Design Flood . . . . .
Sizing the Spillway . . . . . . . . . . . . . . . . .
Outlet Works . . . . . . . . . . . . . . . . . . . . . .
Chapter 18
Real Estate and Right-of-Way
Introduction . . . . . . . . . . . . . . . . . . . .
Definition of Terms . . . . . . . . . . . . . .
Real Estate Acquisition
Policies for Reservoirs . . . . . . . . . . .
Hydrologic Evaluations . . . . . . . . . . .
Water Surface
Profile Computations . . . . . . . . . . . .
Development of an
Envelope Curve . . . . . . . . . . . . . . . .
Evaluations to Determine
Guide Taking Lines (GTL) . . . . . . .
Acquisitions of Lands for
Reservoir Projects . . . . . . . . . . . . . .
Acquisition of Lands
Downstream from Spillways
for Hydrologic Safety
Purposes . . . . . . . . . . . . . . . . . . . . .
Appendix A
EM 1110-2-1420
31 Oct 97
EM 1110-2-1415 Hydrologic Frequency Analysis
Chapter 1
EM 1110-2-1416 River Hydraulics
EM 1110-2-1417 Flood-Runoff Analysis
1-1. Purpose
EM 1110-2-1602 Hydraulic Design
Outlet Works
of Reservoir
This manual provides guidance to field office personnel for
hydrologic engineering investigations for planning and
des i gn of reservoir projects. The manual presents typical
study methods; however, the details of procedures are only
presented if there are no convenient references describing
the methods. Also, publications that contain the theoretical
basis for the methods are referenced. Many of the comput at ional procedures have been automated, and appropriate references are provided.
EM 1110-2-1701 Hydropower
1-2. Applicability
EM 1110-2-4000 Sedimentation Investigation
Rivers and Reservoirs
T his manual applies to all HQUSACE elements and
USACE commands having civil works responsibilities.
a. Scope.
This manual provides information on
hydrologic engineering studies for reservoir projects.
These studies can utilize many of the hydrologic engineeri ng m et hods described in the manuals listed in paragraph 1-4. Hydraulic design of project features are not
included here; they are presented in a series of hydraulic
design manuals.
b. Or gani zation. This manual is divided into four
parts. Part 1 provides basic hydrologic concepts for reservoi rs . Reservoir purposes and basic hydrologic concerns
and methods are presented. Part 2 describes hydrologic
dat a and analytical methods. Part 3 covers storage
requirements for various project purposes, and the last,
Part 4, covers hydrologic engineering studies.
1-3. References
EM 1110-2-1603 Hydraulic Design of Spillways
EM 1110-2-3600 Management
b. Engineer regulations. There are several engineer
regu lations (ER) which prescribe necessary studies associated with reservoir projects. The most relevant ER's are
listed below.
ER 111 0-2-1460 Hydrol ogi c
Reservoir Water Quality Analysis
Engi neeri n g
ER 1110-2-7004
Hydrologic Analysis for Watershed
ER 1110-2-7005
Hydrologic Engineering Requirements for Flood Damage Reduction
ER 1110-2-7008
Hydrologic Engineering in Dam
ER 1110-2-7009
Hydrologic Data Collection and
ER 1110-2-7010
1-4. Related H&H Guidance
EM 1110-2-1201
These manuals provide the technical background for study
proced ures that are frequently required for reservoir analysis. Specific references to these EM's are made throughout
this document.
Required and related publications are listed in Appendix A.
a. Engineer manuals. This engineer manual (EM)
relies on, and references, technical information presented in
ot her guidance documents. Some of the key EM's for
res ervoir studies are listed below. Additionally, there are
rel at ed documents on hydraulic design for project features
associated with reservoir projects. This document does not
present hydraulic design concepts.
of Water Control
ER 1110-8-2(FR) Inflow Design Floods for Dams
and Reservoirs
T hes e and other regulations should be consulted prior to
performing any hydrologic engineering study for reservoirs.
A current index of regulations should be consulted for new
and updated regulations.
EM 1110-2-1420
31 Oct 97
EM 1110-2-1420
31 Oct 97
Chapter 2
Reservoir Purposes
2-1. Congressional Authorizations
a. Authorization of purposes. The United States
Congress authorizes the purposes served by U.S. Army
Corps of Engineers reservoirs at the time the authorizing
legislation is passed. Congress commonly authorizes a
project Asubstantially in accordance with the recommendations of the Chief of Engineers,@ as detailed in a separate
congressional document. Later, additional purposes are
sometimes added, deleted, or original purposes modified,
by subsequent congressional action. When the original
purposes are not seriously affected, or structural or
operational changes are not major, modifications may
be made by the Chief of Engineers (Water Supply Act
b. General legislation. Congress also passes general
legislation that applies to many projects. The 1944
Flood Control Act, for example, authorizes recreational
facilities at water resource development projects. This
authority has made recreation a significant purpose at many
Corps reservoirs. Similar general legislation has been
passed to enhance and promote fish and wildlife (1958) and
wetlands (1976). The Water Resource Development Act of
1976 authorizes the Chief of Engineers, under certain
conditions, to plan and establish wetland areas as part of an
authorized water resource development project.
chronology of the congressional legislation authorizing
various purposes and programs is shown in Figure 2-1
(USACE 1989).
c. Additional authorization. Figure 2-1 illustrates
how additional authorizations have increased the number of
purposes for which the Corps is responsible both in
planning and managing water resource development
projects. The first authorizations were principally for
navigation, hydroelectric power, and flood control. Later
authorizations covered a variety of conservation purposes
and programs. During drought when there is a water
shortage, all purposes compete for available water and are
affected by the shortage. The more purposes and programs
Figure 2-1. Purposes and programs authorized by Congress
EM 1110-2-1420
31 Oct 97
there are to serve, the greater the potential for conflict, and
the more complex the task of managing existing supplies.
AAuthorized and Operating Purposes of Corps of Engineers' [email protected] (USACE 1992) lists the purposes for
which Corps operated reservoirs were authorized and are
2-2. Reservoir Purposes
a. Storage capacity. A cross section of a typical
reservoir is shown in Figure 2-2. The storage capacity is
divided into three zones: exclusive, multiple-purpose, and
inactive. While each Corps reservoir is unique both in its
allocation of storage space and in its operation, the division of storage illustrated by Figure 2-2 is common.
included in this manual, are nonetheless important in water
control management.
d. Inactive capacity. The inactive space is commonly used to maintain a minimum pool and for sediment
storage. Sediment storage may affect all levels of the
reservoir storage. Also, the inactive capacity may sometimes be used during drought when it can provide limited
but important storage for water supply, irrigation, recreation, fish and wildlife, and water quality.
e. Storage space allocation. Reservoir storage
space may not be allocated to specific conservation purposes. Rather, reservoir releases can serve several purposes. However, the amount of water needed to serve
each purpose varies. During drought, with limited multiple-purpose storage available, the purposes requiring
greater releases begin to compete with purposes requiring
less. For example, if the greater releases are not made, the
storage would last longer for the purposes served by the
lesser releases.
f. General information. A brief description of
project purposes is presented below. Additional detail and
a discussion of reservoir operating procedures may be
found in EM 1110-2-3600, from which the following
sections are excerpts.
2-3. Flood Control
Figure 2-2. Typical storage allocation in reservoirs
b. Exclusive capacity.
The exclusive space is
reserved for use by a single purpose. Usually this is flood
control, although navigation and hydroelectric power have
exclusive space in some reservoirs. The exclusive capacity reserved for flood control is normally empty. Some
reservoirs with exclusive flood control space have no
multiple-purpose pool but have a nominal inactive pool
that attracts recreational use. Recreational use is also
common on pools originally established exclusively for
c. Multiple-purpose capacity.
storage serves a variety of purposes. These purposes
include both seasonal flood control storage, often in addition to exclusive storage, and conservation. Conservation
purposes include: navigation, hydroelectric power, water
supply, irrigation, fish and wildlife, recreation, and water
quality. Other conservation purposes such as wetlands,
groundwater supply and endangered species, while not
a. Utilizing storage space. Reservoirs are designed
to minimize downstream flooding by storing a portion or
sometimes the entire runoff from minor or moderate flood
events. Each reservoir's water control plan defines the
goals of regulation. Usually, a compromise is achieved to
best utilize the storage space to reduce flooding from both
major and minor flood events. In special circumstances
where reservoir inflows can be forecast several days or
weeks in advance (for example, when the runoff occurs
from snowmelt), for the best utilization of storage space,
the degree of control for a particular flood event may be
determined on the basis of forecasts. When runoff is
seasonal, the amount of designated flood control storage
space may be varied seasonally to better utilize the reservoirs for multiple-purpose regulation.
b. Releases. Flood control releases are based upon
the overall objectives to limit the discharges at the downstream control points to predetermined damage levels.
The regulation must consider the travel times caused by
storage effects in the river system and the local inflows
between the reservoir and downstream control points.
EM 1110-2-1420
31 Oct 97
c. Int ervening tributary and downstream damage
ar eas. A multiple-reservoir system is generally regulated
for flood control to provide flood protection both in intervening tributary areas and at downstream main stem damage areas. The extent of reservoir regulation required for
protecting these areas depends on local conditions of flood
damage, uncontrolled tributary drainage, reservoir storage
capacity, and the volume and time distribution of reservoir
i nflows. Either the upstream or downstream requirements
m ay govern the reservoir regulation, and usually the
optimum regulation is based on the combination of the two.
d. C oo rdinated reservoir regulation. Water control
with a system of reservoirs can incorporate the concept of a
balanced reservoir regulation, with regard to filling the
res ervoirs in proportion to each reservoir's flood control
capability, while also considering expected inflows and
downstream channel capacities. Evacuation of flood water
stored in a reservoir system must also be accomplished on a
coordinated basis. Each reservoir in the system is drawn
down as quickly as possible, considering conditions at
control points, to provide space for controlling future
fl oods. The objectives for withdrawal of water in the
various zones of reservoir storage are determined to
m i ni m i ze the risk of encroaching into the flood control
storage and to meet other project requirements. Sometimes
the lower portion of the flood control pool must be
evacuated slower to transition to a lower flow to minimize
bank caving and allow channel recovery.
2-4. Nav igation
a. N avigational requirements. Problems related to
t he management of water for navigation use vary widely
among river basins and types of developments. Control
s t ruct u res at dams, or other facilities where navigation is
one of the project purposes, must be regulated to provide
required water flows and/or to maintain project navigation
depths. Navigational requirements must be integrated with
other water uses in multiple-purpose water resource
s ys t em s. In the regulation of dams and reservoirs, the
nav igational requirements involve controlling water levels
in the reservoirs and at downstream locations, and providi ng t he quantity of water necessary for the operation of
l ock s. There also may be navigational constraints in the
regulation of dams and reservoirs with regard to rates of
change of water surface elevations and outflows. There are
numerous special navigational considerations that may
involve water control, such as ice, undesirable currents and
water flow patterns, emergency precautions, boating events,
and launchings.
b. Waterflow requirements. Navigation locks located
at dams on major rivers generally have sufficient water
from instream flows to supply lockage water flow
requirements. Navigation requirements for downstream use
i n open river channels may require larger quantities of
water over a long period of time (from several months to a
year), to maintain water levels for boat or barge transportation.
Usually, water released from reservoirs for
navigation is also used for other purposes, such as hydroel ectric power, low-flow augmentation, water quality,
enhancement of fish and wildlife, and recreation. Seasonal
or annual water management plans are prepared which
defi ne the use of water for navigation. The amount of
stored water to be released depends on the conditions of
wat er storage in the reservoir system and downstream
requirements or goals for low-flow augmentation, as well as
factors related to all uses of the water in storage.
c. Using water for lockage. Navigational constraints
are al so important for short-term regulation of projects to
meet all requirements. In some rivers, supply of water for
lockage is a significant problem, particularly during periods
of l ow flow or droughts. The use of water for lockage is
generally given priority over hydropower or irrigation
usages. However, this is dependent on the storage allocated
to each purpose.
In critical low-water periods, a
curt ailment of water use for lockage may be instituted by
restricting the number of locks used, thereby conserving the
utilization of water through a more efficient use of the
navi g ation system. Water requirements for navigation
canals are sometimes based on lockage and instream flows
as necessary to preserve water quality in the canal.
2-5. Hydroelectric Pow er
a. Res ervoir project categories. Reservoir projects
whi ch incorporate hydropower generally fall into two
distinct categories: storage reservoirs which have sufficient
capacity to regulate streamflow on a seasonal basis and runof-river projects where storage capacity is minor relative to
t he vol u me of flow. Most storage projects are multiplepurpose.
Normally, the upstream reservoirs include
provisions for power production at the site, as well as for
release of water for downstream control. Run-of-river
hyd ropower plants are usually developed in connection
with navigation projects.
b. Int eg ration and control of a power system.
Integration and control of a major power system involving
hydropower resources is generally accomplished by a
cent ralized power dispatching facility.
This facility
EM 1110-2-1420
31 Oct 97
contains the equipment to monitor the entire power system
operation, including individual plant generation, substation
operation, transmission line operation, power loads and
requ irements by individual utilities and other bulk power
users, and all factors related to the electrical system control
for real-time operation. The dispatching center is manned
on a continuous basis, and operations monitor and control
the flow of power through the system, rectify outages, and
perform all the necessary steps to ensure the continuity of
power system operation in meeting system loads.
c. Regulation of a hydropower system. Regulation of
hydropower systems involves two levels of control:
s cheduling and dispatching. The scheduling function is
performed by schedulers who analyze daily requirements
for meeting power loads and resources and all other project
requirements. Schedules are prepared and thoroughly
coordi n ated to meet water and power requirements of the
system as a whole. Projections of system regulation, which
i ndi cate the expected physical operation of individual
pl an ts and the system as a whole, are prepared for one to
fi ve days in advance. These projections are updated on a
dai l y or more frequent basis to reflect the continuously
changing power and water requirements.
2-6. Irrigation
a. Irrigation diversion requirements. Irrigation water
diverted from reservoirs, diversion dams, or natural river
channels is controlled to meet the water duty requirements.
T he requirements vary seasonally, and in most irrigated
areas in the western United States, the agricultural growing
s eason begins in the spring months. The diversion
requirements gradually increase as the summer progresses,
reaching their maximum amounts in July or August. They
then recede to relatively low amounts by late summer. By
t he end of the growing season, irrigation diversions are
terminated, except for minor amounts of water that may be
necessary for domestic use, stock water, or other purposes.
b. Irrigation as project purpose. Corps of Engineers'
res ervoir projects have been authorized and operated
pri marily for flood control, navigation, and hydroelectric
However, several major Corps of Engineers
multiple-purpose reservoir projects include irrigation as a
proj ect purpose. Usually, water for irrigation is supplied
from reservoir storage to augment the natural streamflow as
requ ired to meet irrigation demands in downstream areas.
In some cases, water is diverted from the reservoir by
gravity through outlet facilities at the dam which feed
directly into irrigation canals. At some of the run-of-river
power or navigation projects, water is pumped directly from
the reservoir for irrigation purposes.
c. Meeting irrigation demands. The general mode
for regulation of reservoirs to meet irrigation demands is to
capture all runoff in excess of minimum flow demands and
water rights during the spring and early summer. This
us ually results in refilling the reservoirs prior to the
irrigation demand season. The water is held in storage until
the natural flow recedes to the point where it is no longer of
sufficient quantity to meet all demands for downstream
i rri gat ion. At that time, the release of stored water from
reservoirs is begun and continued on a demand basis until
t he end of the growing season (usually September or
October). During the winter, projects release water as
required for instream flows, stock water, or other project
2-7. Municipal and Industrial Water Supply
a. Muni cipal and industrial use. Regulation of
reservoirs for municipal and industrial (M&I) water supply
i s performed in accordance with contractual arrangements.
S t orag e rights of the user are defined in terms of acre-feet
of s t o red water and/or the use of storage space between
fi xed limits of reservoir levels. The amount of storage
space is adjusted to account for change in the total reservoir
capacity that is caused by sediment deposits. The user has
t he right to withdraw water from the lake or to order
rel eases to be made through the outlet works. This is
s ubj ect to Federal restrictions with regard to overall
regulation of the project and to the extent of available
storage space.
b. T emporary withdrawal. In times of drought,
special considerations may guide the regulation of projects
with regard to water supply. Adequate authority to permit
temporary withdrawal of water from Corps projects is
cont ai n ed in 31 U.S.C. 483a (HEC 1990e). Such withdrawal requires a fee that is sufficient to recapture lost
proj ect revenues, and a proportionate share of operation,
maintenance, and major replacement expenses.
2-8. Water Quality
a. Goa l and objective. Water quality encompasses
t he physical, chemical, and biological characteristics of
water and the abiotic and biotic interrelationships. The
quality of the water and the aquatic environment is significantly affected by management practices employed by the
wat er control manager. Water quality control is an
authorized purpose at many Corps of Engineers reservoirs.
However, even if not an authorized project purpose, water
qual i ty is an integral consideration during all phases of a
proj ect's life, from planning through operation. The
m i nimum goal is to meet State and Federal water quality
EM 1110-2-1420
31 Oct 97
s t an dards in effect for the lakes and tailwaters. The operat ing objective is to maximize beneficial uses of the
resources through enhancement and nondegradation of
water quality.
b. Release requirements. Water quality releases for
downstream control have both qualitative and quantitative
requirements. The quality aspects relate to Corps' policy
and objectives to meet state water quality standards, maint ai n present water quality where standards are exceeded,
and maintain an acceptable tailwater habitat for aquatic life.
T he C orps has responsibility for the quality of water
di s charged from its projects. One of the most important
m easures of quality is quantity. At many projects authori zed for water quality control, a minimum flow at some
downstream control point is the primary water quality
obj ective. Other common objectives include temperature,
di s s o lved oxygen, and turbidity targets at downstream
c. Coordinated regulation. Coordinated regulation of
multiple reservoirs in a river basin is required to maximize
benefits beyond those achievable with individual project
regulation. System regulation for quantitative aspects, such
as flood control and hydropower generation, is a widely
accepted and established practice, and the same principle
appl ies to water quality concerns.
Water quality
m ai ntenance and enhancements may be possible through
coordinated system regulation. This applies to all facets of
quality from the readily visible quantity aspect to traditional
concerns such as water temperature and dissolved oxygen
d. Sys t em regulation. System regulation for water
qual i t y is of most value during low-flow periods when
avai l able water must be used with greatest efficiency to
avo id degrading lake or river quality. Seasonal water
control plans are formulated based on current and forecasted basin hydrologic, meteorologic and quality conditions, reservoir status, quality objectives and knowledge of
wat er quality characteristics of component parts of the
system. Required flows and qualities are then apportioned
to the individual projects, resulting in a quantitatively and
qualitatively balanced system. Computer programs capable
of simulating reservoir system regulation for water quality
provide useful tools for deriving and evaluating water
control alternatives.
2-9. Fish and Wildlife
P roject regulation can influence fisheries both in the reservoi r pool and downstream. One of the most readily
observable influences of reservoir regulation is reservoir
pool fluctuation. Periodic fluctuations in reservoir water
levels present both problems and opportunities to the water
con trol manager with regard to fishery management. The
seasonal fluctuation that occurs at many flood control
res ervoirs, and the daily fluctuations that occur with
hydropower operation often result in the elimination of
s horeline vegetation and subsequent shoreline erosion,
wat er quality degradation and loss of habitat. Adverse
impacts of water level fluctuations also include loss of
s horeline shelter and physical disruption of spawning and
2-10. Recreation
a. Reservoir level. Recreational use of the reservoirs
may extend throughout the entire year. Under most
circumstances, the optimum recreational use of reservoirs
would require that the reservoir levels be at or near full
conservation pool during the recreation season. The degree
to which this objective can be met varies widely, depending
upon the regional characteristics of water supply, runoff,
and the basic objectives of water regulation for the various
project purposes. Facilities constructed to enhance the
recreational use of reservoirs may be designed to be
operable under the planned reservoir regulation guide
curves on water control diagrams, which reflect the ranges
of res ervoir levels that are to be expected during the
recreational season.
b. Dow ns tream river levels. In addition to the
s eas o nal regulation of reservoir levels for recreation,
regulation of project outflows may encompass requirements
for s pecific regulation criteria to enhance the use of the
ri vers downstream from the projects, as well as to ensure
t he s afety of the general public. The Corps has the
responsibility to regulate projects in a manner to maintain
or enhance the recreational use of the rivers below projects
t o t he extent possible (i.e., without significantly affecting
t he project function for authorized purposes). During the
peak recreation season, streamflows are regulated to ensure
t he s afet y of the public who may be engaged in water
rel ated activities, including boating, swimming, fishing,
rafting, and river drifting. Also, the aesthetics of the rivers
m ay be enhanced by augmenting streamflows during the
l ow-water period. Water requirements for maintaining or
enhancing the recreational use of rivers are usually much
smaller than other major project functional uses.
Nevertheless, it is desirable to include specific goals to
enhance recreation in downstream rivers in the water
control plan. The goals may be minimum project outflows
or augmented streamflows at times of special need for
boat i ng or fishing. Of special importance is minimizing
any danger that might result from changing conditions of
outflows which would cause unexpected rise or fall in river
l evels. Also, river drifting is becoming an important
EM 1110-2-1420
31 Oct 97
recreational use of rivers, and in some cases it may be
pos s i bl e to enhance the conditions of stream flow for
relatively short periods of time for this purpose.
2-11. Water Management Goals and Obj ectiv es
a. Water management. ER 1110-2-240 paragraph 6,
defines the goals and objectives for water regulation by the
Corps. Basically, the objective is to conform with specific
provisions of the project authorizing legislation and water
m an agement criteria defined in Corps of Engineers reports
prepared in the planning and design of the project or
system. Beyond this, the goals for water management will
include the provisions, as set forth in any applicable
aut h orities, established project construction, and all
applicable Congressional Acts and Executive Orders
relating to operations of Federal facilities.
b. Water control systems management. EM 1110-23600 provides guidance on water control plans and project
m anagement. A general prime requirement in project
regul ation is the safety of users of the facilities and the
general public, both at the project and at downstream
locations. The development of water control plans and the
s chedu ling of reservoir releases must be coordinated with
appropriate agencies, or entities, as necessary to meet
commitments made during the planning and design of the
project. Additionally, water control plans must be reviewed
and adjusted, when possible, to meet changing local
c. Regi onal management. Regional water managem ent should consider the interaction of surface-groundwat er resources. HEC Research Document 32 provides
exam ples for several regions in the United States (HEC
EM 1110-2-1420
31 Oct 97
Chapter 3
Multiple-Purpose Reservoirs
3-1. Hydrologic Studies for Multipurpose
Proj ects
a. Conception. Multipurpose reservoirs were original l y conceived as projects that served more than one
purpo se independently and would effect savings through
the construction of a single large project instead of two or
more smaller projects. As the concept developed, the joint
us e of water and reservoir space were added as
m ul t i p urpose concepts. Even such competitive uses as
flood control and water supply could use the same reservoir
space at different times during the year.
b. Feasibility. The feasibility of multiple-purpose
development is almost wholly dependent upon the demons t rat ed ability of a proposed project to serve several purpos es simultaneously without creating conditions that
would be undesirable or intolerable for the other purposes.
In order to demonstrate that multipurpose operation is
feas i bl e, detailed analyses of the effects of various combinations of streamflows, storage levels, and water requirements are required. Detailed analyses of these factors may
be overlooked during the planning phase because the
analyses are complex and simplifying methods or assumptions may not consider some details that may be important.
However, ignoring the details of multipurpose operation in
t he pl anning phase is risky because the operation criteria
are critical in determining the feasibility of serving several
purposes simultaneously.
c. Def i ni ng the multipurpose project. One of the
factors that make detailed sequential analyses of multipurpos e operation difficult during planning studies is that
sufficient data on various water demands are either not
available or not of comparable quality for all purposes. To
adequately define the multipurpose operation, the analyses
must include information on the magnitude and seasonal
variations of each demand, long-term changes in demands,
rel ative priority of each use, and shortage tolerances.
Information on magnitude and seasonal variation in
demands and on long-term variations in demands is usually
m ore readily available than information on relative
priorities among uses and on shortage tolerances. If
information on priorities and shortages is not available from
t he various users, one can make several assumptions
concerning the priorities and perform sequential routing
s t udi es for each set of assumptions. The results of these
studies can determine the consequences of various priorities
to potential water users. It may be possible for the potential
users to adopt a priority arrangement based on the value of
the water for the various demands.
d. Success of multipurpose projects. The success of
multipurpose operation also depends on the formulation of
operational rules that ensure that water in the proper
quantities and qualities is available for each of the purposes
at t he proper time and place. Techniques for formulating
operational rules are not fixed, but the logical approach
involves determining the seasonal variation of the
flood-control space requirement, and the seasonal variation
of conservation requirements, formulation of general
operational rules that satisfy these requirements, and
det ai l ed testing of the operational rules to ascertain the
adequacy of the plan for each specific purpose.
e. Multipurpose project rules. The judgment of an
experienced hydrologic engineer is invaluable in the initial
formulation and subsequent development and testing of
operational rules. Although the necessary rules cannot be
completely developed until most of the physical dimensions
of t he project are known, any tendency to discount the
importance of operational rules as a planning variable
should be resisted because of the important role they often
as s um e i n the feasibility of multipurpose projects. As a
m i ni mum, the operational rules used in a planning study
s houl d be sufficiently refined to assist the engineer in
evaluating the suitability of alternative projects to satisfy
water demands for specified purposes.
3-2. Relativ e Priorities of Proj ect Functions
a. Developing project rules. As indicated above, the
use of operational rules based on the relative priorities
am o ng the project purposes appears to offer the best
app roach to multipurpose operational problems. The
degree of success that can be realized depends on a realistic
priority system that accurately reflects the relative value of
water from the project for a given purpose at a given time.
Unless a realistic priority system is used to develop the
operational rules, it will not be possible to follow the rules
duri ng the project life because the true priorities may
con trol the operational decisions and prevent the project
from supplying the services it was designed to provide.
b. Typical system priorities. Priorities among the
various water resource purposes vary with locale, water
rights, the need for various types of water use, the legal and
political considerations, and with social, cultural, and
environmental conditions. Although these variations make
i t i m possible to specify a general priority system, it is
us eful to identify a set of priorities that would be typical
under average conditions. In such a situation, operation for
the safety of the structure has the highest priority unless the
EM 1110-2-1420
31 Oct 97
consequences of failure of the structure are minor (which is
seldom the case). Of the functional purposes, flood control
m us t have a high priority, particularly where downstream
levees, bridges, or other vital structures are threatened. It is
not unusual for conservation operations to cease entirely
during periods of flood activity if a significant reduction in
flooding can be realized thereby. Among the conservation
purposes, municipal and industrial water supply and
hydroelectric power generation are often given a high
priority, particularly where alternative supplies are not
readily available. After those purposes, other project
purpo ses usually have a somewhat lower priority because
temporary shortages are usually not disastrous. It should be
em phas i zed again that there can be marked exceptions to
t hes e relative priorities. There are regional differences in
relative needs and, legal and institutional factors may
greatly affect priorities.
c. C ompl ex system priorities. In complex reservoir
s ys t ems, with competing demands and several alternative
proj ect s to meet the demand, the relative priority among
projects and purposes may not be obvious. The operation
rules, which can be evaluated with detailed simulation, may
not be known or may be subject to criticism. In these
situations, it may be useful to apply a system analysis based
on consistent values for the various project purposes. The
results of the analysis could suggest an operational strategy
which can be tested with more detailed analysis. Chapter 4
presents information and approaches for system analysis.
3 -3. Managing Competitiv e and Complementary
a. Identifying interactions between purposes. Before
operation rules can be formulated, the adverse (competit i ve) and the beneficial (complementary) interactions
bet ween purposes must be identified. The time of occurrence of the interactions is often as important as the degree
of interaction, particularly if one or more of the water uses
has s ignificant variations in water demand. In supplying
water from a single reservoir for several purposes with
s eas o nally varying demands, it is possible for normally
complementary purposes to become competitive at times
due to differences in their seasonal requirements.
b. Allocating storage space. When several purposes
are t o be served from a single reservoir, it is possible to
allocate space within certain regions of the reservoir storage
for each of the purposes. This practice evolved from
projects that served only flood control and one conservation
purpose because it was necessary to reserve a portion of the
reservoir storage for storing floodwater. It is still necessary
have a specific allocation of flood-control
s t orag e space (although the storage reservation can vary
s easonally) because of the basic conflict between reserving
em pty storage space for regulating potential floods and
filling that space to meet future water supply requirements.
However, applying specific storage allocations or
reservations for competing conservation purposes should be
kept to a minimum because it reduces operational
c. Operational conflicts. Allocation of specific storage space to several purposes within the conservation pool
can result in operational conflicts that might make it imposs i bl e or very costly to provide water for the various purposes in the quantities and at the time they are needed. The
con cept of commingled or joint-use conservation storage
for all conservation purposes with operational criteria to
maximize the complementary effects and minimize the
com p etitive effects is far easier to manage and, if carefully
designed, will provide better service for all purposes.
W here t he concept of joint-use storage is used, the
operational criteria should be studied in the planning
process in such a way that the relative priorities of the
various purposes are taken into account. This allows
careful evaluation of a number of priority systems and
operational plans. The operational decisions that result
from such disputes are frequently not studied in enough
detail (from the engineering point of view), and as a result,
t he abi lity of the project to serve some purposes may be
seriously affected.
3-4. Operating Concepts
a. Operating goals. Reservoir operating goals vary
with the storage in the reservoir. The highest zone in the
reservoir is that space reserved at any particular time for the
control of floods. This zone includes the operational floodcont ro l space and the surcharge space required for the
passage of spillway flows. Whenever water is in this zone
it must be released in accordance with flood-control
requi rements. The remaining space can be designated as
conservation space. The top zone of conservation space
may include storage that is not required to satisfy the firm
conservation demands, including recreational use of the
reservoir. Water in this space can be released as surplus to
s erve needs or uses that exceed basic requirements. The
m i ddl e zone of conservation space is that needed to store
wat er to supply firm water needs. The bottom zone of
conservation space can be termed buffer space, and when
operation is in this zone the firm services are curtailed in
order to prevent a more severe shortage later. The bottom
zone of space in the reservoir is designated as the minimum
poo l reserved for recreation, fish, minimum power head,
sediment reserve, and other storage functions.
EM 1110-2-1420
31 Oct 97
b. Storage zone boundaries. The boundaries between
storage zones may be fixed at a constant level or they may
vary seasonally.
In general, the seasonally varying
bou ndaries offer the potential for a more flexible operating
pl an that can result in higher yields for all purposes.
However, the proper location of the seasonal boundaries
requi res more study than the location of a constant
boundary. This is discussed in more detail in Chapter 11.
Furthermore, an additional element of chance is introduced
when the boundaries are allowed to vary, because the joint
use of storage might endanger firm supplies for one or more
s pecific purposes. The location of the seasonally varying
boundaries is determined by a process of formulating a set
of boundaries and attendant operational rules, testing the
s cheme by a detailed sequential routing study, evaluating
the outcome of the study, changing the rules or boundaries
if necessary, and repeating the procedure until a satisfactory
operation results.
c. Demand schedules. Expressing demand schedules
as a function of the relative availability of water is another
means of incorporating flexibility and relative priority in
operational rules. For example, the balance between hydro
and thermal power generation might well be a continuous
function of available storage. As another example, it might
be possible to have two or more levels of navigation service
or l engths of navigation season with the actual level of
s ervi ce or length of season being dependent upon the
availability of water in the reservoir. By regulating the
level of supply to the available water in the reservoir, users
can plan emergency measures that will enable them to
withstand partial reductions in service and thereby avoid
com p lete cessation of service, which might be disastrous.
Terms such as desired flow and minimum required flow for
navigation can be used to describe two levels of service.
d. Levels of service. There can be as many levels of
service as a user desires, but each level requires criteria for
determining when the level is to be initiated and when it is
t o be t erminated. The testing and development of the
cri teria for operating a multipurpose project with several
purposes and several levels of service are accomplished by
detailed sequential routing studies. Because the developm ent and testing of these criteria are relatively difficult,
the number of levels of service should be limited to the
m i n imum number needed to achieve a satisfactory plan of
e. Buf f er storage. Buffer storage or buffer zones are
regi o ns within the conservation storage where operational
rul es effect a temporary reduction in firm services. The
t wo primary reasons for temporarily reducing services are
to ensure service for a high-priority purpose while
el i minating or curtailing services for lower-priority
purpo ses, and to change from one level of service for a
gi ven purpose to a lower level of service for that same
purpose when storage levels are too low to ensure the
continuation of firm supplies for all purposes. As with the
ot her techniques for implementing a multipurpose operat i on, t he amount of buffer storage and the location of the
bou ndaries cannot be determined accurately except by
successive approximations and testing by sequential routing
3-5. Construction and Physical Operation
a. General. In addition to hydrologic determinations
discussed above, a number of important hydrologic
determinations are required during project construction and
during project operation for ensuring the integrity of the
project and its operation.
b. C of f erdams. From a hydrologic standpoint,
during construction the provisions for streamflow diversion
are a primary concern. If a cofferdam used for dewatering
the work area is overtopped, serious delays and additional
construction costs can result.
In the case of high
cofferdams where substantial poundage occurs, it is
possible that failure could cause major damage in
downstream areas. Cofferdams should be designed on the
same principles as are permanent dams, generally on the
basis of balancing incremental costs against incremental
benefits of all types. This will require flood frequency and
hypothetical flood studies, as described in Chapters 6 and 7
of t hi s manual. Where major damage might result from
cofferdam failure, a standard project flood (SPF) or even a
probable maximum flood (PMF) may be used as a primary
basis for design.
c. Over t opping. Where a major dam embankment
m ay be subject to overtopping during construction, the
di version conduit capacity must be sufficient to regulate
floods that might occur with substantial probability during
the critical construction period. It is not necessary that the
regulated releases be nondamaging downstream, but it is
vital that the structure remain intact. An explicit evaluation
of risk of embankment failure and downstream impacts
duri n g construction should be presented in the design
d. Conduits, spillways, and gates. Conduits, spillways, and all regulating gates must be functionally adequat e to accomplish project objectives. Their sizes,
dep endability, and speed of operation should be tested
us i ng recorded and hypothetical hydrographs and anticipated hydraulic heads to ensure that they will perform
properly. The nature of stilling facilities might be dictated
by hydrologic considerations if frequency and duration of
EM 1110-2-1420
31 Oct 97
high outflows substantially influence their design.
T he necessity for multilevel intakes to control the quality
of reservoir releases can be assessed by detailed
reservoir stratification studies under all combinations of
hydrologic and reservoir conditions.
Techniques for
co nducting reservoir stratification studies are discussed in
EM 1110-2-1201.
e. Design. The design of power facilities can be
greatly influenced by hydrologic considerations, as discussed in Chapter 11 of this manual and EM 1110-2-1701.
General considerations in the hydrologic design of
spillways are discussed in Chapter 10 and more detailed
information is presented in Chapter 14 herein.
f. Extreme floods. Regardless of the reservoir purposes, it is imperative that spillway facilities provided will
ens ure the integrity of the project in the event of extreme
fl ood s. Whenever the operation rules of a reservoir are
s ubstantially changed, spillway facilities should be
reviewed to ensure that the change in project operation does
not adversely alter the capability to pass extreme floods
wi t hout endangering the structure. The capability of a
spillway to pass extreme floods can be adversely affected
by changes in operation rules that actually affect the flood
operation itself or by changes that result in higher pool
stages during periods of high flood potential.
g. Speci al operating rules. A number of situations
m i ght require special operating rules. For example, operat ing rules are needed for the period during which a reservoi r is initially filling, for emergency dewatering of a
reservoir, for interim operation of one or more components
in a system during the period while other components are
und er construction, and for unanticipated conditions that
seem to require deviation from established operating rules.
T he need for operation rules during the filling period is
especially important because many decisions must be based
on the filling plan. Among the important factors that are
dependent upon the filling schedule are the on-line date for
power generating units, the in-service dates for various
purpos es such as water supply and navigation, and the
effective date for legal obligations such as recreation
h. Specification of monitoring facilities. One of the
more important considerations in the hydrologic analysis of
any reservoir is the specification of monitoring facilities,
including streamflow, rainfall, reservoir stage, and other
hydrologic measurements. These facilities serve two basic
purp oses:
to record all operations and to provide
information for operation decisions. The former purpose
s atisfies legal requirements and provides data for future
studies. The latter purpose may greatly increase the project
effectiveness by enabling the operating agency, through
rel iable forecasts of hydrologic conditions, to increase
operation efficiency. Hydrologic aspects of monitoring
faci l i ties and forecasts will be presented in a new EM on
hydrologic forecasting.
i. St ream gauges. Because gauged data are most
important during flood events, special care should be taken
in locating the gauge. Stream gauges should not be located
on bri d ges or other structures that are subject to being
washed out. To the extent possible, the gauges should be
capabl e of working up through extreme flood events, and
stage-discharge relationships should be developed up to
that level. The gauge should have reasonable access for
check ing and repair during the flood. Reservoir spilling,
l ocal flooding, and backwater effects from downstream
tributaries should all be considered when finding a suitable
location. More detailed information on stream gauges can
be found in many USGS publications, such as Carter and
Davidian (1968), Buchanan and Somers (1968 and 1969),
or Smoot and Novak (1969).
3-6. General Study Procedure
As i ndicated earlier, there is no fixed procedure for developing reservoir operational plans for multipurpose projects;
however, the general approach that should be common to
all cases would include the following steps:
a. S urvey the potential water uses to be served by
t he project in order to determine the magnitude of each
dem and and the seasonal and long-term variations in the
demand schedule.
b. Develop a relative priority for each purpose and
determine the levels of service and required priority that
will be necessary to serve each purpose. If necessary, make
s equ ential studies illustrating the consequences of various
alternative priority systems.
c. Es t ablish the seasonal variation of flood-control
space required, using procedures discussed in Chapter 10.
d. Es tablish the total power, water supply, and
l ow-flow regulation requirements for competitive purposes
during each season of the year.
e. Establish preliminary feasibility of the project
based on physical constraints.
f. Es t ablish the seasonal variation of the storage
requ irement to satisfy these needs, using procedures
described in Chapter 11.
EM 1110-2-1420
31 Oct 97
g. Determine the amount of storage needed as a
minimum pool for power head, recreation, sedimentation
reserve, and other purposes.
h. Us i n g the above information, estimate the size of
reservoir and seasonal distribution of space for the various
purposes that would satisfy the needs. Determine the
reservoir characteristics, including flowage, spillway,
power plant, and outlet requirements.
i. T es t and evaluate the operation of the project
through the use of recorded hydrologic data in a sequential
routing study to determine the adequacy of the storage
estimates and proposed rules with respect to the operational
objectives for each purpose. If the record is short, supplement it with synthetic floods to evaluate flood storage
reserves. If necessary, make necessary changes and repeat
t es t ing, evaluating,
operation is obtained.
j. Test proposed rules of operation by using sequent i al routing studies with stochastic hydrologic data to
eval uate the possibility of historical bias in the proposed
k. Determine the needs for operating and monitoring
equipment required to ensure proper functional operation of
the project.
l. As detailed construction plans progress, evaluate
cofferdam needs and protective measures needed for the
i nt egrity of project construction, particularly diversion
capacity as a function of dam construction stage and flood
threat for each season.
EM 1110-2-1420
31 Oct 97
Chapter 4
Reservoir Systems
4-1. Introduction
Water resource systems should be designed and operated
for the most effective and efficient accomplishment of
overal l objectives. The system usually consists of reservoirs, power plants, diversions, and canals that are each
con structed for specific objectives and operated based on
existing agreements and customs. Nevertheless, there is
considerable latitude in developing an operational plan for
any water resource system, but the problem is greatly
complicated by the legal and social restrictions that
ordinarily exist.
a. Mat hematical modeling. Water resource system
operation is usually modeled mathematically, rather than
with physical models. The mathematical representation of
a water resource system can be extremely complex. Operations research techniques such as linear programming and
dynam ic programming can be applied to a water resource
s ys t em; however, they usually are not capable of
incorporating all the details that affect system outputs. It is
us ually necessary to simulate the detailed sequential
operation of a system, representing the manner in which
each element in the system will function under realistic
conditions of inputs and requirements on the system. The
s i mulation can be based on the results from the optimization of system outputs or repeated simulations. Success i vel y refining the physical characteristics and operational
rules can be applied to find the optimum output.
b. Inputs and requirements. A factor that greatly
complicates the simulation and evaluation of reservoir
system outputs is the stochastic nature of the inputs and of
t he requirements on the system. In the past, it has been
cu s t o mary to evaluate system accomplishments on the
as s um p tion that a repetition of historical inputs and
requirements (adjusted to future conditions) would adequately represent system values. However, this assumption
has been demonstrated to be somewhat deficient. It is
des irable to test any proposed system operation under a
great many sequences of inputs and requirements. This
requires a mathematical model that will define the frequency and correlation characteristics of inputs and
requirements and that is capable of generating a number of
l ong sequences of these quantities.
Concepts for
accomplishing this are discussed in paragraph 5-5.
4-2. System Description
a. Simulating system operation. Water resource
s ys tems consist of reservoirs, power plants, diversion
structures, channels, and conveyance facilities. In order to
simulate system operation, the system must be completely
described in terms of the location and functional
characteristics of each facility. The system should include
all components that affect the project operation and provide
the required outputs for analysis.
(1) Reservoirs. For reservoirs, the relation of surface
area and release capacity to storage content must be
des cri bed. Characteristics of the control gates on the
outlets and spillway must be known in order to determine
constraints on operation. The top-of-dam elevation must be
s peci fi ed and the ability of the structure to withstand
overtopping must be assessed.
(2) Downstream channels. The downstream channels
must be defined. Maximum and minimum flow targets are
requ ired. For short-interval simulation, the translation of
fl ow through the channel system is modeled by routing
criteria. The travel time for flood flow is important in
determining reservoir releases and potential limits for flood
control operation to distant downstream locations.
(3) Power plants. For power plants at storage reservoi rs , the relation of turbine and generation capacity to
head m us t be determined. To compute the head on the
plant, the relation of tailwater elevation to outflow must be
known. Also, the relation of overall power plant efficiency
t o head is required. Other characteristics such as turbine
l eakage and operating efficiency under partial load are also
(4) Diversion structures. For diversion structures,
m axi m um diversion and delivery capacity must be established. The demand schedule is required, and the consumptive use and potential return flow to the system may be
important for the simulation.
b. Preparing data. While reservoir system data must
be defined in sufficient detail to simulate the essence of the
physical system, preparing the required hydrologic data
m ay req uire far more time and effort. The essential flow
dat a are required for the period of record, for major flood
events, and in a consistent physical state of the system.
Flow records are usually incomplete, new reservoirs in the
system change the flow distribution, and water usage in the
EM 1110-2-1420
31 Oct 97
time. Developing a consistent hydrologic data series, making maximum use of the available information, is discussed
in Chapter 5.
4-3. Operating Obj ectiv es and Criteria
a. Us er services. Usually, there is a fixed objective
for each function in a water resource system. Projects are
cons t ructed and operated to provide services that are
cou nted on by the users. In the case of power generation
and water supply, the services are usually contracted, and it
i s es sential to provide contracted amounts insofar as
pos sible. Services above the contracted amounts are
ordinarily of significantly less value. Some services, such
as fl ood control and recreation, are not ordinarily covered
by con tracts. For these, service areas are developed to
provi de the degree of service for which the project was
Rul es for services. Shortages in many of the
s erv ices can be very costly, whereas surpluses are usually
of m i n or value. Accordingly, the objectives of water
resource system operational are usually fixed for any particular plan of development. These are expressed in terms of
operational rules that specify quantities of water to be
released and diverted, quantities of power to be generated,
reservoir storage to be maintained, and flood releases to be
made. These quantities will normally vary seasonally and
wi t h the amount of storage water in the system. Rule
curves for the operation of the system for each function are
developed by successive approximations on the basis of
performance during a repetition of historical streamflows,
adj u sted to future conditions, or on the basis of synthetic
stream flows that would represent future runoff potential.
4-4. System Simulation
T he evaluation of system operation under specified operat i on rules and a set of input quantities is complex and
requires detailed simulation of the operation for long
peri o ds of time. This is accomplished by assuming that
s t eady-state conditions prevail for successive intervals of
time. The time interval must be short enough to capture the
details that affect system outputs. For example, average
m ont hl y flows may be used for most conservation
purposes; however, for small reservoirs, the flow variation
within a month may be important. For hydropower reservoirs, the average monthly pool level or tailwater elevation
may not give an accurate estimate of energy production.
To simulate the operation during each interval, the simulation solves the continuity equation with the reservoir
release as the decision variable. The system is analyzed in
an upstream-to-downstream direction. At each pertinent
l ocat i o n, requirements for each service are noted, and the
reservoirs at and above that location are operated in such a
way as to serve those requirements, subject to system
constraints such as outlet capacity, and channel capacity,
and reservoir storage capacity.
As the computation
procedure progresses to downstream locations, the tentative
rel ease decisions made for upstream locations become
i ncreasingly constraining. It often becomes necessary to
assign priorities among services that conflict. Where power
generation causes flows downstream to exceed channel
capacity, for example, a determination must be made as to
whether to curtail power generation. If there is inadequate
water at a diversion to serve both the canal and river
requirements, a decision must be made.
4-5. Flood-Control Simulation
Flood discharge can change rapidly with time. Therefore,
s t eady-state conditions cannot be assumed to prevail for
long periods of time (such as one month). Also, physical
constraints such as outlet capacity and the ability to change
gate settings are more important. The time translation for
fl ow and channel storage effects cannot ordinarily be
ignored. Consequently, the problem of simulating the
fl ood-control operation of a system can be more complex
than for conservation.
a. Computational interval. The computation interval
necessary for satisfactory simulation of flood operations is
usually on the order of a few hours to one day at the most.
Sometimes intervals as short as 15 or 30 min are necessary.
It is usually not feasible to simulate for long periods of
time, such as the entire period of record, using such a short
computation interval. However, period-of-record may be
unnecessary because most of the flows are of no
consequence from a flood-control standpoint. Accordingly,
simulation of flood-control operation is usually made only
for important flood periods.
b. St arting conditions. The starting conditions for
simulating the flood-control operation for an historic flood
period would depend on the operation of the system for
conservation purposes prior to that time. Accordingly, the
conservation operation could be simulated first to establish
the state of the system at the beginning of the month during
whi ch t he flood occurred as the initial conditions for the
fl ood simulation. However, the starting storage for flood
operation should be based on a realistic assessment of
likely future conditions. If it is likely that the conservation
pool is full when a flood occurs, then that would be a better
starting condition to test the flood-pool capacity. It may be
possible that the starting pool would be higher if there were
s everal storms in sequence, or if the flood operation does
EM 1110-2-1420
31 Oct 97
not start the instant excessive inflows raise the pool level
into flood-control space.
c. Hi s t oric sequences. While simulating historic
sequences are important, future floods will be different and
occur in different sequences. Therefore, the analysis of
fl ood operations should utilize both historic and synthetic
floods. The possibility of multiple storms, changes in the
upstream catchment, and realistic flood operation should be
i ncl uded in the analysis. Chapter 7 presents flood-runoff
analysis and Chapter 10 presents flood-control storage
d. Ups tream-to-downstream solution. If the operat i on of each reservoir in a system can be based on
conditions at or above that reservoir, an upstream-todownstream solution approach can establish reservoir
rel eases, and these releases can be routed through channel
reaches as necessary in order to obtain a realistic simulation. Under such conditions, a simple simulation model is
capab le of simulating the system operation with a high
degree of accuracy. However, as the number of reservoirs
and downstream damage centers increase, the solution
becomes far more complex. A priority criteria must be
established among the reservoirs to establish which should
release water, when there is a choice among them.
e. Combination releases. The HEC-5 Simulation of
Fl ood Control and Conservation Systems (HEC 1982c)
computer program can solve for the combination of releases
at upstream reservoirs that will satisfy channel capacity
cons traints at a downstream control point, taking into
accoun t the time translation and channel storage effects,
and that will provide continuity in successive time intervals.
T he time translation effects can be modeled with a choice
of hydrologic routing methods. Reservoir releases are
determined for all designated downstream locations, subject
t o operation constraints. The simulation is usually
performed with a limited foresight of inflows and a cont i ng ency factor to reflect uncertainty in future flow values.
T he con cept of pool levels is used to establish priorities
am on g projects in multiple-reservoir systems. Standard
out put includes an indicator for the basis of reservoir
release determination, along with standard simulation output of reservoir storage, releases, and downstream flows.
f. Period-of-record flows. Alternatively, a single
time interval, such as daily, can be used to simulate periodof-record flows for all project purposes. This approach is
routinely used in the Southwestern Division with the comput er program “ Super” (USACE 1972), and in the North
Pacific Division with the SSARR program (USACE 1991).
The SSARR program is capable of simulation on variable
time intervals.
4-6. Conserv ation Simulation
W hi l e the flood-control operation of a reservoir system is
sensitive to short time variations in system input, the
operation of a system for most conservation purposes is
usually sensitive only to long-period streamflow variations.
Hi s torically, simulation of the conservation operation of a
water resource system has been based on a relatively long
com put ation interval such as a month. With the ease of
computer simulation and available data, shorter
com p utational intervals (e.g., daily) can provide a more
accu rate accounting of flow and storage. Some aspects of
t he conservation operation, such as diurnal variations in
power generation in a peaking project, might require even
s hort er computational intervals for selected typical or
critical periods to define important short-term variations.
a. Hydropower simulation. Hydropower simulation
requires a realistic estimate of power head, which depends
on reservoir pool level, tailwater elevation, and hydraulic
energy losses. Depending on the size and type of reservoir,
there can be considerable variation in these variables.
General ly, the shorter time intervals will provide a more
accurate estimate of power capacity and energy
b. Evapo ration and channel losses. In simulating
t he operation of a reservoir system for conservation, the
t i m e of travel of water between points in the system is
usually ignored, because it is small in relation to the typical
computation interval (e.g., monthly or weekly). On the
other hand, evaporation and channel losses might be quite
i m po rtant; and it is sometimes necessary to account for
such losses in natural river channels and diversion canals.
c. Rul e curves. Rule curves for the operation of a
reservoir system for conservation usually consist of standard power generation and water supply requirements that
will be served under normal conditions, a set of storage
levels that will provide a target for balancing storage among
the various system reservoirs, and maximum and minimum
permissible pool levels for each season based on flood
cont rol, recreation, and other project requirements. Often
some criteria for decreasing services when the system
reservoir storage is critically low will be desirable.
4-7. System Pow er Simulation
W here a number of power plants in the water resource
s ys t em serve the same system load, there is usually cons i derab le flexibility in the selection of plants for power
generation at any particular time. In order to simulate the
operat ion of the system for power generation, it is necess ary to specify the overall system requirement and the
EM 1110-2-1420
31 Oct 97
minimum amount of energy that must be generated at each
plant during each month or other interval of time. Because
t he entire system power requirement might possibly be
s upplied by incidental generation due to releases made for
ot her purposes, it is first necessary to search the entire
system to determine generation that would occur with only
minimum power requirements at each plant and with all
requirements throughout the system for other purposes. If
insufficient power is generated to meet the entire system
load in this manner, a search will be made for those power
reservoirs where storage is at a higher level, in relation to
t he rul e curves, than at other power reservoirs. The
additional power load requirement will then be assigned to
t hos e reservoirs in such a manner as to maintain the
reservoir storage as nearly as possible in conformance with
the rule curves that balance storage among the reservoirs in
t he most desirable way. This must be done without
as signing more power to any plant than it can generate at
overl o ad capacity and at the system load factor for that
i nt erval. EM 1110-2-1701 paragraph 5-14, describes
hydropower system analysis.
4-8. Determination of Firm Yield
If the yield is defined as the supply that can be maintained
t hrou ghout the simulation period without shortages, then
t he pro cess of computing the maximum yield can be
expedited. This is done by maintaining a record of the
m i nimum reserve storage (if no shortage has yet occurred)
or of the amount of shortage (if one does occur) in relation
to the total requirement since the last time that all reservoirs
were full. The surplus or shortage that existed at the end of
any computation interval would be expressed as a ratio of
t he supply since the reservoirs were last full, and the
minimum surplus ratio (if no shortage occurs) or maximum
shortage ratio (if a shortage does occur) that occurs during
t he ent ire simulation period would be used to adjust the
target yield for the next iteration. This basic procedure for
com p uting firm yield is included in the HEC-5 computer
program. Additionally, the program has a routine to make
an initial estimate of the critical period and expected yield.
After the yield is determined using the critical period, the
program will evaluate the yield by performing a simulation
wi th the entire input flow record. Chapter 12 describes
storage-yield procedures.
4-9. Deriv ation of Operating Criteria
A plan of development for a water resource system consists
not only of the physical structures and their functional
characteristics but also of the criteria by which the system
wi ll be operated. In order to compare alternative plans of
devel o pment, it is necessary that each plan be operated
optimally. The derivation of optimal operation criteria for a
wat er resource system is probably more difficult than the
derivation of optimum configuration and unit sizes because
any small change in operation rules can affect many
functions in the system for long periods of time and in very
subtle ways.
a. Simulation. Operation criteria generally consist
of release schedules at reservoirs, diversion schedules at
control points, and minimum flows in the river at control
points, in conjunction with reservoir balancing levels that
defi ne the target storage contribution among the various
reservoirs in the system. All of these can vary seasonally,
and target flows can vary stochastically. Once the unit
s i zes and target flows are established for a particular plan
of development, a system of balancing levels must be
devel oped. The system response to a change in these
balancing levels is a complicated function of many system,
input, and requirement characteristics. For this reason, the
devel o pment of a set of balancing levels is an iteration
process, and a complete system simulation must be done for
each iteration.
(1) When first establishing balancing levels in the
reservoir system, it usually is best to simulate system
operation only for the most critical periods of historical
streamflows. The final solution should be checked by
simulation for long periods of time. The balancing levels
defi ni ng the flood-control space are first tentatively establ i s hed on the basis of minimum requirements for flood
cont rol storage that will provide the desired degree of
protection. Preliminary estimates of other levels can be
established on the basis of reserving the most storage in the
smaller reservoirs, in those reservoirs with the least amount
of runoff, and in those reservoirs that supply operation services not producible by other reservoirs.
(2) After a preliminary set of balancing levels is
established, they should be defined approximately in terms
of a minimum number of variables. The general shape and
spacing of levels at a typical reservoir might be defined by
t he use of four or five variables, along with rules for
com pu ting the levels from those variables. Variations in
l evel s among reservoirs should be defined by one or two
variables, if possible, in order to reduce the amount of work
required for optimization to an acceptable quantity.
(3) Optimization of a set of balancing levels for operat i on al rule curves can be accomplished by successive
approximations using a complete system simulation computation for critical drought periods. However, the procedures are limited to the input specifications of demands and
storage allocation. While one can compare simulation
results and conclude one is better than another based on
EM 1110-2-1420
31 Oct 97
performance criteria, there is no way of knowing that an
optimum solution has been achieved.
b. Optimization. While water resource agencies have
generally focused on simulation models for system analysis,
the academic community and research literature have
em p hasized optimization and stochastic analysis
techniques. Research performed at HEC (HEC 1991b) has
found a proliferation of papers on optimization of reservoir
system operations written during the past 25 years,
pri marily by university researchers. There still remains a
cons iderable gap between the innovative applications
reported in the literature and the practices followed by the
agencies responsible for water resource development. One
basic problem is that many of the reported applications are
uniquely formulated to solve a specific problem for a given
There is a general view that the models
performance, or the methods assumptions, would not
sufficiently evaluate a different problem and system.
c. Prescriptive reservoir model. HEC has developed
a s ys t em analysis tool based on a network flow model
(HEC 1991a). The Prescriptive Reservoir Model (HECPRM) will identify the water allocation that minimizes poor
performance for all defined system purposes. Performance
i s measured with analyst-provided functions of flow or
s t orage or both. The physical system is represented as a
net work , and the allocation problem is formulated as a
m i n imum-cost network flow problem. The objective
functions for this network problem are convex, piecewiselinear approximations of the summed penalty functions for
each project purpose (HEC 1991d).
(1) S ys tems have been analyzed in studies on the
M i ssouri River (HEC 1991d) and the Columbia River
(HEC 1991f). A preliminary analysis of the Phase I
Missouri River study has developed initial methodologies
for developing operation plans based on PRM results
(HEC 1992b).
Continued application experience is
required to define generalized procedures for these
The primary advantages for the HEC-PRM
approach are the open state of the system and the required
penal t y functions for each system purpose. There are no
rul e curves or details of storage allocation, only basic
phy sical constraints are defined. The reservoir system
information defines maximum and minimum storage in the
reservoirs and the linking of the system through the
net work of channels and diversions. The other primary
reservoir data is traditional period-of-record monthly flows
for the system.
(3) The development of the penalty functions requires
an economic evaluation of the values to be placed on flow
and storage in the system. The process is difficult and there
are disagreements on the values, due to the difficulty of
defi ning values for some purposes. However, the process
does provide a method for defining and reviewing the
purposes and their relative values.
(4) The primary disadvantage of the HEC-PRM is that
the monthly flow data and lack of channel routing limit its
app lication for short interval simulation, such as flood
cont rol and peaking hydropower.
Additionally, the
optimized solution is provided in terms of period-of-record
flows and storage; however, the basis for the system
operation are not explicitly defined. The post-processing of
t he results requires interpretation of the results in order to
develop an operation plan that could be used in basic
simulation and applied operation. More experience with
this analysis of results is still required to define these
4-10. System Formulation Strategies
a. Determining the best system. A system is best for
the national income criteria if it results in a value for system
net benefits that exceeds that of any other feasible system.
Excep t where noted, the following discussion was
devel op ed in a paper presented at the International Comm i ssions on Large Dams Congress (Eichert and
Davis 1976). For a few components, analysis of the number of alternative systems that are feasible is generally
m anageable, and exhaustive evaluation provides the
strategy for determining the best system. When the number
of components is more than just a few, then the exhaustive
evaluation of all feasible alternative systems cannot
practically be accomplished. In this instance, a strategy is
needed that reduces the number of system alternatives to be
evaluated to a manageable number while providing a good
chance of identifying the best system. System analysis does
not permit (maximum net benefit system) for reasonably
com pl ex systems even with all hydrologic-economic data
known. An acceptable strategy need not make the absolute
guarantee of economic optimum because seldom will the
optimum economic system be selected as best.
b. Incremental test. The incremental test of the value
of an individual system component is definitive for the
economic efficiency criteria and provides the basis for
s everal alternative formulation strategies.
If existing
res ervoir components are present in the system, then they
define the base conditions. If no reservoirs exist, the base
condition would be for natural conditions. The strategies
EM 1110-2-1420
31 Oct 97
des cribed below are extensions of currently used techniques and are based upon the concept of examining in detail
the performance of a selected few alternative systems. The
performance is assumed to be evaluated generally by
traditional simulation methods, like the use of HEC-5.
c. Reasoned thought strategy. This strategy is predicated upon the idea that it is possible to reason out using
judgment and other criteria, reasonable alternative systems.
The strategy consists of devising through rational thought,
sampling, public opinion, literature search, and
brainstorming, a manageable number of system alternatives
that will be evaluated. No more than 15 to 20 alternative
s ys t ems could be evaluated by detailed simulation in a
practical sense.
(1) The total performance of each system in terms of
econo mic (net benefit) and performance criteria is evaluated by a system simulation. A system (or systems if more
t han one have very similar performance) is selected that
m ax imizes the contribution towards the formulation
obj ectives (those that exhibit the highest value of net
benefits while satisfying the minimum performance criteri a). To confirm the incremental justification of each
component, the contribution of each system component in
the last added position is evaluated. The last added value is
t he difference between the value (net benefits) of the
system with all components in operation and the value (net
benefits) of the system with the last added component
removed. If each component is incrementally justified, as
indicated by the test, the system is economically justified,
and form ulation is complete. If any components are not
incrementally justified, they should be dropped and the last
added analysis repeated.
(2) The system selected by this strategy will be a
feas i ble system that is economically justified. Assuming
the method of devising the alternative systems is rational,
t he chances are good that the major worthwhile projects
will have been identified. On the other hand, the chances
that this system provides the absolute maximum net benefits is relatively small. This strategy would require between
30 and 60 system evaluations for a moderately complex
(15 component) system.
d. First added strategy. This strategy is designed
s uch t hat its successive application will yield the formul at ed system. The performance of the systems, including
t he base components (if any), are evaluated with each
potential addition to the system in the “ first added” position. The component that contributes the greatest value (net
benefit) to the system is selected and added to the base
(1) The analysis is then repeated for the next stage by
com puting the first added value of each component to the
s ystem again, the base now including the first component
added. The strategy is continued to completion by successive application of the first added analysis until no more
component additions to the system are justified.
(2) T he strategy does have a great deal of practical
appeal and probably would accomplish the important task
of i den tifying the components that are clearly good additions to the system and that should be implemented at an
early stage. The strategy, however, ignores any system
value that could be generated by the addition of more than
one component to the system at a time, and this could omit
potentially useful additions to the system. For example, the
situation sometimes exists where reservoirs on, say, two
tributaries above a damage center are justified, but either
one analyzed separately is not, i.e., the system effect is
great enough to justify both. The number of system
anal yses required to formulate a system based on this
s t rat egy could range upwards to 120 for a moderately
complex (15 component) system, which is probably close to
being an unmanageably large number of evaluations.
e. Last added strategy. This strategy, similar to first
added strategy, is designed such that successive application
yields the formulated system. Beginning with all proposed
components to the system, the value of each component in
t he l ast added position is computed. The project whose
del et i o n causes the value (net benefit) of the system to
increase the most is dropped out. The net benefits would
i ncrease if the component is not incrementally justified.
T he strategy is continued through successive staged
appl i cations until the deletion of a component causes the
total system value (net benefits) to decrease.
(1) The last added strategy will also yield a system in
whi ch all components are incrementally justified and in
whi ch the total system will be justified. This strategy
would probably identify the obviously desirable projects, as
woul d t he others. However, its weakness is that it is
s l i gh tly possible, though not too likely, that groups of
projects that would not be justified are carried along
because of their complex linkage with the total system. For
example, the situation sometimes exists where reservoirs on
t wo tributaries above a damage center are not justified
t ogether, but deletion of each from a system that includes
both results in such a great loss in system value that individual analysis indicates neither should be dropped
(2) The number of systems analyses required for this
s t rategy would be similar to the first added strategy
EM 1110-2-1420
31 Oct 97
requiring perhaps 10-20 percent more evaluations. Twentyt wo last added analyses were made in the four stages
required to select four new projects out of seven
al t ern atives. This strategy is more efficient than the first
added if the majority of the potential system additions are
good ones.
f. Branch-and-bound enumeration.
“ Branch-andbound enumeration is a general-purpose technique for
identifying the optimal solution to an optimization problem
without explicitly enumerating all solutions,” (HEC 1985a).
T he technique provides a framework to evaluate
independent alternatives by dividing the entire set into
s ubs et s for evaluation. The method has been applied in
res ource planning to problems of sizing, selecting,
sequencing, and scheduling projects. HEC has developed a
t rai ning document illustrating the application to flooddamage-mitigation plan selection (HEC 1985). Additional l y, HEC Research Document No. 35 (Bowen 1987)
i l l us trates an application for reservoir flood control plan
selection using computer program HEC-5 for reservoir
s i m u lation. The procedure can use any criteria for evaluat i on and supports detailed simulation in the analysis
needed to provide a reasonable degree of protection, using
procedures described in Chapter 10. Distribute this storage
in a reasonable way among contemplated reservoirs in order
t o obtain a first approximation of a plan for flood control.
Include approximate rule curves for releasing some or all of
t hi s storage for other uses during the nonflood season
where appropriate.
c. Determine approximately for each tributary,
where appropriate, the total water needed each month for
al l conservation purposes and attendant losses, and, using
procedures described in Chapter 11, estimate the storage
need ed on each principle tributary for conservation
services. Formulate a basic plan of development including
detailed specification of all reservoir, canal, channel, and
powerplant features and operation rules; all flow
requirements; benefit functions for all conservation
s ervi ces; and stage-damage functions for all flood damage
index locations. Although this part of plan formulation is
not entirely a hydrologic engineering function, a satisfactory first approximation requires good knowledge of runoff
characteristics, hydraulic structure characteristics and
l i mitations, overall hydroelectric power characteristics,
engineering feasibility, and costs of various types of structures, and relocations.
4-11. General Study Procedure
There is no single approach to developing an optimum plan
of improvement for a complex reservoir system. Ordinarily
m any s ervices are fixed and act as constraints on system
operation for other services. In many cases, all but one
service is fixed, and the system is planned to optimize the
output for one remaining service, such as power generation.
It should also be recognized that most systems have been
devel op ed over a long period of time and that many
services are in fact fixed, as are many system features.
Nevertheless, an idealized general study procedure is
presented below:
a. Prepare regional and river-system topographic
maps showing locations of hydrologic stations, existing and
cont emplated projects, service and damage areas, and
pertinent drainage boundaries. Obtain all precipitation,
evaporation, snowpack, hydrograph timing and runoff data
pertinent to the project studies. Obtain physical and operational data on existing projects. Construct a normal seasonal isohyetal map for the river basin concerned.
b. F or each location where flood protection is to be
provided, estimate approximately the nondamaging flow
capacity that exists or could be ensured with minor channel
and l ev ee improvements. Estimate also the amount of
storage (in addition to existing storage) that would be
d. Us i n g the general procedures outlined in Part 2,
dev elop flood frequencies, hypothetical flood hydrographs,
and stage-discharge relations for unregulated conditions
and for the preliminary plan of development for flood
control. It may be desirable to do this for various seasons
of the year in order to evaluate seasonal variation of floodcont rol space. Evaluate the flood-control adequacy of the
plan of development, using procedures described in
paragraph 4-5 and Chapter 10, modify the plan, as necess ary, to improve the overall net benefits for flood control
whi l e preserving basic protection where essential. Each
m odification must be followed by a new evaluation of net
benefits for flood control. Each iteration is costly and timeconsuming; consequently, only a few iterations are feasible,
and considerable thought must be given to each plan
e. For system analysis to determine the best allocat i on of flow and storage for conservation purposes,
consider optimization using a tool HEC-PRM (paragraph 4-9c). The program outputs would then be analyzed
t o i nfer an operation policy that could be defined for
simulation and more detailed analysis. The alternative is to
repeatedly simulate with critical low-flow periods to
develop a policy to meet system goals and then perform a
period-of-record simulation to evaluate total system
EM 1110-2-1420
31 Oct 97
f. Consider generating synthetic sequences of flow to
evaluate the system's performance with different flow
s equences (see paragraph 5-5). Future system flows replicate the period-of-record. Also, projected changes in the
basin should be factored into the analysis. Typically,
future conditions are estimated at several stages into the
future. The system analysis should be performed for each
stage. While these analyses will take additional time and
effort , they will also provide some indication of how
responsive the system results are to changing conditions.
EM 1110-2-1420
31 Oct 97
EM 1110-2-1420
31 Oct 97
Chapter 5
Hydrologic Engineering Data
5-1. Meteorological Data
The extent of meteorological observations is determined by
the data needs and use and the availability of personnel and
equi pm ent. Data usually recorded at weather stations
include air (sometimes water) temperature, precipitation,
wind, and evaporation. As indicated below, more extensive
recording of various types of data is often made for special
purpos es. The primary source of meteorological data for
the United States is the National Oceanic and Atmospheric
Administration (NOAA) National Climatic Data Center in
Asheville, North Carolina. Data are available from NOAA
i n com puter-readable form as well as published reports.
NOAA publication “ Selective Guide to Climatic Data
S ources” is an excellent reference for data availability.
P ri vate vendor sources employing compact disc (CD)
technology using NOAA records are also available.
a. Storm meteorology. The determination of runoff
potential, particularly flood potential, in areas where
hydrologic data are scarce can be based on a knowledge of
s t orm meteorology. This includes sources of moisture in
t he paths over which the storm has traveled, as well as a
knowledge of the mechanics of storm activity. Derivation
of hydrologic quantities associated with various storms
m us t t ak e into consideration the type of storm, its path,
potential moisture capacity and stability of the atmosphere,
i s obar, wind and isotherm patterns, and the nature and
intensity of fronts separating air masses. These are usually
des cribed adequately in the synoptic charts that are
prepared at regular (usually 6-hr) intervals for weather
forecasting purposes and associated upper-air soundings.
Where such charts are available, it is important that they be
retained as a permanent record of meteorological activity
for use in supplementing information contained in the
regu larly prepared hemispheric charts. These latter charts
s um m arize the daily synoptic situation throughout the
hem i s phere but do not contain all of the data that are of
interest or that would have direct bearing on the derivation
of design criteria.
(1) NOAA monthly publication “ Storm Data” (1959present) documents the time, location, and the meteorologic
characteristics of all reported severe storms or unusual
weat her phenomena. Synoptic maps are published by
NOAA on a weekly basis in “ Daily Weather Maps, Weekly
Series” and on a monthly basis, “ Synoptic Weather Maps,
Daily Series, Northern Hemisphere Sea-Level
500-Millibar Charts and Data Tabulations.”
(2) Storm data including synoptic charts for selected
hi s t o ric storms are included in the “ Hydrometeorological
R eports” and “ Technical Memorandum” prepared by the
National Weather Service (NWS) Office of Hydrology in
S i l ver Spring, MD. Other sources of meteorological data
include the National Hurricane Center in Coral Gables, FL,
and state climatologists as well as U.S. Geological Survey
and Corps flood reports.
b. Pr ecipitation. Monthly summaries of observed
hourly and daily precipitation data are published by NOAA
i n “ Climatical Data” and “ Hourly Precipitation Data.”
P recipitation data are also available from NOAA in
computer-readable media. Precipitation data for significant
hi storic storms (1870's-1960's) are tabulated in “ Storm
Rainfall in the United States, Depth, Area, Duration Data.”
(1) There are usually local, or state agencies, collecting precipitation data for their own use. These data could
provide additional storm information. However, precipitation measurements at remote unattended locations may not
be consistently and accurately recorded, particularly where
s now and hail frequently occur. For this reason, records
obt ained at unattended locations must be interpreted with
care. When an observer is regularly on-site, the times of
occurren ce of snowfall and hail should be noted to make
accurate use of the data. The exact location and elevation
of t he gauge are important considerations in precipitation
measurement and evaluation. For uniform use, this is best
expressed in terms of latitude and longitude and in meters
or feet of elevation above sea level. Of primary importance
in processing the data is tabulating precipitation at regular
i nt ervals. This should be done daily for non-recording
gauges with the time of observation stated. Continuously
recording gauges should be tabulated hourly. The original
record ing charts should be preserved in order to permit
s t udy of high-intensity precipitation during short intervals
for certain applications.
(2) Procedures to develop standard project and probabl e maximum precipitation estimates are presented in
NW S hydrometeorological reports and technical memorandum. Chapter 7 of this manual provides an overview of
hypothetical storms and their application to flood-runoff
c. Sno wpack. Where snowmelt contributes significant ly to runoff, observations of the snowpack characterist i cs can be of considerable value in the development of
hyd rologic design criteria. The observation of water
EM 1110-2-1420
31 Oct 97
equivalent (weight) of a vertical column sampled from the
snowpack at specified locations and observation times is of
primary importance. As the observations will ordinarily be
used as an index for surrounding regions, the elevation and
exposure of the location must be known. The depth of
s nowpack is of secondary importance, but some observations of the areal extent of the snowpack are often useful.
(1) An important element in processing snowpack
observation data is the adjustment of observations at all
locations to a common date, such as the first of each month
duri n g the snowpack accumulation season. Since these
observations are often made by traveling survey teams, they
are not made simultaneously. Also, they cannot always be
made at a specified time because it is impossible to obtain
(2) Where continuous recordings of snowpack water
equivalent by means of radioactive gauges or snow pillows
are available, these can be used on a basis for adjusting
manual observations at nearby locations. Otherwise, some
j udgment or correlation technique based on precipitation
measurements is required to adjust the observations data to
a uniform date at all locations. It is important to preserve
t he original records whenever such adjustments are made.
However, data that are disseminated for use in design
should be the adjusted systematic quantities.
(3) T he primary agency for the collection and distribut ion of snowpack data is the Soil Conservation Service
(S C S ) (Department of Agriculture, Washington, D.C.).
From January through May each year, the SCS publishes a
m onthly report titled, “ Water Supply Outlook.” These
reports provide snowpack and streamflow forecast data for
each state and region. The SCS also issues “ Basin Outlook
Reports,” a monthly regional summary of snow depth and
water content. Additionally, NOAA distributes an annual
t abu lation of snowpack data in their “ Snow Cover
S urveys.”
Climatoligcal precipitation data published
monthly by NOAA also include information on snowfall
and snow on the ground.
d. Temperatures. In most hydrologic applications of
temperature data, maximum and minimum temperatures for
each day at ground level are very useful. Continuous
records of diurnal temperature variations at selected locations can be used to determine the daily temperature pattern
fai rly accurately at nearby locations where only the
m axi m um and minimum temperatures are known. In
applying temperature data to large areas, it must be recogni zed that temperatures normally decrease with increasing
elevation and latitude. It is also important to preserve all of
the original temperature records. Summaries of daily
maximum and minimum temperatures should be maintained
and, where feasible, published. The NOAA report series
on “ Climatography of the United States” by city, state, or
region also provides information on daily and monthly
normal temperatures.
e. Moisture. Atmospheric moisture is a major factor
i nfluencing the occurrence of precipitation. This moisture
can be measured by atmospheric soundings which record
t em perature, pressure, relative humidity, and other items.
T ot al moisture in the atmosphere can be integrated and
expressed as a depth of water. During storms, the vertical
distribution of moisture in the atmosphere ordinarily
follows a rather definite pattern. Total moisture can
therefore be related to the moisture at the surface, which is
a function of the dew point at the surface. Accordingly, a
record of daily dew points is of considerable value. Here,
again, the elevation, latitude and longitude of the measuring
s t at i on must be known. NOAA publication “ Local
Climatological Data” is a primary source for observed dew
point, pressure, and temperature data.
f. Winds. Probably the most difficult meteorological
el ement to evaluate is wind speed and direction. Quite
commonly, the direction of surface winds reverses
di urnally, and wind speeds fluctuate greatly from hour to
hour and minute to minute. There is also a radical change
of wind speed and direction with altitude. The speed and
direction at lower levels is greatly influenced by
obs t ru ctions such as mountains, and locally by small
obstructions such as buildings and trees. Accordingly, it is
important that great care is exercised in selecting a location
and altitude for wind measurement. For most hydrologic
applications, wind measurements at elevations of 5 to 15 m
above the ground surface are satisfactory. It is important to
pres erve all basic records of winds, including data on the
l ocat i on, ground elevation, and the height of the
anemometer above the ground. An anemometer is an
instrument for measuring and indicating the force or speed
of t he wind. Where continuous records are available,
hourly tabulations of speed and direction are highly
desirable. Total wind movement and the prevailing direction for each day are also useful data. Daily wind data for
each state are published in NOAA publications “ Local
Climatological Data” and “ Climatological Data.”
g. Evaporation. Evaporation data is usually required
for reservoir studies, particularly for low-flow analysis.
R eservoir evaporation is typically estimated by measuring
pan evaporation or computing potential evaporation. There
are several methods of estimating potential evaporation,
bas ed on meteorological information.
Pruitt (1990)
reviewed various approaches in an evaluation of the
m et ho dology and results published in “ A Preliminary
EM 1110-2-1420
31 Oct 97
As sessment of Corps of Engineer Reservoirs, Their Purposes and Susceptibility to Drought,” (HEC 1990e).
(1) Evaporation is usually measured by using a pan
about 4 ft (1.2 m) in diameter filled with water to a depth of
about 8 in. (0.2 m). Daily evaporation can be calculated by
subtracting the previous day's reading from today's reading
and adding the precipitation for the intervening period. The
pan s ho uld be occasionally refilled and this fact noted in
t he record. This volume of added water, divided by the
area of the pan, is equal to the daily evaporation amount
exp ressed in inches or millimeters. A tabulation of daily
evaporation amounts should be maintained and, if possible,
published. It is essential that a rain gauge be maintained at
the evaporation pan site, and it is usually desirable that
temperature, dew point (or wet-bulb temperature) and lowlevel wind measurements also be made at the site for future
study purposes.
(2) NOAA Technical Report “ NWS 33, Evaporation
At l as for the Contiguous United States” (Farnsworth,
Thompson, and Peck 1982) provides maps showing annual
and May-October evaporation in addition to pan
coefficients for the contiguous United States. Companion
report “ NWS 34, Mean Monthly, Seasonal, and Annual Pan
Evaporation for the United States” documents monthly
evaporation data which was used in the development of the
evaporation atlas. Daily observed evaporation data are
published for each state in NOAA publications “ Local
Climatological Data” and in “ Climatological Data.”
h. Upper air soundings. Upper air soundings are
avai lable from NOAA National Climatic Data Center in
Asheville, NC. The soundings provide atmospheric pressure, temperature, dew point temperature, wind speed, and
direction data from which lapse rate, atmospheric stability,
and jet stream strength can be determined. These meteorol ogical parameters are necessary to a comprehensive
storm study.
5-2. Topographic Data
a. Mappi ng. For most hydrologic studies, it is
es s ent i al that good topographic maps be used. It is
important that the maps contain contours of ground-surface
el ev ation, so that drainage basins can be delineated and
i m port ant features such as slopes, exposure, and stream
patterns can be measured. United States Operational
Navigation Charts, with a scale of 1:1,000,000 and contour
i nt ervals of 1,000 ft, are available for most parts of the
However, mapping to a much larger scale
(1: 2 5,000) and smaller contour intervals in the range of
5-20 ft (1.5 to 6 m) are usually necessary for satisfactory
hyd rologic studies. The USGS 7.5 Minute Series (Scale
1:24,000), with a typical contour interval of 5 or 10 ft (1.5
or 3.0 m), provides a good basic map for watershed studies.
The USGS publications “ Catalog of Published Maps” and
“ Index to Maps” are excellent guides to readily available
topographic data for each state. Reports by the USGS are
avai l able through Books and Open File Reports Section,
US GS, Federal Center, Box 25425, Denver, CO, or by
contacting the National Technical Information Service
(NTIS), 5285 Port Royal Road, Springfield, VA 22161.
b. Di gital mapping. Increasingly, topographic data
are avai lable in digital form. One form of computer
des cription of topography is a digital elevation model
(DEM). Geographic information systems (GIS) can link
land attributes to topographic data and other information
concerning processes and properties related to geographic
locations. DEM and GIS representations of topologic data
are part of a general group of digital terrain models (DTM).
S om e of the earliest applications in hydrologic modeling
used grid cell (raster) storage of information. An example
of raster-based GIS is the Corps' Geographic Resource
Analysis Support System (GRASS). An alternate approach
ut i lizes a collection of irregularly spaced points connected
by lines to produce triangles, known as triangular irregular
net work (TIN). The use of DEM and TIN data and
processing software is rapidly changing and may soon
become the standard operation for developing terrain and
related hydrologic models. A review of GIS applications in
hydrology is provided by DeVantier et al. (HEC 1993).
c. Stream patterns and profiles. Where detailed
studies of floodplains are required, computation of waters urface profiles is necessary. Basic data needed for this
computation include detailed cross sections of the river and
overban k areas at frequent intervals. These are usually
obtained by special field surveys and/or aerial photography.
When these surveys are made, it is important to document
and date the data and resulting models, then permanently
preserve the information so it is readily available for future
reference. Observations of actual water-surface elevations
duri n g maximum flood stages (high-water marks) are
i nvaluable for calibrating and validating models for profile
d. L akes and swamps. The rate of runoff from any
watershed is greatly influenced by the existence of lakes,
swamps, and similar storage areas. It is therefore important
t o i ndicate these areas on available maps. Data on the
outlet characteristics of lakes are important because, in the
absence of outflow measurements, the outflow can often be
com pu ted using the relationship between the amount of
water stored in the lake and its outlet characteristics.
EM 1110-2-1420
31 Oct 97
e. Soi l and geology. Certain maps of soils and
geo logy can be very useful in surface-water studies if they
show characteristics that relate to perviousness of the basin.
These can be used for estimating loss rates during storms.
Of particular interest are areas of extensive sandy soils that
do not contribute to runoff and areas of limestone and
volcanic formations that are highly pervious and can store
large amounts of water beneath the surface in a short time.
Addi tionally, watershed sediment yield estimates will
depend on similar information. The SCS soil survey
reports are the primary source of soil and permeability data.
State geologic survey or mineral resource agencies are also
a useful source of geologic data.
f. Veg etal cover. Often the type of vegetation more
accurately reflects variation in hydrologic phenomena than
does the type of soil or the geology. In transposing information to areas of little or no hydrologic data, generalized
m aps of vegetal cover are very helpful. As with soil and
geol og y, vegetation has a significant impact on sediment
yield. In the arid southwest, time since the watershed last
burned is a significant parameter in estimating total sediment yield for a storm event. The U.S. Forest Service, the
Agricultural Stabilization and Conservation Service
(AS C S ) and, in western states, the Bureau of Land Managem ent are sources of vegetal cover maps. State forest,
agricultural agencies, or USGS topographic maps also
provide information on vegetal cover.
g. Existing improvements. Streamflow at any particular location can be greatly affected by hydraulic structures
l ocated upstream. It is important, therefore, that essential
dat a be obtained on all significant hydraulic structures
l ocated in and upstream from a study area. For diversion
structures, detailed data are required on the size of the
di vers i on dam, capacity of the diversion canal, and the
probabl e size of flood required to wash out the diversion
dam. In the case of storage reservoirs, detailed data on the
relation of storage capacity to elevation, location, and size
of outlets and spillways, types, sizes, and operation of
cont rol gates, and sizes of power plant and penstocks
s houl d be known. Bridges can produce backwater effects
which will cause upstream flooding. This flooding may be
produced by the approach roads, constriction of the channel
and floodplain, pier shapes, the angle between the piers and
the streamflow, or the pier length-width ratios.
5-3. Streamflow Data
The availability of streamflow data is a significant factor in
t he selection of an appropriate technical method for
reservoir studies. It is important to be cognizant of the
nat ure, source, reliability, and adequacy of available data.
If estimates are needed, the assumptions used should be
documented, and the effect of errors in the estimates on the
technical procedure and results should be considered.
Measurement. Streamflow data are usually best
obtained by means of a continuous record of river stage,
s upplemented by frequent meter measurements of flows
t hat can be related to corresponding river stages. It is
i m portant that stage measurements be made at a good
cont rol section, even if a weir or other control structures
m ust be constructed. Each meter measurement should
cons i s t of velocity measurements within each of several
(6-20, where practical) subdivisions within the channel
cross section. Velocity for a subdivision is usually taken as
t he velocity at a depth of 60 percent (0.6) of the distance
from the surface to the streambed or as the average of
vel o cities taken at 20 and 80 percent (0.2 and 0.8) of the
depth at the middle of the subdivision. River stage readings
should be made immediately before and after the cross
section is metered. The average of these two stages is the
stage associated with the measurement. The measurement
i s com puted by integrating the rates of flow (m3/s) in all
subdivisions of the cross section.
(1) Measurements of stream velocity and computed
streamflow are usually recorded on standard forms. When
m easurements have been made for a sufficient range of
fl ows , the rating curve of flow versus stage can be
devel oped . The rating curve can be used to convert the
continuous record of stage into a continuous record of flow.
T he fl ows should be averaged for each day in order to
construct a tabulation of mean daily flows. This constitutes
the most commonly published record of runoff.
(2) For flood studies, it is particularly important to
obtain accurate records of short-period variations during
hi gh river stages and to obtain meter measurements at or
near the maximum stage during as many floods as possible.
W here t he river profile is very flat, as in estuaries and
m aj or rivers, it is advisable to obtain measurements
frequently on the rising and on the falling stage to determ i ne if a looped, or hysteresis, effect exists in the rating.
T he reason for this is that the hydraulic slope can change
greatly, resulting in different rating curves for rising and
falling stages.
b. Streamflow data sources. The USGS is the prim ary agency for documenting and publishing flow data in
t he United States. Daily flow data for each state are
published in the USGS annual “ Water Data Report.” The
USGS National Water Data Exchange (NAWDEX) comput erized database identifies sources of water data. The
National Water Data Storage and Retrieval System
(W ATSTORE) provides processing, storage, and retrieval
of surface water, groundwater, and water quality data.
EM 1110-2-1420
31 Oct 97
NAWDEX is only an index of the contents of
WATSTORE. These programs will eventually be integrated into a National Water Information System, which
wi l l also combine the National Water-Use Information
Program and Water Resources Scientific Information
C ent er (Mosley and McKerchar 1993). Commercial data
s ervi ces have also provided convenient access to USGS
daily and peak flow files on CD.
c. Flow conditions. Reservoirs substantially alter the
distribution of flow in time. Many other developments,
such as urbanization, diversions, or cultivation and
irrigation of large areas can also have a significant effect on
watershed yield and the distribution of flow in time. The
degree that flows are modified depends on the scale and
manner of the development, as well as the magnitude, time,
and areal distribution of rainfall (and snowmelt, if
pertinent). Most reservoir evaluations require an assessment based on a consistent flow data set. Various terms are
used to define what condition the data represent:
(1) Natural conditions in the drainage basin are defined
as the hydrologic conditions that would prevail if no regulatory works or other development affecting basin runoff and
s t ream flow were constructed. The effects of natural lakes
and swamp areas are included.
(2) Present conditions are defined as the conditions that
exi s t at, or near, the time of study. If there are upstream
res erv oirs in the basin, the observed flow record would
represent “ regulated flow.”
Flow records, preceding
curren t reservoir projects, would be adjusted to reflect
those project operations in order to have consistent “ present
conditions” flow.
(3) Unregulated conditions reflect the present (or
recen t) basin development, but without the effect of reservoir regulation. Unlike natural conditions which are
di fficult to determine, only the effect of reservoir operation
and major diversions are removed from the historic data.
(4) Without-project conditions are defined as the
condi tions that would prevail if the project under consideration were not constructed but with all existing and future
projects under construction assumed to exist.
(5) W ith-project conditions are defined as the conditions that will exist after the project is completed and after
completion of all projects having an equal or higher priority
of construction.
5-4. Adj ustment of Streamflow Data
The adjustment of recorded streamflows is often required
before the data can be used in water resources development
s t udi es . This is because flow information usually is
required at locations other than gauging stations and for
conditions of upstream development other than those under
which flows occurred historically. In correlating flows
bet ween locations, it is important to use “ natural” flows
(unaffected by artificial storage and diversion) in order that
correlation procedures will apply logically and efficiently.
In generating flows, natural flows should be used because
general frequency functions, characteristic of natural flows,
are employed in this process.
N atural conditions. When feasible, flow data
should be converted to natural conditions. The conversion
is made by adding historical storage changes (plus net
evaporation) and upstream diversions (less return flows) to
historical flows at the gauging stations for each time
interval in turn. Under some conditions, it may be necessary to account for differences in channel and overbank
infiltration losses, distributary flow diversions, travel times,
and other factors.
b. Unregulated conditions. It is not always feasible
to convert flows to natural conditions. Often, required data
are not available. Also, the hydrologic effects and timing
of some basin developments are not known to sufficiently
define the required adjustments. An alternative is to adjust
the data to a uniform basin condition, usually near current
time. The primary adjustments should remove special
i nfl uences, such as major reservoirs and diversions, that
would cause unnatural variations of flow.
c. Res er voir holdouts. The primary effect of reservoi r operation is the storage of excess river flow during
high-flow periods, and the release of stored water during
l ow-flow periods. The flow adjustment process requires
the addition of the change in water stored (hold-outs) in
each time step to the observed regulated flow. Holdouts,
both positive and negative, are routed down the channel to
each gauge and algebraically added to the observed flow.
Hydro logic routing methods, typically used for these
adj us tments, are described in Chapter 9 of EM 1110-21417. The HEC Data Storage System (HEC-DSS) software
(HEC 1995) provides a convenient data management
system and utilities to route flows and add, subtract, or
adjust long time-series flow data.
EM 1110-2-1420
31 Oct 97
d. Reservoir losses. The nonproject inflow represents
t he fl ow at the project site without the reservoir and
i ncl ud es runoff from the entire effective drainage area
above the dam, including the reservoir area. Under nonproject conditions, runoff from the area to be inundated by
t he res ervoir is ordinarily only a fraction of the total
precipitation which falls on that area. However, under
project conditions infiltration losses over the reservoir area
are us ually minimal during a rainfall event. Thus,
practically all the precipitation falling on the reservoir area
will appear as runoff. Therefore, the inflow will be greater
under project conditions than under nonproject conditions,
i f inflow is defined as total contribution to the reservoir
before evaporation losses are considered. In order to
determine the amount of water available for use at the
reservoir, evaporation must be subtracted from project
inflow. In operation studies, nonproject inflow is ordinarily
converted to available water in one operation without computing project inflow as defined above. This is done in one
of two ways: by means of a constant annual loss each year
with seasonal variation or with a different loss each period,
expressed as a function of observed precipitation and
evaporation. These two methods are described in the
following paragraphs.
(1) Constant annual loss procedure consists of estimating the evapotranspiration and infiltration losses over
the reservoir area for conditions without the project, and the
evaporation and infiltration losses over the reservoir area
with the project. Nonproject losses are usually estimated as
the difference between average annual precipitation and
average annual runoff at the location, distributed seasonally
in accordance with precipitation and temperature variations.
These are expressed in millimeters of depth. Under project
conditions, infiltration losses are usually ignored, and
l os ses are considered to consist of only direct evaporation
from the lake area, expressed in millimeters for each period.
T he difference between these losses is the net loss due to
t he project. Figure 5-1 illustrates the differences between
nonproject and project losses.
(2) The variable loss approach uses historical records
of long-term average monthly precipitation and evaporation
data to account for the change in losses due to a reservoir
proj ect. This is accomplished by estimating the average
runo ff coefficient, the ratio of runoff to rainfall, for the
reservoir area under preproject conditions and subtracting
t hi s from the runoff coefficient for the reservoir area under
proj ect conditions. The runoff coefficient for project
condi t i ons is usually 1.0, but a lower coefficient may be
used if substantial infiltration or leakage from the reservoir
i s anticipated. The difference between preproject and
proj ect runoff coefficients is the net gain expressed as a
ratio of precipitation falling on the reservoir. This is often
estimated to be 0.7 for semi-arid regions. This increase in
runoff is subtracted from gross reservoir evaporation, often
estimated as 0.7 of pan evaporation, to obtain a net loss.
e. Ot her losses. In final project studies it is often
necessary to consider other types of project losses which
may be of minor importance in preliminary studies. Often,
these losses cannot be estimated until a project design has
been adopted. The importance of these losses is dependent
upon t heir relative magnitude. That is, losses of 5 m3/s
might be considered unimportant for a stream which has a
minimum average annual flow of 1,500 m3/s. Such losses,
though, would be significant from a stream with a minimum
average annual flow of 25 m3/s. Various types of losses are
discussed in the following paragraphs.
(1) T h e term “ losses” may not actually denote a
physical loss of water from the system as a whole. Usually,
wat er unavailable for a specific project purpose is called a
“ l os s ” for that purpose although it may be used at some
other point or for some other purpose. For example, water
which leaves the reservoir through a pipeline for municipal
water supply or fish hatchery requirements might be called
a loss to power. Likewise, leakages through turbines, dams,
conduits, and spillway gates are considered losses to
hydropower generation, but they are ordinarily not losses to
fl ow requirements at a downstream station. Furthermore,
s uch l osses that become available for use below the dam
s hould be added to inflow at points downstream from the
(2) Leakage at a dam or in a reservoir area is considered a loss for purposes which are dependent upon availability of water at the dam or in the reservoir itself. These
purposes include power generation, pipelines from the
res erv oir, and any purpose which utilizes pump intakes
which are located at or above the dam. As a rule, leakage
through, around, or under a dam is relatively small and is
difficult to quantify before a project is actually constructed.
In s om e cases, the measured leakage at a similar dam or
geologic area may be used as a basis for estimating losses at
a proposed project. The amount of leakage is a function of
the type and size of dam, the geologic conditions, and the
hydrostatic pressure against the dam.
(3) Leakage from conduits and spillway gates is a
function of gate perimeter, type of seal, and head on the
gat e, and it varies with the square root of the head. The
am ou nt of leakage may again be measured at existing
proj ect s with various types of seals, and a leakage rate
com put ed per meter of perimeter for a given head. This
rat e may then be used to compute estimated leakage for a
EM 1110-2-1420
31 Oct 97
Figure 5-1. Project and nonproject reservoir losses
proposed project by using the proposed size and number of
gates and the proposed head on the gates.
(4) If a proposed project will include power, and if the
area demand is such that the turbines will sometimes be
idle, it is advisable to estimate leakage through the
turbines when closed. This leakage is a function of the
type of penstock gate, type of turbine wicket gate, number
of turbines, and head on the turbine. The actual leakage
through a turbine may be measured at the time of acceptance and during annual maintenance inspections, or the
measurements of similar existing projects may be used to
estimate leakage for a proposed project. An estimate of
the percent of time that a unit will be closed may be
obtained from actual operational records for existing units
in the same demand area. The measured or estimated leakage
rate is then reduced by multiplying by the proportion of
time the unit will be closed. For example, suppose that
leakage through a turbine has been measured at 0.1 m3/s,
and the operation records indicate that the unit is closed
60 percent of the time. The average leakage rate would be
estimated at 0.1 × 0.6 or 0.06 m3/s.
(5) The inclusion of a navigation lock at a dam
requires that locking operations and leakages through the
lock be considered. The leakage is dependent upon the lift
or head, the type and size of lock, and the type of gates and
seals. Again, estimates can be made from observed leakage
at similar structures. Water required for locking operations
should also be deducted from water available at the dam
site. These demands can be computed by multiplying the
volume of water required for a single locking operation
EM 1110-2-1420
31 Oct 97
times the number of operations anticipated in a given time
peri od and converting the product to a flow rate over the
given period.
Correlation procedures and suggested
bas in characteristics are described in Chapter 9 of
EM 1110-2-1415.
(6) Water use for purposes related to project operations
is often treated as a loss. Station use for sanitary and drinking purposes, cooling water for generators, and water for
condensing operations have been estimated to be about 0.06
m3/s per turbine at some stations in the southwestern United
S t ates.
Examining operation records for comparable
proj ects in a given study area may also be useful in
es t i m at ing these losses. If house units are included in a
proj ect to supply the project's power requirements, data
s hould be obtained from the designer in order to estimate
water used by the units.
g. Preproject conditions. After project flows for a
specified condition of upstream development are obtained
for all pertinent locations and periods, they must be converted to preproject (nonproject) conditions. Nonproject
conditions are those that would prevail during the lifetime
of the proposed project if the project was not constructed.
This conversion is made by subtracting projected upstream
diversions and storage changes and by accounting for
evaporation, return flows, differences in channel
i nfiltration, and timing. Where nonproject conditions will
vary during the project lifetime, it is necessary to convert to
t wo or more sets of conditions, such as those at the start
and end of the proposed project life. Separate operation
studies would then be made for each condition. This
conversion to future conditions can be made simultaneously
with project operation studies, but a separate evaluation of
nonproject flows is usually required for economic
evaluation of the project.
(7) The competitive use of water should also be cons i dered when evaluating reservoir losses. When initially
estimating yield rates for various project purposes at a
m ul t iple-purpose project, competitive uses of water are
oft en treated as losses. For example, consider a propos ed project on a stream with an average minimum
usable flow of 16 m3/s. The reservoir of this project is
t o s up ply 1.5 m3/s by pipeline for downstream water
s uppl y and 2.0 m3/s for a fish hatchery in addition to
provi ding for hydroelectric power production. The minim um average flow available for power generation is thus,
16 - (1.5 + 2.0) = 12.5 m3/s. Care should be exercised in
accounting for all such competitive uses when making
preliminary yield estimates.
Missing data.
(1) After recorded flows are converted to uniform
conditions, flows for missing periods of record at each
pertinent location should be estimated by correlation with
record ed flows at other locations in the region. Usually,
onl y one other location is used, and linear correlation of
flow logarithms is used. It is more satisfactory, however, to
us e all other locations in the region that can contribute
independent information on the missing data. Although
this would require a large amount of computation, the
computer program HEC-4 Monthly Streamflow Simulation
accomplishes this for monthly streamflow (HEC 1971).
(2) Flow estimates for ungauged locations can be
estimated satisfactorily on a flow per basin area basis in
some cases, particularly where a gauge exists on the same
stream. In most cases, however, it is necessary to correlate
mean flow logarithms (and sometimes standard deviation of
fl ow logarithms) with logarithms of drainage area size,
logarithm of normal seasonal precipitation, and other basin
5-5. Simulation of Streamflow Data
a. Introduction. The term “ simulation” has been
us ed t o refer to both the estimation of historic sequences
and t he assessment of probable future sequences of
s t reamflow. The former reference concerns the application
of continuous precipitation-runoff models to simulate
s t reamflow based on meteorologic input such as rainfall
and temperature. The latter reference concerns the applicat i o n of stochastic (probabilistic) models that employ
Monte Carlo simulation methods to estimate the probable
occu rrence of future streamflow sequences. Assessment of
t he probable reliability of water resource systems can be
made given the assessment of probable future sequences of
flow. Statistical methods used in stochastic models can
al s o be employed to augment observed historic data by
filling in or extending observed streamflow records.
b. Historic sequences from continuous precipitationrunoff models. Many different types of continuous simulation runoff models have been used to estimate the historic
sequence of streamflow that would occur from observed
precipitation and other meteorologic variables. Among the
m os t prominent are the various forms of the Stanford
Watershed Model (Mays and Tung 1992) and the SSARR
Model used by the North Pacific Division (USACE 1991).
F or a further description of the application of the models
see EM 1110-2-1417 Section 8.
EM 1110-2-1420
31 Oct 97
c. Stochastic streamflow models. Stochastic streamflow models are used to assess the probable sequence of
fut u re flows. As with any model, a model structure is
assumed, parameters are estimated from observed data, and
t he m odel is used for prediction (Salas et al. 1980).
Typically, stochastic streamflow models are used to simulate annual and/or monthly streamflow volumes. Stochastic
s t reamflow models have not been successfully developed
for daily streamflow.
(1) Although many different structures have been
proposed in the research literature, regression is most
com m on ly used as the basis for stochastic streamflow
models. The regressions involve both correlation between
fl ows at different sites and the correlation between current
and past flow periods, termed serial correlation. The
correlation between sites is useful in improving parameters
estimates from regional information. The serial correlation
between periods is important in modeling the persistence,
or the tendency for high flow or drought periods. A
random error component is added to the regression to
provide a probabilistic component to the model.
(2) The model parameters are estimated to preserve the
correlation structure observed in the observed data. If the
app ropriate correlation structure is preserved, then the
regression residuals should closely approximate the
behavi or that would be expected from a random error
(3) Model prediction is performed via the application
of Monte Carlo simulation. Monte Carlo simulation is a
numerical integration technique. This numerical technique
i s necessary because the stochastic model effectively represents a complex joint probability distribution of
s t ream flows in time and space that cannot be evaluated
analytically. The simulation is performed by producing
rando m sequences of flows via a computer algorithm that
em pl oy s random number generators. These sequences of
fl ows are analyzed to assess supply characteristics, for
example the probability for a certain magnitude or duration
of drought. The number of flow sequences generated is
sufficient when the estimated probabilities do not change
s i gn ificantly with the number of simulations. For further
expl an ation of this point, see “ Stochastic Analysis of
Drought Phenomena” (HEC 1985b).
d. As sessment of reliability with stochastic streamflow models. The advantage of using a stochastic streamflow model over that of employing only historic records is
that it can be used to provide a probabilistic estimate of a
water resource system's reliability. For example, the
probability that a particular reservoir will be able to meet
certain goals can be estimated by simulating the stochastic
fl ow s equences with a reservoir simulation model. Once
again, the number of flow sequences used are sufficient
when the estimate of the probabilities stabilize.
e. Available software for stochastic streamflow
s i mu lation. HEC-4, “ Monthly Streamflow Simulation”
(HEC 1971), and LAST (Lane 1990) are public domain
s oftware for performing stochastic streamflow simulation.
HEC-4 performs monthly stochastic streamflow simulation.
LAST utilizes a more modern approach where annual and
s horter time period (seasonal, monthly, etc.) stochastic
streamflow can be co-simulated.
f. Extending and filling in historic records. Statist i cal techniques can be used to augment existing historic
records by either “ filling in” missing flow values or
ext en ding the observed record at a gauge based on observations at other gauges. The statistical techniques used are
referred to as MOVE, maintenance of variance extension,
and are a modification of regression based techniques
(Alley and Burns 1983, and Salas 1992).
algorithms have been instituted because the variance of
s eri es augmented by regression alone is underestimated.
T he MOVE technology is only generally applicable when
s erial correlation does not exist in the streamflow records.
However, monthly or annual sequences of streamflow
vol um es usually do exhibit a degree of serial correlation.
In t hese circumstances, the information provided by the
l onger record station may not be useful in extending a
s hort er record station. For a discussion of the impact of
s erial correlation see Matalas and Langbein (1962) and
Tasker (1983).
EM 1110-2-1420
31 Oct 97
Chapter 6
Hydrologic Frequency Determinations
6-1. Introduction
F req uency curves are most commonly used in Corps of
Engineers studies to determine the economic value of flood
reduction projects. Reservoir applications also include the
determination of reservoir stage for real estate acquisition
and reservoir-use purposes, the selection of rainfall
m agn itude for synthetic floods, and the selection of runoff
magnitude for sizing flood-control storage.
a. Annual and partial-duration frequency. There are
t wo bas ic types of frequency curves used in hydrologic
work. A curve of annual maximum events is ordinarily
used when the primary interest lies in the very large events
or when the second largest event in any year is of minor
concern in the analysis.
The partial-duration curve
represents the frequency of all events above a given base
val ue, regardless of whether two or more occurred in the
s am e year. This type of curve is ordinarily used in econom ic analysis when there are substantial damages resulting from the second largest and third largest floods in
extremely wet years. Damage from floods occurring more
frequen tly than the annual event can occur in agricultural
areas, when there is sufficient time between events for
recovery and new investment. When both the frequency
curv e of annual floods and the partial-duration curve are
us ed, care must be exercised to assure that the two are
b. Seasonal frequency curves. In most locations,
there are seasons when storms or floods do not occur or are
not s evere, and other seasons when they are more severe.
Also, damage associated with a flood often varies with the
season of the year. In studies where the seasonal variation
is of primary importance, it becomes necessary to establish
frequency curves for each month or other subdivision of the
year. For example, one frequency curve might represent
t he largest floods that occur each January; a second one
would represent the largest floods that occur each February,
et c. In another case, one frequency curve might represent
fl oods during the snowmelt season, while a second might
repres ent floods during the rainy season. Occasionally,
when seasons are studied separately, an annual-event curve
covering all seasons is also prepared. Care should be
exercised to assure that the various seasonal curves are
consistent with the annual curve.
6-2. Duration Curv es
a. Fl ow duration curve. In power studies, for runof-river plants particularly and in low-flow studies, the
flow-duration curve serves a useful purpose. It simply
represents the percent of time during which specified flow
rat es are exceeded at a given location. Ordinarily, variat i ons within periods less than 1 day are inconsequential,
and the curves are therefore based on observed mean-daily
flows. For the purposes served by flow duration curves, the
ext reme rates of flow are not important, and consequently
t here i s no need for refining the curve in regions of high
b. Pr eparing flow-duration curve. The procedure
ordinarily used to prepare a flow duration curve consists of
counting the number of mean-daily flows that occur within
given ranges of magnitude. The lower limit of magnitude
in each range is then plotted against the percentage of days
of record that mean-daily flows exceed that magnitude. A
flow duration curve example is shown in Figure 6-1.
6-3. Flood-Frequency Determinations
At many locations, flood stages are a unique function of
flood discharges for most practical purposes. Accordingly,
i t i s u sual practice to establish a frequency curve of river
di s charges as the basic hydrologic determination for flood
damage reduction project studies. In special cases, factors
ot her than river discharge, such as tidal action or
accumulated run-off volume, may greatly influence river
s t ages. In such cases, a direct study of stage frequency
based on recorded stages is often warranted.
a. Determination made with available data. Where
runoff data at or near the site are available, flood-frequency
det erm i nations are most reliably made by direct study of
t hes e data. Before frequency studies of recorded flows are
made, the flows must be converted to a uniform condition,
usually to conditions without major regulation or diversion.
Developing unregulated flow requires detailed routing
s t ud ies to remove the effect of reservoir hold-outs and
diversions. As damaging flows occur during a very small
fraction of the total time, only a small percentage of daily
fl ows are used for flood-frequency studies. These consist
of the largest flow that occurs each year and the secondary
peak flows that cause damage.
However, for most
reservoirs studies, the period-of-record flow will be
required for analysis of nonflood purposes and impacts.
EM 1110-2-1420
31 Oct 97
Figure 6-1. Example flow duration curve
b. Historical data. Flood frequency estimates are
subject to considerable uncertainty, even when fairly long
records are available. In order to increase the reliability of
frequency estimates, empirical theoretical frequency
relations are used in specific frequency studies. These
studies require that a complete set of data be used. In
order to comply with this requirement, the basic frequency
study ordinarily is based on the maximum flow for each
year of record. Supplementary studies that include other
damaging events are ordinarily made separately. The
addition of historical information can be very important in
verifying the frequency of large recorded events.
Historical information on large damaging floods can be
obtained through standard sources such as USGS water
supply series or from newspapers and local museums. The
latter sources often are more qualitative but give important
insight into the relative frequency of recorded events.
EM 1110-2-1420
31 Oct 97
c. Selecting and computing frequency curves. The
underlying general assumption made in all frequency
s t udi es is that each observed event represents an approximately equal proportion of the future events that will occur
at the location, if controlling conditions do not change.
Detailed procedures for selecting data and computing
fl oo d-frequency curves are presented in EM 1110-2-1415
and HEC-FFA program (HEC 1992c).
d. Regional correlation of data. Where runoff data at
or near the site do not exist or are too fragmentary to
s upport direct frequency calculations, regional correlation
of frequency statistics may be used for estimating frequencies. These correlations generally relate the mean and
standard deviation of flows to drainage basin characteristics
and l ocation. Techniques of regional correlation are
presented in Chapter 9 of EM 1110-2-1415.
e. Ext reme floods. In the analysis of reservoir projects, the project's performance during floods larger than the
m axi m um recorded events is usually required. Extrapolating derived frequency relations is uncertain, so special
studies of the potential magnitudes of extreme flood events
are us u ally required. The most practical approach is
through examination of rainstorms that have occurred in the
region and determination of the runoff that would result at
t he project location if these storms should occur in the
t ri butary area. This subject is discussed in the following
chapter, “ Flood-Runoff Analysis.”
6-4. Estimating Frequency Curv es
a. Approaches. There are two basic approaches to
estimating frequency curves--graphical and analytical.
Each approach has several variations, but the discussion
herei n will be limited to recommended methods. The
pri mary Corps reference for computing frequency curves is
EM 1110-2-1415.
(1) Graphical. Frequencies are evaluated graphically
by arranging observed values in the order of magnitude and
representing frequencies by a smooth curve through the
array of values. Each observed value represents a fraction
of the future possibilities and, when plotting the frequency
curv e, it is given a plotting position that is calculated to
give it the proper weight. Every derived frequency relation
should be plotted graphically, even though the results can
be obtained analytically. Paragraph 2-4 of EM 1110-21415 presents “ Graphical Frequency Analysis.”
(2) Analytical. In the application of analytical (stat i s t i cal) procedures, the concept of theoretical populations
or distributions is employed. The events that have occurred
are presumed to constitute a random sample and are used
accordingly to make inferences regarding their “ parent
population” (i.e., the distribution from which they were
deri ved ). The procedure is applied to annual maxima of
unregu lated flow, which are assumed to be independent
random events. The fact that the set of observations could
res ul t from any of many sets of physical conditions or
distributions leaves considerable uncertainty in the derived
curve. However, using statistical processes, the most
probabl e nature of the distribution from which the data
were derived can be estimated. Because this in all
probabi lity is not the true parent population, the relative
chance that variations from this distribution might be true
m us t be evaluated.
Each range of possible parent
populations is then weighted in proportion to its likelihood
i n order to obtain a weighted average. A probability
obtained from this weighted average is herein referred to as
the expected probability PN. Chapter 3 of EM 1110-2-1415
covers analytical flood-frequency analysis.
(3) Regional. Because of the shortness of hydrologic
records, frequency determinations for rare events are
relatively unreliable when based on a single record. Also, it
i s oft en necessary to estimate frequencies for locations
where no record exists. For these reasons, regionalized
frequency studies, in which frequency characteristics are
related to drainage-basin features and precipitation charact eri s tics, are desirable. Regionalized frequency studies
usually develop relationships for analytical frequency
statistics. An alternative approach is to develop predictive
equations for the flow for specific recurrence intervals.
Chapter 9 of EM 1110-2-1415 presents regression analysis
and its application to regional studies.
b. Flood volume-duration frequencies. A comprehen sive flood volume-duration frequency series consists of
a set of: 1, 3, 7, 15, 30, 60, 120, and 183-day average flows
for each water year. These durations are normally available
from the USGS WATSTORE files, and they are the default
durations in the computer program STATS (HEC 1987a).
R unoff volumes are expressed as average flows in order
t hat peak flows and volumes can be readily compared and
coordinated. Paragraph 3-8 of EM 1110-2-1415 covers
flood volumes.
c. Low flow frequencies. Reservoir analysis often
requ ires the evaluation of the frequencies of low flows
forvarious durations. The same fundamental procedures
can be used, except that minimum instead of maximum
runo ff values are selected from the basic data. For low
fl ows, the effects of basin development are usually more
s i gni fi cant than for high flow. A relatively moderate
di vers ion may not be very significant during a flood;
however, it may greatly modify or even eliminate low
fl ows. Accordingly, one of the most important aspects of
EM 1110-2-1420
31 Oct 97
low flows concerns the evaluation of past and future effects
of bas in developments. Chapter 4 of EM 1110-2-1415
describes low-flow frequency analysis.
d. Reservoir level frequency. A reservoir frequency
curve of annual maximum storage is ordinarily constructed
grap hically, using the procedures for flood-flow frequency.
Observed storage should be used to the extent available, but
onl y if the reservoir has been operated in the past in
accordan ce with future plans. If historical data are not
avai l able, or if it is not appropriate for future use, then
res ervo ir routings should be used to develop data for
exp ected reservoir operations. Stage-duration curves can
be constructed from historical operation data or from
s i m ulations. These curves are usually constructed for the
entire period-of-record, or for a selected wet or dry period.
F or some purposes, particularly recreational use, the
seasonal variation of reservoir stages is of importance, and
a set of frequency or duration curves for each month of the
year may be required. Reservoir stage (or elevation) curves
s hould indicate significant reservoir levels such as:
m i ni m u m pool, top of conservation pool, top of floodcontrol pool, spillway crest elevation, and top of dam.
6-5. Effect of Basin Dev elopments on Frequency
a. Ef f ects of flood-control works. Most hydrologic
frequency estimates serve some purpose relating to the
pl an ning, design, or operation of water resources managem en t projects. The anticipated effects of a project on
fl ooding can be assessed by comparing the peak discharge
and volume frequency curves with and without the project.
Also, projects that have existed in the past have affected the
rat es and volumes of floods, and recorded values must be
adj us t ed to reflect uniform conditions in order for the
frequency analysis to conform to the basic assumptions of
randomness and common population. For a frequency
curve to conform reasonably with a generalized
mathematical or probability law, the flows must be
es s en tially unregulated by man-made storage or diversion
s t ru ctures. Consequently, wherever practicable, recorded
runoff values should be adjusted to unregulated conditions
before a frequency analysis is made. However, in cases
where the regulation results from a multitude of relatively
small hydraulic structures that have not changed appreci abl y during the period of record, it is likely that the
general mathematical laws will apply as in the case of
nat ural flows, and that adjustment to natural conditions
wou ld be unnecessary. The effects of flood control works
are presented in paragraph 3-9 of EM 1110-2-1415, and
effects of urbanization in paragraph 3-10.
b. Reg ulated runoff frequency curves. If it is practical, the most complete approach to determining frequency
curv es of regulated runoff consists of routing flows for the
ent i re period of record through the proposed management
work s, arranging the annual peak regulated flows in order
of m agnitude, assigning a plotting position to the peak
values, plotting the peak flow values at the assigned
plotting position, and drawing the frequency curve based on
the plotted data. A less involved method consists of routing
t he l argest floods of record, or multiples of a large
hypothetical flood, to estimate the regulated frequency
curve. This approach requires the assumption that the
frequ ency of the regulated peak flow is the same as the
unregu lated peak flow, which is probably true for the
l argest floods.
Paragraph 3-9d of EM 1110-2-1415
describes these methods.
c. Er ratic stage-discharge frequency curves. In
general, cumulative frequency curves of river stages are
det ermined from frequency curves of flow. In cases where
t he stage-discharge relation is erratic, a frequency curve of
stages can be derived directly from stage data. Chapter 6 of
EM 1110-2-1415 presents stage-frequency analysis.
d. Res ervoir or channel modifications. Project
construction or natural changes in streambed elevation may
chang e the relationship between stage and flow at a
location. By forming constrictions, levees may raise river
stages half a meter for some distance upstream. Reservoir
or channel modifications may cause changes in degradation
or aggradation of streambeds, and thereby change rating
curves . Thus, the effect of projects on river stages often
involves the effects on channel hydraulics as well as the
effects on streamflow. Consult EM 1110-2-1416 for
information on modeling these potential changes.
6-6. Selection of Frequency Data
a. Primary considerations. The primary considerat i on in selecting an array of data for a frequency study is
t he objective of the frequency analysis. If the frequency
curve that is developed is to be used for estimating damages
t hat are related to instantaneous peak flows in a stream,
peak flows should be selected from the record. If the
dam ages are related to maximum mean-daily flows or to
maximum 3-day flows, these items should be selected. If
the behavior of a reservoir under investigation is related to
t he 3-day or 10-day rain-flood volume, or to the seasonal
s nowmelt volume, that pertinent item should be selected.
Norm ally, several durations are analyzed along with peak
flows to develop a consistent relationship.
EM 1110-2-1420
31 Oct 97
b. Selecting a related variable. Occasionally, it is
neces s ary to select a related variable in lieu of the one
des i red. For example, where mean daily flow records are
m ore complete than the records of peak flows, it may be
desirable to derive a frequency curve of mean-daily flows
and then, from the computed curve, derive a peak-flow
curve by means of an empirical relation between mean daily
fl ows and peak flows. All reasonably independent values
should be selected, but the annual maximum events should
ordinarily be segregated when the application of analytical
procedures is contemplated.
c. Data selected. Data selected for a frequency study
must measure the same aspect of each event (such as peak
fl ow, mean-daily flow, or flood volume for a specified
duration), and each event must be controlled by a uniform
s et of hydrologic and operational factors. For example, it
would be improper to combine items from old records that
are repo rted as peak flows but are in fact only daily
readi ngs, with newer records where the peak was actually
measured. Similarly, care should be exercised when there
has been significant change in upstream storage regulation
during the period of record so as not to inadvertently
combine unlike events into a single series. In such a case,
the entire flow record should be adjusted to a consistent
condition, preferably the unregulated flow condition.
d. Hydrologic factors. Hydrologic factors and relat i ons hips operating during a winter rain flood are usually
qui te different from those operating during a spring snowmelt flood or during a local summer cloudburst flood.
Where two or more types of floods are distinct and do not
occur predominantly in mutual combinations, they should
not be combined into a single series for frequency analysis.
T hey should be considered as events from different parent
popul ations. It is usually more reliable in such cases to
segregate the data in accordance with type and to combine
only the final curves, if necessary. For example, in the
mountainous region of eastern California, frequency studies
are made separately for rain floods, which occur principally
from November through March, and for snowmelt floods,
whi ch occur from April through July. Flows for each of
these two seasons are segregated strictly by cause, those
predominantly caused by snowmelt and those
predominantly caused by rain. In desert regions, summer
thunderstorms should be excluded from frequency studies
of winter rain flood or spring snowmelt floods and should
be considered separately. Similarly, in coastal regions it
would be desirable to separate floods induced by hurricanes
or typhoons from other general flood events.
e. Dat a adjustments. When practicable, all runoff
dat a s hould be adjusted to unregulated hydrologic condit i ons before making the frequency study because natural
flows are better adapted to analytical methods and are more
easily compared within a region. Frequency curves of
present-regulated conditions (those prevailing under current
practices of regulation and diversion) or of future-regulated
condi tions can be constructed from the frequency curve of
natural flow by means of an empirical or logical
relationship between natural and regulated flows. Where
data recorded at two different locations are to be combined
for construction of a single frequency curve, the data
should be adjusted as necessary to a single location, usually
the location of the longer record, accounting for differences
of drainage area and precipitation and, where appropriate,
channel characteristics between the locations. Where the
stream-gauge location is somewhat different from the
project location, the frequency curve should be constructed
for the stream-gauge location and subsequently adjusted to
the project location.
f. Runoff record interruptions. Occasionally, a
runoff record may be interrupted by a period of one or more
years. If the interruption is caused by the destruction of the
gauging station by a large flood, failure to fill in the record
for that flood would have a biasing effect, which should be
avoi ded. However, if the cause of the interruption is
known to be independent of flow magnitude, the entire
peri o d of interruption should be eliminated from the
frequency array, since no bias would result. In cases where
no runoff records are available on the stream concerned, it
is possible to estimate the frequency curve as a whole using
regional generalizations. An alternative method is to
estimate a complete series of individual floods from
recorded precipitation by continuous hydrologic simulation
and perform conventional frequency analysis on the
simulated record.
6-7. Climatic Variations
S om e hydrologic records suggest regular cyclic variations
i n precipitation and runoff potential. Many attempts have
been made to demonstrate that precipitation or stream flows
display variations that are in phase with various cycles, particularly the well-established 11-year sunspot cycle. There
i s no doubt that long duration cycles or irregular climatic
changes are associated with general changes of land masses
and s eas and with local changes in lakes and swamps.
Also, large areas that have been known to be fertile in the
past are now arid deserts, and large temperate regions have
been covered with glaciers one or more times. Although
t he exi stence of climatic changes is not questioned, their
effect is ordinarily neglected because long-term climatic
chan ges generally have insignificant effects during the
period concerned in water development projects, and shortt erm climatic changes tend to be self-compensating. For
t hes e reasons, and because of the difficulty in
EM 1110-2-1420
31 Oct 97
di fferentiating between fortuitous and systematic changes,
it is considered that, except for the annual cycle, the effect
of natural cycles or trends during the period of useful
proj ect life can ordinarily be neglected in hydrologic
frequency studies.
6-8. Frequency Reliability Analyses
a. Inf l uences. The reliability of frequency estimates
i s i nfl uenced by the amount of information available, the
vari ability of the events, and the accuracy with which the
data were measured.
(1) In general with regard to the amount of information
available, errors of estimate are inversely proportional to
t he s quare root of the number of independent items
cont ai n ed in the frequency array. Therefore, errors of
es t i m ates based on 40 years of record would normally be
half as large as errors of estimates based on 10 years of
record, other conditions being the same.
(2) The variability of events in a record is usually the
m os t i mportant factor affecting the reliability of frequency
es t i m ates. For example, the ratio of the largest to the
smallest annual flood of record on the Mississippi River at
Red River Landing, LA, is about 2.7; whereas the ratio of
the largest to the smallest annual flood of record on the
Ki ngs R iver at Piedra, CA, is about 100 or 35 times as
great. Statistical studies show that as a consequence of this
di fference in variability, a flow corresponding to a given
frequen cy that can be estimated within 10 percent on the
Mississippi River, can be estimated only within 40 percent
on the Kings River.
(3) The accuracy of data measurement normally has
relatively little influence on the reliability of a frequency
es timate, because such errors ordinarily are not systematic
and tend to cancel. The influence of extreme events on
reliability of frequency estimates is greater than that of
measurement errors. For this reason, it is usually better to
i ncl u de an estimated magnitude for a major flood than to
ignore it. For example, a flood event that was not recorded
because of gauge failure should be estimated, rather than to
omit it from the frequency array. However, it is advisable
to always use the most reliable sources of data and to guard
against systematic errors.
b. Errors in estimating flood frequencies. It should
be remembered that possible errors in estimating flood
frequencies are very large, principally because of the
chan ce of having a nonrepresentative sample. Sometimes
t he occurrence of one or two rare flood events can change
t he apparent exceedance frequency of a given magnitude
from once in 1,000 years to once in 200 years. Neverthel es s, the frequency-curve technique is considerably better
t han any other tool available for certain purposes and
represents a substantial improvement over using an array
restricted to observed flows only. Reliability criteria useful
for illustrating the accuracy of frequency determinations are
described in Chapter 8 of EM 1110-2-1415.
6-9. Presentation of Frequency Analysis Results
Information provided with frequency curves should clearly
i ndicate the scope of the studies and include a brief
description of the procedure used, including appropriate
references. When rough estimates are adequate or necess ary, the frequency data should be properly qualified in
order to avoid misleading conclusions that might seriously
affect the project plan. A summary of the basic data
consisting of a chronological tabulation of values used and
indicating sources of data and adjustments made would be
hel pful. The frequency data can also advantageously be
presented in graphical form, ordinarily on probability paper,
along with the adopted frequency curves.
EM 1110-2-1420
31 Oct 97
Chapter 7
Flood-Runoff Analysis
7-1. Introduction
Flood-runoff analysis is usually required for any reservoir
project. Even without flood control as a purpose, a reservoir must be designed to safely pass flood flows. Rarely
are there sufficient flow records at a reservoir site to meet
all analysis requirements for the evaluation of a reservoir
project. This chapter describes the methods used to analyze the flood hydrographs and the application of hypothetical floods in reservoir projects. Most of the details on
methods are presented in EM 1110-2-1417. The dam
safety standards are dependent on the type and location of
the dam. ER 1110-8-2 defines the requirements for design
floods to evaluate dam and spillway adequacy. Requirements for flood development and application are also
7-2. Flood Hydrograph Analysis
a. Unit hydrograph method. The standard Corps
procedure for computing flood hydrographs from catchments is the unit hydrograph method. The fundamental
components are listed below:
(1) Analysis of rainfall and/or snowmelt to determine
the time-distributed average precipitation input to each
catchment area.
(2) Infiltration, or loss, analysis to determine the
precipitation excess available for surface runoff.
(3) Unit hydrograph transforms to estimate the surface
flow hydrograph at the catchment outflow location.
(4) Baseflow estimation to determine the subsurface
contribution to the total runoff hydrograph.
(5) Hydrograph routing and combining to move
catchment hydrographs through the basin and determine
total runoff at desired locations.
b. Rainfall-runoff parameters. Whenever possible
unit hydrographs and loss rate characteristics should be
derived from the reconstitution of observed storm and
flood events on the study watershed, or nearby watersheds
with similar characteristics. The HEC-1 program has
optimization routines to facilitate the determination of bestfit rainfall-runoff parameters for each event. When runoff
records are not available at or near the location of interest,
unit hydrograph and loss characteristics must be
determined from regional studies of such characteristics
observed at gauged locations. Runoff and loss coefficients
can be related to drainage basin characteristics by multiple
correlation analysis and mapping procedures, as
described in Chapter 16, “Ungauged Basin Analysis” of
EM 1110-2-1417.
c. Developing basin models. Flood hydrographs
may be developed for a number of purposes. Basin models
are developed to provide hydrographs for historical events
at required locations where gauged data are not available.
Even in large basins, there will be limited gauged data and
many locations where data are desired. With some gauged
data, a basin model can be developed and calibrated for
observed flood events. Chapter 13 of EM 1110-2-1417
provides information on model development and
d. Estimating runoff. Basin models can estimate the
runoff response under changing conditions. Even with
historical flow records, many reservoir studies will require
estimates of flood runoff under future, changed conditions.
The future runoff with developments in the catchment and
modifications in the channel system can be modeled with a
basin runoff model.
e. Application. For reservoir studies, the most
frequent application of flood hydrograph analysis is to
develop hypothetical (or synthetic) floods. The three
common applications are frequency-storms, SPF and PMF.
Frequency-based design floods are used to develop floodfrequency information, like that required to compute
expected annual flood damage. SPF and PMF are used as
design standards to evaluate project performance under the
more rare flood events.
7-3. Hypothetical Floods
For urban catchments, the kinematic-wave approach is
often used to compute the surface flow hydrograph, instead
of unit hydrograph transforms. Each of the standard flood
runoff and routing procedures is described in Part 2 of
EM 1110-2-1417. HEC-1 Flood Hydrograph Package
(HEC 1990c) is a generalized computer program providing
the standard methods for performing the required
components for basin modeling.
a. General. Hypothetical floods are usually used in
the planning and design of reservoir projects as a primary
basis of design for some project features and to substantiate the estimates of extreme flood-peak frequency. Where
runoff data are not available for computing frequency
curves of peak discharge, hypothetical floods can be used
to establish flood magnitudes for a specified frequency
EM 1110-2-1420
31 Oct 97
from rainstorm events of that frequency. This approach is
not accurate where variations in soil-moisture conditions
and rainfall distribution characteristics greatly influence
flood magnitudes. In general, measured data should be
used to the maximum extent possible, and when approximate methods are used, several approaches should be taken
to compute flood magnitudes.
values for a flood of this magnitude.
Part 2 of
EM 1110-2-1417 provides detailed information on the unit
hydrograph procedure and the simulation of hypothetical
floods is described in Chapter 13. The computer program
HEC-1 Flood Hydrograph Package provides the SPS and
SPF computation procedures, as described in the SPF
determination manual.
b. Frequency-based design floods. In areas where
infiltration losses are small, it may be feasible to compute
hypothetical floods from rainfall amounts of a specified
frequency and to assign that frequency to the flood event.
NOAA publishes generalized rainfall criteria for the United
States. They contain maps with isopluvial lines of point
precipitation for various frequencies and durations. These
point values are then adjusted for application to areas
greater that 10 square miles, based on precipitation
duration and catchment area. Section 13-4 of EM 1110-21417 provides information on simulation with frequencybased design storms.
(3) While the frequency of the standard project flood
cannot be specified, it can be used as a guide in extrapolating frequency curves because it is considered to lie
within a reasonable range of rare recurrence intervals, such
as between once in 200 years and once in 1,000 years.
c. Standard project flood. The SPF is the flood that
can be expected from the most severe combination of
meteorologic and hydrologic conditions that are considered
reasonably characteristic of the region in which the study
basin is located. The SPF, which provides a performance
standard for potential major floods, is based on the Standard Project Storm (SPS).
(1) The SPS is usually an envelope of all or almost all
of the storms that have occurred in a given region. The
size of this storm is derived by drawing isohyetal maps of
the largest historical storms and developing a depth-area
curve for the area of maximum precipitation for each
storm. Depth-area curves for storm rainfall of specified
durations are derived from this storm-total curve by a study
of the average time distribution of precipitation at stations
representing various area sizes at the storm center. When
such depth-area curves are obtained for all large storms in
the region, the maximum values for each area size and
duration are used to form a single set of depth-areaduration curves representing standard project storm
hyetographs for selected area sizes, using a typical time
distribution observed in major storms. EM 1110-2-1411
provides generalized SPS estimates for small and large
drainage basins, and projects for which SPF estimates are
required. The generalized rainfall criteria and recommended procedures for SPS computations for U.S. drainage
basins located east of the 105th longitude are presented.
(2) The SPF is ordinarily computed using the unit
hydrograph approach with the SPS precipitation. The unit
hydrograph and basin losses should be based on reasonable
d. Probable maximum flood. The PMF is the flood
that may be expected from the most severe combination of
critical meteorologic and hydrologic conditions that are
reasonably possible in the region. The PMF is calculated
from the Probable Maximum Precipitation (PMP). The
PMP values encompass the maximized intensity-duration
values obtained from storms of a single type. Storm type
and variations of precipitation are considered with respect
to location, areal coverage of a watershed, and storm
duration. The probable maximum storm amounts are
determined in much the same way as are SPS amounts,
except that precipitation amounts are first increased to
correspond to maximum meteorologic factors such as wind
speed and maximum moisture content of the atmosphere.
(1) Estimates of PMP are based generally on the
results of the analyses of observed storms. More than 600
storms throughout the United States have been analyzed in
a uniform manner, and summary sheets have been
distributed to government agencies and the engineering
profession. These summary sheets include depth-areaduration data for each storm analyzed along with broad
outlines of storm magnitudes and their seasonal and geographical variations. NWS (1977) Hydrometeorlogical
Report No. 51 (HMR 51) contains generalized all-season
estimates for the United States, east of the 105th longitude.
The PMP is distributed in space and time to develop the
PMS, which is a hypothetical storm that produces the PMF
for a particular drainage basin.
(2) NWS (1981) HMR No. 52 provides criteria and
instructions for configuring the storm to produce the PMF.
The precipitation on a basin is affected by the storm
placement, storm-area size, and storm orientation. The
HMR52 PMS (HEC 1984) computer program uses a
procedure to produce maximum precipitation on the basin.
However, several trials are suggested to ensure that the
maximum storm is produced. The PMS is then input to a
rainfall-runoff model to determine the flood runoff.
EM 1110-2-1420
31 Oct 97
(3) The HMR52 User's Manual shows an example
application with the HEC-1 Flood Hydrograph Package.
The storm hyetographs can be written to an output file, in
HEC-1 input format, or to an HEC-DSS file. HEC-1 can
read the DSS file to obtain the basin precipitation.
(4) Hydrometeorlogical criteria are being updated for
various areas of the country. A check should be made for
the most recent criteria. Figure 13-3 in EM 1110-2-1417
shows the regional reports available in 1993. The HMR52
computer program does not apply to U.S. regions west of
the 105th meridian.
(5) In the determination of both the SPF and the PMF,
selection of rainfall loss rates and the starting storage of
upstream reservoirs should be based on appropriate
assumptions for antecedent precipitation and runoff for the
season of the storm. Also, PMF studies should consider
the capability of upstream reservoir projects to safely
handle the PMF contribution from that portion of the
watershed. There could be deficiencies in an upstream
project spillway that significantly affects the downstream
project's performance.
e. Storm duration. Hypothetical storms to be used
for any particular category of hypothetical flood computation must be based on data observed within a region. For
application in the design of local flood protection projects,
only peak flows and runoff volumes for short durations are
usually important. Accordingly, the maximum pertinent
duration of storm rainfall is only on the order of the time of
travel for flows from the headwaters to the location
concerned. After a reasonable maximum duration of
interest is established, rainfall amounts for this duration
and for all important shorter durations must be established.
For standard project storm determinations, this would
consist of the amounts of observed rainfall in the most
severe storms within the region that correspond to area
sizes equal to the drainage area above the project. In the
case of hypothetical storms and floods of a specified
frequency, these rainfall amounts would correspond to
amounts observed to occur with the specified frequency at
stations spread over an area the size of the project drainage
area. Larger rates and smaller amounts of precipitation
would occur for shorter durations, as compared with the
longer durations of interest. Once a depth-duration curve is
established that represents the desired hypothetical storm
rainfall, a time pattern must be selected that is reasonably
representative of observed storm sequences. The HEC-1
computer program has the capability of accepting any
depth-duration relation and selecting a reasonable time
sequence. It is also capable of accepting specified time
sequences for hypothetical storms.
f. Snowmelt contribution. Satisfactory criteria and
procedures have not yet been developed for the computation of standard project and probable maximum snowmelt
floods. The problem is complicated in that deep snowpack
tends to inhibit rapid rates of runoff, and consequently,
probable maximum snowmelt flood potential does not
necessarily correspond to maximum snowpack depth or
water equivalent. Snowpack and snowmelt differ at various
elevations, thus adding to the complexity of the problem.
(1) Where critical durations for project design are
short, high temperatures occurring with moderate snowpack depths after some melting has occurred will probably
produce the most critical runoff. Where critical durations
are long, as is the more usual case in the control of snowmelt floods, prolonged periods of high temperature or
warm rainfall occurring with heavy snowpack amounts will
produce critical conditions.
(2) The general procedure for the computation of
hypothetical snowmelt floods is to specify an initial snowpack for the season that would be critical. In the case of
SPF's a maximum observed snowpack should be assumed.
The temperature sequence for SPF computation would be
that which produces the most critical runoff conditions and
should be selected from an observed historical sequence.
In the case of PMF computation, the most critical
snowpack possible should be used and it should be
considerably larger or more critical than the standard
project snowpack. The temperature pattern should be
selected from historical temperature sequences augmented
to represent probable maximum temperature for the season.
Where simultaneous contribution from rainfall is possible,
a maximum rainfall for the season should be added during
the time of maximum snowmelt. This would require some
moderation of temperatures to ensure that they are
consistent with precipitation conditions. EM 1110-2-1406
covers snowmelt for design floods, standard project and
maximum probable snowmelt flood derivation.
(3) Snowmelt computations can be made in accordance with an energy budget computation, accounting for
radiation, evaporation, conductivity, and other factors, or
by a simple relation with air temperature, which reflects
most of these other influences. The latter procedure is
usually more satisfactory in practical situations. Snowmelt,
loss rate, and unit hydrograph computations can be made
by using a computer program like Flood Hydrograph
Analysis, HEC-1. EM 1110-2-1417 has detailed descriptions of each computational component.
EM 1110-2-1420
31 Oct 97
Chapter 8
Water Surface Profiles
(4) Channel slope less than 0.1 m/m - because the
hydrostatic pressure distribution is computed from the
depth of water measure vertically.
8-1. Introduction
(5) Averaged friction slope - the friction loss between
cross sections can be estimated by the product of the
representative slope and reach length.
a. General. Water surface profiles are required for
most reservoir projects, both upstream and downstream
from the project. Profile computations upstream from the
project define the “backwater” effect due to high reservoir
pool levels. The determination of real estate requirements
are based on these backwater profiles. Water surface
profiles are required downstream to determine channel
capacity, flow depths and velocities, and other hydraulic
information for evaluation of pre- and post-project
b. Choosing a method. The choice of an appropriate
method for computing profiles depends upon the following
characteristics: the river reach, the type of flow hydrograph, and the study objectives. The gradually varied,
steady flow profile computation (e.g., HEC-2), is used for
many studies. However, the selection of the appropriate
method is part of the engineering analysis. EM 1110-21416 provides information on formulating a hydraulic
study and a discussion of the analytical methods in general
use. The following sections provide general guidance on
the methods and the potential application in reservoir
related studies.
8-2. Steady Flow Analysis
a. Method assumptions. A primary consideration in
one-dimensional, gradually varied, steady flow analysis is
that flow is assumed to be constant, in time, for the profile
Additionally, all the one-dimensional
methods require the modeler to define the flow path when
defining the cross-sectional data perpendicular to the flow.
The basic assumptions of the method are as follows:
(1) Steady flow - depth and velocity at a given location
do not vary with time.
(2) Gradually varied flow - depth and velocity change
gradually along the length of the water course.
(3) One-dimensional flow - variation of flow characteristics, other than in the direction of the main axis of flow
may be neglected, and a single elevation represents the
water surface of a cross section perpendicular to the flow.
(6) Rigid boundary - the flow cross section does not
change shape during the flood.
b. Gradually varied steady flow. The assumption of
gradually varied steady flow for general rainfall and
snowmelt floods is generally acceptable.
changes slowly with time and the use of the peak discharge
for the steady flow computations can provide a reasonable
estimate for the flood profile.
Backwater profiles,
upstream from a reservoir, are routinely modeled using
steady flow profile calculations. However, inflow hydrographs from short duration, high intensity storms, e.g.,
thunderstorms, may not be adequately modeled assuming
steady flow.
c. Downstream profile. Obviously, the downstream
profile for a constant reservoir release meets the steady
flow condition. Again, the consideration is how rapidly
flow changes with time. Hydropower releases for a peaking operation may not be reasonably modeled using steady
flow because releases can change from near zero to turbine
capacity, and back, in a short time (e.g., minutes) relative to
the travel time of the resulting disturbances. Dam-break
flood routing is another example of rapidly changing flow
which is better modeled with an unsteady flow method.
d. Flat stream profiles. Another consideration is
calculating profiles for very flat streams. When the stream
slope is less than 0.0004 m/m (2 ft/mile), there can be a
significant loop in the downstream stage-discharge
relationship. Also, the backwater effects from downstream
tributaries, or storage, or flow dynamics may strongly
attenuate flow. For slopes greater than 0.0009 m/m
(5 ft/mile), steady flow analysis is usually adequate.
e. Further information. Chapter 6 of EM 1110-21416 River Hydraulics provides a detailed review of the
assumptions of the steady flow method, data requirements,
and model calibration and application. Appendix D provides information on the definition of river geometry and
energy loss coefficients, which is applicable to all the onedimensional methods.
EM 1110-2-1420
31 Oct 97
8-3. Unsteady Flow Analysis
a. Unsteady flow methods.
unsteady flow methods require the same assumptions listed
in 8-2(a), herein, except flow, depth, and velocity can vary
with time. Therefore, the primary reason for using
unsteady flow methods is to consider the time varying
nature of the problem. Examples of previously mentioned
rapidly changing flow are thunderstorm floods,
hydroelectric peaking operations, and dam-break floods.
The second application of unsteady flow analysis consideration, mentioned above, is streams with very flat slopes.
portions of a study area at the design stage of a project.
The typical river-reservoir application requires both the
direction and magnitude of velocities. Potential model
applications include areas upstream and downstream from
reservoir outlets. Additionally, flow around islands, and
other obstructions, may require two-dimensional modeling
for more detailed design data.
c. Further information. Chapter 4 of EM 1110-21416 provides a review of model assumptions and typical
8-5. Movable-Boundary Profile Analysis
b. Predicting downstream stages. Another application of unsteady flow is in the prediction of downstream
stages in river-reservoir systems with tributaries, or lockand-dam operations where the downstream operations
affect the upstream stage. Flow may not be changing rapidly with time, but the downstream changes cause a time
varying downstream boundary condition that can affect the
upstream stage. Steady flow assumes a unique stagedischarge boundary condition that is stable in time.
c. Further information. Chapter 5, “Unsteady Flow,”
in EM 1110-2-1416 provides a detailed review of model
application including selection of method, data requirements, boundary conditions, calibration, and application.
8-4. Multidimensional Analysis
a. Two- and three-dimensional modeling. Multidimensional analysis includes both two- and
three-dimensional modeling. In river applications, twodimensional modeling is usually depth-averaged. That is,
variables like velocity do not vary with depth, so an average
value is computed. For deep reservoirs, the variation of
parameters with depth is often important (see Chapter 12,
EM 1110-2-1201). Two-dimensional models, for deep
reservoirs, are usually laterally-averaged. Three-dimensional models are available; however, their applications
have mostly been in estuaries where both the lateral and
vertical variation are important.
b. Two-dimensional analysis.
depth-averaged analysis is usually performed in limited
a. Reservoirs. Reservoirs disrupt the flow of sediment when they store or slow down water. At the upper
limit of the reservoir, the velocity of inflowing water
decreases and the ability to transport sediment decreases
and deposition occurs. Chapter 9 herein presents reservoir
sediment analysis. Reservoir releases may be sediment
deficient, which can lead to channel degradation
downstream from the project because the sediment is
removed from the channel.
b. River and reservoir sedimentation. EM 1110-24000 is the primary Corps reference on reservoir sedimentation. Chapter 3 covers sediment yield and includes
methods based on measurement and mathematical models.
Chapter 4 covers river sedimentation, and Chapter 5 presents reservoir sedimentation. Section III, of Chapter 5,
provides an overview of points of caution, sedimentation
problems associated with reservoirs, and the impact of
reservoirs on the stream system. Section IV provides
information on levels of studies and study methods.
b. Further information. Chapter 7 of EM 1110-21416 presents water surface profile computation with
movable boundaries. The theory, data requirements and
sources, plus model development and application are all
covered. The primary math models, HEC-6 Scour and
Deposition in Rivers and Reservoirs (HEC 1993) and
Open-Channel Flow and Sedimentation TABS-2 (Thomas
and McAnally 1985) two-dimensional modeling package,
are also described. The focus for the material is riverine.
EM 1110-2-1420
31 Oct 97
Chapter 9
Reservoir Sediment Analysis
d. Further information. The primary Corps reference for sediment analysis is EM 1110-2-4000. Major
topics include developing a study work plan, sediment
yield, river sedimentation, reservoir sedimentation, and
model studies.
9-1. Introduction
9-2. Sediment Yield Studies
a. Parameters of a natural river. Nature maintains a
very delicate balance between the water flowing in a
natural river, the sediment load moving with the water, and
the stream's boundary. Any activity which changes any one
of the following parameters:
water yield from the watershed.
sediment yield from the watershed.
water discharge duration curve.
depth, velocity, slope or width of the flow.
size of sediment particles.
or which tends to fix the location of a river channel on its
floodplain and thus constrains the natural tendency will
upset the natural trend and initiate the formation of a new
one. The objective of most sediment studies is to evaluate
the impact on the flow system resulting from changing any
of these parameters.
b. Changes caused by reservoirs. Reservoirs interrupt the flow of water and, therefore, sediment. In terms of
the above parameters, the reservoir causes a change in the
upstream hydraulics of flow depth, velocity, etc. by forcing
the energy gradient to approach zero. This results in a loss
of transport capacity with the resulting sediment deposition
in the reservoir. The reservoir also alters the downstream
water discharge-duration relation and reduces the sediment
supply which may lead to the degradation of the
downstream channel.
c. Areas of analysis. Sedimentation investigations
usually involve the evaluation of the existing condition as
well as the modified condition. The primary areas of
reservoir sediment analysis are the estimation of volume
and location of sediment deposits in the reservoir and the
evaluation of reservoir releases' impact on the downstream
channel system. Sediment deposits start in the backwater
area of the reservoir, which increase the elevation of the
bed profile and the resulting water surface profile.
However, reservoirs may also cause sediment deposits
upstream from the project, which affect the upstream water
surface profiles.
a. General. Sediment yield studies determine the
amount of sediment that leaves a basin for an event or over
a period of time. Sediment yield, therefore, involves
erosion processes as well as sediment deposition and
delivery to the study area. The yield provides the necessary
input to determine sedimentation impacts on a reservoir.
b. Required analysis.
Each reservoir project
requires a sediment yield analysis to determine the storage
depletion resulting from the deposition of sediment during
the life of the project. For most storage projects, as
opposed to sediment detention structures, the majority of
the delivered sediment is suspended. However, the data
required for the headwater reaches of the reservoir should
include total sediment yield by particle size because that is
where the sands and gravels will deposit.
c. Further information. Corps of Engineer methods
for predicting sediment yields are presented in Appendix C
of EM 1110-2-4000. A literature review, conducted by the
Hydrologic Engineering Center under the Land Surface
Erosion research work unit, showed numerous mathematical models are available to estimate sediment discharge
rates from a watershed and the redistribution of soil within
a watershed. An ETL on the methods will be issued soon.
9-3. Reservoir Sedimentation Problems
a. Sediment deposition. As mentioned above, the
primary reservoir sediment problem is the deposition of
sediment in the reservoir. The determination of the sediment accumulation over the life of the project is the basis
for the sediment reserve. Typical storage diagrams of
reservoirs, showing sediment (or dead) storage at the
bottom of the pool can be misleading. While the reservoir
storage capacity may ultimately fill with sediment, the
distribution of the deposits can be a significant concern
during the life of the project. The reservoir sedimentation
study should forecast sediment accumulation and
distribution over the life of the project. Sediment deposits
in the backwater area of the reservoir may form deltas,
particularly in shallow reservoirs. A number of problems
associated with delta formations are discussed below.
(1) Deposits forming the delta may raise the water
surface elevation during flood flows, thus requiring special
EM 1110-2-1420
31 Oct 97
consideration for land acquisition. In deep reservoirs, this
is usually not a problem with the reservoir area because
project purposes dictate land acquisitions or easements.
Deltas tend to develop in the upstream direction. In
shallow reservoirs, the increase in water surface elevation
is a problem even within the reservoir area. That is, floods
of equal frequency may have higher water surface
elevations after a project begins to develop a delta deposit
than was experienced before the project was constructed.
Land acquisition studies must consider such a possibility.
(2) Aggradation problems are often more severe on
tributaries than on the main stem. Analysis is complicated
by the amount of hydrologic data available on the
tributaries, which is usually less than on the main stem
itself. Land use along the tributary often includes recreation sites, where aggradation problems are particularly
(3) Reservoir deltas often attract phreatophytes due to
the high moisture level. This may cause water-use problems due to their high transpiration rate.
(4) Reservoir delta deposits are often aesthetically
(5) Reservoir sediment deposits may increase the water
surface elevation sufficiently to impact on the groundwater
table, particularly in shallow impoundments.
(6) In many existing reservoirs, the delta and backwater-swamp areas support wildlife. Because the characteristics of the area are closely controlled by the operation
policy of the reservoir, any reallocation of storage would
need to consider the impact on the present delta and swamp
b. Upstream projects. It is important to identify and
locate all existing reservoirs in a basin where a sediment
study is to be made. The projects upstream from the point
of analysis potentially modify both the sediment yield and
the water discharge duration curve.
The date of
impoundment is important so that observed inflowing
sediment loads may be coordinated with whatever conditions existed in the basin during the periods selected for
calibration and verification. Also, useful information on
the density of sediment deposits and the gradation of
sediment deposits along with sediment yield are often
available from other reservoirs in the basin. Information on
the rate of sediment deposition that has occurred at other
reservoir sites in the region is the most valuable
information when estimating sediment deposition for a new
9-4. Downstream Sediment Problems
a. Channel degradation.
Channel degradation
usually occurs downstream from the dam. Initially, after
reservoir construction, the hydraulics of flow (velocity,
slope, depth, and width) remain unchanged from preproject conditions. However, the reservoir acts as a sink
and traps sediment, especially the bed material load. This
reduction in sediment delivery to the downstream channel
causes the energy in the flow to be out of balance with the
boundary material for the downstream channel. Because of
the available energy, the water attempts to re-establish the
former balance with sediment load from material in the
stream bed, and this results in a degradation trend.
Initially, degradation may persist for only a short distance
downstream from the dam because the equilibrium
sediment load is soon re-established by removing material
from the stream bed.
b. Downstream migratory degradation. As time
passes, degradation tends to migrate downstream. However, several factors are working together to establish a
new equilibrium condition in this movable-boundary flow
system. The potential energy gradient is decreasing
because the degradation migrates in an upstream-todownstream direction. As a result, the bed material is
becoming coarser and, consequently, more resistant to
being moved. This tendency in the main channel has the
opposite effect on tributaries. Their potential energy
gradient is increasing which results in an increase in
transport capacity. This will usually increase sediment
passing into the main stem which tends to stabilize the
main channel resulting in less degradation than might be
anticipated. Finally, a new balance will tend to be established between the flowing water-sediment mixture and the
c. Extent of degradation. The extent of degradation
is complicated by the fact that the reservoir also changes
the discharge duration curve. This will impact for a
considerable distance downstream from the project because
the existing river channel reflects the historical phasing
between flood flows on the main stem and those from
tributaries. That phasing will be changed by the operation
of the reservoir. Also, the reduced flow will probably
promote vegetation growth at a lower elevation in the
channel. The result is a condition conducive to deposition
in the vegetation. Detailed simulation studies should be
performed to determine future channel capacities and to
identify problem areas of excessive aggradation or
degradation. All major tributaries should be included.
EM 1110-2-1420
31 Oct 97
9-5. Sediment Water Quality
a. Sediments and pollutants. When a river carrying
sediments and associated pollutants enters a reservoir, the
flow velocity decreases and the suspended and bed load
sediments start settling down. Reservoirs generally act as
depositories for the sediments because of their high sediment trap efficiency. Due to a high adsorption capacity,
sediments act as sinks for contaminants in the reservoirs
and, in agricultural and industrial areas, may contain PCB's,
chlorinated hydrocarbon pesticides, oil and grease, heavy
metals, coliform bacteria, or mutagenic substances. Burial
of these contaminants by sedimentation may be an
important factor and an effective process in isolating
potentially toxic substances from surface waters and
important biological populations. Toxic inorganic and
organic contaminants associated with the sediments can
also be bioconcentrated by the aquatic organisms present in
of the study. Chapter 1 of EM 1110-2-4000 describes
staged sedimentation studies in Section I. Section II, of
that chapter, provides reporting requirements. Problem
identification and the development of a study work plan are
covered in Chapter 2.
b. Sediment deposits. Considering a dam site as an
important natural resource, it is essential to provide enough
volume in the reservoir to contain anticipated deposits
during the project life. If the objective of a sediment study
is just to know the volume of deposits for use in screening
studies, then trap efficiency techniques can provide a
satisfactory solution. The important information that must
be available is the water and sediment yields from the
watershed and the capacity of the reservoir. Chapter 3 of
EM 1110-2-4000 covers sediment yield. Section 3-7
provides information on reservoir sedimentation, including
trap efficiency.
b. Monitoring chemical contaminants. These incoming sediments and associated pollutants significantly affect
the water quality of the reservoir pool and downstream
releases. Therefore, it is essential that these sediment
reservoir interactions be characterized by their depositional
behavior, particle size distribution, and pollutant
concentrations to successfully plan a management strategy
to quantify contaminant movement within reservoirs.
Analytical and predictive methods to assess the influence
of contaminated sediments in reservoirs have not been
developed enough to be used in Corps field offices, but
WES Instruction Report E-86-1, “General Guidelines for
Monitoring Contaminants in Reservoirs” (Waide 1986),
does provide general guidance on the design and conduct of
programs for monitoring chemical contaminants in
reservoir waters, sediments, and biota.
c. Land acquisition. If the sediment study must
address land acquisition for the reservoir, then knowing
only the volume of deposits is not sufficient. The location
of deposits must also be known, and the study must take
into account sediment movement. This generally requires
simulation of flow in a mobile boundary channel. Sorting
of grain sizes must be considered because the coarser
material will deposit first, and armoring must be considered
because scour is involved. Movable-bed modeling is useful
to predict erosion or scour trends downstream from the
dam, general aggradation or degradation trends in river
channels, and the ability of a stream to transport the bedmaterial load. The computer program, HEC-6 Scour and
Deposition in Rivers and Reservoirs (HEC 1993), is
designed to provide long-term trends associated with
changes in the frequency and duration of the water
discharge and/or stage or from modifying the channel
c. Sedimentation patterns. Sedimentation patterns
can often be associated with water quality characteristics.
There seems to be a relationship between longitudinal
gradients in water quality (a characteristic of many reservoirs) and sediment transport and deposition. High concentrations of inorganic particulates can reduce light availability near inflows and thus influence algal production and
decrease dissolved oxygen. The association of dissolved
substances, such as phosphorus, with suspended solids may
act to reduce or buffer dissolved concentrations, thus
influencing nutrient availability.
d. Details of investigations. The details of reservoir
sedimentation investigations are covered in Chapter 5 of
EM 1110-2-4000. The primary emphasis is on the
evaluation of the modified condition, which includes
consideration of quality and environmental issues. The
levels of sedimentation studies and methods of analysis are
presented in Section IV of Chapter 5. Model studies and a
short review of HEC-6 and the two-dimensional TABS-2
modeling system are covered in Chapter 6.
9-6. Sediment Investigations
a. General. The level of detail required for the
analysis of any sediment problem depends on the objective
EM 1110-2-1420
31 Oct 97
EM 1110-2-1420
31 Oct 97
Chapter 10
Flood-Control Storage
10-1. General Considerations
a. Reservoir flood storage. Where flood damage at a
number of locations on a river can be significantly reduced
by construction of one or more reservoirs, or where a
reservoir site immediately upstream from one damage
center provides more economical protection than local
protection works, reservoir flood storage should be
considered. Whenever such reservoirs can serve needs
other than flood control, the integrated design and operation of the project for multipurpose use should be
b. Flood-control features. In planning and designing
the flood-control features of a reservoir, it is important that
the degree and extent of continuous ensured protection be
no less than that provided by a local protection project, if
the alternatives of reservoir construction or channel and
levee improvement are to be evaluated fairly. This means
that the storage space and release schedule for flood
control must be provided at all times when the flooding
potential exists. In some regions this may be for the entire
year, but more commonly there are dry seasons when the
flood potential is greatly reduced and storage reservation
for flood control can be reduced correspondingly. Except
where spring snowmelt floods can be forecasted reliably or
where safe release rates are sufficient to empty flood space
in a very short time, it is not ordinarily feasible to provide
flood-control space only after a flood is forecasted. Space
must be provided at all times during the flood season unless
it can be demonstrated that the necessary space can be
evacuated on a realistic forecast basis. Also, space may be
reduced if less storage is needed due to low snowpack, or
there is some other reliable basis for long range flood
c. Runoff volume durations. Whereas the peak rates
of runoff are critical in the design of local protection
projects, runoff volumes for pertinent durations are critical
in the design of reservoirs for flood control. The critical
durations will be a function of the degree of flood protection selected and of the release rate or maximum rate of
flow at the key downstream control point. As the proposed
degree of protection is increased and as the proposed rates
of controlled flows at key damage centers are reduced, the
critical duration is increased. If this critical duration
corresponds to the duration of a single rainstorm period or
a single snowmelt event, the computation of hypothetical
floods from rainfall and snowmelt can constitute the
principle hydrologic design element. On the other hand, if
the critical duration is much longer, hypothetical floods and
sequences of hypothetical floods computed from rainfall or
snowmelt become less dependable as guides to design. It
then is necessary to base the design primarily on the
frequency of observed runoff volumes for long durations.
Even when this is done, it will be advisable to construct a
typical hydrograph that corresponds to runoff volumes for
the critical duration and that reasonably characterizes
hydrographs at the location, in order to examine the
operation of the proposed project under realistic
d. Hypothetical flood simulations. When hypothetical floods are selected, they must be routed through the
proposed reservoir under the operation rules that would be
specified for that particular design. In effect, a simulation
study of the proposed project and operation scheme would
be conducted for each flood. It is also wise to simulate the
operation for major floods of historical record in order to
ensure that some peculiar feature of a particular flood does
not upset the plan of operation. With present software, it is
relatively inexpensive to perform a complete period of
record simulation once the flood-control storage is set.
10-2. Regulated Release Rates
a. Flood reduction purposes. For flood reduction
purposes reservoirs must store only the water that cannot
be released without causing major damage downstream. If
more water can be released during a flood, less water needs
to be stored. Thus, less storage space needs to be planned
for flood control. Because reservoir space is costly and
usually in high demand for other purposes, good floodcontrol practice consists of releasing water whenever
necessary at the highest practical rates so that a minimum
amount of space need be reserved for flood control. As
these rates increase, it becomes costly also to improve
downstream channels and to provide adequate reservoir
outlets, so there is an economic balance between release
rates and storage capacity for flood control. In general, it is
economical to utilize the full nondamage capacity of
downstream channels, and it may pay to provide some
additional channel or levee improvements downstream.
However, as described in paragraph f, full channel capacity
may not be available, so analyses should consider the
impact of reduced capacity.
b. Channel capacities. Channel capacities should be
evaluated by examing water-surface profile data from
actual flood events whenever possible. Under natural
channel conditions, it will ordinarily be found that floods
which occur more frequently than once in two years are not
seriously damaging, while larger floods are.
EM 1110-2-1420
31 Oct 97
c. Minor versus major damage releases. In some
cases, it is most economical to sustain minor damage by
releasing flows above nondamaging stages in order to
accommodate major floods and thereby protect the more
important potential damage areas from flooding. In such
situations, a stepped-release schedule designed to protect
all areas against frequent minor floods, with provision to
increase releases after a specified reservoir stage is
reached, might be considered. However, such a plan has
serious drawbacks in practice because protection of the
minor damage areas would result in greater improvements
in those areas; and it soon becomes highly objectionable, if
not almost impossible, to make the larger releases when
they are required for protection of major damage areas. In
any case, it is necessary to make sure that the minor
damage areas are not flooded more frequently or severely
with the project than they would have been without it.
d. Maintenance and zoning. It is important on all
streams in developed areas to provide for proper maintenance of channel capacity and zoning of the floodplain
where appropriate. This is vital where upstream reservoirs
are operated for flood control because proper reservoir
regulation depends as much on the ability to release
without damage as it does on the ability to store. Minor
inadequacies in channel capacity can lead to the loss of
control and result in major flooding. This situation is
aggravated because the reduced frequency of flooding
below reservoirs and the ability to reduce reservoir releases
when necessary often increase the incentive to develop the
floodplain and sometimes even remove the incentive for
maintaining channel capacity.
e. Forecasted runoff. When a reservoir is located
some distance upstream from a damage center, allowance
must be made for any runoff that will occur in the intermediate area. This runoff must be forecasted, a possible
forecast error added, and the resulting quantities subtracted
from project channel capacity to determine per-missible
release rates considering attenuation when routing the
release from the dam to the damage center and the contribution of flow from the intermediate drainage area.
Also, with high intensity rainfall, the added rainfall depth to
the total downstream channel flow should be considered.
f. Delaying flood releases. Experience in the floodcontrol operation of reservoirs has demonstrated that the
actual operation does not make 100 percent use of
downstream channel capacities. Due to many contributing
factors average outflows during floods are less than maximum permissible values. It is usually wise to approach
maximum release rates with caution, in order to ascertain
any changes in channel capacity that have taken place since
the last flood, and this practice reduces operational
efficiency. It may be necessary to delay flood releases to
permit removal of equipment, cattle, etc., from areas that
would be flooded. Releases might be curtailed temporarily
in order to permit emergency repairs to canals, bridges, and
other structures downstream. If levees fail, releases might
be reduced in order to hasten the drainage of flooded areas.
Release can be reduced in order to facilitate rescue
operations. These and various other conditions result in
reduced operation efficiency during floods. To account for
this, less nondamage flow capacity than actually exists
(often about 80 percent) is assumed for design studies. It is
important, however, that every effort be made in actual
operation to effect the full non-damage releases in order to
attain maximum flood-control benefits.
g. Gradually increasing and decreasing releases.
During flood operations, reservoir releases must be
increased and decreased gradually in order to prevent
damage and undue hardship downstream. Gradually
increasing releases will usually permit an orderly evacuation of people, livestock, and equipment from the river
areas downstream. If releases are curtailed too rapidly,
there is some danger that the saturated riverbanks will
slough and result in the loss of valuable land or damage to
10-3. Flood Volume Frequencies
a. Critical durations. Flood volume frequency
studies usually consist of deriving frequency curves of
annual maximum volumes for each of various specified
durations that might be critical in project design. Critical
durations range from a few hours in the case of regulating
“cloudburst” floods to a few months where large storage
and very low release rates prevail. The annual maximum
volumes for a specific duration are usually expressed as
average rates of flow for that duration. It is essential that
these flows represent a uniform condition of development
for the entire period of observation, preferably unregulated
conditions. Procedures for computing the individual
frequency curves are discussed briefly in Chapter 6 herein
and are described in detail in EM 1110-2-1415.
b. Flood-control space requirement. Determination
of the flood-control space needed to provide a selected
degree of protection is based on detailed hydrograph
analysis, but a general evaluation can be made as illustrated
in Figure 10-1. The curve of runoff versus duration is
obtained from frequency studies of runoff volumes or from
SPF studies at the location. The tangent line represents a
uniform flow equal to the project release capacity (reduced
by an appropriate contingency factor). The intercept
represents the space required for control of the flood. The
chart demonstrates that a reservoir capable of storing
EM 1110-2-1420
31 Oct 97
c. Longer duration floods. Where flood durations
longer than the typical single-flood duration are important
in the design, a sequence of flood hydrographs spaced
reasonably in time should be used as a pattern flood. In
order to represent average natural sequences of flood
events, the largest portions of the pattern flood should
ordinarily occur at or somewhat later than the midpoint of
the entire pattern, because rainfall sequences are fairly
random but ground conditions become increasingly wet
and conducive to larger runoff as any flood sequence
Figure 10-1. Flood-control space requirement
155,000 units of water and releasing 30,000 units per day
can control 100-year runoff for any duration, and that the
critical duration (period of increasing storage) is about
5 days. The volume-duration curve would be made for
each damage area and should include more than 100 percent of the local uncontrolled runoff downstream from the
reservoir and above the control point in order to allow for
errors of forecast which would be reflected in reduced
project releases. If this local runoff appreciably exceeds
nondamage flow capacity at the damage centers, the volume over and above the flow capacity is damaging water
that cannot be stored in the project reservoir.
10-4. Hypothetical Floods
a. Two classes. Two classes of hypothetical floods
are important in the design of reservoirs for flood control.
One is a balanced flood that corresponds to a specified
frequency of occurrence; the other is a flood that represents a maximum potential for the location, such as the
SPF or PMF. ER 1110-8-2(FR) sets forth hydrologic
engineering requirements for selecting and accommodating inflow design floods for dams and reservoirs.
b. Specified frequencies.
A hypothetical flood
corresponding to a specified frequency should contain
runoff volumes for all pertinent durations corresponding to
that specified frequency. The derivation of frequency
curves is as discussed in the preceding section. A balanced flood hydrograph is constructed by selecting a
typical hydrograph pattern and adjusting the ordinates so
that the maximum volumes for each selected duration
correspond to the volumes for that duration at the specified
d. Maximum flood potential. Two types of hypothetical floods that represent maximum flood potential are
important in the design of reservoirs. The PMF, which is
the largest flood that is reasonably possible at the location,
is ordinarily the design flood for the spillway of a structure
where loss of life or major property damage would occur
in the event of project failure. The SPF, which represents
the largest flood for that location that is reasonably
characteristic for the region, is a flood of considerably
lesser magnitude and represents a high degree of design
for projects protecting major urban and industrial areas.
These floods can result from heavy rainfall or from
snowmelt in combination with some rainfall.
e. Computing hydrographs. SPF and PMF hydrographs are computed from the storm hyetographs by unit
hydrograph procedures. In the case of the SPF, ground
conditions that are reasonably conducive to heavy runoff
are used. In the case of the PMF, the most severe ground
conditions that are reasonably consistent with storm magnitudes are used. A general description of these analyses
is provided in Chapter 7 of this manual. Detailed methods
for performing these computations are described in EM
1110-2-1417. The computer program HEC-1 Flood
Hydrograph Package contains routines for computing
floods from rainfall and snowmelt and also contains
standard project criteria for the eastern United States.
10-5. Operation Constraints and Criteria
a. General. As stated earlier, whenever flood
releases are required, it is imperative that they be made at
maximum rates consistent with the conditions downstream. This means that the outlets should be designed to
permit releases at maximum rates at all reservoir levels
within the flood-control space. In some cases where
controlled releases are very high, such an outlet design is
not economical, and releases at lower stages might be
restricted because of limited outlet capacity. This constraint, of course, should be taken into account during the
design studies.
EM 1110-2-1420
31 Oct 97
b. Downstream damage centers. Where damage
centers are at some distance downstream from the reservoir, local runoff below the reservoir and above the damage
center must be considered when determining releases to be
made. This will ordinarily require some forecasting of the
local runoff and, consequently, some estimate of the
forecast uncertainty. The permissible release at any time is
determined by adding a safe error allowance to the
forecasted local inflow and subtracting this sum from the
nondamaging flow capacity.
c. Rate-of-change of release. The rate-of-change of
release must be restricted to the maximum changes that
will not cause critical conditions downstream. As a practical matter, these rates-of-change of release should be less
than the rates-of-change of flow that occurred before the
reservoir was built. After the main flood has passed, water
stored in the flood-control space must be released and
maximum rates of release will continue until the desired
amount of water is released, except that the rate of release
should be decreased gradually toward the end of the release
period. This reduction in release must be started while
considerable flood waters remain in the reservoir in order
that water retained for other purposes is not inadvertently
released. Schedules for this operation are discussed in
Part 3.
10-6. Storage Capacity Determinations
a. Determining required storage capacity.
storage capacity required to regulate a specific flood (represented by a flood hydrograph at the dam) to a specified
control discharge immediately downstream of the dam is
determined simply by routing the hydrograph through a
hypothetical reservoir with unlimited storage capacity and
noting the maximum storage. However, there are many
special practical considerations that complicate this
process. Release rates should not be changed suddenly;
therefore, the routing should conform to criteria that
specify the maximum rate of change of release. Also,
outlet capacities might not be adequate to supply full
regulated releases with low reservoir stages. If this is the
case, a preliminary reservoir design is required in order to
define the relation of storage capacity to outlet capacity.
b. Specified flood. In the more common cases, where
damage centers exist at some distance downstream of the
reservoir, the storage requirement for a specified flood is
determined by successive approximations, operating the
hypothetical reservoir to regulate flows at each damage
center to nondamaging capacity, and allowing for local
inflow and for some forecasting error.
c. Detailed operational study. Although there are
approximate methods for estimating storage capacity, it is
essential that the final project design be tested by a detailed
operational study. The analyses are based on actual outlet
capacities and realistic assumptions for limiting rates of
release change, forecast errors, and operational
contingencies, and include various combinations of
reservoir inflow and local flow that can produce a specific
downstream flood event. It is also important to route the
largest floods of record and synthetic floods through the
project to determine that the project design is adequate and
that the project provides the degree of protection for which
it was designed.
d. Seasonal distribution of storage requirements.
Where some of the flood-control space will be made
available for other uses during the dry season, a seasonal
distribution of flood-control storage requirement should be
developed. The most direct approach to this entails the
construction of runoff frequency curves for each month of
the year. The average frequency of the design flood during
the rainy-season months can be used to select flood
magnitudes for other months. These could then serve as a
basis for determining the amount of space that must be
made available during the other months.
e. Further information. Sequential routing in planning, design, and operation of flood-control reservoirs can
be accomplished with the computer program HEC-5 Simulation of Flood Control and Conservation Systems
(HEC 1982c).
10-7. Spillways
Spillways are provided to release floodwater which normally cannot be passed by other outlet works. The spillway
is sized to ensure the passage of major floods without
overtopping the dam. A general discussion of spillways is
provided in Section 4-2 of EM 1110-2-3600. EM 1110-21603 describes the technical aspects of design for the
hydraulic features of spillways and ER 1110-8-2(FR) sets
forth requirements for selecting and accommodating inflow
design floods.
a. Spillway design flood. The spillway design flood
is usually selected as a large hypothetical flood derived
from rainfall and snowmelt. Other methods of estimating
extreme flood magnitudes, such as flood-frequency analysis, are not reliable due to limited observations. The
selection of a spillway design flood depends on the policies
of the construction agency and regulations governing dam
construction. Usually, the spillways for major dams, whose
EM 1110-2-1420
31 Oct 97
failure might constitute a major disaster, are designed to
pass the PMF without a major failure; however, the
spillways for many small dams are designed for smaller
floods such as the SPF.
b. Hydrologic design. The hydrologic design of a
spillway is accomplished by first estimating a design and
then testing it by routing the spillway design flood. In
routing the spillway design flood, the initial reservoir stage
should be as high as reasonably expected at the start of
such a major flood, considering the manner in which the
reservoir is planned to operate or how in the future the
reservoir might operate differently from the planned
operation. In the case of ungated spillways, it is possible
that the outlets of the dam will be closed gradually as the
spillway goes into operation, in order to delay damaging
releases as long as possible and possibly to prevent them.
However, if spillway flows continue to increase, it may be
necessary to reopen the outlets. In doing so, care should be
exercised to prevent releases from exceeding maximum
inflow quantities. The exact manner in which outlets will
be operated should be specified so that the spillway design
will be adequate under conditions that will actually prevail
after project construction. Consideration should be given
to the possibility that some outlets or turbines might be out
of service during flood periods.
c. Large spillway gates. The operation of large
spillway gates can be extremely hazardous, since opening
them inadvertently might cause major flooding at downstream areas. Their operation should be controlled by rigid
regulations. In particular, the opening of the gates during
floods should be scheduled on the basis of inflows and
reservoir storage so that the lake level will continue to rise
as the gates are opened. This will ensure that inflow
exceeds outflow as outflows are increased. The adequacy
of a spillway to pass the spillway design flood is tested for
gated spillways in the same manner as for ungated
spillways described above.
Methods for developing
spillway-gate operation regulations are described in
Chapter 14.
d. Preventing overtopping.
To ensure that the
spillway is adequate to protect the structure from overtopping, some amount of freeboard is added to the dam above
the maximum pool water-surface elevation. This can vary
from zero for structures that can withstand overtopping to
2 m or more for structures where overtopping would
constitute a major hazard. The freeboard allowance
accounts for wind set and wave action. Methods for
estimating these quantities are discussed in Chapter 15.
Risk analysis should be performed to determine the appropriate top-of-dam elevation.
e. Spillway types. While the spillway is primarily
intended to protect the structure from failure, the fact that it
can cause some water to be stored above ordinary full pool
level (surcharge storage) is of some consequence in
reducing downstream flooding. Narrow, ungated spillways
require higher dams and can, therefore, be highly effective
in partially regulating floods that exceed project design
magnitude, whereas wide spillways and gated spillways are
less effective for regulating floods exceeding design
magnitude. Where rare floods can cause great damage
downstream, the selection of spillway type and
characteristics can appreciably influence the benefits that
are obtained for flood control. Accordingly, it is not
necessarily the least costly spillway that yields the most
economical plan of development. In evaluating floodcontrol benefits, computing frequency curves for regulated
conditions should be based on spillway characteristics and
operation criteria as well as on other project features.
10-8. Flood-Control System Formulation
a. Objectives. The objectives of system formulation
are to identify the individual components, determine the
size of each, determine the order in which the system
components should be implemented, and develop and
display the information required to justify the decisions and
thus secure system implementation. Section 4-10 describes
several formulation strategies.
b. Criteria. Criteria for system formulation are
needed to distinguish the best system from among
competing alternative systems. The definition of “best” is
crucial. A reasonable viewpoint would seem to recognize
that simply aggregating the most attractive individual components into a system, while assuring physical compatibility, could result in the inefficient use of resources
because of system effects, data uncertainty, and the
possibility that all components may not be implemented. It
is proposed that the best system be considered to be as
(1) The system that includes the obviously good
components while preserving flexibility for modification of
components at future dates.
(2) The system which could be implemented at a
number of stages, if staging is possible, such that each
stage could stand on its own merits (be of social value) if
no more components were to be added.
c. General guidance. General guidance for formulation criteria are contained in the Principles and Standards
(Water Resources Council 1973).
The criterion of
EM 1110-2-1420
31 Oct 97
economic efficiency from the national viewpoint has been
interpreted to require that each component in a system
should be incrementally justified, that is, each component
addition to a system should add to the value (net benefits)
of the total system. The environmental quality criteria can
be viewed as favoring alternatives that can be structured to
minimize adverse environmental impacts and provide
opportunities for mitigation measures. Additional criteria
that are not as formally stated as U.S. national policy are
important in decisions among alternatives. A formulated
flood-control system must draw sufficient support from
responsible authorities in order to be implemented. In
addition, flood-control systems should be formulated so
that a minimum standard of performance (degree of risk) is
provided so that public safety and welfare are adequately
d. Environmental and other assessments. Of these
criteria, only the national economic efficiency and minimum performance standard have generally accepted
methods available for their rigorous inclusion in formulation studies. Environmental quality analysis and social/
political/institutional analyses related to implementation
have not developed technology applicable on a broad scale.
As a consequence, these criteria must guide the formulation
studies but, as yet, probably cannot directly contribute in a
structured formulation strategy. In discussions that follow,
focus is of necessity upon the economic criteria with
acceptable performance as a constraint, with the assumption that the remaining criteria will be incorporated when
the formulation strategy has narrowed the range of alternatives to a limited number for which the environmental and
other assessments can be performed.
e. Degrees of uncertainty. There will be varying
degrees of uncertainty in the information used in system
formulation. The hydrology will be better defined near
gauging stations than it is in remote areas, and certain
potential reservoirs will have been more thoroughly investigated than others. In addition, the accuracy of economic
data, both costs and value, existing or projected, is generally lower than the more physically based data. Also, since
conditions change over time, the data must be continuously
updated at each decision point. The practical accommodation of information uncertainty is by limited sensitivity
analysis and continuing reappraisal as each component of a
system is studied for implementation.
f. Sensitivity analysis. Sensitivity analysis has, as its
objective, the identification of either critical elements of
data, or particularly sensitive system components, so that
further studies can be directed toward firming up the
uncertain elements or that adjustments in system formulation can be made to reduce the uncertainty. Because
combinations of historic and synthetic floods are typically
used to evaluate reservoir flood-reduction performance
(i.e., to develop regulated conditions frequency relations at
damage index stations), particular attention must be paid to
the selection or development of the system hydrology. The
problem arises when evaluating complex reservoir systems
with many reservoirs above common damage centers. The
problem increases with the size and complexity of the basin
because the storm magnitudes and locations can favor one
reservoir location over another. There are a large number
of storm centerings that could yield similar flows at a
particular control point. Because of this, the contribution
of a specific system component to reduced flooding at a
downstream location is uncertain and dependent upon
storm centering. This makes the selection or development
of representative centerings crucial if all upstream
components are to be evaluated on a comparable basis.
g. Desired evaluation. The desired evaluation for
regulated conditions is the expected or average condition so
that economic calculations are valid. The representative
hydrograph procedure is where several proportions (ratios
of one or more historic or synthetic events used to
represent system hydrology) are compatible with the simulation technique used, but care must be taken to reasonably accommodate the storm centering uncertainty.
Testing the sensitivity of the expected annual damage to
the system hydrology (event centering) is appropriate and
necessary. Even if all historical floods of record are used,
there still may be some bias in computing expected annual
damages if most historical floods were, by chance, centered
over a certain part of the basin and not over others. For
instance, one reservoir site may have experienced several
severe historical floods, while another site immediately
adjacent to the area may, due to chance, not have had any
severe floods.
10-9. General Study Procedure
After various alternative locations are selected for a reservoir site to protect one or more damage centers, the following steps are suggested for conducting the required
hydrologic engineering studies:
a. Obtain a detailed topographic map of the region
showing the locations of the damage areas, of proposed
reservoir sites, and of all pertinent precipitation, snowpack,
and stream-gauging stations. Prepare a larger scale
topographic map of the drainage basin tributary to the most
downstream damage location. Locate damage centers,
project sites, pertinent hydrologic measurement stations,
and drainage boundaries above each damage center, project
site, and stream-gauging station. Measure all pertinent
tributary areas.
EM 1110-2-1420
31 Oct 97
b. Establish stage-discharge relations for each damage reach, relating the stages for each reach to a selected
index location in that reach; procedures for doing this are
described in Flood-Damage Analysis Package User's
Manual (HEC 1990b). Where local protection works are
considered part of an overall plan of improvement, establish the stage-discharge relation for each plan of local
c. Obtain area- and storage-elevation curves for each
reservoir site; select alternative reservoir capacities as
appropriate for each site; select outlet and spillway rating
curves for each reservoir, and develop a plan of
flood-control operation for each reservoir. Determine
maximum regulated flows for each damage center.
d. Estimate the maximum critical duration of runoff
for any of the plans of improvement, considering the
relation of regulated flows at damage centers to
unregulated flood hydrographs of design magnitude at
those damage centers. Prepare frequency curves of
unregulated peak flows and volumes of each of various
representative durations, as described for peak flows in
Chapter 6, for each damage center index location, and for
each reservoir site. If seasonal variation of flood-control
space is to be considered, these curves should be developed
for each season.
e. The two basic approaches for flood-control simulation are complete period-of-record analysis and representative floods analysis. If flooding can occur during any
time of the year, the complete sequential analysis might be
favored. However, if there is a separable flood season, e.g.,
in the western states, then the representative storm
approach may be sufficient. For the storm approach,
develop data for historical floods with storm centerings
throughout the basin and use several proportions of those
floods to obtain flows at the damage centers representing
the full range of the flow-frequency-damage relationship
for base conditions and for regulated conditions. Also,
develop synthetic events that have consistency in volumes
of runoff and peak flows and are reasonably representative
regarding upstream contributions to downstream flows.
f. Perform sequential analysis with the developed
hydrology. The period-of-record simulation provides
simulated regulated flow which can be analyzed directly to
develop flow-frequency relations. The representative flood
approach requires an assumption that the regulated-flow
frequency is the same as the natural-flow frequency.
Frequency curves of regulated conditions at each damage
center can then be derived from frequency curves of
unregulated flows simply by assuming that a given ratio of
the base flood will have the same recurrence frequency
whether it is modified by regulatory structures or not. This
assumption is valid as long as larger unregulated floods
always correspond to the larger regulated flows.
g. Derive a flow-frequency and stage-discharge
curve for the index station at each damage center as
described in Chapters 6 and 8, for unregulated conditions
for each plan of improvement. These can be used for
determining average annual damage for unregulated conditions and for each plan of development and would thus
form the primary basis for project selection.
h. Develop a PMF for each reservoir site, using
procedures described in Chapter 7. These will be used as a
possible basis for spillway design. Route the PMF through
each reservoir, assuming reasonably adverse conditions for
initial storage and available outlet capacity.
EM 1110-2-1420
31 Oct 97
Chapter 11
Conservation Storage
11-1. General Considerations
a. Purposes. Water stored in the conservation pool
can serve many purposes. The primary purposes for
conservation storage are water supply, navigation, low-flow
augmentation, fish and wildlife, and hydroelectric power.
The water requirements for these purposes are discussed in
this chapter along with water quality considerations.
Methods for estimating the conservation storage, or yield,
are presented in Chapter 12.
b. Operational policy. In general, the operational
policy is to conserve available supplies and to release only
when supplemental flow is needed to meet downstream
requirements. Water stored in the conservation pool also
provides benefits within the pool, such as lake recreation
and fish and wildlife habitat.
c. Changing hydrology. When a reservoir is filled,
the hydrology of the inundated area and its immediate
surroundings is changed in a number of respects. The
effects of inflows at the perimeter of the reservoir are
translated rapidly to the reservoir outlet, thus, effectively
speeding the flow of water through the reservoir. Also,
large amounts of energy are stored and must be dissipated
or utilized at the outlet. The reservoir loses water by
evaporation, and this usually exceeds preproject evapotranspiration losses from the lake area. Siltation usually
seals the reservoir bottom, but rising and falling water
levels may alter the pattern of groundwater storage due to
movement into and out of the surrounding reservoir banks.
At high stages, water may seep from the reservoir through
permeable soils into neighboring catchment areas and so be
lost to the area of origin. Finally, sedimentation takes place
in the reservoir and scour occurs downstream.
d. Storage allocation. The joint use of storage for
more than one purpose creates problems of storage allocation for the various purposes. While retained in reservoir
storage, water may provide benefits to recreation, fish,
wildlife, hydropower, and aesthetics. Properly discharged
from the reservoir, similar benefits are achieved downstream. Other benefits that can be derived from the reservoir are those covered in this chapter, including municipal
and industrial water supply, agricultural water supply,
navigation, and low-flow augmentation.
e. Supplemental storage capacity. In most areas,
supplemental storage capacity is required for sediment
deposition; otherwise, the yield capability of the reservoir
may be seriously diminished during the project's economic
life. Sediment storage is determined by estimating the
average annual sediment yield per square mile of drainage
area from observations in the region and multiplying by the
drainage area and the economic life of the project. Trap
efficiency of the reservoir is evaluated and the distribution
of this estimated volume of sediment is determined, using
methods described in EM 1110-2-4000 Sedimentation
Investigations of Rivers and Reservoirs. Sediment surveys
within the reservoir during actual operation will establish
the reliability of these estimates. Storage allocation levels
may then be revised if the sediment surveys show a
significant difference between what was projected and what
was measured.
More complete descriptions of the
techniques used to determine reservoir sedimentation are
presented in EM 1110-2-4000.
f. Minimum pool. A minimum pool at the bottom of
active conservation storage is usually established to
identify the lower limit of normal reservoir drawdown. The
inactive storage below the minimum pool level can be used
for recreation, fish and wildlife, hydropower head, sediment deposition reserve, and other purposes. In rare
instances, it might be used to relieve water supply
g. Reservoir outlets. Reservoir outlets must be
located low enough to withdraw water at desired rates with
the reservoir stage at minimum pool. These outlets can
discharge directly into an aqueduct or into the river. In the
latter case, a diversion dam may be required downstream at
the main canal intake.
h. Computing storage capacity. Because the primary function of reservoirs is to provide storage, their most
important physical characteristic is storage capacity.
Capacities of reservoirs on natural sites must usually be
determined from topographic surveys. The storage capacity can be computed by planimetering the area enclosed
within each elevation contour throughout the full range of
elevations within the reservoir site. The increment of
storage between any two elevation contours is usually
computed by multiplying the average of the areas at the two
elevations by the elevation difference. The summation of
these increments below any elevation is the storage volume
below that level. An alternative to the average-end-area
method is the determination of the storage capacity by the
conic method, which assumes that the volumes are more
nearly represented by portions of a cone. This method is
available in the HEC-1 Flood Hydrograph Package
computer program and is described in the program user's
manual. In the absence of adequate topographic maps,
cross sections of the reservoir area are sometimes
EM 1110-2-1420
31 Oct 97
surveyed, and the capacity is computed from these vertical
cross sections by using the formula for the volume of a
11-2. Water Supply
a. Introduction. Water supply for any purpose is
usually obtained from groundwater or from surface waters.
Groundwater yields and the methods currently in use are
covered in Physical and Chemical Hydrogeology
(Domenico and Schwartz 1990). This discussion is limited
to surface water supplies for low-flow regulation or for
diversion to demand areas.
(1) In some cases, water supply from surface waters
involves only the withdrawal of water as needed from a
nearby stream. However, this source can be unreliable
because streamflows can be highly variable, and the desired
amount might not always be available. An essential
requirement of most water supply projects is that the
supply be available on a dependable basis. Reservoirs play
a major role in fulfilling this requirement. Whatever the
ultimate use of water, the main function of a reservoir is to
stabilize the flow of water, either by regulating a varying
supply in a natural stream or by satisfying a varying
demand by the ultimate consumer. Usually, some overall
loss of water occurs in this process.
(2) In determining the location of a proposed reservoir
to satisfy water needs, a number of factors should be
considered. The dam should be located so that adequate
capacity can be obtained, social and environmental effects
of the project will be satisfactory, sediment deposition in
the reservoir and scour below the dam will be tolerable, the
quality of water in the reservoir will be commensurate with
the ultimate use, and the cost of storing and transporting
the water to the desired location is acceptable. It is
virtually impossible to locate a reservoir site having completely ideal characteristics, and many of these factors will
be competitive. However, these factors can be used as
general guidelines for evaluating prospective reservoir
(3) In the planning and design of reservoirs for water
supply, the basic hydrologic problem is to determine how
much water a specified reservoir capacity will yield. Yield
is the amount of water that can be supplied from the
reservoir to a specified location and in a specified time
pattern. Firm yield is usually defined as the maximum
quantity of water that can be guaranteed with some
specified degree of confidence during a specific critical
period. The critical period is that period in a sequential
record that requires the largest volume from storage to
provide a specified yield. Chapter 12 describes procedures
for yield determination.
b. Municipal and industrial water use. The water
requirement of a modern city is so great that a community
system capable of supplying a sufficient quantity of potable
water is a necessity. The first step in the design of a
waterworks system is a determination of the quantity of
water that will be required, with provision for the estimated
requirements of the future. Next, a reliable source of water
must be located and, finally, a distribution system must be
provided. Water at the source may not be potable, so
water-purification facilities are ordinarily included as an
integral part of the system. Water use varies from city to
city, depending on the population, climatic conditions,
industrialization, and other factors. In a given city, use
varies from season to season and from hour to hour.
Planning of a water supply system requires that the
probable water use and its variations be estimated as
accurately as possible.
(1) Municipal uses of water may be divided into
various classes. Domestic use is water used in homes,
apartment houses, etc., for drinking, bathing, lawn and
garden sprinkling, and sanitary purposes. Commercial and
industrial use is water used by commercial establishments
and industries. Public use is water required in parks, civic
buildings, schools, hospitals, churches, street washing, etc.
Water that leaks from the system, unauthorized
connections, and other unaccounted-for water is classified
as loss and waste.
(2) The average daily use of water for municipal and
industrial purposes is influenced by many factors. More
water is used in warm, dry climates than in humid climates
for bathing, lawn watering, air conditioning, etc. In
extremely cold climates water may be wasted at faucets to
prevent freezing of pipes. Water use is also influenced by
the economic status of the users. The per capita use of
water in slum areas is much less than that in high-cost
residential districts. Manufacturing plants often require
large amounts of water; however, some industries develop
their own water supply and place little or no demand on a
municipal system. The actual amount depends on the
extent of the manufacturing and the type of industry.
Zoning of the city affects the location of industries and may
help in estimating future industrial demands.
(3) About 80 percent of industrial water may be used
for cooling and need not be of high quality, but water used
for process purposes must be of good quality. In some
cases, industrial water must have a lower content of
dissolved salts than can be permitted in drinking water.
EM 1110-2-1420
31 Oct 97
The location of industry is often much influenced by the
availability of water supply. If water costs are high, less
water is used, and industries will often develop their own
supply to obtain cheaper water. In this respect, the installation of water meters in some communities has reduced
water use by as much as 40 percent. The size of the city
being served is a factor affecting water use. Per capita use
tends to be higher in large cities than in small towns. The
difference results from greater industrial use, more parks,
greater commercial use, and, perhaps, more loss and waste
in the larger cities. All of these factors, plus estimated
population projections, should be considered in designing a
waterworks system.
(4) The use of water in a community varies almost
continually. In midwinter the average daily use is usually
about 20 percent lower than the daily average for the year,
while in summer it may be 20 to 30 percent above the daily
average for the year. Seasonal industries such as canneries
may cause wide variation in water demand during the year.
It has been observed that for most communities, the
maximum daily use will be about 180 percent of the
average daily use throughout the year. Within any day,
large variations can be as low as 25 percent to as high as
200 percent of the average for portions of the day. The
daily and hourly variations in water use are not usually
considered in reservoir design, because most communities
use distribution reservoirs (standpipes, etc.) to regulate for
these variations.
c. Agricultural water use. The need for agricultural
water supply is primarily for irrigation. Irrigation can be
defined as the application of water to soil to supplement
deficient rainfall in order to provide moisture for plant
growth. In the United States, about 46 percent of all the
water used is for irrigation. Irrigation is a consumptive use;
that is, most of the water is transpired or evaporated and is
essentially lost to further use.
(1) In planning an irrigation project a number of
factors must be considered. The first step would be to
establish the capability of the land to produce crops that
provide adequate returns on the investment in irrigation
works. This involves determining whether the land is
arable (land which, when properly prepared for agriculture,
will have a sufficient yield to justify its development) and
(2) The amount of water required to raise a crop
depends on the kind of crop and the climate. The plants
that are the most important sources of food and fiber need
relatively large amounts of water. The most important
climatic characteristic governing water need is the length of
the growing season. Other factors that affect water
requirements are the quality of the water, the amount of
land to be irrigated, and, of course, the cost of the water to
the irrigator.
(3) In estimating the amount of storage that will be
required in a reservoir for irrigation, the losses and waste
that occur in the irrigation system must be considered.
Losses and waste are usually divided into conveyance and
irrigation losses and waste. Conveyance losses and waste
are those that occur in the conveyance and distribution
system prior to the application of water to crops. These are
dependent on the design and construction of the system and
also on how the system is operated and maintained.
Irrigation losses and waste are those that occur due to the
slope of the irrigated land, the preparation of the land, soil
condition, the method of irrigation, and the practices of the
(4) Usually, most of the irrigation losses and waste, as
well as a portion of the water applied to the irrigated lands,
return to the stream. If there are requirements for flow
downstream of the reservoir, these return flows can be
important in determining the amount of water that must be
released to meet such requirements.
(5) In most areas, the need for irrigation water is
seasonal and depends on the growing season, the number of
crops per year, and the amount of precipitation. For these
reasons the variation of the demand is often high, ranging
from no water for some months up to 20 to 30 percent of
the annual total for other months. This variation can have a
very large effect on the amount of storage required and the
time of year when it is available.
11-3. Navigation and Low-Flow Augmentation
a. Objective. In designing a reservoir to supply
water for navigation and low-flow augmentation, the
objective is significantly different from objectives for the
other purposes that have been discussed previously in this
chapter. The objective is to supplement flows at one or
more points downstream from the reservoir. For navigation, these flows aid in maintaining the necessary depth of
water and alleviate silting problems in the navigable channel. Low-flow augmentation serves a number of purposes
including recreation, fish and wildlife, ice control, pollution
abatement, and run-of-river power projects. Under certain
conditions, low-flow augmentation provides water for the
other purposes discussed in this chapter. For instance, if
the intake for a municipal and industrial water supply is at
some point downstream of the reservoir, the objective may
be to supplement low flows at that point.
EM 1110-2-1420
31 Oct 97
b. Criteria for navigability. There are no absolute
criteria for navigability and, in the final analysis, economic
criteria control. The physical factors that affect the cost of
waterborne transport are depth of channel, width and
alignment of channel, locking time, current velocity, and
terminal facilities. Commercial inland water transport is,
for the most part, accomplished by barge tows consisting of
1 to 10 barges pushed by a shallow-draft tug. The cost of a
trip between any two terminals is the sum of the fuel costs
and wages, fixed charges, and other operating expenses
depending on the time of transit. Reservoirs aid in
reducing these costs by providing the proper depth of water
in the navigation channel, or by providing a slack-water
pool in lock and dam projects. Storage reservoirs can
rarely be justified economically for navigation purposes
alone and are usually planned as multipurpose projects.
Improving navigation by using reservoirs is possible when
flood flows can be stored for release during low-flow
c. Supplying deficiencies without waste. The ideal
reservoir operation for navigation or low-flow augmentation would provide releases so timed as to supply the
deficiencies in natural flow without waste. This is possible
only if the reservoir is at the head of a relatively short
control reach. As the distance from the reservoir to the
reach is increased, releases must be increased to allow for
uncertainties in estimating intermediate runoff and for
evaporation and seepage enroute to the reach to be served.
Moreover, the releases must be made sufficiently far in
advance of the need to allow for travel time to the reach,
and in sufficient quantity so that after reduction by channel
storage, the delivered flows are adequate. The water
requirement for these releases is considerably greater than
the difference between actual and required flows.
d. Climate. Climate can also affect reservoir operation for low-flow regulation. Depending on the purpose to
be served, the releases may be required only at certain
times of the year or may vary from month to month. For
pollution abatement, the important factors are the quality of
the water to be supplemented, the quality of the water in
the reservoir, and the quality standard to be attained. Also,
the level of the intakes from which releases will be made
can be a very sufficient factor in pollution abatement, since
the quality can vary from one level to another in the
reservoir. Long-term variations can occur due to increased
contamination downstream of a reservoir. This should be
considered in determining the required storage in the
11-4. Fish and Wildlife
a. Added authorized purposes. As shown in Figure 2-1, fish and wildlife and subsequent environmental
purposes have been added as authorized purposes since
1960. Because many of the reservoirs were built prior to
that time, their authorized purposes and regulation plans
may not adequately reflect the more recent environmental
objectives. Therefore, there is an increasing demand and
need for the evaluation of environmental impacts for these
b. Water level fluctuations. The seasonal fluctuation that occurs at many flood control reservoirs and the
daily fluctuations that occur with hydropower operation
often result in the elimination of shoreline vegetation and
subsequent shoreline erosion, water quality degradation,
and loss of habitat for fish and wildlife. Adverse impacts
of water level fluctuations also include loss of shoreline
shelter and physical disruption of spawning and nests.
c. Water level management. Water-level management in fluctuating warm-water and cool-water reservoirs
generally involves raising water levels during the spring to
enhance spawning and the survival of young predators.
Pool levels are lowered during the summer to permit
regrowth of vegetation in the fluctuation zone. Fluctuations may be timed to benefit one or more target species;
therefore, several variations in operation may be desirable.
In the central United States, managers frequently recommend small increases in pool levels during the autumn for
waterfowl management.
d. Fishery management.
Guidelines to meet
downstream fishery management potentials are developed
based on project water quality characteristics and water
control capabilities. To do so, an understanding of the
reservoir water quality regimes is critical for developing the
water control criteria to meet the objectives. For example,
temperature is often one of the major constraints of fishery
management in the downstream reach, and water control
managers must understand the temperature regime in the
pool and downstream temperature requirements, as well as
the capability of the project to achieve the balance required
between the inflows and the releases. Releasing cold water
downstream where fishery management objectives require
warm water will be detrimental to the downstream fishery.
Conversely, releasing warm water creates difficulty in
maintaining a cold-water fishery downstream.
EM 1110-2-1420
31 Oct 97
e. Water temperature management. Water control
activities can also impact water temperatures within the
pool by changing the volume of water available for a
particular layer. In some instances, cold-water reserves
may be necessary to maintain a downstream temperature
objective in the late summer months; therefore, the availability of cold water must be maintained to meet this
objective. For some projects, particularly in the southern
United States, water control objectives include the maintenance of warm-water fisheries in the tailwaters. In other
instances, fishery management objectives may include the
maintenance of a two-story fishery in a reservoir, with a
warm-water fishery in the surface water, and a cold-water
fishery in the bottom waters. Such an objective challenges
water control managers to regulate the project to maintain
the desired temperature stratification while maintaining
sufficient dissolved oxygen in the bottom waters for the
cold-water fishery. Regulation to meet this objective
requires an understanding of operational affects on
seasonal patterns of thermal stratification, and the ability to
anticipate thermal characteristics.
f. Minimum releases. Minimum instantaneous flows
can be beneficial for maintaining gravel beds downstream
for species that require this habitat. However, dramatic
changes in release volumes, such as those that result from
flood-control regulation, as well as hydropower, can be
detrimental to downstream fisheries. Peaking hydropower
operations can result in releases from near zero to very
high magnitudes during operations at full capacity. Maintaining minimum releases and incorporating reregulation
structures are two of the options available to mitigate this
g. Fishing versus peak power. In some instances,
tailwater fishing is at a maximum during summer weekends
and holidays, and this is a time when power generation may
be at a minimum and release near zero. Maintaining
minimum releases during weekend daylight hours may
improve recreational fishing, but may reduce the capability
to meet peak power loads during the week because of lower
water level (head) in the reservoir. In these instances,
water control managers will be challenged to regulate the
project with consideration of these two objectives.
h. Anadromas fish. Regulation for anadromous fish
is particularly important during certain periods of the year.
Generally, upstream migration of adult anadromous fish
begins in the spring of each year and continues through
early fall, and downstream migration of juvenile fish occurs
predominantly during the spring and summer months. The
reduced water velocities through reservoirs, in comparison
with preproject conditions, may greatly lengthen the travel
time for juvenile fish downstream through the impounded
reach. In addition, storage for hydropower reduces the
quantity of spill, and as a result, juvenile fish must pass
through the turbines. The delay in travel time subjects the
juvenile fish to greater exposure to birds and predator fish,
and passage through the powerhouse turbines increases
mortality. To improve juvenile survival, storage has been
made available at some projects to augment river flows,
and flows are diverted away from the turbine intakes and
through tailraces where the fish are collected for
transportation or released back into the river. Barges or
tank trucks can be used to transport juveniles from the
collector dams to release sites below the projects. Other
Corps projects have been modified so the ice and trash
spillways can be operated to provide juvenile fish passage.
i. Wildlife habitat. Project regulation can influence
wildlife habitat and management principally through water
level fluctuations. The beneficial aspects of periodic
drawdowns on wildlife habitat are well documented in
wildlife literature. Drawdowns as a wildlife management
technique can, as examples, allow the natural and artificial
revegetation of shallows for waterfowl, permit the installation and maintenance of artificial nesting structures, allow
the control of vegetation species composition, and ensure
mast tree survival in greentree reservoirs. Wildlife benefits
of periodic flooding include inhibiting the growth of
undesirable and perennial plants, creating access and
foraging opportunities for waterfowl in areas such as
greentree reservoirs, and ensuring certain water levels in
stands of vegetation to encourage waterfowl nesting and
11-5. Hydroelectric Power
General. The feasibility of hydroelectric development is dependent upon the need for electric power, the
availability of a transmission system to take the power from
the point of generation to the points of demand, and the
availability of water from streamflow and storage to produce power in accordance with the capacity and energy
demands in the power market area. Also, the project's
power operations must be coordinated with the operations
for other project purposes to ensure that all purposes are
properly served. Each of these factors must be investigated
to ensure that the project is both feasible and desirable and
to minimize the possibility that unforeseen conflicts will
develop between power and other water uses during the
project life.
(1) The ability of a project to supply power is measured in terms of two parameters: capacity and energy.
Capacity, commonly measured in kilowatts (kw), is the
maximum amount of power that a generating plant can
deliver. Energy, measured in kilowatt-hours (kwh), is the
EM 1110-2-1420
31 Oct 97
amount of actual work done. Both parameters are important, and the operation of a hydroelectric project is sensitive to changes in the demand for either capacity or
(2) Experience has indicated that it is very unlikely
that power demands will remain unchanged during the
project life. Furthermore, the relative priority of various
other water uses can change during the project life, and
there are often legal, institutional, social, or environmental
factors that might affect the future use of water at a particular project. Consequently, the feasibility studies for a
proposed project must not be limited to conditions that are
only representative of the current time or the relatively
near future. Instead, the studies must include considerations of future conditions that might create irreconcilable
conflicts unless appropriate remedial measures are provided for during project formulation.
(3) This section presents general concepts for the
hydrologic analyses associated with the planning, design,
and operation of hydroelectric projects and systems. More
detailed information is provided in EM 1110-2-1701.
Other investigations that influence or affect the hydrologic
studies will be discussed to the extent that their outcome
must be understood by the hydrologic engineer.
b. Types of hydroelectric load. Power developments,
for purposes of this discussion, can be classified with
respect to the type of load served or the type of site
development proposed. The two categories related to the
type of load served are baseload and peaking plants.
(1) Base load. Baseload plants are projects that generate hydroelectric power to meet the baseload demand.
The baseload demand is the demand that exists 100 percent of the time. The baseload can readily be seen in Figure 11-1 as the horizontal dashed line on a typical annual
load duration curve. This curve displays the percent of
time during a given year that a given capacity demand is
equaled or exceeded. The area under this curve represents
the total energy required to meet the load during the year.
Usually, the baseload demand is met by thermal
generating facilities. However, in cases where there is a
relatively abundant supply of water that is available with a
high degree of reliability and where fuel is relatively
scarce, hydroelectric projects may be developed to meet
the baseload demands. These projects would then operate
at or near full capacity 24 hr per day for long periods of
time. This type of development is not feasible where there
is a large seasonal variation in streamflow unless the
baseflow is relatively high or unless there is a provision
for a large volume of power storage in the project.
Figure 11-1. Typical annual load duration curve
(2) Peaking load. Peaking plants are projects that
generate hydroelectric power to supplement baseload
generation during periods of peak power demands. The
peak power demands are the loads that exist primarily
during the daylight hours. The time of occurrence and
magnitude of peak power demands are shown on a load
curve in Figure 11-2. This curve shows the time variation
in power demands for a typical week. Depending upon the
quantity of water available and the demand, a peaking
plant may generate from as much as 18 hr a day to as
little as no generation at all, but it is usually 8 hr a day or
less. Peaking plants must supply sufficient capacity to
satisfy the peak capacity demands of a system and
sufficient energy to make the capacity usable on the load.
This means that energy or water should be sufficient to
supply peaking support for as long and as often
as the capacity is needed. In general, a peaking hydroelectric plant is desirable in a system that has thermal
generation facilities to meet the baseload demands. The
hydroelectric generating facilities are p a r t i c u l a r ly
a d a p t a b l e to the
peaking operation because
EM 1110-2-1420
31 Oct 97
Figure 11-2. Weekly load curve for a large electric system
their loading can be changed rapidly. Also, the factors
that make seasonal variations in streamflow a major
problem in baseload operation are usually quite easily
overcome in a peaking plant if some storage can be
c. Project types. There are three major classifications of hydroelectric projects: storage, run-of-river, and
pumped storage. There are also combinations of projects
that might be considered as separate classifications, but for
purposes of discussing hydrologic analysis it is necessary
to define only these three types.
(1) Storage projects. Storage plants are projects that
usually have heads in the medium to high range (™ 25 m)
and have provisions for storing relatively large volumes of
water during periods of high streamflow in order to provide water for power generation during periods of deficient streamflow. Considerable storage capacity may be
required because the period of deficient flow is quite
frequently more than a year long and, in some instances,
may be several years long. Because use of the stored
water entails drawdown of the power storage, it is desirable that other water uses associated with the development
of a storage plant permit frequent and severe drawdowns
during dry periods. Peaking operation, which is quite
frequently associated with storage projects requires large
and sometimes rapid fluctuations in releases of water
through the generating units. It is often necessary to
provide facilities to re-regulate the power releases if fluctuations of water levels below the project are not tolerable.
Because storage projects are conducive to multipurpose
use and because the power output from a storage plant is a
function of the guaranteed output during a multi-year dry
period, it is usually necessary to make detailed routing
studies to determine the storage requirements, installed
capacity, firm energy, and an operating plan.
(2) Run-of-river projects. Run-of-river plants have
little or no power storage and, therefore, must generate
power from streamflow as it occurs. The projects generally have productive heads in the low to medium range (530 m) and are quite frequently associated with navigational developments or other multipurpose developments
with limitations on reservoir drawdowns. Because of the
absence or near-absence of storage in run-of-river projects,
there is usually very little operational flexibility in these
projects, and it is necessary that all water uses be
compatible. The existence of one or more storage projects
in the upstream portion of a river basin may make a
run-of-river project in the lower portion of the basin feasible where it would not otherwise be feasible. In this
situation, the storage projects provide a regulated outflow
EM 1110-2-1420
31 Oct 97
that is predictable and usable, while the natural streamflow
might be neither.
(a) Run-of-river projects may have provisions for a
small amount of storage, often called pondage. This
pondage detains the streamflow during off-peak periods in
daily or weekly cycles for use in generating power during
peak demand periods. If the cycle of peaking operation is a
single day, the pondage requirements are based on the flow
volume needed to sustain generation at or near installed
capacity for 12 hr. If more storage capacity is available
and large fluctuations in the reservoir surface are
permissible, a weekly cycle of peaking operation may be
considered. Because industrial and commercial consumption of power is significantly lower on weekends than on
week days, an “off-peak” period is created from Friday
evening until Monday morning. If generation from the
hydroelectric peaking plants is not required during this
period, water can be stored in the pondage for use during
the 5-day peak-load period.
(b) Because of the relatively low heads associated with
run-of-river projects, the tailwater fluctuations are usually
quite important, particularly in peaking operations. Also,
flood flows may curtail power generation due to high
tailwater. While flow-duration analysis can be used to
estimate average annual energy production, sequential
analysis may be required for more detailed analysis of
extreme conditions.
(3) Pumped-storage projects. Pumped-storage plants
are projects that depend on pumped water as a partial or
total source of water for generating electric energy. This
type of project derives its usefulness from the fact that the
demand for power is generally low at night and on weekends; therefore, pumping energy at a very low cost will be
available from idle thermal generating facilities or run-ofriver projects. If there is a need for peaking capacity and if
the value of peaking power generation sufficiently exceeds
the cost of pumping energy (at least a ratio of 1.5 to 1.0
because roughly 3 kwh of pumping energy are necessary to
deliver enough water to provide 2 kwh of energy
generation), pumped storage might be feasible. There are
three types of pumped-storage development: diversion,
off-channel, and in-channel, which are detailed in Chapter
7 of EM 1110-2-1701.
(a) In general, pumped storage projects consist of a
high-level forebay where pumped water is stored until it is
needed for generation and a low-level afterbay where the
power releases are regulated, if necessary, and from which
the water is pumped. The pumping and generating are
done by generating units composed of reversible pump
turbines and generator motors located along a tunnel or
penstock connecting the forebay and afterbay. The water is
pumped from the afterbay to the forebay when the normal
power demand is low and least expensive and released
from the forebay to the afterbay to generate power when
the demand is high and most costly. The feasibility of
pumped-storage developments is dependent upon the need
for relatively large amounts of peaking capacity, the
availability of pumping energy at a guaranteed favorable
cost, and a load with an off-peak period long enough to
permit the required amount of pumping.
(b) A unique feature of pumped-storage systems is
that very little water is required for their operation. Once
the headwater and tailwater pools have been filled, only
enough water is needed to take care of evaporation and
seepage. For heads up to 300 m, reversible pump turbines
have been devised to operate at relatively high efficiency as
either a pump or turbine. The same electrical unit serves as
a generator and motor by reversing poles. Such a machine
may reduce the cost of a pumped-storage project by
eliminating the extra pumping equipment and pump house.
The reversible pump turbine is a compromise in design
between a Francis turbine and a centrifugal pump. Its
function is reversed by changing the direction of rotation.
d. Need for hydroelectric power. The need for
power is established by a power market study or survey.
The feasibility of a particular hydroelectric project or
system is determined by considering the needs as established by the survey, availability of transmission facilities,
and the economics of the proposed project or projects.
Although forecasts of potential power requirements within
a region to be served by a project are not hydrologic
determinations, they are essential to the development of
plans for power facilities and to the determination of
project feasibility and justification. The power market
survey is a means of evaluating the present and potential
market for electrical power in a region.
(1) The survey must provide a realistic estimate of the
power requirements to be met by the project and must
show the anticipated rate of load growth from initial operation of the project to the end of its economic life. The
survey also provides information regarding the characteristics of the anticipated demands for power. These
characteristics, which must be considered in hydrologic
evaluations of hydroelectric potential, include the seasonal
variation of energy requirements (preferably on a monthly
basis), the seasonal variation of capacity requirements (also
preferably on a monthly basis), and the range of usable
plant factors for hydroelectric projects under both adverse
and average or normal flow conditions.
EM 1110-2-1420
31 Oct 97
(2) The results of a power market survey might be
furnished to the hydrologic engineer in the form of load
duration or load curves (Figures 11-1 and 11-2) showing
the projected load growth, the portion of the load that can
be supplied by existing generating facilities, and the portion
that must be supplied by future additions to the generating
system. From these curves, the characteristics of planned
hydroelectric generating facilities can be determined.
Because these data are developed from the needs alone
without consideration of the potential for supplying these
needs, the next step is to study the potential for
hydroelectric development, given the constraints established in the study of needs.
kW =
In order to convert a project's power output to energy,
Equation 11-1 must be integrated over time:
kWh =
Q H e dt
11.81 P t t
kW = power available from the project, kW
e. Estimation of hydroelectric power potential.
Traditionally, hydroelectric power potential has been
determined on the basis of the critical hydro-period as
indicated by the historical record. The critical hydroperiod is defined as the period when the limitations of
hydroelectric power supply due to hydrologic conditions
are most critical with respect to power demands. Thus, the
critical period is a function of the power demand, the
streamflow, and the available storage. In preliminary
project planning, the estimates of power potential are often
based on a number of simplifying assumptions because of
the lack of specific information for use in more detailed
analyses. Although these estimates and the assumptions
upon which they are based are satisfactory for preliminary
investigations, they are not suitable for every level of
engineering work. Many factors affecting the design and
operation of a project are ignored in these computations.
Therefore, detailed sequential analyses of at least the
critical hydro-period should be initiated as early as
possible, usually when detailed hydrologic data and some
approximate physical data concerning the proposed project
become available. Because of the availability of computer
programs for accomplishing these sequential routings, they
can be done rapidly and at a relatively low cost.
(1) The manner in which the streamflow at a given site
is used to generate power depends upon the storage
available at the site, the hydraulic and electrical capacities
of the plant, streamflow requirements downstream from the
plant, and characteristics of the load to be served. In
theory, the hydroelectric power potential at a particular
site, based on repetition of historical runoff, can be estimated by identifying the critical hydro-period and obtaining
estimates of the average head and average streamflow
during this critical period. The data can then be used in the
equation below to calculate the power available from the
kWh = energy generated during a time period, kWh
Q = average streamflow during the time period,
H = average head during the time period, m (Head =
headwater elevation - tailwater elevation hydraulic losses)
t = number of hours in the time period
e = overall efficiency expressed as the product of
the generator efficiency and the turbine
In practice, the summation of energy production over the
critical period is performed with a sufficiently small time
step to provide reasonable estimates of head and, therefore,
energy. Two basic approaches are available: flow-duration
and sequential analysis.
(2) For run-of-river projects, where the headwater
elevation does not vary significantly, the flow-duration
approach can be used to estimate average annual energy
production. The duration curve can be truncated at the
minimum flow rate for power production. The curve can
also be truncated for high-flows if the tailwater elevation is
too high for generation. The remaining curve is converted
to capacity-duration and integrated to obtain average
annual energy. Hydropower Analysis Using Flow-Duration
Procedures HYDUR (HEC 1982d) was developed to
perform energy computations based on flow-duration data.
EM 1110-2-1701 describes HYDUR in paragraph C-2b and
the flow-duration method in Section 5-7.
EM 1110-2-1420
31 Oct 97
(3) Sequential streamflow analysis will be applied to
most reservoir studies. The procedure allows detailed
computations of the major parameters affecting hydropower (e.g., headwater and tailwater elevation, efficiency,
and flow release). By performing the analysis in sufficiently small time steps, an accurate simulation of the
reservoir operation, power capacity and energy production
can be obtained. Chapter 5 of EM 1110-2-1701, Sections 5-8 through 5-10, provides a discussion of sequential
routing studies. Appendix C provides information concerning computer programs that are available for use in
these studies.
f. Hydropower effect on other project purposes.
Usually, power generation must have a high priority relative to other conservation uses. Consequently, thorough
investigations of all aspects of the power operation must be
conducted to ensure that the power operations do not create
intolerable situations for other authorized or approved
water uses. Likewise, the power operations must be
coordinated with other high priority purposes such as flood
control and municipal water supply to ensure that the
planned power operation will not interfere with the
operations for these purposes. The operation rules that are
necessary to effect the coordination are usually developed
and tested using engineering judgment and detailed
sequential routing studies. However, it is necessary to
define the interactions between power and other project
purposes before initiating operation studies.
(1) Power generation is generally compatible with most
purposes that require releases of water from a reservoir for
downstream needs. However, power generation usually
competes with purposes that require withdrawal of the
water directly from the reservoir or that restrict fluctuations
in the reservoir level.
Flood-control requirements
frequently conflict with power operations because
flood-control needs may dictate that storage space in a
reservoir be evacuated at a time when it would be beneficial to store water for use in meeting future power
demands. Furthermore, when extensive flooding is anticipated downstream from a reservoir project, it may be
necessary to curtail power releases to accomplish
flood-control objectives. It is often possible to pass part or
all of the flood-control releases through the generating
units, thereby reducing the number of additional outlets
needed and significantly increasing the energy production
over what would be possible if the flood-control releases
were made through conduits or over the spillway. Also,
many of the smaller floods can be completely regulated
within the power drawdown storage, an operation that is
beneficial to power because it provides water for power
generation that might otherwise have been spilled. This
joint use can reduce the exclusive flood-control storage
requirements and also reduce the frequency of use of
flood-control facilities.
(2) Water for municipal, industrial, or agricultural use
can be passed through the generating units with no harmful
effects if the point of withdrawal for the other use is below
the point where the power discharge enters the river.
Re-regulation may be required for hydropower peaking
operations to “smooth out” the power releases. Conflicts
between power and these consumptive uses more likely
occur when the withdrawal for other uses is directly from
the reservoir. When the withdrawal is from the reservoir of
a storage project, the inclusion of power as a project
purpose may require that special attention be given to
intake facilities for the other purposes because of the
relatively large drawdown associated with storage projects.
(3) Low-flow augmentation for navigation, recreation,
or fish and wildlife can be accomplished by releases
through power generating units. In the case of baseload
projects, the power release is ideally suited for this type of
use. With peaking projects, however, a re-regulation
structure may be necessary to provide the relatively
uniform releases that might be required for navigation or
for in-stream recreation. Release of water for quality
enhancement can sometimes be accomplished through the
generating units. Although the intakes for the turbines are
usually located at a relatively low elevation in the reservoir
where dissolved oxygen content might be low, the
oxygenation that occurs in the tailrace and in the stream
below the project may produce water with an acceptable
dissolved oxygen content. The water released from the
lower levels of the reservoir is normally at a relatively low
temperature and, thus, ideal for support or enhancement of
a cold-water fishery downstream. If warm waters are
needed for in-stream recreation, for fishery requirements,
or for any other purpose, a special multilevel intake may be
required to obtain water of the desired temperature.
(4) Recreation values at a reservoir project may be
enhanced, somewhat, by the inclusion of power because a
much larger reservoir is frequently required, and that may
increase opportunities for extensive recreational activities.
Unfortunately, however, the large drawdowns associated
with the big storage projects create special problems with
respect to the location of permanent recreational facilities
and may create mudflats that are undesirable from the
standpoint of aesthetics and public health requirements.
The drawdown may also expose boaters, swimmers, and
other users to hazardous underwater obstacles unless
provisions are made to remove these obstacles to a point
well below the maximum anticipated drawdown.
Obviously the time of occurrence of extreme drawdown
EM 1110-2-1420
31 Oct 97
conditions is an important factor in determining the degree
of conflict with recreation activities.
11-6. Water Quality Considerations
a. Water quality definition. Water quality deals with
the kinds and amounts of matter dissolved and suspended
in natural water, the physical characteristics of the water,
and the ecological relationships between aquatic organisms
and their environment. It is a term used to describe the
chemical, physical, and biological characteristics of water
in respect to its suitability for a particular purpose. The
same water may be of good quality for one purpose or use,
and bad for another, depending on its characteristics and
the requirements for the particular use.
b. Water quality parameters. In general, physical
parameters define the water quality characteristics that
affect our senses while chemical and biological parameters
index the chemical and biological constituents present in
the water resource system. However, these are not
independent, but are actually highly related. For example,
chemical waste discharges may affect such physical factors
as density and color, may alter chemical parameters such
an pH and alkalinity, and may affect the biological
community in the water. Even so, the physical, chemical,
and biological subdivision is a useful way to discuss water
quality conditions. The following sections describe the
water quality parameters frequently associated with
reservoirs. EM 1110-2-1201 provides details on parameters, assessment techniques, plus data collection and
11-7. Water Quality Requirements
A wide variety of demands are made for the use of water
resources. Water of a quality that is unsatisfactory for one
use may be perfectly acceptable for another. The level of
acceptable quality is often governed by the scarcity of the
resource or the availability of water of better quality.
a. Domestic use. The use of water for domestic
purposes such as drinking, culinary use, and bathing is
generally considered to be the most essential use of our
water resources. The regulations for the quality of this
water are likewise higher than for most (but not all) other
beneficial uses of water. In early times, the quality of the
water supply source and the quality at the delivery point
were synonymous; but the general degradation of both
surface water quality and shallow ground water quality has
made it necessary, in most cases, for some degree of water
treatment to be used to produce acceptable water for
domestic use. In recent decades, there has been a strong
trend (which is likely to continue) for the quality of the
source waters throughout the world to decline as a result of
increased urbanization and industrialization and as a result
of changes in agricultural practices. At the same time,
populations are coming to expect a higher standard of
health and well-being; and as a result, the regulations for
acceptable domestic water continue to rise and enlarge the
role of water treatment.
b. Drinking water standards. Drinking water standards for the world as a whole have been set by the World
Health Organization (WHO). One should keep in mind that
these standards do not describe an ideal or necessarily
desirable water, but are merely the maximum values of
contaminant concentration which may be permitted. It is
highly desirable to have water of much better quality. In
the United States, the Environmental Protection Agency
(EPA) sets regulations that legally apply to public drinking
water and water supply systems. The regulations are
divided into three categories: bacterial, physical, and
chemical characteristics. They are defined in terms of
maximum contaminant levels (MCL's). Bacterial quality is
defined by establishing the sampling sequence, the method
of analysis, and the interpretation of test results for the
coliform organisms which serve as presumptive evidence of
bacterial contamination from intestinal sources. Analysis is
generally made for total coliform, fecal coliform, and
streptococci coliform. The limits on biological and
physical parameters, and on chemical elements or
compounds in water are documented in Water Supply and
Sewerage (McGhee 1991).
c. Quality of source waters. The drinking water
standards are the end product of a production line which
begins with the source water as a raw material and proceeds through the various unit processes of water treatment
and finally water distribution. The quality of source waters
for other uses such as agricultural and industrial water
supply, fish and other aquatic life, and recreation are set by
state regulations of receiving waters. Other specific uses
may include regulations for navigation, wild and scenic
rivers, and other state-specific needs. The state regulations
are subject to EPA approval. The regulations of a given
state may take a variety of forms but are often specified by
stream reaches including associated natural or constructed
impoundments. Each reach may be classified for its
various water uses and water quality standards defined for
each use.
(1) Industry uses water as a buoyant transporting
medium, cleansing agent, coolant, and as a source of steam
for heating and power production. Often the quality
required for these purposes is significantly higher than that
required for human consumption. The availability of water
of high quality is often an important parameter in site
EM 1110-2-1420
31 Oct 97
selection by an industry. The needs of a particular industry
as to both quantity and quality of water varies with the
competition for water, the efficiency of the plant process
with regard to water utilization, the recycling of water, the
location of the plant site and the ratio of the cost of the
water to the cost of the product. For economic reasons and
for reasons of quality control and operation responsibility,
industries with high water requirements usually develop
their own supply and treatment facilities.
(2) Farmstead water is that water used by the human
farm population for drinking, food preparation, bathing,
and laundry. It also includes water used for the washing
and hydrocooling of fruits and vegetables, and water used
in the production of milk. The quality of water desired for
farmstead use is generally that required for public water
supplies. It is not feasible to set rigid quality standards for
irrigation waters because of such varied and complex
factors as soil porosity, soil chemistry, climatic conditions,
the ratio of rainwater to irrigation water, artificial and
natural drainage, relative tolerance of different plants, and
interferences between and among constituents in the water.
Examples of the latter are the antagonist influence of
calcium-sodium, boron-nitrates, and selenium-sulfates.
cyanides, oil, solids, turbidity, or insecticides are known to
(5) For an aquatic system to be acceptable for swimming and bathing, it must be aesthetically enjoyable (i.e.,
free from obnoxious floating or suspended substances,
objectionable color and foul odors), it must contain no
substances that are toxic upon ingestion or irritating to the
skin or sense organs, and it must be reasonably free of
pathogenic bacteria. Standards generally do not cover the
first two terms as related to swimming and bathing except
in qualitative terms. In the United States, numerous standards exist for bacteriological quality based on the coliform
count in the water. Generally, the standards range
downward from 1,000 coliform bacteria per 100 ml to as
low as 50 per 100 ml. Such standards are not based on
demonstrated transmittal of disease, but have been established because in these ranges the standards are
economically reasonable and no problems appear. The use
of an aquatic system for boating and aesthetic enjoyment is
generally not so demanding as the requirements for swimming or the propagation of fish and aquatic life although
these three water uses are usually closely linked.
11-8. Reservoir Water Quality Management
(3) Water quality parameters of importance for
irrigation are sodium, alkalinity, acidity, chlorides, bicarbonates, pesticides, temperature, suspended solids, radionuclids, and biodegradable organics. All of these factors
need to be weighed carefully in evaluating the suitability of
water for irrigating a particular crop. There is surprisingly
little data on the effect of water quality on livestock, but
generally they thrive best on water meeting human drinking
water standards. The intake of highly mineralized water by
animals can cause physiological disturbances of varying
degrees of severity. In some cases, particular ions such as
nitrates, fluorides, selenium salts, and molybdenum may be
harmful. Certain algae and protozoa have also been proven
toxic to livestock.
(4) The basic purpose of water quality criteria for
aquatic life is to restore or maintain environmental conditions that are essential to the survival, growth, reproduction, and general well-being of the important aquatic
organisms. These criteria are ordinarily determined without the aid of economic considerations. Generally, a
number of major problems arise in establishing water
quality criteria for an aquatic community because of the
inability to quantify the effects of the pertinent parameters
and reduce them to a conceptual model that describes the
nature of the biological community which will develop
under a given set of conditions. But extensive research
should be considered when unusually high concentrations
of such parameters as alkalinity, acidity, heavy metals,
Reservoirs may serve several purposes in the management
of water quality. If used properly, substantial benefits can
be achieved. On the other hand, unwise use of reservoirs
may cause increased quality degradation. Benefits may
accrue as a result of detention mixing or selective withdrawal of water in a reservoir or the blending of waters
from several reservoirs. The effects of improper management are often far-reaching and long-term. They may
range from minor to catastrophic, and may be as obvious as
a fish kill or subtle and unnoticed. It is essential that all
water control management activity and especially real-time
actions include valid water quality evaluation as a part of
the daily water control decision process. It must be
understood that water quality benefits accumulate slowly,
build on each other, and can become quite substantial over
time. This is in contrast to the sudden benefits that come
from a successful flood-control operation. Water quality
management requirements, objectives, and standards are
presented in EM 1110-2-3600.
a. Reservoirs in streams. The presence of a reservoir in a stream affects the quality of the outflow as compared to the inflow by virtue of the storage and mixing
which takes place in the reservoir. The effect of such an
impoundment may be easily evaluated for conservative
parameters if the waters of the reservoir are sufficiently
mixed that an assumption of complete mixing within an
EM 1110-2-1420
31 Oct 97
analysis time period does not lead to appreciable error.
However, this assumption is limited to relatively small,
shallow reservoirs.
b. Reservoir outflow and inflow. The simplest technique requires the assumption that the reservoir outflow
during a given time period is of constant quality and equal
to the quality of the reservoir storage at the end of the
computation time period. It is then assumed that the inflow
for the time period occurs independently of the outflow,
and reservoir quality is determined by a quality mass
balance at the end of the time period. This approach is
equivalent to the mass balance of water in reservoir
c. Reservoir water quality. Simple mass balance
procedures may be applicable is some situations; however,
usually more comprehensive methods should be
considered. Chapter 4, “Water Quality Assessment Techniques,” in EM 1110-2-1201 describes various techniques
available for assessing reservoir water quality conditions.
There is a hierarchy of available techniques that reflects
increasing requirements of time, cost, and technical expertise. The increasing efforts should provide accompanying
increases in the degree of understanding and resolution of
the problem and causes. This hierarchy includes screening
diagnostic and predictive techniques, which are described.
d. Reservoirs as detention basins. Reservoir mixing
is a continual process where low inflows of poor quality are
stored and mixed with higher inflows of better quality.
Generally, this is accomplished in large reservoirs where
annual or even multiple-year flows are retained, but the
concept extends to small reservoirs in which weekly or
even daily quality changes occur due to variability of
loading associated with the inflow.
(1) The use of a reservoir as a mixing devise should be
considered whenever the inorganic water quality is
unacceptable during some periods but where the average
quality falls within the acceptance level. Lake Texoma on
the Red River is an example of a reservoir which modifies
the quality pattern. Although monthly inflow quality has
equalled 1,950 mg/l chloride concentration, the outflow has
not exceeded 520 mg/l.
(2) Many materials which enter a reservoir are
removed by settling. This applies not only to incoming
settleable solids, but also to colloidal and dissolved
materials which become of settleable size by chemical
precipitation or by synthesis into biological organisms.
Reservoirs are often used to prevent such settleable material from entering navigable rivers where settleable
materials would interfere with desired uses. However,
reservoirs that receive substantial sediment will have a
short useful life. Planning should include evaluation of the
ultimate fate and possible replacement of such reservoirs.
Reservoir sedimentation is covered in EM 1110-2-4000.
e. Reservoirs as stratified systems. Reservoirs
become stratified if density variations caused by temperature or dissolved solids are sufficiently pronounced to
prevent complete mixing. This stratification may be
helpful or harmful depending on the outlet works, inflow
water quality, and the operating procedure of the reservoir.
(1) Temperature stratification can be beneficial for
cold-water fisheries if the water which enters the reservoir
during the cooler months can also be stored and released
during the warmer months. The cooler water released
during the warm months can also be valuable as a cooling
water source, can provide for higher oxygen transfer
(re-aeration) or slower organic waste oxidation (deoxygenation), and can make the water more aesthetically acceptable for water supply and recreational purposes.
(2) Dissolved oxygen stratification usually occurs in
density stratified lakes, particularly during the warmer
months. The phenomenon occurs because oxygen which
has been introduced into the epilimnion by surface
re-aeration does not transfer through the metalimnion into
the hypolimnion at a rate high enough to satisfy the oxygen
demand by dissolved and suspended materials and by the
benthal organisms. Thus, the cool bottom waters which are
sometimes desirable may be undesirable from a dissolved
oxygen standpoint unless energy dissipation structures are
constructed to transfer substantial oxygen into the reservoir
pool or the reservoir discharge. Mechanical reservoir
mixing to equalize temperature and transfer oxygen to
lower reservoir levels is one possible tool for managing
reservoir water quality.
f. Reservoirs as flow management devices. Reservoirs may improve water quality by merely permitting the
management of flow. This management may include
maintenance of minimum flows, blending selective releases
from one or more reservoirs to maintain a given stream
quality, and the exclusion of a flow from a system by
(1) Minimum flow is often maintained in a stream for
navigation, recreation, fish and wildlife, and water rights
purposes. Such flows may also aid in maintaining acceptable water quality.
(2) There is general agreement that water may be
stored and selectively released to help reduce natural water
quality problems where source control is not possible, and
EM 1110-2-1420
31 Oct 97
also that water should not be stored and released solely to
improve water quality where similar improvement may be
achieved by treatment at the source. The use of a water
resource to dilute treatable waste materials is regarded as
the misuse of a valuable resource in most cases.
(3) Selective release of water from one or more reservoirs may help improve quality at one or more downstream
locations. Such releases may be one of the governing
factors in establishing reservoir management rules. The
water to be released may either be good quality water that
will improve the river quality or poor quality water that is
to be discharged when it will do a minimum of harm (e.g.,
during high flow). The water quality version of the HEC-5
reservoir simulation program (HEC-5Q) is designed to
perform quality analysis based on a reservoir simulation for
quantity demands and subsequently determine additional
releases to meet quality objectives (HEC 1986). Section III, Chapter 4 of EM 1110-2-1201 describes various
predictive techniques, including numerical and physical
EM 1110-2-1420
31 Oct 97
Chapter 12
Conservation Storage Yield
12-1. Introduction
a. Purpose. There are three purposes of this chapter:
(1) to provide a descriptive summary of the technical
procedures used in the hydrologic studies to analyze reservoirs for conservation purposes; (2) to furnish background
information concerning the data requirements, advantages,
and limitations of the various procedures; and (3) to
establish guidelines which will be helpful in selecting a
procedure, conducting the studies, and evaluating results.
b. Procedures. The procedures presented are generally used to determine the relationship between reservoir
storage capacity and reservoir yield (supply) for a single
reservoir. The procedures may be used to determine
storage requirements for water supply, water quality control, hydroelectric power, navigation, irrigation, and other
conservation purposes. Although the discussions are
limited to single reservoir analysis, many of the principles
are generally applicable to multi-reservoir systems. Chapter 4, “Reservoir Systems,” presents concepts regarding the
analysis of a multi-reservoir system.
12-2. Problem Description
(5) Perform the analysis, evaluate the results, and
present the information.
b. Evaluating hydrologic aspects of planning.
Many of the methods described in other chapters of this
manual are necessary to develop and provide data to
evaluate the hydrologic aspects of reservoir planning,
design, and operation. In many cases, the methods
required to provide data for a reservoir analysis are more
complex than the method for the reservoir study itself.
However, because the usefulness and validity of the reservoir analysis are directly dependent upon the accuracy and
soundness of basic data, complex methods can often be
justified to develop the data.
c. General information.
This chapter contains
information on types of procedures, considerations of time
interval, storage allocation, project purposes, several types
of studies, and a summary of methods to analyze the results
of reservoir studies. The methods can be characterized as
simplified, including sequential and nonsequential analysis,
and detailed sequential analysis. Emphasis is given to the
sequential routing because:
(1) It is adaptable to study single or multiple reservoir
(2) With an appropriate time interval, the variations in
supply and demand can be directly analyzed.
a. Determining storage yield relationships. The
determination of storage-yield relationships for a reservoir
project is one of the basic hydrologic analyses for reservoirs. The basic objective can be to determine the reservoir
yield given a storage allocation, or find the storage required
to obtain a desired yield. The determination follows a
traditional engineering approach:
(3) It gives results that are easily understood and
explained by engineers familiar with basic hydrologic
(1) Determine the study objectives. This includes the
project purposes, operation goals, and the evaluation
(5) It can be used with sparse basic data for preliminary analyses, as well as with detailed data and analyses.
(4) The accuracy and results of the study can be
closely controlled by the engineers performing and supervising the studies.
12-3. Study Objectives
(2) Determine the physical and hydrologic constraints
for the site. This includes the reservoir storage and outflow
capability, as well as the downstream channel system.
(3) Compile the basic data. The basic data include
demands, flow, and losses. Also, the appropriate time
interval for analysis, which depends on the data and its
variations in time.
(4) Select the appropriate method, one that meets the
study objectives and provides reliable information to
evaluate results based on accepted criteria.
a. Establish and consider objectives. Before any
meaningful storage-yield analysis can be made, it is necessary to establish and consider the objectives of the
hydrologic study. The objectives could range from a
preliminary study to a detailed analysis for coordinating
reservoir operation for several purposes. The objectives,
together with the available data, will control the degree of
accuracy required for the study.
b. Determination of storage required. Basically,
there are two ways of viewing the storage-yield
EM 1110-2-1420
31 Oct 97
relationship. The most common viewpoint involves the
determination of the storage required at a given site to
supply a given yield. This type of problem is usually
encountered in the planning and early design phases of a
water resources development study.
c. Determination of yield. The second viewpoint
requires the determination of yield from a given amount of
storage. This often occurs in the final design phases or in
re-evaluation of an existing project for a more comprehensive analysis. Because a higher degree of accuracy is
desirable in such studies, detailed sequential routings are
usually used.
d. Other objectives.
Other objectives of a
storage-yield analysis include the following: determination
of complementary or competitive aspects of multiple project development, determination of complementary or
competitive aspects of multiple purpose development in a
single project, and analysis of alternative operation rules
for a project or group of projects. Each objective and the
basis for evaluation dictates implicitly the method which
should be used in the analysis.
12-4. Types of Procedures
a. Selecting. The procedures used to determine the
storage-yield relationship for a potential dam site may be
divided into either simplified or detailed sequential analysis. The selection of the appropriate technical procedure
may be governed by the availability of data, study objectives, or budgetary considerations. In general, the simplified techniques are only satisfactory when the study
objectives are limited to preliminary or screening studies.
Detailed methods are usually required when the study
objectives advance to the feasibility and design phases.
b. Simulation and mathematical programming analyses. The detailed sequential methods may be further
subdivided into simulation analyses and mathematical
programming analyses. In simulation analysis, the physical
system is simulated by performing a sequential reservoir
routing with specified demands and supply. In this type of
study, attempts are made to accurately reproduce the
temporal and spatial variation in streamflow and reservoir
storage in a reservoir-river system. This is accomplished
by accounting for as many significant accretions and
depletions as possible. In mathematical programming
analysis, the objective is to develop a mathematical model
which can be used to analyze the physical system without
necessarily reproducing detailed factors. The model
usually provides a simulation that will provide a maximum
(or minimum) value of the objective function, subject to
system constraints.
c. Simplified method. A simplified method can be
used if demands for water are relatively simple (constant)
or if approximate results are sufficient, as in the case of
many preliminary studies. However, it should be emphasized that the objective of the simplified methods is to
obtain a good estimate of the results which could be
achieved by detailed sequential analysis. Simplified
methods consist generally of mass curve and depth duration
analyses, which are discussed later.
d. Computer models.
Computer models have
changed the role of the simplified methods because of the
relatively low cost of a detailed sequential routing. Computer programs like HEC-5 Simulation of Flood Control
and Conservation Systems (HEC 1982c) provide efficient
models of reservoirs, based on the level of data availability.
For preliminary studies, minimum reservoir and demand
information are sufficient. The critical and more complex
problem is the development of a consistent flow sequence,
which is required by all methods of analysis. Simplified
methods still have a role in screening studies or as tools to
obtain good estimates of input data for the sequential
e. Detailed sequential routing. In the past, detailed
sequential routings have been used almost exclusively for
the development of operating plans for existing reservoirs
and reservoir systems. However, the advent of the comprehensive basin planning concept, the growing demand for
more efficient utilization of water resources, and the
increasing competition for water among various project
purposes indicate a need for detailed sequential routings in
planning studies. Also, these complex system studies
provide an opportunity to use optimization to suggest
system operations or allocations (ETL 1110-2-336).
f. Mathematical programming.
programming (optimization) has generally been applied in
water management studies of existing systems. The questions addressed usually deal with obtaining maximum gain
from available resources, e.g., energy production from a
hydropower system. Recent Corps applications include the
review of operation plans for reservoir systems, e.g.,
Columbia and Missouri River Systems (HEC 1991d 1991f,
and 1992a). These studies utilized the HEC Prescriptive
Reservoir Model HEC-PRM (HEC 1991a).
g. Further information. Wurbs (1991) provides a
review of modeling and analysis approaches including
simulation models, yield analysis, stochastic streamflow
models, impacts of basinwide management on yield, and
optimization techniques.
EM 1110-2-1420
31 Oct 97
12-5. Factors Affecting Selection
a. Examining objectives and data availability.
Before initiating a storage-yield study, the study objectives
and data availability should be examined in order to
ascertain: (1) the method best suited for the study
requirements; (2) the degree of accuracy required to produce results consistent with the study objectives; and
(3) the basic data required to obtain the desired accuracy
using the selected method. In preliminary studies, limitations in time and scope might dictate the data and method
to be used and the accuracy. More detailed analyses are
needed when a higher degree of accuracy is desired. A
technical study work plan is very useful in organizing study
objectives, inventory of available data, and the selection of
general procedures.
b. Availability of data. The availability of basic
physical and hydrologic data will quite frequently be a
controlling factor in determining which of the several
technical methods can be used. Obviously, the detailed
methods require more data which may not be available.
However, detailed simulation can be performed with
limited system data if the historic flow data are available.
The simplified methods require less data, but the reliability
of the results decreases rapidly as the length of hydrologic
record decreases. Therefore, it is often desirable to
simulate additional hydrologic data for use with simplified
methods. Hydrologic data and data simulation are discussed in Chapter 5.
c. Significant aspects. The study level and available
data are not the only deciding factors. The study methods
must capture the significant aspects of the prototype system. If simplified methods do not utilize data which have
major influences upon the results, it would be necessary to
utilize a more detailed method which accounts for
variations in these data. For example, the National
Hydropower Study (USACE 1979) used flow-duration
techniques for most reservoirs, but used sequential routing
for reservoirs with significant storage.
12-6. Time Interval
a. Selection. The selection of an appropriate time
interval depends primarily on the type of analysis and the
significance of the data variation over time. Time intervals
of one month are usually adequate for nonsequential and
preliminary sequential analyses. For more detailed studies,
shorter routing intervals will ordinarily be required.
Average daily flow data are increasingly used because they
are readily available, and computer speed is sufficient to
process the data in a reasonable time. Only in exceptional
cases will routing intervals of less than one day be required
for conservation studies. Considerable work is involved
with shorter intervals, and the effects of time translation,
which are usually ignored in conservation routing studies,
become important with shorter intervals. Shorter intervals
are necessary and should be used during flood periods or
during periods when daily power fluctuations occur.
b. Using short time intervals. Ordinarily, when
using short time intervals of one day or less, it is necessary
to obtain adequate definition of the conditions under study.
Periods selected for analysis should exhibit critical
combinations of hydrologic conditions and demand
characteristics. For example, analysis of hourly power
generation at a hydroelectric plant under peaking conditions might be studied for a one-month period where
extremely low flows could be assumed to coincide with
extremely high power demands. As a rule, studies involving short time intervals are supplementary to one or more
studies of longer periods using longer time intervals. The
results of the long period study are often used to establish
initial conditions such as initial reservoir storage for the
selected periods of short-interval analysis.
c. Selecting a routing interval. In sequential conservation routing studies, the selection of a routing interval
is dependent upon four major factors: (1) the demand
schedule that will be utilized in determining the yield;
(2) the accuracy required by the study objectives; (3) the
data available for use in the study; and (4) the phase
relationship between periods of high and low demands and
high and low flow. If the water demand schedule is
relatively uniform, it is ordinarily possible to estimate the
amount of storage required for a specified yield by graphical analysis using the Rippl diagram or the nonsequential analysis discussed later herein. Demand schedules
which show marked seasonal variations usually preclude
the use of graphical techniques alone in determining storage requirements. This is especially true when the demand
is a function which cannot be described in terms of a
specific amount of water, as in the cases of hydroelectric
power and water quality. In order to obtain accurate
estimates of storage requirements when the demand
schedule is variable, it is necessary to make sequential
routing studies with routing intervals short enough to
delineate important variations in the demand schedule.
Simplified techniques may be utilized in obtaining a first
estimate of storage requirements for the detailed sequential
d. More accurate results. As a general rule, shorter
routing intervals will provide more accurate results. This is
due to many factors, such as better definition of relationships between inflow and releases, and better estimates
of average reservoir levels for evaporation and power-head
EM 1110-2-1420
31 Oct 97
calculations. Average flow for longer routing intervals
tends to reduce the characteristic variations of streamflows,
thus producing a “dampened” storage requirement. This
will tend to overestimate yield, or underestimate required
storage for reservoirs with small storage. Therefore, the
volume of conservation storage, compared to the average
flow in a time period, is an additional consideration in
selecting the time interval. For example, if a monthly
interval is used and there is no sufficient conservation
storage to control the variation of flow within the month,
the use of average monthly will conceal that fact.
e. Effects on storage requirements. When fluctuations in streamflows or demands have a significant effect
on storage requirements, computations should be refined
for critical portions of the studies, or shorter routing intervals should be used. However, the routing interval should
not be shorter than the shortest period for which flow and
demand data are available. Attempts to “manufacture”
flow or demand data are usually time consuming and may
create errors or give a false impression of accuracy unless
reliable information is available for subdivision of basic
f. Nonsequential methods. The selection of the flow
interval for analysis by nonsequential methods is usually
not as critical as for a sequential analysis. Because the
nonsequential analysis is restricted to uniform demands, it
does not produce results as accurate as those obtained by
sequential methods. Therefore, there is very little gain in
accuracy with short intervals. Flow intervals of one month
are usually suitable for nonsequential methods.
12-7. Physical Constraints
Physical constraints which should be considered in
storage-yield studies include conservation storage available, minimum pool, outlet capacities, and channel capacities. The addition of hydropower as a purpose will require
the inclusion of constraints to power generation, e.g.,
maximum and minimum head, penstock capacity, and
power capacity. If flood control is to be included as a
project purpose, the maximum conservation storage
feasible at a given site will be affected by the flood-control
12-8. Priorities
In order to determine optimum yield in a multiple-purpose
project, some type of priority system for the various purposes must be established. This is necessary when the
competitive aspects of water use require a firm basis for an
operating decision. Safety of downstream inhabitants and
cities are of utmost importance, which makes flood
reduction the highest priority in a multiple-purpose project
during actual operation. During periods of flood operations, conservation requirements might be reduced in order
to provide the best flood operation. Although this chapter
is not concerned directly with flood-control operation or
criteria, it is necessary to integrate flood-control constraints
with the conservation study to ensure that operating
conditions and reservoir levels for conservation purposes
do not interfere with flood-control operation. Priorities
among the various conservation purposes vary with locale,
water rights, and with the need for various types of water
utilization. In multipurpose projects, every effort should be
made to develop operation criteria which maximize the
complementary uses for the various conservation purposes.
12-9. Storage Limitations
One of the reasons for making sequential conservation
routing studies is to determine the effect of storage limitations on yield rates. Simplified yield methods cannot
account for operational restrictions imposed by storage
limitations in a multiple-purpose project. As shown in
Figure 2-2, three primary storage zones, any or all of which
may exist in a given reservoir project, may generally be
described as follows:
Exclusive capacity, generally for flood control, in
the uppermost storage space in the reservoir.
Multiple-purpose capacity, typically conservation
storage, immediately below the flood control
Inactive capacity, or dead storage, the lowest
storage space in the reservoir.
An additional space, called surcharge, exists between the
top of the flood-control space and the top of the dam.
Surcharge storage is required to pass flood waters over the
spillway. The boundaries between the storage zones and
operational boundaries within the zones may be fixed
throughout the year, or they may vary from season to
season as shown on Figure 12-1. The varying boundaries
usually offer a more flexible operational plan which may
result in higher yields for all purposes, although an additional element of chance is often introduced when the
boundaries are allowed to vary. The purpose of detailed
sequential routing studies is to produce an operating
scheme and boundary arrangement which minimizes the
chance of failure to satisfy any project purpose while
optimizing the yield for each purpose. The three storage
zones and the effect of varying their boundaries are discussed in the following paragraphs.
EM 1110-2-1420
31 Oct 97
Figure 12-1. Seasonally varying storage boundaries
a. Flood-control storage. The inclusion of floodcontrol storage in a multiple-purpose project may
adversely affect conservation purposes in two ways. First,
storage space which might otherwise be utilized for conservation purposes is reserved for flood-control usage.
Second, flood-control operations may conflict with conservation goals, with a resultant reduction or loss of
conservation benefits. However, detailed planning and
analysis of criteria for flood-control and conservation
operations can minimize such adverse effects. Even without dedicated flood storage, conservation projects must be
able to perform during flood events.
(1) Where competition between flood-control and
conservation requirements exists, but does not coincide in
time, the use of a seasonally varying boundary between
flood-control storage and conservation storage may be
used. The general procedure is to hold the top of the
conservation pool at a low level when conservation
demands are not critical in order to reserve more storage
space for flood-control regulation. Then, as the likelihood
of flood occurrence decreases, the top of the conservation
pool is raised to increase the storage available for conservation purposes. Water management criteria are then
tested by detailed sequential routing for the period of
recorded streamflow. Several alternative patterns and
magnitudes of seasonal variations should be evaluated to
determine the response of the storage-yield relationship
and the flood-control efficiency to the seasonal variation
of the boundary. A properly designed seasonally-varying
storage boundary should not reduce the effectiveness of
flood-control storage to increase the conservation yield.
EM 1110-2-1420
31 Oct 97
(2) Flood-control operation is generally simplified in
conservation studies because the routing interval for such
studies is frequently too long to adequately define the
flood-control operation. Nevertheless, flood-control constraints should be observed insofar as possible. For
example, channel capacities below the reservoir are considered for release purposes, and storage above the top of
flood-control pool is not utilized.
b. Conservation storage. The conservation storage
may be used to regulate minor floods in some multipurpose
projects, as well as to supply water for conservation
purposes. In addition to seasonal variations in its upper
boundary between flood control and conservation, the
lower conservation storage boundary may also vary seasonally. If several conservation purposes of different
priorities exist, there may be need for a buffer zone in the
conservation storage. The seasonal variation in the boundary between conservation storage and buffer zone would
be determined by the relationship between seasonal
demands for the various purposes.
c. Buffer storage. Buffer storage may be required for
one or two reasons. First, it may be used in multipurpose
projects to continue releases for a higher priority purpose
when normal conservation storage has been exhausted by
supplying water for both high and low priority purposes.
Second, it may be used in a single-purpose project to
continue releases at a reduced rate after normal
conservation storage has been exhausted by supplying
water at a higher rate. In either case, the boundary between
the normal conservation storage and buffer storage is used
to change the operational criteria. The location of this
boundary and its seasonal variation are important factors in
detailed sequential routing because of this change in water
management criteria. The amount of buffer storage and,
consequently, the location of the seasonally varying
boundary between the buffer zone and the remainder of the
conservation storage is usually determined by successive
approximations in sequential routing studies. However, a
simplified procedure, which produces a satisfactory
estimate in cases without seasonally varying boundaries, is
described in Section 12-11.
d. Inactive or dead storage. The inactive storage
zone is maintained in the reservoir for several purposes,
such as a reserve for sedimentation or for fish and wildlife
habitat. As a rule, the reservoir may not be drawn below
the top of the inactive storage.
Although it may be
possible to vary the top of the reserve pool as shown in
Figure 12-1, it is seldom practical to do so. This could
reflect the desire to maintain a higher recreation pool
during the summer.
12-10. Effects of Conservation and Other
As previously indicated, the seasonal variation of demand
schedules may assume an important role in the determination of required yield. The effect of seasonal variation is
most pronounced when the varying demand is large with
respect to other demands, as is often the case when
hydroelectric power or irrigation is a large demand item.
The quantity of yield from a specified storage may be
overestimated by as much as 30 percent when a uniform
yield rate is used in lieu of a known variable yield rate.
Also, variable demand schedules often complicate the
analysis of reservoir yield to the extent that it is impossible
to accurately estimate the maximum yield or the optimum
operation by approximate methods. Because detailed
sequential routing is particularly adaptable to the use of
variable demand schedules, every effort should be made to
incorporate all known demand data into the criteria for
routing. Thus, successive trials using detailed sequential
analysis must often be used to determine maximum yield.
Computer programs such as HEC-5 provide yield
determination for reservoirs by performing the successive
sequential routing until a firm yield is determined. Firm
yield is the amount of water available for a specific use on
a dependable basis during the life of a project. The project
purposes which often require analysis of seasonal
variations in demand are discussed in more detail in the
following subparagraphs.
a. Low-flow regulation. The operation of a reservoir
for low flow regulation at a downstream control point is
difficult to evaluate without a detailed sequential routing,
because the operation is highly dependent upon the flows
which occur between the reservoir and the control point,
called intervening local flow. As these flows can vary
significantly, a yield based on long period average intervening flows can be subject to considerable error. A
detailed sequential routing, in which allowance is made for
variations of intervening flows within the routing interval,
produces a more reliable estimate of storage requirements
for a specified yield and reduces the chance of
overestimating a firm yield. Ordinarily, the yield and the
corresponding operation of a reservoir for low-flow
regulation are determined by detailed sequential routing of
the critical period and several other periods of low flow.
The entire period of recorded streamflow may not be
required, unless summary-type information is needed for
functions such as power.
b. Diversion and return flows. The analysis of yield
for diversions is complicated by the fact that diversion
requirements may vary from year to year as well as from
EM 1110-2-1420
31 Oct 97
season to season. Furthermore, the diversion requirements
may be stated as a function of the natural flow and water
rights rather than as a fixed amount. In addition, diversion
amounts may often be reduced or eliminated when storage
in the reservoir reaches a certain critical low value. When
any one of these three items is important to a given
reservoir analysis, detailed sequential analysis for the entire
period of flow record should be made to determine
accurately the yield and the water management criteria.
Coordination of the water management criteria for other
purposes with the diversion requirements may also be
achieved with the detailed sequential analysis results.
c. Water quality control. Inclusion of water quality
control and management as a project purpose almost
always dictates that sequential routing studies be used to
evaluate project performance. Practically every variable
under consideration in a water quality study will vary
seasonally. Following are the variables which must be
considered in a water quality study: (1) variation in quality
of the inflow, (2) subsequent change in quality of the
reservoir waters due to inflow quality and evaporation,
(3) variation in quality of natural streamflow entering the
stream between the reservoir and the control station,
(4) variation in effluent from treatment plants and storm
drainage outflow between the reservoir and the control
station, and (5) variation in quality requirements at the
control station. Accurate evaluation of project performance must consider all of these variations which pertain
to water quality control. Additionally, there are several
quality parameters which may require study, and each
parameter introduces additional variations which should be
evaluated. For example, if temperature is an important
parameter, the level of the reservoir from which water is
released should be considered in addition to the above
variables. Likewise, if oxygen content is important, the
effects of release through power units versus release
through conduits must be evaluated.
Hydroelectric power generation.
(1) If hydroelectric power is included as a project
purpose, detailed sequential routings are necessary to
develop water management criteria, to coordinate power
production with other project purposes, and to determine
the project's power potential. As a rule, simplified methods
are usable for power projects only for preliminary or
screening studies, reservoirs with very little power storage,
or when energy is a by-product to other operations. Flowduration analysis, described in Chapter 11, is typically
applied in these situations. Power production is a function
of both head and flow, which requires a detailed sequential
study when the conservation storage is relatively large and
the head can be expected to fluctuate significantly.
(2) Determination of firm power or firm energy is
usually based on sequential routings over the critical
period. The critical period must consider the combination
of power demand and critical hydrologic conditions.
Various operational plans are used in an attempt to maximize power output while meeting necessary commitments
for other project purposes. When the optimum output is
achieved, a water management guide curve can be developed. The curve is based on the power output itself and on
the plan of operation followed to obtain the maximum
output. Critical period analysis and curve development are
described in Section 12-11. Additional sequential routings
for the entire period of flow record are then made using the
rule curve developed in the critical period studies. These
routings are used to coordinate power production with
flood-control operation and to determine the average
annual potential energy available from the project.
(3) In areas where hydroelectric power is used primarily for peaking purposes, it is important that storage
requirements be defined as accurately as possible because
the available head during a period of peak demand is
required to determine the peaking capability of the project.
An error in storage requirements, on the other hand, can
adversely affect the head with a resultant loss of peaking
(4) Tailwater elevations are also of considerable
importance in power studies because of the effect of head
on power output. Several factors which may adversely
affect the tailwater elevation at a reservoir are construction
of a reregulation reservoir below the project under
consideration, high pool elevations at a project immediately
downstream from the project under consideration, and
backwater effects from another stream if the project is near
the confluence of two streams. If any of these conditions
exist, the resultant tailwater conditions should be carefully
evaluated. For projects in which peaking operation is
anticipated, an assumed “block-loading” tailwater should
be used to determine reservoir releases for the sequential
reservoir routing. The “block-loading” tailwater elevation
is defined as the tailwater elevation resulting from
sustained generation at or near the plant's rated capacity
which represents the condition under which the project is
expected to operate most of the time. Although in reality
the peaking operation tailwater would fluctuate
considerably, the use of block-loading tailwater elevation
ensures a conservative estimate of storage requirements
and available head.
(5) Reversible pump-turbines have enhanced the
feasibility of the pumped-storage type of hydroelectric
development. Pumped storage includes reversible pumpturbines in the powerhouse along with conventional
EM 1110-2-1420
31 Oct 97
generating units, and an afterbay is constructed below the
main dam to retain water for pumping during nongenerating periods. Sequential routing studies are required for
analyses of this type because of the need to define storage
requirements in the afterbay, pumping requirements and
characteristics, and the extent to which plan should be
developed. Many of the existing and proposed pumpedstorage projects in the United States, however, are single
purpose projects which do not have conventional units and
often utilize off-channel forebays.
a. Sequential mass curve. The most commonly
used simplified sequential method is the sequential mass
curve analysis, sometimes referred to as the Rippl Method.
This method produces a graphical estimate of the storage
required to produce a given yield, assuming that the
seasonal variations in demand are not significant enough to
prohibit the use of a uniform draft (demand) rate. The
sequential mass curve is constructed by accumulating
inflows to a reservoir site throughout the period of record
and plotting these accumulated inflows versus the sequential
time period as illustrated in Figure 12-2.
12-11. Simplified Methods
If demands for water are relatively constant or if approximate results are sufficient, as in the case of many preliminary studies, a simplified method can be used to save time
and effort. The use of simplified techniques which do not
consider sequential variations in streamflow or demand are
generally limited to screening studies or developing first
estimates of storage or yield. The following procedures
will generally produce satisfactory results and continue to
have a role in storage-yield determinations.
The desired yield rate, in this example
38,000 m3/year, is represented by the slope of a straight line.
Straight lines are then constructed parallel to the desired
yield rate and tangent to the mass curve at each low point
(line ABC) and at the preceding high point that gives the
highest tangent (line DEF). The vertical distance between
these two lines (line BE) represents the storage required to
provide the desired yield during the time period between the
two tangent points (points D and B). The maximum vertical
Figure 12-2. Storage determination using a sequential mass curve
EM 1110-2-1420
31 Oct 97
storage to meet the desired yield, during the given flow
(2) The critical period is the duration of time from
point D, when conservation storage drawdown begins, to
point F, when the reservoir conservation storage fills. The
critical drawdown is from point D to point B, while during
the time from B to F the reservoir would be refilling.
(3) The sequential mass curve method does not indicate the relative frequency of a shortage. However, by
using nonsequential methods, a curve of yield versus
shortage frequency can be determined.
b. Nonsequential mass curve. Several nonsequential
methods can be used to develop a relation for storage yield
versus shortage frequency.
The application of this
procedure is limited, however, to water supply demands
that are uniform in time. These methods involve the
development of probability relations for varying durations
of streamflow. The historical flows, supplemented by
simulated flows where needed, are used to determine
frequency tables of independent low-flow events for several durations. A series of low-flow events for a particular
duration is selected by computing and arranging in order of
magnitude, the independent minimum-flow rates for that
duration for the period of record.
(1) After the frequency tables of independent low-flow
events are computed for various durations, low-flow
frequency curves are obtained by plotting the average flow
on log-probability paper. Chapter 4 of EM 1110-2-1415
describes the procedure and presents an example table and
frequency plot.
(2) Care must be exercised in the interpretation of the
low-flow curves because the abscissa is “nonexceedance
frequency per 100-years,” or the number of events within
100-years that have a flow equal to or less than the indicated flow. Thus, when low-flow durations in excess of
one year are evaluated, the terminology cannot be used
interchangeably with probability. For instance, during a
100-year period, the maximum number of independent
events of 120 months (10-years) duration is 10. Therefore,
the 120-month duration curve cannot cross the value of 10
on the “nonexceedance frequency per 100-year” scale.
(3) Minimum runoff-duration curves for various
frequencies, as shown in Figure 12-3 are obtained by
plotting points from the low-flow frequency curves on
logarithmic paper. The flow rates are converted to volumes
(millions of cubic meters in this example). The logarithmic
scales simply permit more accurate interpolation between
durations represented by the frequency curves.
(4) The nonsequential mass curve (Figure 12-4) is
developed by selecting the desired volume-duration curve
from Figure 12-3 and plotting this curve on arithmetic grid.
The desired yield is then used to determine the storage
requirement for the reservoir. The storage requirement is
determined by drawing a straight line, with slope equivalent
to the required gross yield, and by plotting this line tangent
to the mass curve. The absolute value of the negative
vertical intercept represents the storage requirement. The
application of this procedure is severely limited everywhere
in the case of seasonal variations in runoff and yield
requirements because the nonsequential mass curve does
not reflect the seasonal variation in streamflows, and the
tangent line does not reflect seasonal variations in demand.
However, the method does provide an estimate of yield
c. Evaporation losses. Another disadvantage of
these simplified types of storage-yield analysis is the
inability to evaluate evaporation losses accurately. This
may not be critical in humid areas where net evaporation
(lake evaporation minus pre-project evapotranspiration) is
relatively small, but it can cause large errors in studies for
arid regions. Also, these procedures do not permit consideration of seasonal variations in requirements, system
nonlinearities, conflicting and complementing service
requirements, and several other factors.
12-12. Detailed Sequential Analysis
a. Sequential analysis. Sequential analysis is currently the most accepted method of determining reservoir
storage requirements. Many simplified methods have given
way to the more detailed computer simulation approaches.
In many instances, the computer solution provides more
accurate answers at a lower cost than the simple hand
b. Accounting for reservoir water. Sequential analysis applies the principle of conservation of mass to account
for the water in a reservoir. The fundamental relationship
used in the routing can be defined by:
I O = S
I = total inflow during the time period, in volume
O = total outflow during the time period, in volume
S = change in storage during the time period, in
volume units
EM 1110-2-1420
31 Oct 97
Figure 12-3. Minimum runoff-duration curves
The inflow and outflow terms include all types of inflow
and outflow. The inflows should include natural streamflow, releases from upstream reservoirs, local inflow to the
reservoir, precipitation falling on the reservoir surface
(sometimes included in computation of net evaporation),
and diversions into the reservoir. Outflows consist of
reservoir releases plus evaporation losses, leakage, and
diversions out of the reservoir. Sequential routing provided the framework for accounting for all water in the
system. The application can be as detailed as required.
The development of the required data constitutes the major
effort in most sequential routing studies.
c. Sequential routing. Sequential routing uses a
repetitive solution of Equation 12-1 in the form of:
St = St&1 % It & Ot
St = storage at the end of time t, volume units
St-1 = storage at the end of time t-1, volume units
Figure 12-4. Nonsequential mass curve from
Figure 12-3
EM 1110-2-1420
31 Oct 97
It = average inflow during time step t, converted to
volume units
Ot = average outflow during time step t, converted
to volume units
The primary input includes reservoir storage capacity and
allocation, requirements, losses, flow at all model locations
for the simulation period, and system connectivity and
constraints. The primary output for the reservoir is the
average reservoir release for each time step and the
resulting reservoir storage at the end of each time step.
The releases are made to meet specified requirements,
subject to all specified constraints such as storage allocation and maximum release capability.
accounting of flows adds reservoir releases and subtracts
diversions and losses to the local downstream flows to
compute regulated flow at desired locations. If a short time
interval is used, the flow travel time must be considered.
Channel routing is usually done with hydrologic routing
d. Multipurpose reservoir routings. The HEC-5
Simulation of Flood Control and Conservation Systems
(HEC 1982c) computer program performs multipurpose
reservoir routings for reservoir systems providing for
services at the reservoirs and downstream control points.
Releases from a reservoir are determined by the specified
requirements for project purposes. Reservoir releases may
be controlled at the dam site by hydroelectric power
requirements, downstream control for flow, diversion,
water rights, or quality. Additional diversions may be
made directly from the reservoir.
(1) The program operates to meet the downstream flow
requirement, considering available supplies and supplement
flow from the intervening area. The storage allocation and
most demands can be defined as constant, monthly varying,
or seasonally varying.
Historic simulations can be
performed with period-by-period demands for low flow and
(2) Computer program HEC-5 has a firm yield routine
called optimization of conservation storage in the program
user's manual. The routine can either determine the
required storage to meet a specified demand or the
maximum reservoir yield that can be obtained from a
specified amount of storage. While designed for a single
reservoir, it can use up to six reservoirs in a single run,
provided the reservoirs operate independently. Optimization can be accomplished on monthly firm energy requirements, minimum monthly flow, monthly diversions, or all
of the conservation purposes. The routine can estimate the
critical period and make a firm yield estimate based on that
period. After the firm yield is estimated, the program will
perform a period of record simulation to ensure that the
firm yield can be met. Several cycles of critical period and
period of record simulations can be performed in one
computer run, based on user input specifications.
12-13. Effects of Water Deficiencies
a. Water storages. Absolute guarantees of water
yield are usually not practical, and the designer should
therefore provide estimates of shortages that could
reasonably develop in supplying the demands with available storage. If nonsequential procedures have been used,
information on future shortages is limited to the probability
or frequency of occurrence, and the duration or severity of
shortages will not be known. In using the Rippl Method,
the computations are based on just meeting the demand;
therefore, no shortages are allowed during the period of
analysis. The result gives no information on the shortages
that might be expected in the future. Only in the detailed
sequential analysis procedure is adequate information on
expected future shortages obtainable.
b. Amount and duration of water shortage. The
amount and duration of shortage that can be tolerated in
serving various project purposes can greatly influence the
amount of storage required to produce a firm yield. These
tolerances vary a great deal for different project purposes
and should be analyzed carefully in reservoir design. Also,
changes in reservoir operation should be considered to
meet needs during drought (HEC 1990a).
c. Intolerable shortages. Shortages are generally
considered to be intolerable for purposes such as drinking
water. However, some reduction in the quantity of municipal and industrial water required can be tolerated without
serious economic effects by reducing some of the less
important uses of water such as lawn watering, car washing, etc. Shortages greater than 10 percent may cause
serious hardship. Most designs of reservoir storage for
municipal and industrial water supply are based on supplying the firm yield during the most critical drought of
record. Typically, drought contingency plans are developed to meet essential demands during drought conditions
that may be more severe than the historic critical period.
ETL 1110-2-335 provides guidance for developing and
updating plans.
d. Irrigation shortages. For irrigation water supply,
shortages are usually acceptable under some conditions.
Often the desired quantity can be reduced considerably
during the less critical parts of the growing season without
great crop loss. Also, if there is a reliable forecast of a
drought, the irrigator may be able to switch to a crop
EM 1110-2-1420
31 Oct 97
having less water requirements or use groundwater to make
up the deficit. Shortages of 10 percent usually have
negligible economic effect, whereas shortages as large as
50 percent are usually disastrous.
e. Water supply for navigation and low flow augmentation. In designing a reservoir to supply water for
navigation or low-flow augmentation, the amount and
duration of shortages are usually much more important
than the frequency of the shortages. Small shortages might
only require rescheduling of fully-loaded vessels, whereas,
large shortages might stop traffic altogether. The same
thing is true for such purposes as fish and wildlife where
one large shortage during the spawning season, for
example, could have serious economic effects for years to
f. Effects of shortages. Each project purpose should
be analyzed carefully to determine what the effects of
shortages will be. In many cases, this will be the criterion
that determined the ultimate amount of reservoir storage
needed for water supply and low-flow regulation.
12-14. Shortage Index
a. Definition.
A general approach to shortage
definition is to use a shortage index. The shortage index is
equal to the sum of the squares of the annual shortages
over a 100-year period when each annual shortage is
expressed as a ratio to the annual requirements, as shown
100 (
SI =
SI = shortage index
N = number of years in routing study
SA = annual shortage (annual demand volume minus
annual volume supplied)
DA = annual demand volume
This shortage index reflects the observation that economic
and social effects of shortages are roughly proportional to
the square of the degree of shortage. For example, a
shortage of 40 percent is assumed to be four times as
severe as a shortage of 20 percent. Similarly, as illustrated
in Table 12-1, shortages of 50 percent during 4 out of
Table 12-1
Illustration of Shortage Index
No. of Annual Shortages per 100 Years
Annual Shortage
(SA/DA) In %
100 years are assumed four times as severe as shortages of
10 percent during 25 out of 100 years.
The shortage index has considerable merit over shortage
frequency alone as a measure of severity because shortage
frequency considers neither magnitude nor duration. The
shortage index can be multiplied by a constant to obtain a
rough estimate of associated damages.
b. Additional criteria needed. There is a definite
need for additional criteria delineating shortage acceptability for various services under different conditions. These
criteria should be based on social and economic costs of
shortages in each individual project study, or certain standards could be established for the various services and
conditions. Such criteria should account for degree of
shortage as well as expected frequency of shortage.
12-15. General Study Procedures
a. Water supply. After alternative plans for one or
more water supply reservoirs have been established, the
following steps can be followed in performing hydrologic
studies required for each plan:
(1) Obtain all available daily and monthly streamflow
records that can be used to estimate historical flows at each
reservoir and diversion or control point. Compute monthly
flows and adjust as necessary for future conditions at each
pertinent location. A review of hydrologic data is
presented in Chapter 5.
(2) Obtain area-elevation data on each reservoir site to
be studied and compute storage capacity curves.
Determine maximum practical reservoir stage from physiographic and cultural limitations.
(3) Estimate monthly evapotranspiration losses from
each site and monthly lake evaporation that is likely to
occur if the reservoir is built.
(4) Determine seasonal patterns of demands and total
annual requirements for all project purposes, if applicable,
EM 1110-2-1420
31 Oct 97
as a function of future time. Synthesize stochastic variations in demands, if significant.
(5) Establish a tentative plan of operation, considering
flood control and reservoir sedimentation as well as conservation requirements, and perform an operation study
based on runoff during the critical period of record. The
HEC-5 Simulation of Flood Control and Conservation
Systems computer program can be used for this purpose.
(6) Revise the plan of operation, including sizes of
various facilities, as necessary to improve accomplishments
and perform a new operation study. Repeat this process
until a near-optimum plan of development is obtained.
(7) Depending on the degree of refinement justified in
the particular study, test this plan of development using the
entire period of estimated historical inflows and as many
sequences of synthetic streamflows and demands as might
be appropriate. Methods for developing synthetic flow
sequences are presented in Chapter 12 of the Hydrologic
Frequency manual (EM 1110-2-1415).
(8) Modify the plan of development to balance yields
and shortages for the maximum overall accomplishment of
all project objectives.
b. Hydroelectric power. The study procedure for
planning, designing, and operating hydroelectric developments can be summarized as follows:
(1) From an assessment of the need for power generation facilities, obtain information concerning the feasibility and utility of various types of hydroelectric projects.
This assessment could be made as part of the overall study
for a given project or system, or it could be available from
a national, regional, or local power authority.
(2) From a review of the physical characteristics of a
proposed site and a review of other project purposes, if
any, develop an estimate of the approximate amount of
space that will be available for either sole- or joint-use
power storage. This determination and the needs developed under step (1) will determine whether the project will
be a storage, run-of-river, or pumped-storage power project
and whether it will be operated to supply demands for
peaking or for baseload generation.
(3) Using information concerning seasonal variation in
power demands obtained from the assessment of needs,
and knowing the type of project and the approximate
storage usable for power production, determine the historical critical hydro-period by review of the historical
hydrologic data.
(4) An estimate of potential hydroelectric energy for
the assumed critical hydro-period is made using Equation 11-2. If the energy calculated from this equation is for
a period other than the basic marketing contract period
(usually a calendar year), the potential energy during the
critical hydro-period should be converted to a firm or
minimum quantity for the contract period (minimum annual
or annual firm in the case of a calendar year).
(5) Because the ability of a project to produce
hydroelectric energy and peaking capacity is a complex
function of the head, the streamflow, the storage, and
operation for all other purposes, the energy estimate
obtained in step (4) is only an approximation. Although
this approximation is useful for planning purposes it should
be verified by simulating the operation of the project for all
authorized purposes by means of a sequential routing
study. Chapter 11 provides methods for performing and
analyzing sequential routing studies.
(6) From the results of detailed sequential routing
studies, the data necessary for designing power-generating
units and power-related facilities of the project should be
developed. The design head and design output of the
generating units, approximate powerhouse dimensions,
approximate sizes of water passages, and other physical
dimensions of the project depend on the power installation.
(7) Operation rules for the project must be developed
before construction is completed. These rules are developed and verified through sequential routing studies that
incorporate all of the factors known to affect the project's
operation. For many multipurpose projects, these operation
rules are relatively complex and require the use of
computerized simulation models to facilitate the computations involved in the sequential routing studies.
(8) If the project is to be incorporated into an existing
system or if the project is part of a planned system, system
operation rules must be developed to define the role of the
project in supplying energy and water to satisfy the system
demands. These rules are also developed and tested using
sequential routing studies. Sequential routing studies for
planning or operating hydroelectric power systems are best
accomplished using a computer program such as HEC-5.
EM 1110-2-1420
31 Oct 97
Chapter 13
Reservoir Sedimentation
13-1. Introduction
AThe ultimate destiny of all reservoirs is to be filled with
sediment,@ (Linsley et al. 1992). The question is how long
will it take? Also, as the sediment accumulates with time,
will it adversely affect water control goals?
13-2. Reservoir Deposition
a. Total available sediment. The first step is to
estimate the total sediment that will be available for deposition during the design life of the project. Required data
include design life of the reservoir, reservoir capacity,
water and sediment yield from the watershed, the composition of the sediment material, and the unit weight of
sediment deposits. With this information, the trap efficiency can be determined.
a. Transport capacity. A reservoir changes the
hydraulics of flow by forcing the energy gradient to
approach zero. This results in a loss of transport capacity
with the resulting deposition. The smaller the particles,
the farther they will move into the reservoir before depositing. Some may even pass the dam. Deep reservoirs are
not fully mixed and are conducive to the formation of
density currents.
b. Trap efficiency. Trap efficiency is the percent of
inflowing sediment that remains in the reservoir. Some
proportion of the inflowing sediment leaves the reservoir
through the outlet works. The proportion remaining in the
reservoir is typically estimated based on the trap efficiency. Trap efficiency is described in Section 3-7(a) of
EM 1110-2-4000, and the calculations are described in an
appendix therein. The efficiency is primarily dependent
on the detention time, with the deposition increasing as the
time in storage increases.
b. Sediment deposits. The obvious consequence of
sediment deposits is a depletion in reservoir storage
capacity. Figure 13-1 illustrates components of sediment
deposition in a deep reservoir. The volume of sediment
material in the delta and the main reservoir depends on the
inflowing water and sediment, reservoir geometry, project
operation and life among other things. The delta will
continue to develop, with time, and the reservoir will
eventually fill with sediment.
c. Existing reservoirs.
Existing reservoirs are
routinely surveyed to determine sediment deposition, and
resulting loss of storage. Section 5-30 and Appendix K of
EM 1110-2-4000 describe the Corps program. This
historic deposition data can be useful for checking computed estimates. ASediment Deposition in U.S. Reservoirs
(Summary of Data Reported 1981-85)@ provides
Figure 13-1. Conceptual deposition in deep reservoirs
EM 1110-2-1420
31 Oct 97
data on reservoir locations, drainage areas, survey dates,
reservoir storage capacities, ratios of reservoir capacities to
average annual inflows, specific weights (dry) of sediment
deposits, and average annual sediment-accumulation rates
(U.S. Geological Survey 1992). Reservoirs are grouped by
drainage basins.
13-3. Distribution of Sediment Deposits in the
The planning or design of a reservoir requires an analysis
to determine how sediment deposits will be distributed in
the reservoir. This is a difficult aspect of reservoir sedimentation because of the complex interaction between
hydraulics of flow, reservoir operating policy, inflowing
sediment load, and changes in the reservoir bed elevation.
The traditional approach to analyzing the distribution of
deposits has relied on empirical methods, all of which
require a great deal of simplification from the actual
physical problem.
a. Main channel deposition. Conceptually, deposition starts in the main channel. As flow enters a reservoir,
the main channel fills at the upstream end until the
elevation is at or above the former overbank elevations on
either side. Flow then shifts laterally to one side or the
other, but present theory does not predict the exact location. During periods of high water elevation, deposition
will move upstream. As the reservoir is drawn down, a
channel is cut into the delta deposits and subsequent
deposition moves material farther into the reservoir. The
lateral location of the channel may shift from year to year,
but the hydraulic characteristics will be similar to those of
the natural channel existing prior to impounding the
reservoir. Vegetation will cover the exposed delta deposits
and thus attract additional deposition until the delta takes
on characteristics of a floodplain.
b. Sediment diameters. The diameter of sediment
particles commonly transported by streams ranges over five
log cycles. Generally, the coarse material will settle first in
the outer reaches of the reservoir followed by progressively
finer fractions farther down toward the reservoir dam.
Based on this depositional pattern, the reservoir is divided
into three distinct regions: top-set, fore-set, and bottom-set
beds. The top-set bed is located in the upper part of the
reservoir and is largely composed of coarse material or bed
load. While it may have a small effect on the reservoir
storage capacity, it could increase upstream stages. The
fore-set region represents the live storage capacity of the
reservoir and comprises the wash load. The bottom-set
region is located immediately upstream of the dam and is
primarily composed of suspended sediments brought from
upstream by density currents. The region is called the
reservoir dead storage and generally does not affect the
storage capacity. Some of the finest material may not settle
out and will pass through the dam. In order to calculate the
volume of material which will deposit as a function of
distance, grain size must be included as well as the
magnitude of the water discharge and the operating policy
of the reservoir.
c. Reservoir shape. Reservoir shape is an important
factor in calculating the deposition profile. For example,
flow entering a wide reservoir spreads out, thus reducing
transport capacity, but the path of expanding flow does not
necessarily follow the reservoir boundaries. It becomes a
2-dimensional problem to calculate the flow distribution
across the reservoir in order to approximate transport
capacity and, therefore, the resulting deposition pattern.
On the other hand, flow entering a narrow reservoir has a
more uniform distribution across the section resulting in
hydraulic conditions that are better approximated by 1dimensional hydraulic theory.
d. Flood waves. Flood waves attenuate upon entering a reservoir. Therefore, their sediment transport capacity decreases from two considerations: (1) a decrease in
velocity due to the increase in flow area and (2) a decrease
in velocity due to a decrease in water discharge resulting
from reservoir storage. As reservoir storage is depleted by
the sediment deposits in the delta, the impact of attenuation
on transport capacity diminishes. The resulting configuration, therefore, is assumed to depend upon the first
consideration, whereas, the time for delta development is
influenced somewhat by the second consideration.
e. Flood-pool index method. If flood control is a
project purpose, the next level of detail in reservoir sedimentation studies is to divide the total volume of predicted
deposits into that volume settling into the flood-control
pool and that volume settling in the remainder of the
reservoir. The flood-pool index method requires the depth
of flood-control pool, depth of reservoir, and the percent of
time the reservoir water level is at or above the bottom of
the flood-control pool. Based on the index, the percent of
sediment trapped in the flood-control pool is estimated by a
general empirical relationship. Appendix H of EM 1110-24000 describes the index method and provides several
other methods for estimating the distribution of sediment
deposits in reservoirs. Chapter 5, Section IV, EM 1110-24000, provides an overview of levels of sedimentation
studies and methods of analysis.
EM 1110-2-1420
31 Oct 97
EM 1110-2-1420
31 Oct 97
Chapter 14
Spillways and Outlet Works
operational pool. The outlets may also serve to empty the
reservoir to permit inspection, to make needed repairs, or to
maintain the upstream face of the dam or other structures
normally inundated.
14-1. Function of Spillways and Outlet Works
f. Outlets as flood-control regulators. Outlet works
may act as a flood-control regulator to release waters
temporarily stored in flood control storage space or to
evacuate storage in anticipation of flood inflows. In this
case, the outflow capacity should be able to release channel
capacity, or higher. The flood control storage must be
evacuated as rapidly as safely possible, in order to maintain
flood reduction capability.
Spillways and outlet works are necessary to provide
capability to release an adequate rate of water from the
reservoir to satisfy dam safety and water control regulation
of the project. Sections 4-2 and 4-3 in EM 1110-2-3600
provide general descriptions of types and operation
requirements for spillways and outlet works, respectively.
a. Spillway adequacy. While the outflow capability
must be provided throughout the operational range of the
reservoir, the focus of hydrologic studies is usually on the
high flows and spillway adequacy. Dam failures have been
caused by improperly designed spillways or by insufficient
spillway capacity. Ample capacity is of great importance
for earthfill and rockfill dams, which are likely to be
destroyed if overtopped; whereas concrete dams may be
able to withstand moderate overtopping.
b. Spillway classification. Spillways are ordinarily
classified according to their most prominent feature, either
as it pertains to their shape, location, or discharge channel.
Spillways are often referred to as controlled or
uncontrolled, depending on whether they are gated or
ungated. EM 1110-2-1603 describes a variety of spillway
types and provides hydraulic principles, design criteria, and
results from laboratory and prototype tests.
c. Outlet works. Outlet works serve to regulate or
release water impounded by a dam. It may release
incoming flows at a reduced rate, as in the case of a
detention dam; divert inflows into canals or pipelines, as in
the case of a diversion dam; or release stored-water at such
rates as may be dictated by downstream needs, evacuation
considerations, or a combination of multiple-purpose
d. Outlet structure classification. Outlet structures
can be classified according to their purpose, their physical
and structural arrangement, or their hydraulic operation.
EM 1110-2-1602 provides information on basic hydraulics, conduits for concrete dams, and conduits for earth
dams with emphases on flood-control projects. Appendix IV of EM 1110-2-1602 provides an illustrative example
of the computation of a discharge rating for outlet works.
e. Low level outlets. Low level outlets are provided
to maintain downstream flows for all levels of the reservoir
14-2. Spillway Design Flood
a. Spillway design flood analyses. Spillway design
flood (SDF) analyses are performed to evaluate the adequacy of an existing spillway or to size a spillway. For a
major project, the conservative practice in the United
States is to base the spillway design flood on the probable
maximum precipitation (PMP). The PMP is based on the
maximum conceivable combination of unfavorable meteorological events. While a frequency is not normally
assigned. a committee of ASCE has suggested that the
PMP is perhaps equivalent to a return period of
10,000 years.
b. Probable maximum flood. The PMF inflow
hydrograph is developed by centering the PMP over the
watershed to produce a maximum flood response. The unit
hydrograph approach, described in Chapter 7 of this
manual, is usually applied. Section 13-5 of EM 1110-21417 contains information on PMP determination and
computation of the PMF.
c. Flood hydrographs. The inflow design flood
hydrographs are usually for rainfall floods. Normally, such
floods will have the highest peak flows but not always the
largest volumes. When spillways of small capacities in
relation to these inflow design flood peaks are considered,
precautions must be taken to ensure that the spillway
capacity will be sufficient to evacuate storage so that the
dam will not be overtopped by a recurrent storm, and
prevent the flood storage from being kept partially full by a
prolonged runoff whose peak, although less than the inflow
design flood, exceeds the spillway capacity. To meet these
requirements, the minimum spillway capacity should be in
accord with the following general criteria (Hoffman 1977):
(1) In the case of snow-fed perennial streams, the
spillway capacity should never be less than the peak discharge of record that has resulted from snowmelt runoff.
EM 1110-2-1420
31 Oct 97
(2) The spillway capacity should provide for the
evacuation of sufficient surcharge storage space so that in
routing a succeeding flood, the maximum water surface
does not exceed that obtained by routing the inflow design
flood. In general, the recurrent storm is assumed to begin 4
days after the time of peak outflow obtained in routing the
inflow design flood.
(3) In regions having an annual rainfall of 40 in. or
more, the time interval to the beginning of the recurrent
storm in criterion (2) should be reduced to 2 days.
(4) In regions having an annual rainfall of 20 in. or
less, the time interval to the beginning of the recurrent
storm in criterion (2) may be increased to 7 days.
14-3. Area and Capacity of the Reservoir
a. Reservoir capacity and operations. Dam designs
and reservoir operating criteria are related to the reservoir
capacity and anticipated reservoir operations. The reservoir capacity and reservoir operations are used to properly
size the spillway and outlet works. The reservoir capacity
is a major factor in flood routings and may determine the
size and crest elevation of the spillway. The reservoir
operation and reservoir capacity allocations will determine
the location and size of outlet works for the controlled
release of water for downstream requirements and flood
b. Area-capacity tables. Reservoir area-capacity
tables should be prepared before the final designs and
specifications are completed. These area-capacity tables
should be based on the best available topographic data and
should be the final design for administrative purposes until
superseded by a reservoir resurvey. To ensure uniform
reporting of data for design and construction, standard
designations of water surface elevations and reservoir
capacity allocations should be used.
14-4. Routing the Spillway Design Flood
a. Discharge facilities. The facilities available for
discharging inflow from the spillway design flood depend
on the type and design of the dam and its proposed use. A
single dam installation may have two or more of the
following discharge facilities:
uncontrolled overflow
spillway, gated overflow spillway, regulating outlet, and
power plant. With a reservoir full to the spillway crest at
the beginning of the design flood, uncontrolled discharge
will begin at once. Surcharge storage is created when the
outflow capacity is less than the inflow and the excess
water goes into storage, causing the pool level to rise above
spillway crest. The peak outflow will occur at maximum
pool elevation, which should always be less, to some
degree, than the peak inflow.
b. Gated spillway. With a gated spillway, the
normal operating level is usually near the top of the gates,
although at times it may be drawn below this level by other
outlets. A gated spillway's main purpose is to maximize
available storage and head, while at the same time limiting
backwater damages by providing a high initial discharge
capacity. In routing the spillway design flood, an initial
reservoir elevation at the normal full pool operating level is
assumed. Operating rules for spillway gates must be based
on careful study to avoid releasing discharges that would be
greater than would occur under natural conditions before
construction of the reservoir. By gate operation, releases
can be reduced and additional water will be held in storage,
which is called “induced-surcharge storage.” The release
rates should be made in accordance with spillway gate
regulation schedules developed for each gated reservoir.
EM 1110-2-3600 Section 4-5 describes induced surcharge
storage and the development and testing of the regulation
c. Surcharge storage. The important factor in the
routing procedure is the evaluation of the effect of storage
in the upper levels of the reservoir, surcharge storage, on
the required outflow capacity. In computing the available
storage, the water surface is generally considered to be
level. There will be a sloping water surface at the head of
the reservoir due to backwater effect, and this condition
will create an additional “wedge storage.” However, in
most large and deep reservoirs this incremental storage can
be neglected.
d. Drawdown. If a reservoir is drawn down at the
time of occurrence of the spillway design flood, the initial
increments of inflow will be stored with the corresponding
reduction in ultimate peak outflow.
Therefore, for
maximum safety in design it is generally assumed that a
reservoir will be full to the top of flood-control pool at the
beginning of the spillway design flood.
e. Large flood-control storage reservations. There
may be exceptions to the above criteria in the case of
reservoirs with large reservations for flood-control storage.
However, even in such cases, a substantial part
(> 50 percent) of the flood-control storage should be
considered as filled by runoff from antecedent floods. The
effect on the economics and safety of the project should be
analyzed before adopting such assumptions. ER 1110-82(FR) contains guidance on inflow design flood development and application.
EM 1110-2-1420
31 Oct 97
f. Release rates. Assuming a reservoir can be significantly drawn down in advance of the spillway design
flood by using a short-term flood warning system is generally not acceptable for several reasons. The volume that
can be released is the product of the total rate of discharge
at the dam times the warning time. Because the warning
time is usually short, except on large rivers, the release rate
must be the greatest possible without flood damage
downstream. Even under the most favorable conditions, it
is unlikely that the released volume will be significant,
relative to the volume of the spillway design flood.
14-5. Sizing the Spillway
a. Storage and spillway capacity. In determining the
best combination of storage and spillway capacity to
accommodate the selected inflow design flood, all pertinent
factors of hydrology, hydraulics, design, cost, and damage
should be considered. In this connection and when applicable, consideration should be given to the following factors:
(1) The characteristics of the flood hydrograph.
(2) The damage which would result if such a flood
occurred without the dam.
(3) The damage which would result if such a flood
occurred with the dam in-place.
(4) The damage which would occur if the dam or
spillway were breached.
(5) Effects of various dam and spillway combinations
on the probable increase or decrease of damages above or
below the dam.
(6) Relative costs of increasing the capacity of the
(7) The use of combined outlet facilities to serve more
than one function.
b. Outflow characteristics. The outflow characteristics of a spillway depend on the particular device selected
to control the discharge. These control facilities may take
the form of an overflow weir, an orifice, a tube, or a pipe.
Such devices can be unregulated, or they can be equipped
with gates or valves to regulate the outflow.
c. Flood routing. After a spillway control of certain
dimensions has been selected, the maximum spillway
discharge and the maximum reservoir water level can be
determined by flood routing. Other components of the
spillway can then be proportioned to conform to the
required capacity and the specific site conditions, and a
complete layout of the spillway can be established. Cost
estimates of the spillway and dam can then be made.
Estimates of various combinations of spillway capacity and
dam height for an assumed spillway type, and of alternative
types of spillways, will provide a basis for the selection of
the most economical spillway type and the optimum
relation of spillway capacity to the height of the dam.
d. Maximum reservoir level. The maximum reservoir level can be determined by routing the spillway design
flood hydrograph using sequential routing procedures and
the proposed operation procedures. This is a basic step in
the selection of the elevation of the crest of the dam, the
size of the spillway, or both.
e. Peak rate of inflow. Where no flood storage is
provided, the spillway must be sufficiently large to pass the
peak of the flood. The peak rate of inflow is then of
primary interest, and the total volume in the flood becomes
less important. However, where a relatively large storage
capacity above normal reservoir level can be made
economically available by a higher dam, a portion of the
flood volume can be retained temporarily in reservoir
surcharge space, and the spillway capacity can be reduced
considerably. If a dam could be made sufficiently high to
provide storage space to impound the entire volume of the
flood above normal storage level, theoretically, no spillway
other than an emergency type would be required, provided
the outlet capacity could evacuate the surcharge storage in
a reasonable period of time in anticipation of a recurring
flood. The maximum reservoir level would then depend
entirely on the volume of the flood, and the rate of inflow
would be of no concern. From a practical standpoint,
however, relatively few sites will permit complete storage
of the inflow design flood by surcharge storage.
f. Overall cost. The spillway length and corresponding capacity may have an important effect on the
overall cost of a project because the selection of the spillway characteristics is based on an economic analysis. In
many reservoir projects, economic considerations will
necessitate a design utilizing surcharge.
The most
economical combination of surcharge storage and spillway
capacity requires flood routing studies and economic
studies of the costs of spillway-dam combinations. Among
the many economic factors that may be considered are
damage due to backwater in the reservoir, cost-height
relations for gates, and utilization in the dam of material
excavated from the spillway channel.
consideration must still be given to the minimum size
spillway which must be provided for safety.
EM 1110-2-1420
31 Oct 97
g. Comprehensive study. The study may require
many flood routings, spillway layouts, and spillway and
dam estimates. Even then, the study is not necessarily
complete because many other spillway arrangements could
be considered. A comprehensive study to determine
alternative optimum combinations and minimum costs may
not be warranted for the design of some dams. Judgment
on the part of the designer would be required to select for
study only those combinations which show definite
advantages, either in cost or adaptability. For example,
although a gated spillway might be slightly cheaper than an
ungated spillway, it may be desirable to adopt the latter
because of its less complicated construction, its automatic
and trouble-free operation, its ability to function without an
attendant, and its less costly maintenance.
14-6. Outlet Works
a. Definition. An outlet works consists of the
equipment and structures which together release the
required water for a given purpose or combination of
purposes. Flows through river outlets and canal or pipeline
outlets change throughout the year and may involve a wide
range of discharges under varying heads. The accuracy and
ease of control are major considerations and a great amount
of planning may be justified in determining the type of
control devices that can be best utilized.
b. Description. Usually, the outlet works consist of
an intake structure, a conduit or series of conduits through
the dam, discharge flow control devices, and an energy
dissipating device where required downstream of the dam.
The intake structure includes a trash-rack, an entrance
transition, and stop-logs or an emergency gate. The control
device can be placed at the intake on the upstream face, at
some point along the conduit and be regulated from
galleries inside the dam, or at the downstream end of the
conduit with the operating controls placed in a gate-house
on the downstream face of the dam. When there is a power
plant or other structure near the face of the dam, the outlet
conduits can be extended farther downstream to discharge
into the river channel beyond these features. In this case, a
control valve may be placed in a gate structure at the end of
the conduit.
c. Discharge. Discharges from a reservoir outlet
works fluctuate throughout the year depending upon
downstream water needs and reservoir flood control
Therefore, impounded water must be
released at specific regulated rates. Operating gates and
regulating valves are used to control and regulate the outlet
works flow and are designed to operate in any position
from closed to fully open. Guard or emergency gates are
designed to close if the operating gates fail, or where
dewatering is desired to inspect or repair the operating
d. Continuous low-flow releases. Continuous lowflow releases are usually required to satisfy the needs of
fish, wildlife and existing water rights downstream from the
dam. When the low-flow release is small, one or two
separate small bypass pipes, with high-pressure regulating
valves, are provided to facilitate operations. Floodregulating gates may be used for making low-flow releases
when those low-flow releases require substantial gate
openings (EM 1110-2-1602).
e. Uses of an outlet works. An outlet works may be
used for diverting the river flow or portion thereof during a
phase of the construction period, thus avoiding the
necessity for supplementary installations for that purpose.
The outlet structure size dictated by this use rather than the
size indicated for ordinary outlet requirements may
determine the final outlet works capacity.
f. Intake level. The establishment of the intake
level is influenced by several considerations such as maintaining the required discharge at the minimum reservoir
operating elevation, establishing a silt retention space, and
allowing selective withdrawal to achieve suitable water
temperature and/or quality. Dams which will impound
waters for irrigation, domestic use, or other conservation
purposes must have the outlet works intake low enough to
be able to draw the water down to the bottom of the
allocated storage space. Further, if the outlets are to be
used to evacuate the reservoir for inspection or repair of
the dam, they should be placed as low as practicable.
However, it is usual practice to make an allowance in a
reservoir for inactive storage for silt deposition, fish and
wildlife conservation, and recreation.
g. Elevation of outlet intake. Reservoirs become
thermally stratified, and taste and odor vary between elevations. Therefore, the outlet intake should be established
at the best elevation to achieve satisfactory water quality
for the purpose intended. Downstream fish and wildlife
requirements may determine the temperature at which the
outlet releases should be made. Municipal and industrial
water use increases the emphasis on water quality and
requires the water to be drawn from the reservoir at the
elevation which produces the most satisfactory combination of odor, taste, and temperature. Water supply
releases can be made through separate outlet works at
different elevations if requirements for the individual water
uses are not the same and the reservoir is stratified.
EM 1110-2-1420
31 Oct 97
h. Energy-dissipating devices. The two types of
energy dissipating devices most commonly used in conjunction with outlet works on concrete dams are hydraulic
jump stilling basins and plunge pools. On some dams, it is
possible to arrange the outlet works in conjunction with the
spillway to utilize the spillway-stilling device for
dissipating the energy of the water discharging from the
river outlets. Energy-dissipating devices for free-flow
conduit outlet works are essentially the same as those for
EM 1110-2-1420
31 Oct 97
Chapter 15
Dam Freeboard Requirements
15-1. Basic Considerations
a. Freeboard.
Freeboard protects dams and
embankments from overflow caused by wind-induced tides
and waves. It is defined as the vertical distance between
the crest of a dam and some specified pool level, usually
the normal operating level or the maximum flood level.
Depending on the importance of the structure, the amount
of freeboard will vary in order to maintain structural
integrity and the estimated cost of repairing damages
resulting from overtopping. Riprap or other types of slope
protection are provided within the freeboard to control
erosion that may occur even without overtopping.
b. Estimating freeboard. Freeboard is generally
based on maximum probable wind conditions when the
reference elevation is the normal operating level. When
estimating the freeboard to be used with the probable
maximum reservoir level, a lesser wind condition is used
because it is improbable that maximum wind conditions
will occur simultaneously with the maximum flood level.
A first step in wave height determinations is a study of
available wind records to determine velocities and related
durations and directions. Three basic considerations are
generally used in establishing freeboard allowance. These
are wave characteristics, wind setup, and wave runup.
c. Further information. The Corps of Engineers
Coastal Engineering Research Center (CERC) has developed criteria and procedures for evaluating each of the
above areas. The primary references are EM 1110-2-1412
and EM 1110-2-1414. The procedures presented in these
manuals have received general acceptance for use in estimating freeboard requirements for reservoirs.
d. Applications. In applications for inland reservoirs,
it is necessary to give special consideration to the
influences that reservoir surface configuration, surrounding
topography, and ground roughness may have on wind
velocities and directions over the water surface. The
effects of shoreline irregularities on wave refraction and
influences of water depth variations on wave heights and
lengths must be accounted for. Although allowances can
only be approximated, the estimates of wave and wind tide
characteristics in inland reservoirs can be prepared
sufficiently accurate for engineering purposes.
15-2. Wind Characteristics over Reservoirs
a. General. The more violent windstorms experienced in the United States are associated with tropical
storms (hurricanes) and tornadoes. Hurricane wind characteristics may affect reservoir projects located near Atlantic and Gulf coastlines, but winds associated with tornadoes
are not applicable to the determination of freeboard
allowances for wave action. In mountainous regions, the
flow of air is influenced by topography as well as meteorological factors. These “orographic” wind effects, when
augmented by critical meteorological patterns, may produce high wind velocities for relatively long periods of
time. Therefore, they should be given special consideration
in estimating wave action in reservoirs located in
mountainous regions. In areas not affected by major
topographic influences, air movement is generally the result
of horizontal differences in pressure which in turn are due
primarily to large-scale temperature differences in air
masses. Wind velocities and durations associated with
these meteorological conditions, with or without major
influences of local topography, are of major importance in
estimating wave characteristics in reservoirs.
b. Isovel patterns. Estimates of wind velocities and
directions near a water surface at successive intervals of
time, as a windstorm passes the area, may be established by
deriving “isovel” patterns.
Sequence relations can
represent wind velocities at, say, one-half hour intervals
during periods of maximum winds, and one-hour or longer
intervals thereafter. The “isovel” lines connect points of
equal wind speeds, resembling elevation contour maps.
Wind directions are indicated by arrows. EM 1110-2-1412,
Sections 1.9 and 1.10 describe storms and the storm surge
generation process. Figure 1-1a shows an example wind
isovel pattern and pertinent parameters.
c. Relation of wind duration to wave heights. If
wind velocity over a particular fetch remains constant,
wave heights will progressively increase until a limiting
maximum value is attained, corresponding approximately
to relations dependent on fetch distance, wind velocity, and
duration. Accordingly, wind velocity-duration relations
applicable to effective reservoir fetch areas are needed for
use in computing wave characteristics in reservoirs.
d. Wind velocity-duration relations. In some cases
it is desired to estimate wave characteristics in existing
reservoirs in order to analyze causes of riprap damage or
for other reasons. Wind records, supplemented by meteorological studies are usually required. Data on actual
EM 1110-2-1420
31 Oct 97
windstorms of record have been maintained at many
U.S. Weather Bureau stations. Index values, such as the
fastest mile, 1-min average or 5-min average velocities,
with direction indications, are usually presented in climatological data publications. Some data collected by other
agencies and private observers may be available in published or unpublished form.
However, information
regarding wind velocities sustained for several hours or
days is not ordinarily published in detail. Accordingly,
special studies are usually required to determine wind
velocity-duration relations applicable to specific effective
fetch areas involved in wave computations. Basic records
for such studies are usually available from the U.S. Weather Bureau offices or other observer stations. Some
summaries of wind velocities over relatively long periods
of time have been published by various investigators, and
others may be available in project reports related to water
resources development.
e. Generalized wind velocity-duration relations.
Studies show that maximum wind velocities in one general
direction during major windstorms, in most regions of the
United States, have averaged approximately 40 to 50 mph
for a period of 1 hr. Corresponding velocities in the same
general direction for periods of 2 hr and 6 hr have averaged
95 percent and 88 percent, respectively, of the maximum
1-hr average velocity. In EM 1110-2-1414, Figure 5-26
provides the ratio of wind speed of any duration to the 1-hr
wind speed. Extreme wind velocities for brief periods,
normally referred to as “fastest mile” or 1-min average,
have been recorded as high as 150 to 200 percent of
maximum 1-hr averages in most regions. In EM 1110-21414, Figures 5-18 through 5-20 provide the annual
extreme fastest-mile speed 30 ft (9.1 m) above ground for
the 25-year, 50-year, and 100-year recurrence intervals,
respectively. However, these extreme values are seldom of
interest in computing wave characteristics in reservoirs.
Generalized wind velocity-duration relations are considered
to be fairly representative of maximum values that are
likely to prevail over a reservoir in generally a single
direction for periods up to 6 hr (excluding projects located
in regions that are subject to severe hurricanes or
orographic wind-flow effects). Special studies of wind
characteristics associated with individual project areas
should be made when determinations of unusual
importance, or problems requiring consideration of wind
durations exceeding 6 hr, are involved.
f. Ratio of wind velocities over water and land areas.
The wind velocities described in paragraph d are for over
land. Under comparable meteorological conditions, wind
velocities over water are higher than over land surfaces
because of smoother and more uniform surface conditions.
Winds blowing from land tend to increase with passage
over reservoir areas, and vice versa. The relationships are
not constant, but vary with topographic and vegetative
cover of land areas involved, reservoir configurations, and
other conditions affecting air flow. However, on the basis
of research and field studies (Technical Memorandum
No. 132, USACE 1962), the following ratios represent
averages that are usually suitable for computing wave
characteristics in reservoirs that are surrounded by terrain
of moderate irregularities and surface roughness:
Fetch (Fe) in Miles
Wind ratio Over Water
Over Land
5 (or over)
Computation of Wave and Wind Tide
a. Effective wind fetch (F e) for wave generation.
The characteristics of wind-generated waves are influenced
by the distance that wind moves over the water surface in
the “fetch” direction. The generally narrow irregular
shoreline of inland reservoirs will have lower waves than
an open coast because there is less water surface for the
wind to act on. The method to compensate for the reduced
water surface for an enclosed body of water is computation
of an effective fetch. The effective fetch (F e) adjusts radial
lines from the embankment to various points on the
reservoir shore. The radials spanning 45 deg on each side
of the central radial are adjusted by the cosign of their
angle to the central radial to estimate an average effective
fetch. The computation procedure is shown in EM 1110-21414, Figure 5-33 and Example Problem 7-2. Generalized
relations are based on effective fetch distances derived in
this manner.
b. Fetch distance for wind tide computations. Fetch
distances for use in estimating wind tide (set-up) effects are
usually longer than effective fetch distances used in
estimating wave heights. In as much as wind tide effects in
deep inland reservoirs are relatively small, extensive
studies to refine estimates are seldom justified. For practical purposes, it is usually satisfactory to assume that the
wind tide fetch is equal to twice the effective fetch (F e). If
wind tide heights determined in this manner are relatively
large in relation to overall freeboard requirements, more
detailed analyses are advisable using methods as generally
discussed in Chapter 3 of EM 1110-2-1414.
EM 1110-2-1420
31 Oct 97
c. Generalized diagrams for wave height and wave
period in deep inland reservoirs. EM 1110-2-1414, Figure 5-34, presents generalized relations between significant
wave height, wave period, fetch (F e), and wind velocities
corresponding to critical durations. These diagrams were
developed from research and field studies based on wind
speed at 10 m (33 ft). If wind speeds are for a different
level, Equation 5-12 can be used to adjust to the 10 m level
(EM 1110-2-1414).
d. Wave characteristics in shallow inland lakes and
reservoirs. In the analyses of wave characteristics, lakes
and reservoirs are considered to be shallow when depths in
the wave-generating area are generally less than about onehalf the theoretical deep-water wave length (L o)
corresponding to the same wave period (T). Curves presented in Figures 5-35 through 5-44 represent relations
between wave characteristics, fetch distances (in feet) and
constant water depths in the wave-generating area, ranging
from 5 to 50 ft (1.5 to 15.2 m) (EM 1110-2-1414).
many slopes, wave conditions, and embankment porosity
provide sufficient data to make estimates of wave runup on
a prototype embankment.
b. Relative runup relations.
EM 1110-2-2904
Plate 25 presents generalized relations on wave runup on
rubble-mound breakwaters and smooth impervious slopes.
Plate 26 provides similar curves for various embankment
slopes for water depths greater than three-times wave
height. The curves correspond to statistical averages of a
large number of small and large-scale hydraulic model test
results, and have been adjusted for model scale effects to
represent prototype conditions. The relations were based
primarily on tests involving mechanically generated waves
and may differ somewhat from relations associated with
individual waves in natural wind-generated spectrums of
However, general field observations and
comparisons with wave experiences support the conclusion
that relations presented.
Runup of waves on sloping embankments.
Wind tides (set-up) in inland waters.
(1) When wind blows over a water surface, it exerts a
horizontal stress on the water, driving it in the direction of
the wind. In an enclosed body of water, this wind effect
results in a piling up of water at the leeward end, and a
lowering of water level at the windward end. This effect is
called “wind tide” or “wind set-up.” Wind set-up can be
reasonably estimated for lakes and reservoirs, based on the
following equation:
S =
U2 F
1400 D
in which S is wind tide (set-up) in feet above the stillwater
level that would prevail with zero wind action; U is the
average wind velocity in statute miles per hour over the
fetch distance (F) that influences wind tide; D is the
average depth of water generally along the fetch line
(EM 1110-2-1414). The fetch distance (F) used in the
above formula is usually somewhat longer than the effective fetch (Fe) used in wave computations, as indicated in
paragraph 15-3b. Refer to EM 1110-2-1414, Section 3-2
for a discussion of prediction models.
15-4. Wave Runup on Sloping Embankment
a. Introduction. Most dam embankments are fronted
by deep water, have slopes between 1 on 2 and 1 on 4, and
are armored with riprap. Rock-fill dams are considered as
permeable rubble slopes and earth-fill dams with riprap
armor are considered impermeable. Laboratory tests of
(1) If waves generated in deep water (i.e., depths
exceeding about one-third to one-half the wave length),
reach the toe of an embankment without breaking, the
vertical height of runup may be computed by multiplying
the deep-water wave height (H o) by the relative runup
ration (R/H) obtained from EM 1110-2-2904, Plate 26, for
the appropriate slope and wave steepness (H o/L o). In this
case, deep-water values of Ho and Lo should be used as
indices, even though wave heights and lengths are modified
by passing through areas in which water depths are less
than Lo/3 (provided the depth is not small enough to cause
the wave to break before reaching the embankment). That
is, the height of runup may be computed by using the deepwater steepness H o/L o, whether the structure under study is
located in deep water or in shallow water, provided the
wave does not break before reaching the toe of the
(2) If waves are generated primarily by winds over
open-water areas where the relative depth (d/L) is
appreciably less than 0.3, the wave heights and periods
should be computed by procedures applicable to shallow
(3) Waves generated by wind over open-water areas
of a particular depth change characteristics when they
reach areas where the constant depth is substantially less,
the height (H) tending to increase while the length (L)
decreases. The distribution of wave energy changes as a
wave enters the shallow water, the proportion of total
energy which is transmitted forward with the wave toward
the shore increasing, although the actual amount of this
EM 1110-2-1420
31 Oct 97
translated energy remains constant except from minor
frictional effects. If the depth continues to decrease, the
steepness ratio (H/L) increases, until finally the wave
becomes unstable and breaks, resulting in appreciable
energy dissipation. Theoretically, the maximum wave
cannot exceed 0.78 D, where D is the depth of water
without wave action. After breaking, the waves will tend to
reform with lower heights within a distance equal to a few
wave lengths. For most engineering applications, it is
satisfactory to assume that the wave height after breaking
will equal approximately 0.78 D in the shallow area and
that L will be the same as before the wave broke. Plate 26
would then be entered with a wave-steepness ratio equal to
0.78/L to determine the relative runup ration (R/H), and
this ratio would be multiplied by 0.78 D to obtain the
estimated runup height (R). This procedure should provide
conservative results under circumstances in which the
distance between the point where waves reach breaking
depths and location of the structure under study is long
enough to permit waves to reform, and short enough to
preclude substantial build-up by winds prevailing over the
shallower area. More accurate values could be obtained by
using this breaking height (0.78 D) and period to obtain
comparable deep water values of H o and L o.
d. Adjustments in wave runup estimates for variations in riprap.
(1) A rough riprap layer on an embankment tends to
reduce the height of runup after a wave breaks. If the
riprap layer thickness is small in comparison with wave
magnitudes and the underlying surface is relatively impermeable, so that the void spaces in the riprap remain mostly
filled with water between successive waves during severe
storm events, the height of runup may closely approach
heights attained on smooth embankments of comparable
slope. However, if the riprap layer is sufficiently rough,
thick, and free draining to quickly absorb the water that
impinges on the embankment as each successive wave
breaks, further wave runup will be almost completely
(2) The design of riprap to absorb most of the energy
of breaking waves is practicable if waves involved will be
relatively small or moderate, but costs and other practical
considerations usually preclude such design where large
waves are encountered.
Accordingly, the design
characteristics of riprap layers are usually somewhere
between the two extremes described above.
15-5. Freeboard Allowances for Wave Action
a. Purpose. In connection with the design of dams
and reservoirs, the estimate of freeboard is required to
establish allowances needed to provide for wave action that
is likely to affect various project elements, as follows:
(1) Main embankment of the dam, and supplemental
dike sections.
(2) Levees that protect areas within potential flowage
limits of the reservoir.
(3) Highway and railroad embankments that intersect
the reservoir limits.
(4) Structures located within the reservoir area.
(5) Shoreline areas that are subject to adverse effects
of wave action.
b. Freeboard on dams. The establishment of freeboard allowances on dams includes not only the consideration of potential wave characteristics in a reservoir, but
several other factors of importance, including certain policy
c. Freeboard allowances for wave action on
embankments and structures within reservoir flowage
(1) Wave action effects must be taken into account in
establishing design grades and slope protection measures
for highway, railroad, levee, and other embankments that
intersect or boarder a reservoir. The design of operating
structure, boat docks, recreational beaches, and shoreline
protection measures at critical locations involves the
consideration of wave characteristics and frequencies under
a range of conditions. Estimates of wave characteristics
affecting the design of these facilities can have a major
influence on the adequacy of design and costs of
relocations required for reservoir projects, and in the
development of supplemental facilities.
(2) The freeboard reference level selected as a base
for estimating wave effects associated with each of the
several types of facilities referred to above will be governed by considerations associated with the particular
facility. Otherwise, procedures generally as described with
respect to the determination of freeboard allowances for
dams should be followed, and stage hydrographs and
related wave runup elevations corresponding to the selected
wind criteria should be prepared. However, the freeboard
reference level and coincident wind velocity-duration
relations selected for these studies usually correspond to
conditions that would be expected with moderate
frequency, instead of the rare combinations assumed in
estimating the height of dam required for safety.
EM 1110-2-1420
31 Oct 97
(3) In estimating effects of wave action on embankments and structures, the influences of water depths near
the facility should be carefully considered. If the shallow
depths prevail for substantial distances from the embankment or structure under study, wave effects may be greatly
reduced from those prevailing in deep-water areas. On the
other hand, facilities located where sudden reductions in
water depths cause waves to break are likely to be
subjected to greater dynamic forces than would be imposed
on similar facilities located in deep water.
consideration is particularly important in estimating the
effects waves may have on bridge structures that are
partially submerged under certain reservoir conditions.
(4) Systematic analyses of wave effects associated
with various key locations along embankments that cross or
border reservoirs provide a practical basis for varying
design grades and erosion protection measures to establish
the most economical plan to meet pertinent operational and
maintenance standards.
EM 1110-2-1420
31 Oct 97
Chapter 16
Dam Break Analysis
16-1. Introduction
a. Corps policy. It is the policy of the Corps of Engineers to design, construct, and operate dams safely
(ER 1110-8-2(FR)). When a dam is breached, catastrophic
flash flooding occurs as the impounded water escapes
through the gap into the downstream channel. Usually, the
response time available for warning is much shorter than
that for precipitation-runoff floods, so the potential for loss
of life and property damage is much greater.
b. Hazard evaluation. A hazard evaluation is the
basis for selecting the performance standards to be used in
dam design or in evaluating existing dams. When flooding
could cause significant hazards to life or major property
damage, the design flood selected should have virtually no
chance of being exceeded. ER 1110-8-2(FR) provides dam
safety standards with respect to the appropriate selection of
an inflow design flood. If human life is at risk, the general
requirement is to compute the flood using PMP. If lesser
hazards are involved, a smaller flood may be selected for
design. However, all dams should be designed to withstand
a relatively large flood without failure even when there is
apparently no downstream hazard involved under present
conditions of development.
e. Flood emergency documents. In support of the
National Dam Safety Program, flood emergency planning
for dams was evaluated in the 1980's, and a series of documents were published: Emergency Planning for Dams,
Bibliography and Abstracts of Selected Publications (HEC
1982a), Flood Emergency Plan, Guidelines for Corps
Dams, Research Document 13 (HEC 1980), Example
Emergency Plan for Blue Marsh Dam and Lake (HEC
1983a), and Example Plan for Evacuation of Reading,
Pennsylvania in the Event of Emergencies at Blue Marsh
Dam and Lake (HEC 1983b). The development of an
emergency plan requires the identification of the type of
emergencies to be considered, the gathering of needed data,
performing the analyses and evaluations, and presenting the
results. HEC Research Document 13 (HEC 1980) provides
guidelines for each step of the process.
16-2. Dam Breach Analysis
a. Causes of dam failures. Dam failures can be
caused by overtopping a dam due to insufficient spillway
capacity during large inflows to the reservoir, by seepage or
piping through the dam or along internal conduits, slope
embankment slides, earthquake damage and liquification of
earthen dams from earthquakes, or landslide-generated
waves within the reservoir. Hydraulics, hydrodynamics,
hydrology, sediment transport mechanics, and geotechnical
aspects are all involved in breach formation and eventual
dam failure.
HEC Research Document 13 lists the
prominent causes as follows:
c. Safety design. Safety design includes studies to
ascertain areas that would be flooded during the design
flood and in the event of dam failure. The areas downstream from the project should be evaluated to determine
the need for land acquisition, flood plain management, or
other methods to prevent major damage. Information
should be developed and documented suitable for releasing
to downstream interests regarding the remaining risks of
(1) Earthquake.
d. National Dam Safety Act. The potential for catastrophic flooding due to dam failures in the 1960's and
1970's brought about passage of the National Dam Safety
Act, Public Law 92-367. The Corps of Engineers became
responsible for inspecting U.S. Federal and non-Federal
dams, which met the size and storage limitations of the act,
in order to evaluate their safety. The Corps inventoried
dams; surveyed each State and Federal agency's
capabilities, practices, and regulations regarding the design,
construction, operation, and maintenance of the dams;
developed guidelines for the inspection and evaluation of
dam safety; and formulated recommendations for a
comprehensive national program.
(6) Structural damage.
(2) Landslide.
(3) Extreme storm.
(4) Piping.
(5) Equipment malfunction.
(7) Foundation failure.
(8) Sabotage.
b. Dam breach characteristics. The breach is the
opening formed in the dam when it fails. Despite the fact
that the main modes of failure have been identified as
piping or overtopping, the actual failure mechanics are not
well understood for either earthen or concrete dams. In
previous attempts to predict downstream flooding due to
dam failures, it was usually assumed that the dam failed
EM 1110-2-1420
31 Oct 97
completely and instantaneously. These assumptions of
instantaneous and complete breaches were used for reasons
of convenience when applying certain mathematical
techniques for analyzing dam-break flood waves. The
presumptions are somewhat appropriate for concrete archtype dams, but they are not suitable for earthen dams and
concrete gravity-type dams.
(1) Earthen dams, which exceedingly outnumber all
other types of dams, do not tend to completely fail, nor do
they fail instantaneously. Once a developing breach has
been initiated, the discharging water will erode the breach
until either the reservoir water is depleted or the breach
resists further erosion. The fully formed breach in earthen
dams tends to have an average width (b) in the range
(hdyby3hd) where, hd is the height of the dam. Breach
widths for earthen dams are therefore usually much less
than the total length of the dam as measured across the
valley. Also, the breach requires a finite interval of time
for its formation through erosion of the dam materials by
the escaping water. The total time of failure may range
from a few minutes to a few hours, depending on the height
of the dam, the type of materials used in construction, and
the magnitude and duration of the flow of escaping water.
Piping failures occur when initial breach formation takes
place at some point below the top of the dam due to erosion
of an internal channel through the dam by escaping water.
As the erosion proceeds, a larger and larger opening is
formed. This is eventually hastened by caving-in of the top
portion of the dam.
(2) Concrete gravity dams also tend to have a partial
breach as one or more monolith sections formed during the
dam construction are forced apart by the escaping water.
The time for breach formation is in the range of a few
(3) Poorly constructed earthen dams and coal-waste
slag piles which impound water tend to fail within a few
minutes and have average breach widths in the upper range
or even greater than those for the earthen dams mentioned
c. Dam breach parameters. The parameters of failure depend on the dam and the mode of failure. For flood
hydrograph estimation, the breach is modeled assuming
weir conditions, and the breach size, shape, and timing are
the important parameters. The larger the breach opening
and the shorter the time to total failure, the larger the peak
outflow. HEC Research Document 13, Table 1, lists
suggested breach parameters for earth-fill, concretegravity, and concrete-arch dams. There are two basic
approaches used to determine possible breach sizes and
(1) The first approach uses statistically derived
regression equations, like those formulated by MacDonald
and Langridge-Monopolis (1984) and by Froelich (1987).
Both sets of equations are based on actual data from
dozens of historic dam failures. The MacDonald, and
Langridge-Monopolis study was based on data from
42 constructed earth- and rock-fill dams. The Froelich
study included data from constructed and landslide-formed
earthen dams. Both studies resulted in a set of graphs and
equations that can be used to predict the approximate size
of the breach and the time it takes for the breach to reach
its full size.
(2) The second approach is a physically based computer model called BREACH, developed by Dr. Danny
Fread (1989) for the National Weather Service. The breach
model uses sediment transport and hydraulic routing
equations to simulate the formation of either a piping or
overtopping type of failure. The model requires information about the physical dimensions of the dam, as well
as a detailed description of the soil properties of the dam.
Soils information includes D50 (mm), porosity, unit weight
(lb/ft 3), internal friction angle, cohesive strength (lb/ft 2),
and D90/D30.
These parameters can be specified
separately for the inner-core and outside-bank materials of
a dam.
16-3. Dam Failure Hydrograph
a. Flow hydrograph. The flow hydrograph from a
breached dam may be computed using traditional methods
for flow routing through a reservoir and downstream
channel. The reservoir routing approach is the same as
routing for the spillway design flood, described in Chapter 14. Generally, a short time step is required because the
breach formation and resulting reservoir outflow change
rapidly with time.
b. Routing methods. The choice between hydraulic
and hydrologic routing depends on many factors, including
the nature of available data and accuracy required. The
hydraulic method is the more accurate method of routing
the unsteady flow from a dam failure flood through the
downstream river.
This technique simultaneously
computes the discharge, water surface elevation, and
velocity throughout the river reach.
Chapter 9 of
EM 1110-2-1417 describes the routing methods and
applicability of routing techniques. Chapter 5 of EM 11102-1416 describes unsteady flow computations.
c. Geometry and surface area. The geometry and
surface area of the reservoir can also affect the choice of
method. For very narrow and long reservoirs where the
dam is relatively large, the change of water level at the
EM 1110-2-1420
31 Oct 97
failed dam is rapid, and the unsteady flow method is useful.
However, for very large reservoirs where the dam is small
compared to the area of the lake, the change in water level
is relatively slow and the storage routing method (Modified
Puls) is economical in developing the failure hydrograph.
Because of the rapid change in water level, small time
periods are required for both methods.
d. Height of downstream water. The height of the
water downstream of a dam (tailwater) also affects the
outflow hydrograph in a failure analysis. It also affects the
formation or nonformation of a bore in front of the wave.
e. Deriving the peak outflow. By assuming a rectangular cross section, zero bottom slope, and an instantaneous failure of a dam, the peak outflow can be derived
by the mathematical expression originally developed by St.
Venant, as follows:
Qmax =
W g Yo3/2
27 b
(1) Level-pool reservoir routing to determine timehistory of pool elevation.
(2) Breach shape is a generalized trapezoid with
bottom width and side slopes prespecified by the analyst.
(3) Bottom of the breach moves downward at a
constant rate.
(4) Breach formation begins where the water surface
in the reservoir reaches a prespecified elevation.
(5) Breach is fully developed when the bottom
reaches a prespecified elevation.
Yo is the initial depth, Wb is the width of the breach, g is
the gravity coefficient, and the water depth, y, just
downstream of the dam is
y =
g. Potential for overtopping.
The Hydrologic
Engineering Center's HEC-1 Flood Hydrograph Package
(HEC 1990c) can be used to determine the potential for
overtopping of dams by run off resulting from various
proportions of the PMF. This technique is most appropriate for simulating breaches in earthen dams caused by
overtopping. Other conditions may be approximated,
however, such as instantaneous failure. This method
makes six assumptions:
9 o
This equation is applicable only for relatively long and
narrow rectangular channels where the dam is completely
removed. Guidelines for Calculating and routing a DamBreak Flood, HEC Research Document No. 5 (HEC 1977)
describes this approach.
f. Failed dam outflow hydrograph. The outflow
hydrograph from a failed dam may also be approximated
by a triangle. For instantaneous failure, a right triangle is
applicable. The base represents the time to empty the
reservoir volume, and the height represents the
instantaneous peak outflow. In erosion analysis, the Office
of Emergency Services, after consultation with other agencies, suggested an isosceles triangle. The rising side of the
isosceles triangle is developed by assuming that half of the
reservoir storage is required to erode the dam to natural
ground level. The apex of the triangle represents the peak
flow through the breach under the assumption that the flow
occurs at critical depth.
(6) Discharge through the breach can be calculated
independently of downstream hydraulics, i.e., critical
depth occurs at or near the breach. A tailwater rating curve
or a single cross section (assuming normal-depth for a
rating) can be used to simulate submergence effects.
The total discharge from the dam at any instant is calculated by summing the individual flows through the low
level outlet, over the spillway and top of the dam, and
through the breach.
h. Peak flow values. With several calculations of
theoretical flood peaks from assumed breaches, peak flow
values may seem either too low or too high. One way of
checking the reasonableness of the assumption is to compare the calculated values with historical failures. An
envelope of estimated flood peaks from actual dam failures
prepared by the Bureau of Reclamation is a good means of
comparing such values. HEC Research Document No. 13,
Figure 2, provides an envelope of experienced outflow
rates from breached dams, as a function of hydraulic depth.
16-4. Dam Break Routing
a. Dam-break flood hydrographs. Dam-break flood
hydrographs are dynamic, unsteady flow events.
EM 1110-2-1420
31 Oct 97
Therefore, the preferred routing approach is to utilize a full
unsteady flow routing model. The HEC-1 Flood Hydrograph package provides the capability to compute and route
the inflow design flood and compute the breach and
resulting hydrograph, but its channel routing is limited to
hydrologic methods.
The most appropriate HEC-1
approach is the Muskingum-Cunge option. The option uses
a simple cross section plus reach slope and length to define
a routing reach. No downstream backwater effects are
considered. If simplified representations of the downstream river reaches are acceptable, an adequate routing
may be obtained.
b. St. Venant equations. The St. Venant equations
apply to gradually varied flow with a continuous profile. If
features which control or interrupt the water surface profile
exist along the main stem of the river or its tributaries,
internal boundary conditions are required. These features
include dams, bridges, roadway embankments, etc. If the
structure is a dam, the total discharge is the sum of spillway
flow, flow over the top of the dam, gated-spillway flow,
flow through turbines, and flow through a breach, should a
breach occur. The spillway flow and dam overtopping are
treated as weir flow, with corrections for submergence.
The gated outlet can represent a fixed gate or one in which
the gate opening can vary with time. These flows can also
be specified by rating curves which define discharge
passing through the dam as a function of upstream water
surface elevation.
c. Unsteady flow computer programs. There are an
increasing number of available unsteady flow computer
programs. The FLDWAV program is a generalized
unsteady-flow simulation model for open channels. It
replaces the DAMBRK, DWOPER, and NETWORK
models, combining their capabilities and providing new
hydraulic simulation procedures within a more userfriendly model structure (DeVries and Hromadka 1993).
Given the long history of application by the National
Weather Service, this program is likely the most capable
for this purpose.
d. FLDWAV. FLDWAV can simulate the failure of
dams caused by either overtopping or piping failure of the
dam. The program can also represent the failure of two or
more dams located sequentially on a river. The program is
based on the complete equations for unsteady open-channel
flow (St. Venant equations). Various types of external and
internal boundary conditions are programmed into the
model. At the upstream and downstream boundaries of the
model (external boundaries), either discharges or water
surface elevations, which vary with time, can be specified.
e. Special features. The following special features
and capacities are included in FLDWAV: variable t and
x computational intervals; irregular cross-sectional geometry; off-channel storage; roughness coefficients that vary
with discharge or water surface elevation, and with distance
along the waterway; capability to generate linearly
interpolated cross sections and roughness coefficients
between input cross sections; automatic computation of
initial steady flow and water elevations at all cross sections
along the waterway; external boundaries of discharge or
water surface elevation time series (hydrographs), a singlevalued or looped depth-discharge relation (tabular or
computed); time-dependent lateral inflows (or outflows);
internal boundaries enable treatment of time-dependent
dam failures, spillway flows, gate controls, or bridge flows,
or bridge-embankment overtopping flow; short-circuiting
of floodplain flow in a valley with a meandering river;
levee failure and/or overtopping; a special computational
technique to provide numerical stability when treating
flows that change from supercritical to subcritical , or
conversely, with time and distance along the waterway; and
an automatic calibration technique for determining the
variable roughness coefficient by using observed
hydrographs along the waterway.
f. UNET. The unsteady flow program UNET (HEC
1995) has a dam-break routing capability. However, there
has been limited application of this feature. UNET could
be used to route the outflow hydrograph computed in an
HEC-1 runoff-dam break model. Both programs can read
and write hydrographs using the HEC Data Storage System,
HEC-DSS (HEC 1995a).
16-5. Inundation Mapping
a. Preparation of maps. To evaluate the effects of
dam failure, maps should be prepared delineating the area
which would be inundated in the event of failure. Land
uses and significant development or improvements within
the area of inundation should be indicated. The maps
should be equivalent to or more detailed than the USGS
quadrangle maps, 7.5-min series, or of sufficient scale and
detail to identify clearly the area that should be evacuated if
there is evident danger of failure of the dam. Copies of the
maps should be distributed to local government officials for
use in the development of an evacuation plan. The intent
of the maps is to develop evacuation procedures in case of
collapse of the dam, so the travel time of the flood wave
should be indicated on every significant habitation area
along the river channel.
EM 1110-2-1420
31 Oct 97
b. Evaluation of hazard potential. To assist in the
evaluation of hazard potential, areas delineated on inundation maps should be classified in accordance with the
degree of occupancy and hazard potential. The potential
for loss of life is affected by many factors, including but
not limited to the capacity and number of exit roads to
higher ground and available transportation. Hazard potential is greatest in urban areas. The evaluation of hazard
potential should be conservative because the extent of
inundation is usually difficult to delineate precisely.
c. Hazard potential for recreation areas. The hazard
potential for affected recreation areas varies greatly,
depending on the type of recreation offered, intensity of
use, communications facilities, and available transportation. The potential for loss of life may be increased
where recreationists are widely scattered over the area of
potential inundation because they would be difficult to
locate on short notice.
d. Industries and utilities. Many industries and utilities requiring substantial quantities of water are located on
or near rivers or streams. Flooding of these areas and
industries, in addition to causing the potential for loss of
life, can damage machinery, manufactured products, raw
materials and materials in process of manufacture, plus
interrupt essential community services.
e. Least hazard potential. Rural areas usually have
the least hazard potential. However, the potential for loss
of life exists, and damage to large areas of intensely cultivated agricultural land can cause high economic loss.
Evacuation plans.
(1) Evacuation plans should be prepared and implemented by the local jurisdiction controlling inundation
areas. The assistance of local civil defense personnel, if
available, should be requested in preparation of the evacuation plan. State and local law enforcement agencies
usually will be responsible for the execution of much of the
plan and should be represented in the planning effort. State
and local laws and ordinances may require that other state,
county, and local government agencies have a role in the
preparation, review, approval, or execution of the plan.
Before finalization, a copy of the plan should be furnished
to the dam agency or owner for information and comment.
(2) Evacuation plans will vary in complexity in accordance with the type and degree of occupancy in the
potentially affected area.
The plans may include
delineation of the area to be evacuated; routes to be used;
traffic control measures; shelter; methods of providing
emergency transportation; special procedures for the evacuation and care of people from institutions such as hospitals, nursing homes, and prisons; procedures for securing
the perimeter and for interior security of the area; procedures for the lifting of the evacuation order and reentry to
the area; and details indicating which organizations are
responsible for specific functions and for furnishing the
materials, equipment, and personnel resources required.
HEC Research Documents 19 and 20 provide example
emergency plans and evacuation plans, respectively (HEC
1983a and b).
EM 1110-2-1420
31 Oct 97
Chapter 17
Channel Capacity Studies
17-1. Introduction
a. General. Channel capacity studies tend to focus
on high flows. Flood operations for a reservoir will require
operational downstream targets for nondamaging flows
when excess water must be released. Nondamaging
channel capacity may be defined at several locations, and
the target flow may be defined at several levels. There may
be lower targets for small flood events and, under extreme
flood situations, the nondamaging target may cause some
minor damage. Also, the nondamaging flow target may
vary seasonally and depend on floodplain land use.
b. Withstanding release rates. Channel capacity is
also concerned with the capability of the channel to withstand reservoir release rates. Of particular concern is the
reach immediately downstream from the reservoir. High
release rates for hydropower or flood control could damage
channel banks and cause local scour and channel
c. Channel capacity. While flood operation may
focus on maximum channel capacity, planning studies
usually require stage-discharge information over the entire
range of expected operations. Also, low-flow targets may
be concerned with maintaining minimum downstream flow
depth for navigation, recreation, or environmental goals.
Channel capacity studies typically provide information on
safe channel capacity and stage-discharge (rating) curves
for key locations.
17-2. Downstream Channel Capacity
a. Downstream channel erosion. Water flowing over
a spillway or through a sluiceway is capable of causing
severe erosion of the stream bed and banks below the dam.
Consequently, the dam and its appurtenant works must be
so designed that harmful erosion is minimized. The outlet
works for a dam usually require an energy-dissipating
structure. The design may vary from an elaborate multiplebasin arrangement to a simple head wall design, depending
on the number of conduits involved, the erosion resistance
of the exit channel bed material, and the duration, intensity,
and frequency of outlet flows. A stilling basin may be provided for outlet works when such downstream uses as
navigation, irrigation, and water supply, require frequent
operation or when the channel immediately downstream is
easily eroded. Sections 4-2b and 4-3j of EM 1110-2-3600
provide a general discussion of energy dissipators for
spillways and outlet works, respectively.
b. Adequate capacity. The channel downstream
should have adequate capacity to carry most flows from
reservoir releases. After the water has lost most of its
energy in the energy-dissipating devices, it is usually
transported downstream through the natural channel to its
destination points. With the expected release rates, the
channel should be able to resist excessive erosion and
scour, and have a large enough capacity to prevent downstream flooding except during large floods.
c. River surveys. River surveys of various types
provide the basic physical information on which river
engineering planning and design are based. Survey data
include information on the horizontal configuration (planform) of streams; characteristics of the cross sections
(channel and overbank); stream slope; bed and bank materials; water discharge; sediment characteristics and discharge; water quality; and natural and cultural resources.
d. Evaluating bank stability. To evaluate bank
stability, it is essential to understand the complex historical
pattern of channel migration and bank recession of the
stream and the relationship of channel changes to streamflow. Studies of bank caving, based on survey data and
aerial photographs, provide information on the progressively shifting alignment of a stream and are basic to laying
out a rectified channel alignment. The concepts and
evaluation procedures presented in “Stability of Flood
Control Channels” (USACE 1990) are applicable to the
channel capacity evaluation.
e. Interrupted sediment flow. A dam and reservoir
project tends to interrupt the flow of sediment, which can
have a significant impact on the downstream channel
capacity. If the project is relatively new, the affect may not
be seen by evaluating historic information or current
channel conditions. The future channel capacity will
depend on the long-term trends in aggradation and degradation along the river. General concepts on sediment
analysis are presented in Chapter 9. Sediment Investigations of Rivers and Reservoirs, EM 1110-2-4000, is the
primary reference for defining potential problems and
analyses procedures.
f. Downstream floodplain land use. Channel capacity also depends on the long-term trends in downstream
floodplain land use. While it is not a hydrologic problem,
channel capacity studies should recognize the impact of
floodplain encroachments on what is considered the
nondamaging channel capacity. Anecdotal history has
EM 1110-2-1420
31 Oct 97
shown that many Corps’ projects are not able to make
planned channel-capacity releases due to development and
encroachments downstream.
deviations which increase with time, but a flood can flush
away sediment and aquatic weed and cause a sudden
reversal of the rating curve shift.
17-3. Stream Rating Curve
e. Flow magnitude and bed material. Stream bed
configuration and roughness in alluvial channels are a
function of the flow magnitude and bed material. Bed
forms range from ripples and dunes in the lower regime
(Froude number y 1.0) to a smooth plane bed, to antidunes
with standing waves (bed and water surface waves in
phase) and with breaking waves and, finally, to a series of
alternative chutes and pools in the upper regime as the
Froude number increases.
a. Stage-discharge relationship. The relationship
between stage and discharge, the “rating” at a gauging station, is based on field measurements with a curve fitted to
plotted data of stage versus discharge. For subcritical flow,
the stage-discharge relationship is controlled by the stream
reach downstream of the gauge; for supercritical flow, the
control is upstream of the gauge. The stage-discharge
relationship is closely tied to the rate of change of
discharge with time, and the rating curve for a rising stage
can be different from that for the falling stage in alluvial
b. Tailwater rating curve. The tailwater rating curve,
which gives the stage-discharge relationship of the natural
stream below the dam, is dependent on the natural conditions along the stream and ordinarily cannot be altered by
the spillway design or by the release characteristics.
Degradation or aggradation of the river below the dam,
which will affect the ultimate stage-discharge conditions,
must be recognized in selecting the tailwater rating curves
to be used for design. Usually, river flows which approach
the maximum design discharges have never occurred, and
an estimate of the tailwater rating curve must either be
extrapolated from known conditions or computed on the
basis of assumed or empirical criteria. Thus, the tailwater
rating curve at best is only approximate, and factors of
safety to compensate for variations in tailwater must be
included in dependent designs.
c. Extrapolation. Extrapolation of rating curves is
necessary when a water level is recorded below the lowest
or above the highest gauged level. Where the cross section
is stable, a simple method is to extend the stage-area and
stage-velocity curve and, for given stage values, take the
product of velocity and cross-section area to give discharge
values beyond the stage values that have been gauged.
Generally, water-surface profiles should be computed to
develop the rating beyond the range of observed data.
d. Rating curve shifts. The stage-discharge relationship can vary with time, in response to degradation,
aggradation, or a change in channel shape at the control
section, deposition of sediment causing increased approach
velocities in a weir pond, vegetation growth, or ice
accumulation. Shifts in rating curves are best detected
from regular gauging and become evident when several
gaugings deviate from the established curve. Sediment
accumulation or vegetation growth at the control will cause
f. Upper and lower rating portions. The large changes in resistance to flow that occur as a result of changing
bed roughness affect the stage-discharge relationship. The
upper portion of the rating is relatively stable if it represents the upper regime (plane-bed, transition, standing
wave, or antidune regime) of bed form. The lower portion
of the rating is usually in the dune regime, and the stagedischarge relationship varies almost randomly with time.
Continuous definition of the stage-discharge relationship at
low flow is a very difficult problem, and a mean curve for
the lower regime is frequently used for gauges with shifting
g. Break up of surface material. In gravel-bed
rivers, a flood may break up the armoring of the surface
gravel material, leading to general degradation until a new
armoring layer becomes established and ratings tend to
shift between states of quasi-equilibrium. It may then be
possible to shift the rating curve up or down by the change
in the mean-bed level, as indicated by plots of stage and
bed level versus time.
h. Ice. Ice at the control section may also affect the
normal stage-discharge relationship. Ice effects vary with
the quantity and the type of ice (surface ice, frazil ice, or
anchor ice). When ice forms a jam in the channel and
submerges the control or collects in sufficient amounts
between the control and the gauge to increase resistance to
flow, the stage-discharge relationship is affected; however,
ice may form so gradually that there is little indication of
its initial effects. Surface ice is the most common form and
affects station ratings more frequently than frazil ice or
anchor ice. The major effect of ice on a rating curve is due
to backwater and may vary from day to day.
17-4. Water Surface Profiles
a. Appropriate methods. For most channel-capacity
studies, water surface profiles will be computed to develop
the required information. Given the technical concerns
EM 1110-2-1420
31 Oct 97
described in the preceding section on rating curves, the
selection of the appropriate method requires some
evaluation of the physical system and the expected use of
the information. The modeling methods are described in
Chapter 8 and are presented in EM 1110-2-1416. While
steady-flow water surface profiles are used in a majority of
profile calculations, the unsteady flow aspects of reservoir
operation or the long-term effects of changes in sediment
transport may require the application of methods that
capture those aspects.
Appendix D, “River Modeling - Lessons
Learned” (EM 1110-2-1416), provides an overview of
technical issues and modeling impacts that apply to profile
calculations. Stability of Flood Control Channels (USACE
1990) provides case examples of stream stability problems,
causes, and effects. While the focus is not on reservoirs,
the experience reflects the high flow conditions that are a
major concern with reservoir operation. And EM 1110-24000 provides procedures for problem assessment and
modeling. All of these documents should be reviewed prior
to formulating and performing technical studies.
b. Further information. The Corps, and other agencies, have accumulated considerable experience with river
EM 1110-2-1420
31 Oct 97
Chapter 18
Real Estate and Right-of-Way Studies
18-1. Introduction
a. General. This chapter provides guidance on the
application of hydrologic engineering principles to determine real estate acquisition requirements for reservoir
projects. Topics include selection of the analysis method,
potential problems, evaluation criteria, and references
associated with the acquisition of real estate for reservoir
projects developed by the Corps of Engineers.
b. Related documents. Real estate reporting requirements associated with feasibility reports, General Design
Memoranda, and Real Estate Design Memoranda are set
forth in ER 405-1-12. Real estate reporting requirements
associated with the acquisition of lands downstream from
spillways are set forth in ER 1110-2-1451, paragraph 9.
18-2. Definition of Terms
A list of terms and definitions used in this chapter is as
a. Project design sediment. The volume and distribution of sediment deposited in a reservoir over the life of
the project.
b. Land acquisition flood.
A hypothetical or
recorded flood event used to determine requirements for
real estate acquisition.
c. Full pool. The maximum reservoir elevation for
storing water for allocated project purposes.
d. Induced surcharge. Storage created in a reservoir
above the top of flood control pool by regulating outflows
during flood events.
e. Envelope curve. A curve which connects the high
points of intersection of preproject and postproject watersurface profiles.
f. Guide taking line. A contour line used as a guide
for land acquisition in the reservoir area. (Also referred to
as the guide contour line or guide acquisition line.)
18-3. Real Estate Acquisition Policies for
a. Basic policies. Basic policies and procedures
related to the acquisition of lands for reservoirs are presented in ER 405-1-12. Paragraph 2-12 of ER 405-1-12
states that, “Under the Joint Policy the Corps will take an
adequate interest in lands, including areas required for
public access, to accomplish all the authorized purposes of
the project and thereby obtain maximum public benefits
therefrom.” Paragraph 2-12.a(2) further states that land to
be acquired in fee shall include, “lands below a guide
contour line...established with a reasonable freeboard
allowance above the top pool elevation for storing water for
flood control, navigation, power, irrigation, and other
purposes, referred to in this paragraph as “full pool” elevation.
In nonurban areas generally, this freeboard
allowance will be established to include allowances for
induced surcharge operations plus a reasonable additional
freeboard to provide for adverse effects of saturation, wave
action and bank erosion.”
b. Considering factors. Factors such as estimated
frequency of occurrence, probable accuracy of estimates,
and relocation costs will be taken into consideration.
Where freeboard does not provide a minimum of 300 ft
horizontally from the conservation pool, defined as the top
of all planned storage not devoted exclusively to floodcontrol storage, then the guide acquisition line will be
increased to that extent. In the vicinity of urban communities or other areas of highly concentrated developments, the total freeboard allowance between the full pool
elevation and the acquisition line may be greater than
prescribed for nonurban areas generally. Also, there should
be sufficient distance to assure that major hazards to life or
unusually severe property damages would not result from
floods up to the magnitude of the SPF. In such
circumstances, however, consideration may be given to
easements rather than fee acquisition for select sections if
found to be in the public interest. However, when the
project design provides a high level spillway, the crest of
which for economy of construction is considerably higher
than the storage elevation required to regulate the reservoir
design flood, the upper level of fee acquisition will
normally be at least equal to the top elevation of spillway
gates or crest elevation of ungated spillway, and may
exceed this elevation if necessary to conform with other
criteria prescribed.
EM 1110-2-1420
31 Oct 97
18-4. Hydrologic Evaluations
a. Development of land acquisition flood.
establish a reasonable surcharge allowance above the top
pool elevation, a land acquisition flood, which includes the
effects of any upstream reservoirs, should be selected and
routed through the project to determine the impact on the
establishment of the guide acquisition line.
b. Nonurban areas. In nonurban areas, the land
acquisition flood should be selected from an evaluation of a
range of floods with various frequencies of occurrence.
The impact of induced surcharge operations on existing
and future developments, hazards to life, land use, and
relocations must be evaluated. The land acquisition flood
will be chosen based on an evaluation of the risk and
uncertainty associated with each of these frequency events.
Basic considerations to be addressed during the land
acquisition flood selection process should include the
credibility of the analysis, identification and significance of
risk, costs and benefits, and legal, social, and political
c. Urban areas. In urban areas or other areas with
highly concentrated areas of development, the SPF will be
used for the land acquisition flood.
d. Project design sedimentation volume. Project
capacity data should be adjusted for projected sediment
volumes when routing the land acquisition flood. Project
design sediment should be based on appropriate rates of
sedimentation for the project area for the life of the project.
18-5. Water Surface Profile Computations
a. Backwater model development. A basic backwater
model should be developed for the project area from the
proposed flat pool area through the headwater area where
impacts of the proposed reservoir are expected to be
significant. The model should reflect appropriate crosssectional data and include parameters based on historical
flood discharges and high water marks. EM 1110-2-1416
presents the model requirements and calibration
b. Preproject profiles. A series of preproject water
surface profiles should be developed utilizing preproject
cross-section geometry, calibrated Manning's “n” values,
and appropriate starting water surface elevations for the
initial cross section. Flow rates used in the water surface
profile computations should be selected from the peak and
recession side of the land acquisition flood hydrograph.
c. Postproject profiles. A series of water surface
profiles shall be developed utilizing the postproject cross
sections which are adjusted to reflect project design
sedimentation over the life of the project. Manning's
roughness coefficients are based on adjusted preproject
roughness coefficients to account for factors such as vegetation and land use changes which decrease hydraulic conveyance. Agricultural lands existing in the headwater areas
prior to land purchases will likely revert to forested areas
some years after the reservoir is filled. Preproject flow
rates and coincident reservoir pool elevations from land
acquisition flood routing should be used to compute
postproject profiles.
d. Project design sedimentation distribution. Postproject cross-section geometry must be adjusted to reflect
the impacts of sedimentation over the life of the project.
Sedimentation problems associated with reservoir projects
and methods of analysis to address sediment volumes and
distributions are given in Chapter 5 of EM 1110-2-4000.
18-6. Development of an Envelope Curve
The development of an envelope curve is based on preproject and postproject water-surface profiles. A selected
discharge from the land acquisition flood is used to compute a preproject and a postproject profile. A point of
intersection is established where the profiles are within 1 ft
of each other. The point of intersection is placed at the
elevation of the higher of the two profiles. A series of
points of intersection are derived from water-surface
profile computations utilizing a range of selected discharges from the land acquisition flood. A curve is drawn
through the series of points of intersection to establish the
envelope curve.
18-7. Evaluations to Determine Guide Taking
Lines (GTL)
a. Land acquisition flat pool. The land acquisition
flat pool of a reservoir project is established by the maximum pool elevation designated for storing water for allocated project purposes to include induced surcharge storage
and is not impacted by the backwater effects of main
stream or tributary inflows. In flat pool areas, the elevations of the GTL are based on the flat pool elevation and a
freeboard allowance to account for adverse effects of
saturation, bank erosion, and wave action.
b. Headwater areas. In headwater areas, the GTL
may be based on the envelope curve elevations and
appropriate allowances to prevent damages associated with
saturation, bank erosion, and wave action.
EM 1110-2-1420
31 Oct 97
c. Flood-control projects.
The selection of an
appropriate land acquisition flood for flood-control projects
located in rural areas should be based on an elevation of a
range of frequency flood events. The land acquisition
flood selection for flood-control projects in rural locations
must include regulation by upstream reservoirs and reflect
postproject conditions which minimize adverse impacts
within the project area resulting from induced flood
elevations and duration of flooding. In highly developed
areas along the perimeter of flood-control projects, the SPF
should be used for land acquisition. An envelope curve can
be developed from the land acquisition flood routings and
water-surface profile computations for preproject and
postproject conditions. The land acquisition GTL may be
established from the envelope curve and appropriate
allowances for reservoir disturbances.
d. Nonflood-control projects.
Nonflood control
projects may be any combination of purposes such as water
supply, hydropower, recreation, navigation and irrigation.
The land acquisition flood selection process for nonfloodcontrol projects located in rural areas is based on an
evaluation of a range of frequency floods and is used to
determine postproject flood elevations and duration of
flooding in the project area. As with flood-control projects,
regulation of flows by upstream reservoirs must be
incorporated in the development process. The land
acquisition flood used to evaluate real estate acquisitions in
rural areas should reflect postproject conditions which
minimize adverse impacts. The land acquisition flood for
developed areas should be the SPF. The maximum pool
elevation designed for storing water for allocated project
purposes is used in the development of the land acquisition
flood routing. An envelope curve based on preproject and
postproject water-surface profiles utilizing project design
sedimentation and distribution should be developed. The
envelope curve and appropriate allowances for reservoir
disturbances may be used to establish the land acquisition
Acquisitions of Lands for Reservoir
Land acquisition policies of the Department of the Army
governing acquisition of land for reservoir projects is
published in ER 405-1-12, Change 6, dated 2 January 1979.
Paragraph is as follows:
Joint Land Acquisition Policy for Reservoir Projects.
The joint policies of the Department of the Interior and
Department of the Army, governing the acquisition of land
for reservoir projects, are published in the Federal
Register, dated 22 February 1962, Volume 27, page 1734.
On July 1966, the Joint Policy was again published in 31,
F.R. 9108, as follows:
“A joint policy statement of the Department of the Interior
and the Department of the Army was inadvertently issued
as a notice in 27 F.R. 1734. Publication should have been
made as a final rule replacing regulations then appearing
in 43 CFR part 8. The policy as it appears in 27 F.R. 1734
has been the policy of the Department of the Interior and
the Department of the Army since its publication as a
Notice and is now codified as set forth below.
Lands for reservoir construction and
Additional lands for correlative purposes
Blocking out
Mineral rights
Authority: The provisions of this Part 8
issued under Sec. 7, 32 Stat., 389, Sec. 14, 53
Stat. 1197, 43 U.S.C. 421, 389.
8.0 Acquisition of Lands for Reservoir Projects.
Insofar as permitted by law, it is the policy of the Departments of the Interior and of the Army to acquire, as a part
of reservoir project construction, adequate interest in
lands necessary for the realization of optimum values for
all purposes including additional land areas to assure full
realization of optimum present and future outdoor recreational and fish and wildlife potentials of each reservoir.
8.1 Lands for Reservoir Construction and Operation.
The fee title will be acquired to the following:
EM 1110-2-1420
31 Oct 97
a) Lands necessary for permanent structures.
b) Lands below the maximum flowage line of the
reservoir including lands below a selected freeboard where
necessary to safeguard against the effects of saturation,
wave action, and bank erosion and to permit induced
surcharge operation.
c) Lands needed to provide for public access to the
maximum flowage line, as described in Paragraph 1b, or
for operation and maintenance of the project.
8.2 Additional Lands for Correlative Purposes. The
fee title will be acquired for the following:
a) Such lands as are needed to meet present and future
requirements for fish and wildlife as are determined pur
suant to the Fish and Wildlife Coordination Act.
b) Such lands as are needed to meet present and future
public requirements for outdoor recreation, as may be
authorized by Congress.
8.3 Easements. Easements in lieu of fee title may be
taken only for lands that meet all of the following
a) Lands lying above the storage pool,
8.6 Buildings. Buildings for human occupancy as
well as other structures which would interfere with the
operation of the project for any project purpose will be
prohibited on reservoir project lands."
18-9. Acquisition of Lands Downstream from
Spillways for Hydrologic Safety Purposes
a. General. A real estate interest will be acquired
downstream of dam and lake projects to assure adequate
security for the general public in areas downstream from
spillways. Real estate interests must be obtained for
downstream areas where spillway discharges create or
significantly increase a hazardous condition.
b. Evaluation criteria.
Combinations of flood
events and flood conditions which result in a hazardous
condition or increase the hazard from the preproject to
postproject flood conditions are determined for areas
downstream from the spillway. These combinations of
flood events and flood conditions are identified as critical
c. Flood events and conditions. Flood events up to
the magnitude of the spillway design flood are evaluated
for preproject and postproject conditions for areas downstream from the spillway. Flood conditions to be analyzed
include flooded area, depth of flooding, duration,
velocities, debris, and erosion.
b) Lands in remote portions of the project area,
c) Lands determined to be of no substantial value for
protection or enhancement of fish and wildlife
resources, or for public outdoor recreation,
d. Hazardous and nonhazardous conditions. The
imposed critical conditions are analyzed to determine if
these conditions are hazardous or nonhazardous. Nonhazardous areas are characterized by the following criteria:
d) It is to the financial advantage of the Government to
take easements in lieu of fee title.
(1) Flood depths do not exceed 2 ft in urban and rural
8.4 Blocking Out. Blocking out will be accomplished
in accordance with sound real estate practices, for example, on minor sectional subdivision lines: and normally,
land will not be acquired to avoid severance damage if the
owner will waive such damage.
(2) Flood depths are essentially nondamaging to
urban property.
8.5 Mineral Rights. Mineral, oil and gas rights will
not be acquired except where the development thereof
would interfere with project purposes, but mineral rights
not acquired will be subordinated to the Government's
right to regulate their development in a manner that will
not interfere with the primary purposes of the project,
including public access.
(3) Flood durations do not exceed 3 hr in urban areas
and 24 hr in agricultural areas.
(4) Velocities do not exceed 4 fps.
(5) Debris and erosion potential are minimal.
(6) Imposed flood conditions would be infrequent.
The exceedance frequency should be less than 1 percent.
EM 1110-2-1420
31 Oct 97
Appendix A
EM 1110-2-1419
Hydrologic Engineering Requirements for Flood Damage
Reduction Studies
A-1. Required Publications
EM 1110-2-1602
Hydraulic Design of Reservoir Outlet Works
Flood Control Act of 1944.
EM 1110-2-1603
Hydraulic Design of Spillways
National Dam Safety Act, Public Law 92-367.
EM 1110-2-1701
Water Resources Development Act of 1976.
Water Supply Act of 1958.
EM 1110-2-2904
Design of Breakwater and Jetties
ER 405-1-12
Real Estate Handbook
EM 1110-2-3600
Management of Water Control Systems
ER 1110-2-240
Water Control Management
EM 1110-2-4000
Sedimentation Investigations of Rivers and Reservoirs
ER 1110-2-1451
Acquisition of Lands Downstream from Spillways for
Hydrologic Safety Purposes
ETL 1110-2-335
Development of Drought Contingency Plans
ER 1110-8-2(FR)
Inflow Design Floods for Dams and Reservoirs
ETL 1110-2-336
Operation of Reservoir Systems
EM 1110-2-1201
Reservoir Water Quality Analyses
A-2. Computer Program/Document References
Fread 1989
Fread, D. 1989. BREACH: An Erosion Model for Earthen
Dam Failures, User’s Manual, National Weather Service,
Hydrologic Research Laboratory, Silver Spring, MD.
EM 1110-2-1406
Runoff from Snowmelt
EM 1110-2-1411
Standard Project Flood Determinations
Hydrologic Engineering Center (HEC) 1971
Hydrologic Engineering Center (HEC). 1971. HEC-4
“Monthly streamflow simulation,” User's Manual,
U.S. Army Corps of Engineers, Davis, CA.
EM 1110-2-1414
Water Levels and Wave Heights for Coastal Engineering
HEC 1982c
HEC. 1982c. HEC-5, “Simulation of flood control and
conservation systems,” User's Manual, U.S. Army Corps of
Engineers, Davis, CA.
EM 1110-2-1412
Storm Surge Analysis
EM 1110-2-1415
Hydrologic Frequency Analysis
EM 1110-2-1416
River Hydraulics
HEC 1982d
HEC. 1982d. HYDUR, “Hydropower analysis using flowduration procedures,” User's Manual, U.S. Army Corps of
Engineers, Davis, CA.
EM 1110-2-1417
Flood Run-off Analysis
EM 1110-2-1420
31 Oct 97
HEC 1984
HEC. 1984. HMR52, “Probable maximum storm (eastern
United States),” User's Manual, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1986
HEC. 1986. HEC-5(Q), “Simulation of flood control and
conservation systems,” Appendix on Water Quality Analysis, U.S. Army Corps of Engineers, Davis, CA.
HEC 1987a
HEC. 1987a. STATS, “Statistical analysis of time series
data,” Input Description, U.S. Army Corps of Engineers,
Davis, CA.
Lane 1990
Lane. 1990. LAST, “Applied stochastic techniques,”
User’s Manual, U.S. Bureau of Reclamation, Denver, CO.
USACE 1972
USACE. 1972. SUPER, “Regulation simulation and
analysis of simulation for a multi-purpose reservoir system,” Southwestern Division, Dallas, TX.
USACE 1991
USACE. 1991. SSARR, “Model streamflow synthesis and
reservoir regulation,” User's Manual, North Pacific
Division, Portland, OR.
A-3. Related Publications
HEC 1990b
HEC. 1990b. “Flood damage analysis package on the
microcomputer,” Installation and User's Guide, Training
Document 31, U.S. Army Corps of Engineers, Davis, CA.
HEC 1990c
HEC. 1990c. HEC-1, “Flood hydrograph package,” User's
Manual, U.S. Army Corps of Engineers, Davis, CA.
HEC 1990d
HEC. 1990d. HEC-2, “Water surface profiles,” User's
Manual, U.S. Army Corps of Engineers, Davis, CA.
HEC 1991a
HEC. 1991a. HEC-PRM, “Prescriptive reservoir model,
program description,” User's Manual, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1992c
HEC. 1992c. HEC-FFA, “Flood flow frequency analysis,”
User's Manual, U.S. Army Corps of Engineers, Davis, CA.
HEC 1993
HEC. 1993. HEC-6, Scour and deposition in rivers and
reservoirs,” User's Manual, U.S. Army Corps of Engineers,
Davis, CA.
HEC 1995a
HEC. 1995a. HEC-DSS, “Users guide and utility program
manual,” U.S. Army Corps of Engineers, Davis, CA.
HEC 1995b
HEC. 199b. UNET, “One-dimensional unsteady flow
through a full network of open channels,” User's Manual,
U.S. Army Corps of Engineers, Davis, CA.
Alley and Burns 1983
Alley, W. M., and Burns, A. W. 1983. “Mixed station
extension of monthly streamflow records,” Journal of
Hydraulic Engineering, ASCE, HY109(10), October 12721284.
Bowen 1987
Bowen, T. H. 1987. “Branch-bound enumeration for
reservoir flood control plan selection,” Research Document
35, HEC, Davis, CA.
Buchanan and Somers 1968
Buchanan, T. J., and Somers, W. P. 1968. “Stage measurements at gauging stations,” U.S. Geological Survey,
TWI 3-A7.
Buchanan and Somers 1969
Buchanan, T. J., and Somers, W. P. 1969. “Discharge
measurements at gauging stations,” U.S. Geological Survey, TWI 3-A8.
Carter and Davidian 1968
Carter, R. W., and Davidian, J. 1968. “General procedure
for gauging streams,” U.S. Geological Survey, TWI 3-A6.
DeVries and Hromadka 1993
DeVries, J. J., and Hromadka, T. V. 1993. “Computer
models for surface water,” Handbook of hydrology D. R.
Maidment, ed., McGraw-Hill, New York.
Domenico and Schwartz 1990
Domenico, P. A., and Schwartz, F. W. 1990. Physical and
Chemical Hydrogeology. John Wiley and Sons, New York.
EM 1110-2-1420
31 Oct 97
Eichert and Davis 1976
Eichert, B. S., and Davis, D. W. 1976. “Sizing flood
control reservoir systems by systems analysis,” Technical
Paper 44, Hydrologic Engineering Center, Davis, CA.
HEC 1985b
HEC. 1985b. “Stochastic Analysis of Drought Phenomena,” Training Document 25, U.S. Army Corps of
Engineers, Davis, CA.
Farnsworth, Thompson, and Peck 1982
Farnsworth, R. K., Thompson, E. S., and Peck, E. L. 1982.
“Evaporation atlas for the contiguous 48 United States,”
Technical Report NWS 33, National Oceanic and
Atmospheric Administration (NOAA), Washington, DC.
HEC 1990a
HEC. 1990a. “Modifying reservoir operations to improve
capabilities for meeting water supply needs during
droughts,” Research Document 31, U.S. Army Corps of
Engineers, Davis, CA.
Froelich 1987
Froelich, D. C. 1987. “Embankment-dam breach parameters,” Proceedings of the 1987 National Conference on
Hydraulic Engineering at Williamsburg, VA. American
Society of Civil Engineers, New York.
HEC 1990e
HEC. 1990e. “A preliminary assessment of Corps of
Engineers' reservoirs, their purposes and susceptibility to
drought,” Research Document 33, U.S. Army Corps of
Engineers, Davis, CA.
Hoffman 1977
Hoffman, C. J. 1977. “Design of spillways and outlet
works,” Handbook of Dam Engineering A. R. Golzé, ed.,
Van Nostrand Reinhold, New York.
HEC 1991b
HEC. 1991b. “Optimization of multiple-purpose reservoir
system operations: A review of modeling and analysis
approaches,” Research Document 34, U.S. Army Corps of
Engineers, Davis, CA.
Hydrologic Engineering Center (HEC) 1977
Hydrologic Engineering Center (HEC). 1977. “Guidelines
for calculating and routing a dam break flood,” Research
Document 5, U.S. Army Corps of Engineers, Davis, CA.
HEC 1980
HEC. 1980. “Flood emergency plan guidelines for Corps
dams,” Research Document 13, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1982a
HEC. 1982a. “Emergency planning for dams, bibliography and abstracts of selected publications,” Research
Document 17, U.S. Army Corps of Engineers, Davis, CA.
HEC 1983a
HEC. 1983a. “Example emergency plan for Blue Marsh
Dam and Lake,” Research Document 19, U.S. Army Corps
of Engineers, Davis, CA.
HEC 1983b
HEC. 1983b. “Example plan for evacuation of Reading,
Pennsylvania, in the event of emergencies at Blue Marsh
Dam and Lake,” Research Document 20, U.S. Army Corps
of Engineers, Davis, CA.
HEC 1985a
HEC. 1985a. “Flood-damage-mitigation plan selection
with branch-and-bound enumeration,” Training Document
23, U.S. Army Corps of Engineers, Davis, CA.
HEC 1991c
HEC. 1991c. “Importance of surface-ground water interaction to Corps total water management: Regional and
national examples,” Research Document 32, U.S. Army
Corps of Engineers, Davis, CA.
HEC 1991d
HEC. 1991d. “Missouri River System Analysis Model Phase I,” Project Report 15, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1991f
HEC. 1991f. “Columbia River System Analysis Model Phase I,” Project Report 16, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1992a
HEC. 1992a. “Missouri River System Analysis Model Phase II,” Project Report 17, U.S. Army Corps of
Engineers, Davis, CA.
HEC 1992b
HEC. 1992b. “Developing Operation Plans from HEC
Prescriptive Reservoir Model, Results for the Missouri
Preliminary Results,” Project Report 18,
U.S. Army Corps of Engineers, Davis, CA.
EM 1110-2-1420
31 Oct 97
HEC 1993
HEC. 1993. “Review of GIS Applications in Hydrologic
Modeling,” Technical Paper 144, U.S. Army Corps of
Engineers, Davis, CA.
Pruitt 1990
Pruitt, W. O. 1990. Evaluation of Reservoir Evaporation
Estimates, submitted to USACE, Sacramento District,
Sacramento, CA.
Linsley et al. 1992
Linsley, R. K., Franzini, J. B., Freyberg, D. L., and
Tchobanoglous, G. 1992. Water resources engineering,
4th ed., McGraw-Hill, New York.
Salas 1992
Salas, J. D. 1992. “Analysis and modeling of hydrologic
time series,” Handbook of Hydrology, Maidment, ed.
MacDonald and Langridge-Monopolic 1984
MacDonald, T. C., and Langridge-Monopolis, J. 1984.
“Breaching characteristics of dam failures,” Journal of
Hydraulic Engineering, American Society of Civil Engineers, 110 (5), 567-586.
Matalas and Langbein 1962
Matalas, N. C., and Langbein, W. 1962. “Information
Content of the Mean,” in Journal of Geophysical Research,
67(9), 3441-3448.
Mays and Tung 1992
Mays, L. W., and Tung, Y-K. 1992. Hydrosystems engineering & management, McGraw-Hill, New York.
Salas, Delleur, Yevjevich, and Lane 1980
Salas, J. D., Delleur, J. W., Yevjevich, V., and Lane, W. L.
1980. “Applied modeling of hydrologic time series,” Water
Resources Publications, Littleton, CO, 484.
Smoot and Novak 1969
Smoot, G. F., and Novak, C. E. 1969. “Measurement of
discharge by moving-boat method,” U.S. Geological Survey, TWI 3-A11.
Tasker 1983
Tasker, G. D. 1983. “Effective record length of T-year
event,” Journal of Hydrology, 64, 39-47.
McGhee 1991
McGhee, T. J. 1991. Water supply and sewerage,
6th Edition, McGraw-Hill, New York.
Thomas and McAnally 1985
Thomas, W. A., and McAnally, W. H. 1985. “OpenChannel Flow and Seidmentation TABS2,” Instruction
Report HL-85-1, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.
Mosley and McKerchar 1993
Mosley, M. P., and McKerchar, A. I. 1993. “Streamflow,”
Handbook of Hydrology. D. R. Maidment ed., McGrawHill, New York.
USACE 1962
USACE. 1962. “Waves in Inland Reservoirs: Summary
Report on CWI Projects,” CW-164 and CW-165.
Washington, DC.
National Weather Service 1960
National Weather Service. 1960. “Generalized estimates
of PMP for the U.S. west of the 105th Meridian for areas
less than 400 square miles and durations to 24 hr,” Technical Paper 38, Silver Spring, MD.
USACE 1979
USACE. 1979. National hydroelectric power study,
Institute of Water Resources, Washington, DC.
National Weather Service 1977
National Weather Service. 1977. “Probable maximum
precipitation estimates,” United States east of 105th Meridian, Hydrometeorological Report No. 51, Silver Spring,
National Weather Service 1981
National Weather Service. 1981. “Application of probable
maximum precipitation estimates - United States east of the
105th meridian,” Hydrometeorological Report No. 52,
Silver Spring, MD.
USACE 1989
USACE. 1989. Digest of water resources policies and
authorities, 1165-2-1, Washington, DC.
USACE 1990
USACE. 1990. “Stability of flood control channels,” EC
1110-8-1(FR), Washington, DC.
USACE 1992
USACE. 1992. “Authorized and operating purposes of
Corps of Engineers' reservoirs,” Washington, DC.
EM 1110-2-1420
31 Oct 97
U.S. Geological Survey 1992
United States Geological Survey. 1992. “Sediment deposition in U.S. Reservoirs - summary of data reported 198185,” Interagency Advisory Committee on Water Data Subcommittee on Sedimentation.
Water Resources Council 1973
Water Resources Council. 1973. “Principles and standards
for planning water and related land resources,” The Federal
Register, The National Archives of the United States,
September 10, 1973.
Waide 1986
Waide, J. B. 1986. “General guidelines for monitoring
contaminants in reservoirs,” Instruction Report E-86-1,
U.S. Army Engineers Waterways Experiment Station,
Vicksburg, MS.
Wurbs 1991
Wurbs, R. A. 1991. “Optimization of multiple-purpose
reservoir system operations: A review of modeling and
analysis approaches,” Research Document 34, Hydrologic
Engineering Center, Davis, CA.
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

Related manuals

Download PDF