PHD_THESIS_MASIH-202pag-vs310511

PHD_THESIS_MASIH-202pag-vs310511
UNDERSTANDING
HYDROLOGICAL
VARIABILITY FOR
IMPROVED WATER
MANAGEMENT
in the semi-arid Karkheh Basin, Iran
ilyas masih
UNDERSTANDING HYDROLOGICAL
VARIABILITY FOR IMPROVED WATER
MANAGEMENT IN THE SEMI-ARID
KARKHEH BASIN, IRAN
UNDERSTANDING HYDROLOGICAL
VARIABILITY FOR IMPROVED WATER
MANAGEMENT IN THE SEMI-ARID
KARKHEH BASIN, IRAN
DISSERTATION
Submitted in fulfillment of the requirements of
the Board for Doctorates of Delft University of Technology
and of the Academic Board of the UNESCO-IHE
Institute for Water Education
for the Degree of DOCTOR
to be defended in public on
Tuesday, 21 June 2011 at 15:00 hours
in Delft, the Netherlands
by
Ilyas MASIH
Master of Philosophy in Water Resources Management,
Centre of Excellence in Water Resources Engineering, U.E.T.,
Lahore, Pakistan
born in Baddomalhi, District Narowal, Pakistan
This dissertation has been approved by the supervisor:
Prof. dr. S. Uhlenbrook
Committee members:
Chairman
Vice-chairman
Prof. dr. S. Uhlenbrook
Prof. dr. ir. H.H.G. Savenije
Prof. dr. ir. P. van der Zaag
Prof. dr. M. Karamouz
Prof. dr. W. Bauwens
Advisor: dr. S. Maskey
Prof. dr. ir. N.C. van de Giesen
Rector Magnificus TU Delft, the Netherlands
Rector UNESCO-IHE, Delft, the Netherlands
UNESCO-IHE/TU Delft, the Netherlands
TU Delft/UNESCO-IHE, the Netherlands
UNESCO-IHE/TU Delft, the Netherlands
University of Tehran, Iran
Vrije Universiteit Brussel, Belgium
UNESCO-IHE, Delft, the Netherlands
TU Delft, the Netherlands, reserve member
The research reported in this dissertation has been sponsored by International Water
Management Institute, Colombo, Sri Lanka.
CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa
business
© 2011, Ilyas Masih
All rights reserved. No part of this publication or the information contained herein
may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, by photocopying, recording or otherwise, without
written prior permission from the publisher.
Although care is taken to ensure the integrity and quality of this publication and the
information therein, no responsibility is assumed by the publishers nor the author for
any damage to property or persons as a result of operation or use of this publication
and/or the information contained herein.
Published by:
CRC Press/Balkema
PO Box 447, 2300 AK Leiden, The Netherlands
e-mail: [email protected]
www.crcpress.com – www.taylorandfrancis.co.uk – www.balkema.nl
ISBN 978-0-415-68981-6 (Taylor & Francis Group)
FOREWORD
The right atmosphere required to blaze the way for the desire of obtaining higher
education and more and more effectively contributing to the water-related issues was
put in my way when I joined the International Water Management Institute (IWMI)
as a Junior Hydrologist at IWMI, Lahore, Pakistan in 2001. IWMI’s mission to
improve the management of water and land resources for food, livelihoods and the
environment very much important and captivating and, very soon, I was fully
devoted to contribute to the achievement of this sublime mission. During the few
years of work at IWMI, I realized how crucial it was to improve the efficiency of
water use and raise the productivity of land and water resources to improve food
security, ease sectoral competition for water use and safeguard the environment.
Furthermore, I found that cross-disciplinary knowledge and understanding
constituted an important element of my personal and professional capabilities. At
IWMI, where diversity, team spirit and excellence are much appreciated, I learned
the importance of striving to attain the best quality of work in the designated area,
and tried to understand how small pieces of the puzzle fit together to complete a big
picture.
The personal discussion with many colleagues at IWMI, Lahore, Pakistan and
Colombo, Sri Lanka, further emphasized the importance of doing PhD studies to
better understand and contribute to the abovementioned issues. Dr. Mobin-ud-Din
Ahamad remained the cornerstone in this regard, especially because he significantly
motivated, guided and recommended me to do PhD studies. The series of
discussions with IWMI supervisors and management finally culminated in the form
of an offer of a PhD fellowship and to join the IWMI team in Iran on the Karkheh
Basin Focal Project (BFP) as a PhD researcher. Dr. Frank Rijsberman, former DG of
IWMI and former Professor at UNESCO-IHE Institute for Water Education, Delft,
the Netherlands, kindly agreed to be my promoter together with Prof. Stefan
Uhlenbrook, Professor of Hydrology at UNESCO-IHE, who also very kindly
accepted me as his PhD student.
The BFPs have been very important initiatives of the CGIAR Challenge Program
on Water and Food (CPWF), started in several basins worldwide, i.e., Andean, IndoGangetic, Karkheh, Limpopo, Mekong, Niger, Nile, Sao Francisco, Volta, and
Yellow River, with the main purpose of strengthening the basin focus of the CPWF
program. The main aims of the BFPs, including Karkheh BFP, were to provide more
comprehensive and integrated understanding of the water, food and environmental
issues in a basin; and to understand the extent and nature of poverty within each
selected basin and determine where water-related constraints are a major
determinant of the poverty factor and where those constraints can be addressed. The
adopted research framework was underpinned by the use of sound scientific
methods, interdisciplinary knowledge and rigorous research/evaluation
methodologies. The scientifically sound knowledge on hydrology and water
resources was a substantially important component of the Karkheh BFP beside other
vi
____________________________________________________________________
areas related to the assessment of water productivity, poverty, institutions and
policies. I was designated with the role of conducting a comprehensive assessment
of the surface water hydrology that is underpinning the sustainable management of
water for food, environment and poverty alleviation.
I strongly consider that the research documented in this thesis has significantly
contributed to achieving the project aims and objectives. Moreover, I view that this
piece of research is very relevant and beneficial for the hydrological and water
management community in Iran and worldwide. This thesis provides an example of
understanding issues in local and global contexts, wisely using and further
developing existing methods and (scarce) data sets, seeking for innovations to
overcome constraints of data, methods and information, and finally realizing the
need for having more knowledge and understanding of the variability of
hydrological processes and water availability and its proper inclusion in water
resources planning and management that envision the well-being of humans and
nature.
Ilyas Masih
UNESCO-IHE, Delft, the Netherlands
May 2011
ACKNOWLEDGEMENTS
I give honor and thanks to God, who is the source of knowledge and wisdom,
initiator of every good work and leads it to completion, for the provision of
necessary intellect and every other resource required for the successful execution of
this PhD study.
I am extremely grateful to my promoter Prof. Dr. Stefan Uhlenbrook for his wise
guidance, critical and innovative insights, wealth of broad knowledge and
understanding, and very strong commitment towards this study that played a pivotal
role in the success of this endeavor. I have learned a lot from him professionally as
well as personally, which has significantly improved my professional capabilities
and greatly enriched my life. His quest for advancing hydrological sciences and their
applications, and his dynamic and humble personality, kind and gentle behavior, and
ability to stimulate critical thinking and express divergent views in an appealing and
inoffensive way are some of the most notable virtues that will remain as exemplary
and worth following in the future.
I would like to extend my sincere thanks to my supervisors for all the help they
gave me and the guidance they kindly extended to me during this study. I owe many
thanks to Dr. Shreedhar Maskey for significantly contributing to this research and
admire his consistent encouragement to pursue excellence in every component of
this study. I exceedingly benefited from his understanding of hydrological modeling
and uncertainty assessment, critical thinking, technical writing ability and computer
programming skills.
Many thanks are due to my PhD supervisors from IWMI. Sincere gratitude is due
to Dr. Smakhtin for his overall contribution in this research, especially for providing
valuable professional insights, guiding in writing skills, kind and generous behavior,
and considerate response to all the professional and administrative issues, all of
which greatly helped in the successful completion of this study. I would also like to
thank my former PhD supervisors from IWMI, Dr. Mobin-ud-Din Ahmad, and Dr.
Frank Rijsberman, for making this research directly relevant to IWMI’s research
portfolio. Both of them had to step out of my PhD supervision, because they left
IWMI and joined other organizations. Their supervision and guidance greatly helped
achieve a good balance between advancement of hydrological sciences and the
practical use of it in the integrated water resources management.
During this study, I worked at IWMI offices in Lahore, Pakistan, Karaj, Iran and
Colombo, Sri Lanka, and at UNESCO-IHE, Delft, the Netherlands. I received
immeasurably great cooperation from many friends and colleagues at IWMI and
UNESCO-IHE. I extend my heartiest thanks to many friends and colleagues of
IWMI Pakistan for their support in one way or another, notably to Khalid
Mohtadullah, Abdul Hakeem Khan, Zhongping Zhou, Asghar Hussain, Aamir
Nazeer, Tabriz Ahmad, Moghis Ahmad, Pervaiz Ramzan, Riaz Wicky and Sidique
Akbar. The precious help from a number of people in settling down in Iran, Sri
Lanka and the Netherlands, discovering the culture and beauty of these impressive
viii
____________________________________________________________________
lands is an everlasting treasure for me and my family. I am thankful to Dr. Asad
Sarwar Qureshi for his mentoring and extending professional advice and practical
support provided during my stay at IWMI Iran and at IWMI Pakistan. Cooperation
from Poolad Karimi, in learning Persian and understanding key literature in Persian,
data collection, field visits, adjusting in Iran was remarkable and is highly
appreciated. I extend my sincere thanks to many colleagues at IWMI Colombo, Sri
Lanka who supported me and my family in many ways. Special thanks are due to
Dr. Hugh Turral, Dr. Mark Giordano, Dr. Francis Gichuki, Gamage Nilantha, Lal
Muthuwatta, Amin-ul- Islam, Mir Matin and Poorna de Silva.
I am thankful for the ever available support and kind interaction with many
UNESCO-IHE colleagues. Thanks are due to Susan Graas and Marloes Mul for
translating propositions and summary into Dutch. Many thanks to Ali Dastgheib for
his urgent help in translating summary into Persian. Dr. Ir. Ann van Griensven is
acknowledged for her helpful discussions on the SWAT model application. Special
thanks to Sylvia van Opdorp-Stijlen, Maria Laura Sorrentino, and Mariëlle van
Erven for their cooperation on logistics and counseling issues. I value and appreciate
the time spent with many MSc and PhD participants at UNESCO-IHE, and would
like to particularly appreciate time spent with Sarfraz Munir, which provided me an
excellent opportunity to share my feelings and concerns more openly with someone
from my own country, Pakistan. The interaction with many people from various
cultures and nationalities at UNESCO-IHE was a unique and highly enriching
experience, which has brought added respect for the diversity and difference of
opinion and cultures in my life. The friendship with a few Dutch families further
helped feel more at home and reduced the agony of missing family and close friends
back home. Sincere thanks to Marcel van Genderen and his family, Henk Jansen and
his family, Prof. Bill Rosen and many other people from the IREF Church, Delft, for
sharing with us in the times of our joys and sorrows and helping us in better
understanding the Dutch and European societies.
Funds for this research were generously provided by IWMI through its CapacityBuilding Program and from PN 57 ‘Basin Focal Project for the Karkheh,’ a project
of the CGIAR Challenge Program on Water and Food, implemented by IWMI in
collaboration with several Iranian partners. Additional funds were made available by
UNESCO-IHE for supporting me for a few months of additional stay at UNESCOIHE and for a conference attendance. These funding institutes and their donors are
gratefully acknowledged. Special thanks are due to David van Eyck, IWMI
Capacity-Building officer, for his gentle and highly professional attitude in
executing administrative and financial issues, which greatly facilitated the progress
of this study. Similarly cooperation from Ms. Jolanda Boots, PhD Fellowship
Officer, at UNESCO-IHE was remarkable and is sincerely acknowledged. Her
guidance on social and logistics issues is also highly appreciated.
Main data sets used in this study were accumulated from IWMI data
management program, for which kind cooperation of the data management team at
IWMI Colombo, Sri Lanka is highly appreciated. Thanks are also due to staff
members from IWMI Iran and Sri Lanka offices who exerted every effort in
collecting these data sets from the primary and secondary sources. Special gratitude
ix
____________________________________________________________________
is due to the Ministry of Energy and the Meteorological Organization, Iran, for the
provision of necessary data sets on hydrology and climate. Thanks are also due to
the Ministry of Jihad-e-Agriculture, in particular to the Agricultural Research and
Extension Organization of Iran (AREO) and Soil Conservation and Watershed
Management Research Institute (SCWMRI) for their support during data collection,
field visits, and providing useful insights on the research issues and results.
I would like to specially thank my parents, wife Huma Ilyas and daughter Sarah
Ilyas for their deep love, good care and earnest prayers for me, which provided me
with the necessary support, comfort and energy to successfully complete this
challenging venture. The sweet company of Huma and Sarah made this tough
journey a very pleasant and memorable experience of my life to the extent that I feel
very happy to dedicate this work to the two of them.
SUMMARY
The escalating growth of water resources utilization for human purposes,
particularly agriculture, is mounting increasing pressure on freshwater resources.
Although the human appropriation of water has helped mankind in many ways such
as improving food production and socioeconomic well-being, it has also caused
damages to the environment and its related services. Balancing water uses for
humans and nature is seen as the major challenge of this century. This issue is far
more complex for the semi-arid to arid regions of the world, like the Islamic
Republic of Iran, where water is generally scarce and demands from agriculture,
industry, urbanization and the growing population are rapidly swelling. The high
climatic variability and expected ongoing climate change further add to the pressing
issues.
Under the condition of water scarcity and competing water uses, improved
knowledge of basin-wide hydrology and resource availability are pivotal to instruct
informed policy formulation and sustainable development of the water sector. This
study is carried out in semi-arid to arid Karkheh Basin of Iran, where massive water
allocation planning is on the way, but a comprehensive knowledge on basin
hydrology and impact of these developments on different water uses and users
across the basin are lacking. The main objective of this research is to provide a
hydrology-based assessment of (surface) water resources of the Karkheh Basin and
study its continuum of variability and change at different spatio-temporal scales. The
methodological framework used in this study was underpinned by the combined use
of rigorous system investigation and hydrological modeling techniques. The spatial
investigations were carried out at the levels of the river basin, catchment (subbasin)
and subcatchment whereas the temporal resolutions were daily, monthly, annually
and in long-term time series.
The comprehensive assessment of spatio-temporal variability of surface water
hydrology was carried out by using long-term daily streamflow data available for the
period 1961 to 2001 for seven important gauging stations located at the Karkheh
River and its major tributaries. The analysis was carried out applying techniques,
such as measure of central tendency and dispersion, base flow separation and flow
duration analysis. Additionally, basin-level water accounting was done for the year
1993-94, for which requisite data sets were available.
The study shows that the hydrology of the Karkheh Basin has high inter- and
intra-annual variability, mainly driven by high spatio-temporal variability of climate
and spatially diverse soil, land use and hydrogeological characteristics of its
drainage area most of which is part of the Zagros mountains. The increase in the
streamflows starts in October and lasts till April. Peak flows are normally observed
during March-April, but flooding may occur any time between November and April.
These high flows are caused by the combined effect of snowmelt and rainfall. The
period May through September represent low flows mainly replenished from the
base flow contribution from subsurface storages. Moreover, the runoff regime of the
xii
____________________________________________________________________
middle part of the basin (Kashkan River) is notably different from the upper parts
(Gamasiab and Qarasou), with the former showing more runoff per unit area and
comparatively higher base flow contributions. The issue of variability is
substantiated here by the estimates of mean annual flow and its variability for the
Karkheh River gauged at the Paye Pole stations (just downstream of the Karkheh
Dam). The mean discharge at this location is 5.83 × 109 m3/yr., whereas the annual
flow was just about one-third (1.916 × 109 m3/yr.) in the extremely dry year 19992000 and as high as 12.60 × 109 m3/yr. during the highly wet year 1968-69. Under
such highly variable conditions, the understanding of the reliability of the water
availability becomes more meaningful for better resources use and allocation
decisions. The flow duration analysis conducted in this study provides such
estimates of streamflow reliability for the Karkheh Basin at daily, monthly and
annual time resolutions.
The synthesis of the results on hydrological variability, water availability, and
water accounting suggests that the Karkheh Basin was an open basin during the
study period (1961-2001), and there is further room for water resources allocations,
i.e., in the range of 1-4 × 109 m3/yr. depending on the amount of water left for
environmental flows. However, the allocation should be done after a careful impact
assessment and trade-off analysis for multiple and highly competing uses and users
across the basin. The evaluation of ongoing water allocation planning appeared as
nonsustainable given the limitations of resources availability and its high variability.
If the current water policy is implemented the basin will soon approach the closure
stage in the near future (latest by 2025), and then, meeting demands of all users will
be extremely difficult, especially during low flow months and dry years. The
environmental sector is likely to suffer the most which, so far, has been given low
priority, but other sectors such as agriculture and domestic uses are also likely to
face reductions in their allocated water rights.
The changes in the hydro-climatic variables and their linkages were also
explored as part of the system analysis. Streamflow records from five mainstream
stations were used for the period 1961-2001 to examine trends in a number of
streamflow variables representing a range of flow variability, i.e., mean annual and
monthly flows, 1 and 7 days maximum and minimum flows, timing of the 1-day
maxima and minima, and the number and duration of high- and low-flow pulses.
Similarly, the precipitation and temperature data from six synoptic climate stations
were used for the period 1950s to 2003 to examine trends in climatic variables and
their correlation with the streamflow. The Spearman rank test was used for the
detection of trends, and the correlation analysis was based on the Pearson method.
The results revealed a number of significant trends in streamflow variables both
increasing and decreasing. Moreover, the observed trends were not spatially
uniform. The decline in low-flow characteristics were more significant in the upper
parts of the basin (particularly for Qarasou River), whereas increasing trends in high
flows and winter flows were noteworthy in the middle parts of the basin (Kashkan
River). Most of these trends were mainly attributed to precipitation changes. The
results showed that the decline in April and May precipitation caused decline in the
low flows while increase in winter (particularly March) precipitation coupled with
xiii
____________________________________________________________________
temperature changes led to an increase in the flood regime. The observed trends at
the Jelogir station on the Karkheh River reflect the combined effect of the upstream
catchments. The significant trends observed for the number of streamflow variables
at Jelogir, e.g., 1-day maximum, December flow and low pulse count and duration,
indicated alterations of the hydrological regime of the Karkheh River and were
mainly attributed to the changes in the climatic variables.
Regionalization of hydrological parameters emerged as an important issue for the
Karkheh Basin because streamflow records were not available for many
subcatchments, and many streamflow gauging stations were abandoned. A new
regionalization method was developed in this study to estimate streamflow time
series for poorly gauged catchments. The proposed method is based on the
regionalization of a conceptual rainfall-runoff model based on the similarity of flow
duration curves (FDC). The performance of this method was compared with three
other methods based on drainage area, spatial proximity and catchment
characteristics. The data of 11 gauged catchments (475 to 2,522 km2) were used to
develop the regionalization procedures. The widely used HBV model was applied to
simulate daily streamflow with parameters transferred from gauged catchment
counterparts. The study indicated that transferring HBV model parameters based on
the FDC similarity criterion produced better runoff simulation compared to the other
three methods. Furthermore, it was demonstrated that the parameter uncertainty of
the model has little impact on the regionalization outcome. The results of this novel
method compared very well with most of the promising regionalization techniques
developed and applied elsewhere. Therefore, the FDC-based model regionalization
method developed in this study is a valuable addition to existing regionalization
methods. The proposed method is easy to replicate in other river basins, particularly
those facing a declining streamflow network.
Furthermore, a semi-distributed, process-based model – Soil Water Assessment
Tool (SWAT) – was used to understand and quantify the hydrological fluxes, and to
test different scenarios. It was recognized that the widely used SWAT model offers a
range of possibilities for defining the model structure, but the input of climatic data
is still rather simplistic. SWAT uses the data of a precipitation gauge nearest to the
centroid of each subcatchment as an input for that subcatchment. This may not
represent overall catchment precipitation conditions well, and may lead to increased
uncertainty in the modeling results. In this study, an alternative method for
precipitation input was evaluated. In particular, the input of interpolated areal
precipitation was tested against the standard SWAT precipitation input procedure.
The extent of the modeling domain was 42,620 km2, located in the mountainous
semi-arid part of the study basin, from where almost all of the basin’s runoff is
generated. The model performance was evaluated at daily, monthly and annual
scales using a number of performance indicators at 15 streamflow gauging stations,
each draining an area in the range of 590-42,620 km2. The comparison suggested
that the use of areal precipitation improves model performance particularly in small
subcatchments with drainage area in the range of 600-1,600 km2. The areal
precipitation input results in increased reliability of simulated streamflows in the
areas of low rain gauge density and poor spatial distribution of the rain gauge(s).
xiv
____________________________________________________________________
Both precipitation input methods result in reasonably good simulations for larger
catchments (over 5,000 km2), which was attributed to the averaging out effect of
precipitation at larger spatial resolution.
The understanding of catchment hydrology through the abovementioned studies,
field visits and literature review, and rigorous parameter estimation procedures
helped achieve reasonably good calibration, validation and uncertainty analysis of
the SWAT model for the Karkheh River Basin. This provided adequate confidence
for using the SWAT model for the analysis of water use scenarios in the basin. Three
scenarios, related to increased water use in rain-fed agriculture, were evaluated. The
tested scenarios are: upgrading rain-fed areas to irrigated agriculture (S1), improving
soil water availability through rainwater harvesting (S2), and a combination of S1
and S2 (S3). The results of these scenarios were compared against the baseline case
over the study period 1988-2000. The baseline simulations were carried out using
the finally adopted model structure and a parameter set obtained from the used
calibration procedure. The results of the first scenario (S1) indicated a reduction of
10% in the mean annual flows at the basin level, which ranged from 8 to 15% across
the main catchments across the basin. The reductions in the mean monthly flows
were in the range of 3-56% at the basin level. The months of May-July sustained
high impacts, with June witnessing the highest percentage of flow reductions. Flow
reductions in these months were more alarming in the upper parts of the basin which
was mainly attributed to relatively higher potential of developing rain-fed area to the
irrigation, coupled with comparatively lower amounts of runoff available in these
months. The impacts of S2 were generally small at the catchment as well as basin
scale, with reductions in the range of 2-5% and 1-10% in the mean annual and mean
monthly flows, respectively. The estimated flow reductions at the annual scale
remain well within the available water resources development potential in the basin.
However, avoiding excessive flow reductions in May-July would require adoption of
additional measures, such as practicing supplementary irrigation and augmenting
supplies through developing a range of water storage options, and considering less
than the potential rain-fed area for upgrading to irrigated farmland (particularly in
upper parts of the basin).
The study concludes that understanding of the prevalent high level of variability
in hydrology and water resources, a sound foundation of which has been laid by this
study, and inclusion of a range of variability of the water resources into planning and
management does play a pivotal role in the sustainable use and management of
available water resources of the Karkheh Basin. The ongoing water allocation
planning is not sustainable and a thorough revision of it is recommended, which will
essentially require the reduction in water allocations to human uses (particularly
agriculture) and leaving more water for the environment. The climate variability and
change have significantly altered the hydrological regime of the Karkheh River
system, warranting immediate mitigation efforts, i.e., structural measures and
programs to reverse catchment degradation to manage intensified flood regime in the
middle parts of the basin and considering how to reduce water withdrawals during
low-flow months (May to September) in upper parts of the basin in order to mitigate
the impacts of declining low flows in these areas. The impact evaluation study
xv
____________________________________________________________________
conducted herein have shown that the improving water use in rain-fed agriculture
could be promoted in the basin, with consideration of in-situ soil and water
conservation interventions all across the basin as they pose minimal impacts on
downstream water availability. However, the conversion of rain-fed areas to
irrigation requires a cautious approach to ensure reasonable levels of flow reduction
on monthly time resolution, which calls for upgrading limited rain-fed areas to
irrigation (particularly in upper parts of the basin), practicing supplementary
irrigation and developing a range of water storage options. Strengthening hydroclimatic data- monitoring networks is recommended to improve available data and
consequent application of hydrological and water management models for more
informed decision-making processes. In this regard, rehabilitation of abandoned
hydro-climatic stations and consideration of installation of more monitoring stations
in the mountainous parts are recommended. Planning and managing all water
resources in the river basin context should be promoted in the study basin.
In general, the knowledge generated in this case study is very much relevant for
other river basins of Iran, and worldwide.
TABLE OF CONTENTS
FOREWORD ............................................................................ V
ACKNOWLEDGEMENTS ...................................................... VII
SUMMARY .............................................................................. XI
1.
INTRODUCTION ............................................................. 1
1.1. Background ................................................................................................... 1
1.1.1.
Increasing pressure on earth’s water resources ................................. 1
1.1.2.
Adapting sustainable solutions ............................................................ 3
1.1.3.
Managing water by river basin............................................................ 4
1.1.4.
Need for a hydrological synthesis ........................................................ 5
1.2.
Hydrological and Water Management Issues in the Karkheh Basin, Iran
........................................................................................................................ 7
1.2.1.
An overview of the water issues of Iran .............................................. 7
1.2.2.
Description of the Karkheh Basin and problem statement ............. 10
1.3. Research Framework ................................................................................. 19
1.3.1.
Research motivation ........................................................................... 19
1.3.2.
Research objectives and questions .................................................... 20
1.3.3.
Contribution of the proposed research ............................................. 20
2.
MATERIALS AND METHODS ...................................... 23
2.1. Methodological Framework ....................................................................... 23
2.1.2.
System investigation ........................................................................... 24
2.1.3.
Hydrological modeling ....................................................................... 27
2.2.
Data Collection ............................................................................................ 30
3. STREAMFLOW VARIABILITY AND WATER ALLOCATION
PLANNING .......................................................................... 33
3.1
Introduction ................................................................................................ 33
3.2.
Data and Methods ....................................................................................... 34
xviii
____________________________________________________________________
3.3. Results and Discussion ............................................................................... 36
3.3.1.
Spatial and temporal variability of daily streamflow regimes ........ 36
3.3.2.
Spatial and temporal variability of monthly streamflows ............... 41
3.3.3.
Long-term variability in annual surface water availability ............ 45
3.3.4.
Overview of the basin-level water accounting .................................. 48
3.4.
Concluding Remarks .................................................................................. 51
4.
STREAMFLOW TRENDS AND CLIMATE LINKAGES 53
4.1.
Introduction ................................................................................................ 53
4.2. Data and Methods ....................................................................................... 54
4.2.1.
Hydrological and climate data and indices ....................................... 54
4.2.2.
Trend and correlation analysis .......................................................... 58
4.3. Results and Discussion ............................................................................... 58
4.3.1.
Characterizing the streamflow regime.............................................. 58
4.3.2.
Streamflow trends ............................................................................... 60
4.3.3.
Trends in the climatic data ................................................................ 63
4.3.4
Streamflow trends and climate linkages ........................................... 65
4.3.5
The impact of NAO index on the local climate ................................. 70
4.4.
Concluding Remarks .................................................................................. 71
5. REGIONALIZATION OF A CONCEPTUAL RAINFALLRUNOFF MODEL BASED ON SIMILARITY OF THE FLOW
DURATION CURVE ............................................................ 73
5.1. Introduction ................................................................................................ 73
5.1.1.
Problem statement .............................................................................. 73
5.1.2.
Review of regionalization methods using conceptual rainfall-runoff
models .................................................................................................. 74
5.1.3.
Scope and objective............................................................................. 78
5.2. Materials and Methods............................................................................... 78
5.2.1
Study catchments and available data ................................................ 78
5.2.2
Naturalization of the streamflows ..................................................... 80
5.2.3.
Model calibration and validation at the gauged catchments........... 83
5.2.4.
Regionalization of model parameters based on catchment similarity
analysis................................................................................................. 85
5.2.5
Assessment of the impact of parameter uncertainty on the
regionalization results......................................................................... 87
xix
____________________________________________________________________
5.3. Results and Discussion ............................................................................... 88
5.3.1.
Model results of automatic parameter estimation ........................... 88
5.3.2.
Regionalization results based on drainage area, spatial proximity
and catchment characteristics ........................................................... 90
5.3.3.
Regionalization results based on FDC .............................................. 91
5.3.4.
Impact of parameter uncertainty on the regionalization results .... 93
5.3.5.
Comparison of the FDC- based regionalization results with other
studies .................................................................................................. 94
5.4.
Concluding Remarks .................................................................................. 96
6. IMPACT OF AREAL PRECIPITATION INPUT ON
STREAMFLOW SIMULATIONS ......................................... 97
6.1.
Introduction ................................................................................................ 97
6.2. Data and Methods ....................................................................................... 99
6.2.1.
Data used in the model setup ............................................................. 99
6.2.2.
Formulation of precipitation input scenarios ................................. 101
6.2.3.
Model calibration .............................................................................. 105
6.3. Results and Discussions ............................................................................ 108
6.3.1.
Comparison of precipitation input .................................................. 108
6.3.2.
Comparison of streamflow simulations........................................... 111
6.4.
Concluding Remarks ................................................................................ 120
7. QUANTIFYING SCALE-DEPENDENT IMPACTS OF
UPGRADING RAIN-FED AGRICULTURE ....................... 123
7.1
Introduction .............................................................................................. 123
7.2. Methodology .............................................................................................. 124
7.2.1.
Model used for the scenario simulation .......................................... 124
7.2.2.
Tested scenarios ................................................................................ 127
7.3. Results and Discussion ............................................................................. 130
7.3.1.
Downstream impact of upgrading rain-fed areas to irrigated
agriculture (S1) ................................................................................. 130
7.3.2.
Downstream impact of improved soil water availability through
rainwater harvesting Scenario (S2) ................................................. 133
7.3.3.
Combined impact of S1 and S2 (S3) ................................................ 134
7.3.4.
Consideration of prediction uncertainty of the model ................... 134
xx
____________________________________________________________________
7.4.
Summary and Concluding Remarks ....................................................... 137
8. SYNTHESIS, CONCLUSIONS AND RECOMMENDATIONS .
...................................................................................... 139
8.1.
Nature and Causes of a High Level of Hydrological Variability .......... 139
8.2.
Water Allocations, Water Availability and Sustainability .................... 140
8.3.
Streamflow Trends and Their Underlying Causes ................................ 140
8.4.
Addressing Methodological and Data Scarcity Issues in the Hydrological
Modeling .................................................................................................... 141
8.5.
Consideration of the Impacts on Downstream Water Availability while
Upgrading Rain-fed Agriculture ............................................................. 142
8.6.
Contribution and Innovative Aspects of This Research ........................ 142
8.7.
Major Recommendations and Future Directions .................................. 143
SAMENVATTING .................................................................. 145
SUMMARY IN PERSIAN ....................................................... 151
REFERENCES ...................................................................... 157
LIST OF FIGURES ................................................................ 175
APPENDIX ............................................................................ 179
Appendix A.
Short description of the Hargreaves method and its application
in the study basin ...................................................................... 179
ABOUT THE AUTHOR ......................................................... 181
1.
INTRODUCTION
1.1.
Background
1.1.1. Increasing pressure on earth’s water resources
Water plays a key role in sustaining life on our planet earth. We use water not only
for our basic survival (e.g., for drinking, cooking, bathing and sanitation) but also for
many other purposes such as hydropower generation, industry, navigation and
recreation. Water is essential not only for meeting human needs but for nature where
it is essentially required to maintain fisheries, wildlife, riparian vegetation, river
deltas and aquatic biodiversity.
The freshwater resources of the earth are finite and are distributed into
hydrological storages as glaciers, groundwater, freshwater lakes and wetlands, soil
moisture, atmospheric water and river waters (Shiklomanov and Rodda 2003).
Balonishnikova et al. (2006) have estimated that the total renewable freshwater
resources of the world are about 42,700 km3/yr.. The spatio-temporal distribution of
water is very much nonuniform across the globe. Also the full amount of renewable
water is not accessible to human uses due to different reasons such as the fact that a
major part of the rainwater flows as flood runoff during short period of time. This
high spatio-temporal variability together with extreme climatic events in the form of
floods and droughts, and localized high demands from intensive agriculture and big
cities make water management a very complicated task.
Large investments in infrastructure (e.g., dams and irrigation facilities) have
resulted in a rapid increase in the uses of water for human purposes during the last
century (Figure 1) (Shiklomanov 1999). The major share of the total water
withdrawals and consumptions pertain to the agriculture sector (about 70%)
followed by industrial and municipal sectors. The world water withdrawals have
increased over 7 times during the last century, i.e., from 578 km3/yr. in the year 1900
to about 3,788 km3/yr. in the year 1995. This trend is projected to continue in future,
though with comparatively lower rates. As a consequence, the freshwater resources
of the world are under ever-increasing pressure due to escalating demands. The main
driving forces behind this rising pressure are: population growth; major demographic
changes as people move from rural to urban environments; higher demands for food
security and socioeconomic well-being; increased competition between uses and
usages; and pollution from industrial, municipal and agricultural sources, climate
variability and change, and land use change (e.g., WWAP 2006).
Despite the immense progress in water development the demands are still very
difficult to meet in many regions of the world. There are about 1.1 billion people
who still do not have access to improved water supply, and about 2.4 billion, i.e.,
40% of the world population, lack access to improved sanitation (WHO 2000).
2
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Irrigated agriculture has to expand further to meet the food needs of growing
populations and hence withdrawals to irrigated agriculture will keep increasing
(Seckler et al. 1998). Even if irrigation efficiency could be improved dramatically at
some places, meeting these water demands will be a big challenge in many parts of
the world, especially in developing countries of Asia and Africa as these regions will
be facing severe water scarcity in the coming decades (Rijsberman 2006).
Water withdrawals (km 3/yr.)
6000
Agriculture
5000
Municipal
Industrial
4000
World Total
3000
2000
1000
0
1900
1940
1950
1960
1970
1980
1990
1995
2000
2010
2025
Year
Figure 1. Trends in the global water withdrawals by sector of economic activity.
(Data source: Shiklomanov, 1999, cited in Cosgrove and Rijsberman 2000a.)
A historical overview depicts that the human appropriation of freshwater water
resources has helped in many ways such as preventing food crises in the world,
provision of water and sanitation, generating electric power and mitigation of
damage from hydrological hazards such as flood and drought. But it is now well
recognized that water resources strategies of the last century have largely worked
against nature and have resulted in environmental degradation as many rivers no
longer reach the sea for extended periods of time, river delta regions are ruining,
groundwater in the world’s key aquifers are depleting, water pollution is increasing
and aquatic ecosystems are being increasingly damaged (Rijsberman and Molden
2001; Gleick 2003; Postel and Richter 2003). Many countries of the world are facing
this conflicting situation at present and are searching for sustainable solutions to
achieve a balance among human and ecosystem uses of water. However, most of the
restoration examples are limited to USA, Australia, South Africa and Europe
(Tharme 2003; Smakhtin et al. 2004). Balancing water for human needs and for
nature is a big challenge faced by many countries at present and has been regarded
as one of the greatest challenges of this century all across the globe (Rijsberman and
Introduction
3
____________________________________________________________________
Molden 2001; Zehnder et al. 2003; Postel 2003; Loucks 2006; Palmer and Bernhardt
2006).
1.1.2. Adapting sustainable solutions
Water issues in the world are diverse in nature, governed by a large array of natural
and anthropogenic forces such as climatic conditions, land features, hydrological
behavior, variability of water resources, socio-political conditions, economic factors,
technological capacity and ecosystems needs. Integrated Water Resources
Management (IWRM) has been advocated as the better way forward for addressing
the complex and dynamic nature of the water-related issues (e.g., Bouwer 2000;
Karamouz et al. 2001; Snellen and Schrevel 2004; van der Zaag 2005; Savenije and
van der Zaag 2008).
Emphasizing the adoption of an integrated approach to water resources
management, Cosgrove and Rijsberman (2000a) suggested that limiting the
expansion of irrigated agriculture, increasing water productivity, developing
biotechnology for agriculture, increasing storage, reforming water resource
management institutions, increasing cooperation in international basins, valuing
ecosystem functions and supporting innovation would be the key areas of
interventions contributing towards addressing the global water crisis and,
consequently, would help achieving the Millennium Development Goals (MDGs).
Gleick (2003) argued that the “hard path1” solutions of the past are no longer
better choices and we need to follow the “soft path2” solutions. Therefore, we need
to rely on carefully planned and managed centralized infrastructure complimented
by small-scale decentralized facilities; strive for improving the productivity of water
rather than seeking for endless sources of new supply; deliver water services and
qualities matched to users’ needs rather than just delivering quantities of water;
apply economic tools for promoting efficient water use; and include local
communities in decisions about water management, allocation and use.
Vörösmarty et al. (2000) have recommended that an integrated research on
climate change, water resources and socioeconomic aspects would be essential for
making progress as the population growth and economic development will be the
main forces escalating the water demands in the future. Investments in
socioeconomic and hydrometric data are important and should be enhanced for
making adequate progress.
Improving productivity of water in agriculture is regarded as one of the most
promising solutions (CAWMA 2007). It is argued that producing more food with
1
Hard path refers to the approach based on the construction of massive infrastructure in the
form of dams, aqueducts, pipelines, and complex centralized treatment plants, which
dominated the water agenda of the twentieth century.
2
Soft path refers to the approach based on the carefully selected centralized physical
infrastructure with lower-cost community-scale systems, decentralized and open decision
making, water markets and equitable pricing, application of efficient technology, and
environmental protection (Gleick 2003).
4
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
less or with the same amount of water (more crop per drop) will lead to more food
security, less infrastructural requirements, reduced competition for water as less
water will be needed for agriculture and more can be diverted for domestic,
industrial and environmental purposes (Cosgrove and Rijsberman 2000a, b; Postel
2000; Rijsberman and Molden 2001; CPWF 2002). Improving productivity of water
both from rain-fed and irrigated lands is a key focus of the new blue-green water
paradigm (Falkenmark and Rockström 2006).
The list of potential solutions is quite long. Just to mention a few more: creating
awareness among all the stakeholders about water-food-environment nexus
(DIALOGUE 2002) and developing and adopting new technologies and changing
lifestyles (e.g., changing dietary patterns, improving education and reducing
population growth rates) would be very essential for matching the water supplies and
demands in the future (Gallopin and Rijsberman 2000; Cosgrove and Rijsberman
2000b). There is a need to change mindsets, policies and practices and to overhaul
water policies and practices in a way that will protect freshwater ecosystems and
their valuable services (Postel 2003, 2005). While we update water policies, the
highest priority should be given to the following three policy areas (Postel 2005): a)
securing drinking water supplies through increased investments in the catchment
protection; b) inventorying and setting ecological goals for the health of rivers,
lakes, and other freshwater ecosystems and establishing caps on the degree to which
human activities are allowed to modify river flows, deplete groundwater, and
degrade catchments; and c) improving water productivity both from agriculture and
nature through a combination of efficient water use and implementation of caps on
water use.
1.1.3. Managing water by river basin
Scale consideration is very important both for the understanding and simulation of
hydrology (Blöschl and Sivapalan 1995) and for the management of water resources
(Zehnder et al. 2003; van der Zaag and Gupta 2008). Water issues and water
management could be viewed in many different spatial scales such as global,
continent, country, river basin, catchment (subbasin), irrigation system, city,
wetland, farmer field, etc. The temporal scales could be every minute, hour, daily,
month, season, decade, year or even every specified longer period. It is now well
recognized that the river basin is the most appropriate scale for the sustainable
management of water resources (WWAP 2006; Molle 2006). The European Frame
Work directive is a well known example in this regard which states that the rivers,
lakes and groundwater resources need to be managed by the river basin which is a
natural hydrological unit, instead of only according to administrative or political
boundaries (Ringeltaube 2002).
It must be noted that there are a lot of unknown processes and facts pertaining to
all of the abovementioned scales. It is extremely important to understand the present
Introduction
5
____________________________________________________________________
state of a river basin with respect to the degree of “basin closure,”3 i.e., whether it is
an open basin, closed basin or closing basin, as this knowledge has implications for
many water management polices (Keller et al. 1996, 1998; Seckler, 1996;
Falkenmark and Molden 2008). For instance, adapting water conservation and
irrigation efficiency improvement strategies aiming at water savings may not really
save water in a closed basin and may merely reflect the reappropriation among
different users/uses. In such cases, improving overall productivity of water is a more
plausible alternative. Furthermore, understanding various factors, such as
hydrological, water management, socio-political and economic, governing the river
basin transformations and water uses are also essential (Molle 2003).
Management of water resources from a river-basin perspective requires
comprehensive interdisciplinary analysis, evaluation of present and future
conditions, and formulation of multiple management plans (Schultze 2001). But,
there are several scientific and technical obstacles that prohibit us from
understanding, predicting and ultimately guiding the management of water
resources. The major scientific issues are the lack of understanding of hydrological
processes at the basin scale and inadequate understanding of the coupling between
hydrological, ecological and climate systems (Uhlenbrook 2006; Uhlenbrook et al.
2006).
1.1.4. Need for a hydrological synthesis
The need for hydrological investigations was at the core of the hard path solutions of
the last century. The hydrology-based assessment of water resources was then
integrated with the information from other disciplines (such as geology, soil science,
atmospheric science, sociology/anthropology and various engineering disciplines)
for implementing water resources development and management strategies. The
pivotal role of hydrology in implementing the hard path solutions is quite evident
and has been very well internalized in the planning, construction and operational
phases of the water resources development projects. However, in the past when
structural measure were the main options for solving water availability issues, this
role of hydrology was much simpler, as water was abundant and the effects of
3
A water resource system is "closed" when there is no usable water leaving the
system other than that necessary to meet minimum instream and outflow
requirements (Keller et al. 1998). From the agricultural standpoint, either all of the
initial available water supply has been lost to beneficial evaporation and crop
evapotranspiration (ETc), plus unavoidable nonbeneficial evaporation and ETc, or it
has such high concentrations of salts and other pollutants that it is unusable.
Conversely, an integrated water resource system is "open" when excess usable water
does leave the system and there is nonbeneficial evaporation and ET that can be
avoided. According to a recent definition by Falkenmark and Molden (2008), a river
basin is termed closed when additional water commitments for domestic, industrial,
agricultural, or environmental uses cannot be met during all or part of a year, while
in an open basin more water can be allocated and diverted.
6
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
anthropogenic causes, climate and land use changes were not that prominent. But the
scientific role of hydrology is much more demanding and challenging now and in
the future when water challenges are more pronounced, diverse and complicated.
Hydrological investigations are essentially needed before formulating both hard
and soft path solutions and should be continuously updated in view of changing
needs and conditions. Understanding of hydrological processes and hydrology-based
assessment of water resources and water balance is, in fact, an integral part or a
basic requirement for most of the abovementioned solutions. For instance, it is one
of the essential components of water productivity estimations (Molden 1997) and
water scarcity studies (Seckler et al. 1998; Sullivan et al. 2000). Hydrological
investigations are essentially required to devise action plans for the policy areas as
proposed by Postel (2005) on sustainable uses of water by humans and ecosystems
(see section 2.1). For example, putting caps on water withdrawals requires a
quantitative assessment of the availability of water resources. We need to
study/model the hydrological response of the catchments before making catchment
restoration measures and investment decisions and we essentially require detailed
information on the spatio-temporal pattern of water flows for the restoration of the
natural hydrological regime of the rivers.
Similarly, hydrological analyses at basin and smaller levels are required to obtain
the knowledge of the degree of a basin closure and flow paths of water which can
then help in guiding the kind of appropriate interventions. Without such hydrological
assessments much of the debate on the real water savings, upstream- downstream
impacts (Keller et al. 1996 and 1998; Seckler 1996; Molle et al. 2004) or whether a
demand-side or supply-side intervention is better (Molle and Turral 2004) remains
mere conceptual and, therefore, a qualitative assessment of complex realities may
lead to erroneous planning and consequently nonsustainable management of water
resources.
The issues of water resources management are becoming increasingly important
almost everywhere in the world and water-related problems are becoming
increasingly complex. The role of hydrological investigations remains pivotal in
exploring sustainable solutions for the present and emerging water issues. A
hydrologic synthesis in at least three respects is essentially required, which are
across a) processes, where the challenge is how to represent complex interacting
dynamic systems including feedback between system components, b) places, where
the challenge is how to synthesize the plethora of case studies around the world in
the past decades, and c) scales, where one is interested in the general characteristics
of processes as a function of spatio-temporal scales for the same site or an ensemble
of sites (Blöschl 2006).
Therefore, there is need to increase understanding of hydrological processes in
spatio-temporal scales and their interaction with humans and ecosystems. Then the
challenging undertaking, for future research and practices in hydrology and water
resources, is how to induce and deduce sustainable water management strategies
based on the hydrological research.
Introduction
7
____________________________________________________________________
1.2.
Hydrological and Water Management Issues in the Karkheh Basin, Iran
1.2.1. An overview of the water issues of Iran
The Islamic Republic of Iran is located in Southwest Asia and is situated between
approximately 25-40 degrees northern latitudes and 44-64 degrees eastern
longitudes. The total area of Iran is about 1.65 million km2, out of which about 52%
is mountainous and desert terrain and about 16% is terrain with an elevation of over
2,000 meters above sea level (masl) (FAO 1997). The two largest and highest
mountain systems are Zagros and the northern highlands (Talish and Alburz), the
former extending from northwest to southeast, while the later stretches from west to
east along the southern Caspian Sea.
Forests and woodlands comprises only 7% (11.4 million ha) of the total land area
and about 27% (44 million ha) is under pastures and meadows (FAO 2006). The
arable land and permanent crops are estimated to be 16.1 and 2.1 million ha,
respectively. The agricultural area under irrigation has grown from 4.7 million ha in
1961 to 7.7 million ha in 2003, indicating a growth of about 63% over this period.
Despite tremendous increase in irrigated area, rain-fed farming is a very important
feature of the country’s food security and agricultural economy.
The climate of Iran depicts extreme variations due to its geographic locations and
varied topography. Generally, it is regarded as a country of dry conditions and its
climate is mostly arid to semi-arid. Precipitation (P) patterns show large spatial and
temporal variations, caused mainly by Zagros and the northern mountain ranges. The
average annual P over Iran is about 240 mm/year. (/yr.) (Dinpashoh et al. 2004).
Over half of the country’s area receives less than 200 mm/yr. and over 75% receives
less than 300 mm/yr.. Only 8% of the area receives more than 500 mm/yr.. The
seasonal distribution in winter (January-March), spring (April-June), summer (JulySeptember) and autumn (October-December) is about 53, 20, 4 and 23%,
respectively, of the annual P.
Iran has several large rivers, among which Kurun, Dez and Karkheh are the three
major ones. Most of the rivers and streams are steep and irregular and end up in the
marshes/wetlands. Most of the marshes and wetlands of Iran have high significance
for their biodiversity, environmental and socioeconomic values. Water is also stored
naturally underground both in confined and unconfined aquifers, finding its outlet in
qanats (subterranean water canals), springs and streams. Vakili et al. (1995)
analyzed the different estimates of water resources of Iran and suggested that the
total quantity of renewable water resources is about 135 km3 /yr.. According to FAO
(1997), the internal renewable water resources of Iran are estimated at 128.5 km³/yr..
It receives 6.7 km³/yr. of surface water from Helmand River having a drainage area
in Pakistan and Afghanistan. The flow of the Arax River, at the border with
Azerbaijan, is estimated at 4.63 km³/yr.. Surface runoff represents a total of 97.3
km3/yr. whereas groundwater recharge is estimated at around 49.3 km³/yr. of which
12.7 km³/yr. are obtained from infiltration from the riverbeds.
Consistent with the global trends shown in Figure 1, the increasing water
withdrawals continue to amplify pressure on the water resources of Iran (Figure 2).
8
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Water withdrawals have doubled over the last 3 decades, rising from 45 km3/yr. in
1975 to 93.3 km3/yr. in 2004 (FAO 2009). Most of this increase is due the increased
allocations to the agriculture sector.
Total withdrawals
Agriculture
Municipal
Industrial
3
Water withdrawals (10^9 m/yr.)
100
80
60
40
20
0
1975
1995
2000
2004
Year of assessment
Figure 2. Water withdrawals by sector in Iran.
(Data source: FAO 2009, Aquastat database.)
The major driver of this trend is the country’s policy to attain food selfsufficiency illustrated in Figure 3, showing the increase in cereal area and cereal
production in the country. However, despite the increasing trend in the production
over time, the cereal import was imperative to meet demands (FAO 2009). The high
variations in cereal area, yield and imports could be attributed to the variable nature
of the climate and water resources. For instance, the food production faced serious
decline in the dry years, 1999-2001, and therefore about 10 million tonnes of cereals
were imported costing about 1.5 billion US dollars in 2000.
Various sources project that Iran would be facing serious water stress and water
scarcity problems by the first quarter of this century (Seckler et al. 1998; Wallace
2000; Alcamo et al. 2000; Sullivan et al. 2000; Yang et al. 2003). Figure 4, showing
trend in population increase and corresponding decline in per capita water
availability, demonstrates the simplest representation of water scarcity. The water
availability/capita/yr. was about 6,057 m3/person/year in 1961, which showed a
sharp decline of about 70% over the period 1961 to 2009, reaching about 1,820
m3/year/person in 2009. Given the rising trends in population, the per capita water
availability is projected to fall below the water stress threshold value of 1,700
m3/person/year in the coming decade by as early as 2015.
Introduction
9
____________________________________________________________________
Indicated values are in 10^6
25
Cereals production, 10^6 tonnes
Cereal import, 10^6 tonnes
20
Cereals Area, 10^6 hectares
15
10
5
0
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Year
Figure 3. Cereal area, production and import in Iran during 1961-2007.
(Data source: FAO 2009.)
Total population
Urban population
Rural population
Per capita water availability
Population (10^6)
100
6,000
80
4,500
60
3,000
40
1,500
20
0
1960
1970
1980
1990
2000
2010
2020
2030
2040
Per capita water availability
( m 3/person/yr.)
7,500
120
0
2050
Year
Figure 4. Overview of trends in per capita availability of renewable water
resources and population growth of Iran (1961-2050).
(Data for population estimate and projections are taken from FAO 2009 whereas the value of 135 km3/yr.
is used as annual renewable water resources after Vakili et al. 1995.)
10
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Falkenmark et al. (1989) proposed 1,700 m3/capita/yr. of renewable water
resources as the threshold below which a country experience water stress; this
threshold is based on estimates of the water requirements in the household,
agriculture and energy sectors, and the needs of the environment. In fact, the signs of
water scarcity and water stress (e.g., reduction in river flows, groundwater overdraft,
environmental degradation, water shortages for urban users) in Iran have been
already evident during the past few decades, exacerbating the water management
issues (Foltze 2002).
The semi-arid and arid environments, as in many areas in Iran, are highly
sensitive to (local and global) changes, mainly due to scarcity and variable
distribution of water and nutrients (Newman et al. 2006). Soil erosion, salinization,
groundwater depletion and desertification are the most common environmental
changes that have occurred in these water limited environments (De Fries et al.
2004) and dry conditions of Iran are not exceptions. Increasing water demands for
agriculture, industry and domestic uses continue to put more pressure on the scarce
water resources in these water-limited environments. The expected regional climate
change (Christensen et al. 2007) poses yet another challenge to the sustainable
management of natural resources and the environment for the benefit of the society.
In summary, the water crisis of Iran is likely to intensify given the increasing
competition of water for human uses and the environment. There are many other
governing factors ranging from natural and anthropogenic climate changes to the
complex socioeconomic, institutional and hydrological factors. This stresses the
importance of increasing knowledge of the basin hydrology and water availability
for constructing a sound and sustainable water regime. Further studies on hydrology
and water management are also required to achieve the national water policy
objectives, which stress the need to establish a comprehensive water management
system that incorporates natural elements of the total water cycle as part of
principles of sustainable development (Ardakanian 2005). Therefore, there is an
urgent need to increase knowledge and understanding of the hydrology and water
resources systems that can, in turn, help address the water and related issues.
1.2.2. Description of the Karkheh Basin and problem statement
The Karkheh Basin is located in the western part of Iran (Figure 5). The drainage
area of the basin is about 50,764 km2, out of which 80% falls in the Zagros mountain
ranges. Administratively, Karkheh Basin area is distributed into seven provinces as
shown in Figure 5. Hydrologically, it is divided into five main catchments
(subbasins) (Figure 5), namely Gamasiab, Qarasou, Saymareh, Kashkan and South
Karkheh. These catchments are named after the main river passing through the
respective areas. The Karkheh River eventually terminates in the Hoor-Al-Azim
swamp, a large transboundary wet land shared with Iraq, which is connected to
Euphrates-Tigris system.
Introduction
11
____________________________________________________________________
Figure 5. Location of Karkheh Basin in Iran and its hydrological and
administration units.
12
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The details on the study basin can be found in Sutcliffe and Carpenter (1968),
JAMAB (1999; 2006a, 2006b), UNEP (2001), Ashrafi et al. (2004), , Karamouz et
al. (2006; 2008, 2011), Heydari (2006), Absalan et al. (2007), Keshavarz et al.
(2007), Ghafouri et al. (2007), Ahmad et al. (2009), Ahmad and Giordano (2010),
Marjanizadeh (2008), Marjanizadeh et al. (2009; 2010) and Muthuwatta et al.
(2010). The salient features and problem statement are described below.
The topography depicts large spatial variation with elevations ranging from 3 to
more than 3,000 masl (Figure 6). The elevation of about 60% of the basin area is
1,000-2,000 masl and about 20% is below 1,000 masl (Ashrafi et al. 2004). The
highest peak in the basin has a height of 3,645 masl. In the upper part of the basin (in
northern parts), a number of wide alluvial plains lie at an elevation of about 1,500
masl within complex faulted and overthrust limestone or metamorphic mountain
masses whose summit exceeds 3,000 masl at several points. In the central part of the
basin, upstream of the Khuzestan plains, the Karkheh and its tributaries flow through
the remote and sparsely inhabited region of the Lorestan and Ilam provinces, an area
of extremely elongated and uniform mountain folds, oriented northwest to southeast
and again predominantly of limestone (Sutcliffe and Carpenter 1968). In the lower
parts, the Karkheh River runs through mostly flat and irrigable regions of the basin,
through several meanders, before draining into the Hoor-Al-Azim Swamp.
As in all other areas of Iran, the Ministry of Energy (MOE) is in charge of water
resource assessment and development in the Karkheh Basin. Through its provincial
water and power development authorities the MOE is responsible for large
hydraulics works, including dam and irrigation and drainage canals for distribution
of water. MOE and its water-related department oversee and coordinate planning,
development, management and conservation of water resources. The responsibility
of operation and maintenance of primary and secondary irrigation and drainage
canals lies within the water-related department of MOE. The Khuzestan Water and
Power Development Authority (KWPA) is among the key institutions dealing with
water issues in the Karkheh Basin. The Ministry of Jihad-e-Agriculture, through its
provincial organizations, is responsible for on-farm water management, on-farm
irrigation and drainage networks, rain-fed and irrigated crops, catchment
management and other related issues. Many other social and nonformal institutions
are functioning in the basin; working for the local water management activities.
These local organizations have derived their water allocation and management
principles through the rich history of Iranian cultures.
The population living in the basin is about 4 million (in 2002), and about onethird resides in the rural areas (JAMAB 1999; Ashrafi et al. 2004). The annual
population growth rate is about 2.6%. Historically, the Karkheh Basin had been the
cradle of ancient civilization of Mesopotamia and a boundary between Arab and
Persian cultures. The Karkheh Basin, once called the “breadbasket of Southwest
Asia” now faces many challenges such as low water and land productivity, poverty,
land degradation, groundwater depletion and growing competition for water among
upstream and downstream areas and among different sectors of water use such as
irrigation, domestic, hydropower and environment (CPWF 2003).
Introduction
13
____________________________________________________________________
Figure 6. Digital elevation map of the Karkheh Basin and the streamflow
monitoring network.
14
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Hydrological features of the Karkheh Basin are complex and heterogeneous
because of its diverse topography, and natural settings of geology, climate and
ecology. Generally, the basin is characterized by a Mediterranean climate having
cool and wet winters and hot and dry summers. The main sources of P are the
Mediterranean depressions and Mediterranean cyclones (Domroes et al. 1998). The
former are mainly responsible for the P over the basin areas falling under Zagros
mountain ranges and are later the main source of P in the arid plains of the South
Karkheh Region. The P pattern depicts large spatial and intra- and inter-annual
variability across the basin. The mean annual P ranges from 150 mm/yr. in the lower
arid plains to 750 mm/yr. in the mountainous parts (JAMAB 1999). This variability
is demonstrated by Figure 7 indicating the mean monthly climate of Kermanshah
(the Upper Karkheh), Khorramabad (the Middle Karkheh) and Ahwaz (the Lower
Karkheh). On average, the middle part receives higher P than the upper and lower
parts as illustrated by the records of Kermanshah (450 mm/yr.), Khorramabad (510
mm/yr.) and Ahwaz (230 mm/yr.) (Figure 7). Most of the P (about 65%) falls during
the winter months from December to March and almost no P during summer season,
i.e., June to September. In the mountainous parts during winter, due to temperatures
often falling below 0 °C, the winter P falls as snow and rain. A recent study on snow
cover in the Zagros mountains by Saghafian and Davtalab (2007) has shown that the
snow water equivalent for the mountainous parts of the Karkheh basin is about 75
mm/yr., which is about 17% of the long-term annual P in the basin. The amount and
distribution of snow are strongly influenced by elevation, varying from 44 mm/yr.
for elevations less than 1,500 masl to 245 mm/yr. with elevation more than 3,500
masl.
Both temperature (T) and potential evapotranspiration (ETP) increase from north
to south of the basin, as indicated in Figure 7. The temperature shows large intraannual variability, with January being the coolest (e.g., mean temperature at
Kermanshah, Khorramabad and Ahwaz are about 2, 5 and 12 oC, respectively) and
July the hottest month (e.g., mean temperature at Kermanshah, Khorramabad and
Ahwaz are about 27, 29, 37 oC, respectively). The ETP largely follows a similar
pattern as the T with the highest in the southern arid plains (e.g., 1,930 mm/yr. at
Ahwaz) and the lowest at the mountainous semi-arid region (e.g., 1,515 mm/yr. at
Kermanshah). There is a large gap between ETP and P in most of the months, which
widens as we move from upper northern semi-arid regions to the lower southern arid
parts of the basin. The hydrological analysis and assessment of water resources in
such semi-arid to arid regions with high climatic variability is a challenging
undertaking compared to humid areas where P exceeds the ETP in most of the
months (Sutcliffe 2004).
The spatial variability of soil and land use types is demonstrated in Figure 8. The
valley soils are mainly fine-to-medium in texture, whereas the hilly areas are
composed of shallow soils generally classified as rock outcrops. The rain-fed
farming, rangelands, forests and irrigation farming are the main land use types.
Introduction
15
____________________________________________________________________
Mean monthly climate at Kermanshah
300
40
P and ETP
(mm/month)
250
ETp
T
35
30
200
25
150
20
15
100
10
50
T (oC/month)
P
5
0
0
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Month
Mean monthly climate at Khorramabad
40
300
35
250
25
20
150
15
100
10
50
T (oC/month)
P and ETP
(mm/month)
30
200
5
0
0
Jan. Feb. Mar. Apr.
May June July Aug. Sept. Oct. Nov. Dec.
Month
Mean monthly climate at Ahwaz
300
40
35
250
25
150
20
15
100
10
50
T (oC/month)
P and ETP
(mm/month)
30
200
5
0
0
Jan.
Feb. Mar. Apr.
May June July
Aug. Sept. Oct.
Nov. Dec.
Month
Figure 7. Mean monthly climate of the Karkheh Basin, illustrated by precipitation
(P), temperature (T) and potential evapotranspiration (ETP) at the three climatic
stations Kermanshah, Khorramabad and Ahwaz. Data source: Meteorological
Organization of Iran.
(The averages are for the period of 1950s to 2003. potential evapotranspiration was estimated using
Hargreaves method, Hargreaves et al. 1985).
16
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Figure 8. Spatial variability of soil and land use types in the Karkheh Basin.
(Source: Soil map: Soil and Water Research Institute (SWRI), Iran; Land use map: International Water
Management Institute (IWMI), Sri Lanka.)
The rain-fed farming and rangelands are mainly scattered throughout almost the
whole mountainous region with varying degrees of coverage. Forested areas are
mainly found in the middle parts of the basin. Most of the irrigated farming is
concentrated in the lower region (South Karkheh catchment) and in the upper
northern regions (Gamasiab catchment). Over the past few decades, there has been a
trend of shifting rangelands to rain-fed or irrigated crop production (CPWF 2003;
Ashrafi et al. 2004; Qureshi et al. 2005). The degradation of rangelands is increasing
due to overgrazing. These anthropogenic land use changes together with natural
factors (low forest canopy covers and high erosion rates) add to the complexity of
the hydrology and water resources of the basin considered in the wider spatiotemporal perspective.
The cultivation of food grain crops, mainly wheat, dominates the agricultural
land use, besides other crops including fodder, vegetables, maize, sugar beat, pulses
and orchards. The dominance of wheat cultivation reflects the country’s policy of
attaining self-sufficiency in wheat production. The wheat is grown all over the basin,
both in rain-fed and irrigated conditions. The land and water productivity of wheat
and other major crops is generally low and has a large variation across the basin
(Ahmad et al. 2009). For instance, the land productivity of rain-fed wheat is about
1,460 ± 580 kg/ha and its water productivity is 0.46 ± 0.22 kg/m3, indicating
considerable scope for improvement. The water scarcity and the high variability of
the rainfall within a crop-growth season could be regarded as the major constraints
Introduction
17
____________________________________________________________________
to the crop production under rain-fed conditions, besides other factors such as soil
fertility and management-related issues.
The improved availability of water through adopting soil and water conservation
techniques and/or by means of providing (supplemental) irrigation could help
improve land and water productivity in the rain-fed agricultural systems. These
techniques could also contribute to addressing the catchment degradation issue, as
they are likely to promote land cover growth and reduce soil erosion. However, a
proper understanding their impacts is required for the informed agricultural and
water policy formulation process.
The MOE and other institutions have been engaged in the assessment,
development and management of the water resources. For instance, a vast network
of hydrological stations was established by MOE in the 1950s for monitoring river
discharges, climatic variables, sediment yields and water-quality parameters across
the whole river system (Figure 6). There were about 50 streamflow gauging stations
installed after 1950, but only half of them are used continuously. Consequently,
long-term streamflow data are not available for many catchments and the existing
records have gaps and quality issues. Filling these data gaps by estimating missing
streamflow time series for the poorly gauged catchments is essential for the proper
understanding of the spatio-temporal variability of hydrology and water availability
in the basin.
JAMAB (1999) undertook assessment of the hydrology and water resources of
the basin with the main motivations of developing the available renewable water
resources to expand irrigated lands, provide water to increasing populations and
industry, control floods and produce hydroelectricity. The basin-level water balance
analysis conducted for the hydrological year4 1993-1994 shows that, on average,
annual precipitation in the basin totals about 25×109 m3/yr. About 66% (16.4×109
m3/yr.) of total precipitation is returned to the atmosphere through ET. The
renewable water of the basin accounts for 34% of the total precipitation, equivalent
to about 8.6×109 m3/yr., and represents the sum of the amounts of surface water and
groundwater. Groundwater exists often in karsts (hard rock aquifers) and alluvial
aquifers, with the presence of both unconfined and confined conditions. The aquifers
have large variations in area and thickness, which have largely been attributed to the
tectonic factors, lithology, climate conditions and topography (e.g., JAMAB 1999;
2006a; Tizro et al. 2007). Generally, subsurface water storage in porous aquifers in
the northern mountainous regions of the basin is limited to valley floors
characterized by relatively large depths, high infiltration rates and good water
quality. In the southern arid plains, while the area of porous groundwater bodies
increases, the thickness and infiltration decrease and the salinity of groundwater
increases.
Out of 7.4×109 m3/yr. of the total streamflows 2.5×109 m3/yr. (or 34%) were
diverted to various uses in 1993-94. The direct diversions and pumping from the
streams constituted the main mode of water withdrawals in the basin. Groundwater
4
A hydrological year in the Karkheh Basin corresponds to October-September.
18
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
contributed about 1.7×109 m3/yr. to agriculture, domestic and industrial uses.
Groundwater withdrawals are mainly provided through pumping from deep and
shallow aquifers besides natural flow through springs and qanats. The total amount
of irrigation water diverted from the surface and subsurface resources in the basin
was estimated as 3.9×109 m3/yr., with 63 and 37% contributed through surface and
groundwater resources, respectively. Groundwater exploitation is a major source of
irrigation in Gamasiab and Qarasou subbasins. Based on the study by JAMAB
(1999), the year 1993-94 has been taken as the main reference for the water
availability and allocation planning in the Karkheh Basin. The detailed water
allocations for different sectors are summarized in Table 1 (JAMAB 1999; 2006b).
Table 1. Current and planned water allocations in the Karkheh Basin, Iran.
Sectors
2001
Rural areas
Urban areas
Mining
Industry
Agriculture
Fish farming
Environment
Total
59
203
0
23
4,149
14
500
4,949
Water allocation in different years (106 m3/yr.)
2006
2011
2016
2021
70
69
66
62
278
259
242
231
2
1
1
1
93
76
57
30
7,476
7,135
6,814
6,879
477
379
249
119
500
500
500
500
8,896
8,419
7,929
7,822
2025
67
295
2
113
7,416
510
500
8,902
Notes: The water sources are surface water, groundwater and reservoirs. Data source: JAMAB 1999;
2006b.
The Karkheh Basin remained unregulated by large dams before the completion
of the Karkheh Dam in 2001 (details on dams can be found at:
http://daminfo.wrm.ir/dam-secondary-fa.html). The Karkheh Dam, having a
designed storage capacity of about 7.5×109 m3 (and live storage capacity of about
4.7×109 m3), is a multipurpose dam aimed at providing irrigation water to about
350,000 ha in the Khuzestan plains (in the Lower Karkheh Region) besides the other
objectives of hydropower generation and flood control. The various dams and
irrigation schemes are currently under construction/study, most notably the
construction of another large multipurpose dam, namely Saymareh Dam on the
Saymareh River. These massive water works are turning this basin into a largely
regulated one.
The ongoing water resources development strategies in the Karkheh Basin have
impacted the distribution of water within the basin and will be continuously
impacting the basin hydrology. The earlier studies attempted to provide accounts of
water resources availability and their development potential but the implications of
water development strategies on the basin hydrology and on the different users and
uses of water across the basin are not properly investigated. The needs for reserving
water to environmental uses have not been adequately assessed in the earlier studies.
The upstream-downstream linkages of the water uses are not evaluated and,
Introduction
19
____________________________________________________________________
therefore, are poorly understood and not internalized in the water policies. There is a
lack of understanding of the realities of basin hydrology and linkages with water
management at the river-basin scale. With such information gaps, a sound
knowledge of basin hydrology is essential for effective water development policies
so that their negative impacts on different uses and users can be avoided, minimized
or mitigated. Therefore, there is a dire need for increasing the knowledge and
understanding of basin hydrology in view of the changing phases of water
management in the Karkheh Basin. A sound knowledge of spatio-temporal
hydrology is also imperative for addressing the pressing water management issues
revealed by close consultations with key stakeholders in the Karkheh Basin. These
issues are enumerated below (CPWF 2003, 2005; Ashrafi et al. 2004; Qureshi et al.
2005):
•
•
•
•
•
•
•
1.3.
Improving understanding of the hydrology and water management at the
river-basin scale.
Assessing the impacts of present irrigation development strategies on
different users in upstream-downstream locations, and how they are
influencing the basin hydrology.
Assessing the environmental water demands in the basin at different
scales through developing appropriate assessment methodologies.
Minimizing land erosion and reducing sedimentation yields to the
Karkheh Dam by exploring and implementing better catchment
management practices and reversing land degradation caused by
different reasons such as overgrazing and increasing agricultural area.
Managing salinity and waterlogging in the lower parts of the basin.
Improving the productivity of agricultural water use in irrigated, rainfed and pastoral systems.
Finding how water and poverty in the basin are interlinked and
determining the potential water-related interventions that can lead to
poverty reduction in the basin.
Research Framework
1.3.1. Research motivation
The role of hydrological analysis remains pivotal in formulating policies and
strategies for water resources development and management when stakeholders
require more precise assessment of the state of their water resources for making
tough water allocation decisions for highly competing water needs such as
agriculture, environment and other uses. This will be intensified due to increased
stress on the water resources as a result of global changes (climate change, land use
change, escalating population and food demands, etc.). The hydrological analysis
20
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
that can provide a reliable assessment of the state of the water resources, and is able
to integrate cause(s) and effect(s) of relationships of natural and human-induced
changes on hydrology and water resources across multiple spatio-temporal scales
and among multiple users in a river basin remains highly imperative in making
sustainable water-related decisions.
1.3.2. Research objectives and questions
The main objective of this research is to provide a hydrology-based assessment of
surface water resources of the Karkheh Basin, and study its continuum of variability
and change at different spatio-temporal scales.
The specific research questions are as follows:
•
•
•
•
•
•
•
What is the state of spatio-temporal variability of surface water
hydrology and water availability?
What are the major natural and anthropogenic factors governing the
streamflow regime?
What are the main features of the natural streamflow variability
including both high- and low-flow regimes?
What is the nature of the observed trends in streamflow (if any) and
how are the observed trends associated with climate?
How can scientifically sound and reliable assessments of rainfall-runoff
relationships be made using hydrological models with limited amount of
data?
How can regionalization procedures contribute to catchment modeling
under a data-limited environment?
What are the impacts of developing rain-fed agriculture on downstream
flows under different scenarios?
1.3.3. Contribution of the proposed research
This research contributes to Basin Focal Project (BFP) for Karkheh Basin. The BFPs
were the initiative of Challenge Program on Water and Food (CPWF) in order to
strengthen the basin focus of the program. The main goal of the BFPs were to
provide a more comprehensive and integrated understanding of the water, food and
environment issues in a basin; and to understand the extent and nature of poverty
within each selected basin and determine where water related constraints are both a
major determinate of poverty factor and where those constraints can be addressed
(CPWF 2005).
The IWMI executed Karkheh BFP in collaboration with several Iranian partners
during 2005-2009. The project followed the new IWMI research framework, which
focused on analysing water availability, mapping water productivity, mapping water
poverty, analysing high potential interventions and assessing impacts (IWMI 2005).
Introduction
21
____________________________________________________________________
The IWMI research framework and Karkheh BFP research methodology were
underpinned by the interdisciplinary knowledge and research/evaluation
methodologies for which hydrology and water resources assessments were the
important components. This PhD research contributes directly to improve
understanding of the Karkheh Basin hydrology and water availability. In general,
this research contributes in improving understanding of basin scale hydrological
processes exhibited in macro scale semi-arid basin which is quite diverse in hydroclimatic features and is data scarce. The knowledge generated by this study is
helpful to improve understanding of spatio-temporal variability of the basin
hydrology and its use in the sustainable management of water resources in a river
basin context for the Karkheh Basin, and similar regions of Iran and elsewhere.
2.
MATERIALS AND METHODS
2.1.
Methodological Framework
The methodological framework followed in this study is schematised in Figure 9.
The spatio-temporal details of analysis depend on the specific research issue,
application of a particular method and data availability. These details are specified in
the relevant chapters of this thesis. The hydrological variability and water
availability were investigated using various state-of-the-art methods of hydrological
analysis (termed as system investigation). The hydrological modeling was carried
out to understand hydrological process and their variability and to test the impact of
the water management interventions. The hydrological analysis and modeling
consequently provide a sound scientific basis for guiding the water resources
management in the river-basin context. A brief description of the methods used in
this study is presented below.
System investigation
Hydrological modeling
Understanding hydrological processes and their
spatio-temporal variability and change
Hydrological synthesis for water management in the river-basin context
Figure 9. Methodological framework followed in this research study
24
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
2.1.2. System investigation
The following methods were used in the system investigation activities.
Measure of central tendency and dispersion
The statistical measures were calculated to understand the central tendency of the
streamflow time series. For this the arithmetic means were estimated for monthly
and annual flows. Since in semi-arid and arid river basins, like the Karkheh Basin,
the arithmetic mean may be biased by a relatively small number of extreme values,
median statistics were also computed to get a better understanding of the average
conditions. The range of variability was measured by estimating the dispersion in the
data by computing the standard deviation and the coefficient of variation (CV).
Flow duration analysis
The flow duration curve (FDC) is a widely used measure in water resources
assessment and management for the investigation of water availability for designing
hydropower or irrigation schemes, streamflow requirements for riverine ecosystems,
etc. The FDC is a cumulative distribution of flows at a site showing the flow
assurance of how often any flow is equaled or exceeded. The details on the concept
and applications can be found in literature (Linsley et al. 1949; Vogel and Fennessey
1995; Smakhtin 2001a; Gupta 2008; Niadas and Mentzelopoulos 2008). The FDC
analysis was carried out for the daily, monthly and annual time scales and various
exceeding percentiles representing high, median and low flows (e.g., Q1, Q5, Q10,
Q25, Q50, Q75, Q90, and Q95) were derived.
Base flow index
The hydrograph separation into quick and base (slow) flow was carried out using the
commonly used digital filter method expressed by the following equations (e.g.,
Smakhtin 2001b):
qt = α × qt −1 +
(1 + α ) × (Q
2
Qbaseflow = Qt − qt
BFI =
Qbaseflow
Qtotal
t
− Qt −1 )
(1)
(2)
(3)
Here, qt is the filtered direct runoff at time step t (m3/s); qt-1 is the filtered direct
runoff at time step t-1 (m3/s); α is the filter parameter (-); Qt is the total runoff at
time step t (m3/s); and Qt-1 is the total runoff at time step t-1 (m3/s). Then the base
Materials and Methods
25
____________________________________________________________________
flow (Qbaseflow) is estimated as the difference of Qt and qt (equation 2). The Webbased Hydrograph Analysis Tool (WHAT) was used to do the calculations (Lim et
al. 2005; http://cobweb.ecn.purdue.edu/~what/). The value of α was set to 0.995
after Smakhtin (2001b) for all of the investigated gauging stations. The main
purpose of this exercise was to estimate the Base flow Index (BFI) which is the ratio
of the base flow to the total streamflow (equation 3). The BFI estimates were used to
characterize the base flow contribution to the streamflows and as well as its spatiotemporal variability. Further details on base flow analysis and some of its
applications can be found at Nathan and McMahon (1990a), Arnold and Allen
(1999), Larocque et al. (2010) and Welderufael and Woyessa (2010).
Water accounting
The water accounting framework developed by Molden and Sakthivadivel (1999)
was applied for the basin-level water accounting. This framework provides a unique
way of distinguishing different water use categories such as net inflow, process
depletion, non-process depletion, committed water and uncommitted outflows. The
key terms of the water accounting methodology, used in this study are defined
below. The details can be found in Molden (1997) and Molden and Sakhtivadivel
(1999).
•
•
•
•
•
•
•
Gross inflow: the total amount of inflow crossing the boundaries of a
domain.
Net inflow: the gross inflow less the change in storage over the time period
of interest within the domain. Net inflow is larger than gross inflow when
water is removed from storage.
Process depletion: that amount of water diverted and depleted (or
consumptively used) to produce an intended good.
Non-process depletion: depletion of water by uses other than the process
that the diversion was intended for.
Committed water: the part of outflow that is reserved for other uses such as
the environment.
Uncommitted outflow: outflow from the domain that is in excess of
requirements for downstream uses.
Available water: the amount of water available for a service or use, which
is equal to the inflow less the committed water.
Trend and correlation analysis
Trends in the long-term streamflow and climatic data were examined using the
Spearman’s Rank (SR) test (e.g., McCuen 2003; Yue et al. 2002). The SR test is a
nonparametric rank-order test. Given a sample data set {Xi, i = 1, 2,…, n}, the null
hypothesis H0 of the SR test is that all the Xis are independent and identically
distributed. The alternative hypothesis is that Xi increases or decreases with i, so
that, consequently, a trend exists. The calculation of the SR statistics Rsp requires
26
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
that the original observations Xis are transformed to ranks kis by arranging them in
the increasing order of magnitude and computed the quantity Di as Di = ki – i, where
i ranges from 1 to n and n is the number of observations. The Rsp and test statistics t
were calculated using the following equations:
n
Rsp = 1 −
t = Rsp
6 Di
i =1
(
2
)
n n2 − 1
n−2
2
1 − Rsp
(4)
(5)
If the computed t value lies within the desired confidence limits, we can
conclude that there is no trend in the series. We used a 90% confidence interval for
the evaluating presence or absence of trends. The water limited semi-arid to arid
environments, like the Karkheh Basin, are sensitive to changes; therefore, trends that
are significant at the 90% level could have quite serious implications. The
relationship between streamflow and climatic variables was studied by performing a
correlation analysis among them. For this purpose Pearson correlation coefficient, r,
was estimated (e.g., McCuen 2003). It is pertinent to note that defining the
significance of r values varies with the number of observations and selecting the
confidence bound, i.e., in the case of 40 observations, where the values outside the
range of ± 0.304 are defined as significant at the 95% confidence interval. However,
there is no strict approach for the interpretation of the correlation values and it
largely depends on the context and purposes. In this study, the variables were
considered having a good correlation if the r values fall outside the critical range of
± 0.304.
Serial correlation
Before applying the trend test, the studied data series were checked for the presence
of serial correlation. Previous studies have shown that the existence of serial
correlation can complicate the detection and evaluation of trends when applying a
nonparametric trend test and, thus, may have strong influence on the null hypothesis
about the presence of trends (e.g., von Storch and Navarra 1995; Yue and Wang
2002). The widely used method, termed as “pre-whitening,” is used to remove the
serial correlation, if present, from the data series before examining the trends. The
pre-whitening approach involves calculating the serial correlation and removing the
correlation if the calculated serial correlation is significant at the 95% confidence
Materials and Methods
27
____________________________________________________________________
interval (e.g., Douglas et al. 2000; Yue and Wang 2002). The following equation
was applied for this purpose:
Yt = X t − r1 X t −1
(6)
where, Yt is the pre-whitened series value for time interval t, Xt is the original
time series value for time interval t, and r1 is the estimated first serial correlation
coefficient. Data were normalized before pre-whitening was carried out, by
subtracting the mean and dividing the result by the standard deviation.
In this study, most of the studied variables did not show significant serial
correlation. However, when a significant serial correlation was noted, the trend
results for that particular case were mentioned for the pre-whitened data.
2.1.3. Hydrological modeling
The two hydrological models Hydrologiska Byråns Vattenbalansavdelning (HBV)
and Soil Water Assessment Tool (SWAT) were used. A brief description of these
models is given below.
The HBV model
The HBV model was used for regionalization purpose to estimate time series of
streamflow at poorly gauged sites. This model was selected for the following
reasons: a) its model structure is simple but flexible and can be adapted to local
conditions. For instance, a catchment can be subdivided into different elevation and
vegetation zones, which fact was important to model catchments in the mountainous
Karkheh Basin, b) it is not very data-intensive and most of the data needed are
readily available, c) it has been widely used worldwide, in particular in snowinfluence climates, but recent studies demonstrate its applicability in semi-arid
environments too (e.g., Lidén and Harlin 2000; Love et al. 2010), and d) a number of
studies have demonstrated its suitability in regionalization studies (e.g., Seibert
1999; Merz and Blöschl 2004; Götzinger and Bárdossy 2007).
The HBV model (Bergström 1992) is a conceptual rainfall-runoff model which
simulates daily discharge using as input variables daily rainfall, temperature and
daily or monthly estimates of reference evapotranspiration (ETo). The model
consists of different routines representing the snow accumulation and snowmelt by a
degree-day method, recharge and actual ET as functions of the actual water storage
in a soil box, runoff generation by two linear reservoirs with three possible outlets
(i.e., runoff components), and channel routing by a simple triangular weighting
function. Further descriptions of the model can be found elsewhere (Bergström
1992; Seibert 1999; 2002; Uhlenbrook et al. 1999). The version of the model used in
this study, “HBV light” (Seibert 2002), corresponds to the version HBV-6 described
by Bergström (1992) with only two slight changes. Instead of starting the simulation
with some user-defined initial state values, this version uses a warming-up period
28
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
during which the state variables evolve from standard initial values to more
appropriate values for the given hydro-meteorological conditions. The length of the
warming-up period generally depend on the catchment response characteristics and
length of the available data. For this modeling study, the visual inspection of the
observed and simulated runoff for the study catchments revealed that a warm-up
period should be more than 6 months, and after this period observed and simulated
runoff starts matching satisfactorily. Furthermore, the restriction that only integer
values are allowed for the routing parameter MAXBAS has been removed, which
enables a somewhat more realistic parameterization of the runoff routing processes.
The function of MAXBAS is to distribute runoff generated during a time period into
the following time steps. In the original version of the HBV model (Bergström 1992)
computations in both the snow and soil routine are performed individually for each
elevation zone before the groundwater recharge of all zones is lumped in the
response routine. In the model version used in this study, the upper box in the
response function is treated individually for each elevation zone in addition to the
separate computations in the snow and soil routines. This version is considered more
logical than the standard HBV versions, especially for use in a mountainous area like
the Karkheh Basin.
The SWAT model
The Soil Water Assessment Tool (SWAT) was used for a detailed investigation of
hydrological processes and water resources variability and to assess the impact of
various water management interventions. This model was selected because: a) it
possesses adequate representation of physical processes governing hydrology and is
particularly suitable for application to large river basins, b) it is well suited to the
proposed research questions on understanding the hydrological processes with
limited amount of data, c) it provides a wide range of options for testing “what if”
scenarios related to agricultural water management, climate and land use changes
etc., and d) it is freely available.
SWAT is a widely used process-based semi-distributed catchment model
developed by the Agricultural Research Service of the United States Department of
Agriculture (USDA) over the last 30 years and is available free of charge as a public
domain model (Arnold et al. 1998; Srinivasan et al. 1998; Arnold and Fohrer 2005;
Neitsch et al. 2005; Gassman et al. 2007). SWAT is developed to predict the impact
of land and water management practices and of climate change on water, sediment
and agricultural chemical yields in large complex watersheds with varying soils,
land use and management conditions over long periods of time. It has gained
international acceptance as an interdisciplinary tool suitable for applications in large
river basins with varying degree of biophysical, climatic and water management
settings (e.g., Gassman et al., 2007). The model has been widely applied throughout
the world for dealing with a wide range of issues related to hydrology, water
management, climate change impacts, land use impacts, best management practices,
conservation agriculture, sedimentation and water quality etc. (e.g., Weber et al.
2001; van Griensven and Bauwens 2003; Arnold and Fohrer 2005; Chaplot et al.
2005; Jayakrishnan et al. 2005;Vandenberghe et al. 2005; Faramarzi et al. 2009,
Materials and Methods
29
____________________________________________________________________
Githu et al. 2009a, 2009b). A comprehensive review on the SWAT model and its
applications can be found in Gassman et al. 2007. A brief description of the model is
presented here. The detailed theoretical documentation can be found in Neitsch et al.
2005.
In the SWAT model, a river basin is subdivided into a number of subcatchments,
each subcatchment consisting of at least one representative stream. The
subcatchments are further divided into hydrologic response units (HRUs), which are
lumped land areas within the catchment comprising unique land cover, soil, and
slope combinations. The hydrology in SWAT is divided into two major divisions,
the first being the land phase of the hydrologic cycle, which controls the amount of
water, sediment, nutrient and pesticide loadings to the main channel in each
catchment and the second being the water routing phase of the hydrologic cycle,
which can be defined as the movement of water, sediments, nutrients, etc., through
the channel network.
The hydrologic cycle as simulated by the SWAT model is based on the following
water balance equation:
t
SWt = SWo +  (Rday − Qsurf − Ea − Wseep − Qgw )
(7)
i =1
where, SWt is the final soil water content (mm), SW0 is the initial soil water
content on day i (mm), t is the time (days), Rday is the amount of precipitation on day
i (mm), Qsurf is the amount of surface runoff on day i (mm), Ea is the amount of
evapotranspiration (ET) on day i (mm), Wseep is the amount of water entering the
vadose zone from the soil profile on day i (mm), and Qgw is the amount of return
flow on day i (mm). The surface runoff volume is calculated by using the Soil
Conservation Service (SCS) curve number equation. Potential evapotranspiration
can be estimated by one of the three methods: Penman–Monteith, Priestly and
Taylor or Hargreaves method. The actual ET is estimated on the basis of simulated
plant growth and soil water availability. The model calculates percolation when the
soil-water content exceeds the soil-field capacity and determines the amount of
water moving from one soil layer to the next by using a storage routing method. In
each subcatchment, the SWAT model simulates two groundwater aquifers: a shallow
aquifer that contributes to streamflow and a deeper aquifer that does not add to
streamflow within the modeled subcatchment. Streamflow is routed by using either
the variable storage or the Muskingum routing method.
There are numerous other processes represented in SWAT, such as water balance
for lakes/ponds/reservoirs, sediment erosion and sediment transport processes,
industrial and municipal pollution added through point sources, processes related to
transformation and movement of nitrogen and phosphorus.
The SWAT 2005 version is well linked to geographic information system (GIS),
ARC-GIS, which have further enhanced its abilities to deal with spatial information
for management, query, visualization and analysis. The model also has added
30
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
features for auto-calibration, sensitivity and uncertainty analysis. In this study, the
SWAT 2005 modeling system, version ARCSWAT 2.0 (Winchell et al. 2008), was
used.
Statistics used in the performance evaluation
The model performance was assessed by using three performance measures; NashSutcliffe Efficiency (NSE) (Nash and Sutcliffe 1970), Coefficient of Determination
(R2) and the mean annual volume balance (VB). These criteria are most widely
recommended and commonly used in hydrological modeling (e.g., ASCE 1993;
Gupta et al. 2009). The VB is estimated as a percentage difference between the
observed and simulated mean annual runoff. The equations for estimating NSE and
R2 are given below:
 (Q
NSE = 1 −
 (Q
obs
− Qsim )
obs
− Qobs
2
(8)
)
2
( (Q − Q )(Q − Q ))
=
 (Q − Q )  (Q − Q )
2
R
2
obs
obs
obs
2
obs
sim
sim
2
sim
(9)
sim
where: Qobs and Qsim refer to the observed and simulated discharges,
respectively. and Qobs and Qsim refer to the mean of the observed and simulated
discharges, respectively. The observed and simulated streamflows and their means
will have same units, i.e., expressed as m3/s in this study.
2.2.
Data Collection
The International Water Management Institute (IWMI: http://www.iwmi.cgiar.org/ )
in collaboration with local partners has conducted research through the Karkheh
Basin Focal Project (BFP) in Iran to address some of the issues and challenges
discussed in the previous chapter. This project was funded through the Challenge
Program on Water and Food (CPWF: http://www.waterandfood.org/). The main
aims of the Karkheh BFP were to provide a comprehensive and integrated
understanding of the water, food and environment issues in the river-basin context.
Various data sets were accumulated from global and local sources under this Project.
All of these data sets were managed under the Integrated Database Information
Materials and Methods
31
____________________________________________________________________
System (IDIS) - a database management project of IWMI and CPWF based in
Colombo, Sri Lanka. This research study was an integral component of the Karkheh
BFP, and thereby, a number of data sets collected through primary and secondary
sources under this project were used in this research. The major data sets used in this
study are listed in Table 2. Further details are provided in the relevant chapters.
The quality of collected data sets was checked in a number of ways, mainly
depending on the type of the data set and perceived uncertainties. For instance, land
use map prepared under Karkheh BFP by IWMI was considered of reasonably good
quality, because it used sound scientific basis in its preparation and was extensively
validated through field observations. The quality of hydro-climatic data sets was
examined by visual inspection of the tabular data and their graphical presentations.
Moreover, double mass analysis was used to check the consistency of the
hydrological time series (Change and Lee 1974; Linsley et al. 1982). In the double
mass analysis, if a linear relationship is found between an individual station and the
mean of its neighbours, or the remainder of the set within a basin, then it is inferred
that the data series has been recorded consistently over its history. The results of the
double mass analysis applied on the time series data of the selected flow gauging
stations found no deviations from the corresponding linear plots, indicating that
records were consistent.
Table 2. An overview of main data sets used in this study.
Category
Hydrology
Climate
Topography
Data
River discharge
Precipitation, temperature, relative
humidity, sunshine hours, wind speed
Digital elevation model ( DEM)
Soils
Digital map of the soils and soil
properties
Land use
Irrigation
Land use map
Irrigation diversions and
on-farm irrigation practices
Crops, yields, agronomic practices,
agricultural statistics
Agriculture
Data source
MOE, Iran
Meteorological organization, Iran,
MOE, Iran
Shuttle Radar Topography Mission
(SRTM) of USGS
Soil and Water Research Institute
(SWRI),
Iran,
other
relevant
departments, and FAO 1995 soil map
of the world
IWMI Karkheh BFP
IWMI Karkheh BFP and relevant
Iranian sources
IWMI Karkheh BFP and relevant
Iranian sources
3. STREAMFLOW VARIABILITY AND WATER
ALLOCATION PLANNING5
3.1
Introduction
Arid, semi-arid and subhumid regions are called water limited environments and
occupy about half of the global land area (Parsons and Abrahams 1994). Changes in
water availability can have serious repercussions on the sustainability of these
sensitive environments. The pressure on water and other natural resources is
increasing in these areas as demands for water for human uses are growing rapidly
(e.g., Newman et al. 2006). For instance, in the dryland Mediterranean regions, large
increases in population, development of irrigated agriculture and rise of living
standards have drastically increased the water use and in many basins future needs
are hard to satisfy as many aquifers are already overexploited and surface water
resources are endangered (Cudennec et al. 2007). Southern Africa faces similar
challenges (e.g., van der Zaag 2005). The expected regional climate change
(Christensen et al. 2007) poses yet another dangerous alteration of the hydrological
regimes in these regions. This will also cause change in the water demand pattern,
with an expected two-thirds of the world facing an increase in irrigation demand
(Döll 2002).
The semi-arid to arid Karkheh Basin has a fragile balance between
environmental and human uses of natural resources and demands for water are
increasing and sustainable management of water resources has become an important
issue. The main challenges related to land and water resources are land degradation,
soil erosion, low water and land productivity, groundwater depletion and growing
competition for water among upstream and downstream areas and among different
sectors of water use such as irrigation, domestic, hydropower and environment
(CPWF 2005). In this river basin, massive irrigation development is on the way, but
the knowledge and understanding of basin hydrology (including the spatio-temporal
water balance variations) and impacts of these developments on other users and
water uses across the basin are patchy.
Quantitative and holistic knowledge of basin hydrology becomes essential as
water-management needs become complex. Molle et al. (2004) concluded that, as
water demands increase and more and more water is allocated to different uses, the
management of water resources becomes increasingly complex due to the huge
5
This chapter is mainly based on, but not limited to, the paper Analysing streamflow
variability and water allocation for sustainable management of water resources in the semiarid Karkheh River Basin, Iran by Masih, I., Ahmad, M.D., Uhlenbrook, S., Turral, H. and
Karimi, P. 2009. Physics and Chemistry of the Earth 34 (4-5): 329-340.
34
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
number of interacting factors such as upstream-downstream impacts, increasing
impacts on the environment and changes in de facto water rights. They have argued
that under such conditions, increasing the knowledge of the basin hydrology is
essential for constructing a sound and sustainable water management. A sound
knowledge of basin hydrology is essential for effective water allocation policies so
that negative impacts can be avoided, minimized or mitigated (Green and Hamilton
2000). Hydrological analysis provides the basis for detailed accounting of water use
and productivity (Molden and Sakthivadivel 1999). It is a basic requirement for
water resources development and management evaluations and decision making
related to a) assessing water availability, b) understanding the balance between the
actual use resource availability, c) improving water allocation decisions, d)
monitoring the performance of water use, and e) formulating environmental flow
requirements and working out ecosystem restoration strategies.
This chapter provides a comprehensive analysis of spatio-temporal variability of
the surface water hydrology over the period of 1961 to 2001 in the Karkheh Basin.
Additionally, basin-level water accounts are evaluated for the year 1993-94 and
challenges for sustainable management of water resources are highlighted.
3.2.
Data and Methods
For this study seven streamflow gauging stations of the main rivers (as shown in
Figure 6 and Table 3) were selected. The rationale for selecting these stations
includes their geographical importance, availability of consistent length and quality
of the records. Out of the seven stations, three stations namely Pole Chehre at the
Gamasiab River, Ghore Baghestan at the Qarasou River and Pole Dokhtar at the
Kashkan River are located at the outlet of their respective sub-basins. The Holilan at
the Saymareh River represents the combined effect of the hydrologic characteristics
of the upstream sub-basins Gamasiab and Qarasou. The Jelogir at the Karkheh River
is located upstream of Karkheh dam and the Paye Pole station located downstream
of the Karkheh dam is important for water supplies for hydropower and downstream
flows for irrigation and environment. The Hamedieh station is the last gauging
station before the Karkheh River routes towards Hoor-Al-Azim swamp and hence is
important for environmental flows further downstream, i.e., towards Hoor-Al-Azim
swamp.
The analysis was conducted using daily streamflow data for the period 19612001. This data set was used for the analysis of central tendency and dispersion,
flow duration analysis, base flow separation and water accounting. Basin-level
accounting of water use was carried out using available data for the year 1993-94. A
description of these methods is provided in section 2.2 of this thesis.
The data on water accounting components were accumulated from the study of
JAMAB (1999) who conducted comprehensive water balance investigations for the
hydrological year 1993-94. A brief description of their methodology is provided
here; further details can be found in JAMAB 1999. The estimates were mainly based
Streamflow Variability and Water Allocation Planning
35
____________________________________________________________________
on the data related to climate, runoff and water uses. For the purpose of a detailed
water balance analysis, the basin was divided into 47 subcatchments, and the results
were aggregated to the basin level. The inflow components of the water accounting
were composed of precipitation, inflow from outside of the basin and changes in
surface and subsurface storage. The precipitation data were based on 61 climatic
stations distributed in or close to the basin. Changes in the subsurface storage were
estimated based on the groundwater measurements related to changes in the water
level, specific yield and domain area. Since there was no major storage dam in the
basin during 1993-94, the surface storage was considered zero. Inflow from outside
of the basin was zero, as no water was diverted to the Karkheh from outside of the
basin. The actual ET was estimated through empirical equations calibrated for
selected locations in the basin where detailed data on climate and water balance
were available. The actual ET from diversions for agricultural and other purposes
was estimated as the difference between the total abstraction and return flows. The
return flows were estimated for industrial, domestic and agricultural sectors for each
of the 47 subcatchments, and were based on field observations. The outflow from
the basin was composed of outflow from rivers, drains and subsurface outflow. The
subsurface outflow was regarded as zero, whereas, outflow from rivers and drains
was based on the observed records. The data on committed and uncommitted water
were not available, but are necessary to complete the water accounting exercise. We
estimated the committed water to the range of 10 to 50% of the available annual
streamflows. This estimation was based on the study of Tennant (1976) who
suggested that allocating 50% of the available streamflows to the river ecosystems
can maintain healthy ecosystems whereas the minimum flows should be 10%,
though the ecosystem degradation will be inevitable at this level of allocations.
Table 3. Geographical characteristics of the selected river stations.
Name
river
of
Name of station
Longitude
(degrees East)
Latitude
(degrees
North)
Elevation
(masl)
Drainage
area (km2)
Gamasiab
Pole Chehre
47.43
34.33
1,280
Qarasou
Ghore Baghestan
47.25
34.23
1,268
5,370
Saymareh
Holilan
47.25
33.73
1,000
20,863
Kashkan
Pole Dokhtar
47.72
33.17
650
9,140
Karkheh
Jelogir
47.80
32.97
450
39,940
Karkheh
Paye Pole
48.15
32.42
125
42,620
Karkheh
Hamedieh
48.43
31.50
20
46,121
Data source: MOE, Iran.
10,860
36
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
3.3.
Results and Discussion
3.3.1. Spatial and temporal variability of daily streamflow regimes
The daily streamflows show large variability within a year and between years, as
exemplified in Figures 10 and 11. However, the general temporal patterns are quite
synchronized, with rising and falling limbs of the hydrographs most often
corresponding to similar timings, when different streams or flow behaviors at
different locations on a single river are compared. For tributary rivers (Figure 10),
the highest streamflow is observed at Holilan, which aggregates the streamflows
coming downstream from Pole Chehre and Ghore Baghestan. It is pertinent to note
that despite higher drainage area, Pole Chehre has lower streamflows than Pole
Dokhtar. This could be mainly attributed to comparatively lower precipitation and
higher agricultural water use in the case of Pole Chehre. For the Karkheh River at
Jelogir, Paye Pole and Hamedieh (Figure 10), the flows are very similar to one
another. This could be attributed to the fact that most of the streamflows are
generated before Jelogir. Although there are some abstractions used for irrigation
downstream of Jelogir and Paye Pole, these were not high enough to cause major
differences in the flow regimes, mainly due the absence of any major water
infrastructural project (e.g., large dams) during the period under study. However, it
could be anticipated that due to operations of the Karkheh Dam (operational since
2001) and new irrigation schemes, the streamflow regimes of these three stations
would be markedly different from one another.
The high flow events are mainly concentrated in the months from November to
May, particularly in February and March. The duration of these events varies largely
depending on the precipitation timing and snowmelt conditions. Generally, high
flow events of small duration (1-5 days) occur due to high rainfall events, but the
high flows prevailing for a few weeks to a couple of months, mainly observed from
February to May, are caused by the snowmelt and the combined effects of snowmelt
and rainfall. The low flow regime is marked from June to September. The high
spatial and intra-annual variability in the streamflows is mainly governed by the
seasonality of climate and by factors such as land use, geology, soils and
topography. Most of the precipitation occurs during winter, both in terms of rain and
snow, and in spring mainly as rain. The snowfall occurs from December to March,
with the highest amounts in January and February. The amount of snow varies
within the basin, with upper parts receiving more than the middle parts and no
snowfall in the lower arid region. A recent study on snow cover in the Zagros
mountains by Saghafian and Davatalab (2007) showed that the snow water
equivalent to the mountainous parts of the Karkheh Basin is about 75 mm/yr., which
is about 17% of the long-term mean annual precipitation in the basin. Moreover, the
amount and distribution of snow are strongly influenced by elevation varying from
44 mm/yr. in locations less than 1,500 masl to 245 mm/yr. in locations over 3,500
masl.
Streamflow Variability and Water Allocation Planning
37
____________________________________________________________________
Precipitation at Kermanshah
Pole Chehre
Holilan
Pole Dokhtar
Ghore Baghestan
0
750
25
500
50
250
75
0
Precipitation (mm/d)
Streamflow (m 3/s)
1,000
9/26/1963
8/27/1963
7/28/1963
6/28/1963
5/29/1963
4/29/1963
3/30/1963
2/28/1963
1/29/1963
12/30/1962
11/30/1962
10/31/1962
10/1/1962
100
Date
Precipitation at Kermanshah
Jelogir
Paye Pole
Hamedieh
0
2,000
25
1,500
50
1,000
75
500
0
Precipitation (mm/d)
Streamflow (m 3/s)
2,500
9/26/1963
8/27/1963
7/28/1963
6/28/1963
5/29/1963
4/29/1963
3/30/1963
2/28/1963
1/29/1963
12/30/1962
11/30/1962
10/31/1962
10/1/1962
100
Date
Figure 10.
Intra-annual variability of mean daily streamflows, illustrated for the
data of the hydrological year 1962-63.
(Note: The precipitation at Kermanshah was used as an example in this figure and in few other
figures in this chapter. The main aim is to illustrate the general pattern of the precipitation in the study
basin. The major reason for selecting this station was due to availability of good quality long-term daily
data series at this site. It is important to note that the precipitation values observed at this station are not
fully representative of the amount and distribution of precipitation in the whole basin. More details on
precipitation dynamics and their influence on runoff are given in chapter 4 and 6.)
38
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Precipitation at Kermanshah
Streamflow at Jelogir
0
6,000
50
4,000
100
3,000
2,000
150
Precipitation (mm/d)
Streamflow (m 3/s)
5,000
1,000
200
10/1/2000
10/1/1997
10/1/1994
10/1/1991
10/1/1988
10/1/1985
10/1/1982
10/1/1979
10/1/1976
10/1/1973
10/1/1970
10/1/1967
10/1/1964
10/1/1961
0
Date
Figure 11. View of the inter-annual variability of mean daily streamflows, illustrated
for the streamflows at Jelogir station (1961-2001).
This seasonality of precipitation and its distribution into rain and snow have a
major influence on the streamflows, indicated by the high flows from March to May
resulting from the combined effect of snowmelt and rainfall. The precipitation
recharged to subsurface, later supports the streamflows and, therefore, despite very
less precipitation during summer (July to September), the streams still flow, though
flows are less than those occurring in winter and spring seasons. In terms of interannual variability, precipitation depicts the most notable variability compared to
other physiographic catchment characteristics and, hence, is likely to have a strong
influence on the inter-annual variability of the streamflows. Figure 11, clearly shows
that higher streamflows correspond to more precipitation and vice versa.
The spatio-temporal differences in the flow regime can be investigated further
through the FDC analysis, as illustrated in Figure 12 showing selected exceeding
percentiles of the streamflows normalized by the drainage area. The actual
streamflows are also provided in Table 4. It is worth noting from Figure 12 that the
FDC of Pole Dokhtar plots higher compared to all other stations, even to those with
higher streamflows (e.g., Paye Pole), whereas the FDC of Pole Chehre plots the
lowest. This is attributed to higher specific runoff for Pole Dokhtar than for Pole
Chehre and other locations. Furthermore, the steeper slopes of FDCs observed at
Pole Chehre and Ghore Baghestan indicate less stable flow regimes having lower
proportions of the base flow than at Pole Dokhtar. The flow regime at Holilan and
Jelogir demonstrates the net effects of the upstream tributaries.
Streamflow Variability and Water Allocation Planning
39
____________________________________________________________________
The base flow constitutes a quite significant part of the total streamflows all
across the basin, particularly for Pole Dokhtar, Jelogir, Paye Pole and Hamedieh, as
indicated in Figure 12 and by the annual Base Flow Index (BFI) values close to 0.5
for these gauging stations (Table 5). These BFI estimates suggest that the role of
slow flow part of the hydrograph is quite significant in sustaining the streamflows in
middle and lower parts of the basin as compared to upper parts showing lower
values of BFI (e.g., Pole Chehre’s BFI is 0.36), where quick flow dominates
volumetrically the overall flow regime.
% of time indicated streamflow is equaled or exceeded
0
10
20
30
40
50
60
70
80
90
100
10.00
Streamflow (mm/d)
1.00
0.10
0.01
Pole Chehre
Ghore Baghestan
Holilan
Pole Dokhtar
Jelogir
Paye Pole
Hamedieh
0.00
Figure 12. The flow duration curves (FDCs) of selected gauging stations.
(FDC plots are based on selected flow percentiles extracted from the daily streamflow data for the period
October 1, 1961 to September 30, 2001.)
The main reasons for the stable base flow regime and a less steeper FDC slope in
the case of Pole Dokhtar than in o Pole Chehre and Ghore Baghestan are likely to be
the higher precipitation amounts in the middle parts of the basin, higher proportion
of forest area which leads to higher infiltration of precipitation that later slowly
discharges to rivers via subsurface routes and comparatively fewer irrigated areas in
the middle parts of the basin (which mean less water withdrawals from streams and
aquifers). The flow regimes of the Karkheh River at Jelogir, Paye Pole and
Hamedieh are largely similar to one another, with slightly more stable base flows in
the case of Paye Pole. This could be attributed to the presence of the Karkheh Lake
just above the Paye Pole station providing some attenuation to the streamflows and
then contributing stored water as base flows. The impact of the Karkheh Dam is not
evident in this analysis because the dam started operations only in 2001. Evaluating
the impact of dams on flow variability was beyond the scope of this research.
40
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
However, it is anticipated that the natural flow regime of the Karkheh River below
the Karkheh dam would be changed as a result of the reservoir operations. Further
details on dams impacts and operational strategies can be found at number of studies
conducted in Iran (e.g., Manouchehri and Mahmoodian 2002; Mousavi et al. 2004;
Karamouz et al. 2003, 2006, 2008, 2011; Ganji et al. 2007; Kerachian and Karamouz
2007; Zahraie et al. 2008; Zahraie and Hosseini 2009). For instance, the research
studies conducted by Karamouz et al. (2006, 2008, 2011) for the Karkheh dam have
highlighted the possible conflicts in downstream water availability between various
sectors of water use below the Karkheh dam. These investigations emphasized the
need of development and adoption of sound reservoir operating policies, such as
those recommended by these modeling based studies, in order to ensure adequate
water supplies in terms of quantity and quality for the different downstream uses
related to environment, agriculture and domestic sectors.
The presented BFI values are quite sensitive to the filter parameter ‘α’ (equation
1). Therefore, uncertainty of the estimated BFI was high, which warrant caution in
the interpretation and use of the results on the base flow estimation. Nevertheless,
the estimated values provide useful insights to understand the role of the base flow
as part of the total streamflow and for understanding its spatio-temporal variability
in particular. Further studies are recommended to investigate this important aspect of
the flow regime, for instance, using different hydrograph separation techniques such
as tracers, different analytical methods and hydrological modeling (e.g., Uhlenbrook
et al. 2002; Lim et al. 2005; Gallart et al. 2006; Mul et al. 2008).
Table 4. Various exceedance percentiles of daily streamflow (m3/s) for selected
locations in the Karkheh Basin, Iran.
Q1
Q5
Q10
Q20
Q30
Q40
Q50
Q60
Q70
Q80
Q90
Q95
Q99
Pole
Chehre
245.0
126.9
87.1
51.9
34.9
25.0
16.4
8.9
5.5
3.5
1.8
1.1
0.0
Ghore
Baghestan
149.0
80.8
55.2
34.0
21.9
15.0
11.0
7.9
5.9
4.2
2.5
1.7
0.4
Holilan
518.0
266.0
192.0
113.0
73.6
53.9
35.6
24.0
16.0
11.5
7.0
4.7
2.4
Pole
Dokhtar
325.0
171.4
116.2
71.8
48.0
36.0
28.6
22.9
19.3
15.5
12.0
9.7
5.6
Jelogir
924.0
516.0
365.0
233.0
156.0
118.0
89.6
68.8
54.0
42.0
31.0
24.4
12.9
Paye
Pole
1,120.0
588.9
418.0
265.0
183.0
135.0
101.0
80.0
66.5
52.0
42.0
34.9
25.0
Hamedieh
1,018.0
563.0
393.0
245.0
173.0
122.0
84.0
61.6
47.6
36.0
24.0
16.5
8.7
Note: These exceedance percentiles are extracted from the FDC analysis of the daily data for the
period of October 1, 1961 to September 30, 2001.
Streamflow Variability and Water Allocation Planning
41
____________________________________________________________________
3.3.2. Spatial and temporal variability of monthly streamflows
Mean monthly discharges at the selected river stations are shown in Figure 13. The
hydrograph peaks occur in March and April, roughly one month in lag of
precipitation. This could be attributed to contributions of snowmelt in the late winter
and early spring seasons as well as contributions of water into streams after passing
through different hydrological pathways (such as groundwater). The peak flows are
observed in April at all the examined stations whereas minimum flows occur in
September. Although most of the discharge takes place in winter (about 41%) and
spring (about 39%), all the main tributaries of the Karkheh River have some flow all
around the year. The hydrograph separation analysis indicates that the base flow
contributions are mainly responsible for keeping the streams flowing during half of
the water year, particularly from June through September (Table 5).
The river flows show quite high variability both with respect to space and time,
as indicated by high CV (Figure 14). The maximum values of CV are observed for
November and it corresponds to river flows at all of the seven selected stations in the
basin ranging from 0.96 for Pole Dokhtar to 1.77 for Pole Chehre. Minimum values
of CV are observed in February with the spatial variation ranging from 0.44 to 0.53.
For rest of the months, the values are in the range of 0.4 to 0.9.
Pole Chehre
Pole Dokhtar
Hamedieh
Ghore Baghestan
Jelogir
800
0
700
25
600
50
500
75
400
100
300
125
200
150
100
175
0
200
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug. Sept.
Month
Figure 13. Mean monthly discharge at selected locations in the Karkheh Basin,
Iran.
Mean monthly precipitation at
Kermanshah (mm/month)
Mean monthly streamflow (m 3/s)
Precipiation at Kermanshah
Holilan
Paye Pole
42
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Table 5. Base Flow Index (BFI) for selected locations in the Karkheh Basin, Iran.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Annual
Pole
Chehre
0.37
0.22
0.25
0.32
0.33
0.28
0.39
0.72
0.94
0.90
0.88
0.71
0.36
Ghore
Bagestan
0.64
0.43
0.38
0.40
0.34
0.28
0.39
0.71
0.95
0.93
0.93
0.85
0.41
Pole Chehre
Jelogir
Holilan
0.55
0.34
0.31
0.36
0.33
0.28
0.40
0.70
0.94
0.98
0.97
0.86
0.38
Pole
Dokhtar
0.72
0.55
0.50
0.51
0.43
0.35
0.46
0.71
0.96
0.97
0.96
0.90
0.49
Ghore Baghestan
Paye Pole
Jelogir
Paye Pole
0.65
0.47
0.41
0.45
0.39
0.34
0.46
0.74
0.98
0.97
0.96
0.89
0.47
Holilan
Hamedieh
Hamedieh
0.69
0.49
0.42
0.45
0.41
0.38
0.49
0.78
0.97
0.94
0.90
0.87
0.49
0.64
0.43
0.38
0.41
0.39
0.36
0.47
0.74
0.93
0.87
0.89
0.84
0.46
Pole Dokhtar
Coefficient of variation, CV (-)
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Month
Figure 14. The variability of mean monthly streamflows, indicated by the
Coefficient of Variation (CV) at selected locations in the Karkheh Basin, Iran.
This high level of variability stresses the importance of understanding the
reliability of flow for meeting the needs of different users. The exceeding percentiles
of flow reveal the assurance level associated with various flow values. The selected
flow percentiles representing the reliability of mean monthly flows are given in
Table 6. The full range of flow percentiles is given for the Paye Pole station as an
example (Figure 15). For example, the minimum value of Q70 at Paye Pole
corresponds to the month of September, indicating that the mean monthly
streamflow of 41.4 m3/s is likely to be available for 28 out of 40 months according
to the study period (70% of the time or 7 out of 10 months). The maximum mean
monthly flow of 285.5 m3/s with a reliability of 70% is available in March at the
Paye Pole.
Streamflow Variability and Water Allocation Planning
43
____________________________________________________________________
200
1,500
150
1,200
October
November
900
100
600
50
300
0
0
0
25
50
75
100
800
0
25
50
75
100
800
600
600
Decem ber
400
400
200
200
0
January
0
0
25
50
75
100
800
600
0
1,500
25
50
1,200
February
100
March
900
400
75
600
200
300
0
0
0
25
50
75
100
0
25
50
75
100
800
1,500
1,200
600
April
900
May
400
600
200
300
0
0
0
25
50
75
100
0
25
50
75
100
200
300
250
200
150
100
50
0
150
June
July
100
50
0
0
25
50
75
100
150
0
25
50
75
100
150
August
100
Septem ber
100
50
50
0
0
0
25
50
75
100
0
25
50
75
100
Figure 15. The reliability of the mean monthly surface water availability, indicated
by the monthly FDCs at the Paye Pole station at the Karkheh River.
(Note: The x-axis shows percentage of time mean monthly flow was equaled or exceeded, whereas the yaxis shows mean monthly streamflows (m3/s). These exceedance percentiles are extracted from the FDC
analysis of the flow data for the period October 1, 1961 to September 30, 2001.)
44
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Table 6. Selected values of the streamflow percentiles (m3/s) for each month at the
studied stations across the Karkheh River system.
Indicator
Oct.
Nov.
Dec.
Jan.
Pole Chehre station at the Gamasiab River
Q10
16.0 41.8
55.3
57.3
Q30
8.0
26.2
35.2
37.8
Q50
5.5
14.9
26.7
31.2
Q70
4.0
11.5
21.1
26.5
Q90
2.3
6.6
14.3
18.3
Ghore Baghestan station at the Qarasou River
38.2
34.2
20.3
9.1
Q10
26.3
16.1
10.5
5.9
Q30
16.4
13.3
9.1
5.2
Q50
12.0
10.7
6.5
4.1
Q70
8.9
6.4
4.3
2.4
Q90
Holilan station at the Saymareh River
101.2 124.8
28.7 81.3
Q10
79.3
71.8
18.8 46.9
Q30
60.8
55.8
15.7 30.8
Q50
52.4
42.3
11.7 24.2
Q70
33.8
26.0
14.2
7.5
Q90
Pole Dokhtar station at the Kashkan River
Q10
31.1 53.5
85.6
90.4
Q30
21.9 35.5
57.7
57.6
Q50
19.2 26.2
37.3
41.4
Q70
15.4 20.9
29.1
35.2
Q90
11.4 17.2
21.5
26.5
Jelogir station at the Karkheh River
81.4 153.0 246.4 303.4
Q10
64.9 107.2 168.7 162.7
Q30
126.8 136.8
48.7 82.6
Q50
114.9
92.7
42.1 69.3
Q70
79.8
67.8
33.5 49.6
90
Paye Pole station at the Karkheh River
Q10
92.1 207.0 323.3 378.5
Q30
74.6 143.5 200.2 233.6
Q50
58.2 100.1 150.9 178.2
Q70
51.3 75.9
102.3 137.5
Q90
43.8 57.9
81.4
82.0
Hamedieh station at the Karkheh River
79.9 183.2 298.3 356.6
Q10
59.5 125.5 183.2 214.9
Q30
128.5 160.1
48.0 82.3
Q50
120.8
87.4
35.3 61.5
Q70
61.5
60.0
19.2 40.5
Q90
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
78.7
57.4
44.8
34.6
23.7
171.5
104.4
78.6
63.2
33.2
186.9
110.1
84.3
59.9
36.5
139.0
61.6
39.6
29.7
11.3
24.6
12.4
8.6
5.6
2.9
10.4
5.9
4.0
2.4
1.2
6.0
4.0
2.3
1.5
0.9
5.4
4.0
2.6
1.6
0.7
55.9
40.5
25.7
19.7
13.7
124.0
66.8
55.0
35.9
22.4
123.5
77.5
51.2
37.7
23.5
80.4
43.7
34.0
21.7
12.5
23.6
16.4
11.2
6.9
4.8
12.0
8.1
5.9
3.5
2.4
9.5
5.5
4.0
2.3
1.4
7.0
4.7
3.5
2.3
1.7
178.9
127.4
96.1
74.0
47.6
395.2
251.2
175.0
133.6
72.9
435.6
241.7
179.0
124.6
82.4
281.2
127.2
98.6
65.9
32.8
76.7
37.6
26.5
15.1
11.0
29.1
18.7
15.3
7.9
5.6
17.4
12.6
10.2
5.8
3.2
15.4
11.9
9.8
5.9
3.8
129.1
85.9
66.0
48.7
33.8
224.5
142.3
110.7
83.3
51.2
222.7
153.9
113.1
70.6
48.1
151.9
86.2
66.5
41.8
22.3
52.7
32.8
27.9
18.3
11.8
34.4
23.1
19.1
13.8
9.2
26.4
19.5
15.4
12.4
8.6
22.5
18.3
14.9
12.3
9.5
375.6
275.6
194.0
168.4
112.6
678.2
423.9
326.9
244.3
162.7
739.5
432.7
374.5
236.4
148.8
495.4
275.4
200.5
137.8
69.8
152.1
102.3
76.3
52.9
34.8
98.8
61.7
47.4
34.5
24.0
71.6
49.4
40.5
30.6
20.9
63.3
44.5
38.5
31.1
21.4
445.0
334.1
248.8
191.8
114.3
729.8
474.5
363.4
285.5
154.7
847.8
485.1
406.9
261.8
162.4
540.8
314.5
219.0
157.3
91.8
178.3
117.7
96.0
61.7
46.4
104.2
80.0
61.9
48.9
36.7
86.9
63.5
50.8
43.3
32.0
79.4
60.9
46.6
41.4
31.4
442.2
301.3
240.7
163.5
94.3
757.9
483.1
331.2
245.2
130.8
796.8
444.4
373.2
237.7
143.6
531.1
318.8
202.7
148.8
68.8
192.6
105.0
65.5
48.6
23.8
92.7
57.8
49.6
31.8
19.5
67.0
42.8
38.7
27.0
14.3
60.2
45.0
36.4
26.1
13.4
Streamflow Variability and Water Allocation Planning
45
____________________________________________________________________
3.3.3. Long-term variability in annual surface water availability
The long-term temporal behavior in the annual river flows has similar patterns
throughout all the subbasins, where wet and dry years prevail over all areas
simultaneously (Figure 16). The annual values of CV fluctuate around 0.47 within a
range of 0.41 to 0.54. A comparison of mean and median annual water availability
indicates that the mean values are 0-7% higher than the median estimates (Table 7).
This exhibits the classic arid and semi-arid hydrology characteristic that the mean is
greater than the median but, in this case, not by a large margin at an annual scale
(only 4% on average). The maximum flow of 12.59×109 m3/yr. occurred in the wet
year of 1968-69 whereas the minimum flow of 1.92×109 m3/yr. corresponds to the
drought year of 1999-2000, at the Paye Pole station. In the time period of this
analysis, i.e., 1961 to 2001, the severest drought occurred from 1999 to 2001 though
the longer-term time series depicts both high and low flow years throughout the
study period. During this persistent drought the Gamasiab River ceased to flow
during part of the year, indicated by zero flow of 44 to 77 days in a year observed at
Pole Chehre but other examined locations recorded some flows.
Pole Chehre
Pole Dokhtar
Hamedieh
Ghore Baghestan
Jelogir
20,000
0
18,000
250
500
14,000
12,000
750
10,000
1,000
8,000
1,250
6,000
1,500
4,000
2000-01
1997-98
1994-95
1991-92
1988-89
1985-86
1982-83
1979-80
1976-77
1973-74
2,000
1970-71
0
1967-68
1,750
1964-65
2,000
1961-62
Annual surface water
availability (10 6 m3 /yr.)
16,000
Precipitation at Kermanshah
(mm/yr.)
Precipitation at Kermanshah
Holilan
Paye Pole
Hydrological year
Figure 16. Long-term variability in annual surface water availability across the
Karkheh Basin.
These large temporal variations indicate the high level of supply insecurity for
current and future increased withdrawals for human uses. The analysis of flow
46
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
duration curves (Table 7 and Figure 17) provides further insights into the assurance
levels of the annual availability of surface water across the Karkheh Basin. For
instance, the value of Q75 at Paye Pole is 4.08×109 m3/yr., which shows that this
much volume of surface water could be available for 75% of the time, i.e., 30 out of
40 years as per duration of the study. Further examination was done to ascertain the
assurance levels associated with mean annual water availability. For this, FDC plots
were generated, using annual data, (Figure 17) and the exceedance level of means
was noted for each station. This analysis indicated that mean annual surface water
availability has an assurance level of about 45% at the basin level, ranging from
40% for Pole Chehre to 52% for Pole Dokhtar. This shows that the annual mean is
biased towards hydrological years with high values for Pole Chehre (and also for
Ghore Baghestan) and, therefore, the median is a better measure of central tendency
for these stations. Furthermore, due to the construction of the Karkheh Dam and
ongoing irrigation schemes in downstream parts, one can anticipate that, during the
below-average/low-flow years, the most serious conflict would concern retention of
water in the Karkheh Dam for hydropower generation and reduced supplies to the
downstream agricultural users whose situation will be exacerbated by soil salinity
problems. This would also be accompanied by the diminished flows to river
ecosystem and floodplains as well as to the Hoor-Al-Azim Swamp further
downstream.
Table 7. Some basic streamflow statistics (106 m3/yr.) derived from the annual time
series of the streamflows of the period 1961-2001 at the selected flow gauges.
Mean
Standard
deviation
Minimum
Maximum
Median
Q5
Q10
Q25
Q50
Q75
Q90
Q95
Pole
Chehre
1,080
Ghore
Baghestan
722
Holilan
2,431
Pole
Dokhtar
1,639
540
198
2,851
1,003
2,416
1,684
1,303
1,022
766
549
294
392
104
1,914
712
1,844
1,183
957
716
419
353
268
1,277
607
6,193
2,292
6,042
4,250
2,977
2,343
1,499
1,168
871
667
645
3,206
1,637
3,081
2,455
2,064
1,645
1,113
854
778
Jelogir
4,974
Paye
Pole
5,827
2,115
1,790
10,773
4,692
8,958
8,227
6,193
4,836
3,562
2,601
2,230
2,512
1,916
12,594
5,590
10,755
9,280
7,756
5,651
4,082
3,020
2,404
Hamedieh
5,153
2,476
1,068
11,324
4,852
9,280
8,641
7,555
4,873
3,447
2,254
1,648
Streamflow Variability and Water Allocation Planning
47
____________________________________________________________________
3,000
3,000
Qarasou River gauged at Ghore Baghestan
Gamasiab River gauged at Pole Chehre
Annual flow (10 6 m 3 /yr.)
Annual flow (10 6 m 3/yr. )
2,500
2,000
1,500
1,000
500
6
3
Mean: 1,080×10 m /yr.
2,500
2,000
1,500
1,000
500
6
0
0
10
20
30
40
50
60
70
80
90 100
0
10
% of time flow is equaled or exceeded
3,500
Annual flow (10 6 m 3 /yr.)
Annual flow (10 6 m 3 /yr.)
Saymareh River gauged at Holilan
6,000
5,000
4,000
3,000
2,000
6
1,000
20
30
40
50
60
70
80
90 100
% of time flow is equaled or exceeded
7,000
3
Mean: 2,431×10 m /yr.
0
Kashkan River gauged at Pole Dokhtar
3,000
2,500
2,000
1,500
1,000
6
3
Mean: 1,639×10 m /yr.
500
0
0
10
20
30
40
50
60
70
80
90 100
0
% of time flow is equaled or exceeded
10
20
30
40
50
60
70
80
90 100
% of time flow is equaled or exceeded
14,000
14,000
Karkheh River gauged at Jelogir
12,000
Annual flow (10 6 m 3/yr.)
Annual flow (10 6 m 3 /yr.)
3
Mean: 722×10 m /yr.
0
10,000
8,000
6,000
4,000
2,000
6
3
Mean: 4,974×10 m /yr.
0
Karkheh River gauged at Paye Pole
12,000
10,000
8,000
6,000
4,000
2,000
6
3
Mean: 5,827×10 m /yr.
0
0
10
20
30
40
50
60
70
80
90 100
% of time flow is equaled or exceeded
0
10
20
30
40
50
60
70
80
90 100
% of time flow is equaled or exceeded
Annual flow (10 6 m 3/yr.)
14000
Karkheh River gauged at Hamedieh
12000
10000
8000
6000
4000
2000
6
3
Mean: 5,153×10 m /yr.
0
0
10
20
30
40
50
60
70
80
90 100
% of time flow is equaled or exceeded
Figure 17. The reliability of the annual surface water availability, indicated by
annual FDCs at the selected gauging stations across the Karkheh River system.
48
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
3.3.4. Overview of the basin-level water accounting
The results of the basin-level water accounting are provided in Figure 18 and Table
8. The gross inflow, net inflow and total depletion are 24.96×109, 25.08×109, and
19.94×109 m3/yr., respectively. The net inflow is slightly higher than gross inflow
due to the net addition of water from the subsurface storage, as discharge from the
groundwater reservoir was higher than the recharge by an amount of 0.12×109 m3/yr.
Evaporation from precipitation constitutes 82% (or 16.39×109 m3/yr.) of the total
depleted water (19.94×109 m3/yr.) in the basin. The portion of irrigation diversions
depleted as ET from irrigated areas is estimated as 3.21×109 m3/yr. The depletion of
water in municipal and industrial sectors is very small (only about 0.05×109 m3/yr.),
as most of the water diverted to these sectors generates return flows (about 76%).
The total outflow from rivers is 54% or 3.99×109 m3/yr. of the total annual
streamflow volume of 7.37×109 m3/yr. available in 1993-94.
Figure 18. Finger diagram presentation of the basin level water accounts of the
Karkheh river basin for the hydrological year 1993-94.
Streamflow Variability and Water Allocation Planning
49
____________________________________________________________________
Table 8.
Basin-level water accounts of the Karkheh Basin for the year 1993-94.
Water accounting indicators
Inflow
Gross inflow
Precipitation
inflow from outside of the basin
Storage Change
Surface
Sub surface
Net Inflow
Depletion/Consumption
Actual evapotranspiration (ET)
ET from plains and hills (including all land uses)
ET from Irrigation diversions to agriculture
ET from lakes and wetlands
ET from groundwater evaporation
Municipal and Industrial
Outflow from basin
Total outflow
Surface outflow from rivers
Surface outflow from drains
Subsurface outflow
Committed water (assumed for environment)
Uncommitted outflow (Total outflow–Committed water)
Value
(109 m3/yr.)
Total
(109 m3/yr.)
24.96
24.96
0
-0.12
0
-0.12
25.08
19.99
19.94
16.39
3.21
0.030
0.31
0.05
5.09
3.99
1.10
0.00
5.09-3.69 to 5.09-0.74
0.74 to 3.69*
1.40 to 4.35
Notes: Data Source: JAMAB 1999. * Values are calculated based on 10 and 50%, respectively, of the
total annual streamflows (7.374×109 m3/yr.) required for in-streamflows, as suggested in Tennant 1976.
Tennant (1976) suggested that 50% of the available freshwater flows are required
to maintain excellent conditions in associated river ecosystems and the level of the
minimum environmental flow requirements is 10%, though degradation of
ecosystems will be inevitable at this level of appropriation. In many countries, the
flow equivalent to Q90 (e.g., in Brazil and Canada) or Q95 (Australia and United
Kingdom) are taken as the minimum environmental flow requirements (Tharme
2003). Based on the values suggested by Tennant (1976), we estimated committed
water essentially required to support river ecosystem functions in the range of
0.74×109 to 3.69×109 m3/yr. It should be noted this is a very simple way to estimate
environmental flow requirements and does not account for specific species/life phase
habitat requirements, short-long-term changes in flow rates, and seasonal variability
or channel geometry. Most of the environmental flow assessment studies
recommend that to keep healthy, resilient and productive river ecosystems, water
management policies should aim to restore the natural flow regime of the rivers,
including flow variability, as much as possible (e.g., Poff et al. 1997; Richter et al.
1997). This requires detailed assessment of the flow characteristics of the Karkheh
Basin streams (e.g., magnitude, timing, frequency and duration, rate of change,
floods and low flows, etc.) and to explore further how to make balanced allocations
50
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
to environment and human demands under varying present and future flow
conditions.
However, based on these simple assumptions on uncommitted outflow from
rivers, in a year like 1993-94, available for further allocation to various uses, would
be in the range of 1.070×109 to 4.02×109 m3/yr. The situation in 1993-94, when
viewed in terms of future water allocation planning (Table 1), clearly highlighted the
high level of competition between environmental and human demand. The water
allocation to different sectors for 2001 was 4.95 ×109 m3/yr., which is about 60% of
the total renewable water resources available during the reference year 1993-94. The
allocation to different sectors will be 8.90×109 m3/yr. by the year 2025. Among them
the irrigation share will be the biggest (7.42×109 m3/yr.), which is almost equal to
the renewable water supplies in an average year. The flow duration analysis suggests
that planning on the basis of mean annual flows cannot provide the required
streamflows every year. The anticipated situation in low flow years may be more
stressful to the ecosystem health, if water allocations to human uses remain at the
same levels. This also highlights the increasing stress on groundwater resources that
are already overexploited, in some areas, particularly in the Gamasiab subbasin
(JAMAB 2006a) and greater challenge for managing dam supplies for hydropower
generation, irrigation and environment. Water allocated to the environmental sector
is fixed to around 0.5×109 m3/yr. (Table 1), which is even below 10% of the
streamflows available in the reference year 1993-94. This indicates that further
studies are required to assess the reasonable allocations for the environment, also
looking into the temporal patterns of streamflows whereby streamflows should
follow, to some extent, the natural patterns of flow variability. The management of
releases from the newly constructed Karkheh Dam and other reservoirs would be
critical to attain that, and will require more detailed scientific studies. Although the
Karkheh Dam is a carryover dam, and therefore, water stored during high flow years
can be used to meet demands during dry years. However, meeting the demands of all
sectors would be extremely difficult in the future, particularly during dry years. Its
additional complications were studied by Karamouz et al. (2006) who examined the
possibilities of conflicts arising among urban, agricultural and environmental sectors
located downstream of the Karkheh Dam due to deterioration of water quality as a
result of increased water allocation to agriculture and urban sectors under the current
water development and allocation policies. They have shown that if the current
water development planning is followed, then by the year 2021 the quality of water
flowing to the Hoor-Al-Azim Swamp would be deteriorated to the unacceptable
levels during most of the time in a year as a result of the decreased quantity of flows
and high salinity and agrochemical loads coming from agricultural return flows.
The water accounting exercise has generated useful information on the
availability of water and different pathways by which water resources were depleted
or moved out of the basin. The estimation of committed and uncommitted outflows
provided practical insights into the degree to which water resources can be further
developed. The analysis also highlighted trade-offs between different uses of water,
for instance, increasing allocations to irrigation will increase the depleted portion of
the water accounting and consequently reduce outflow from the basin that are likely
Streamflow Variability and Water Allocation Planning
51
____________________________________________________________________
to have a negative impact on the environment. In sum, this exercise is a simple way
of viewing current pathways of water in the basin, and comparing it with variability
in water supplies and future water allocations indicated trade-offs among different
sectors of water use.
3.4.
Concluding Remarks
This study demonstrates that the hydrology of the Karkheh Basin is governed by the
natural climatic characteristics of a semi-arid to arid region, which has unique
interactions with its diverse drainage areas, mostly located in the Zagros mountains.
High spatio-temporal variability is a strong feature of the hydrology of the Karkheh
Basin. For instance, the variability of streamflows within a month and between the
months is quite high, as indicated by the range of CV values ranging from 0.44 to
1.77. The highest variability is found in November whereas the lowest variability is
associated with February. In spatial terms, the highest variability is observed for
Pole Chehre and Ghore Baghestan, both located in the upper parts of the Karkheh
Basin.
The flow duration analysis presented in this thesis has generated further insights
into the hydrological variability, surface water availability and its expected water
security levels. For instance, the analysis has clearly shown that the flow regime of
Pole Chehre and Ghore Baghestan (i.e., upper parts of the basin) is dominated by
quick flow, whereas, base flow contributions are higher for Pole Dokhtar (i.e.,
middle parts of the basin) indicating a stabler flow regime for the latter station. The
FDC analysis at the annual scale further reveals that the mean annual surface water
availability has a security level of about 45%, ranging from 40 to 52% at the studied
gauging stations across the Karkheh Basin. For example, the mean and median
surface water availability at the Paye Pole station at the Karkheh River was
estimated as 5,827 × 106 m3/yr. and 5,590 × 106 m3/yr. Like all other stations, the
minimum and maximum had a wide range at Paye Pole, with values of 1,916 × 106
m3/yr. observed during 1999-2000 and 12,596 × 106 m3/yr. observed during 196869. The FDC analysis reveals that the amount of surface water available for 30 out
of 40 years over the period 1961-2001 (e.g., 75th percentile, Q75 ) at Paye Pole was
4,082 × 106 m3/yr. Furthermore, the FDC analysis has generated information on the
values of various exceeding percentiles of streamflows, which could serve as the
basis for water allocation planning.
The examination of water availability, variability, water accounting, and
allocation planning suggested that, on the whole, water allocations to different
sectors were lower than the totally available resources and, hence, the competition
among different sectors of water use was not alarming during the study period. This
was exemplified, for instance, by the facts that about half the total renewable
streamflows was flowing out of the basin during 1993-94, the amount which is
generally considered sufficient to maintain healthy ecosystems, as indicated by the
records at the Hamedieh station. However, considering the high range of variability
52
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
of the streamflows, changes in climate, land use and future water allocation
planning, it would be extremely difficult to meet the demands in future, i.e., by
2025, as planned allocation will reach close to the annual renewable water resources
available in an average climatic year. The competition between irrigation and other
sectors will increase further, particularly during dry years. The analysis conducted in
this study is helpful in gaining further insights into the hydrological variability of
surface water resources and can, in turn, be instructive for the (re)formulation of a
sustainable water resources development and management regime for the Karkheh
Basin.
4.
STREAMFLOW TRENDS AND CLIMATE LINKAGES6
4.1.
Introduction
Examining streamflow records for the detection of trends has received increased
attention of the scientific community over the last two decades (e.g., Lettenmaier et
al. 1994; Zhang et al. 2001) due to the growing need to secure water for human uses
such as hydropower and irrigation, as well as for aquatic ecosystems. In addition,
rising concern about climate change and its impacts on streamflow has been an
important driver of such studies (e.g., Cullen et al. 2002; Birsan et al. 2005).
Zhang et al. (2001), Burn and Elnur (2002) and George (2007) have illustrated
that significant changes in the hydrological regime of Canadian rivers were strongly
related to changes in precipitation and temperature. However, the observed trends
and climate linkages were not uniformly distributed spatio-temporally. Similar
observations were made by Lettenmaier et al. (1994) for the Continental United
States. Cullen et al. (2002) studied the relationship of monthly streamflow,
precipitation and temperature with the North Atlantic Oscillation (NAO) Index for
five Middle East rivers for 1938-1984. The study indicated that changes in NAO
strongly influence winter streamflows and climate. Strong linkages of NAO with
temperature in the Middle East were also reported by Mann (2002). Cullen et al.
(2002) stressed that, as increased greenhouse gases promote NAO’s upward trend,
future precipitation and winter flows will continue to decline in the study region.
Tu (2006) analyzed streamflow trends for the Meuse River Basin in Europe and
suggested that the streamflows were stable at an annual time scale, though some
significant increases were observed in spring flows and the flood regime. The study
concluded that most of these trends were related to climate variability and were
linked to changes in precipitation, which were strongly influenced by the changes in
the NAO and European atmospheric circulation patterns. Ceballos-Barbancho et al.
(2008) studied trends in annual and seasonal records of streamflow, precipitation and
temperature for the Duero River Basin, Spain for the period 1957-2003. They found
a decreasing trend in streamflow, which was strongly correlated to precipitation.
They also attempted to relate changes in streamflow with land use changes (forest
cover), but concluded that changes in plant cover were too far below the level of
making a significant impact on the streamflow. They also noted that time-dependant
6
This chapter is based on paper Streamflow trends and climate linkages in the Zagros
Mountains, Iran by Masih, I., Uhlenbrook, S., Maskey, S. and Smakhtin, V. 2011. Climatic
Change 104: 317-338. DOI: 10.1007/s10584-009-9793-x.
54
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
changes in the catchment characteristics other than climate were masked by the high
inter-annual variability of precipitation in the studied Mediterranean Region. Birsan
et al. (2005) studied streamflow trends in Switzerland for the period 1931-2000.
Their study concluded that the mountain basins are the most vulnerable
environments from the point of view of climate change. These findings are in line
with those of Beniston (2003) who emphasized the importance of, and need for,
more research and policy adaptation on the environmental change in the
mountainous basins across the world.
Most of the climatic studies in Iran have focused on studying precipitation
variability and classifying the country into different climatic regions (e.g., Domroes
et al. 1998; Dinpashoh et al. 2004; Alijani et al. 2008), trend detection in the
observed climatic data (e.g., Modarres and da Silva 2007) and studying the largescale atmospheric circulations and their linkages with the local climate (e.g.,
Nazemosadat and Cordery 2000; Alijani 2002). To date streamflow trends and their
linkages with climate are not well understood in Iran. Filling this gap is important
because a) Iran is primarily an arid country with high climatic variability, b) four of
its major rivers (Dez, Karun, Karkheh and Zayandeh Rud) originate from the Zagros
mountains and are, thus, vulnerable to climate change with potentially adverse
subsequent impacts for hydropower, agriculture and environment in the country.
Therefore, exploring the fundamental questions on the nature and scale of the
changes in climate and water availability is critical for informed water management
and adaptation.
The main objective of this work is to identify, quantify and analyze recent trends
in streamflow, precipitation and temperature using the mountainous, semi-arid
Karkheh River Basin as an example. The relationship between the NAO index with
the local climate (precipitation and temperature) is also investigated.
4.2.
Data and Methods
4.2.1. Hydrological and climate data and indices
For the analysis of streamflows, five stations located in the main rivers, namely Pole
Chehre, Ghore Baghestan, Holilan, Kashkan and Jelogir, were selected (Figure 19
and Table 3). The daily streamflow records used in this study cover a 40-year period
from October 1961 to September 2001. The months October and September refer to
the start and end of the hydrological year, respectively. The stations below the
Karkheh Dam were not included in this analysis because the presence of large
hydrological storages and diversions are likely to obscure the relationship between
streamflow and climate (e.g., George 2007). The stations used in this study generally
represent the natural flow variability induced by the climatic and other
physiographic factors. At some locations, water is diverted directly from streams for
agricultural purposes.
Streamflow Trends and Climate Linkages
55
____________________________________________________________________
Figure 19. Location of the Karkheh Basin in Iran and some of its important
features.
56
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
However, the scale of these abstractions could be considered negligible in
influencing the mainstream rivers under study, as the irrigated areas are reasonably
small compared to the catchment area, e.g., the total irrigated areas in Qarasou,
Gamasiab, Saymareh and Kashkan subbasins were 5, 12, 3 and 6%, respectively, of
the total subbasin area during 1993-94 (JAMAB 1999).
In this study, the examined streamflow variables were the mean annual and
monthly flows and the indicators describing the hydrological extremes which
included 1 and 7 days maximum and minimum flows, timing of the 1-day maxima
and minima, and the number and duration of high- and low-flow pulses. Low and
high pulses are defined as those periods during which daily mean flows drop below
the 25th percentile and exceed the 75th percentile, respectively. The threshold values
of 25th and 75th flow percentiles are derived from a flow duration analysis. The
streamflow variables selected reflect different aspects of a natural river flow regime,
i.e., magnitude, timing, duration, frequency and rate of change (Richter et al. 1997).
For the analysis of the climate, monthly climatic data on precipitation and
temperature for six synoptic stations, two located in the Karkheh Basin and four
located in the vicinity of the basin were used (Figure 19 and Table 9). There were a
few other climatic stations located inside and close to the basin, but they were not
used in this study because of shorter and incomplete records. The precipitation
analysis was confined to the months of October through May, when about 99 % of
the total annual precipitation occurs, and data for other months with almost
negligible rainfall were not analyzed. The indicators used in the study were total
monthly precipitation, number of precipitation days, number of days with
precipitation equal to, or greater than, 10 mm/d, maximum daily precipitation,
number of snow and sleet days and mean monthly temperature. These indicators
were selected from the list of available climatic indices at the selected stations,
mainly due to their importance in the hydrological processes. For instance,
maximum daily precipitation is likely to influence flood response and groundwater
recharge processes. Similarly, precipitation values of more than 10 mm/d are also
important for runoff generation process. The monthly NAO index for the same
period
was
obtained
from
the
CRU
website
(http://www.cru.uea.ac.uk/ftpdata/nao.dat). Since the NAO index is known to impact
the winter climate of the Middle East (e.g., Cullen et al. 2002; Mann 2002), we
examined its relationships with the local climate (precipitation and temperature) for
the four winter months of December to March in the study area. In addition to
monthly correlations, the relationship of the composite NAO index, averaged over
December-March, with the corresponding values of precipitation and temperature
was also investigated.
Although the studied station covered about 50 years, from the 1950s to 2003, it is
worth looking at climatic patterns spanning over the last century. For this purpose,
CRU data on monthly precipitation and temperature were used for the period 1900
to 2002 which is available at a 0.5 degree scale (New et al. 2000; Mitchell et al.
2004) available through http://www.cru.uea.ac.uk/~timm/grid/CRU_TS_2_1.html.
of
35.12
33.29
34.6
31.20
48.43
48.22
49.46
48.40
1,125.0
1,708.0
22.5
1,322.0
1,373.4
1,679.7
Elevation
(masl)
1951-2003
1955-2003
1957-2003
1951-2003
1960-2003
1955-2003
Record
length,
temperature
1951-2003
1951-2003
1955-2003
1957-2003
1951-2003
1960-2003
Record
length,
precipitation
17.2
13.8
25.3
14.2
13.4
11.0
Mean
1.1
1.1
0.8
Standard
deviation
0.8
0.9
1.0
Annual
temperature (oC)
1,602
1,364
1,930
56
57
57
Annual
evaporation
demand (mm/yr.)
Mean Standard
deviation
1,515
58
1,437
60
1,362
58
Annual
precipitation
(mm/yr.)
Mean Standard
deviation
447
134
464
122
335
87
510
128
342
102
229
87
Notes: Potential evapotranspiration was calculated using the Hargreaves method (Hargreaves et al. 1985). cMean and standard deviations for temperature, evaporation and
precipitation are based on the annual values.
Data source: Islamic Republic Iran, Meteorological organization (http://www.irimo.ir/english/).
34.17
35.20
Latitude
(degrees
North)
47.70
47.00
Longitude
(degrees
East)
Geographic and climatic characteristics of the selected synoptic climatic stations.
Kermanshah
Sanandaj
Hamedan
Nozheh
Khorramabad
Arak
Ahwaz
Name
station
Table 9.
Streamflow Trends and Climate Linkages
57
_____________________________________________________________________________________________________________________
58
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The records of the available stations had very few missing values. The missing
values were replaced by average values in case of climatic data. The missing values
in daily streamflow were filled by taking average of the days before and after the
day having missing record. The missing records for longer periods were filled by
using statistical relationship based on correlation analysis with the neighboring
station(s).
4.2.2. Trend and correlation analysis
Trends were examined using the Spearman’s Rank (SR) test (e.g., McCuen 2003;
Yue et al. 2002). The studied data were examined for the presence of serial
correlation before conducting the trend analysis. The serial correlation, if found
significant, was removed using the pre-whitening method, before application of the
trend test. The relationship between streamflow and climatic variables was studied
by performing a correlation analysis among them. This analysis was mainly focused
on the two catchments, Qarasou and Kashkan, mainly because of the better
representative climatic data sets. The climatic data of Kermanshah and Khorramabad
were used to study the linkages between streamflows at Ghore Baghestan and Pole
Dokhtar, respectively.
4.3.
Results and Discussion
4.3.1. Characterizing the streamflow regime
A brief description of the salient features of the streamflow regime, in terms of study
variables, is presented in this section. Table 10 shows the mean and CV (given in
parenthesis) of the studied streamflow variables. The results substantiate that the
studied streamflow variables generally have high variability. For instance, the
differences between the peak and low flows within a year are quite large. For
example, at Jelogir, the mean monthly streamflow in April (386 m3/s) is nearly ten
times higher than in September (41 m3/s). Although the extreme floods (i.e., 1-day
maximum) are generally observed in spring, particularly in March, these can occur
any time from November to April (Figure 20). On an average, 2 to 5 high pulses are
observed in a year, with the mean duration ranging from 8 to 17 days. The peaks are
generated as a result of the large amount of precipitation as well as from snowmelt
contributions or the combined effect of snowmelt and rainfall. The low flows are
recorded from June to September. During this period the magnitudes are quite low,
though there remains some water in all the main rivers throughout the year. On an
average, 2 low pulses in a hydrological year, with a mean duration ranging from 63
to 79 days is observed across the examined stations.
Streamflow Trends and Climate Linkages
59
____________________________________________________________________
Table 10. Streamflow characteristics indicating mean and CV (given in parenthesis)
values at selected locations in the Karkheh River Basin.
Streamflow
indicators
Annual
October
November
December
January
February
March
April
May
June
July
August
September
1-day minimum
7-day minimum
1-day maximum
7-day maximum
Date of minimum
Date of maximum
Low pulse count
Low pulse duration
High pulse count
High pulse duration
Units
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
m3/s
Julian daya
Julian day
No.
Day
No.
Day
Pole
Chahre
34(0.50)
7(0.76)
25(1.77)
32(0.76)
34(0.54)
46(0.44)
95(0.70)
98(0.63)
52(0.81)
11(0.84)
5(0.85)
3(0.87)
3(0.85)
2(0.71)
2(0.67)
274(0.73)
197(0.69)
251(0.07)
82(0.10)
2(0.83)
63(0.78)
3(0.70)
13(0.94)
Ghore
Baghestan
23(0.54)
5(0.47)
12(1.26)
16(0.68)
20(0.62)
30(0.53)
65(0.86)
64(0.69)
36(0.66)
13(0.66)
6(0.70)
4(0.67)
4(0.58)
2(0.58)
2(0.53)
183(1.06)
135(0.95)
254(0.09)
87(0.08)
2(0.72)
68(0.76)
2(0.68)
15(0.98)
Holilan
77(0.52)
16(0.53)
46(1.42)
62(0.69)
72(0.55)
105(0.46)
222(0.89)
220(0.71)
117(0.74)
32(0.74)
16(0.77)
10(0.66)
10(0.60)
6(0.56)
7(0.48)
613(0.91)
446(0.91)
248(0.09)
103(0.17)
2(0.75)
79(0.67)
3(0.91)
17(1.09)
Pole
Dokhtar
52(0.41)
20(0.45)
35(0.94)
47(0.61)
51(0.54)
74(0.52)
123(0.52)
123(0.59)
74(0.74)
28(0.55)
19(0.46)
16(0.39)
15(0.35)
12(0.34)
12(0.32)
553(0.63)
265(0.52)
250(0.09)
68(0.13)
2(1.14)
71(0.86)
5(0.58)
8(1.05)
Jelogir
158(0.42)
55(0.40)
108(1.18)
143(0.60)
157(0.53)
221(0.43)
379(0.57)
386(0.59)
230(0.65)
85(0.57)
52(0.52)
41(0.45)
39(0.39)
28(0.44)
31(0.39)
1,093(0.64)
751(0.60)
248(0.12)
71(0.13)
2(1.18)
68(0.77)
4(0.61)
11(0.88)
Note: aJulian day is calculated for a calendar year based on the notion of taking 1st January as 1st
Julian day and 31st of December as 365th or 366th Julian day.
Date of 1-day maximum streamflow
(Julian day)
360
270
180
90
0
1961-62
1966-67
1971-72
1976-77
1981-82
1986-87
1991-92
1996-97
Hydrological year
Figure 20. Timing of the 1-day maximum streamflow, illustrated by the records at
Pole Dokhtar.
60
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
4.3.2. Streamflow trends
The results of the trend analysis for the streamflow variables are presented in Table
11. The sign of the t statistics indicates the nature of the trend, with positive and
negative values indicating increase and decrease, respectively, and its magnitude
indicates the level of significance. Trends are not significant for the mean annual
streamflows across any of the examined locations. However, a number of significant
trends (both increasing and decreasing) have been found with respect to other
indicators. This stresses the need for investigating streamflow trends at finer
temporal resolutions, as most of the important trends may not be identified at the
coarser time scales.
Streamflows in May depict a decreasing trend for all examined stations, though
more noticeable in upper parts of the basin. The trend is observed significant at
Ghore Baghestan and Holilan as shown in Figure 21. The slope of the linear trend
indicated a decrease of about 0.61 m3/s/yr. at Ghore Baghestan and 1.94 m3/s/yr. at
Holilan. A strong decline at Holilan is due the compound effect of the upper two
stations, both experiencing a decline. The visual inspection of the inter-annual
patterns from Figure 21 suggests that increases and decreases follow a pattern of 2 to
5 years, with a more significant decline after the late 1970s.
However, the years with high and low flows follow similar patterns for both
locations, e.g., highest and lowest values were observed during 1969 and 2000,
respectively. Further to the decline in May flows at these two locations, all the low
flow periods from June to September indicate a declining pattern with significant
trends in August. It is noteworthy that this pattern is similar, though with varying
significance, for four out of five observed stations, with the exception of Pole
Dokhtar. The impact of these patterns is clearly seen in the three upper stations, for
instance, indicating declining patterns in extreme low flows with significant
decreases in 1 and 7 days minimum at Ghore Baghestan.
On the contrary, increasing trends are observed for December and March flows.
In particular, December flows are significant at 90% confidence level at Holilan,
Pole Dokhtar and Jelogir. March flows indicate noticeable increases at Pole Dokhtar,
significant at the 85% confidence level. These trends are illustrated in Figure 22,
demonstrating temporal distributions for December (slope: 0.65 m3/s/yr.) and March
(slope: 0.89 m3/s/yr.) streamflows at Pole Dokhtar where the highest changes are
observed. Consistent with these observations, the flood regime at Pole Dokhtar
shows discernible intensification, with significantly increasing trends for 1 and 7
days maxima. The observed slopes of the trends for these extremes are 10.11
m3/s/yr. and 3.59 m3/s/yr., respectively. Similar patterns, though less significant, are
observed for all other stations with the exception of Ghore Baghestan.
The low pulse count indicates an increasing trend across all stations, significant
in the case of Ghore Baghestan, Holilan and Jelogir and, consequently the duration
of the low pulses followed decreasing trends. This could be related to the decline in
the low-flow regimes because of their interrelationships. Similarly, increasing
patterns of high pulse duration and high pulse count are generally consistent with the
increasing trends in the flood regime. Therefore, recognition of the interdependence
Streamflow Trends and Climate Linkages
61
____________________________________________________________________
of the studied streamflow variables is helpful while interpreting the consistency of
the trend results. For example, the 7-days maximum flow is strongly correlated with
1-day maximum, and the 7-days minimum flow follows pattern similar to that of 1day minimum. Moreover, these maximum and minimum flows are governed by the
variability in the corresponding streamflows during high- and low-flow periods,
which also have an influence on the dynamics of associated high- and low-flow
pulses.
Table 11. Results of the trend analysis showing calculated t statistics for streamflow
indicators.
Streamflow
variables
Annual
October
November
December
January
February
March
April
May
June
July
August
September
1-day minimum
7-day minimum
1-day maximum
7-day maximum
Date of minimum
Date of maximum
Low pulse count
Low pulse duration
High pulse count
High pulse duration
Pole
Chehre
0.479
-0.594
-0.575
0.460
-0.212
-0.459
0.671
0.345
-1.193
-0.808
-0.987
-0.837
-0.590
-0.578
-0.132
1.269
1.063
0.382
-1.282
1.085
0.227
0.332
1.791*
Ghore
Baghestan
0.0860
0.471
0.153
0.840
0.329
-0.354
0.328
-0.049
-1.535*
-1.119
-1.129
-1.446*
-1.020
-1.763*
-1.560*
-0.234
0.163
-0.566
-0.012
3.307*
-1.446*
-0.428
1.185
Holilan
0.180
-0.075
0.204
1.315*
0.690
0.104
0.796
0.068
-1.543*
-1.251
-1.160
-1.461*
-0.911
-1.01
-0.624
0.081
0.200
-0.534
-0.826
2.991*
-0.906
0.178
0.881
Pole
Dokhtar
1.140
1.467*
0.732
2.070*
0.993
0.463
1.248
0.631
-0.014
0.278
0.428
0.293
0.579
-0.378
-0.069
1.996*
2.298*
0.156
-0.196
0.843
-1.469*
1.670*
0.930
Notes:
a
Numbers in italics refer to the trend results based on the pre-whitened data.
* Indicate a significant trend at 90% confidence level (one tailed).
Jelogir
0.680
0.689
0.271
1.385*
0.325
0.315
1.152
0.522
-0.806
-0.408
-0.314
-0.174
0.045
-0.625
-0.238
1.347*
1.052
-0.548
-0.085
2.368*
-1.722*
1.293
0.636
62
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Holilan
Ghore Baghestan
375
Streamflow ( m3/s)
Streamflow ( m3/s)
100
75
50
25
0
1962
1972
1982
300
225
150
75
0
1962
1992
1972
1982
Year
May
1992
Year
Linear (May)
May
Linear (May)
Figure 21. Declining trend in May streamflow at Ghore Baghestan and Holilan.
Pole Dokhtar
Pole Dokhtar
250
Streamflow ( m3/s)
Streamflow ( m3/s)
300
200
150
100
50
0
1962
1972
1982
150
100
50
0
1961
1992
Year
Linear (March)
March
1971
Decem ber
Pole Dokhtar
800
Streamflow ( m3/s)
Streamflow ( m3/s)
1991
Pole Dokhtar
1,500
1,200
900
600
300
0
1962
1981
Year
Linear (Decem ber)
1972
1-day m ax
1982
1992
Year
Linear (1-day m ax)
600
400
200
0
1962
1972
1982
1992
Year
7-day m ax
Linear (7-day m ax)
Figure 22. Increasing trends observed at Pole Dokhtar, illustrated by December,
March, 1 and 7 days maximum streamflows.
Streamflow Trends and Climate Linkages
63
____________________________________________________________________
It is important to note that the serial correlation was mainly depicted by the
extreme low flow indicators (e.g., 1 and 7 days minimum flows at four out of five
examined stations indicated significant r1). The rest of the studied indicators did not
depict persistence in them, with the exception of January flows at Pole Chehre and
February flows at Pole Dokhtar. The strong persistence in the extreme low flow
indicators could be due to the groundwater flow processes which are slow in nature
and, consequently, manifest a carry over effect over time. This observation of
persistence in the low flow indicators is in agreement with the literature (e.g.,
Douglas et al. 2000). Moreover, the presence of serial correlation did not alter the
trend results as the number of significant trends remains the same with and without
the pre-whitening. Although the magnitudes of the calculated t statistics were
different in the case of the pre-whitened series when compared to their
corresponding values without pre-whitening, these were not distinctive enough to
change the indicated significance level. For example, t values for 7-days minimum
flow at Ghore Baghestan were estimated as 1.560 and 1.670 in the cases of with and
without pre-whitening, respectively. Both of these values constitute a significant
category at 90% confidence level. Moreover, insignificant impact of pre-whitening
could be due to the lower values of the first serial correlation (r1) observed in this
study, which falls around 0.3 in most cases of the pre-whitened streamflow
indicators. This point is substantiated by the study of von Storch and Navarra
(1995). They demonstrated that the false rejection of the null hypothesis also
depends on the magnitude of r1. They showed that while applying the Mann-Kendall
trend test the chances of false rejection of the null hypothesis increase from 15% at
r1 value of about 0.3 to more than 30% when r1 exceeds 0.6.
4.3.3. Trends in the climatic data
The trend investigation results for precipitation and temperature are given in Table
12. As in the streamflow indicators, the number of significant trends remains the
same before and after the pre-whitening of some of the studied climatic data series,
most likely due to the reasons similar to those mentioned before. The results indicate
that the total annual precipitation has not significantly changed at most of the
examined stations, with the exception of Arak where a downward trend is
significant. However, a number of upward and downward trends are observed in
other studied indicators at an annual scale (Table 12). These trends were quite
consistent with the changes observed in the studied indicators at the monthly scale at
the respective stations. On the whole, reasonably uniform trends in precipitation are
observed on various indicators for precipitation in April, May, March, October and
December precipitation. Among them, the decreasing trend in total monthly
precipitation in April is the most striking, which is significant for the four out of six
stations. Similarly, a decreasing pattern is observed for May, though less significant
compared to that for April. The results indicate that the precipitation regime shows a
general decreasing trend during these 2 months for the Karkheh Basin as well as for
the neighboring areas in the Zagros mountains.
64
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Table 12. Trend results for precipitation (P) and temperature (T) data showing
calculated t statistics for the studied climatic indicators.
Variable/
Climatic station
Total P
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
No of P days
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
No of days P ≥10 mm
Kermanshah
Sanandaj
Hamedan
Khormabad
Arak
Ahwaz
No of snow/sleet days
Kermanshah
Sanandaj
Hamedan
Khormabad
Arak
Greatest daily P
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
Mean T
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Annual
1.247
0.302
1.463*
0.799
1.238
0.565
-0.044
0.025
-0.376
-0.933
-0.076
1.010
1.229
0.688
0.685
0.531
0.529
0.896
1.476*
-1.839*
0.292
-0.752
-3.122*
0.744
0.129
-1.144
-0.745
0.341
-1.61*
0.320
0.947
-0.572
0.164
1.279
0.39
2.551*
-2.458*
-0.548
-1.615*
-1.379*
-1.368*
-0.728
-0.618
-1.203
-0.939
-0.637
-0.061
-0.973
0.788
-1.242
-1.239
-0.613
-1.972*
1.221
1.636*
0.79
1.503*
1.495*
1.185
1.769*
0.509
0.816
0.252
-0.491
1.102
1.548*
2.126*
0.951
1.778*
0.7
2.016*
3.589*
2.586*
-0.781
1.595*
1.285
0.745
3.244*
0.652
-0.888
0.728
0.171
1.131
3.555*
0.465
-0.598
-0.063
0.76
1.518*
3.046*
-1.802*
-2.257*
-0.666
-0.738
-0.388
1.673*
0.708
-0.918
0.281
-0.042
0.288
1.657*
1.813*
-1.048
1.506*
0.557
1.934*
4.572*
1.733*
0.81
1.995*
1.371*
0.831
1.352*
-0.46
-0.01
0.122
0.028
-0.079
1.902*
0.611
0.884
1.236
0.278
0.569
0.966
1.356*
-1.431*
0.393
-0.289
-2.015*
0.479
0.359
-0.203
-0.761
1.518*
-1.821*
0.951
0.753
0.207
1.253
1.348*
0.217
2.133*
-1.95
-0.321
-1.258
-0.63
-1.42*
0.518
-0.706
-0.933
-0.728
-0.181
0.354
0.714
0.254
-0.672
-1.09
-0.227
-2.373*
1.543*
-0.336
0.111
0.676
0.793
1.399*
1.483*
-0.594
1.824
0.69
0.484
0.742
0.178
1.019
0.455
1.524*
-0.529
-1.223
-0.025
1.216
1.189
0.319
-0.451
1.109
0.264
1.259
1.158
0.305
1.433*
0.800
1.277
0.757
-0.052
-0.401
-0.452
-0.609
0.192
0.992
1.709*
0.354
0.641
0.370
0.420
0.841
1.062
-1.472*
-0.767
-1.093
-3.055*
1.172
0.281
-2.180*
0.081
0.193
-1.250
0.054
1.576*
-0.608
0.017
2.735*
0.345
2.378*
-1.117
-0.541
-2.845*
0.900
-0.388
-0.950
-0.818
-1.051
-0.689
-0.780
0.619
-1.063
-0.273
-0.936
-1.582*
0.85
-2.042*
0.746
2.586*
0.597
-0.365
-2.52*
-0.688
2.830*
3.531*
0.309
-0.030
-2.62*
-0.899
2.101*
2.266*
1.008
-0.558
-1.726*
-0.247
1.853*
0.750
0.875
-0.897
-2.065*
0.127
0.826
0.935
0.764
-1.149
-2.020*
-0.304
1.197
1.389*
0.249
-1.574*
-2.193*
-1.804*
0.760
3.896*
3.260*
1.744*
-1.752*
0.829
3.070*
4.003*
0.337
-0.403
-2.312*
-0.945
4.055*
3.904*
1.421*
-0.175
-2.279*
-0.427
3.382*
Notes:
a
Numbers in italics refer to the trend results based on the pre-whitened data.
* indicates a significant trend at the 90% confidence level (one tailed).
Streamflow Trends and Climate Linkages
65
____________________________________________________________________
Conversely, precipitation in March shows an increasing trend in most cases,
except for Sanandaj. For the Karkheh Basin, the trend is noteworthy for
Khorramabad where significant trends are observed for total precipitation
(significant at 85% confidence level), number of days with precipitation ≥ 10 mm/d
(significant at 90% confidence level) and the amount of greatest daily precipitation
(significant at 90% confidence level). December precipitation also exhibits an
increasing trend in terms of monthly totals, with the number of days with
precipitation showing an increasing trend at most of the stations. Precipitation totals
in October show an increasing pattern, with significant trends for Hamedan. Mean
monthly temperature showed nonuniform trends across the examined locations, and
most noteworthy for the study region are the increasing trends at Kermanshah and
decreasing trends at Khorramabad.
Some of the abovementioned trends regarding precipitation show similarity with
an earlier study conducted for semi-arid to arid regions of Iran by Modarres and da
Silva (2007). They studied trends in total annual precipitation, total monthly
precipitation and total number of rainy days in a year. They used data of 20 climatic
stations located all across Iran with varying recorded lengths of time of 32 to 50
years. It is important to note that their study period was almost similar to ours but
the subset of climatic stations was different from those used in our study and none of
their stations fall within the Karkheh Basin and its proximity. Nonetheless, their
findings are largely consistent with those of our study. Their results show no
significant trends in total annual precipitation and number of rainy days in a year at
18 out of 20 stations. They reported the presence of increasing and decreasing trends
in spring and winter months across a few locations, most notably a significant
decrease in precipitation during April at four stations and significant increases at
four stations in March.
4.3.4 Streamflow trends and climate linkages
The correlation analysis indicates that the mean temperature showed negative
correlation with streamflow variables, but these correlations were generally weak by
themselves (less than 0.3 in most cases) and also in comparison with those of
precipitation with streamflow. Therefore, further discussion is mainly focused on
streamflow and precipitation. Streamflows are strongly correlated with precipitation
at the annual scale as indicated by the correlation analysis between average annual
streamflow at Ghore Baghestan and total annual precipitation at Kermanshah (r =
0.81). A similar inference was drawn from the correlation analysis between
streamflow at Pole Dokhtar and precipitation at Khorramabad (r = 0.84).
Furthermore, both variables did not exhibit significant trends at annual scale at the
studied locations (Tables 11 and 12). The presence of a strong correlation at annual
scale is in agreement with an earlier study in the Karkheh Basin by Sutcliffe and
Carpenter (1968).
However, one can anticipate that the relationship of other streamflow variables,
e.g., monthly flows and extremes, with precipitation may not be straightforward
66
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
because of the complexities of various hydrological processes. For instance, mean
flow in a month may not be entirely dependent on the precipitation in that month,
but is often the result of a combined effect of precipitation during that and earlier
months. For the two selected subbasins of the study area (e.g., Qarasou subbasin
represented by the streamflow measured at Ghore Baghestan station and Kashkan
subbasin represented by the streamflow measured at Pole Dokhtar station), the
correlation analysis with monthly streamflows and precipitation suggests that the
precipitation in a month can influence monthly streamflows with a lag time varying
from 1 to 7 months. The main causes are the occurrence of snowfall in winter which
mainly melts in spring and the contribution of (delayed) runoff from the subsurface
storage. For instance, at Ghore Baghestan, streamflows in November are influenced
by precipitation in November (r = 0.79) and October (r = 0.38), whereas
streamflows in June are governed by the precipitation in January to May (r ranging
from 0.32 to 0.57). The presence of a good correlation among most of the
streamflow and precipitation variables enabled us to develop the linkages between
them, which are discussed in detail further below. We also conducted a correlation
analysis by using the Spearman and Kendall methods (e.g., McCuen 2003), not
shown here because the results were in good agreement with those of the Pearson
method.
The Case of Ghore Baghestan
The decline in May streamflow could be attributed to the decline in precipitation
during April and May. Both these months strongly influence the streamflow in May,
as indicated by the strong correlation between streamflows and precipitation (r =
0.56 for April and r = 0.49 for May). The analysis of the groundwater contribution
to the total streamflow during May could further help understand the dependency of
May streamflow on the precipitation in the previous months. Masih et al. (2009)
have estimated a base flow index of 0.7 for May at Ghore Baghestan using longterm data of the period 1961-2001. This clearly depicts a very high contribution
from groundwater storage in the low flow regime. Considering no significant change
in precipitation during winter periods (December to March) and a noticeable decline
in May and April precipitation in the area, it can be argued that these changes in the
precipitation are the main trigger of declining streamflows in May. This point is
further substantiated in Figure 23, which clearly shows the streamflow in May
following similar patterns of precipitation in April and May. The CRU data helped
to look at trends over the last century (1901 to 2002). Further examination of CRU
data produced similar evidence that the precipitations during April and May have
significantly decreased over time, notably, after the 1980s.
Since the low flow regime of the Ghore Baghestan is strongly governed by the
precipitation in April (r ranging from 0.54 for June to 0.44 for September) and May
(r ranging from 0.36 for June to 0.23 for September), the decline in monthly flows
(June through September) and extreme low flow conditions (1 and 7-days minima)
could also be attributed to the decline in precipitation during April and May. But,
there might be some other complementary factors as well. For instance, the
Streamflow Trends and Climate Linkages
67
____________________________________________________________________
increasing trend in temperature observed for Kermanshah might be another reason
for the decline in streamflow during low flow periods observed at Ghore Baghestan.
This is likely to impact the snowfall and snowmelt dynamics in the area and could
induce early snowmelt causing an increase in winter flows and a subsequent
decrease in spring and early summer flows (e.g., Arnell 1999; Bouraoui et al. 2004).
Although these observations were not sufficiently supported by the observed trends
in the streamflow at Ghore Baghestan, they do signal towards these changes as
depicted by positive t values for December, though not significant at this station
(Table 11). Nonetheless, the observed upward trend in December flow at Holilan is
significant (t = 1.315) which indicates the combined effect of the abovementioned
changes observed at Ghore Baghestan and Pole Chehre. The rise in temperature is
likely to accelerate ET processes and hence could reduce the streamflow (e.g., Nash
and Gleick 1991; Ficklin et al. 2009). Additionally, the increase in crop water
demand is likely to enhance water consumptions (e.g., more irrigation applications
by the farmers). Consequently, this is likely accelerate irrigation withdrawals from
the streams, which would be significant during low flow periods when streamflows
are already at their lowest level (e.g., about 5 m3/s from July to September at Ghore
Baghestan, Table 10).
Generally, the trends observed at Ghore Baghestan are consistent with those
observed at Pole Chehre, most notably in terms of the low flow indicators (Table
11). The composite impact of these changes observed at Ghore Baghestan and Pole
Chehre is also clearly evident from the observations at Holilan station on the
Saymareh River, which is mainly sourced through these two upper subbasins.
However, the observed changes in these three stations were not uniform across all
other examined stations. For instance, changes in the low flow regime were not
significant at the Pole Dokhtar. On the other hand, the flood regime showed
significant upward trends in the middle parts of the basin, which are further
discussed in the following section.
68
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
P in April_CRU data
P in April_IRIMO data
P in May_CRU data
Streamflow in May
P in May_IRIMO data
180
70
60
140
50
120
100
40
80
30
60
20
40
5-Year mean streamflow
(m 3/s)
5-Year mean precipitation
(mm/month)
160
10
20
2001
1991
1981
1971
1961
1951
1941
1931
1921
1911
0
1901
0
Year
Figure 23. The linkages of trends in streamflow in May and precipitation in April
and May, illustrated by the case of Ghore Baghestan.
The Case of Pole Dokhtar
The increasing trends observed in the streamflows at Pole Dokhtar are consistent
with the climatic alterations observed in the area. This is evident by the increasing
trends in precipitation during winter and decreasing trends in temperature observed
at Khorramabad (Table 12).
The flood regime at this location, i.e., 1 and 7-days maxima mainly depends on
the winter precipitation, with major influence of precipitation in March and
February, as indicated in Figure 24. Therefore, intensification of the precipitation
regime in March will possibly be the main cause of the increasing trends in the flood
events as most of them occur in March, as indicated by Figure 20. This point is
further supported by concurrent trends in the March streamflow (t = 1.248,
significant at 85% confidence level) and total precipitation in March (t = 1.279,
significant at 85% confidence level).
The increasing trends in December streamflows could be mainly linked to the
increasing trends in the precipitation regime in December (r = 0.67). The streamflow
in October is significantly correlated with precipitation in March (r = 0.44),
February (r = 0.41) and October (r = 0.38). This suggests that the increasing trends
in October streamflow are due to the increases in the precipitation regime in these
months, most notably March and October. It can be further explained from the point
of view of a hydrological process, as it is very likely that more frequent precipitation
Streamflow Trends and Climate Linkages
69
____________________________________________________________________
events and of greater magnitude produce more surface runoff as well as more
recharge to the subsurface flows. The increased subsurface recharge contributes to
the streamflows in the latter part of the year via base flow. This means that the
precipitation recharging subsurface flow in March and February has an influence on
October streamflows in this case. Since temperature data showed negative
correlations with the streamflows, the significant decline in temperature observed for
this region are likely to reduce ET and, therefore, might be another contributing
factor towards the general increase in streamflows at Pole Dokhtar. Another factor
contributing to these increasing trends could be the watershed degradation that has
taken place in the study area over the last few decades (Ghafouri et al. 2007). Some
studies suggest that the decrease in forest cover increases the flood potential (e.g.,
Guo et al. 2008). For the Karkheh Basin, this point is further supported by the study
of Mirqasemi and Pauw (2007) that compared the land use maps derived from
Landsat data for the years 1975 and 2002, and found that a decline of about 25% has
occurred in the forest cover during this period in the Karkheh Basin. Therefore, this
change is likely to cause increasing trends in the flood regime, particularly in the
middle parts of the basin where forests are a major land cover, but this warrant
further research.
The changes in the low flow regime (e.g., indicated by flows from May through
September and 1 and 7 days minimum) at Pole Dokhtar were not significant, despite
declining trends in precipitation in April and May for the region indicated by the
climatic station at Khorramabad. This could be due to the counter-effect of the
increase in precipitation during March. Another reason could be the decrease in
evaporation demand, caused by the decreasing trend in temperature observed in the
region, which is likely to contribute positively to the streamflow generation
processes.
As expected, the observed changes in the streamflows at Jelogir station on the
Karkheh River generally concur with those at the upstream stations (e.g., Pole
Dokhtar and Holilan). This is particularly evident by the significant trends in
December flow, 1-day maximum flow, low pulse count, and low pulse duration
observed at Jelogir (Table 11).
70
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
P in February
7-day max
500
125
400
5-year mean precipitation
(mm/month)
3
150
5-year mean streamflow ( m /s)
P in March
100
300
75
200
50
100
25
1997
1992
1987
1982
1977
1972
1967
0
1962
0
Year
Figure 24. Linkages of extreme floods with precipitation, indicated by the 7-days
maximum streamflows at Pole Dokhtar and precipitation in March and February at
Khorramabad.
4.3.5 The impact of NAO index on the local climate
Further examination of the observed trends in relation to the changes in the global
circulation patterns generated useful insights into the study region. The earlier
studies in the Middle East have shown the influence of NAO on controlling the
temperature and precipitation regime during winter and early spring (e.g., Cullen et
al. 2002; Mann 2002; Zangvil et al. 2003; Evans et al. 2004). We also attempted to
study the correlation between monthly NAO index with the monthly precipitation
and temperature during the winter months from December to March. These
relationships were also investigated for the whole winter period by averaging the
data sets from December through March. The results indicated very weak
correlations with precipitation (Table 13). Nonetheless, the NAO index showed
comparatively better correlations with temperature (Table 13), with all the stations
located in the Zagros mountain area depicting significant correlations for the
composite NAO index from December to March as well as for most of the winter
months.
Our findings regarding precipitation are different from those of Cullent et al.
(2002) who found a strong impact of changes in the NAO on the streamflow,
precipitation and temperature during December-March for the neighboring
Euphrates-Tigris River system. However, our findings correspond with those of
Evans et al. (2004) who found that NAO index alone could not be a predictor of the
local climate in the Zagros mountains, Iran. They used climate models to simulate
the climate of the Middle East, including the Zagros mountain ranges, Iran, and
illustrated that local factors related to storm tracks, topography, and atmospheric
Streamflow Trends and Climate Linkages
71
____________________________________________________________________
stability have a strong control over climate of the Zagros mountains as compared to
NAO. A study by Alijani (2002) on the linkages of 500 hpa (hectopascals) flow
patterns and the climate of Iran also indicated the importance of the local climatic
factors. Therefore, more detailed studies are required on linkages between long-term
changes in the local climate (i.e., precipitation and temperature) and global as well
as local circulation patterns.
Table 13. Correlation (r) of NAO index with winter precipitation and temperature
for the study area.
Variable/Climatic
station
Precipitation
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
Temperature
Kermanshah
Sanandaj
Hamedan
Khorramabad
Arak
Ahwaz
December
January
February
March
Composite
(December to
March)
-0.209
-0.213
0.028
-0.151
-0.098
-0.039
-0.062
-0.403*
-0.013
-0.033
0.001
-0.044
0.040
-0.070
0.108
0.031
-0.018
0.053
-0.161
-0.139
-0.049
-0.062
-0.025
0.008
-0.031
-0.247
-0.002
-0.090
-0.277
-0.044
-0.322*
-0.315*
-0.407*
-0.302*
-0.346*
-0.252
-0.301*
-0.341*
-0.294*
-0.376*
-0.337*
-0.385*
-0.350*
-0.238
-0.274
-0.432*
-0.249
-0.223
-0.322*
-0.403*
-0.396*
-0.411*
-0.363*
-0.460*
-0.344*
-0.291*
-0.312*
-0.484*
-0.346*
-0.297
Note: * indicates significant correlation at 95% confidence level
4.4.
Concluding Remarks
The study provided an overview of the changes in the streamflows in the Karkheh
Basin and identified a number of trends, both increasing and decreasing. Most of
these trends were found triggered by climatic factors - mainly by changes in
precipitation. The most notable trends were declines in May streamflows, which can
be attributed to the decline in precipitation in April and May. The two upstream
catchments displayed declining trends in low flow regimes, demonstrated by
monthly streamflows, 1 and 7 days minima and the number and duration of low flow
pulses. In the middle part of the basin (at Pole Dokhtar) the increasing trends were
reflected by 1 and 7 days maxima, and March, December and October flows. These
trends can be attributed to the intensification of the precipitation regime in these
72
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
months, with the March precipitation having the highest influence on the flood
regime.
The observed trends for Holilan at the Saymareh River and Jelogir at the
Karkheh River were a reflection of the combined effects of their upstream
catchments. Similar to the observed patterns at Ghore Baghestan and Pole Chehre,
the Holilan indicated declining patterns in monthly streamflows from May through
September, as well as a decline in the low flow regime.
All trends were not reflected in the flow regime of the Karkheh River because of
the varying changes in the upper and middle parts of the basin. The changes at
Jelogir were significant when the patterns were similar for most of the upstream
tributaries. For instance, consistent with the upper parts of the basin, the declining
patterns from May through August were observed at Jelogir, but were not as
significant as in the case of the upper stations. This is because the more stable
response of Pole Dokhtar during these months counterweighted these declining
patterns were observed in the upper parts of the basin. Nevertheless, the significant
trends in streamflows at Jelogir, i.e., an increase in the 1-day maximum, December
flows, low pulse count and a decrease in low pulse duration, indicated alterations of
the hydrological regime of the Karkheh River due to the changes in climate during
the study period.
Contrary to expectations, North Atlantic Oscillation Index did not show a good
correlation with the precipitation in the Zagros mountains because its impact might
be masked by the strong topographic controls and other local climatic factors, which
deserve further research.
Since most of the observed changes in streamflow, precipitation and temperature
were not uniformly distributed across the Karkheh Basin, the adaptation response
should be different for different parts of the basin. If the observed trends will persist,
the major policy concerns about water management would be how to a) stabilize
declining streamflows during low flow periods in the upper parts of the basin, and b)
control intensification of the flood regime in the middle parts of the basin.
5.
REGIONALIZATION OF A CONCEPTUAL RAINFALLRUNOFF MODEL BASED ON SIMILARITY OF THE FLOW
DURATION CURVE7
5.1.
Introduction
5.1.1. Problem statement
Streamflow data are a prerequisite for planning and management of water resources
such as the design of dams and hydropower plants, assessment of water availability
for irrigation and other water uses, assessment of flood and drought risks and
ecological health of a river system. However, in many cases, observed streamflow
data are not available or are insufficient in terms of quality and quantity. This
undermines the informed planning and management of water resources at a specific
site as well as at the river-basin scale.
Hydrologists have responded to this challenge by developing various predictive
tools, which are commonly referred to as regionalization methods (e.g., Blöschl and
Sivapalan 1995; Sivapalan et al. 2003; Yadav et al. 2007). These methods can be
broadly classified into two groups based on their temporal dimension. The first
group deals with the estimation of continuous time series of streamflows (e.g.,
Magette et al. 1976; Merz and Blöschl 2004). The second group estimates selected
hydrological indices, such as the mean annual flow and base flow index (e.g.,
Nathan and McMahon 1990b), or various percentiles of the flow instead of
continuous time series (e.g., regionalization of the flow duration curve – FDC)
(Castellarin et al. 2004). Further classification can be done within each group. For
example, Castellarin et al. (2004) classified regionalization methods for FDC into
statistical, parametric and graphical approaches. The methods used for estimating the
time series of streamflows can be further categorized into three subgroups: a) model
parameter estimation by developing regression relationships between model
parameters and catchment characteristics (e.g., Magette et al. 1976); b) transfer of
model parameters, whereby a catchment similarity analysis is conducted and
7
This chapter is based on the paper Regionalization of a conceptual rainfall-runoff model
based on similarity of the flow duration curve: a case study from the semi-arid Karkheh Basin,
Iran” by Masih, I.; Uhlenbrook, S.; Maskey, S.; Ahmad, M. D. 2010. Journal of Hydrology
391: 188-201. DOI:10.1016/j.jhydrol.2010.07.018.
74
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
parameters of gauged catchments are used in simulations for similar ungauged or
poorly gauged catchment (e.g., Kokkonen et al. 2003; Wagener et al. 2007); and c)
other regionalization techniques such as spatial interpolation of parameters (e.g.,
Merz and Blöschl 2004) or regional pooling of data for parameter estimation for
ungauged catchments (e.g., Goswami et al. 2007).
Despite considerable progress in hydrological research, the prediction of
streamflow for ungauged or poorly gauged catchments still remains a major
challenge (Sivapalan et al. 2003; Wagener and Wheater 2006). A brief review of
some key studies involving commonly used regionalization methods applying
conceptual rainfall-runoff models for streamflow estimations in ungauged or poorly
gauged catchments is presented in the following section. We defined a catchment as
ungauged when no streamflow records exist, whereas a data limited or poorly
gauged catchment is defined as a catchment where some measured streamflow
records are available that are usually short, have many gaps and are of poor quality.
These records are not enough to achieve a satisfactory level of model calibration for
streamflow simulation.
5.1.2. Review of regionalization methods using conceptual rainfall-runoff models
An overview of some applications of the rainfall-runoff models for regionalization
in different parts of the world is given in Table 14 and briefly discussed below. The
selected studies estimated continuous time series of streamflows using a rainfall
runoff model and reported the performance measures in terms of at least one of the
three evaluation criteria, namely, Nash–Sutcliffe efficiency (NSE), coefficient of
determination (R2) and the mean annual volume balance (VB). These points were
considered in the selection for consistency in comparison of this study and the
presented literature in Table 14, Moreover, in selecting the literature for discussion
we attempted to represent a wide range of hydro-climatic environments and provide
reasonably good coverage of most of the regionalization methods.
Magette et al. (1976) used 21 catchments (0.02–12 km2) in USA for
regionalization of six selected parameters of the Kentucky Watershed Model
(KWB). They used 15 catchment characteristics in developing regression equations
and found that a multiple regression technique used in stepwise manner was
successful in developing equations to estimate model parameters from catchment
characteristics, but that simple linear regression models were totally unsuccessful.
They randomly selected five out of 21 catchments for validation. Although the
validation results showed significant variations, they concluded that the approach
was useful and should be further developed. Vandewiele et al. (1991) used 24
catchments (16-2160 km2) in Belgium for developing regression equations to
estimate three parameters of a monthly conceptual rainfall-runoff model using the
basin lithological characteristics. They concluded that their regionalization approach
was capable of generating reliable monthly time series for ungauged sites within the
region.
Servat and Dezetter (1993) evaluated the performance of two conceptual rainfallrunoff models (GR3 and CREC models) for possible applications to ungauged
Model Regionalization Based on the Flow Duration Curve
75
____________________________________________________________________
catchments in the north-western part of the Ivory Coast. They were able to relate all
model parameters to catchment characteristics (rainfall and land cover) with varying
degrees of success. The regionalization results in terms of R2 and NSE were variable,
particularly for the NSE which was quite low (i.e., close to zero) in some cases.
Ibrahim and Cordery (1995) applied a monthly water balance model for
predicting streamflows in New South Wales, Australia. The used model had four
parameters, of which three were estimated from rainfall data. Abdulla and
Lettenmaier (1997) regionalized seven of the nine parameters of a large-scale model
(VIC-2L) for Red and White river basins in USA. They estimated two of the model
parameters from STATSGO soil data. For other parameters, they used 28 catchment
variables, related to soil and climate, for developing multiple regression equations
between model parameters and catchment variables. Their regionalization results
were generally good in most cases, although they noticed better performance in
humid and subhumid catchments than in semi-arid to arid catchments.
Seibert (1999) used the HBV model for a regionalization study using 11
catchments in Sweden and found that six of the 13 model parameters could be
estimated from the land cover features (i.e., forest and lake areas). However, the
application to ungauged catchments was achieved with varying degrees of success,
with daily NSE ranging from 0.23 to 0.72. Merz and Blöschl (2004) compared eight
regionalization methods using the HBV model with data sets from 308 catchments in
Austria. Parajka et al. (2005) conducted a follow-up of the Merz and Blöschl 2004
study by improving the model structure (i.e., by dividing catchments into elevation
bands of 200 m interval), adding snow cover data and conducting similarity analysis
on the basis of catchment attributes. They concluded that the methods based on
similarity approaches produce reasonably good regionalization results. This finding
is also consistent with that of Kokkonen et al. (2003) who concluded that “When
there is reason to believe that, in the sense of hydrological behaviour, a gauged
catchment resembles the ungauged catchment, then it may be worthwhile to adopt
the entire set of calibrated parameters from the gauged catchment instead of
deriving quantitative relationships between catchment descriptors and model
parameters.”
McIntyre et al. (2005) proposed a regionalization method of ensemble modeling
and model averaging and tested it using a five parameter version of the probability
distributed model (PDM) on 127 catchments (1-1,700 km2) in the United Kingdom.
They selected donor catchments based on catchment similarity analysis for which
three catchment characteristics, i.e., catchment area, permeability and rainfall were
used. In this approach more than one donor catchment is selected, which is different
from the usual approaches of using a single donor catchment for streamflow
simulations at an ungauged site. Then the full parameter set of each of the donor
catchments is used to predict streamflows at the ungauged catchment, thereby,
generating an ensemble of flow values. Then the average streamflow could be taken
from the weighted average with weights defined based upon the relative similarity.
They found that the proposed method performs reasonably well as compared to the
established procedure of regressing parameter values from the catchment
descriptors. However, they also noted that the new method estimated the low flows
76
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
better than high flows. They recommended further testing of the model, especially to
test different model types and improved definition of similarity.
Goswami et al. (2007) developed a methodology that uses a regionalization and
multi-model approach for simulating streamflows in ungauged catchments. Like
other methods, their methodology did not involve transfer of model parameters from
gauged catchment to ungauged catchment, and model parameters need not be related
to physical catchment descriptors. They used seven different models for
regionalization and for each model three methods were tested that involved the use
of the discharge series by taking regional averages, regional pooling of data and
transposition of discharge data of the nearest neighbor. They used 12 gauged
catchments in France to illustrate their methodology and each time considered one of
them as ungauged for the application of the method and then compared the results
with observed time series of daily discharge using the NSE criterion. The results
indicated a mix of success and failure for the individual catchments and tested
methods. However, they concluded that the pooling method of regionalization
coupled with the conceptual soil moisture accounting and routing model (SMAR)
was the best approach for simulating flows in ungauged catchments in that region.
The second best method was the transposition of data from the nearest neighbor
provided the catchments are similar in the hydro-meteorological, physiographic
characteristics and drainage area.
Oudin et al. (2008) compared three widely used regionalization approaches for (a
large number of) 913 French catchments (10-9,390 km2) by using two conceptual
rainfall-runoff models (GR4J and TOPMO models). They showed that
regionalization based on the spatial proximity performed the best for their sample of
catchments. They also noted that the dense network of tested catchments used in
their study might have resulted in favor of spatial proximity approach and
recommended that this approach should also be tested in other regions, particularly
where less number of gauged catchments are available.
The presented studies reveal that considerable progress has been made to
estimate streamflows at ungauged catchments and quite a number of promising
methods have been developed over the past few decades. However, the studies also
depict a mix of success and failure of the available methods within a study region or
while comparing outcomes from the different regions. Moreover, the tested
regionalization approaches indicate large variability in the achieved performance
statistics, which shows considerable scope for further improvement. Therefore, there
is every motivation to make further progress on this important subject of
regionalization in hydrology.
16 (5)
20(4)
11 (5)
18 (8)
34 (40)
11 (7)
308 (308)
12 (11)
USA
Belgium
Ivory Cost
Australia
USA
Sweden
Austria
France
10 to 1870 (156 to 1792)
168 to 5226 (442 to 6894)
7 to 950 (7 to 1284)
3 to 5000 (3 to 5000)
32 to 371 (32 to 371)
0.04 to 12 (0.02 to 10)
16 to 2163 (73 to 148)
100 to 4500
Drainage area, km2
Monthly
Daily
Daily
Daily
Daily
Hourly
Monthly
Daily
Simulation
time step
NA (-1 to 4)
NA (-11 to 134)
NA
NA
NA
0.73 to 0.94 (0.67 to 0.76)
0.41 to 0.97 (0.05 to 0.81)
NA
NA
NA
Evaluation measures for gauged and test catchments
Volume balance,
Coefficient of
VB, mm/yr.
Determination, R2 (-)
NA (-372 to 155)
NA
-8 to 12 (-29 to 54)
NA
NA
0.23 to 0.99 (0.62 to 0.99)
0.69 to 0.94 (0.62 to 0.89)
NA
0.70 to0. 88 (0.23 to 0.72)
0.67 (0.32 to 0.56)*
NA (-27.66 to 0.94)
Nash-Sutcliffe Efficiency,
NSE, (-)
NA
NA
0.02 to 1 (0.02 to 0.45)
Ibrahim and Cordery 1995
Abdulla and Lettenmaier 1997
Seibert 1999
Merz and Blöschl 2004
Goswami et al. 2007
Magette et al. 1976
Vandewiele et al. 1991
Servat and Dezetter 1993
Reference
Notes:
Figures in parentheses correspond to the test catchments.
NA refers to information not available.
* Efficiency values refer to median of all 308 catchments during calibration phase and (in parenthesis) minimum and maximum median values of tested regionalization methods.
Catchments
Country
Table 14. An overview of some studies related to regionalization of conceptual rainfall-runoff models.
Model Regionalization Based on the Flow Duration Curve
77
_____________________________________________________________________________________________________________________
78
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
5.1.3. Scope and objective
The main research question examined in this paper is whether or not the parameters
of a conceptual hydrological model applied to a gauged catchment can be
successfully transferred for simulating streamflows in a hydrologically similar but
data-limited or poorly gauged catchment. In this study, the HBV model (Bergström
1992) is used for streamflow simulations in the Karkheh River Basin, Iran. The
hydrologic similarity is defined based on four measures, i.e., drainage area, spatial
proximity, catchment characteristics and flow duration curve (FDC). FDCs are
frequently used for comparing the response of gauged catchments, but their potential
use for the regionalization of conceptual rainfall-runoff models for flow estimation
for the poorly gauged catchments needs to be explored and is a main objective of
this study. Streamflow data are required for the construction of an FDC. However,
an FDC could be established from the catchment characteristics for ungauged
catchments using available FDC regionalization methods (e.g., Castellarin et al.
2004). For poorly gauged catchments, the available records, though short, could be
used for the FDC construction. These insufficient records may not be used directly
for rainfall-runoff modeling as indicated in the previous section. Another limitation
in their direct use for modeling purpose is the unavailability of other corresponding
data sets required for modeling, e.g., climatic data for the same period as runoff data
may not be available. These typical limitations were faced for the poorly gauged
catchments in the Karkheh Basin providing the main motivation for this
regionalization study.
The abovementioned methods evaluated in this study require very limited data
resources and were most suitable in the context of the data-limited region under
study. The other commonly used methods, such as regionalization of the model
parameters, generally require data sets from a large number of gauged catchments
for developing statistically sound relationships between model parameters and
catchment characteristics. Due to limited availability of gauged catchments and
necessary data sets, these data-intensive methods were not tested for the study area.
Nevertheless, the results of this study were compared with those published in the
literature from some widely recommended methods tested in other regions of the
world.
5.2.
Materials and Methods
5.2.1 Study catchments and available data
In the Karkheh Basin streamflow data are not available for many catchments and the
existing records have gaps. There were about 50 streamflow gauging stations
installed after 1950 out of which only 24 have been measured continuously. Filling
these data gaps by estimating missing streamflow time series for the poorly gauged
Model Regionalization Based on the Flow Duration Curve
79
____________________________________________________________________
catchments was required for a good understanding of the hydrology and its spatiotemporal variability, which in turn should guide informed water management
decisions.
Eleven gauged catchments, draining tertiary-level streams (475-2,522 km2),
located in the upper mountainous parts of the Karkheh Basin were selected for this
study (Figure 25 and Table 15).
Figure 25. Salient features of the study area and location of the study catchments
and used climatic stations.
80
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The study period from January 1987 to September 2001 was selected considering
the data availability/quality and representation of dry, wet and average climatic
conditions. Time series of daily precipitation data for the study period were available
for 41 climatic stations, well scattered across the study domain (Figure 25). The
areal precipitation estimates were used in the model simulations, which were
obtained by interpolation of the available station data by using inverse distance and
elevation weighting (IDEW) technique (see chapter 6). Temperature data from eight
climatic stations (Figure 25) were available and the station nearest to the catchment
was used in the simulations for that respective catchment. The missing values in the
data sets were estimated based on the values from neighboring stations. The missing
values in the temperature data sets were few (less than 1% in most cases), with the
exception of one station where records were available only for 1996-2001.
Generally, temperature data of a station showed very good correlation with
corresponding data from the neighboring stations (R2 > 0.90) used for infilling of the
missing records. In case of precipitation data, seven out of 41 stations had no
missing records. On average, there were 5% in-filled precipitation events, ranging
from 0 to 16%. Hargreaves equation (Hargreaves et al. 1985) was used to estimate
the reference ET using daily data of maximum, minimum and mean temperatures.
Further details on Hargreaves method and its application in the study basin are given
in Appendix A.
Table 15. Salient features of the selected streamflow gauges.
Name of river
Name
station
of
Khorram Rod
Toyserkan
Gamasiab
Qarasou
Abe Marg
Bad Avar
Abe Chinare
Chalhool
Khorramabad
Doab Aleshtar
Har Rod
Aran
Firoz Abad
Sange Sorakh
Doabe Merek
Khers Abad
Noor Abad
Dartoot
Afarineh
Cham Injeer
Sarab Seidali
Kaka Raza
ID
Long
Lat
1
2
3
4
5
6
7
8
9
10
11
47.92
48.12
48.23
46.78
46.73
47.97
46.40
47.88
48.23
48.22
48.27
34.42
34.35
34.03
34.55
34.52
34.08
35.45
33.30
33.45
33.80
33.72
Elevation,
masl
Drainage
area (km2 )
1,440
1,450
1,800
1,310
1,320
1,780
1,110
800
1,140
1,520
1,530
2,320
844
475
1,260
1,460
590
2,522
800
1,590
776
1,130
Observed
flow
(mm/yr.)
59
55
254
148
34
202
71
160
223
345
355
Naturalized
flow
(mm/yr.)
87
102
294
148
34
315
95
170
341
516
428
Notes: Long = Longitude (degrees East); Lat = Latitude (degrees North).
Data source: Ministry of Energy, Iran, with the exception of station ID and naturalized flow.
5.2.2 Naturalization of the streamflows
The abstraction of river water for irrigation purposes influenced the river flows in
some of the study catchments. Therefore, naturalization of streamflows was carried
out by adding abstraction rates, if any, to the observed streamflows. The main aim of
doing naturalization of the streamflow was to improve the consistency of the
Model Regionalization Based on the Flow Duration Curve
81
____________________________________________________________________
regionalization procedures used in this study. The naturalization of the streamflows
was considered helpful in reducing uncertainties arising due to the abstractions in the
parameter estimation and consequent transfer from one catchment to the other. The
direct pumping from the streams is the main mode of irrigation diversions by the
farmers. However, no pumping records or data for other means of surface water
diversions were available. Therefore, abstractions were estimated using the available
information on crop ET, cropping patterns and cropped area, estimates of irrigation
efficiencies and total annual abstractions. The procedure used is summarized below.
Calculation of crop water demand. The daily potential crop evapotranspiration
(ETc) was calculated using the following equation:
n
ETc =  A j Kc j ET0
(10)
j =1
Where, ETc is the total potential crop ET in m3/d, Aj is the area under the jth crop
in m2, ETo is the reference ET expressed in m/d estimated using Hargreaves method
(Hargreaves et al. 1985), Kcj is the crop coefficient for the jth crop (according to
Allen et al. 1998), and n is the number of crop types, which are mainly wheat,
barley, alfa alfa, sugar beat, maize and orchards. The data on cropping patterns and
cropped area were obtained from JAMAB 1999 whereas sowing and harvesting
dates were based on field surveys. The total ETc was obtained by the summation of
the values for the individual crops.
Calculation of irrigation demand and streamflow abstractions. The irrigation
demand was estimated using the following equation:
 epP 

I d = ETc 1 −

 ET0 
(11)
where, Id is irrigation demand in m3/d, P is the precipitation in mm/month and ep
(-) is the fraction of the precipitation effectively used as ET. The ratio of effective
precipitation and reference ET was computed using monthly data of precipitation
and ETo. For the whole Karkheh Basin, JAMAB (1999) estimated that 66% of the
annual precipitation is consumed as ET and 34% forms the renewable water
resources. For this study conducted in the upper catchments of the Karkheh Basin,
the value of ep was assumed as 0.5, since the evaporation rates are lower in upper
mountainous part of the basin compared to the lower arid plains.
The abstractions from the streams were estimated using the following equation:
82
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
I sw = f sw
Id
η
(12)
where, Isw is the surface water withdrawals, m3/d, fsw is the fraction of surface
supplies in the total irrigation withdrawals and η (-) is the irrigation efficiency. The
used values of η were in the range of 0.3 to 0.7 (JAMAB 1999). The lower values of
η correspond to catchments with higher surface water withdrawals and vice versa.
The annual values of fsw were also available from the study of JAMAB (1999) who
estimated total irrigation withdrawals from surface water and groundwater sources in
the study catchments for the period 1993-94. The catchments where surface water
was the main source of irrigation (i.e., fsw > 0.9), the same value of fsw was used for
each day of the year. For catchments where conjunctive use of surface water and
groundwater was present, the annual value of fsw was distributed into monthly values
following the supply-demand principle whereby higher values were assigned to the
months having higher streamflows (i.e., March to June) and lower values to the
months having lower streamflows (i.e., August to October). This way, fsw was varied
for each month but was kept constant for each day of a month. The estimated values
of Isw were compared with the available estimates at annual scale for the year 199394. If the difference was more than 15%, the procedure was repeated by modifying
the values of η and monthly distribution of fsw. The threshold of 15% was considered
appropriate given the limitations related to the used data as well as full
representations of the involved processes by (simplified) equations used in this
method. Finally, Isw values were added to the observed streamflow to get the
naturalized streamflows. The observed and naturalized streamflows are given in
Table 15, which indicates the extent of the influence of naturalization for each study
catchment. As an example, Figure 26 shows the observed and naturalized
streamflows of one catchment (Aran). This illustrates the streamflow differences in
particular during the late spring and summer, when the crop water requirements are
the largest. After discussions with local experts it was concluded that these
corrections are reasonable and reflect the impact of local practices.
Model Regionalization Based on the Flow Duration Curve
83
____________________________________________________________________
Precipitation
Observed streamflow
Naturalized streamflow
250
0
30
3
Streamflow ( m /s)
20
150
40
50
100
60
70
50
Precipitation (mm/d)
10
200
80
0
1/1/2001
1/1/2000
1/1/1999
1/1/1998
1/1/1997
1/1/1996
1/1/1995
1/1/1994
1/1/1993
1/1/1992
1/1/1991
1/1/1990
1/1/1989
1/1/1988
1/1/1987
90
Date
Figure 26. Naturalized and observed daily time series of streamflows of Aran
catchment.
5.2.3. Model calibration and validation at the gauged catchments
The HBV model was applied to each of the 11 gauged catchments and was
calibrated using daily climatic and streamflow data from January 1987 to September
2001. The data were split into calibration (October 1987 to September 1994) and
validation (October 1994 to September 2001) periods. Before calibration, a
warming-up period of 273 days was used for initialization so that model parameters
attained appropriate initial values. Each catchment was divided into a number of
elevation zones at an interval of 200 m. This interval was selected in order to
balance the total number of elevation bands that could be accommodated in the HBV
and SWAT (see next chapter) modeling set up. This threshold was also appropriate
to avoid having too many or too less divisions of the study catchments. Each
elevation zone was divided into three vegetation zones, namely forest (zone 1),
cropland (zone 2) and range/bare lands (zone 3). Since the elevation is known to
have major impacts on the distribution of rainfall and temperature, which have
already been studied in the region, the values of the two parameters for lapse rates of
precipitation and temperature were based on the earlier studies of Sutcliffe and
Carpenter (1968), JAMAB (1999) and Muthuwatta et al. (2010). The values of lapse
rates were kept constant for all catchments and set to an increase of 5.5% per 100 m
84
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
increase in elevation for precipitation and to a decrease of 0.4 0C per 100 m increase
in elevation in case of temperature.
A Genetic Algorithm (GA) based automatic calibration method, which is in-built
in the present version of the model by Seibert (2002), was applied during model
calibration. Similar calibration methods have been widely used as a global
optimization tools (e.g., Wang 1991; Seibert 2000; Maskey et al. 2004). The ranges
of parameter values (Table 16) were selected based on our understanding of the
study region, experiences of other studies (Seibert 1999; Uhlenbrook et al. 1999;
Uhlenbrook and Leibundgut 2002) and initial model runs for the study catchments.
Table 16. Model parameters and their ranges used during the GA-based automatic
calibration procedure.
Parameter
Snow routine
TT
CFMAX
SFCF
CFR
CWH
Soil routine
FC
LP
BETA
Response routine
PERC
UZL
K0
K1
K2
Routing routine
MAXBAS
Unit
Explanation
Range
o
C
mm oC-1d-1
-
Threshold temperature
Degree-day factor
Snowfall correction factor
Refreezing coefficient
Water-holding capacity
-2.5 to 2.5
1 to 6
0.8 to 1.25
0.05 to 0.05
0.1 to 0.1
mm
-
Maximum of SM (storage in soil box)
Evaporation reduction threshold (SM/FC)
Shape coefficient for soil storage/percolation
50 to 500
0.5 to 0.7
1 to 6
mm d-1
mm
d-1
d-1
d-1
Maximal flow from upper to lower box
Threshold for Q0 outflow in upper box
Recession coefficient (upper in upper box)
Recession coefficient (lower in upper box)
Recession coefficient (lower box)
0.1 to 6
10 to 100
0.05 to 0.5
0.01 to 0.15
0.001 to 0.05
d
Routing, length of weighting function
1 to 5
For instance, the threshold temperature (TT) for snow was set to fall in the range
of −2.5 to 2.5 0C. The optimized threshold value of this parameter defines whether
the precipitation falls in the form of rain or snow. During winter months, the
temperature may fall below the optimized snow temperature threshold causing
precipitation to occur in the form of snowfall apart from the rain events during this
period. The parameters of the snow and soil routines were estimated, using the
abovementioned GA-based optimization procedure, in a distributed manner, thus
having different values for each of the three vegetation zones. The parameters of the
response and routing routines could only be estimated uniformly in the current
version of the HBV model and were, therefore, representative of the whole
catchment. The Nash-Sutcliffe Efficiency (NSE) estimated at the daily time step
(equation 8) was used as an objective function to estimate the model performance
(Nash and Sutcliffe 1970). The NSE is considered as a robust approach to assess the
model goodness of fit in hydrological modeling and is widely used (e.g., ASCE
1993). However, it is also worth noting that the results based on NSE optimization
Model Regionalization Based on the Flow Duration Curve
85
____________________________________________________________________
could be biased towards high flows, which fact warrants caution in interpretations
(e.g., Wagener et al. 2004). Other commonly used measures also have their own
merits and constraints. For instance, the widely used performance measure,
Coefficient of Determination (R2), may reflect higher values (good performance) if
the variability of two data sets is well synchronized despite their volumetric
difference. Therefore, for having a better picture of the results, in addition to NSE,
we examined R2. The difference in the mean annual runoff, termed as volume
balance (VB) was also examined.
5.2.4. Regionalization of model parameters based on catchment similarity
analysis
In this study, the hydrological similarity was defined based on four similarity
measures: drainage area, spatial proximity, catchment characteristics and flow
duration curve (FDC). Once the similarity was established among 11 gauged
catchments, the best parameter set of one catchment was transferred to another
catchment (temporarily considered as ungauged, termed as pseudo-ungauged) for
streamflow simulations. The whole parameter set was adopted from a donor
catchment. The main advantage of adopting a complete parameter set is that the
parameter interdependencies are not neglected. The results were then compared, in
terms of NSE, R2 and VB, by using the observed streamflow time series of the
pseudo-ungauged catchment.
In terms of similarity in area, each of the 11 catchments was compared with other
catchments and was rendered similar to the one which had the closest drainage area.
Similarly for spatial proximity, the two catchments located nearest to each other
were defined as similar. In cases where more than one catchment were available in
the neighborhood, the catchment with the least distance from the centroid and/or
having the longest common boundary was considered the most similar one. The
similarity based on catchment characteristics was defined comparing the climate
(ratio of mean annual precipitation and reference ET), topography (average
catchment slope, elevation and stream density), land use (area under forest and crop
land), soil (area under rock outcrop type soils) and geology (area under limestonedominated geology). These characteristics are generally considered as the major
drivers of the hydrological processes and catchment runoff response (Nathan and
McMahon 1990b; Wagener et al. 2007). The similarity index (S) was calculated by
using Equation (13) and the variables given in Table 17.

M
S = 1−
i =1
αi
ΔVi
Max(ΔVi ,Vi )
(13)
86
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
where, S is the similarity index (-) which takes a value between 0 and 1 and
defines the degree to which catchment 1 is similar to catchment 2, M is the number
of catchment characteristics (variables) used for computing the similarity index. The
αi are the weights (-) between 0 and 1 for the given characteristics such that sum of
the weights is equal to 1. In this study, equal weights are used for all the
characteristics. The variables V, ΔV , and V refer to the value of the respective
catchment characteristics, the absolute difference between catchment 1 and 2, and
the average value of catchment 1 and 2, respectively.
Table 17. Catchment characteristics used in calculating the similarity index.
Catchment
ID
Name
1
2
3
4
5
6
7
8
9
10
11
Aran
Firoz Abad
Sange Sorakh
Doabe Merek
Khers Abad
Noor Abad
Dartoot
Afarineh
Cham Injeer
Sarab Seidali
Kaka Raza
Catchment characteristics
P/ETo
Slope
Elevation
(-)
(%)
(masl)
0.292
0.292
0.379
0.383
0.312
0.319
0.342
0.391
0.370
0.353
0.357
15
17
15
13
10
16
15
23
20
27
23
1,768
1,949
2,081
1,522
1,529
2,037
1,533
1,643
1,652
2,100
2,024
Stream
density
(km/km2 )
0.061
0.063
0.032
0.060
0.076
0.056
0.084
0.094
0.078
0.061
0.084
Rock
outcrop
soils (%)
54
56
55
48
49
44
63
100
55
71
63
Fore
st
(%)
10
10
15
8
10
8
33
50
29
8
13
Cropland
(%)
48
30
17
87
73
59
54
5
38
45
34
Limestone
dominated
geology
(%)
52
27
59
47
20
62
22
48
39
61
60
In the fourth approach, similarity in the FDCs was compared both by means of
visual inspection and by using a statistical criterion, Relative Root Mean Square
Error (RRMSE). FDCs are very useful for comparing the hydrological response of
catchments (e.g., Linsley et al. 1949; Hughes and Smakhtin, 1996; Yilmaz et al.
2008). Their shape is an indicator of catchment response to rainfall and also depicts
the storage characteristics of the catchments and influence of topography, geology,
vegetative cover and land use. In this study, the FDCs were plotted using daily
discharge data which were normalized by the drainage area to facilitate comparison.
The shape of the FDC for each catchment was visually compared with the FDCs of
the other catchments; the catchments showing best match for both high and low flow
percentiles were considered hydrologically similar. A commonly used objective
criterion based on RRMSE, termed here as ε (–), Equation 14 was used to define the
similarity between the FDCs.
Model Regionalization Based on the Flow Duration Curve
87
____________________________________________________________________
ε=
1
N
N
 (Q
i
i =1
Q
− Qˆ ) 2
(14)
where, Qi is the ith flow percentile (mm/d) of one FDC and i ranges from 1 to N;
Q̂i is the corresponding ith flow percentile (mm/d) of another FDC; and Q is the
mean discharge of the first (base line) FDC. The ε values were calculated for the
whole FDC corresponding to the flow percentiles Q0 to Q100 using daily discharge
data.
5.2.5 Assessment of the impact of parameter uncertainty on the regionalization
results
The issue of parameter uncertainty is well recognized in hydrological modeling
(Uhlenbrook et al. 1999; Beven 2001; Wagener et al. 2004; McIntyre et al. 2005).
Generally, parameter values are not unique, and results in large uncertainty bands in
the discharge predictions. Furthermore, similar model simulations can be achieved
by using different combinations of parameter values, which is generally termed in
hydrology as equifinality or nonuniqueness of the model parameters (Beven 2001).
In this study, the impact of parameter uncertainty on the regionalization results was
also investigated. First, the best parameter set of a study catchment in the
regionalization procedure was used, as indicated in the previous section. Then to
check the consistency of the results, we selected 50 different parameter sets of a
catchment that yielded the highest NSE values during the automatic calibration
process, and used them for the regionalization in a way similar to that of using the
single best parameter set. As mentioned in section 5.2.3, the automatic calibration
was based on the GA-based optimization procedure. Therefore, the 50 best
parameter sets are the ones resulting in the highest NSE out of the many good
parameter sets that the GA-based optimization method generates. More parameter
sets may be used for the purpose of this investigation, but we consider this number is
reasonably good to test our hypothesis on the effect of parameter uncertainty on
regionalization outcome. The regionalization results were considered reliable given
the results remain consistent in terms of studied performance indicators (NSE, R2
and VB) while using different parameter sets (e.g., both in case of the best parameter
set and the 50 other good parameter sets).
88
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
5.3.
Results and Discussion
5.3.1. Model results of automatic parameter estimation
The calibration results showing the comparison of observed and simulated
streamflows are provided in Table 18, summarizing the daily NSE, R2 and VB
estimates. The NSE values were quite good for most of the catchments (i.e., >0.6),
with the exception of two catchments indicating values in the range of 0.41 to 0.46.
Similar patterns were indicated by R2 and VB, depicting reasonably good model
performance in most cases. Although, during the validation period, NSE and R2
values were lower than their corresponding values during the calibration period, the
values were reasonably good in most cases (i.e., NSE >0.5). Furthermore, the
performance results obtained in this study are in good agreement with those of other
model regionalisation studies (e.g., Abdulla and Lettenmaier 1997; Merz and
Blöschl 2004).
The calibration and validation results suggest that the optimized parameter sets
could simulate the rainfall-runoff relationships reasonably well in most cases.
However, it should be noted that the models are not perfect and may involve
uncertainties resulting from uncertainties in the model structure, input data and
parameter values (further discussed in section 5.3.3). Therefore, the results should be
interpreted cautiously. For example, in the case of the Sange Sorakh (ID: 3)
catchment the low performance was attributed to the high influence of groundwater
discharge of a spring which the model was not able to simulate well given the high
uncertainties in locating the boundaries of the karstified recharge area and
complexity of the hydrological processes. The low performance of the Afarineh (ID:
8) could be mainly attributed possibly to high uncertainty in the climatic input data
for this particular catchment due to less density of the climatic gauges in this area. In
this catchment, the model consistently overestimated the average flows resulting in a
high volume error and underestimated the high flood peaks.
Model Regionalization Based on the Flow Duration Curve
89
____________________________________________________________________
Table 18. HBV model calibration and validation results, showing daily NashSutcliffe efficiency (NSE), daily coefficient of determination (R2) and annual
volume balance (VB).
Catchment
ID
Name
Nash-Sutcliffe
Efficiency
(NSE, -)
Coefficient of
determination
(R2 -)
Volume Balance (VB)
Observed
Simulated
(mm/yr.).
(mm/yr.)
Difference
( %)
Calibration
-5
90
95
0.91
0.91
1 Aran
-12
104
118
0.78
0.76
2 Firoz Abad
0
332
332
0.46
0.46
3 Sange Sorakh
-13
148
171
0.89
0.88
4 Doabe Merek
0
39
39
0.67
0.66
5 Khers Abad
-7
326
349
0.70
0.64
6 Noor Abad
17
111
95
0.81
0.80
7 Dartoot
50
294
196
0.48
0.41
8 Afarineh
-5
349
367
0.80
0.80
9 Cham Injeer
-11
498
560
0.76
0.73
10 Sarab Seidali
-16
405
483
0.84
0.83
11 Kaka Raza
Validation
95
79
0.81
0.67
1 Aran
20
85
0.64
0.45
2 Firoz Abad
11
94
271
0.71
0.56
3 Sange Sorakh
-12
238
129
0.69
0.66
4 Doabe Merek
-26
96
30
0.69
0.68
5 Khers Abad
23
37
279
0.57
0.44
6 Noor Abad
11
309
94
0.46
0.25
7 Dartoot
13
106
144
0.58
0.11
8 Afarineh
107
298
315
0.66
0.56
9 Cham Injeer
5
331
471
0.68
0.59
10 Sarab Seidali
-4
450
371
0.77
0.75
11 Kaka Raza
0
370
Notes:
Dartoot and Sange Sorakh had missing streamflow data. For Dartoot the calibration and validation
results refer to the period October 1, 1994 to September 30, 2001 and October 1, 1990 to September
30, 1992, respectively. For Sange Sorakh the calibration and validation results refer to the period
October 1, 1987 to September 30, 1994 and October 1, 1999 to September 30, 2001, respectively. For
all other catchments the calibration and validation periods refer to October 1, 1987 to September 30,
1994 and October 1, 1994 to September 30, 2001, respectively.
90
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
5.3.2. Regionalization results based on drainage area, spatial proximity and
catchment characteristics
The summary of the catchment similarity analysis is presented in Table 19,
indicating most similar catchments whose parameters were transferred for the
regionalization purpose under each of the four tested methods.
Table 19. Results of the catchment similarity analysis for the four tested methods.
Catchment
Drainage area
ID
1
2
3
4
5
6
7
8
9
10
11
Name
Aran
Firoz Abad
Sange Sorakh
Doabe Merek
Khers Abad
Noor Abad
Dartoot
Afarineh
Cham Injeer
Sarab Seidali
Kaka Raza
Similar
catchment
Dartoot
Afarineh
Noor Abad
Kaka Raza
Cham Injeer
Sange Sorakh
Aran
Sarab Seidali
Khers Abad
Afarineh
Doabe Merek
Catchment similarity based on the studied methods
Spatial
Similarity index
Flow duration curve
proximity
Similar
Similar
Value Similar
Value
catchment
catchment
of S
catchment
of ε
0.28
0.85 Firoze Abad
Noor Abad
Firoz Abad
0.25
0.82 Aran
Aran
Aran
0.37
0.70 Cham Injeer
Kaka Raza
Sarab Seidali
0.84
0.81 Firoze Abad
Noor Abad
Khers Abad
2.32
0.75 Aran
Doabe Merek
Dartoot
0.29
0.85 Cham Injeer
Aran
Sarab Seidali
1.39
0.79 Aran
Cham Injeer
Khers Abad
0.99
0.67 Doabe Merek
Cham Injeer
Cham Injeer
0.27
0.79 Noor Abad
Dartoot
Kaka Raza
0.39
0.83 Cham Injeer
Kaka Raza
Noor Abad
0.61
0.83 Sarab Seidali
Sarab Seidali
Cham Injeer
The regionalization results for the calibration period are presented in Figure 27.
The results of transferring the model parameters based on similarity in area show
that in most cases the simulations were far away from the observed values in terms
of NSE, R2 and VB, with the exception of Kaka Raza (ID: 11) where the results were
reasonably good. The regionalization based on spatial proximity showed much better
simulations compared to those based on drainage area. Promising results were
obtained for four catchments, namely, Aran (ID: 1), Firoz Abad (ID: 2), Doabe
Merek (ID: 4) and Sarab Seidali (ID: 10), with NSE in the range of 0.51 to 0.78. But
a large number of catchments resulted in poor simulations, i.e., four catchments had
negative NSE values (ranging from -3.4 to -0.10). Similar to drainage area and
spatial proximity, the regionalization results based on catchment characteristics were
not better in most cases (Figure 27). Four out of 11 catchments produced
comparatively better results with NSE and R2 values in the range of 0.24 to 0.64 and
0.69 to 0.77, respectively. Rest of the catchments yielded poor results, particularly in
terms of VB and NSE. On the whole, the results suggest that the above mentioned
regionalization approaches are likely to produce unacceptable results in most cases.
Therefore, none of them could be recommended for the regionalization purposes in
the study region.
Model Regionalization Based on the Flow Duration Curve
91
____________________________________________________________________
5.3.3. Regionalization results based on FDC
The FDC plots for all the study catchments are shown in Figure 28 and their
similarities in terms of RRMSE (ε) are given in Table 19. In general, visual
comparison and the used objective criteria indicated good correspondence with each
other. Both visual comparison and ε values indicate that 7 out of 11 studied
catchments revealed good similarity with at least one catchment in the study group.
The ε values in these seven cases ranged from 0.25 to 0.61. The FDC-based
regionalization results for these catchments were reasonably good, with five out
seven catchments resulting in the NSE values in the range of 0.23 to 0.78 (Figure
27).
The R2 values were also good, ranging from 0.54 to 0.87. Similarly, most of them
depicted reasonably good performance in terms of VB. For instance, only two out of
these seven catchments produced, negative NSE values, but still could simulate
annual yields reasonably well (e.g., VBs for Sange Sorakh and Noor Abad were 1
and 24%, respectively). It is important to note that the Sange Sorakh catchment
yielded lower NSE and R2 values even during calibration. The lower performance,
during calibration, validation and regionalization could be attributed to the
significant contribution from a perennial spring, which the model was unable to
simulate well given the high uncertainties in locating the geographical boundaries of
the recharge area and the complexities in the hydrological processes in this region.
The FDCs of the remaining four catchments were not very similar to the rest of
the study catchments. However, for consistency in the number of catchments used in
all of the tested regionalization methods, we also executed FDC-based
regionalization for these catchments by transferring the parameters from the
catchment having the least value of ε. As expected, the results were not very good
when compared to those catchments where similarity was adequately defined.
Nevertheless, the outcome was comparable to the other three methods.
Furthermore, in most cases, the good regionalization results in case of tested
methods other than the FDC-based method correspond to the pair of catchments
having quite similar FDCs. For example, three out of four good performing
catchments in case of spatial proximity (e.g., Aran, Firoz Abad and Sarab Seidali)
also depicted similarity in the FDC of the corresponding neighbor.
NSE (-)
92
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
1.0
0.0
-1.0
-2.0
-3.0
-4.0
-5.0
-6.0
-7.0
-8.0
Area
Proximity
Similarity index
FDC
1 2 3 4 5 6 7 8 9 10 11
Catchments
1.0
0.6
Area
2
R (-)
0.8
Proximity
0.4
Similarity index
0.2
FDC
0.0
1 2
3 4 5
6 7 8 9 10 11
Catchments
400
VB (% )
300
200
Area
100
Proximity
Similarity index
0
FDC
-100
1 2 3 4 5 6 7 8 9 10 11
Catchments
Figure 27. Regionalization results of the four tested methods.
(The used catchment numbers in the x-axis correspond to the names as follow: 1: Aran; 2: Firoz Abad; 3:
Sange Sorakh; 4: Doabe Merek; 5: Khers Abad; 6: Noor Abad; 7: Dartoot; 8: Afarineh; 9: Cham Injeer;
10: Sarab Seidali; 11: Kaka Raza.)
Model Regionalization Based on the Flow Duration Curve
93
____________________________________________________________________
Daily streamflow (mm/d)
100.000
10.000
Aran
Firoz Abad
Sange Sorakh
1.000
Doabe Merek
Khers Abad
Noor Abad
0.100
Dartoot
Afarineh
Cham Injeer
0.010
Sarab Seidali
Kaka Raza
0.001
0
10
20
30
40
50
60
70
80
90
100
% of time indicated value exceeded
Figure 28. Comparison of FDCs for the similarity analysis.
5.3.4. Impact of parameter uncertainty on the regionalization results
The summary of the regionalization results using 50 best parameter sets for the FDC
based regionalization method is presented in Table 20 and Figure 29. The resulting
statistics given in Table 20 are reported in terms of median, 25th and 75th percentile,
minimum and maximum. The presented statistics were obtained by arranging the
results in descending order and then calculating various exceedance percentiles in a
way similar to well-known flow duration analysis. This analysis helped to quickly
view the degree of consistency when different parameter sets were used in the
regionalization. For instance, if the range of different percentiles is small, then the
impact of parameter uncertainty could be considered negligible. The results reveal
that, despite different parameter sets, the regionalization results were reasonably
consistent. This suggests that parameter uncertainty did not have considerable
impact on the regionalization outcome. For example, maximum NSE values,
achieved using the best parameter sets (as discussed in the previous sections 5.3.2
and 5.3.3) were not markedly different in most cases. This is further supported by
the fact that the good-performing catchments continue to perform well for all of the
50 tested parameter sets (Table 20 and Figure 29). Moreover, none of the low
performing catchments showed significant improvement as a result of using different
parameter sets. The similar inferences were drawn regarding impact of parameter
uncertainty on the regionalization results of the other three tested methods (not
shown here).
94
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Table 20. Impact of parameter uncertainty on regionalization results, illustrated by
the Nash-Sutcliffe efficiency (NSE) and coefficient of determination (R2) results
achieved for the 50 parameter sets used for the FDC-based regionalization method.
Catchment
ID
Name
1 Aran
2 Firoz Abad
3 Sange Sorakh
4 Doabe Merek
5 Khers Abad
6 Noor Abad
7 Dartoot
8 Afarineh
9 Cham Injeer
10 Sarab Seidali
11 Kaka Raza
Nash-Sutcliffe Efficiency (NSE,-)
Median
P25
P75
Min
0.79
0.80
0.79
0.78
0.66
0.67
0.65
0.57
-0.94
-0.80
-1.33
-1.77
0.36
0.39
0.35
0.24
0.26
0.44
0.11
-0.05
-0.59
-0.48
-0.64
-0.88
0.03
0.11
0.00
-0.11
-1.23
-1.22
-1.24
-1.31
0.39
0.41
0.37
0.29
0.26
0.28
0.23
0.16
0.59
0.60
0.58
0.56
Max
0.80
0.69
-0.41
0.45
0.47
-0.18
0.26
-1.11
0.57
0.31
0.62
Coefficient of determination (R2 , -)
Median
P25
P75
Min Max
0.87
0.86
0.86 0.86 0.84
0.75
0.73
0.74 0.73 0.69
0.32
0.25
0.27 0.25 0.20
0.81
0.79
0.80 0.79 0.77
0.56
0.54
0.54 0.53 0.51
0.29
0.26
0.26 0.25 0.21
0.59
0.27
0.28 0.26 0.25
0.12
0.11
0.12 0.11 0.11
0.57
0.62
0.59
0.60 0.59
0.63
0.61
0.62 0.59 0.55
0.74
0.73
0.74 0.73 0.71
Nash-Sutcliffe efficiency (NSE )
1.0
0.5
Aran
Firoz Abad
0.0
Sange Sorakh
Doabe Merek
Khers Abad
Noor Abad
-0.5
Dartoot
Afarineh
Cham Injeer
Sarab Seidali
-1.0
Kaka Raza
-1.5
-2.0
0
5
10 15
20
25
30 35
40
45
50 55
60
65
70 75
80
85
90 95 100
% of time NSE exceeded
Figure 29. Impact of parameter uncertainty on regionalization results, illustrated
by the exceeding percentiles of Nash-Sutcliffe efficiency (NSE) obtained from the 50
parameter sets used during regionalization based on similarity in the FDC.
5.3.5. Comparison of the FDC- based regionalization results with other studies
The results of this study indicate that the performance of the regionalization based
on the similarity of the FDC is superior to that of the other three tested methods.
Although, we could not test more methods due to limitations of the available data,
we compared our findings with related studies conducted elsewhere using other
Model Regionalization Based on the Flow Duration Curve
95
____________________________________________________________________
methods. The comparison was made between the results of the FDC-based
regionalization (Figure 27) with those of the studies presented in Table 14. The main
aim of this comparison is to obtain an overview of the comparative position of the
proposed FDC-based regionalization method among other widely recommended
regionalization methods. Moreover, this comparison cannot replace a rigorous
comparative assessment and is recommended as a future research activity.
Therefore, it is acknowledged that this comparison should be interpreted cautiously
because of inherent differences in the studies, i.e., differences in the amount and
quality of the data sets used and varying hydro-climatic environments, among
others.
The comparison reveals that the FDC-based regionalization approach stands very
well among the most promising techniques developed elsewhere. For instance, the
regionalization results based on the estimation of model parameters using catchment
characteristics, indicated variable degrees of success, as demonstrated by the wide
range of calculated performance measures (Table 14). The reported daily NSE values
for the parameter regionalization studies of Servat and Dezetter (1993) and Seibert
(1999) were in the range of 0.02-0.45 and 0.23-0.72, respectively. Similarly, the
studies of Servat and Dezetter (1993) and Abdulla and Lettenmaier (1997) reported
R2 values in the range of 0.62-0.99 and 0.05-0.81, respectively. A similar trend of
variable performance can be seen in many methods other than parameter
regionalization. For example, Merz and Blöschl (2004) achieved median NSE values
in the range of 0.32-0.56 for their eight regionalization methods tested for the 308
catchments, and Goswani et al. (2007) indicated NSE values in the range of -27.660.94 for their regional pooling method. The reported FDC-based regionalization
results of this study for five out of seven catchments (where FDC similarity was well
established) were in the range of 0.54-0.87 in terms of daily R2 values and 0.23-0.78
in terms of daily NSE values. These encouraging results suggest that model
regionalization based on the FDC similarity is a very good addition to the available
regionalization methods.
However, all of the tested methods, including the FDC-based regionalization,
resulted in some cases where the performance was not good. This suggests that the
problem of achieving successful outcomes for all model applications for the poorly
gauged or ungauged catchments still remains a challenging undertaking and, thus,
needs further research. This could be attributed to the nature of the problem at hand
as the degree of variability in the hydrological processes among different catchments
is very high. Therefore, supporting the regionalization results through other sources
of data and qualitative information is extremely desirable to avoid erroneous results.
Nonetheless, the chances of invalid results drawn by applying the FDC-based
regionalization method to poorly gauged catchments are likely to be small because at
least some estimates of the streamflow characteristics are available for comparison
in such cases (e.g., mean annual and monthly flows; various exceeding percentiles of
flow).
96
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
5.4.
Concluding Remarks
This study examined the application of the HBV model for the generation of
streamflow time series in data-limited catchments of the Karkheh Basin using model
parameters transferred from similar gauged catchments. The similarities of the
catchments for model parameter transfer were determined based on drainage area,
spatial proximity, catchment characteristics and flow duration curve (FDC).
Although the streamflow validation results based on spatial proximity and catchment
characteristics are better than those based on geographical area, the overall results
remain unsatisfactory in most cases. The study has shown that catchment similarity
analysis based on FDCs provides a sound basis for transferring model parameters
from gauged catchments to poorly gauged catchments in the Karkheh Basin. In most
cases, the simulated time series of streamflows resulted in reasonably good values of
the examined performance indicators (i.e., NSE, R2 and VB) with negligible impact
of the parameter uncertainty on the regionalization outcome. Furthermore, this new
method also compares well with the studies conducted elsewhere using other
promising methods. These demonstrations suggest that the new FDC-based
regionalization method is a valuable addition to the available regionalization
methods. The proposed method could be recommended for the practical applications
for estimating time series of streamflows for the poorly gauged catchments in the
mountainous parts of the Karkheh Basin. However, the poor performance in some
cases for the promising regionalization methods indicates the complexity of the
hydrological issues and of the regionalization problem and clearly highlights the
scope for further improvements. This essentially requires more effort on better
understanding the hydrology of ungauged or poorly gauged catchments and further
developments in the regionalization procedures, in particular with regard to widely
tested, and improving existing, methods, finding new regionalization approaches and
exploring innovative ways of using available (scarce) data sets.
The methodology presented in this thesis is easy to replicate in other river basins
of the world. Moreover, it can work well in the river basins, like the Karkheh Basin
of Iran, facing a decline in streamflow monitoring networks and/or having a limited
number of gauged catchments. Further testing of the proposed FDC-based
regionalization method is highly recommended, i.e., by using different rainfallrunoff models, application under different hydro-climatic conditions, and for
different extents of water resources development in the catchments (e.g., from more
pristine to more regulated catchments).
6.
IMPACT OF AREAL PRECIPITATION INPUT ON
STREAMFLOW SIMULATIONS8
6.1.
Introduction
The use of hydrological models in planning and management of water resources has
become the norm, and a wide array of hydrological models (including freeware) is
now available. The Soil Water Assessment Tool (SWAT) (Arnold et al. 1998;
Neitsch et al. 2005; Gassman et al. 2007) is one such model. The main data sets
required to formulate and run the model include a Digital Elevation Model (DEM),
land use, soil, climatic and land use management data sets. The quality of these
inputs has a significant impact on the model formulation process and on the results.
Many studies have investigated the impact of the resolution of DEM, soil and land
use data on the SWAT simulations (Chaplot 2005; Dixon and Earl 2009). Research
has also been devoted to examine the impact of catchment subdivisions on the
SWAT simulations (Jha et al. 2004; Tripathi et al. 2006). The studies on evaluating
the impact of climatic data input on SWAT simulations (discussed below) are
gaining increased attention, given the fact that climatic data are a major driver of
hydrological and other processes simulated by the model. The current way of
climatic data input in the SWAT is rather simplistic. Climatic data of a rain gauge
located nearest to the centroid of a subcatchment are used for that subcatchment.
This may not be accurate enough, particularly in regions where spatial heterogeneity
is high (e.g., mountainous terrains), or where data are sparse but spatial variability of
processes is not. This, in turn, has an impact on the model formulation process (e.g.,
parameterization) and quality of the simulated results (Oudin et al. 2006; Mul et al.
2009). For example, in response to over-predicted rainfall, the model
parameterization process may tend to increase ET to match the observed and
predicted streamflows. In many cases, finding the appropriate model structure and
parameter sets may not work well in delivering acceptable model simulations if the
input precipitation is inaccurate. Hence, improved precipitation input is very
important to obtain good results (Oudin et al. 2006; Mul et al. 2009; Tobin and
Bennet 2009).
8
This chapter is based on the paper Assessing the impact of areal Precipitation input on
streamflow simulations using the SWAT Model by Masih, I.; Maskey, S.; Uhlenbrook, S.;
Smakhtin, V. 2011. Journal of the American Water Resources Association 47(1):179-195.
DOI: 10.1111/j.1752-1688.2010.00502.x.
98
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The impact of different spatio-temporal resolution of rainfall input on simulated
runoff, using hydrological models other than SWAT, was examined in many studies
(e.g., Faurès et al. 1995; Maskey et al. 2004; Tetzlaff and Uhlenbrook 2005).
Although the results of these vary, they agree on the need to better represent
precipitation input in modeling. A brief review of the studies attempted to address
the issues of climatic data input in SWAT modeling is presented below.
Chaplot et al. (2005) studied the impact of rain gauge density on water, and
sediment and nitrogen fluxes in two small catchments (51 and 918 km2) in USA.
Their study indicated that the use of higher rain gauge densities resulted in better
simulations, particularly for sediment fluxes. Jayakrishnan et al. (2005) compared
monthly and annual streamflows simulated by SWAT for the four catchments (196
to 2,227 km2) in Texas, USA, by using rain gauge and radar (Next Generation
Weather Radar, NEXRAD) data sets. They found that input of areal rainfall
measured by radar performed better than that of the rain gauge data, despite some
inherent limitations of the latter, particularly problems of accuracy at daily time
scale. Watson et al. (2005) compared performances of three daily rainfall generation
models using SWAT for a 308 km2 catchment in Australia. They concluded that all
three rainfall inputs produced reasonably good simulations of mean annual runoff.
However, runoff variability was not well simulated given poor generation of rainfall
variability. Cho and Olivera (2009) evaluated the impact of the resolution of land
use, soil and precipitation data on simulated streamflows in three catchments (277 to
1005 km2) in the USA. They formulated 18 models of each catchment by combining
three land use, three soil types and two precipitation input scenarios. The two
precipitation scenarios used were: a) using data from all available rain gauges, and
b) using data from a single rain gauge for the whole catchment area. Each model was
independently calibrated and validated. All models produced comparable values of
daily Nash-Sutcliffe efficiencies. The main conclusion was that more refined
representation of spatial data may not necessarily result in improved SWAT
streamflow simulations in small catchments. Tobin and Bennett (2009) compared
monthly streamflows simulated by SWAT using precipitation data collected through
rain gauges, radar (NEXRAD stage III) and satellites (Tropical Rainfall
Measurement Mission, TRMM) at the outlet of the two rivers in USA (Middle
Nueces River catchment, 7,720 km2, and the Middle Rio Grande River catchment,
8,905 km2). Their findings revealed that streamflows were better simulated using
radar data compared to the other two sources of precipitation input. Starks and
Moriasi (2009) compared SWAT streamflow simulations using four resolutions of
precipitation data on three experimental catchments (75 to 342 km2). The number of
rain gauges in three scenarios varied from one to seven. The fourth scenario used the
radar precipitation data available at 4 km grid. The study indicated a satisfactory
calibration of the SWAT model in all four cases, although the data set with higher
rain gauge density and the radar-based precipitation produced comparatively better
streamflow simulations.
These studies have strongly pointed out the need for more research on finding
ways and means of improved precipitation input in SWAT simulations. Although
previous investigations are very helpful steps in this direction, they remained limited
Impact of Areal Precipitation Input on Streamflow Simulations
99
____________________________________________________________________
in many features highlighting the need for further research. For instance, most of
them were carried out in regions of USA which are generally considered data-rich
compared to other countries, particularly the developing world (e.g., radar data are
not available in many developing countries) and, therefore, remain limited to draw
general conclusions. The studies represent small- to medium-sized catchments (50 to
9,000 km2) and do not represent large river basins. Another important limitation is
the lack of spatio-temporal coverage. The model performance has been evaluated at
the catchment outlet in all cases, which prohibit explaining the spatial variations of
the studied processes within a catchment. Similarly, few of the abovementioned
studies using SWAT compared the performance at a daily time resolution, but were
mostly limited to annual and monthly time scales. These shortfalls limit our
understanding of the spatio-temporal impact of the improved input data on
hydrological and other processes. Furthermore, these knowledge gaps hamper the
informed basin-wide/regional planning and management of water resources (Santhi
et al. 2008). Therefore, there is also a clear need for studies highlighting the spatiotemporal variability of the studied processes when comparing the impact of different
sources of precipitation data on the model performance (Chaplot et al. 2005;
Jayakrishnan et al. 2005; Watson et al. 2005; Cho and Olivera 2009; Starks and
Moriasi 2009; Tobin and Bennett 2009).
The main research question addressed in this chapter is how improved
precipitation input influences the hydrological simulations and, hence, impact water
resources assessment across a large river basin. The specific objectives are: a) to
compare the SWAT performance achieved by using different areal precipitation
input, obtained by interpolation of the available rain gauge data and by using the rain
gauge data as per SWAT’s standard procedure; and b) to examine spatio-temporal
performance of the model simulations under both precipitation input scenarios. The
model was applied to the upper mountainous part of the Karkheh Basin (Figure 30a)
covering an area of 42,620 km2 from where almost all of the basin’s runoff is
generated (Figure 30b). The SWAT 2005 modeling system, version ARCSWAT 2.0
(Winchell et al. 2008) was used.
6.2.
Data and Methods
6.2.1. Data used in the model setup
The Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) of
90 m resolution was used for subcatchment definition. A drainage area of 300 km2
was used as the threshold for the delineation of subcatchments. This threshold was
chosen to balance the resolution of the available information. This way, the study
area was divided into 71 subcatchments (Figure 30b). The delineated subcatchments
were divided into different elevation bands using an elevation interval of 200 m.
This helped account for the topographic impacts on the climate. The value of
temperature lapse rate was set in the range of -2 to -5 0C/km, which is in close
100
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
agreement with the estimated values for the study area (JAMAB 1999). The used
values of precipitation lapse rate correspond to the annual lapse rate of 150-300
mm/km. This range is also in close agreement with earlier studies for the Karkheh
Basin (Sutcliffe and Carpenter 1968; JAMAB 1999; Muthuwatta et al. 2010). The
advantage of using ranges of precipitation and temperature lapse rates was to
account for the likely differences in the topography and orographic impacts across
the study basin. Since SWAT needs a daily precipitation lapse rate the annual values
were translated into the required format using a procedure similar to that described
by Fontaine et al. (2002). In the followed approach the annual values of lapse rates
are distributed among all of the precipitation events in a year.
Figure 30. The Karkheh basin and location of the selected streamflow gauges (a);
and the location of studied subcatchments and used climatic data stations (b).
The hydrological response units (HRUs) were defined based on information on
land cover, soil and slope. The land cover map was prepared (Ahmad et al. 2009)
using field data, GIS coverage and NDVI images based on remote sensing data from
Moderate Resolution Imaging Spectroradiometer (MODIS) of 250 m resolution. It
distinguishes 10 land use/land cover classes, with rain-fed farming (33%), forest
(23%), rangelands (18%) and bare lands (15%) constituting about 90% of the study
area. A digitized soil classification map was available from the Department of Soil
Impact of Areal Precipitation Input on Streamflow Simulations
101
____________________________________________________________________
and Water Research Institute (SWRI), Iran. This digital map had an original scale of
1: 1,000,000. The available soil map indicates fine to medium texture soils and rock
outcrops (shallow soils) being the dominant soils in the study area.
The initial values of most of the model parameters for the soil classes were
defined based on the results of the field tests conducted by the Iranian soil
department at various field locations in the Karkheh Basin. Information was
available about the soil texture, water content, soil depth, bulk density and some
other parameters. The soil albedo information was defined based on Mathew’s
seasonal integrated surface albedo (Mathews 1983). The information on soil albedo
was extracted from IWMI Integrated Data and Information System (IDIS) basin kit
product for the Karkheh Basin (http://dw.iwmi.org/idis_DP/home.aspx). The other
soil parameters were defined based on the SWAT soil data base, literature and field
information. The topographic slope was derived from the DEM by using SWAT’s
HRU definition tool. The three categories of slope were defined to be used in the
HRU definition, i.e., a) 0-8%; b) 8-30%; and c) > 30%. These slope categories
represent level to undulating lands (0-8% slope), steep lands (8-30% slope) and
mountains area (>30% slope) (FAO 1995). Finally, the HRUs were defined using
the land use, soil and slope information. A threshold value of 5% for land use, soil
and slope was used in the HRU definition. A threshold value of 5 to 10% is
commonly used in HRU definitions to avoid small HRUs, reduce total number of
HRUs and improve the computational efficiency of the model (Starks and Moriasi
2009; Tobin and Bennett 2009).
Daily climatic data for the period from January 1987 to September 2001 were
used for the model simulations. Precipitation data from 41 stations and temperature
data from 11 climatic stations were available. Locations of the used climatic gauges
are shown in Figure 30b. The missing data were patched by using data of other
stations based on a regression analysis. The study period was divided into a
calibration period from October 1987 to September 1994 and a validation period
from October 1994 to September 2001. In both periods, a warm-up period of 273
days was used to initialize the model.
6.2.2. Formulation of precipitation input scenarios
Climatic data of a station nearest to the centroid of a subcatchment was used in the
model simulations as per SWAT’s standard setup. This scenario of station
precipitation input is hereafter referred to as Case I. The interpolated precipitation
data were used as the model input in the second scenario (Case II). The inverse
distance and elevation weighting (IDEW) technique was used for the interpolation of
the available station data (see next section). The resulting precipitation was
aggregated at the subcatchment level. Then a virtual precipitation gauge having the
interpolated catchment precipitation was assigned for each of the 71 subcatchments.
In this scenario (Case II), model simulations were performed by changing the
precipitation data but keeping the rest of the data and the model structure the same
as in Case I. The formulated SWAT models for both scenarios were independently
102
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
calibrated using an automatic calibration procedure, discussed in detail under the
following section on the model calibration.
The model performance was evaluated at 15 streamflow gauges. The selected
gauges had a catchment area in the range of 590-42,620 km2 (Figure 30a and Table
21). These stations were well distributed across the Karkheh River system. The
studied gauges represent the primary- (Saymareh and Karkheh rivers), secondary(Gamasiab, Qarasou, and Kashkan rivers) and tertiary-level streams. Finally, a
comparison was made between the model performance achieved by Case I and Case
II. The hydrological performance was evaluated using the Coefficient of
Determination (R2) and the Nash-Sutcliffe Efficiency (NSE) (Nash and Sutcliffe
1970) measures at daily and monthly time scales (equations 13 and 14, respectively).
The relative difference in the observed and simulated mean annual streamflows was
also compared.
Table 21. Geographical characteristics of the selected streamflow gauges in the
Karkheh Basin.
Name of river
Name of station
Long
Lat
Khorram Rod
Toyserkan
Gamasiab
Qarasou
Abe Marg
Qarasou
Har Rod
Doab Aleshtar
Khorramabad
Chalhool
Kashkan
Bad Avar
Saymareh
Karkheh
Karkheh
Aran
Firoz Abad
Pole Chehre
Doabe Merek
Khers Abad
Ghore Baghestan
Kaka Raza
Sarab Seidali
Cham Injeer
Afarineh
Pole Dokhtar
Noor Abad
Holilan
Jelogir
Paye Pole
47.92
48.12
47.43
46.78
46.73
47.25
48.27
48.22
48.23
47.88
47.72
47.97
47.25
47.80
48.15
34.42
34.35
34.33
34.55
34.52
34.23
33.72
33.80
33.45
33.30
33.17
34.08
33.73
32.97
32.42
Elevation
(masl)
1,440
1,450
1,280
1,310
1,320
1,268
1,530
1,520
1,140
800
650
1,780
1,000
450
125
Drainage
area (km2 )
2,320
844
10,860
1,260
1,460
5,370
1,130
776
1,590
800
9,140
590
20,863
39,940
42,620
Number
of rain
gauges
(No.)
2
1
11
2
2
7
2
1
1
1
6
1
20
32
32
Rain gauge
density, (1
station per
km2 )
1,160
844
987
630
730
767
565
776
1,590
800
1,523
590
1,043
1,248
1,332
Notes: Long = Longitude (degrees East) and Lat = Latitude (degrees North).
Data source: Ministry of Energy, Iran, barring the last two columns.
Preparation of the precipitation input for Case II
The earlier studies for the Karkheh Basin have demonstrated that topography has a
strong influence on the spatial distribution of precipitation in this mountainous
region (Sutcliffe and Carpenter 1968; JAMAB 1999; Muthuwatta et al. 2010).
Elevation is known to be an important factor governing the spatial variability. These
findings are in general agreement with those of other mountainous regions of the
world (e.g., Daly et al. 2002). Moreover, the rain gauge data may not adequately
Impact of Areal Precipitation Input on Streamflow Simulations
103
____________________________________________________________________
represent the precipitation over an entire catchment. This issue is further exacerbated
for catchments where rain gauge density is lower, such as for the region under study.
Under such conditions, areal precipitation is likely to represent catchment conditions
better compared to the station data.
For Case II, the daily station data were interpolated and aggregated at the
subcatchment level using the IDEW technique. The hydrological data processing
software called HyKit developed at UNESCO-IHE was used (Maskey 2007). The
distance weighting method has already proven to perform well as compared to some
other standard methods of perception regionalization for the Karkheh and its
neighboring basins in the Zagros mountains, Iran (Saghafian and Davtalab 2007).
The method has also been successfully used in other regions of Iran (Modallaldoust
et al. 2008). The HyKit also offers the possibility of defining elevation weighting
along with the distance weighting, making it more suitable for mountainous regions
where topographic impacts on precipitation are important. The mathematical form of
the equation used for interpolation is as follows:
N
pˆ k = WD 
i =1
N
1
1
w(d ) i pi + WZ  w( z ) i pi
D
i =1 Z
(15)
where, p̂ in mm per time step is the interpolated precipitation for a grid cell, WD
(-) and WZ (-) are the total weighting factors for distance and elevations,
respectively, pi is the precipitation value in mm per time step of the ith gauge station
and N is the number of gauges used in the interpolation for the current grid cell.
Similarly, w(d)i (-) and w(z)i (-) are the individual gauge weighting factors for
distance and elevation, respectively, and D (-) and Z (-) are the normalization
quantities given by the sum of individual weighting factors w(d)i and w(z)i,
respectively, for all the gauges used in the interpolation. The weighting factors w (d)i
and w(z)i based on inverse of distance and elevation, are given by the following
equations:
w(d ) = 1 / d a
for d > 0
b
1 / z min
for z ≤ z min
 b
w( z ) = 1 / z
for z min < z < z max

for z ≥ z max
0
(16)
(17)
104
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
where, d is the distance in km between the current grid and the gauge station
used for interpolation, z is the absolute elevation difference (expressed in m)
between the current grid cell and the gauge station used for interpolation, a and b are
exponent factors for distance and elevation weightings, respectively. The exponents
(a and b) and the abovementioned weighting factors are dimensionless numbers and
zmin and zmax (expressed in m) are the minimum and maximum limiting values of
elevation differences for computing elevation weightings (Daly et al. 2002). The use
of zmin helps avoid the dominance of the stations having very small elevation
difference (e.g., 10s of meters) from the target cells. The typical value of zmin varies
from 100 to 300 m. The limit on maximum elevation difference enables data point
inclusion to be restricted to a local elevation range. A typical zmax ranges from 500 to
2,500 m.
Note that in this interpolation, no grid cell contains more than one gauge station
and that the grid cell which contains a gauge station will retain the same
precipitation as that of the gauge station. The main advantage of distance weighting
technique is its simplicity and ease of application to large data sets (e.g., daily time
series). The inclusion of elevation weighting is helpful for improving the results in
the mountainous regions where elevation could play a major role in the precipitation
distribution. The method also has some limitations that mainly relate to the careful
choice of the sensitive parameters.
Daily time series of precipitation from all of the 41 available gauges were used
for interpolation in 5×5 km2 grids, which are then aggregated to subcatchments as
defined in the SWAT model. The parameters used in the interpolation were defined,
primarily based on recommendations from the available literature and carrying out a
cross validation exercise. The final parameter values were: a = 2, b = 1, d = 70, WD =
0.8, WZ = 0.2, zmin = 100 m and zmax = 1,500 m. The used parameter values were in
good agreement with the literature (Daly et al. 2002). The interpolated results were
cross-validated at 10 selected rain gauge locations/grid cells. The validation was
done using the Jack-knife cross-validation approach (Quenouille 1956). In this
method, interpolation runs were carried out for each of the 10 validation stations
using data of all other stations excluding the current validation station (e.g., 41-1 =
40 in this case). Then, the interpolated and observed data for that station were
compared by estimating R2 between them. The interpolated values were in good
agreement with the observed ones. The mean and standard deviations of monthly R2
were 0.91 and 0.04, respectively. As expected, the daily R2 values were
comparatively lower than the monthly ones (with mean R2 of 0.62 and standard
deviation of R2 of 0.13). However, considering high spatial variability of
precipitation in this mountainous terrain, the achieved R2 values were considered
satisfactory.
Moreover, a correlation analysis among the rain gauge stations was also carried
out to further evaluate the used radius of influence. The analysis indicated that most
of the stations falling within a distance of 70 km exhibited a good correlation with
one another (e.g., greater than 0.8 at monthly time scale). The 70 km radius of
influence ensured the use of 2-15 stations in the interpolation for a subcatchment.
Generally, the interpolation used more stations in the subcatchments located in the
Impact of Areal Precipitation Input on Streamflow Simulations
105
____________________________________________________________________
upper part of the study area because of the high station density compared to the
middle and lower parts (Figure 30b).
6.2.3. Model calibration
The main options used in the SWAT model set up included: a) Soil Conservation
Services Curve Number (SCS-CN) method, with crack flow not active, for
estimating surface runoff (Soil Conservation Service Engineering Division 1986), b)
Hargreaves method for daily potential evapotranspiration calculation (Hargreaves et
al. 1985), and c) variable storage method for water routing in the streams (Williams
1969).
The formulated SWAT models for Case I and Case II scenarios were
independently calibrated using an auto-calibration procedure. The SWAT-CUP
software was used for this purpose (Abbaspour 2008). The Sequential Uncertainty
Fitting algorithm (SUFI-2) was applied for the parameter optimization (Abbaspour
et al. 2004; 2007). The SUFI-2 optimization follows 9 major steps, discussed in
detail by Abbaspour et al. (2007), which are enumerated below.
1.
2.
3.
4.
5.
6.
7.
8.
9.
An objective function is selected from the given options (e.g., R2 or NSE
etc.).
Physically meaningful ranges of the parameters are defined. Generally,
wide ranges are suggested at this first step, which are revised in the
following rounds of analysis.
A sensitivity analysis is performed to get a first hand view of the
sensitive parameters.
Initial uncertainty ranges are assigned to parameters for the first round
of Latin Hypercube sampling.
A Latin Hypercube sampling is carried; leading to n parameter
combinations, where n is the number of desired simulations.
The simulations are assessed by estimating the objective function
values.
A series of measures (e.g., sensitivity matrix) is calculated to evaluate
each sampling round.
Measures assessing the uncertainties are calculated.
Because parameter uncertainties are initially large, the value of
uncertainty measures tends to be quite large during the first sampling
round. Hence, further sampling rounds are needed with updated
parameter ranges. In this step, new parameter ranges are suggested,
which are generally narrower than those defined in step 2. Then the
whole procedure is repeated until desired results on parameter
optimization are achieved.
This computationally efficient procedure is being increasingly used in the recent
SWAT applications (e.g., Faramarzi et al. 2009) and is known to produce
comparable results with widely used auto-calibration methods (Yang et al. 2008).
106
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The parameters were optimized using R2 as the objective function. Using a
procedure similar to that adopted by Faramarzi et al. (2009), the SWAT model was
simultaneously calibrated using daily streamflow data of the abovementioned 15
streamflow gauging stations. Hence, the best parameter set was the one which
produced maximum value of the average R2. The NSE was not used as an objective
function in this study, mainly due to the possibility of a badly simulated station (with
a large negative value) dominating the optimization process. However, the results
were also evaluated in terms of NSE and annual volume balance (VB), in addition to
R2. Using more than one performance evaluation measure was considered helpful in
evaluating the robustness of the calibration process. This was also useful in
compensating for the specific limitation of a specific performance evaluation
criterion.
The SWAT-CUP offers the possibility of selecting an objective function from the
six different available options (e.g., sum of squares, R2, weighted R2 associated with
slope-termed as bR2, and NSE). Each of the abovementioned performance measure
has its own merits and limitations (e.g., ASCE 1993; Krause et al. 2005; Gupta et al.
2009). Generally, these objective functions tend to better fit the simulated
hydrographs to the high flows to achieve a higher value of the objective function,
which often comes at the expense of relatively poor simulation of the low flows
(e.g., Krause et al. 2005). The use of most widely applied and well- recommended
performance measures, i.e., R2, NSE and VB, was considered appropriate for the
purpose of this study.
Before applying auto-calibration, a rigorous manual calibration exercise was
performed. This helped in defining suitable initial values/ranges of the parameters,
which were based on information from various sources that included measured data,
global data sources, the SWAT soil and land cover database, literature, discussion
with the local experts and field visits. For instance, the initial values of the
parameters of the snow routine were defined in a way to obtain resultant snowfall
values in good agreement with a recent study in the region (Saghafian and Davtalab
2007). Moreover, the used parameter values/ranges were in line with the literature
(Fontaine et al. 2002; Jones et al. 2008). The SCS curve number values were varied
for each of the land use categories and the used values were in close agreement with
the literature (Soil Conservation Service Engineering Division 1986). Similarly, the
selected parameters of the groundwater routine were spatially varied based on the
information available from earlier studies (JAMAB 1999; Tizro et al. 2007; Masih et
al. 2009). The finally used values and/or ranges resulted from the auto-calibration
procedure, for both scenarios are presented in Table 22.
Impact of Areal Precipitation Input on Streamflow Simulations
107
____________________________________________________________________
Table 22. Appropriate values and/or ranges of the selected parameters used in
setting up SWAT model for the cases I and II.
Parameter
Snowfall temperature, SFTMP (oC)
Snowmelt temperature, SMTMP (oC)
Maximum melt rate of snow during a year
(occurring in summer solstice), (mm oC-1 d-1)
Minimum snowmelt rate during a year
(occurring in winter solstice), (mm oC-1 d-1)
Snowpack temperature lag factor (TIMP)
Minimum snow water content that corresponds
to 100% snow cover, SNOCOVMX (mm)
Snow water equivalent that corresponds to
50% snow cover
Soil available water capacity (SOL_AWC),
mm/mm
Soil
saturated
hydraulic
conductivity
(SOL_K), (mm/hr)
Maximum Soil Depth (SOL_ZMX), (mm)
Soil evaporation compensation factor, ESCO
Plant uptake compensation factor, EPCO
Soil albedo, SOL_ALB
Soil bulk density, SOL_BD (g/cm3)
Curve number, CN
Manning’s n value for overland flow
Surface runoff lag coefficient (SURLAG)
Maximum canopy storage, CANMX (mm)
Base flow recession constant (ALPHA_BF)
Groundwater delay from soil to groundwater,
(GW_DELAY), (d)
Fraction of total aquifer recharge percolated to
deep groundwater, RCHRG_DP
Manning’s n value for the main channel,
CH_N2
Effective hydraulic conductivity in the main
channel alluvium, CH_K2 (mm/hr)
Final
value or
ranges
used for
Case I
Final
value or
ranges
used for
Case II
-5-5
-5-5
0-10
Mode
of
change
during
autocalibration
runsa
v
v
v
2.8
1.2
1.9
1.06
0.16
4.3
0.71
0.66
0.46
0-10
v
1.5
2.1
0.81
0-1
0-500
v
v
0.7
127
0.96
492
1.22
2.12
0-1
v
0.3
0.2
6.10
0-1
r
0.06-0.21
0.11-0.41
1.06
0-2000
r
4.5-36
4.8-38.4
7.79
0-3500
0-1
0-1
0-0.25
0.9-2.5
30-100
0.01-30
1-24
0-100
0-1
0-500
r
v
v
r
r
r
r
v
r
r
r
150-2400
0.61
0.78
0.09-0.11
1.70-1.99
57-90
0.02-1.61
1
0-4.65
0.1-0.3
9-129
118-1180
0.87
0.65
0.08-0.10
1.62-1.82
56-88
0.01-0.94
4.65
0-4.85
0.08-0.24
24-144
3.76
0.003
0.56
0.04
3.10
3.20
1.53
6.41
1.29
2.98
5.96
0-1
r
0.07-0.70
0.08-0.75
1.57
-0.01-0.3
v
0.19
0.21
17.64
-0.01-500
v
2.26
1.11
0.53
Suggested
ranges
in
SWAT
Parameter
sensitivity
indicated
by t valueb
Notes:
a
v refers to the absolute change in the parameter made by replacing a parameter by a given value; r refers
to the relative change in the parameter made by multiplying the parameter by 1 plus the factor in the given
range (Abbaspour et al. 2007)
b
t-value indicates parameter sensitivity; higher values refer to more sensitive parameters and vice versa.
108
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The overall automatic calibration process done for this study consisted of three
SWAT-CUP iterations, each composed of 250 simulations. As per the SWAT
structure some of the parameters can only be defined for the whole study area and
therefore will only have one global value for all subcatchments (e.g., parameters of
the basin file including snow routine and SURLAG). The sensitivity analysis of the
selected parameters was also conducted using the sensitivity analysis tool included
in the SWAT-CUP. This helped understand the relative importance of the selected
parameters for the study region. The results are presented in Table 22, indicating the
relative sensitivity of each parameter. The results of the sensitivity analysis are
generally in agreement with the literature (van Griensven et al. 2006; Faramarzi et
al. 2009; Tobin and Bennet 2009).
It is important to recognize here that good model simulations can be achieved using
various combinations of the model parameters. Therefore, the calibrated parameter
values given in Table 22 do not necessarily represent the uniquely best parameter
combination. This issue is well comprehended in hydrology and often termed as
equifinality or nonuniqueness of the parameters (Uhlenbrook et al. 1999; Beven
2001).
6.3.
Results and Discussions
6.3.1. Comparison of precipitation input
Figure 31a presents the mean annual precipitation under Case II. The results showed
substantial variations in annual totals ranging from 370 to 640 mm/yr. Generally, the
subcatchments located in the northeast, central and southern parts of the study area
depict comparatively lower precipitation whereas, the catchments located in the
southeast parts of the study basin have the highest precipitation. The topography
seems to be the major driver of these spatial variations. Westerly winds are the main
source of moisture in the study area (Demroes et al. 1998), which are strongly
influenced by the topographic features causing high spatial variability. In general,
the presented pattern of spatial variability is in good agreement with the available
precipitation maps of the region (Saghafian and Davtalab 2007; Muthuwatta et al.
2010).
Impact of Areal Precipitation Input on Streamflow Simulations
109
____________________________________________________________________
Figure 31. Mean annual precipitation for Case II (a) and percentage difference
between Case II and Case I (b).
Figure 31b shows the percentage difference in the mean annual precipitation in
Case II compared with Case I. The comparison revealed significant differences,
indicating both increases and decreases in the range of -38 to 42%. The overall
precipitation dynamics in Case II could be different from Case I in a number of
ways. This is illustrated by the comparison of daily, monthly and annual
precipitation for three selected subcatchments (Figure 32). Since SWAT simulates
streamflow and other processes at daily time scale, it is more important to closely
examine the differences in daily precipitation to understand the impact on simulated
results. The daily values in Case II could be higher, lower or similar as compared to
Case I. However, they show a clear pattern in extreme values. Generally, lower
precipitation events can be totally missed out by a single rain gauge. These events
are better accounted for in Case II. This is shown in Figure 32 by the daily
precipitation values extending up to 20 mm/d in Case II (along the x-axis) when
Case I shows no precipitation (zero value for the y-axis as indicated by some of the
data points falling on the x-axis line). The high precipitation extremes in Case II are
comparatively smaller in most cases when compared to Case I, though they could be
the other way round for some subcatchments and events because these outcomes
largely depend upon the influence of the neighboring gauge records.
110
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
The abovementioned differences in the precipitation under Case I and Case II are
substantiated by Figure 33, which presents daily precipitation under both cases for a
selected area (subcatchment ID: 1) for a short but significant period in a year. The
figure indicates that, in most cases, the high precipitation values were lower in Case
II than in Case I, with the exception of a few with opposite results. For example, 1
mm of precipitation was recorded on March 2, 1996 by the rain gauge used for the
selected area. In contrast, neighboring gauges recorded a quite high precipitation on
that day (up to 52 mm). Consequently, the interpolated precipitation based on
records of several stations had a higher value for that day compared to the record of
only one station. Similarly, under Case II, some precipitation values were attributed
to the days which show no rainfall at all under Case I. This could be due to
precipitation occurring in the neighboring areas but not on the area where the rain
gauge used under Case I was located.
Subcatchm ent ID: 33
80
60
40
20
0
0
20
40
60
80
Subcatchm ent ID: 33
300
250
200
150
100
50
0
100
0
P Case II (m m/day)
60
40
20
0
20
40
60
80
100
50
0
0
P Case II (m m /day)
600
400
200
0
0
100
250
200
150
100
50
0
0
50 100 150 200 250 300
P Case II (m m/month)
200
400 600
800 1000
P Case II (m m /year)
Subcatchm ent ID: 1
300
800 1000
800
100 150 200 250 300
1000
P Case I (mm/year)
20
80
50
400 600
Subcatchm ent ID: 43
1000
150
0
P Case I (mm/month)
P Case I (mm/day)
40
60
200
P Case II (m m /year)
P Case II (m m /m onth)
60
40
0
0
200
100
80
20
200
50 100 150 200 250 300
250
Subcatchm ent ID: 1
0
400
Subcatchm ent ID: 43
300
P Case II (m m /day)
100
600
P Case I (mm/year)
P Case I (mm/month)
P Case I (mm/day)
80
0
800
P Case II (m m /m onth)
Subcatchm ent ID: 43
100
Subcatchm ent ID: 33
1000
P Case I (mm/year)
P Case I (mm/month)
P Case I (mm/day)
100
Subcatchm ent ID: 1
800
600
400
200
0
0
200 400
600 800 1000
P Case II (m m /year)
Figure 32. Comparison of daily, monthly and annual precipitation among Case I and
Case II, illustrated by three selected subcatchments.
Impact of Areal Precipitation Input on Streamflow Simulations
111
____________________________________________________________________
27
Precipitation (mm/d)
Case I
Case II
18
9
0
1
4
7
10
13
16
19
22
25
28
31
Days in March 1996
Figure 33. Comparison of daily precipitation for Cases I and II for a selected month
(March 1996).
6.3.2. Comparison of streamflow simulations
The summary of the results on the studied performance measures are presented in
Table 23a and 23b. Scatter plots of NSE and R2 are presented in Figure 34, providing
a quick overview of the comparison between Case I and Case II. Moreover, the
magnitude of changes in daily NSE and R2 in Case II compared to Case I is reflected
in Figure 35. The monthly differences follow a pattern similar to that of the daily
results (not shown in Figure 35 but could be inferred from Table 23a and 23b). The
results indicate variable performance under both cases, indicating both increases and
decreases. Nonetheless, better results were obtained under Case II in most cases, in
particular for the smaller catchments. In general, streamflow regimes of the Karkheh
River and its major tributaries were modeled reasonably well under both cases
during calibration as well as validation periods. For example, monthly NSE values
during calibration and validation periods for the two selected gauges on the Karkheh
River, Jelogir and Paye Pole, were 0.76-0.89 and 0.77-0.91 under Case I and Case II,
respectively. The corresponding NSE values at daily time scale were 0.69-0.83 and
0.65-0.81.
A paired t-test was applied to investigate the statistical significance of the
observed differences on the whole. The significance of the resultant test statistics is
noted in Table 23a and 23b. The results depict significant improvement for the NSE
under Case II compared with Case I both at daily and monthly time resolutions,
indicated by significant results at 95% confidence level. Similarly, NSE values under
Case II were significantly better under both calibration and validation periods than
under Case I.
Significant improvements in monthly R2 during the calibration period were
noted, but in the rest of the cases R2 values were comparable under both scenarios.
The simulated volume balance was significantly better in Case II than in Case I
112
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
during validation period (in particular). On the whole, the results of the used paired
t-test indicated significant improvement, in general, in Case II than in Case I.
However, these statistical inferences should be interpreted cautiously given the
limitations. The sample size under study is very small (only 15 cases), which limits
inferring sound general conclusions.
Because of the considerable improvements, though few in number, it would be
worthwhile applying SWAT to large river basins such as the one under the current
study. The marked improvement even in the few subcatchments would improve the
overall reliability of the results, if the cases of deteriorated performance remain
comparatively negligible. Defining the considerable level of change will depend on
the context of the study. For example, in case of change in NSE, an improvement
will not matter much if the NSE values remain negative in both cases, i.e., in general
the model performs poorer than the average scenario under both conditions.
Therefore, it is important to carefully examine the differences under each case. A
further discussion on the cases with similar, improved and deteriorated performance
is presented below.
Impact of Areal Precipitation Input on Streamflow Simulations
113
____________________________________________________________________
Table 23a.
Comparison of the model simulations during calibration under Case I
and Case II.
ID
Name of Gauge
Drainage
area (km2 )
Daily
2
Case I
Aran
1.1
Firoz Abad
1.2
Pole Chehre
1.3
Doabe Merek
2.1
Khers Abad
2.2
Ghore Baghestan
2.3
Kaka Raza
3.1
Sarab Seidali
3.2
Cham Injeer
3.3
Afarineh
3.4
Pole Dokhtar
3.5
Noor Abad
4.1
Holilan
4.2
Jelogir
4.3
Paye Pole
4.4
Case II
1.1
Aran
1.2
Firoz Abad
1.3
Pole Chehre
2.1
Doabe Merek
2.2
Khers Abad
2.3
Ghore Baghestan
3.1
Kaka Raza
3.2
Sarab Seidali
3.3
Cham Injeer
3.4
Afarineh
3.5
Pole Dokhtar
4.1
Noor Abad
4.2
Holilan
4.3
Jelogir
4.4
Paye Pole
Paired t-test significancea
Calibration (October 1987 to September 1994)
Monthly
Mean annual flow (m3/s)
Simulated
Difference
R2
NSE
Observe
(m3/s)
(%)
d (m3/s)
R
NSE
2,320
844
10,860
1,260
1,460
5,370
1,130
776
1,590
800
9,140
590
20,863
39,940
42,620
0.74
0.51
0.82
0.61
0.45
0.81
0.65
0.68
0.78
0.51
0.81
0.35
0.83
0.87
0.76
0.72
0.17
0.75
0.36
-0.12
0.81
0.47
-0.95
0.64
-0.01
0.78
-0.58
0.81
0.83
0.74
0.87
0.72
0.91
0.81
0.71
0.90
0.73
0.87
0.90
0.63
0.92
0.57
0.91
0.95
0.93
0.85
0.55
0.88
0.71
-0.21
0.89
0.50
0.25
0.76
-1.36
0.87
-0.09
0.90
0.89
0.89
5.0
1.9
41.5
6.7
1.8
25.3
14.7
9.6
12.6
4.7
64.7
4.5
86.7
184.5
209.6
4.1
2.1
37.6
7.1
2.6
23.3
8.3
10.6
12.8
7.9
64.5
3.6
74.6
181.8
183.2
-18
11
-9
6
43
-8
-44
10
1
67
0
-21
-14
-1
-13
2,320
844
10,860
1,260
1,460
5,370
1,130
776
1,590
800
9,140
590
20,863
39,940
42,620
0.77
0.52
0.81
0.65
0.51
0.82
0.79
0.66
0.78
0.52
0.79
0.53
0.84
0.83
0.72
NS
0.77
0.48
0.80
0.63
0.14
0.80
0.63
-0.24
0.58
0.52
0.78
0.33
0.82
0.81
0.70
**
0.90
0.71
0.90
0.90
0.81
0.93
0.88
0.84
0.90
0.62
0.93
0.72
0.92
0.93
0.91
**
0.90
0.71
0.88
0.88
0.02
0.91
0.67
0.54
0.75
0.56
0.91
0.60
0.90
0.91
0.88
**
5.0
1.9
41.5
6.7
1.8
25.3
14.7
9.6
12.6
4.7
64.7
4.5
86.7
184.5
209.6
NA
5.2
1.9
36.6
5.4
2.7
21.0
9.9
8.7
13.1
4.6
56.4
4.3
73.7
177.1
179.1
**
4
-1
-12
-19
48
-17
-33
-10
4
-3
-13
-5
-15
-4
-15
NS
Note:
a
The significance of the paired t-test refers to: NS: not significant; **: significant at 95% confidence
level; NA: not applicable
114
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Table 23b. Comparison of the model simulations during validation under Case I
and Case II.
ID
Name of Gauge
Case I
1.1
Aran
1.2
Firoz Abad
1.3
Pole Chehre
2.1
Doabe Merek
2.2
Khers Abad
2.3
Ghore Baghestan
3.1
Kaka Raza
3.2
Sarab Seidali
3.3
Cham Injeer
3.4
Afarineh
3.5
Pole Dokhtar
4.1
Noor Abad
4.2
Holilan
4.3
Jelogir
4.4
Paye Pole
Case II
1.1
Aran
1.2
Firoz Abad
1.3
Pole Chehre
2.1
Doabe Merek
2.2
Khers Abad
2.3
Ghore Baghestan
3.1
Kaka Raza
3.2
Sarab Seidali
3.3
Cham Injeer
3.4
Afarineh
3.5
Pole Dokhtar
4.1
Noor Abad
4.2
Holilan
4.3
Jelogir
4.4
Paye Pole
Paired t-test significancea
Drainage
area (km2 )
Daily
NSE
R
2
Validation (October 1994 to September 2001)
Monthly
Mean annual flow (m3/s)
Simulated Difference
R2
NSE
Observed
(m3/s)
(%)
(m3/s)
2,320
844
10,860
1,260
1,460
5,370
1,130
776
1,590
800
9,140
590
20,863
39,940
42,620
0.73
0.40
0.76
0.23
0.67
0.73
0.69
0.67
0.61
0.41
0.73
0.57
0.86
0.87
0.73
0.30
-2.48
0.71
0.08
-0.12
0.72
0.61
-1.35
-0.63
-0.94
0.56
0.37
0.84
0.74
0.69
0.83
0.49
0.86
0.30
0.88
0.80
0.84
0.73
0.72
0.73
0.88
0.71
0.88
0.93
0.85
0.40
-1.72
0.80
0.21
-0.40
0.79
0.74
-1.06
-0.75
-1.68
0.52
0.34
0.87
0.76
0.78
3.7
1.0
28.3
4.7
1.3
17.0
10.7
7.3
9.9
3.4
47.1
3.0
63.7
140.1
167.7
5.3
1.9
31.3
3.9
2.3
15.2
7.8
9.2
13.2
7.0
60.8
3.7
58.9
156.6
160.1
42
85
10
-18
73
-11
-27
26
34
106
29
21
-8
12
-5
2,320
844
10,860
1,260
1,460
5,370
1,130
776
1,590
800
9,140
590
20,863
39,940
42,620
0.72
0.50
0.72
0.49
0.56
0.71
0.74
0.68
0.67
0.55
0.74
0.54
0.85
0.82
0.66
NS
0.63
-0.04
0.71
0.48
-0.29
0.70
0.67
-0.06
-0.20
0.44
0.72
0.35
0.84
0.78
0.65
**
0.81
0.57
0.82
0.70
0.88
0.84
0.85
0.73
0.80
0.70
0.89
0.74
0.88
0.90
0.79
NS
0.70
0.11
0.82
0.66
-0.07
0.83
0.78
0.15
-0.15
0.62
0.82
0.40
0.88
0.84
0.77
**
3.7
1.0
28.3
4.7
1.3
17.0
10.7
7.3
9.9
3.4
47.1
3.0
63.7
140.1
167.7
NA
5.3
1.6
30.1
3.5
2.3
14.9
9.0
7.4
12.4
4.5
51.8
3.8
58.6
148.2
151.0
**
43
57
6
-26
75
-12
-16
1
26
33
10
23
-8
6
-10
**
Note:
a
The significance of the paired t-test refers to: NS: not significant; **: significant at 95% confidence
level; NA: not applicable
Impact of Areal Precipitation Input on Streamflow Simulations
115
____________________________________________________________________
Daily R 2
Daily NSE
1.0
1.0
0.8
Case II
Case II
0.5
0.0
-0.5
0.6
0.4
0.2
0.0
-1.0
-1.0
-0.5
0.0
0.5
0
1.0
0.2
Case I
0.6
0.8
1
0.8
1.0
Case I
Monthly R 2
Monthly NSE
1.0
1.0
0.5
0.8
0.0
0.6
Case II
Case II
0.4
-0.5
0.4
0.2
-1.0
0.0
-1.5
-1.5
-1.0
-0.5
0.0
Case I
0.5
1.0
0.0
0.2
0.4
0.6
Case I
Figure 34. Scatter plots of NSE and R2, highlighting the comparative performance
under cases I and II.
116
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Difference in daily NSE
3.0
Calibration
2.0
Validation
1.0
0.0
1,260
1,130
844
800
776
590
1,260
1,130
844
800
776
590
1,460
1,590
2,320
5,370
9,140
10,860
20,863
39,940
42,620
-1.0
2
Catchment area (km )
Difference in daily R
2
1.0
0.5
0.0
-0.5
1,460
1,590
2,320
5,370
9,140
10,860
20,863
39,940
42,620
-1.0
2
Catchment area (km )
Figure 35. difference in the daily NSE and R2 in Case II as compared to Case I for
the calibration and validation periods.
Stations indicating good performance in both cases
The simulated results corresponded well with the observed values under both cases
in six out of 15 studied flow gauges. These stations are: Paye Pole, Jelogir, Holilan,
Pole Chehre, Pole Dokhtar and Ghore Baghestan. The daily NSE during the
calibration period in Case I and Case II ranged from 0.74 to 0.83 and from 0.70 to
0.82, respectively. The corresponding values during the validation period were in the
range of 0.56-0.84 and 0.65-0.84, respectively. Furthermore, there were no marked
differences in the values of R2 and NSE in all of these gauges, though with the
exception of Pole Dokhtar where a noteworthy improvement of 0.3 in the monthly
NSE for the validation period was observed. All of these stations represent primaryand secondary-level streams. This suggests that the change in precipitation input had
minimal impact for larger catchments (in general). Daily hydrographs (observed and
Impact of Areal Precipitation Input on Streamflow Simulations
117
____________________________________________________________________
simulated) for the Jelogir station at the Karkheh River are shown in Figure 36a, as an
example, indicating an almost similar pattern of simulations under both cases.
Mean Daily Discharge (m 3/s)
(a)
Jelogir Station at the Karkheh River (39,940 km 2)
1500
Observed
1200
Simulated Case I
900
Simulated Case II
600
300
0
1
16
31
46
61
76
91
106
121
136
151
166
181
136
151
166
181
136
151
166
181
Days from 1 January 1996
Mean Daily Discharge (m 3/s)
(b)
Sarab Seidali Station at the Doab Aleshtar River (776 km 2)
250
200
150
100
50
0
1
16
31
46
61
76
91
106
121
Days from 1 January 1996
Mean Daily Discharge (m 3/s)
(c)
Khers Abad Station at the Abe Marg (1,460 km 2)
50
40
30
20
10
0
1
16
31
46
61
76
91
106
121
Days from 1 January 1996
Figure 36. Observed and simulated daily hydrographs for Cases I and II for a
selected period January to June 1996 at three stations: (a) Jelogir, (b) Sarab seidali,
and (c) Khers Abad.
118
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Furthermore, there were no striking differences in the performance changes in
Case II when compared with Case I with respect to the spatial location of these
gauges. Similarly, no major differences were observed in the daily and monthly NSE
and R2, indicating similar performances under both cases in the temporal respect as
well. This was indicated by the similar pattern of differences in the NSE and R2. The
similar performance in both cases at the larger spatial scales could be attributed to
the averaging-out effect of the precipitation with the increase of drainage area. This
point is further substantiated by Figure 35, indicating minimal changes in NSE and
R2, when comparing corresponding values in Case II and Case I, for the flow gauges
draining larger areas. The tertiary-level streams draining comparatively smaller areas
(e.g., less than 1,600 km2) appear to be more sensitive to the changes in the
precipitation input, as depicted by large changes in the studied performance
indicators shown in Figure 35.
Stations indicating better performance in Case II
Better simulations were achieved in Case II compared to Case I at eight out of 15
studied gauges (Table 23a and 23b), namely Aran, Firoz Abad, Kaka Raza, Sarab
Seidali, Cham Injeer, Afarineh, Noor Abad, and Doabe Merek. All these flow
gauges represent tertiary-level streams (about 600-1,600 km2). Furthermore, most
significant improvements were witnessed in NSE as compared to R2 and the annual
volume balance, though they also showed generally better results in Case II. It is
noteworthy that most of these streamflow gauges represent regions where rain gauge
density and spatial distribution of rain gauges were poor. These characteristics are
more noticeable for the catchments located in the southeast of the study region
(Figure 30 and Table 21). For example, there was only one rain gauge located within
the drainage area of Firoz Abad, Sarab Seidali, Cham Injeer, Afarineh and Noor
Abad. In all these cases, the rain gauge was located at the outlet of the catchment
area and therefore is less likely to accurately catch the spatio-temporal variability of
rainfall within the catchment area. The interpolated rainfall made use of data from
the available neighboring rain gauges and thus tends to improve the spatio-temporal
distribution of rainfall. This resulted in considerable improvements in the simulated
streamflows under alternative precipitation input scenario (Case II). In temporal
perspectives, most of these gauges showed better performance in Case II for all the
examined time scales (e.g., daily, monthly and annual).
The precipitation input used in Case II helped improve the model simulations in
a number of ways. For instance, in the case of Sarab Seidali, overestimation of flood
peaks were the main cause of low NSE in Case I. This point is illustrated in Figure
36b, indicating observed and simulated streamflows under Case I and Case II for a
(short) selected time period. The results show that the Case I simulated a flood peak
of 237 m3/s on April 14, 1996 (Julian day: 105) against the observed flow of 32 m3/s
for that day. Examination of precipitation events showed that higher precipitation
was observed for the gauge used for this catchment compared to neighboring
gauges. The used gauge in Case I recorded a precipitation value of 63 mm/d for
Impact of Areal Precipitation Input on Streamflow Simulations
119
____________________________________________________________________
April 14. But the neighboring gauges recorded comparatively lower precipitation
values, less than 40 mm/d for that day. This suggests that this precipitation event
triggered this very high peak flow. However, the large difference in observed and
simulated flows during this event could be attributed to very high rain in a small
localized area as recorded at the gauge used for Sarab Seidali. The records from
neighboring stations indicate that the used station precipitation did not represent the
whole catchment well. The areal average of precipitation used in Case II
considerably helped better simulate this peak flow, though it still did not perfectly
match the observed flow on that day. However, the generally better estimation of
flood peaks during the calibration and validation periods under Case II as compared
to Case I could be attributed as the main reason for the improved performance under
Case II for this catchment.
The other major reason for the better simulations in Case II could be the better
representation of the overall precipitation amounts besides good depiction of the
temporal dynamics of the individual precipitation events. For instance, in the case of
Kaka Raza the station nearest the centroid recorded consistently lower precipitation
volumes compared to the neighboring stations. The precipitation totals and the
overall rainfall dynamics were better represented after interpolation (Case II) that, in
turn, enhanced the accuracy of the simulated results. On the other hand, stations used
in Case I for the catchment gauged at Afarineh observed higher precipitation
compared to its neighbors causing poor simulated results. For this catchment, the
annual volume balance during calibration and validation periods under Case I was
67 and 106% when compared with the observed streamflows. The simulated values
in Case II considerably reduced this departure from the observed records and
resulted in a volume balance of -3% (calibration) and 33% (validation).
Furthermore, NSE values were also better in Case II than in Case I. The daily NSE
under Case I were -0.01 and -0.94 during calibration and validation periods,
respectively. On the other hand, the corresponding NSE results in Case II were 0.52
and 0.44, respectively.
Stations indicating good performance in Case I
No station performed considerably better in Case I than in Case II. However, there
were two catchments, Firoz Abad and Khers Abad, where simulated results were
generally poor in both cases, as indicated by inferior values of R2, NSE and volume
balance (Table 23a and 23b). In the case of Khers Abad, the percentage difference in
the mean annual runoff was considerably higher in both cases, i.e., 43% under Case I
and 45% under Case II during calibration and 73% under Case I and 75% under
Case II during validation. Under both cases, R2 and NSE ranged from 0.45 to 0.67
and -0.40 to 0.14, respectively. These points are substantiated by Figure 36c,
indicating comparison of the observed and simulated discharge under Case I and
Case II for the Khers Abad catchment. The poor performance could be due to a
number of reasons. The input precipitation may not be well represented in both
cases. However, considering comparatively higher number of stations located within
this catchment and close to it, precipitation may not be a major reason for lower
120
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
performance in this case. The observed mean annual water yield at Khers Abad
(about 30 mm/yr.) is very low compared to the neighboring areas. This was unusual
when considering largely similar climatic and physiographic patterns in this and
neighboring catchments. During field observations and discussion with the local
experts, it was noted that surface water-groundwater interactions were very complex
and poorly known in this region. The area is highly influenced by karst formations
and there are many springs in the neighboring regions. Therefore, the incoming
precipitation might be heavily recharged to the deep groundwater/karst formations,
which could become the source of spring flows outside of the topographic boundary
of this catchment. These aspects were considered very important for proper
modeling of this region and warrant in-depth scientific investigations. In the case of
Firoz Abad, streamflows were heavily influenced by the water withdrawals for
irrigation uses, and this was considered as the major factor for poor performance,
besides uncertainties in other input data including precipitation under both of the
tested scenarios. The poor performance of the hydrological models applied to the
heavily regulated catchments is generally in line with the literature (e.g., Faramarzi
et al. 2009).
6.4.
Concluding Remarks
This study compared the SWAT model performance under a) standard SWAT
precipitation input procedure (using records of the station nearest to the centroid of
a subcatchment - (Case I) and b) modified areal precipitation input obtained through
spatial interpolation (Case II). The model performance was assessed by using R2,
NSE at daily and monthly temporal resolutions and by comparing mean annual
runoff at 15 selected streamflow gauges located in the Karkheh Basin.
The results show that, in general, the model performance was almost similar in
both cases when evaluated in terms of R2. However, a notable improvement was
observed in the NSE criterion in Case II compared to Case I for eight out of 15
studied gauges (600 to 1,600 km2). For these catchments, the performance in terms
of R2 and annual volume balance in Case II was either comparable or better when
compared with Case I. Most of these catchments represent regions with limited
climatic data, i.e., either rain gauge density was comparatively low or the
distribution of the rain gauges within the catchment was poor (e.g., used gauge in
Case I was often located at the outlet of the catchment). The improvement in the
simulated streamflows in Case II was attributed to the improved representation of
the precipitation regime and its spatial variability.
However, the results from both cases were comparable, in terms of all the studied
performance measures, for the gauges located on the larger Karkheh River and its
major tributaries with drainage areas larger than 5,000 km2. Furthermore, for these
gauges, no significant differences could be identified in terms of their spatial
location (e.g., among the streams draining the upper, middle or lower parts of the
study area) or between the studied time scale (e.g., daily, monthly and annual). The
Impact of Areal Precipitation Input on Streamflow Simulations
121
____________________________________________________________________
similar performance at the large spatial extent under both precipitation input
procedures could be attributed to the averaging-out effect of the precipitation input
while simulating the hydrological processes.
It can be concluded that the use of areal precipitation, obtained through
interpolation of the available station data, improved the SWAT model simulated
streamflows in the study basin. The results were strongly influenced by the spatial
extent of the investigations as well as by the station density and spatial distribution
of the available rain gauge data used in the interpolation. Further testing of (semi-)
distributed hydrological models such as SWAT, using areal precipitation as an input
(e.g., obtained through interpolation of rain gauge records, radar data and satellite
observations), for its added value to the streamflow simulations and other processes
is highly recommended. Future investigations should focus on the spatio-temporal
aspects of the hydrological simulations, particularly in the large river basins, and
should also study the impact of calibration strategies, i.e., by using various objective
functions for the parameter optimization and performance evaluation, and by
following different calibration, validation and uncertainty analysis procedures.
Development of an optional component for the interpolation of climatic data within
the existing SWAT model will benefit multiple SWAT users.
Although this study focuses on improvement of precipitation input in the SWAT
model, the procedures and results are instructive for rainfall-runoff modeling in
general.
7.
QUANTIFYING SCALE-DEPENDENT IMPACTS OF
UPGRADING RAIN-FED AGRICULTURE9
7.1
Introduction
Improvements of rain-fed agriculture are required to ensure global food security.
Improved rain-fed agriculture also contributes to the global poverty reduction as the
majority of the world’s rural poor depend on rain-fed agriculture for livelihoods. It is
also beneficial for environment, e.g., to reduce soil erosion. Yet, a proper
understanding of trade-offs resulting from such interventions is essential too (e.g.,
CAWMA 2007; Rockström et al. 2010).
Wakindiki and Ben-Hur (2002) conducted a field-scale evaluation of the
indigenous soil and water conservation techniques in a semi-arid rain-fed region of
Kenya and concluded that the techniques they investigated, i.e., building trash lines
of various sizes and materials, significantly reduced soil erosion and improved crop
yields. The study also noted significant reduction in the surface runoff under the
studied techniques. Makurira et al. (2010) suggested that the food and livelihood
security of the farmers in semi-arid to arid regions could be significantly improved
by promoting rainwater harvesting. Their field scale experimentations conducted in
the Makanya Basin, Tanzania, demonstrated that the combined use of conservation
agriculture, diverting runoff onto field plots and enhancement of in-field soil
moisture through trenching and soil bunding (locally called fanya juus) could help in
managing erratic distribution and scarce quantity of rainfall. The study showed that
these methods could significantly increase plant transpiration resulting in higher
crop yields and water productivity. Oweis and Hachum (2009) reported examples of
successful implementation of various water harvesting techniques (e.g., contour
ridges, semi-circular and trapezoidal bunds, small runoff basins, terraces, wadi-bed
cultivation and tanks) from West Asia and North Africa region. They reported that
the widespread adoption of water harvesting and supplementary irrigation
techniques helped improve land cover growth and raise productivity levels, but
required careful evaluation of factors such as available technical skills at the local
level, characterization of climate, water and land use systems, prevailing
institutional and policy environment and possible conflicts in the water uses and
users among upstream-downstream areas.
9
This chapter is based on the paper Quantifying scale-dependent impacts of upgrading rainfed agriculture in a semi-arid basin by Masih, I.; Maskey, S.; Uhlenbrook, S.; Smakhtin, V.
2011. Agricultural Water Management (in review)
124
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Xiubin et al. (2003) compared the observed runoff and precipitation records for
two periods, representing hydrological conditions without implementation of soil
and water management interventions (1959-1969) and the period (1990-1995) with
the interventions in the three subbasins of the Yellow River Basin, China. They
noted a reduction of about 50% in the mean annual runoff, which was mainly
attributed to various interventions, such as building earth dams, planting trees or
grass, terraces, and irrigation projects. They highlighted that the benefits of
increased food production and reduced soil erosion realized from the
abovementioned interventions came at the cost of reduction in the downstream
flows. Lacombe et al. (2008) investigated the impact of water and soil conservation
works (WSCW), mainly contour ridges and hillside reservoirs, on runoff response of
the Merguellil Basin (1,183 km2) in Tunisia. The observed rainfall and runoff
records over 1981 to 2005 were used to investigate the changes in the runoff regime.
The study indicated runoff reduction of 28-32% due to WSCW at the basin scale.
They further noted that harvested soil moisture and stored water in the small dams
were not efficiently used for the benefit of increased crop production, and argued
that the adopted WSCW contributed to the loss of water through enhanced (nonbeneficial) ET in the region.
The brief review of the recent studies presented above shows the need for much
better understanding of the impact of upgrading rain-fed agriculture on hydrology
and water availability at subbasin to basin scale. The main objective of this chapter
is to investigate such impacts in the semi-arid Karkheh Basin, Iran. More
specifically, this study aims to a) investigate the potential for upgrading rain-fed
agriculture to irrigated agriculture and associated impacts on streamflow, b) evaluate
the impact of soil and water conservation on streamflow, and c) assess the predictive
uncertainty of the model used and its implications.
7.2.
Methodology
7.2.1. Model used for the scenario simulation
A semi-distributed process-based hydrological model Soil Water Assessment Tool
(SWAT) (Arnold et al. 1998; Gassman et al. 2007) was used to simulate various
scenarios (discussed below). The model covers an area of 42,620 km2 up to the
outlet of the study basin at Paye Pole (Figure 30). We adopted the model with areal
precipitation input obtained through interpolation, as it performs better compared to
the standard SWAT precipitation input procedure of using data of a rain gauge
located nearest the centroid of a subcatchment (see chapter 6). The modeling details
such as on the input data, parameterization, calibration and validation can be found
in chapter 6. An assessment of prediction uncertainty of the model was carried out in
this study, and is discussed below.
The uncertainty analysis was carried out using the Sequential Uncertainty Fitting
algorithm (SUFI-2) (Abbaspour et al. 2007) available in the SWAT-CUP software
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
125
____________________________________________________________________
(Abbaspour 2008). In this approach, all uncertainties, i.e., that of model, parameters
and input data, are mapped on to the model parameter ranges. In SUFI-2, the
prediction uncertainty of the model is evaluated using two measures: P-factor and Rfactor. The P-factor which may vary from 0 to 100% indicate the percentage of
observed data falling within the 95 percent prediction uncertainty (95PPU) band
calculated at 2.5 and 97.5 percentiles of the model output based on Latin hypercube
sampling. The R-factor is the average width of the 95PPU band divided by the
standard deviation of the observed data.
The results presented in Figure 37 show the simulated streamflows along with
the uncertainty band and the observed data. The results are shown here for six
selected stations averaged over the 13 years period from 1988 to 2000. These
stations were selected to keep consistency with the study objectives, with a focus on
basin to subbasin scale impacts. The Paye Pole station at the Karkheh River (outlet
of the study domain corresponding to river reach ID 71) reflects basin level
implications of tested scenarios. The changes in streamflows at this location are
pivotal to understand the water availability for the multipurpose Karhkeh Dam and
its downstream area. The other selected stations are important to reflect the spatial
variations within the basin, and represent the Karkheh River and its all major
subbasins. Locations of these gauges are marked in Figure 6, and some basic
features are given in Table 3.
The calibration and uncertainty results shown in Figure 37 reveal that most of the
observed streamflow data fall well within the model’s prediction uncertainty band,
with about 2 months falling slightly outside of the 95PPU band in most cases. The
P- and R- factors on daily values are also reasonably good (e.g., P-factor > 0.5, and
R-factor <0.5). Similarly, the Nash-Sutcliffe Efficiency (NSE) (Nash and Sutcliffe
1970) and co-efficient of determination (R2) on daily streamflows are also in good
range (NSE: 0.65-0.80; R2: 0.66-0.81).The percentage difference between observed
and simulated mean annual flows was also quite small in most cases ranging from 13 to 8%. The reported performance statistics (NSE, R2 and volume balance) were
estimated from the simulations of the baseline simulation against the observed flows,
which represented the final parameter set adopted in this study. The achieved
performance statistics compare very well with other SWAT applications in Iran
(Faramarzi et al. 2009) and elsewhere (e.g., Gassman et al. 2007).
126
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Pole Chehre
Base Scenario
125
Observed
100
Ghore Baghestan
100
Streamflow (m 3/s)
75
50
25
80
60
40
20
Sept.
Sept.
Sept.
July
Aug.
July
Aug.
July
Aug.
May
June
June
June
Apr.
Holilan
350
Streamflow (m 3/s)
150
100
50
300
250
200
150
100
50
Apr.
Month
Month
Jelogir
600
Mar.
Feb.
Jan.
Oct.
Sept.
July
Aug.
June
May
Apr.
Mar.
Feb.
Jan.
Dec.
Oct.
Nov.
Dec.
0
Nov.
Paye Pole
600
Streamflow (m 3/s)
500
400
300
200
100
0
500
400
300
200
100
Month
Apr.
Feb.
Jan.
Dec.
Nov.
Oct.
Sept.
July
Aug.
June
May
Apr.
Mar.
Feb.
Jan.
Dec.
Oct.
Nov.
0
Mar.
Streamflow (m 3/s)
May
Pole Dokhtar
0
Streamflow (m 3/s)
May
Month
Month
200
Mar.
Jan.
Feb.
Oct.
Sept.
July
Aug.
May
June
Apr.
Mar.
Jan.
Feb.
Dec.
Oct.
Nov.
Dec.
0
0
Nov.
Streamflow (m 3/s)
150
Month
Figure 37. Monthly summary of the calibration and uncertainty analysis results.
(The 95PPU band is shown by thin green bars)
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
127
____________________________________________________________________
7.2.2. Tested scenarios
Three scenarios related to increased water consumption in the rain-fed agriculture
were simulated and their impacts on streamflows evaluated against the baseline
simulation. The changes in the mean annual and mean monthly streamflows were
the main assessment indicators.
Scenario 1 (S1): Upgrading rain-fed areas to irrigated agriculture. In this scenario,
the impact of upgrading rain-fed farming to irrigated agriculture was investigated.
The rain-fed lands located in the valleys close to rivers with soils favorable for
agriculture (e.g., alluvial soils) were considered as potential areas for irrigation. The
GIS based overlay analysis conducted through SWAT interface was used to estimate
the potential areas. The analysis revealed that a total area of about 0.5 million ha
could be potentially upgraded from rain-fed to irrigated agriculture in the study basin
(Table 24). This accounts for about 11% of the total study area. It is recognized that
exact estimation of total irrigable area might vary, depending upon various
physiographic, chemical, hydrological, topographic, social and economic factors.
Moreover, we also consider that it will not be possible to convert all the rain-fed
systems to irrigated ones due to several reasons as aforementioned. We assume that
investments in developing surface water use through gravity based systems, lift
irrigation schemes, direct pumping from the rivers and building small tanks and
dams for small scale irrigation could contribute to the proposed land use shift
(upgradation of rain-fed systems to irrigated ones) in future. Impact of such a shift
on downstream water availability is not well known and will be investigated in this
scenario.
S1 was represented in SWAT by using its water use routine, which provides
options to specify average daily water consumption rates for each month. The water
use can be defined from every subbasin through four possible sources of water, i.e.,
rivers, shallow aquifers, deep aquifers and ponds. We specified water consumption
from the rivers only. The other options were not tested due to limitations mainly
related to data availability.
The average daily water consumption from each of the 71 river reaches was
defined based on the irrigation water demand from the potential rain-fed area
considered for upgradation to irrigation. The water consumption was estimated for
every month using the following equation.
IWC = Air (E pot − Eact )
(18)
where, IWC is the average monthly irrigation water consumption rate in m3/d, Air
is the area upgraded to irrigation in m2 and Epot and Eact are the average monthly
potential and actual evapotranspiration rates expressed in m/d. The Epot was
estimated using Hargreaves method (Hargreaves et al., 1985), whereas the actual
evapotranspiration (Eact) was estimated based on the available moisture in the soil
profile and evapotranspiration demand from rain-fed systems under wheat
128
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
cultivation, as per SWAT water balance calculations under the baseline scenario.
The Air was estimated by using information on land use, soil and slope data as
mentioned above. The GIS based overlay analysis conducted through SWAT
interface was used to estimate Air. The analysis revealed that a total area of about
5,000 km2 (0.5 million ha) could be potentially upgraded from rain-fed to irrigated
agriculture in the study area (Table 24).
The monthly estimates of Epot, Eact, and difference between them used in the
formulation of S1 are presented in Figure 38. On the whole, the required IWC was
estimated around 275 mm for the period November to June, which generally reflects
the growth season of winter wheat. Wheat is the main crop cultivated in the study
area. The other crops include barley, chickpea, alfa alfa, maize, sugarbeat and
vegetables. Wheat is cultivated from November to June. Maize and sugerbeat are
cultivated during June to October, and are mainly grown in regions where irrigation
supplies are available. Fodder (alfa alfa) and vegetables are grown in both cold and
warm seasons. The calculations of irrigation requirements were constrained to a
single cropping season reflecting wheat growth period (winter-spring). A double
cropping system was not considered mainly due to the limitations of surface water
supplies during summer season (Masih et al. 2009).
Table 24. The extent of area under drainage, rain-fed systems and irrigable rain-fed
systems at the basin and sub-basin levels.
River
reach/
catchment
ID
Name
river
20
24
38
60
66
71
Gamasiab
Qarasou
Saymareh
Kashkan
Karkheh
Karkheh
of
Name
of
streamflow gauge
Drainage
area
(DA)
(million
ha)
Pole Chehre
Ghore Baghestan
Holilan
Pole Dokhtar
Jelogir
Paye Pole
1.078
0.544
2.042
0.952
3.952
4.237
Area
under
rain-fed
systems (million
ha)
million % of
ha
DA
22
0.240
61
0.333
36
0.736
30
0.282
35
1.399
33
1.402
Rain-fed
area
convertible
to irrigated area,
(million ha)
million
% of DA
ha
8
0.086
23
0.124
12
0.253
10
0.099
12
0.468
11
0.468
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
129
____________________________________________________________________
Evapotranspiration demand and
consumption (mm/month)
300
Epot
250
Eact
Epot - Eact
200
150
100
50
Sept.
Aug.
July
June
May
Apr.
Mar.
Feb.
Jan.
Dec.
Nov.
Oct.
0
Month
Figure 38. Values used in the development of S1 for the monthly potential
evapotranspiration (Epot), actual evapotranspiration (Eact), and the difference
between Epot and Eact .
Scenario 2 (S2): Improved soil water availability through rainwater harvesting.
Various in-situ water harvesting systems and soil and water conservation techniques
are generally recommended as discussed in the introductory section of this chapter
(e.g., micro-basins, terracing, bunds, and mulching etc.) to increase soil water
retention and foster plant water availability. Such interventions can be represented in
the SWAT model by modifying the Available Soil Water Capacity (AWC)
parameter. The AWC parameter controls retention of water in the soil profile for
consumption by plants. An increase in AWC generally leads to an increased soil
water retention and thereby indirectly represents a soil and water conservation
practice. Under S2, we assume that the recommended soil and water management
interventions collectively increase AWC of the soils under rain-fed agriculture by
20%. A similar study was reported by Faramarzi et al. (2010), who investigated the
impact of 20% increase in AWC on the irrigation requirement in Iran. However,
their study did not evaluate the impacts on streamflows, and present study helps
filling this important knowledge gap.
Scenario 3 (S3): Combined impact of S1 and S2. Under this scenario, the combined
impact of the two scenarios (S1 and S2) was evaluated. S3 was represented in the
SWAT model by keeping the water consumptions in the water use routine same as in
the case of S1. Then the soil routine was modified by increasing the AWC parameter
by 20% as done under S2. In this way, S3 simulated the combined effect of
scenarios S1 and S2.
130
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
7.3.
Results and Discussion
7.3.1. Downstream impact of upgrading rain-fed areas to irrigated agriculture
(S1)
The impact of the tested scenarios on mean monthly and annual flows at the selected
locations across the basin is presented in Table 25 as percentage difference in the
streamflows compared to the baseline period. The mean monthly flows under
baseline and three tested scenarios including the uncertainties of the predictions
(95PPU bands) are shown in Figure 39, indicating basin level impacts noted at Paye
Pole station at the Karkheh River. The results at the basin scale show about 10%
reduction in the mean annual flow, which is about 17 m3/s or 537 × 106 m3/yr.. The
range of inter-annual reduction in the mean annual flow is 7-15%. The variation in
the reduction in the monthly flows is very high. Month June appears to be the most
affected with 56% reduction on average compared to the baseline period, whereas
the reduction in the October flow is negligible (just about 1%). In some dry years,
the reduction in the flow in June is as high as 65%.
The impacts vary notably among study subbasins (Table 25). These differences
were mainly governed by the relative area brought under S1 and the amount of
available flows. The highest impact was noted for Ghore Baghestan (Qarasou
subbasin) where the highest proportion of area under S1 falls (Table 24). This
subbasin indicated a decline of 15% in the mean annual flows. Monthly flow
reductions were in the range of 0-92%. The inter-annual variation is also quite high,
with annual flow reduction in the range of 10-43%. Monthly flow reductions could
escalate further, reaching zero flow in June for some dry years. The Kashkan
subbasin witnessed comparatively lower impact, where annual reductions were
around 8%, varying from 0-52% between months, as shown by the estimates at Pole
Dokhtar. The inter-annual variability was also modest here, with the annual flow
reduction in the range of 6-11% and maximum monthly decline of around 66%.
In general, the highest flow reduction corresponds to June for all examined
locations. The other months with high impacts are May, July, November and
December. This pattern of monthly impact is somewhat similar across the basin. The
considerable flow reductions were observed in July despite no water abstractions
from streams during this month. This shows that a reduction in the streamflow in a
month is likely to contribute into diminishing streamflow in the following month(s).
This carryover impact is due to hydrological processes related to water storage in the
river channel and subsequent contribution of stored water to the river flows.
However, this impact was not prominent beyond July, as noted by the negligible
change in August, September and October. Part of the irrigation abstractions
generally contributes back to the streamflows in the form of return flows. A detailed
investigation of the return flow processes was beyond the scope of this study, mainly
due to the limitations related to the model and the available data.
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
131
____________________________________________________________________
Table 25. Difference in the mean monthly and annual streamflows, expressed in %,
under the three tested scenarios as compared to baseline simulation.
Scenario/
Time level
S1
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Annual
S2
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Annual
S3
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Annual
Pole
Chehre
(Reach ID
20)
Ghore
Baghestan
(Reach ID
24)
Holilan
(Reach 38)
Pole
Dokhtar
(Reach ID
60)
Jelogir
(Reach ID
66)
Paye
Pole
(Reach
ID 71)
0
20
9
5
3
1
1
21
89
21
0
0
10
0
23
15
8
5
2
3
31
92
22
0
0
15
0
16
14
7
4
2
2
23
77
37
2
0
12
0
14
6
3
2
1
1
13
53
10
0
0
8
0
12
12
6
3
2
2
18
61
30
3
1
10
1
10
12
6
3
2
2
17
56
38
4
2
10
2
3
3
4
6
4
2
2
1
1
1
1
3
3
5
6
8
9
6
4
3
3
3
3
3
5
3
4
4
6
7
5
3
2
3
3
2
2
4
1
2
3
4
3
3
2
2
2
2
1
1
2
2
3
4
5
6
4
3
3
3
3
3
2
4
2
3
4
5
5
4
3
3
3
3
3
2
4
2
23
12
10
9
5
3
23
89
22
1
1
13
3
27
21
15
14
9
7
33
93
25
3
3
20
3
20
17
12
11
7
5
25
77
39
4
2
16
1
16
9
7
5
4
3
15
54
11
2
1
10
2
15
15
11
8
6
5
20
61
32
6
3
14
3
13
15
11
8
6
5
19
56
40
7
4
14
132
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Paye Pole
600
BS
3
Streamflow ( m /s)
500
S1
S2
S3
400
300
200
100
Sept.
Aug.
July
June
May
Apr.
Mar.
Feb.
Jan.
Dec.
Nov.
Oct.
0
Month
Figure 39.
Simulated streamflows for the baseline period and the three scenarios
(S1, S2 and S3) at the basin level (Paye Pole station). Also shown in the figure are
95PPU uncertainty bands.
The results of S1 are instructive to guide the desired level of irrigation
development by examining the corresponding spatio-temporal impacts. The
expected reductions in the mean annual flows of 10% at the basin level and 8-15%
across the subbasins seem quite reasonable. For instance, this would translate into an
annual flow reduction of about 537 × 106 m3/yr. at the basin level. The water
development potential of the basin was estimated to around 1-4 × 109 m3/yr.,
considering different levels of water allocations for the environment (Masih et al.,
2009). This shows that the annual flow reduction at the basin level as a result of the
S1 is well within the available water development potential of the basin.
However, the major concern in adopting S1 is related to excessively high
percentage of flow reductions from May to July, most notably in June when
reductions could be in the range of 50-100%. In general, the reductions exceeding
50% may severely impact downstream water needs. For instance, it would largely
alter the natural flow regime of the river and is likely to have severe negative
repercussions for the environment (e.g., Poff et al. 1997).
Thus, adoption of S1 would require additional considerations to avoid excessive
decline in flows and consequent impacts on downstream uses and users. This could
be achieved by reducing the abstractions, which could be done through decreasing
the rain-fed area considered for upgradation. This option was further studied, and the
results are substantiated by Figure 40, which shows the basin level impact on
streamflows associated with a certain level of rain-fed area upgraded to irrigation for
critical months of May, June and July. Such analysis could guide the choice of
appropriate level of rain-fed agricultural development. In general, the results show
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
133
____________________________________________________________________
that development of about one-fourth of the available potential rain-fed area to
irrigation (e.g., about 0.1 million ha or 1,000 km2) may be considered safe as it will
keep the flow reduction in the most affected month June to below 30% ensuring
reasonable levels of downstream water availability throughout the year.
The other complementary options for mitigating flow reductions may include: a)
various forms of water storage to augment supplies during the most affected months
(May and June), and b) practicing supplementary irrigation. The studies conducted
by van der Zaag and Gupta (2008), and McCartney and Smakthin (2010) discussed a
number of water storage options and pointed out that such options are also likely to
address the issues of high variability in water availability due to observed and
predicted climate variability and change. A considerable benefit of supplementary
irrigation in terms of improving rain-fed agriculture in semi-arid to arid
environments was shown by Oweis and Hachum (2009).
Paye Pole
100
May
June
July
Flow reduction (%)
80
60
40
20
0
0
20
40
60
80
100
Rain-fed area converted to irrigation (%)
Figure 40. Impact on monthly streamflows (May-July) due to proportion of area
upgraded from rain-fed to irrigated agriculture at the basin level.
7.3.2. Downstream impact of improved soil water availability through rainwater
harvesting Scenario (S2)
Under S2, a reduction of about 4% or 6 m3/s (194 × 106 m3/yr.) was observed in the
mean annual streamflow at the basin level (Table 25 and Figure 39). The impact at
monthly time resolution was also small (2 to 5% decline) at the basin level. At the
subbasin level, the annual flow reduction was in range of only 2 to 5%. Monthly
flow reductions vary from 1-9 % across the basin. However, reductions in FebruaryApril were slightly higher in case of S2 than S1. This difference could be attributed
134
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
to change in precipitation partitioning processes under S2, whereby, enhanced water
retention in the soil and increased evapotranspiration diminish surface and
subsurface runoff. This affect appeared to have more impact on streamflows than
that of water diverted from rivers for irrigation during these months. The interannual pattern of changes indicated a decline in the range of 2-10 % and 0-20 % at
annual and monthly time scales, respectively. The results suggested that the
expected reductions in streamflows are reasonably small compared to the amount of
water available for further development (as noted in the previous section).
Therefore, it could be recommended to increase attention for the promotion of
improved soil and water conservation practices in the basin.
7.3.3. Combined impact of S1 and S2 (S3)
The results of S3 (Table 25 and Figure 39) reflected the combined impact of S1 and
S2, although dominated by S1. The results showed a decline of about 14% or 23
m3/s (718 × 106 m3/yr.) in the mean annual streamflow at the basin level. Flow
reductions at the subbasin level were in the range of 10-20%. Similar to findings
noted for S1 and S2, the highest impact was observed for two upper subbasins, most
notably the Qarasou subbasin, whereas, comparatively lower impact was found in
the Kashkan subbasin in the middle parts. The observation of the annual flow
reductions under S3 shows that the scenario is still well within the available water
resources development potential in the basin. However, similar to the findings under
S1, the impact could be very severe from May to July, as highlighted by Table 25.
This substantiates the need for additional attention, e.g., considering reduced area
under S1, practicing supplementary irrigation, and developing storage options that
can tap water during high flow months (e.g., February-April) to provide additional
supplies in May and June.
7.3.4. Consideration of prediction uncertainty of the model
It is well recognized in hydrology that there are uncertainties involved in
hydrological modeling arising from limitations of model structure, input data and
parameterization (e.g., Beven 2001). These uncertainties not only influence the
model calibration process but could also impact the predictions made by the model.
For this study, a reasonable effort was designated to access the model uncertainty as
per used procedure of SUFI-2. The reasonably good values of the uncertainty
descriptors (e.g., P- factor>0.5 and R-factor<0.5) were obtained beside good results
on commonly used performance measures in hydrological modeling (e.g., NSE:
0.65-0.80, R2 : 0.66-0.81 and volume balance ranging from -13 to 8%). Despite
reasonably good values of the reported performance measures, the range of 95PPU
band obtained from the final 500 parameter sets is still not small. Therefore, the
choice of the parameter set used for simulating the baseline case may have
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
135
____________________________________________________________________
implications for the studied impact scenarios. In this study, we adopted most
commonly used approach of selecting a parameter set for baseline that produces
simulations close to observed streamflows and results in reasonably good values of
studied performance indicators (NSE, R2 and volume balance). Therefore, the impact
results of tested scenario reported in the previous sections were considered
appropriate to guide decision-making process. However, the outcome should be
considered cautiously, i.e., particularly considering the associated uncertainty in the
results.
Furthermore, to get an overview of the prediction uncertainty, the range of
prediction uncertainty associated with the simulated scenarios was estimated by
comparing the values of upper and lower 95PPU band achieved under baseline case
with the corresponding values of the tested scenarios. Figure 41 shows the
percentage reduction in the mean annual streamflows along with the expected
uncertainty. The results for S1 showed that the reduction of 10% in the mean annual
streamflow at the basin level could have uncertainty range of 8-16%. Monthly
results are shown in Figure 42, which, for instance, indicated that the uncertainty
range of 50-77% could be associated with the values of flow reduction of 56% in
June at the basin level. Furthermore, uncertainties of the different scenarios (e.g., S1
vs S2) and that of their flow reductions are markedly different. The uncertainty of S1
is dominating S3 uncertainty as well.
The major implications of the inclusion of prediction uncertainty results in
impact evaluation and consequently consideration in the decision-making process
could be: a) The range of flow reduction at annual scale including the uncertainty
band still remains well within the available development potential in the study basin,
e.g., 8-16% reduction at the annual scale at the basin level under S1., b) The range of
monthly flow reductions including uncertainty clearly reveal much higher likelihood
of flow reduction, e.g., the expected flow reduction in June could be in the range of
50-77% at the basin scale under S1. Therefore, considering the high range of
uncertainty in the predicted impacts on monthly flows under S1, it could be stressed
that the decisions regarding rain-fed area development would require additional
considerations so that streamflows may not be excessively depleted during critical
months (e.g., May-July)., and c) The impact of including prediction uncertainty in
the case of S2 is small, which further strengthen the argument about promoting soil
and water conservation techniques in the study basin.
136
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Reduction in streamflow (%)
40
S1
S2
S3
30
20
10
Paye Pole
Jelogir
Pole Dokhtar
Holilan
Pole Chehre
Ghore
Baghestan
0
Name of the flow gauge
Figure 41. Assessment of the prediction uncertainty of the modeling results at
annual time resolution under the three tested scenarios.
Paye Pole
Reduction in streamflow (%)
100
80
60
40
20
Sept.
Aug.
July
June
May
Apr.
Mar.
Feb.
Jan.
Dec.
Nov.
Oct.
0
Month
Figure 42. Assessment of the prediction uncertainty of the modeling results at
monthly time resolution at the basin level.
Quantifying Scale-Dependent Impacts of Upgrading Rain-Fed Agriculture
137
____________________________________________________________________
7.4.
Summary and Concluding Remarks
The upgradation of the rain-fed systems through improving soil and water
conservation practices and providing water for irrigation is of critical importance for
the global food security, particularly for the rural poor living in water-scarce semiarid to arid regions. However, a proper understanding of the benefits harnessed by
developing water in upstream areas and the consequent impacts on downstream
regions are important for informed planning and sustainable management of natural
resources. This study contributes to such understanding by evaluating the impact of
three scenarios of upgrading rain-fed areas using SWAT model. The tested scenarios
were: upgrading rain-fed areas to irrigated agriculture (S1), improving soil water
availability through rainwater harvesting (S2), and a combination of S1 and S2 (S3).
The impacts on monthly and annual streamflows were investigated.
The basin scale impact of the tested scenarios suggested a decline in mean annual
flows of about 17 m3/s (537 × 106 m3/yr.), 6 m3/s (194 × 106 m3/yr.) and m3/s 23
(718 × 106 m3/yr.) under scenarios S1, S2 and S3, respectively, when compared with
the baseline case. This would mean a reduction of about 10%, 4% and 14% in case
of S1, S2 and S3, respectively. The results revealed that the conversion of rain-fed
areas to irrigation (S1) would have comparatively higher reductions in the
downstream flows as compared to conserving water for plant uptake in the soil root
zone through rain water harvesting (S2). In general, the estimated reductions in the
mean annual streamflows at the basin scale fall well within the limits of available
estimates of water resources development potential. However, under S1, monthly
flows would be severely reduced from May to July, with highest impact in June
when flows could reduce more than half of the available flows at the basin scale.
This situation is much more alarming for two upstream subbasins (Gamasiab and
Qaraou) where reduction in June flows could reach over 90 %. The noted high levels
of impact on streamflows suggested the need of additional measures to avoid
excessive proportions of flow reduction in these months. The excessive impacts
could be minimized by reducing the area brought under irrigation (particularly in
upper parts of the basin), developing a range of storage options to augment supplies
and supplying less water than actually required (practicing supplementary
irrigation).
The consideration of model prediction uncertainty reveal that the range of annual
flow reduction is not large under all the tested scenarios (e.g., only 8-16% at the
basin scale under S1) and, thus, model uncertainty is less likely to have any major
implications in decision-making process in the context of annual flows. However,
the range of monthly flow reductions was quite large when considering model
prediction uncertainty, particularly for May-July, which further substantiate the need
of adopting policy options to mitigate excessive flow reductions during these
months.
Based on the results of this study, it could be recommended that the upgrading
rain-fed areas should be concentrated in the middle parts of the basin (e.g., Kashkan
River subbasin), through introducing irrigation and range of water storage options as
well as promoting soil and water conservation techniques. The water management
138
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
interventions in the upper subbasin areas of Gamasiab and Qarasou Rivers should
mainly limit to promoting soil and water conservation techniques. In these areas, the
conversion of potential rain-fed lands to irrigated ones should be promoted very
cautiously with much attention given to additional considerations of developing
fewer amounts of rain-fed areas, providing means of water storage to augment
supplies and practicing supplementary irrigation.
In general, the methods used and the findings of this study are instructive for the
other basins, and have demonstrated the importance of a rigorous spatio-temporal
investigation of impacts of agricultural water management interventions across a
large river basin.
8.
SYNTHESIS, CONCLUSIONS AND
RECOMMENDATIONS
8.1.
Nature and Causes of a High Level of Hydrological Variability
A comprehensive analysis of daily streamflow records available over the period
1961-2001 at the seven very important flow gauging stations across the Karkheh
River and its major tributaries revealed that the streamflows have large intra-annual
and inter-annual variability across the basin. The high flows are observed from
November to May, with peak flows occurring in March-April. The high flood events
(1-day maximum) can occur anytime from November to April, though most often
they occur in February and March. The low flow period corresponds to June through
October. There are large differences between the amount of water available during
high and low flow periods. For example, at the Jelogir station at the Karkheh River,
mean monthly streamflow in April (386 m3/s) is nearly ten times higher than in
September (41 m3/s). The observed spatio-temporal variations could be substantiated
by the values of Coefficient of Variation (CV). Monthly CV values range from 0.4
to 1.77 across the examined streamflow gauges. In temporal terms, the minimum
and maximum CV values correspond to February and November, respectively,
whereas, in spatial terms, the Gamasiab River indicated higher variability and the
Kashkan River the lowest variability. The mean annual streamflow indicated CV
values in the range of 0.41 to 0.54, indicating marked differences in water available
in the long-term perspective. For example, the mean and median surface water
availability at the Paye Pole station at the Karkheh River was estimated as 5.83 × 109
m3/yr. and 5.59 × 109 m3/yr. As in all other examined stations, the minimum and
maximum had a wide range at Paye Pole, with values of 1.916 × 109 m3/yr. observed
during 1999-2000 and 12.60 × 109 m3/yr. observed during 1968-69. Under such
highly variable conditions, the understanding of the reliability of the water
availability becomes more meaningful for better resources use and allocation
planning. The flow duration analysis conducted in this study provides such estimates
of streamflow reliability for the Karkheh Basin at daily, monthly and annual time
resolutions. For instance, the value of mean annual streamflow with a reliability
level of 75% (indicated by 75th percentile of streamflow derived from the flow
duration analysis) at Paye Pole was 4.10 × 109 m3/yr., which is about 30% lower
than the mean annual flow estimated for this location.
High climate variability is considered as the major driving factor of the observed
spatio-temporal variability of the streamflows, among other factors such as soil, land
use and geological characteristics. The streamflow regime also depicts notable
differences with regard to spatial location in the study basin. For instance, more
runoff is generated from the middle parts of the basin (e.g., Kashkan River
140
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
catchment) compared to the upper parts of the basin (e.g., Gamasiab and Qarasou
River catchments). These differences were mainly attributed to higher precipitation
rates and lower water use by the agriculture sector in the catchment areas of the
Kashkan River compared to the two upper catchments. Furthermore, the contribution
of base flow in the total streamflow is higher for the Kashkan River compared to the
other two upper catchments (Gamasiab and Qarasou), which could be due to the
differences in land use, soil and geological characteristics, and therefore warrant
further research.
8.2.
Water Allocations, Water Availability and Sustainability
The study reveals that the Karkheh Basin still appears to be an open basin, indicating
some room for further water resources development. The estimated range of further
water resources allocations was 1-4 × 109 m3/yr., depending on the amount of water
left for environmental flows. However, the water allocations should be done after a
careful impact assessment and trade-off analysis for multiple and highly competing
uses and users across the basin, which warrant a proper impact assessment before
implementation. The review of ongoing water allocation planning shows that the
allocation to different sectors of water use will be 8.90 × 109 m3/yr. by the year
2025, among which the irrigation share will be the biggest (7.42 × 109 m3/yr.),
almost equaling the renewable water supplies available in an average year.
Therefore, considering the water availability and its variability and water resources
development plans, it is evident that the basin will very likely approach closure stage
during the first quarter of this century. Meeting the demands of all users (particularly
agriculture, hydropower and environment) will then be an extremely challenging
task, particularly during dry years.
8.3.
Streamflow Trends and Their Underlying Causes
The investigation of trends in the hydro-climatic variables revealed a number of
significant trends, both increasing and decreasing. The observed changes were
nonuniform in term of their spatio-temporal prevalence. The upper parts of the basin,
particularly Qarasou River, faced a notable decline in the low flow regime. A
significant decline in the streamflow was observed in most of the studied low flow
indicators, i.e., May, August, 1 and 7 days minima, low pulse count and duration for
Ghore Baghestan station at the Qarasou River. On the other hand, the flood regime
and winter flows indicated intensification in the middle parts of the basin, indicated
by significant upward trends observed in 1 and 7 days maxima, high pulse count,
October and December flows for Pole Dokhtar station at the Kashkan River.
Further, downstream propagation of the observed trends was found dependent on the
combined effect of the upstream drainage areas. For instance, the declining pattern
during low flow months was not significant for the Karkheh River, because the
Synthesis, Conclusions and Recommendations
141
____________________________________________________________________
declining trends that emerged from the upper catchments (Gamasiab and Qarasou)
were counterweighted by the stable low flow behavior of the middle catchments
(e.g., Kashkan). However, the significant trends observed in a number of streamflow
variables at Jelogir, 1-day maximum, December flow, and low pulse count and
durations indicated changing hydrological regime of the Karkheh River. Most of the
observed trends were found triggered by the changing climatic behavior, observed at
a number of studied synoptic climatic stations. The study suggests that the decline in
April and May precipitation caused the decline in the low flows while the increase in
winter (particularly March) precipitation coupled with temperature changes led to an
increase in the flood regime. The catchment degradation could be a complementary
factor in the intensification of the flood regime, while increased water abstractions
might have additionally contributed in the declining low flow regime.
8.4. Addressing Methodological and Data Scarcity Issues in the Hydrological
Modeling
The use of hydrological models is generally seen pivotal in better understanding the
hydrological processes and has become a norm to test various “what if” scenarios,
which otherwise could not be well investigated on the basis of observed data alone.
This study reveals that data scarcity remains a major challenge in the basin-wide
hydrological modeling and water resources assessment, but could be addressed
through developing innovative and tailor-made methodologies and innovative
solutions for a study basin.
The Karkheh Basin noted abandoning of various flow gauging stations during the
course of time. Estimation of streamflow records for these poorly gauged catchments
emerged as an important issue in the study area, and is also generally seen as a major
challenge in hydrology. A new regionalization method was developed in this study.
The proposed method is based on the regionalization of a conceptual rainfall-runoff
model (the HBV model) parameters whereby model parameters could be transferred
from gauged catchment to the poorly gauged catchment depicting hydrological
similarity defined, based on the similarity in their flow duration curves (FDC). It
was demonstrated that the FDC-based regionalization method worked well in the
data limited Karkheh Basin as compared to other widely recommended methods
(e.g., spatial proximity and catchment similarity defined from the physiographic and
climatic characteristics). Moreover, the new FDC-based regionalization method is
regarded as a good addition to the available regionalization method, as it compared
very well with most of the available methods tested in other countries.
Better representation of precipitation data in the hydrological modeling also
emerged as an important consideration in the hydrological modeling of the Karkheh
Basin, besides its general recognition and research needs in hydrological modeling.
It is well recognized that the climatic data are the major driver of the hydrological
and other processes simulated by a model. In this regard, the benefit of using areal
precipitation derived from the observed station records by using inverse distance and
elevation weighting method was evaluated against the usual way of using station
142
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
data in the semi-distributed SWAT model. The study showed that the SWAT
streamflow simulations improved for smaller catchments (600-1,600 km2), most of
them having poor density and distribution of rain gauges, whereas larger catchments
(>5,000 km2) could be modeled equally well under both cases, mainly attributed to
the averaging-out effect of precipitation at larger catchment-to-basin scales. The
study demonstrated that the use of areal precipitation improved simulated
streamflows; however, the results were influenced by the spatial scale of the
investigation and distribution and density of rain gauges.
The gained understanding of the hydro-climatic variables and processes, i.e.,
through system investigation, innovative ways of improving available but scarce
streamflow and precipitation data, field visits and discussions with the stakeholders,
and a literature review, appeared to be instrumental in the good calibration of the
SWAT model for the upper mountainous parts of the Karkheh Basin (42,620 km2)
from where almost all of the basin’s runoff is generated. Moreover, better
understanding of the observed processes and improved quality of the input data
together with rigorous exercise of parameter estimation (based on both manual and
automatic procedures) and uncertainty analysis helped reduce the prediction
uncertainty of the used model, thereby providing a reasonably good confidence in
the modeled hydrological processes and investigation of various “what if” scenarios.
8.5. Consideration of the Impacts on Downstream Water Availability while
Upgrading Rain-fed Agriculture
The well-calibrated SWAT model was used to test the impacts of soil and water
management interventions in the rain-fed systems in the upstream areas on the
streamflows in the downstream areas. The impacts of upgrading rain-fed systems to
irrigated agriculture (S1), soil and water conservation practices (S2), and a
combination of S1 and S2 (S3) were studied on the mean annual and monthly
streamflows at catchment to basin levels. The results reveal that the expected
reduction in the mean annual flows (e.g., about 10% under S1, 4% in S2 and 14% in
S3) remains well within the available development potential in the basin, even after
consideration of the model’s prediction uncertainty. However, excessive decline in
flows during May-July (in particular during June) warrants additional measures to
ensure downstream water availability and environmental integrity throughout the
year. These excessive impacts could be minimized by reducing the area brought
under irrigation (particularly in upper parts of the basin), supplying less water than
actually required (practicing supplementary irrigation), and developing a range of
storage options to augment supplies.
8.6.
Contribution and Innovative Aspects of This Research
This PhD research addresses some key issues related to basin wide spatio-temporal
assessment of hydrology and water resources and highlights its importance and use
Synthesis, Conclusions and Recommendations
143
____________________________________________________________________
in water resources planning and management in the river-basin context. An attempt
has been made to specifically address various issues. The relevance of the study
objectives and research questions to the international literature, specific contribution
to the scientific debate, importance with respect to the case study basin, and
innovative aspects are highlighted in each of the chapter on results and discussions
(e.g., Chapter 3-7). The main innovative aspects and contributions of this research
include, but are not limited to: (i) development of a new FDC based regionalization
approach, (ii) innovative use of areal precipitation input and evaluation of its
implications on streamflow simulations in a large basin using the SWAT model, (iii)
contribution to improve understanding of the streamflow trends and their linkages
with climate, and (iv) improved knowledge on spatio-temporal variability of
hydrology and water resources, through use of rigorous state-of-the-art methods,
including development and application of new innovative techniques.
8.7.
Major Recommendations and Future Directions
Based on the study findings, the following major policy actions are recommended.
The ongoing water allocation planning is not sustainable and its thorough revision is
recommended. The sectoral water allocation needs to be revised in the light of
resource availability and variability, a sound foundation of which has been laid in
this study. In view of the high share of water allocation for human demands
(particularly agriculture), the environment is likely to suffer the most in the near
future. Therefore, further assessments of the environmental water needs for instream, floodplain and Hoor-Al-Azim Swamp are highly recommended. Although a
detailed assessment of environmental flow requirements was beyond the scope of
this study, the preliminary estimates could be based from the hydrological
assessments carried out in this study (see chapters 3 and 4), before more detailed
information on environmental needs and those of other sectors become available.
The changing climate and hydrological regime in the basin further added to the
complexity of hydrological and water management issues and require immediate
attention. Considering the nonuniform nature of the observed trends, the adaptation
response should be underpinned by concurrent nonuniform but basin-wide
approaches that include a sound understanding of spatio-temporal differences in the
observed trends as well as their interactions. For instance, the declining low flows
(e.g., May through September) in the upper parts of the basin (Qarasou and
Gamasiab subbasins) could be tackled through various strategies, i.e., introducing
restriction on the use of surface water during low flow months in these areas.
Moreover, the mitigation of an intensified flood regime, particularly in the middle
parts of the basin (e.g., Kashkan subbasin) should receive high priority in the short
term, to avoid negative repercussions on life, infrastructure and socioeconomics in
the Karkheh Basin. This should also remain a major policy focus in the long-term
strategy, as the predicted climate change is expected to increase frequency and
magnitude of floods in the study region (e.g., Abbaspour et al. 2009). Efforts should
also be enhanced to reduce the degradation of land cover (particularly forest and
144
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
rangelands in the middle parts of the basin), as the improved land cover is likely to
help stabilize the runoff response.
In situ soil and water conservation techniques should be promoted across the
rain-fed systems, as they are likely to pose a small impact on the downstream water
availability. However, the upgradation of rain-fed systems to full-scale irrigated
agriculture should be carried out partly and very cautiously to avoid jeopardizing the
downstream demands from the environment, hydropower, irrigation and other uses
during low flow months in particular.
Although the study has demonstrated that various innovative solutions could help
cover the data gaps, more studies and investments should be made on data collection
and better use of available (scarce) data sets. The abandoning hydro-climatic
monitoring network across the Karkheh River system should be overhauled.
Recently, more climatic stations have been installed in small cities, but the coverage
generally remains poor for the mountainous parts, undermining proper hydrological
investigations.
The SWAT model application demonstrated in this study should be extended,
i.e., by including sedimentation and water-quality processes, and by testing other
“what if” scenarios (e.g., related to storage options, land use changes, and
environmental flows).
In general, the study provided a scientifically important and practically relevant
example of hydrological assessment and its use in the water resources planning and
management in the river basin context, which is instructive for the Karkheh and
other river basins of Iran, and worldwide.
SAMENVATTING10
De escalerende toename van watergebruik voor menselijke doeleinden, met name
voor landbouw, leidt tot toenemende druk op de zoetwatervoorraden. Alhoewel de
toe-eigening van het water de mensheid op vele manieren heeft geholpen zoals het
verbeteren van de voedselvoorziening en het social-economisch welzijn, heeft het
ook tot schade geleid aan het milieu en de daaraan gerelateerde voorzieningen. Het
evenwicht tussen mens en natuur met betrekking tot watergebruik wordt gezien als
de grootste uitdaging van deze eeuw. Dit is nog veel ingewikkelder voor de semiaride tot aride gebieden in de wereld, zoals de Islamitische Republiek Iran, waar
water over het algemeen schaars is en de vraag vanuit de landbouw, industrie,
verstedelijking en de groeiende bevolking snel toenemen. Door de aanwezige
variatie in klimaat en de verwachte klimaatveranderingen zullen de problemen alleen
nog maar toenemen.
In het geval van waterschaarste en concurrerend watergebruik, is een betere
kennis van de stroomgebiedshydrologie en waterbeschikbaarheid essentieel voor
beleidsvorming en duurzame ontwikkeling binnen de watersector. Deze studie is
uitgevoerd in het semi-aride tot aride Karkheh stroomgebied in Iran, waar
grootschalige waterverdeelplannen voor handen zijn, maar waar een uitgebreide
kennis van de stroomgebiedshydrologie en het effect van de waterbeheersplannen op
het watergebruik en de watergebruikers in het stroomgebied ontbreekt. De
belangrijkste doelstelling van dit onderzoek is om een hydrologische studie van de
(oppervlakte) waterbeschikbaarheid van het Karkheh stroomgebied te maken en het
omvang van de variatie en de verandering te bestuderen op verschillende tijds- en
plaatsafhanklijke schalen. De gebruikte methode bestaat uit een gecombineerd
gebruik van een gedegen systeem analyse en hydrologische modeleer technieken. De
plaatsafhankelijke dimensie is bestudeerd op stroomgebied, deelstroomgebied en
substroomgebied, terwijl de tijdsafhankelijke dimensie is bestudeerd aan de hand
van dag, maand, jaarwaarden en lange tijdsreeksen.
De uitgebreide hydrologische studie van de plaats- en tijdsafhankelijke variatie
in het oppervlaktewater is uitgevoerd op basis van een lange tijdsreeks van
dagelijkse afvoergegevens tussen 1961 en 2001 voor zeven belangrijke
afvoermeetstations gelegen in de Karkheh rivier en haar belangrijkste zijtakken. De
analyses zijn uitgevoerd met behulp van verschillende technieken, waaronder
centrale tendentie en dispersie, afvoerscheiding en debietduuranalyse. Tevens is
stroomgebied boekhouding toegepast voor het jaar 1993-94, waarvoor de benodige
gegevens beschikbaar waren.
Het onderzoek laat zien dat de hydrologie van het Karkheh stroomgebied grote
inter- en intra-jaarlijkse variatie heeft, voornamelijk veroorzaakt door grote plaats10
Summary in Dutch was translated by Susan Graas and Marloes Mul, UNESCO-IHE, Delft,
the Netherlands.
146
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
en tijdsafhankelijke variatie in klimaat en lokale verschillen in bodem, landgebruik
en hydrogeologische eigenschappen van het stroomgebied welke grotendeels
onderdeel uitmaakt van het Zagros gebergte. De afvoer neemt toe vanaf begin
oktober en duurt tot aan april. Piekafvoeren vinden normaliter plaats in maart en
april, maar overstromingen kunnen te allen tijden tussen november en april
plaatsvinden. Deze hoge afvoeren worden veroorzaakt door een combinatie van
smeltwater en neerslag. In de periode mei tot en met september overheersen lage
afvoeren afkomstig van de basis afvoer vanuit de ondiepe grondwatervoorraden.
Opvallend is dat het afvoerregiem in het midden gedeelte van het stroomgebied
(Karkheh rivier) verschilt van de bovenstroomse gedeelten (Gamasiab en Qarasou),
waarbij de eerste meer afvoer per oppervlakte-eenheid genereert en een hogere
basisafvoer heeft. De kwestie van variatie is hier onderbouwd middels ramingen van
de gemiddelde jaarafvoeren van de Karkheh rivier gemeten bij het Paye Pole station
(net benedenstrooms van de Karkheh dam). De gemiddelde afvoer op deze locatie is
5,83 x 109 m3/jaar, terwijl de jaarafvoer in het extreem droge jaar 1999-2000 slechts
iets meer dan eenderde ervan bedroeg (1,916 x 109 m3/jaar) en het hoogst gemeten
jaarafvoer was gemeten gedurende het extreem natte jaar 1968-69 en bedroeg circa
12,60 x 109 m3/jaar. Onder dergelijk sterk varierende omstandigheden, is het begrip
van de betrouwbaarheid van informatie met betrekking tot waterbeschikbaarheid
zeer belangrijk voor beter watergebruik en beslissingen omtrend watertoewijzing.
Voor het Karkheh stroomgebied is de infornatie over de variatie in afvoer en
betrouwbaarheid op dag, maand en jaarbasis verkregen door middel van de
debietsduuranalyse.
De synthese van de resultaten over de hydrologische variatie,
waterbeschikbaarheid, en water boekhouding suggereert dat het Karkheh
stroomgebied open was gedurende de onderzoeksperiode (1961-2001), en dat er
ruimte is voor verdere watertoewijzing, tot circa 1-4 x 109 m3/jaar, afhankelijk van
de hoeveelheid water toegewezen aan de natuur. Echter, de toewijzing dient pas te
gebeuren na een zorgvuldige effectenstudie en trade-off analyse van meerdere en
zeer concurrende toepassingen en gebuikers in het stroomgebied. De op hande zijnde
waterbeheerplannen lijken niet duurzaam te zijn gelet op de beschikbare hoeveelheid
water en zijn grote variatie. Indien het huidige waterbeleid ook in de toekomst zal
worden uitgevoerd dan zal het stroomgebied uiterlijk in 2025 gesloten zijn en dan
zal tegemoetkomen aan alle watervragen zeer moeilijk zijn, met name gedurende de
maanden met lage afvoer en tijdens droge jaren. Het milieu zal waarschijnlijk het
meeste schade ondervinden aangezien deze, tot op heden, de laagste prioirteit heeft
gekregen, maar ook andere sectoren waaronder de landbouw en huishoudelijk
gebruik zullen waarschijnlijk ook te maken krijgen met vermindering van hun
toegewezen waterrechten.
Als onderdeel van de systeemanalyse zijn tevens de veranderingen in de hydroklimatologische variabelen en hun afhankelijkheid onderzocht. Afvoergegevens van
vijf meetstations zijn gebruikt voor de periode 1961-2001 om trends in een aantal
afvoer variabelen te onderzoeken welke de omvang van afvoervariatie laten zien, bv
gemiddelde jaar en maandafvoer, 1- en 7-daagse maximum en minimum afvoer,
datum van de 1-daagse maxima en minima en het aantal en duur van piek- en lage
Samenvatting
147
____________________________________________________________________
afvoeren. Voor de neerslag- en temperatuurgegevens van zes synoptische
klimaatstations in de periode 1950 tot 2003 is een vergelijkbaar onderzoek naar
trends in klimatologische variabelen uitgevoerd alsmede de correlatie met de afvoer.
De Spearman rank test is gebruikt voor het vaststellen van de trends en de correlatie
analyse is gebaseerd op de Pearson methode. De resultaten laten een aantal
significante trends in afvoervariabelen zien, zowel toenemend als afnemend.
Bovendien zijn de gevonden trends niet uniform qua plaats. De afname in de
basisafvoer karakteristieken zijn significanter in de bovenstroomse delen van het
stroomgebied (met name de Qarasou rivier), terwijl de toenemende trends in hoge
afvoeren en winter afvoeren met name in het middengedeelte van het stroomgebied
(Kashkan rivier) plaatsvinden. De meeste van deze trends worden voornamelijk
veroorzaakt door veranderingen in neerslag. De resultaten laten zien dat de
vermindering van neerslag in april en mei leidt tot vermindering in de basisafvoer,
terwijl toename van de neerslag in de winter (met name in maart) samen met
temperatuursveranderingen leidt tot een toename in het overstromingspatroon. De
gevonden trends van het Jelogir meetstation aan de Karkheh rivier reflecteren het
gecombineerde effect van het bovenstrooms gelegen stroomgebied. De gevonden
significante trend voor een aantal afvoer karakteristieken waaronder het 1-daagse
maximum, de december afvoer en het aantal en de duur van de lage afvoer, wijzen
op veranderingen van het hydrologische regiem van de Karkheh rivier en worden
voornamelijk toegewezen aan veranderingen in de klimatologische variabelen.
In het Karkheh stroomgebied zijn afvoergegevens van vele deelstroomgebieden
niet beschikbaar zijn en veel afvoermeetstations zijn verlaten, hierdoor is de
methode van het regionaliseren van de hydrologische parameters geschikt voor het
bepalen van de waterbeschikbaarheid in die lokaties. In dit onderzoek is een nieuwe
regionaliseringsmethode ontwikkeld om afvoertijdreeksen te schatten voor slecht
bemeten stroomgebieden. De voorgestelde methode is gebaseerd op de
regionalisering van een conceptueel neerslag-afvoer-model gebaseerd op de
gelijkenis met de debietsduuranalyse (DDA). De resultaten van deze methode zijn
vergeleken met drie andere methoden die gebaseerd zijno op de groote van het
stroomgebied, de ruimtelijke nabijheid en de stroomgebiedskenmerken. De gegevens
van 11 bemeten stroomgebieden (475 tot 2.522 km2) zijn gebruikt om de
regionaliseringsprocedures te ontwikkelen. Het op grote schaal toegepaste model
HBV is gebruikt om dagelijkse afvoeren te simuleren middels overgedragen
parameters van bemeten vergelijkbare stroomgebieden. Het onderzoek laat zien dat
het baseren van de HBV parameters op de DDA vergelijkingsfactor een betere
afvoer simuleert dan de drie andere methodes. Bovendien is aangetoond dat de
onzekerheid van de parameters een klein effect heeft op het
regionaliseringsresultaat. De resultaten van deze nieuwe methode verhouden zich
zeer goed met de meeste elders ontwikkeld en toegepaste regionaliseringsmethoden.
Daarom is de in deze studie ontwikkelde DDA regionaliseringsmethode een
waardevolle aanvulling op bestaande regionaliseringsmethoden. De voorgestelde
methode is eenvoudig te repliceren in andere stroomgebieden, met name die
stroomgebieden waarvan het afvoerwaarnemingsnetwerk vermindert.
148
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Hiernaast is een semi-gedistribueerd, proces-gebaseerd model – Soil Water
Assessment Tool (SWAT) – gebruikt om de hydrologische fluxen te begrijpen en te
kwantificeren en om verschillende scenarios te testen. Het is bekend dat het veel
gebruikte SWAT model een groot verscheidenheid aan mogelijkheden biedt om de
modelstructuur te definieren, maar de invoer van klimatologische gegevens is nog
redelijk simplistisch. SWAT gebruikt de gegevens van het neerslagstation welke het
dichst bij het zwaartepunt van dat deelstroomgebied gelegen is. Dit is wellicht niet
representatief voor de neerslag in het hele deelstroomgebied en kan leiden tot
toenemende onzekerheid in de model resultaten. In dit onderzoek is een alternatieve
methode voor de invoer van de neerslag gegevens geevalueerd. Specifiek is de
invoer van geinterpoleerde gebiedsneerslag getest versus de standaard SWAT invoer
procedure voor neerslag. Het modelleergebied beslaat 42.620 km2 en is gelegen in
het bergachtige, semi-aride deel van het onderzoeksstroomgebied, welke het grootste
deel van de stroomgebiedsafvoer genereert. De modelresultaten zijn beoordeeld op
dag-, maand- en jaarbasis aan de hand van een aantal indicatoren van 15
afvoerstations, met een stroomgebied van 590 tot 42.620 km2. De vergelijking
suggereert dat het gebruik van gebiedsneerslag de modelprestaties verbetert, met
name in kleine sub-stroomgebieden van 600 tot 1.600 km2. De invoer van
gebiedsneerslag resulteert in een toenemende betrouwbaarheid van gesimuleerde
afvoer in gebieden met een kleine neerslagstations dichtheid en een slechte verdeling
van de neerslagmeter(s). Beide methodes voor de invoer van neerslaggegevens
resulteren in redelijk goede simulaties voor grotere stroomgebieden (meer dan 5.000
km2), wat verklaard kan worden door het uitmiddelen van de neerslagvariatie voor
grotere gebieden.
Het begrip van de stroomgebiedshydrologie aan de hand van de bovengenoemde
studies, veldbezoeken en literatuuronderzoek en gedegen parameterschatting
procedures heeft geholpen om een redelijk goede calibratie, validatie en
onzekerheidsanalyse van het SWAT model voor het Karkheh stroomgebied te
krijgen. Dit levert voldoende zekerheid op om het model toe te passen voor het
analyseren van watergebruik scenarios in het stroomgebied. Drie scenarios,
gerelateerd aan een toename van watergebruik in de regenafhankelijke landbouw,
zijn geevalueerd. De onderzochte scenarios zijn: opwaarderen van regenafhankelijke
gebieden naar geirrigeerde landbouw (S1), verbeteren van het bodemvochtgehalte
door opvang van regenwater (S2) en een combinatie van S1 en S2 (S3). De
resultaten van deze scenarios zijn vergeleken met de baseline in de periode 19882000. De baseline simulaties zijn uitgevoerd met de uiteindelijk vastgestelde
modelstructuur en de parameters verkregen uit de gebruikte calibratie procedure. De
resultaten van het eerste scenario (S1) geven een vermindering van 10% van de
gemiddelde jaarafvoer op stroomgebiedsniveau, welke varieert van 8 tot 15% voor
de belangrijkste deelgebieden van het stroomgebied. De afname in de gemiddelde
maandafvoeren varieert tussen de 3 en 56% op stroomgebiedsniveau. De maanden
mei – juli vertonen een groot effect, met in juni de grootste afname in afvoer. De
afnames in afvoer in deze maanden zijn groter in de bovenstroomse gebieden van
het stroomgebied wat hoofdzakelijk veroorzaakt wordt door een relatief groter
potentieel te ontwikkelen irrigatiegebied in combinatie met relatief lagere afvoeren
Samenvatting
149
____________________________________________________________________
in deze maanden. De effecten van S2 zijn over het algemeen klein op deelgebied en
stroomgebiedsniveau, met afnames van 2-5% en 1-10% in respectievelijk de
gemiddelde jaar- en gemiddelde maandafvoeren. De geschatte afvoerafnames op
jaarbasis blijft ruim binnen het beschikbare waterontwikkelingspotentieel van het
stroomgebied. Echter, het voorkomen van buitensporige afvoerafnames in mei-juli
zal aanvullende maatregelen vereisen, zoals aanvullende irrigatie en het vergroten
van de aanvoer door een reeks van bergingsmogelijkheden en rekening houden met
het opwaarderen van minder landbouwgrond naar irrigatie dan potentieel mogelijk is
(met name in de bovenstroomse gebieden van het stroomgebied).
Het onderzoek concludeert dat kennis van de variatie in hydrologie en
waterbeschikbaarheid, en het inachtnemen van de variatie van de
waterbeschikbaarheid in de waterbeheerplannen een belangrijke rol speelt in het
duurzaam gebruik and management van het beschikbare water in het Karkheh
stroomgebied. De huidige watertoewijzing is niet duurzaam en een grondige
herziening wordt aanbevolen. Deze zal uiteindelijk een vermindering in
waterrechten voor menselijk gebruik (met name de landbouw) vereisen en leiden tot
meer water voor het milieu. De klimatologische variatie en veranderingen hebben
het afvoerregiem van de Karkheh rivier significant veranderd, wat onmiddelijke
ingrijpen zou rechtvaardigen, door bijvoorbeeld structurele maatregelen en
programma’s om de stroomgebieddegradatie ten behoeve van het beheersen van
overstromingen in het middengedeelte van het stroomgebied te herzien en het
bekijken hoe wateronttrekkingen gedurende de lage afvoermaanden (mei tot
september) in de bovenstroomse gebieden verminderd kunnen worden om de
gevolgen van de afname in lage afvoeren in deze gebieden te compenseren. De
uitgevoerde effectenstudie laat zien dat efficienter watergebruik in de
neerslagafhankelijke landbouw gepromoot zou kunnen worden, met inachtneming
van bodembeschermende en waterbesparende technieken in het hele stroomgebied
aangezien deze minimale gevolgen hebben voor de waterbeschikbaarheid
benedenstrooms. Echter, de omzetting van gebieden metgrotendeels
neerslagafhankelijke landbouw naar irrigatie vereist een voorzichtige aanpak om
redelijke limieten van afvoer afnames op maandbasis te verzekeren. Dit vereist het
opwaarderen van beperkte gebieden met neerslagafhankelijke landbouw naar
irrigatie (met name in het bovenstroomse gedeelte van het stroomgebied), het
toepassen van irrigatie enkel bij tekorten en het ontwikkelen van een reeks aan
waterbergingsmogelijkheden.
Het
versterken
van
hydro-klimatologische
meetnetwerken wordt aanbevolen om de beschikbaarheid van gegevens en daarmee
de toepassing van hydrologische en waterbeheermodellen voor beter geinformeerde
besluitvorming te verbeteren. Hiermee samenhangend wordt het herstellen van
verlaten hydro-klimatologische meetstations en het overwegen om meer
meetstations in het berggebied te installeren aanbevolen. Het integraal waterbeheer
dient gepromoot te worden in het onderzoeksgebied.
De kennis vergaard tijdens dit onderzoek kan als zeer relevant gezien worden
voor andere stroomgebieden in Iran en wereldwijd.
‫‪SUMMARY IN PERSIAN11‬‬
‫ﺧﻼﺻﻪ ﺑﻪ ﻓﺎرﺳﻲ‬
‫ﺗﻮﺳﻌﻪ روزاﻓﺰون اﺳﺘﻔﺎده اﻧﺴﺎﻧﻬﺎ از ﻣﻨﺎﺑﻊ آب ‪ ،‬ﺑﻪ وﻳﮋه ﺑﺮاي ﻛﺸﺎورزي ‪ ،‬ﺑﺎﻋﺚ اﻳﺠﺎد ﻓﺸﺎر ﻓﺰاﻳﻨﺪه ﺑﺮ ﻣﻨﺎﺑﻊ آب ﺷﻴﺮﻳﻦ ﺷﺪه‬
‫اﺳﺖ‪ .‬ﺑﺎ وﺟﻮد اﻳﻨﻜﻪ دﺳﺘﺮﺳﻲ اﻧﺴﺎن ﺑﻪ آب و اﺳﺘﻔﺎده از آن در ﺑﺴﻴﺎري از زﻣﻴﻨﻪ ﻫﺎ ﻣﺎﻧﻨﺪ ﺑﻬﺒﻮد ﺗﻮﻟﻴﺪ ﻣﻮاد ﻏﺬاﻳﻲ و رﻓﺎه‬
‫اﺟﺘﻤﺎﻋﻲ و اﻗﺘﺼﺎدي ﺑﻪ اﻧﺴﺎن ﻛﻤﻚ ﻛﺮده‪ ،‬وﻟﻲ ﻣﻨﺠﺮ ﺑﻪ ﺻﺪﻣﻪ دﻳﺪن ﻣﺤﻴﻂ زﻳﺴﺖ ﻧﻴﺰ ﺷﺪه اﺳﺖ‪ .‬اﻳﺠﺎد ﺗﻮازن در ﺑﻬﺮه‬
‫ﺑﺮداري اﻧﺴﺎن و ﻃﺒﻴﻌﺖ از ﻣﻨﺎﺑﻊ آب ﺑﻪ ﻋﻨﻮان ﭼﺎﻟﺶ ﺑﺰرگ اﻳﻦ ﻗﺮن ﻧﺎﻣﻴﺪه ﻣﻲ ﺷﻮد‪ .‬اﻳﻦ ﻣﺴﺌﻠﻪ ﺑﺮاي ﻣﻨﺎﻃﻖ ﻧﻴﻤﻪ ﺧﺸﻚ و‬
‫ﺧﺸﻚ ﺟﻬﺎن ‪ ،‬ﻫﻤﭽﻮن ﺟﻤﻬﻮري اﺳﻼﻣﻲ اﻳﺮان ‪ ،‬ﻛﻪ در آن از ﻳﻜﺴﻮ آب ﺑﻪ ﻃﻮر ﻛﻠﻲ ﻛﻤﻴﺎب اﺳﺖ و ازﺳﻮي دﻳﮕﺮ ﺗﻘﺎﺿﺎ‬
‫ﺑﺮاي آب در ﻛﺸﺎورزي ‪ ،‬ﺻﻨﻌﺖ و ﻣﻨﺎﻃﻖ ﺷﻬﺮي ﺑﻪ ﺳﺮﻋﺖ در ﺣﺎل اﻓﺰاﻳﺶ اﺳﺖ‪ ،‬ﺑﻪ ﻣﺮاﺗﺐ ﭘﻴﭽﻴﺪه ﺗﺮ اﺳﺖ‪ .‬ﺗﻨﻮع آب و‬
‫ﻫﻮاﻳﻲ و ﺗﻐﻴﻴﺮات اﻗﻠﻴﻤﻲ ﻧﻴﺰ ﺑﺮ اﻳﻦ ﭘﻴﭽﻴﺪﮔﻲ ﻣﻲ اﻓﺰاﻳﺪ‪.‬‬
‫درﺷﺮاﻳﻂ ﻛﻤﺒﻮد آب و رﻗﺎﺑﺘﻲ ﺷﺪن اﺳﺘﻔﺎده از آب‪ ،‬ﺷﻨﺎﺧﺖ ﻫﻴﺪروﻟﻮژي ﺣﻮﺿﻪ آﺑﺮﻳﺰو ﻣﻨﺎﺑﻊ آب در دﺳﺘﺮس ﻣﺤﻮر اﺻﻠﻲ‬
‫ﺗﺪوﻳﻦ ﺳﻴﺎﺳﺘﻬﺎي ﺗﻮﺳﻌﻪ ﭘﺎﻳﺪار در ﺑﺨﺶ آﺑﺮﺳﺎﻧﻲ ﻣﻲ ﺑﺎﺷﺪ‪ .‬ﻣﻄﺎﻟﻌﻪ ﺣﺎﺿﺮ در ﺣﻮﺿﻪ آﺑﺮﻳﺰ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ اﻳﺮان اﻧﺠﺎم ﺷﺪه‬
‫اﺳﺖ‪ .‬اﻳﻦ ﻣﻨﻄﻘﻪ از ﻣﻨﺎﻃﻖ ﺧﺸﻚ ﺗﺎ ﻧﻴﻤﻪ ﺧﺸﻜﻲ ﺑﻪ ﺣﺴﺎب ﻣﻲ آﻳﺪ‪ ،‬ﻛﻪ ﻓﺎﻗﺪ ﺷﻨﺎﺧﺖ ﻫﻴﺪروﻟﻮژي ﻣﻨﺎﺳﺐ اﺳﺖ و ﺗﺎﺛﻴﺮ‬
‫ﺗﻐﻴﻴﺮات ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﺑﺮ اﺳﺘﻔﺎده از آب در اﻳﻦ ﺣﻮﺿﻪ آﺑﺮﻳﺰ ﺑﺮرﺳﻲ ﻧﺸﺪه اﺳﺖ‪ .‬ﻫﺪف اﺻﻠﻲ از اﻳﻦ ﭘﮋوﻫﺶ ارزﻳﺎﺑﻲ‬
‫ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﻣﻨﺎﺑﻊ آب )ﺳﻄﺤﻲ( در ﺣﻮﺿﻪ آﺑﺮﻳﺰ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ و ﺑﺮرﺳﻲ ﺗﻨﻮع و ﺗﻐﻴﻴﺮات آن در ﻣﻘﻴﺎس ﻫﺎي زﻣﺎﻧﻲ و‬
‫ﻣﻜﺎﻧﻲ ﻣﺘﻔﺎوت اﺳﺖ‪ .‬ﭼﺎرﭼﻮب روش ﻣﻮرد اﺳﺘﻔﺎده در اﻳﻦ ﭘﮋوﻫﺶ ﺗﺮﻛﻴﺒﻲ ازﺑﺮرﺳﻲ دﻗﻴﻖ ﺳﻴﺴﺘﻢ ﻫﺎ ﻫﻤﺮاه ﺑﺎ ﻣﺪل ﺳﺎزي‬
‫ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﻣﻲ ﺑﺎﺷﺪ‪ .‬ﻣﻄﺎﻟﻌﺎت اﻧﺠﺎم ﺷﺪه در اﻳﻦ ﭘﮋوﻫﺶ در ﻣﻘﻴﺎس ﻫﺎي ﻣﻜﺎﻧﻲ رودﺧﺎﻧﻪ‪ ،‬ﺣﻮﺿﻪ آﺑﺮﻳﺰ و زﻳﺮ ﺣﻮﺿﻪ‬
‫آﺑﺮﻳﺰو ﻣﻘﻴﺎس ﻫﺎي زﻣﺎﻧﻲ روزاﻧﻪ‪ ،‬ﻣﺎﻫﺎﻧﻪ‪ ،‬ﺳﺎﻻﻧﻪ و ﺑﻠﻨﺪ ﻣﺪت ﺗﺮ اﻧﺠﺎم ﺷﺪه اﺳﺖ‪.‬‬
‫ارزﻳﺎﺑﻲ ﺟﺎﻣﻊ ﺗﻨﻮع زﻣﺎﻧﻲ و ﻣﻜﺎﻧﻲ ﻫﻴﺪروﻟﻮژي آب ﺳﻄﺤﻲ ﺑﺎ اﺳﺘﻔﺎده از داده ﻫﺎي ﻣﻮﺟﻮد ﺟﺮﻳﺎن درﻫﻔﺖ اﻳﺴﺘﮕﺎه ﻣﻬﻢ‬
‫اﻧﺪازه ﮔﻴﺮي واﻗﻊ در رودﺧﺎﻧﻪ ﻛﺮﺧﻪ و ﺳﺮ ﺷﺎﺧﻪ ﻫﺎي آن ﺑﻴﻦ ﺳﺎﻟﻬﺎي ‪ 2001-1961‬اﻧﺠﺎم ﭘﺬﻳﺮﻓﺘﻪ اﺳﺖ ‪ .‬اﻳﻦ ﺗﺠﺰﻳﻪ و‬
‫ﺗﺤﻠﻴﻞ ﺑﺎ اﺳﺘﻔﺎده از روش ﻫﺎي ﻣﺎﻧﻨﺪ اﻧﺪازه ﮔﻴﺮي ﺗﻤﺎﻳﻞ ﻣﺮﻛﺰي و ﭘﺮاﻛﻨﺪﮔﻲ ‪ ،‬ﺟﺪاﻳﻲ ﺟﺮﻳﺎن ﭘﺎﻳﻪ و آﻧﺎﻟﻴﺰ ﻣﺪت زﻣﺎن ﺟﺮﻳﺎن‬
‫اﻧﺠﺎم ﺷﺪه اﺳﺖ‪.‬‬
‫‪11‬‬
‫‪Summary in Persian was translated by Ali Dastgheib, UNESCO-IHE, Delft, the‬‬
‫‪Netherlands.‬‬
‫ﺧﻼﺻﻪ ﭘﺎﻳﺎن ﻧﺎﻣﻪ ﺗﻮﺳﻂ ﻋﻠﻲ دﺳﺖ ﻏﻴﺐ‪ ،‬ﻣﺪرس ﻳﻮﻧﺴﻜﻮ ‪ ،IHE -‬دﻟﻔﺖ‪ ،‬ﻫﻠﻨﺪ ‪،‬ﺑﻪ ﻓﺎرﺳﻲ ﺗﺮﺟﻤﻪ ﺷﺪه اﺳﺖ‪.‬‬
‫‪152‬‬
‫‪Understanding Hydrological Variability for Improved Water Management‬‬
‫____________________________________________________________________‬
‫اﻳﻦ ﻣﻄﺎﻟﻌﺎت ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ ﻫﻴﺪروﻟﻮژي ﺣﻮﺿﻪ آﺑﺮﻳﺰ ﻛﺮﺧﻪ داراي ﺗﻐﻴﻴﺮات ﺷﺪﻳﺪي ﻫﻢ در ﻣﺪت ﻳﻜﺴﺎل و ﻫﻢ در‬
‫ﺳﺎﻟﻬﺎي ﻣﺘﻮاﻟﻲ ﻣﻲ ﺑﺎﺷﺪ‪ .‬اﻳﻦ ﺗﻐﻴﻴﺮات ﺑﻪ ﻋﻠﺖ ﺗﻐﻴﻴﺮات و ﺗﻨﻮع زﻣﺎﻧﻲ و ﻣﻜﺎﻧﻲ آب و ﻫﻮا و ﺗﻐﻴﻴﺮات ﻣﻜﺎﻧﻲ ﺟﻨﺲ ﺧﺎك و‬
‫ﻧﺤﻮه ﺑﻬﺮه ﺑﺮداري از زﻣﻴﻦ و ﻣﺸﺨﺼﺎت ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﺣﻮﺿﻪ آﺑﺮﻳﺰ اﻳﻦ رودﺧﺎﻧﻪ ﻛﻪ ﻗﺴﻤﺘﻲ از ﻛﻮه ﻫﺎي زاﮔﺮس اﺳﺖ‪ ،‬ﻣﻲ‬
‫ﺑﺎﺷﺪ‪ .‬اﻓﺰاﻳﺶ ﺟﺮﻳﺎن از اﻛﺘﺒﺮ ﺷﺮوع ﺷﺪه و ﺗﺎ آورﻳﻞ اداﻣﻪ ﭘﻴﺪا ﻣﻲ ﻛﻨﺪ‪ .‬ﻣﻌﻤﻮﻻ ﺣﺪ اﻛﺜﺮ ﺟﺮﻳﺎن در ﻣﺎﻫﺎي ﻣﺎرس و آورﻳﻞ‬
‫ﻣﺸﺎ ﻫﺪه ﻣﻲ ﺷﻮد وﻟﻲ ﺳﻴﻼب در ﻫﺮ زﻣﺎﻧﻲ ﺑﻴﻦ ﻧﻮاﻣﺒﺮ و آورﻳﻞ ﻣﻤﻜﻦ اﺳﺖ اﺗﻔﺎق ﺑﻴﺎﻓﺘﺪ‪ .‬اﻳﻦ ﺟﺮﻳﺎن زﻳﺎد ﺑﺮاﺛﺮ ﻫﻤﺰﻣﺎﻧﻲ‬
‫ذوب ﺑﺮف و ﺑﺎرش ﺑﺎران اﻳﺠﺎد ﻣﻲ ﺷﻮد‪ .‬دوره زﻣﺎﻧﻲ ﺑﻴﻦ ﻣﻲ و ﺳﭙﺘﺎﻣﺒﺮ دوره ﻛﻢ آﺑﻲ ﺑﻪ ﺣﺴﺎب ﻣﻲ آﻳﺪ و ﻣﻨﺒﻊ ﺟﺮﻳﺎن در اﻳﻦ‬
‫دوره ﻣﻌﻤﻮﻻ ﻣﻨﺎﺑﻊ آب زﻳﺮ ﺳﻄﺤﻲ اﺳﺖ‪ .‬ﻋﻼوه ﺑﺮ اﻳﻦ ‪ ،‬رژﻳﻢ رواﻧﺎب در ﺑﺨﺶ ﻣﻴﺎﻧﻲ ﺣﻮﺿﻪ )رودﺧﺎﻧﻪ ﻛﺸﻜﺎن( ﺑﻪ ﻃﻮر‬
‫ﻗﺎﺑﻞ ﻣﻼﺣﻈﻪ اي ﺑﺎ ﺑﺎﻻ دﺳﺖ ﺣﻮﺿﻪ )ﻗﺎﻣﺎﺳﻴﺎب و ﻗﺮه ﺳﻮ( ﻣﺘﻔﺎوت اﺳﺖ‪ .‬در ﻧﻮاﺣﻲ ﻣﻴﺎﻧﻲ ﺣﻮﺿﻪ رواﻧﺎب ﺑﻴﺸﺘﺮي در واﺣﺪ‬
‫ﺳﻄﺢ دﻳﺪه ﻣﻲ ﺷﻮد‪ .‬ﺗﻐﻴﻴﺮات ﻣﻮرد ﺑﺤﺚ ﺑﺮ اﺳﺎس ﺑﺮ آورد ﺟﺮﻳﺎن ﺳﺎﻻﻧﻪ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ و ﺗﻐﻴﻴﺮات آن در اﻳﺴﺘﮕﺎه اﻧﺪازه‬
‫ﮔﻴﺮي ﭘﺎﻳﻪ ﭘﻞ ) ﭘﺎﻳﻴﻦ دﺳﺖ ﺳﺪ ﻛﺮﺧﻪ( ﻧﺸﺎن داده ﺷﺪه اﺳﺖ‪ .‬ﻣﺘﻮﺳﻂ دﺑﻲ آب در اﻳﻦ ﻧﻘﻄﻪ ‪ 83,5 x109‬ﻣﺘﺮ ﻣﻜﻌﺐ در‬
‫ﺳﺎل اﺳﺖ‪ ،‬در ﺣﺎﻟﻴﻜﻪ در ﺳﺎل ﺑﺴﻴﺎر ﺧﺸﻚ ‪ 2000 -1999‬ﺟﺮﻳﺎن ﺳﺎﻻﻧﻪ ﻳﻚ ﺳﻮم اﻳﻦ ﻣﻘﺪار)‪ (916,1 x109‬ﺑﻮده‬
‫اﺳﺖ و در ﺳﺎل ﺑﺴﻴﺎر ﺗﺮ ‪ 1968-1969‬ﺟﺮﻳﺎن ﺳﺎﻻﻧﻪ ﺑﻪ ‪ 60,12 x109‬ﻣﺘﺮ ﻣﻜﻌﺐ در ﺳﺎل رﺳﻴﺪه اﺳﺖ‪ .‬ﺑﺎ وﺟﻮد‬
‫ﺗﻐﻴﻴﺮاﺗﻲ ﺑﻪ اﻳﻦ ﺷﺪت‪ ،‬درك درﺳﺘﻲ از دﺳﺘﺮس ﺑﻮدن ﻣﻨﺎﺑﻊ آب ﻣﻄﻤﺌﻦ ﺑﺮاي اﺳﺘﻔﺎده ﺑﻬﺘﺮ ﻣﻌﻨﺎي ﺑﻴﺸﺘﺮي ﭘﻴﺪا ﻣﻲ ﻛﻨﺪ‪ .‬آﻧﺎﻟﻴﺰ‬
‫ﻣﺪت زﻣﺎن ﺟﺮﻳﺎن اﻧﺠﺎم ﺷﺪه در اﻳﻦ ﭘﮋوﻫﺶ‪ ،‬اﻳﻨﭽﻨﻴﻦ ﺑﺮآوردﻫﺎﻳﻲ را ﺑﺮاي ﺣﻮﺿﻪ ﻛﺮﺧﻪ در ﻣﻘﻴﺎس ﻫﺎي زﻣﺎﻧﻲ روزاﻧﻪ‪،‬‬
‫ﻣﺎﻫﺎﻧﻪ و ﺳﺎﻻﻧﻪ ﺑﻪ دﺳﺖ ﻣﻲ دﻫﺪ ‪.‬‬
‫ﺑﺮرﺳﻲ ﻧﺘﺎﻳﺞ در ﻣﻮرد ﺗﻐﻴﻴﺮات ﻫﻴﺪروﻟﻮژﻳﻜﻲ‪ ،‬دﺳﺘﺮﺳﻲ ﺑﻪ آب و ﻣﺤﺎﺳﺒﻪ ﺣﺠﻢ آب ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ ﺣﻮﺿﻪ ﻛﺮﺧﻪ در ﻃﻮل‬
‫زﻣﺎن ﻣﻮرد ﺑﺮرﺳﻲ )‪ (1961-2001‬ﻳﻚ ﺣﻮﺿﻪ ﺑﺎز ﺑﻮده اﺳﺖ و اﻣﻜﺎن ﺗﺨﺼﻴﺺ ﻣﻨﺎﺑﻊ آب ﺟﺪﻳﺪ در اﻳﻦ ﺣﻮﺿﻪ در ﺣﺪود‬
‫‪4-1 x109‬ﻣﺘﺮ ﻣﻜﻌﺐ در ﺳﺎل ﺑﺎ در ﻧﻈﺮ ﮔﺮﻓﺘﻦ آب ﻣﻮرد ﻧﻴﺎز ﺟﺮﻳﺎﻧﻬﺎي زﻳﺴﺖ ﻣﺤﻴﻄﻲ‪ ،‬ﻣﻴﺴﺮ اﺳﺖ‪ .‬ﺑﺎ اﻳﻦ وﺟﻮد‬
‫ﺗﺨﺼﻴﺺ ﻣﻨﺎﺑﻊ آب ﺟﺪﻳﺪ ﺑﺎﻳﺪ ﭘﺲ از ﺑﺮرﺳﻲ دﻗﻴﻖ اﺛﺮ ﻣﺼﺮف ﻛﻨﻨﺪه ﻫﺎي ﻣﺘﻔﺎوت آب ﺑﺮ ﻳﻜﺪﻳﮕﺮ‪ ،‬اﻧﺠﺎم ﺷﻮد‪ .‬ارزﻳﺎﺑﻲ ﺑﺮﻧﺎﻣﻪ‬
‫ﻫﺎي ﻛﻨﻮﻧﻲ ﺗﺨﺼﻴﺺ آب ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ اﻳﻦ ﺑﺮﻧﺎﻣﻪ ﻫﺎ ﭘﺎﻳﺪار ﻧﺨﻮاﻫﺪ ﺑﻮد و اﮔﺮ ﺳﻴﺎﺳﺘﻬﺎي ﻛﻨﻮﻧﻲ در ﺑﺨﺶ آب اداﻣﻪ ﭘﻴﺪا‬
‫ﻛﻨﺪ‪ ،‬در آﻳﻨﺪه ﻧﺰدﻳﻚ ) ﺣﺪاﻛﺜﺮ ﺗﺎ ‪ (2025‬اﻳﻦ ﺣﻮﺿﻪ ﺑﻪ ﻳﻚ ﺣﻮﺿﻪ ﺑﺴﺘﻪ ﺗﺒﺪﻳﻞ ﺧﻮاﻫﺪ ﺷﺪ و از آن ﭘﺲ ﺗﻬﻴﻪ آب ﺑﺮاي‬
‫ﻣﺼﺮف ﻛﻨﻨﺪﮔﺎن ﺑﻮﻳﮋه در ﻣﺎﻫﻬﺎي ﻛﻢ آب و ﻳﺎ ﺳﺎﻟﻬﺎي ﺧﺸﻚ ﺑﺴﻴﺎر ﻣﺸﻜﻞ ﺧﻮاﻫﺪ ﺑﻮد و ﻣﺤﻴﻂ زﻳﺴﺖ ﻛﻪ ﺗﺎ ﻛﻨﻮن از‬
‫اﻫﻤﻴﺖ ﭘﺎﻳﻴﻦ ﺗﺮي در ﺳﻴﺎﺳﺖ ﮔﺰارﻳﻬﺎ ﺑﺮﺧﻮردارﺑﻮده اﺳﺖ‪ ،‬ﺑﻴﺸﺘﺮﻳﻦ ﺻﺪﻣﻪ را ﺧﻮاﻫﺪ دﻳﺪ‪ .‬ﺳﺎﻳﺮ ﺑﺨﺶ ﻫﺎ ﻣﺎﻧﻨﺪ ﻛﺸﺎورزي و‬
‫ﻣﺼﺎرف آب ﺧﺎﻧﮕﻲ ﻧﻴﺰ از ﺣﻖ آب ﻛﻤﺘﺮي ﺑﺮﺧﻮردار ﺧﻮاﻫﻨﺪ ﺑﻮد‪.‬‬
‫ﺗﻐﻴﻴﺮات ﻣﺘﻐﻴﺮﻫﺎي آب و ﻫﻮاﻳﻲ و اﺛﺮات ﻧﺎﺷﻲ از آن ﻧﻴﺰ ﻣﻮرد ﺑﺮرﺳﻲ ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬ﺑﺮاي ﺑﺮرﺳﻲ روﻧﺪ ﺗﻐﻴﻴﺮات ﺟﺮﻳﺎن‪،‬‬
‫داده ﻫﺎي ﭘﻨﺞ اﻳﺴﺘﮕﺎه اﻧﺪازه ﮔﻴﺮي اﺻﻠﻲ در ﺑﺎزه زﻣﺎﻧﻲ ‪ 1961-2001‬ﻣﻮرد اﺳﺘﻔﺎده ﻗﺮار ﮔﺮﻓﺘﻪ و ﺑﺮاﺳﺎس آﻧﻬﺎ ﺗﻌﺪادي از‬
‫ﻣﺘﻐﻴﺮﻫﺎي ﺟﺮﻳﺎن ﻣﺎﻧﻨﺪ ﻣﻴﺎﻧﮕﻴﻦ ﺟﺮﻳﺎن ﺳﺎﻻﻧﻪ و ﻣﺎﻫﺎﻧﻪ‪ ،‬ﺣﺪاﻗﻞ و ﺣﺪاﻛﺜﺮ ﺟﺮﻳﺎن ﻳﻜﺮوزه و ﻫﻔﺖ روزه‪ ،‬زﻣﺎن وﻗﻮع ﺣﺪاﻛﺜﺮ‬
‫و ﺣﺪاﻗﻞ ﺟﺮﻳﺎن در ﻳﻚ روز و زﻣﺎن ﺗﻨﺎوب ﺣﺪاﻗﻞ و ﺣﺪاﻛﺜﺮ ﺟﺮﻳﺎن ﻣﺤﺎﺳﺒﻪ ﺷﺪه اﺳﺖ‪ .‬ﻫﻤﭽﻨﻴﻦ داده ﻫﺎي ﺑﺎرﻧﺪﮔﻲ و‬
‫ﺗﺒﺨﻴﺮ از ﺷﺶ اﻳﺴﺘﮕﺎه ﺳﻴﻨﭙﺘﻴﻚ ﻫﻮاﺷﻨﺎﺳﻲ در ﺳﺎﻟﻬﺎي ‪ 1950‬ﺗﺎ ‪ 2003‬ﺟﻤﻊ آوري ﺷﺪه و ارﺗﺒﺎط روﻧﺪ ﺗﻐﻴﻴﺮات ﺟﻮي ﺑﺎ‬
‫‪Summary in Persian‬‬
‫‪153‬‬
‫____________________________________________________________________‬
‫ﺗﻐﻴﻴﺮات ﺟﺮﻳﺎن آب ﺳﻄﺤﻲ ﻣﻮرد ﺑﺮرﺳﻲ ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬ﺑﺮاي ﺗﻌﻴﻴﻦ روﻧﺪ ﺗﻐﻴﻴﺮات‪ ،‬آزﻣﻮن ﺳﭙﻴﺮﻣﻦ و ﺑﺮاي ﺗﻌﻴﻴﻦ ﺿﺮﻳﺐ‬
‫ﻫﻤﺒﺴﺘﮕﻲ‪ ،‬روش ﭘﻴﺮﺳﻮن ﻣﻮرد اﺳﺘﻔﺎده ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬ﻧﺘﺎﻳﺞ اﻳﻦ ﺑﺮرﺳﻲ روﻧﺪ ﺗﻐﻴﻴﺮات ﻗﺎﺑﻞ ﻣﻼﺣﻈﻪ اي را در ﻣﺘﻐﻴﺮﻫﺎي‬
‫ﺟﺮﻳﺎن ﻧﺸﺎن داد‪ .‬ﻋﻼوه ﺑﺮاﻳﻦ ﻧﺸﺎن داد ﻛﻪ اﻳﻦ روﻧﺪ ﺗﻐﻴﻴﺮات در ﻧﻘﺎط ﻣﺨﺘﻠﻒ ﻳﻜﺴﺎن ﻧﻴﺴﺖ‪ .‬ﻛﺎﻫﺶ ﺟﺮﻳﺎن در زﻣﺎن ﻛﻢ‬
‫آﺑﻲ در ﺑﺎﻻ دﺳﺖ ﺣﻮﺿﻪ )ﺑﻪ ﺧﺼﻮص در رودﺧﺎﻧﻪ ﻗﺮه ﺳﻮ( ﺑﺴﻴﺎر ﻗﺎﺑﻞ ﺗﻮﺟﻪ اﺳﺖ‪ ،‬در ﺣﺎﻟﻴﻜﻪ اﻓﺰاﻳﺶ ﺟﺮﻳﺎن در زﻣﺎن ﭘﺮ آﺑﻲ‬
‫و ﺟﺮﻳﺎﻧﻬﺎي زﻣﺴﺘﺎﻧﻲ در ﻣﻨﺎﻃﻖ ﻣﻴﺎﻧﻲ ﺣﻮﺿﻪ ) رودﺧﺎﻧﻪ ﻛﺸﻜﺎن( ﻗﺎﺑﻞ ﻣﻼﺣﻈﻪ اﺳﺖ‪ .‬ﺑﻴﺸﺘﺮ اﻳﻦ ﺗﻐﻴﻴﺮات را ﻣﻲ ﺗﻮان ﺑﻪ‬
‫ﺗﻐﻴﻴﺮات در ﺑﺎرﻧﺪﮔﻲ ﻧﺴﺒﺖ داد‪ .‬ﻧﺘﺎﻳﺞ ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ ﻛﺎﻫﺶ ﺑﺎرﻧﺪﮔﻲ در آورﻳﻞ و ﻣﻲ ﻣﻮﺟﺐ ﻛﺎﻫﺶ ﺟﺮﻳﺎن در اﻳﻦ ﻣﺎه‬
‫ﻫﺎ ﻣﻲ ﺑﺎﺷﺪ در ﺣﺎﻟﻴﻜﻪ در زﻣﺴﺘﺎن )ﺑﻪ ﺧﺼﻮص در ﻣﺎرس( ﺑﺎرﻧﺪﮔﻲ ﻫﻤﺮاه ﺑﺎ ﺗﻐﻴﻴﺮات دﻣﺎ ﺑﺎﻋﺚ اﻓﺰاﻳﺶ ﺳﻴﻼب ﻣﻲ ﺷﻮد‪.‬‬
‫ﺗﻐﻴﻴﺮات ﻣﺸﺎﻫﺪه ﺷﺪه در اﻳﺴﺘﮕﺎه ﺟﻠﻮﮔﻴﺮ در رودﺧﺎﻧﻪ ﻛﺮﺧﻪ ﻣﻨﻌﻜﺲ ﻛﻨﻨﺪه ﺗﺮﻛﻴﺐ اﺛﺮات زﻳﺮ ﺣﻮﺿﻪ ﻫﺎي ﺑﺎﻻ دﺳﺖ اﺳﺖ‪.‬‬
‫ﺗﻐﻴﻴﺮات ﻋﻤﺪه ﻣﺸﺎﻫﺪه ﺷﺪه در ﺗﻌﺪادي از ﻣﺘﻐﻴﻴﺮﻫﺎي ﺟﺮﻳﺎن در اﻳﺴﺘﮕﺎه ﺟﻠﻮﮔﻴﺮ ﻣﺎﻧﻨﺪ ﺣﺪاﻛﺜﺮ ﺟﺮﻳﺎن روزاﻧﻪ‪ ،‬ﺟﺮﻳﺎن ﻣﺎه‬
‫دﺳﺎﻣﺒﺮ و ﺗﻌﺪاد و ﻣﺪت ﻛﻢ آﺑﻲ‪ ،‬ﻧﺸﺎن دﻫﻨﺪه ﺗﻨﺎوب رژﻳﻢ ﻫﻴﺪروﻟﻮژﻳﻜﻲ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ اﺳﺖ ﻛﻪ آن ﻫﻢ در اﺛﺮ ﺗﻐﻐﻴﺮات‬
‫ﻣﺘﻐﻴﺮﻫﺎي ﺟﻮي ﻣﻲ ﺑﺎﺷﺪ‪.‬‬
‫ﺑﻪ دﻟﻴﻞ آﻧﻜﻪ داده ﻫﺎي ﺟﺮﻳﺎن ﺑﺮاي ﺑﺴﻴﺎري از زﻳﺮﺣﻮﺿﻪ ﻫﺎي رودﺧﺎﻧﻪ ﻛﺮﺧﻪ وﺟﻮد ﻧﺪارد و ﺗﻌﺪاد زﻳﺎدي از اﻳﺴﺘﮕﺎﻫﻬﺎي‬
‫اﻧﺪازه ﮔﻴﺮي ﺑﻲ اﺳﺘﻔﺎده رﻫﺎ ﺷﺪه اﻧﺪ‪ ،‬ﻣﻨﻄﻘﻪ اي ﻛﺮدن ﭘﺎراﻣﺘﺮﻫﺎي ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﻳﻜﻲ از ﻣﻮﺿﻮﻋﺎت ﻣﻬﻢ در ﺣﻮﺿﻪ ﻛﺮﺧﻪ‬
‫ﻣﻲ ﺑﺎﺷﺪ‪ .‬در اﻳﻦ ﭘﮋوﻫﺶ روش ﺟﺪﻳﺪي ﺑﺮاي ﺗﺨﻤﻴﻦ ﺗﺎرﻳﺨﭽﻪ ﺟﺮﻳﺎن درﺣﻮﺿﻪ ﻫﺎي ﺑﺎ اﻧﺪازه ﮔﻴﺮي ﻛﻢ اﺑﺪاع ﺷﺪه اﺳﺖ‪.‬‬
‫اﺳﺎس اﻳﻦ روش ﺟﺪﻳﺪ ﻣﻨﻄﻘﻪ اي ﻛﺮدن ﻳﻚ ﻣﺪل ﺑﺎرﻧﺪﮔﻲ ‪ -‬رواﻧﺎب ﺑﺮﭘﺎﻳﻪ ﻣﻨﺤﻨﻲ ﻫﺎي ﻣﺪت ﺟﺮﻳﺎن اﺳﺖ‪ .‬ﻛﺎراﻳﻲ اﻳﻦ روش‬
‫ﺑﺎ ﺳﻪ روش دﻳﮕﺮ ‪ :‬ﻣﺴﺎﺣﺖ ﻣﻨﻄﻘﻪ زﻫﻜﺸﻲ‪ ،‬ﻧﺰدﻳﻜﻲ ﻣﻜﺎﻧﻲ و ﻣﺸﺨﺼﺎت ﺣﻮﺿﻪ ﻣﻘﺎﻳﺴﻪ ﺷﺪه اﺳﺖ‪ .‬اﻧﺪازه ﮔﻴﺮي ﻫﺎي‬
‫ﻣﻮﺟﻮد در ‪ 11‬ﺣﻮﺿﻪ ) ‪ 475‬ﺗﺎ ‪ 2522‬ﻛﻴﻠﻮﻣﺘﺮ ﻣﺮﺑﻊ( دراﺑﺪاع اﻳﻦ روش ﻣﻮرد اﺳﺘﻔﺎده ﻗﺮار ﮔﺮﻓﺘﻪ و ﻣﺪل ﻣﺸﻬﻮر ‪HBV‬‬
‫ﺑﺮاي ﺷﺒﻴﻪ ﺳﺎزي ﺟﺮﻳﺎن روزاﻧﻪ ﺑﺎ اﺳﺘﻔﺎده از ﭘﺎراﻣﺘﺮﻫﺎي ﺣﻮﺿﻪ ﻫﺎي ﺷﺒﻴﻪ ﺑﻪ ﻫﻢ اﺳﺘﻔﺎده ﺷﺪه اﺳﺖ‪ .‬اﻳﻦ ﻣﻄﺎﻟﻌﺎت ﻧﺸﺎن داد‬
‫ﻛﻪ اﻧﺘﻘﺎل ﭘﺎراﻣﺘﺮﻫﺎي ﻣﺪل ‪ HBV‬ﺑﺮاﺳﺎس ﺷﺒﺎﻫﺘﻬﺎي ‪ FDC‬ﺷﺒﻴﻪ ﺳﺎزي رواﻧﺎب ﺑﻬﺘﺮي ﻧﺴﺒﺖ ﺑﻪ ﺳﻪ روش دﻳﮕﺮ ﺑﻪ دﺳﺖ‬
‫ﻣﻲ دﻫﺪ‪ .‬ﻋﺪم ﻗﻄﻌﻴﺖ در ﭘﺎراﻣﺘﺮﻫﺎي ﻣﺪل ﺗﺎﺛﻴﺮ ﭼﻨﺪاﻧﻲ ﺑﺮﻧﺘﺎﻳﺞ اﻳﻦ روش ﻧﺪارد‪ .‬ﻧﺘﺎﻳﺞ اﻳﻦ روش ﻣﻄﺎﺑﻘﺖ ﺑﺴﻴﺎر ﺧﻮﺑﻲ ﺑﺎ ﺳﺎﻳﺮ‬
‫ﺗﻜﻨﻴﻚ ﻫﺎي ﻣﻮﺟﻮد ﻛﻪ در ﻣﻨﺎﻃﻖ دﻳﮕﺮ ﻣﻮرد اﺳﺘﻔﺎده ﻗﺮارﮔﺮﻓﺘﻪ‪ ،‬دارد‪ .‬اﻳﺠﺎد دوﺑﺎره اﻳﻦ ﻣﺪل ﺑﺮاي اﺳﺘﻔﺎده درﺣﻮﺿﻪ آﺑﺮﻳﺰ‬
‫ﺳﺎﻳﺮ رودﺧﺎﻧﻪ ﻫﺎ ﺑﻪ ﺧﺼﻮص رودﺧﺎﻧﻪ ﻫﺎﻳﻲ ﻛﻪ ﺑﺎ ﻛﺎﻫﺶ ﺟﺮﻳﺎن ﻣﻮاﺟﻬﻨﺪ‪ ،‬ﺑﺴﻴﺎر آﺳﺎن ﻣﻲ ﺑﺎﺷﺪ‪.‬‬
‫ﻫﻤﭽﻨﻴﻦ ﻳﻚ ﻣﺪل ﻣﺒﺘﻨﻲ ﺑﺮﻓﺮآﻳﻨﺪ )‪ ( Soil Water Assessment Tool‬ﺑﺮاي ﺑﺮرﺳﻲ و ﺗﺨﻤﻴﻦ ﺷﺎرﻫﺎي‬
‫ﻫﻴﺪروﻟﻮژﻳﻜﻲ در ﺳﻨﺎرﻳﻮﻫﺎي ﻣﺘﻔﺎوت‪ ،‬ﻣﻮرد اﺳﺘﻔﺎده ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬درﻫﻨﮕﺎم اﺳﺘﻔﺎده ازﻣﺪل ‪ SWAT‬ﻣﺸﺨﺺ ﺷﺪ ﻛﻪ‬
‫ﻫﺮﭼﻨﺪ اﻳﻦ ﻣﺪل اﻣﻜﺎﻧﺎت زﻳﺎدي ﺑﺮاي ﺳﺎﺧﺘﺎر ﻣﺪل در اﺧﺘﻴﺎر ﻗﺮارﻣﻲ دﻫﺪ‪ ،‬وﻟﻲ ورودي ﻫﻮاﺷﻨﺎﺳﻲ ﻣﺪل ﺑﻴﺶ ازﺣﺪ ﺳﺎده‬
‫اﺳﺖ‪ SWAT .‬داده ﻫﺎي ﺑﺎرﻧﺪﮔﻲ ﻧﺰدﻳﻜﺘﺮﻳﻦ اﻳﺴﺘﮕﺎه ﺑﻪ ﻣﺮﻛﺰﻫﺮزﻳﺮﺣﻮﺿﻪ را ﺑﻪ ﻋﻨﻮان ورودي آن زﻳﺮﺣﻮﺿﻪ در ﻧﻈﺮ ﻣﻲ‬
‫ﮔﻴﺮد‪ .‬اﻳﻦ ورودي ﻣﻤﻜﻦ اﺳﺖ ورودي ﻣﻨﺎﺳﺒﻲ ﺑﺮاي ﻛﻞ ﺑﺎرﻧﺪﮔﻲ ﺣﻮﺿﻪ ﻧﺒﺎﺷﺪ و ﻣﻮﺟﺐ اﻓﺰاﻳﺶ ﻋﺪم ﻗﻄﻌﻴﺖ در ﻧﺘﺎﻳﺞ ﻣﺪل‬
‫ﺷﻮد‪ .‬دراﻳﻦ ﻣﻄﺎﻟﻌﺎت ﻳﻚ روش ﺟﺎﻳﮕﺰﻳﻦ ﺑﺮاي ورودي ﺑﺎرش ﻣﻮرد ارزﻳﺎﺑﻲ ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬ﺑﻪ ﻃﻮر ﻣﺸﺨﺺ ورودي‬
‫درون ﻳﺎﺑﻲ ﺷﺪه ﻣﻜﺎﻧﻲ ﺑﺎ ورودي اﺳﺘﺎﻧﺪارد ‪ SWAT‬ﻣﻘﺎﻳﺴﻪ ﺷﺪه اﺳﺖ‪ .‬داﻣﻨﻪ ﻣﺪل ﻣﺤﺪوده اي ﺑﻪ ﻣﺴﺎﺣﺖ ‪42620‬‬
‫‪154‬‬
‫‪Understanding Hydrological Variability for Improved Water Management‬‬
‫____________________________________________________________________‬
‫ﻛﻴﻠﻮﻣﺘﺮﻣﺮﺑﻊ در ﻣﻨﺎﻃﻖ ﻛﻮﻫﺴﺘﺎﻧﻲ ﻧﻴﻤﻪ ﺧﺸﻚ‪ ،‬ﻛﻪ ﺗﻘﺮﻳﺒﺎ ﻣﻨﺸﺎ ﺗﻤﺎم رواﻧﺎب ﺣﻮﺿﻪ اﺳﺖ‪ ،‬ﻣﻲ ﺑﺎﺷﺪ‪ .‬ﻛﺎراﻳﻲ ﻣﺪل‬
‫درﻣﻘﻴﺎﺳﻬﺎي زﻣﺎﻧﻲ روزاﻧﻪ‪ ،‬ﻣﺎﻫﺎﻧﻪ و ﺳﺎﻻﻧﻪ ﻣﻮرد ارزﻳﺎﺑﻲ ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ‪ .‬ﺑﺮاي اﻳﻦ ارزﻳﺎﺑﻲ از ﺷﺎﺧﺺ ﻛﺎراﻳﻲ ﻣﺪل در ‪15‬‬
‫اﻳﺴﺘﮕﺎه اﻧﺪازه ﮔﻴﺮي ﺟﺮﻳﺎن اﺳﺘﻔﺎده ﺷﺪه اﺳﺖ‪ .‬ﻣﺴﺎﺣﺖ ﻣﻨﻄﻘﻪ زﻫﻜﺸﻲ دراﻳﻦ اﻳﺴﺘﮕﺎﻫﻬﺎ ﺑﻴﻦ ‪ 590‬ﺗﺎ ‪42620‬‬
‫ﻛﻴﻠﻮﻣﺘﺮﻣﺮﺑﻊ ﻣﻲ ﺑﺎﺷﺪ‪ .‬اﻳﻦ ﺑﺮرﺳﻲ ﻧﺸﺎن داد ﻛﻪ اﺳﺘﻔﺎده از ورودي ﻣﺘﻐﻴﺮ ﻣﻜﺎﻧﻲ ﺑﻪ ﺧﺼﻮص درزﻳﺮﺣﻮﺿﻪ ﻫﺎي ﻛﻮﭼﻚ‬
‫)‪ 600‬ﺗﺎ ‪1600‬ﻛﻴﻠﻮﻣﺘﺮﻣﺮﺑﻊ( ﻛﺎراﻳﻲ ﻣﺪل را اﻓﺰاﻳﺶ ﻣﻲ دﻫﺪ در ﻧﺘﻴﺠﻪ اﻋﺘﺒﺎر ﺟﺮﻳﺎن ﺷﺒﻴﻪ ﺳﺎزي ﺷﺪه ﺑﺎ اﻳﻦ ورودي‬
‫درﺣﻮﺿﻪ ﻫﺎي ﺑﺪون اﻧﺪازه ﮔﻴﺮي ﺑﺎران و ﻳﺎ ﺑﺎ اﻧﺪازه ﮔﻴﺮي ﻛﻢ ﺑﻴﺸﺘﺮ اﺳﺖ‪ .‬ﻫﺮدوﻧﻮع ورودي ﺑﺎرش ﺑﺮاي ﺣﻮﺿﻪ ﻫﺎي‬
‫ﺑﺰرﮔﺘﺮ )ﺑﺰرﮔﺘﺮاز ‪ 5000‬ﻛﻴﻠﻮﻣﺘﺮﻣﺮﺑﻊ( ﻧﺘﺎﻳﺞ ﻳﻜﺴﺎﻧﻲ ﺑﻪ دﺳﺖ ﻣﻲ دﻫﺪ‪.‬‬
‫ﺷﻨﺎﺧﺖ آﺑﺸﻨﺎﺳﻲ ﺣﻮﺿﻪ ازراه ﻣﻄﺎﻟﻌﺎت ﻣﻮرد اﺷﺎره‪ ،‬ﺑﺮرﺳﻲ ﻫﺎي ﻣﻴﺪاﻧﻲ و روﺷﻬﺎي دﻗﻴﻖ ﺗﺨﻤﻴﻦ ﭘﺎراﻣﺘﺮﻫﺎي ﻫﻴﺪروﻟﻮژﻳﻜﻲ‬
‫ﺑﻪ ﻛﺎﻟﻴﺒﺮاﺳﻴﻮن وآﻧﺎﻟﻴﺰﻋﺪم ﻗﻄﻌﻴﺖ ﻣﺪل ‪ SWAT‬ﺑﺮاي ﺣﻮﺿﻪ آﺑﺮﻳﺰرودﺧﺎﻧﻪ ﻛﺮﺧﻪ ﻛﻤﻚ زﻳﺎدي ﻛﺮده اﺳﺖ‪ .‬داﺷﺘﻦ‬
‫اﻳﻨﭽﻨﻴﻦ ﻣﺪﻟﻲ اﻃﻤﻴﻨﺎن ﻻزم را ﺑﺮاي اﺳﺘﻔﺎده از آن در ﺑﺮرﺳﻲ ﺳﻨﺎرﻳﻮﻫﺎي ﻣﺘﻔﺎوت اﺳﺘﻔﺎده ازآب دراﻳﻦ ﺣﻮﺿﻪ اﻳﺠﺎد ﻣﻲ ﻛﻨﺪ‪.‬‬
‫ﺳﻪ ﺳﻨﺎرﻳﻮ ﺑﺮاي اﻓﺰاﻳﺶ اﺳﺘﻔﺎده از آب در ﻣﺰارع دﻳﻢ ﻣﻮرد ﺑﺮرﺳﻲ ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ ‪ :‬ﺗﺒﺪﻳﻞ ﻣﺰارع دﻳﻢ ﺑﻪ ﻣﺰارع آﺑﻲ )‪(S1‬‬
‫اﻓﺰاﻳﺶ آب ﻣﻮﺟﻮد در ﺧﺎك ﺑﺎ اﺳﺘﻔﺎده از ﺟﻤﻊ آوري آب ﺑﺎران )‪ (S2‬و ﺗﺮﻛﻴﺒﻲ ازدو ﺳﻨﺎرﻳﻮ ﻓﻮق )‪ .(S3‬ﻧﺘﺎﻳﺞ ﺷﺒﻴﻪ ﺳﺎزي‬
‫اﻳﻦ ﺳﻨﺎرﻳﻮﻫﺎ ﺑﺎ ﻧﺘﺎﻳﺞ ﻣﺪل ﭘﺎﻳﻪ )ﻣﺪل ﻛﺎﻟﻴﺒﺮه ﺷﺪه ﻣﻨﻄﻘﻪ( در ﺑﺎزه زﻣﺎﻧﻲ ‪ 1988‬ﺗﺎ ‪ 2000‬ﻣﻘﺎﻳﺴﻪ ﺷﺪه اﻧﺪ‪ .‬ﺳﻨﺎرﻳﻮاول ﺑﺎﻋﺚ‬
‫ﻛﻢ ﺷﺪن ‪ 10‬درﺻﺪي ﻣﻴﺎﻧﮕﻴﻦ ﺟﺮﻳﺎن ﺳﺎﻻﻧﻪ درﻛﻞ ﺣﻮﺿﻪ ﻣﻲ ﺷﻮد ﻛﻪ در زﻳﺮﺣﻮﺿﻪ ﻫﺎ اﻳﻦ ﻣﻴﺰان ﺑﻴﻦ ‪ 8‬ﺗﺎ ‪ 15‬درﺻﺪ ﻣﺘﻐﻴﺮ‬
‫اﺳﺖ‪ .‬ﻛﺎﻫﺶ ﻣﻴﺎﻧﮕﻴﻦ ﻣﺎﻫﺎﻧﻪ ﺟﺮﻳﺎن در ﻛﻞ ﺣﻮﺿﻪ ﺑﻴﻦ ‪ 3‬ﺗﺎ ‪ 56‬درﺻﺪ ﻣﻲ ﺑﺎﺷﺪ‪ .‬ﺗﺎﺛﻴﺮ اﺻﻠﻲ در ﻣﺎﻫﻬﺎي ﻣﻲ ﺗﺎ ﺟﻮﻻي دﻳﺪه‬
‫ﻣﻲ ﺷﻮد و ﺣﺪاﻛﺜﺮ ﻛﺎﻫﺶ در ﻣﺎه ژوﺋﻦ اﺗﻔﺎق ﻣﻲ اﻓﺘﺪ‪ .‬ﻛﺎﻫﺶ ﺟﺮﻳﺎن در اﻳﻦ ﻣﺎه ﻫﺎ درﻧﻮاﺣﻲ ﺑﺎﻻ دﺳﺘﻲ ﺣﻮﺿﻪ ﺑﻴﺸﺘﺮ ﻣﺸﻬﻮد‬
‫اﺳﺖ ﻛﻪ ﻋﻠﺖ اﺻﻠﻲ آن ﭘﺘﺎﻧﺴﻴﻞ ﺑﻴﺸﺘﺮ اﻳﻦ ﻧﻮاﺣﻲ ﺑﺮاي ﺗﺒﺪﻳﻞ ﻣﺰارع دﻳﻢ ﺑﻪ ﻣﺰارع آﺑﻲ و ﻫﻤﭽﻨﻴﻦ ﻛﻢ آﺑﻲ ﻃﺒﻴﻌﻲ اﻳﻦ ﻣﺎه‬
‫ﻫﺎﺳﺖ‪ .‬اﺛﺮ ﺳﻨﺎرﻳﻮ دوم ﺑﻪ ﻃﻮر ﻛﻠﻲ ﻛﻤﺘﺮازﺳﻨﺎرﻳﻮ اول ﻣﻲ ﺑﺎﺷﺪ‪ ،‬در اﻳﻦ ﺳﻨﺎرﻳﻮ ﻛﺎﻫﺶ ﻣﻴﺎﻧﮕﻴﻦ ﺳﺎﻻﻧﻪ و ﻣﺎﻫﺎﻧﻪ ﺟﺮﻳﺎن ﺑﻪ‬
‫ﺗﺮﺗﻴﺐ ‪ 2‬ﺗﺎ ‪ 5‬و ‪ 1‬ﺗﺎ ‪ 10‬درﺻﺪ اﺳﺖ‪ .‬ﻛﺎﻫﺶ ﺟﺮﻳﺎن ﺗﺨﻤﻴﻦ زده ﺷﺪه ﺑﻪ ﺧﻮﺑﻲ در ﻣﺤﺪوده ﻣﻨﺎﺑﻊ آب ﻗﺎﺑﻞ اﺳﺘﺤﺼﺎل‬
‫درﺣﻮﺿﻪ ﻣﻲ ﺑﺎﺷﺪ‪ .‬ﺑﺎ اﻳﻦ وﺟﻮد ﺟﻠﻮﮔﻴﺮي از ﻛﺎﻫﺶ زﻳﺎد ﺟﺮﻳﺎن در ﻣﺎﻫﻬﺎي ﻣﻲ ﺗﺎ ﺟﻮﻻي ﻧﻴﺎزﻣﻨﺪ اﻗﺪاﻣﺎت ﭘﻴﺸﮕﻴﺮاﻧﻪ اي‬
‫ﻣﺎﻧﻨﺪ ‪ :‬ﮔﺴﺘﺮش ﻣﺨﺎزن ذﺧﻴﺮه آب و ﺗﺒﺪﻳﻞ ﻣﺴﺎﺣﺖ ﻛﻤﺘﺮي از ﻣﺰارع دﻳﻢ ﺑﻪ ﻣﺰارع آﺑﻲ ﺑﻪ ﺧﺼﻮص در ﺑﺎﻻدﺳﺖ ﺣﻮﺿﻪ‬
‫اﺳﺖ‪.‬‬
‫اﻳﻦ ﭘﮋوﻫﺶ ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ درك درﺳﺘﻲ از ﺗﻨﻮع زﻳﺎد ﻫﻴﺪروﻟﻮژﻳﻜﻲ‪ ،‬ﻛﻪ از ﻧﺘﺎﻳﺞ اﻳﻦ ﻣﻄﺎﻟﻌﺎت ﺑﻪ دﺳﺖ آﻣﺪه‪ ،‬و در ﻧﻈﺮ‬
‫ﮔﺮﻓﺘﻦ ﺗﻨﻮع ﻣﻨﺎﺑﻊ آب‪ ،‬ﻧﻘﺶ ﺑﺴﻴﺎر ﻣﻬﻤﻲ در ﻣﺪﻳﺮﻳﺖ و ﺑﺮﻧﺎﻣﻪ رﻳﺰي ﭘﺎﻳﺪار ﻣﻨﺎﺑﻊ آب در ﺣﻮﺿﻪ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ دارد‪ .‬ﺑﺮﻧﺎﻣﻪ‬
‫ﻓﻌﻠﻲ ﺗﺨﺼﻴﺺ ﻣﺼﺮف آب در ﻃﻮﻻﻧﻲ ﻣﺪت ﭘﺎﻳﺪار ﻧﺨﻮاﻫﺪ ﺑﻮد و اﺣﺘﻴﺎج ﺑﻪ ﻳﻚ ﺑﺎزﻧﮕﺮي ﻛﺎﻣﻞ دارد ﻛﻪ ﺑﺎﻳﺪ ﺷﺎﻣﻞ ﻛﺎﻫﺶ‬
‫ﺗﺨﺼﻴﺺ آب ﺑﻪ ﻣﺼﺎرف اﻧﺴﺎﻧﻲ ) ﺑﻪ ﺧﺼﻮص ﻛﺸﺎورزي( و ﺗﺨﺼﻴﺺ آب ﺑﻪ ﻣﺤﻴﻂ زﻳﺴﺖ ﺑﺎﺷﺪ‪ .‬ﺗﻨﻮع و ﺗﻐﻴﻴﺮات آب و‬
‫ﻫﻮاﻳﻲ ‪ ،‬رژﻳﻢ ﻫﻴﺪروﻟﻮژﻳﻜﻲ ﺣﻮﺿﻪ رودﺧﺎﻧﻪ ﻛﺮﺧﻪ را دﺳﺘﺨﻮش ﺗﻐﻴﻴﺮﻛﺮده اﺳﺖ ﺑﻨﺎﺑﺮاﻳﻦ اﻗﺪام ﻫﺎي ﺳﺮﻳﻊ ﺟﻬﺖ ﺗﻌﺪﻳﻞ اﻳﻦ‬
‫ﺗﻐﻴﻴﺮات ﺑﺴﻴﺎر ﻻزم ﺑﻪ ﻧﻈﺮ ﻣﻲ رﺳﺪ‪ .‬اﻳﻦ اﻗﺪاﻣﺎت ﻣﻲ ﺗﻮاﻧﺪ ﺗﻐﻴﻴﺮات ﺳﺎزه اي‪ ،‬ﺑﺮﻧﺎﻣﻪ رﻳﺰي ﺑﺮاي ﻣﺪﻳﺮﻳﺖ رژﻳﻢ ﺗﺸﺪﻳﺪ ﺷﺪه‬
‫ﺳﻴﻼب در ﻗﺴﻤﺖ ﻣﻴﺎﻧﻲ ﺣﻮﺿﻪ ﺑﻮﺳﻴﻠﻪ ﻣﻌﻜﻮس ﻛﺮدن روﻧﺪ ﺗﺨﺮﻳﺐ ﺣﻮﺿﻪ و ﻛﻢ ﻛﺮدن ﺑﺮداﺷﺖ آب در ﻣﺎﻫﻬﺎي ﻛﻢ آب‬
‫‪Summary in Persian‬‬
‫‪155‬‬
‫____________________________________________________________________‬
‫)ﻣﻲ ﺗﺎ ﺳﭙﺘﺎﻣﺒﺮ( درﺑﺎﻻدﺳﺖ ﺣﻮﺿﻪ ﺑﺎﺷﺪ‪ .‬ارزﻳﺎﺑﻲ ﻫﺎي اﻧﺠﺎم ﺷﺪه در اﻳﻦ ﭘﮋوﻫﺶ ﻧﺸﺎن ﻣﻲ دﻫﺪ ﻛﻪ ﺗﺮوﻳﺞ اﺳﺘﻔﺎده از آب در‬
‫ﻛﺸﺎورزي ﺑﻪ ﺟﺎي ﻛﺸﺎورزي دﻳﻢ در اﻳﻦ ﺣﻮﺿﻪ ﺑﺎ در ﻧﻈﺮﮔﺮﻓﺘﻦ اﻳﻦ ﻧﻜﺘﻪ ﻛﻪ ﺣﺪاﻗﻞ ﺗﺎﺛﻴﺮ را در ﻣﻴﺰان آب در دﺳﺘﺮس‬
‫درﭘﺎﻳﻴﻦ دﺳﺖ ﺣﻮﺿﻪ داﺷﺘﻪ ﺑﺎﺷﺪ‪ ،‬اﻣﻜﺎن ﭘﺬﻳﺮ اﺳﺖ‪ .‬ﺑﺎ اﻳﻦ وﺟﻮد ﺗﺒﺪﻳﻞ ﻣﺰارع دﻳﻢ ﺑﻪ ﻣﺰارع آﺑﻲ ﺑﺎﻳﺪ ﻣﺤﺘﺎﻃﺎﻧﻪ اﻧﺠﺎم ﺷﻮد ﺑﻪ‬
‫ﺷﻜﻠﻲ ﻛﻪ ﻛﺎﻫﺶ ﺟﺮﻳﺎن ﻣﺎﻫﺎﻧﻪ ﺑﻪ ﻣﻴﺰان ﻣﻌﻘﻮﻟﻲ اﺗﻔﺎق ﺑﻴﺎﻓﺘﺪ‪ .‬ﺑﺮاي اﻳﻦ ﻣﻨﻈﻮر ﺳﻄﺢ ﻣﺤﺪودي از ﻣﺰارع )ﺑﻪ ﺧﺼﻮص در‬
‫ﺑﺎﻻدﺳﺖ ﺣﻮﺿﻪ( ﺑﺎﻳﺪ ﺑﻪ ﻣﺰارع آﺑﻲ ﺗﺒﺪﻳﻞ ﺷﻮد‪ .‬ﻫﻤﭽﻨﻴﻦ در اﻳﻦ ﻣﺰارع ﺑﺎﻳﺪ از ﺗﻜﻨﻴﻚ ﻛﻢ آﺑﻴﺎري اﺳﺘﻔﺎده ﺷﻮد و ﻃﻴﻒ‬
‫ﮔﺴﺘﺮده اي از ﮔﺰﻳﻨﻪ ﻫﺎي ذﺧﻴﺮه ﺳﺎزي آب ﻣﻮرد ﺗﻮﺟﻪ ﻗﺮار ﮔﻴﺮد‪ .‬ﭘﻴﺸﻨﻬﺎد ﻣﻲ ﺷﻮد ﻛﻪ ﺷﺒﻜﻪ اﻧﺪازه ﮔﻴﺮي داده ﻫﺎي ﺟﻮي‬
‫و ﻫﻴﺪروﻟﻴﻜﻲ ﺑﻬﺴﺎزي ﺷﻮد ﻛﻪ ﺑﺎﻋﺚ اﻓﺰاﻳﺶ اﻃﻼﻋﺎت ﻣﻮﺟﻮد و در ﻧﺘﻴﺠﻪ ﺑﻬﺒﻮد ﻣﺪل ﻫﺎي ﻫﻴﺪروﻟﻮژﻳﻜﻲ و ﻣﻨﺎﺑﻊ آب ﻣﻲ‬
‫ﺷﻮد ﻛﻪ ﺑﻪ ﻧﻮﺑﻪ ﺧﻮد ﻣﻮﺟﺐ ﺗﺼﻤﻴﻢ ﮔﻴﺮي ﻫﺎي ﺻﺤﻴﺢ ﺗﺮ ﺧﻮاﻫﺪ ﺑﻮد‪ .‬در اﻳﻦ راﺑﻄﻪ ﺑﻪ ﻛﺎر اﻧﺪازي اﻳﺴﺘﮕﺎه ﻫﺎي ﻣﺘﺮوك و‬
‫اﻓﺰاﻳﺶ اﻳﺴﺘﮕﺎﻫﻬﺎي اﻧﺪازه ﮔﻴﺮي در ﻣﻨﺎﻃﻖ ﻛﻮﻫﺴﺘﺎﻧﻲ ﭘﻴﺸﻨﻬﺎد ﻣﻲ ﮔﺮدد‪ .‬ﺑﺮﻧﺎﻣﻪ رﻳﺰي و ﻣﺪﻳﺮﻳﺖ ﺗﻤﺎم اﻧﻮاع ﻣﻨﺎﺑﻊ آب در‬
‫ﺳﻄﺢ ﺣﻮﺿﻪ آب رﻳﺰ رودﺧﺎﻧﻪ ﺑﺎﻳﺪ ﺗﺮوﻳﺞ ﺷﻮد‪.‬‬
‫ﺑﻪ ﻃﻮر ﻛﻠﻲ داﻧﺶ اﻳﺠﺎد ﺷﺪه در ﻣﻄﺎﻟﻌﻪ ﻣﻮردي اﻳﻦ ﭘﮋوﻫﺶ ﺑﺮاي ﺳﺎﻳﺮ ﺣﻮﺿﻪ ﻫﺎي آﺑﺮﻳﺰ اﻳﺮان و ﺟﻬﺎن ﺗﺎ ﺣﺪ ﺑﺴﻴﺎر زﻳﺎدي‬
‫ﻗﺎﺑﻞ اﺳﺘﻔﺎده اﺳﺖ‪.‬‬
REFERENCES
Abbaspour, K.C.; Johnson, C.A.; Genuchten, M.T. 2004. Estimating uncertain flow
and transport parameters using a sequential uncertainty fitting procedure.
Vadose Zone Journal 3(4): 1340-1352.
Abbaspour, K.C.; Yang, J.; Maximov, I.; Siber, R.; Bogner, K.; Mieleitner, J.;
Zobrist, J.; Srinivasan, R. 2007. Modelling hydrology and water quality in
the pre-Alpine/Alpine Thur Watershed using SWAT. Journal of Hydrology
333: 413-430.
Abbaspour, K.C. 2008. SWAT-CUP2: SWAT calibration and uncertainty programs A user manual. Department of Systems Analysis, Integrated Assessment
and Modelling (SIAM). Duebendorf, Switzerland: Eawag, Swiss Federal
Institute of Aquatic Science and Technology.
Abbaspour, K.C.; Faramarzi, M.; Ghasemi, S.S.; Yang, H. 2009. Assessing the
impact of climate change on water resources in Iran. Water Resources
Research 45: W10434, doi:10.1029/2008WR007615, 2009
Abdulla, F.A.; Lettenmaier, D.P. 1997. Development of regional parameter
estimation equations for a macro scale hydrologic model. Journal of
Hydrology 197: 230-257.
Absalan, S.; Heydari, N.; Abbasi, F.; Farahani, H.; Siadat, H.; Oweis, T. 2007.
Determination and evaluation of water productivity in the saline areas of
lower Karkheh River Basin (KRB), Iran. In: Extended abstracts.
International workshop on improving water productivity and livelihood
resilience in Karkheh River Basin, Ghafouri, M. (ed.), Tehran, Iran: Soil
Conservation and Watershed Management Research Institutes (SCWNRI),
September 10–11, 2007.
Ahmad, M.D.; Islam, Md. A.; Masih, I.; Muthuwatta, L.P.; Karimi, P.; Turral, H.
2009. Mapping basin-level water productivity using remote sensing and
secondary data in the Karkheh River Basin, Iran. Water International 34
(1): 119-133.
Ahmad, M.D.; Giordano, M. 2010. The Karkheh River basin: The food basket of
Iran under pressure. Water International 35(5): 522-544.
Alcamo, J.; Henrichs, T.; Rosch, T. 2000. World water in 2025: Global modeling
and scenario analysis. In: World water scenarios analyses, ed., Rijsberman
Marseille: World Water Council.
Allen, R.G.; Pereira, L.S.; Raes, D.; Smith, M. 1998. Crop evapotranspiration:
Guidelines for computing crop water requirements. FAO Irrigation and
Drainage Paper No. 56. Rome, Italy: FAO, Water Resources, Development
and Management Service.
Alijani, B. 2002. Variations of 500 hPa flow patterns over Iran and surrounding
areas and their relationship with the climate of Iran. Theoretical and
Applied Climatology 72: 41-54.
158
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Alijani, B.; Ghohroudi, M.; Arabi, N. 2008. Developing a climate model for Iran
using GIS. Theoretical and Applied Climatology 92: 103-112.
Ardakanian, R. 2005. Overview of water management in Iran. In: Water
conservation, reuse, and recycling, Proceeding of an Iranian American
workshop, pp. 153-172. Washington, DC: The National Academies Press.
Arnell, N.W. 1999. The effect of climate change on hydrological regimes in Europe:
A continental perspective. Global Environmental Change 9: 5-33.
Arnold, J.G.; Srinivasan, R.; Muttiah, R.S.; Williams, J.R. 1998. Large area
hydrologic modeling and assessment part 1: Model development. Journal
of the American Water Resources Association 34(1): 73-89.
Arnold, J.G.; Allen, P.M. 1999. Validation of automated methods for estimating
baseflow and groundwater recharge from stream flow records, Journal of
the American Water Resources Association 35 (2): 411–424.
Arnold, J.G.; Fohrer, N. 2005. SWAT2000: Current capabilities and research
opportunities in applied watershed modeling. Hydrological Processes 19:
563-572.
ASCE Task Committee on Definition of Criteria for Evaluation of Watershed
Models of the Watershed Management Committee Irrigation and Drainage
Division. 1993. Criteria for evaluation of watershed models. Journal of
Irrigation and Drainage Engineering 119 (3): 429-442.
Ashrafi, S.; Qureshi, A.S.; Gichuki, F. 2004. Karkheh Basin profile: Strategic
research for enhancing agricultural water productivity. Draft Report.
Challenge Program on Water and Food. (Duplicated).
Balonishnikova, J.; Heal, K.; Guobin, F.; Karambiri, H.; Oki, T. 2006. World water
resources, water use and management. Chapter 2, Hydrology 2020: An
integrating science to meet world water challenges, ed. Oki, T.; Valeo, C.;
Heal, K. IAHS Publication 300. Wallingford, UK: IAHS Press.
Beniston, M. 2003. Climate change in mountain regions: A review of possible
impacts. Climatic Change 59:5-31.
Bergström, S. 1992. The HBV model—Its structure and applications. SMHI, no. 4.
Norrköping, Sweden.
Beven, K.J. 2001. Rainfall-runoff modelling: The primer. Chichester: John Wiley &
Sons.
Birsan, M.V.; Molnar, P.; Burlando, P.; Pfaundler, M. 2005. Streamflow trends in
Switzerland. Journal of Hydrology 314:312-329.
Blöschl, G. 2006. Hydrologic synthesis: Across processes, places, and scales. Water
Resources Research 42: W03S02, doi:10.1029/2005WR004319.
Blöschl, G.; Sivapalan, M. 1995. Scale issues in hydrological modelling: A review.
Hydrological Processes 9 (3-4): 251-290.
Bouraoui, F.; Grizzetti, B.; Granlund, K.; Rekolainen, S.; Bidoglio, G. 2004. Impact
of climate change on the water cycle and nutrient losses in a Finish
catchment. Climatic Change 66:109-126.
Bouwer, H. 2000. Integrated water management: Emerging issues and challenges.
Agricultural Water Management 45: 217-228
References
159
____________________________________________________________________
Burn, D.H.; Elnur, M.A.H. 2002. Detection of hydrologic trends and variability.
Journal of Hydrology 255: 107-122.
Castellarin, A.; Galeati, G.; Brandimarte, L.; Montanari, A.; Brath, A. 2004.
Regional flow–duration curves: Reliability for ungauged basins. Advances
in Water Resources 27: 953-965.
CAWMA (Comprehensive Assessment of Water Management in Agriculture).
(2007). Water for food, water for life: A comprehensive assessment of
water management in agriculture. Colombo, Sri Lanka: International Water
Management Institute (IWMI); London, UK: Earthscan.
Ceballos-Barbancho, A.; Morán-Tejeda, E.; Luengo-Ugidos, M.Á.; Llorente-Pinto,
J.M. 2008. water resources and environmental change in a Mediterranean
environment: The south west sector of the Duero River Basin (Spain).
Journal of Hydrology 351: 126-138.
CPWF (Challenge Program on Water and Food). 2002. Challenge program on water
and food: Full proposal. Challenge Program on Water and Food (CPWF).
http://www.waterandfood.org/
CPWF. 2003. Karkheh River Basin: Protecting dry land under environmental siege.
Karkheh Basin Brochure. Challenge Program on Water and Food (CPWF).
http://www.waterandfood.org/
CPWF. 2005. Basin focal project: Karkheh proposal. Challenge Program on Water
and Food (CPWF). http://www.waterandfood.org/
Chang, M.; Lee, R. 1974. Objective Double-Mass Analysis. Water Resources
Research 10 (6): 1123-1126.
Chaplot, V. 2005. Impact of DEM mesh size and soil map scale on SWAT runoff,
sediment, and NO3–N loads predictions. Journal of Hydrology 312: 207222.
Chaplot, V.; Saleh, A.; Jaynes, D.B. 2005. Effect of the accuracy of spatial rainfall
information on the modeling of water, sediment, and NO3–N loads at the
watershed level. Journal of Hydrology 312: 223-234.
Cho, H.; Olivera, F. 2009. Effect of the spatial variability of land use, soil type, and
precipitation on streamflows on small watersheds. Journal of the American
Water Resources Association 45(3): 673-686.
Christensen, J. H.; Hewitson, B.; Busuioc, A.; Chen, A.; Gao, X.; Held, I.; Jones, R.;
Kolli, R.K.; Kwon, W.-T.; Laprise, R.; Magaña Rueda, V.; Mearns, L.;
Menéndez, C.G.; Räisänen, J.; Rinke, A.; Sarr, A.; Whetton, P. 2007.
Regional climate projections. In: Climate Change 2007: The Physical
Science Basis. Contribution of working group I to the fourth assessment
report of the Intergovernmental Panel on Climate Change, pp. 847-940, ed.
Solomon, S.; Qin, D.; Manning, M.; Chen, Z.; Marquis, M.; Averyt, K.B.;
Tingnor, M.; Miller, H.L. Cambridge, UK: Cambridge University Press.
Cosgrove, W.J.; Rijsberman, F.R. 2000a. Challenge for the 21st century: Making
water everybody's business. Sustainable Development International 2: 149156.
Cosgrove, W.J.; Rijsberman, F.R. 2000b. World water vision: making water
everybody’s business. London, UK: Earthscan Publications.
160
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Cudennec, C.; Leduc, C.; Koutsoyiannis, D. 2007. Dryland hydrology in
Mediterranean regions-A review. Hydrological Sciences Journal 52 (6):
1077-1087.
Cullen, H.M.; Kaplan, A.; Arkin, P.A.; Demenocal, P.B. 2002. Impact of the North
Atlantic oscillation on the Middle Eastern climate and streamflow. Climatic
Change 55: 315-338.
Daly, C.; Gibson, W.P.; Taylor, G H.; Johnson, G.L.; Pasteris, P. 2002. A
knowledge-based approach to the statistical mapping of climate. Climate
Research 22: 99-113.
De Fries, R.; Asner, G.P.; Houghton, R., eds. 2004. Ecosystems and land use
change. Geophys. Monogr. Ser. 153: 308, Washington, DC: American
Geophysical Union.
Domroes, M.; Kaviani, M.; Schaefer, D. 1998. An analysis of regional and intraannual precipitation variability over Iran using multivariate statistical
methods. Theoretical and Applied Climatology 61: 151-159.
DIALOGUE. 2002. DIALOGUE on water, food and environment: Proposal.
Dialogue Consortium: FAO, GWP, ICID, IFAP, IWMI, IUCN, UNEP,
WHO,WWC and WWF. http://www.cgiar.org/iwmi/dialogue/dialogue.htm.
Dinpashoh, Y.; Fakheri-Fard, A.; Moghaddam, M.; Jahanbakhsh, S.; Mirnia, M.
2004. Selection of variables for the purpose of regionalization of Iran’s
precipitation climate using multivariate methods. Journal of Hydrology
297: 109-123.
Dixon, B.; Earls, J. 2009. Resample or not? Effects of resolution of DEMs in
watershed modeling. Hydrological Processes 23: 1714-1724.
Döll, P. 2002. Impact of climate change and variability on irrigation requirements: A
global perspective. Climatic Change 54: 269-293.
Douglas, E.M.; Vogel, R.M.; Kroll, C.N. 2000. Trends in floods and low flows in
the United States: Impact of spatial correlation. Journal of Hydrology 240:
90-105.
Evans, J.P.; Smith, R.B.; Oglesby, R.J. 2004. Middle East climate simulation and
dominant precipitation processes. International Journal of Climatology 24:
1671-1694.
Falkenmark, M.; Lundquist, J.; Widstrand, C. 1989. Macro-scale water scarcity
requires micro-scale approaches: Aspects of vulnerability in semi-arid
development. Natural Resources Forum 13(4): 258-267.
Falkenmark, M.; Rockström, J. 2006. The new blue and green water paradigm:
Breaking new ground for water resources planning and management.
Editorial. Journal of Water Resources Planning and Management 132(3):
129-132.
Falkenmark, M.; Molden, D. 2008. Wake up to realities of river basin closure. Water
Resources Development 24(2): 201-15.
Faramarzi, M.; Abbaspour, K.C.; Schulin, R.; Yang, H. 2009. Modelling blue and
green water resources availability in Iran. Hydrological Processes 23: 486501.
References
161
____________________________________________________________________
Faramarzi, M.; Yang, H.; Schulin, R.; Abbaspour, K.C. 2010. Modeling wheat yield
and crop water productivity in Iran: Implications of agricultural water
management for wheat production. Agricultural Water Management 97:
1861-1875.
Faurès, J.M.; Goodrich, D.C.; Woolhiser, D.A.; Sorooshian, S. 1995. Impact of
small-scale spatial rainfall variability on runoff modelling. Journal of
Hydrology 173: 309–326.
Ficklin, D.L.; Luo, Y.; Luedeling, E.; Zhang, M. 2009. Climate change sensitivity
assessment of a highly agricultural watershed using SWAT. Journal of
Hydrology 374: 16-29.
FAO (Food and Agriculture Organization of the United Nations. 1995. The Digital
Soil Map of the World and Derived Soil Properties. CD-ROM, Version 3.5.
Rome, Italy: FAO. http://www.fao.org/
FAO. 1997. Iran water resources. Rome, Italy: FAO.
http://www.fao.org/ag/agl/aglw/aquastat/countries/iran/index.stm
FAO. 2006. AQUASTAT-FAO’s information system on water and agriculture.
Rome, Italy: FAO. http://www.fao.org/
FAO. 2009. AQUASTAT-FAO’s information system on water and agriculture.
Rome, Italy: FAO. http://www.fao.org/
Foltze, R.C. 2002. Iran’s water crisis: Cultural, political and ethical dimensions.
Journal of Agricultural and Environmental Ethics 15: 357-380.
Fontaine, T.A.; Cruickshank, T.S.; Arnold, J.G.; Hotchkiss, R.H. 2002.
Development of a snowfall-snowmelt routine for mountainous terrain for
the Soil Water Assessment Tool (SWAT). Journal of Hydrology 262: 209223.
Gallart, F.; Latron, J.; llorens, P.; Beven, K. 2006. Using internal catchment
information to reduce the uncertainty of discharge and baseflow
predictions,
Advances
in
Water
Resources,
doi:10.1016/j.advwatres.2006.06.005.
Gallopin, G.C.; Rijsberman, F.R. 2000. Three global water scenarios. International
Journal of Water 1(1): 16-40.
Ganji, A.; Khalili, D.; Karamouz, M. 2007. Development of stochastic dynamic
Nash game model for reservoir operation. I. The symmetric stochastic
model with perfect information. Advances in Water Resources 30: 528-542.
Gassman, P.W.; Reyes, M.R.; Green, C.H.; Arnold, J.G. 2007. The soil water
assessment tool: Historical development, applications, and future research
directions. Transactions of the ASABE 50(4): 1211-1250.
George, S.S. 2007. Streamflow in the Winnipeg River Basin, Canada: Trends,
extremes and climate linkages. Journal of Hydrology 332: 396-411.
Ghafouri, M.; Turkleboom, F.; Sarreshtehdhari, A. 2007. Integrated catchments
management – Lessons learned and Iranian vision. In: Extended abstracts.
International workshop on improving water productivity and livelihood
resilience in Karkheh River Basin, Ghafouri, M. (ed.) Tehran, Iran: Soil
Conservation and Watershed Management Research Institutes (SCWNRI),
September 10–11, 2007.
162
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Githui, F.; Mutua, F.; Bauwens, W. 2009a. Estimating the impacts of land-cover
change on runoff using the soil and water assessment tool (SWAT): case
study of Nzoia catchment, Kenya. Hydrological Sciences Journal 54(5):
899-908.
Githui, F.; Gitau, W.; Mutua, F.; Bauwens, W. 2009b. Climate change impact on
SWAT simulated streamflow in Western Kenya. International Journal of
Climatology 29(12): 1623-1834. DOI 10.1002/joc.1828.
Gleick, P. 2003. Global freshwater resources: Soft path solutions for the 21st
century. Science 302: 1524-1528.
Goswami, M.; O’Connor, K.M.; Bhattarai, K.P. 2007. Development of
regionalization procedures using a multi-model approach for flow
simulation in an ungauged catchment. Journal of Hydrology 333: 517-531.
Götzinger, J.; Bárdossy, A. 2007. Comparison of four regionalization methods for a
distributed hydrologic model. Journal of Hydrology 333, 374-384.
Green, G.P.; Hamilton, J.R. 2000. Water allocation, transfers and conservation: links
between policy and hydrology. Water Resources Development 16(2): 197208.
Guo, H.; Hu, Q.; Jian, T. 2008. Annual and seasonal streamflow responses to climate
and land-cover changes in the Poyang Lake Basin, China. Journal of
Hydrology 355: 106-122.
Gupta, A.D. 2008. Implications of environmental flows in river basin management.
Physics and Chemistry of the Earth 33 (5): 298-303.
Gupta, H.V.; Kling, H.; Yilmaz, K.K.; Martinez, G.F. 2009. Decomposition of the
mean squared error and NSE performance criteria: Implications for
improving hydrological modelling. Journal of Hydrology 377: 80-91.
Hargreaves, G.L.; Hargreaves, G.H.; Riley, J.P. 1985. Agricultural benefits for
Senegal River Basin. Journal of Irrigation and Drainage Engineering 111:
113-124.
Heydari, N. 2006. Water and irrigation management in the water-stressed ZayandehRud and Karkheh River Basins, Islamic Republic of Iran. In proceedings of
the International Forum on water resources management and irrigation
modernization in Shanxi Province, China, 22-24 November, 2006, Taiyuan,
China.
Hughes, D.A.; Smakhtin, V.Y. 1996. Daily flow time series patching or extension: A
spatial interpolation approach based on FDCs. Hydrological Sciences
Journal 41(6): 851–71.
Ibrahim, A.B.; Cordery, I. 1995. Estimation of recharge and runoff volumes from
ungauged catchments in eastern Australia. Hydrological Sciences Journal
40: 499-515.
IWMI. 2005. Annual Report 2004/2005. Colombo, Sri Lanka, International Water
Management Institute (IWMI).
JAMAB. 1999. Comprehensive Assessment of National Water Resources: Karkheh
River Basin. JAMAB Consulting Engineers in association with Ministry of
Energy, Iran. (In Persian).
References
163
____________________________________________________________________
JAMAB. 2006a. A report on the development of ground waters and utilization
thereof in Karkheh Drainage Basin: An analysis of specifications. JAMAB
Consulting Engineers in association with Ministry of Energy, Iran.
JAMAB 2006b. Water balance report of Karkheh River Basin area: Preliminary
analysis. JAMAB Consulting Engineers in association with Ministry of
Energy, Iran.
Jayakrishnan, R.; Srinivasan, R.; Santhi, C.; Arnold, J.G. 2005. Advances in the
application of the SWAT model for water resources management.
Hydrological Processes 19: 749-762.
Jha, M.; Gassman, P.W.; Secchi, S.; Gu, R.; Arnold, J.G. 2004. Effect of watershed
subdivision on SWAT flow, sediment, and nutrient predictions. Journal of
the American Water Resources Association 40(3): 811-825.
Jones, C.; Sultan, M.; Yan, E.; Milewski, A.; Hussein, M.; Al-Dousari, A.; Al-kaisy,
S.; Becker, R. 2008. Hydrologic impacts of engineering projects on the
Tigris-Euphrates systems and its marshlands. Journal of Hydrology 353:
59-75.
Karamouz, M.; Zahraie, B.; Araghi-Nejhad, Sh.; Shahsavari, M.; Torabi, S. 2001.
An integrated approach to water resources development of the Tehran
region in Iran. Journal of the American Water Resources Association 37
(5): 1301-1311.
Karamouz, M.; Zaharie, B.; Khodatalab, N. 2003. Reservoir optimization: A nonstructural solution for control of seepage from Lar Reservoir in Iran. Water
International 28(1): 19-26.
Karamouz, M.; Akhbari, M.; Moridi, A.; Kerachian, R. 2006. A system dynamicsbased conflict resolution model for river water quality management.
Iranian Journal of Environment and Health Science and Engineering 3 (3):
147-160.
Karamouz, M.; Moridi, A.; Fayyazi, H.M. 2008. Dealing with conflict over water
quality and quantity allocation: A case study. Scientia Iranica 15(1): 34-49.
Karamouz, M.; Akhbari, M.; Moridi, A. 2011. Resolving disputes over reservoirriver operations. Journal of Irrigation and Drainage Engineering 137(5):
327-339.
Keller, A.; Keller, J.; Seckler, D. 1996. Integrated water resource systems: Theory
and policy implications. Research Report 3. Colombo, Sri Lanka:
International Water Management Institute.
Keller, J.; Keller, A.; Davids, G. 1998. River basin development phases and
implications of closure. Journal of Applied Irrigation Science 33(2): 145164.
Kerachian, R.; Karmouz, M. 2007. A stochastic conflict resolution model for water
quality management in reservoir-river systems. Advances in Water
Resources 30: 866-882.
Keshavarz, A.; Asadollahi, A.; Pazira, E. 2007. Assessment of agricultural water
policies, laws, and regulations influencing water productivity across KRB.
In: Extended abstracts. International workshop on improving water
productivity and livelihood resilience in Karkheh River Basin, Ghafouri, M.
164
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
(ed.) Tehran, Iran: Soil Conservation and Watershed Management Research
Institutes (SCWNRI), September 10–11, 2007.
Kokkonen, T.S.; Jakeman, A.J.; Young, P.C.; Koivusalo, H.J. 2003. Predicting daily
flows in ungauged catchments: Model regionalization from catchment
descriptors at the Coweeta Hydrologic Laboratory, North Carolina.
Hydrological Processes 17: 2219–2238.
Krause, P.; Boyle, D.P.; Base, F. 2005. Comparison of different efficiency criteria
for hydrological model assessment. Advances in Geosciences 5: 89-97.
Lacombe, G.; Cappelaere, B.; Leduc, C. 2008. Hydrological impact of water and soil
conservation works in the Merguellil catchment of Central Tunisia. Journal
of Hydrology 359: 210-224.
Larocque, M.; Fortin, V.; Pharand, M.C.; Rivard, C. 2010. Groundwater contribution
to river flows using hydrograph separation, hydrological models in a
southern Quebec aquifer. Hydrology and Earth System Sciences Discussion
7: 7809-7838.
Lettenmaier, D.P.; Wood, E.F.; Wallis, J.R. 1994. Hydro-climatological trends in the
Continental United States, 1948-88. Journal of Climate 7: 586-607.
Lidén, R.; Harlin, J. 2000. Analysis of conceptual rainfall-runoff modelling
performance in different climates. Journal of Hydrology 238: 231-247.
Linsley, R.K.; Kohler, M.A.; Paulhus, J.L.H. 1949. Applied Hydrology. McGrawHill, New York.
Linsley, R.K.; Kohler, M.A.; Paulhus, J.L.H. 1982. Hydrology for engineers (3rd ed).
McGraw-Hill series in Water Resources and Environmental Engineering,
McGraw-Hill Inc., New York.
Lim, J.K.; Engel, B.A.; Tang, Z.; Kim, J.C. 2005. Automated web GIS based
hydrograph analysis tool, WHAT. Journal of the American Water
Resources Association 41(6): 1407-1416.
Loucks, D.P. 2006. Modeling and managing the interactions between hydrology,
ecology and economics. Journal of Hydrology 328: 408- 416.
Love, D.; Uhlenbrook, S.; Corzo-Perez, G.; Twomlow, S. 2010. Rainfallinterception-evaporation-runoff relationships in a semi-arid mesocatchment, northern Limpopo Basin, Zimbabwe. Hydrological Sciences
Journal 55 (5): 687-703.
Magette, W.L.; Shanholtz, V.O.; Carr, J.C. 1976. Estimating selected parameters for
the Kentucky watershed model from watershed characteristics. Water
Resources Research 1: 472-476
Makurira, H.; Savenije, H.H.G.; Uhlenbrook, S. 2010. Modeling field scale water
partitioning using on-site observations in sub-Saharan rainfed agriculture.
Hydrology and Earth System Sciences 14, 627-638.
Mann, M.E. 2002. Large-scale climate variability and connections with the Middle
East in the past centuries. Climatic Change 55: 287-314.
Manouchehri, G.R.; Mahmoodian, S.A. 2002. Environmental Impacts of dams
constructed in Iran. International Journal of Water Resources Development
18(1): 179-182.
References
165
____________________________________________________________________
Marjanizadeh, S. 2008. Developing a “best case scenario”for Karkheh River Basin
management (2025 horizon); a case study from Karkheh River Basin, Iran.
PhD dissertation, Department of Water, Atmosphere and Environment,
University of Natural Resources and Applied Life Sciences, Vienna,
Austria.
Marjanizadeh, S.; Qureshi, A.S.; Turral, H.; Talebzadeh, P. 2009. From
Mesopotamia to the third millennium: The historical trajectory of water
development and use in the Karkheh River Basin, Iran. IWMI Working
Paper 135. Colombo, Sri Lanka: International Water Management Institute.
. doi:10.3910/2010.206.
Marjanizadeh, S.; de Fraiture, C.; Loiskandl, W. 2010. Food and water scenarios for
the Karkheh River Basin, Iran. Water International 35 (4): 409-424.
Masih, I.; Ahmad, M.D.; Turral, H.; Uhlenbrook, S.; Karimi, P. 2009. Analysing
streamflow variability and water allocation for sustainable management of
water resources in the semi-arid Karkheh River Basin, Iran. Physics and
Chemistry of the Earth 34 (4-5): 329–340.
Masih, I.; Uhlenbrook, S.; Maskey, S.; Ahmad, M.D. 2010. Regionalization of a
conceptual rainfall-runoff model based on similarity of the flow duration
curve: A case study from the semi-arid Karkheh basin, Iran, Journal of
Hydrology 391: 188-201. DOI:10.1016/j.jhydrol.2010.07.018
Masih, I.; Maskey, S.; Uhlenbrook, S.; Smakhtin, V. 2011a. Assessing the impact of
areal precipitation input on streamflow simulations using the SWAT model.
Journal of the American Water Resources Association .(JAWRA)
47(1):179-195. DOI: 10.1111/j.1752-1688.2010.00502.x.
Masih, I.; Uhlenbrook, S.; Maskey, S.; Smakhtin, V. 2011b. Streamflow trends and
climate linkages in the Zagros Mountain, Iran. Climatic Change 104: 317338. DOI 10.1007/s10584-009-9793-x.
Masih, I.; Maskey, S.; Uhlenbrook, S.; Smakhtin, V. 2011c. Quantifying scaledependent impacts of upgrading rain-fed agriculture in a semi-arid basin.
Agricultural Water Management (in review).
Maskey, S.; Guinot, V.; Price, R.K. 2004. Treatment of precipitation uncertainty in
rainfall-runoff modelling: A fuzzy set approach. Advances in Water
Resources 27: 889-898.
Maskey, S. 2007. HyKit - A tool for grid based interpolation of hydrological
variables, user's guide. Delft, the Netherlands: UNESCO-IHE Institute for
Water Education.
Mathews, E. 1983. Global vegetation and land use: new high-resolution data bases
for climate studies, Journal of Climate and Applied Meteorology 22: 474–
487.
McCartney, M.; Smakhtin, V. 2010. Water storage in an era of climate change:
Addressing the challenge of increasing rainfall variability. Blue Paper.
International Water Management Institute, Colombo, Sri Lanka.
McCuen, R.H. 2003. Modelling hydrologic change: Statistical methods. Boca Raton,
Florida: Lewis Publishers.
166
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
McIntyre, N.; Lee, H.; Wheater, H.S.; Young, A.; Wagener, T. 2005. Ensemble
Predictions of runoff in ungauged catchments. Water Resources Research,
41: W12434, doi:10.1029/2005WR004289.
Merz, B.; Blöschl, G. 2004. Regionalization of catchment model parameters.
Journal of Hydrology 287: 95-123.
Mirqasemi, S.A.; Pauw, E. De. 2007. Land use change detection in the Karkheh
Basin, Iran by using multi-temporal satellite images. In: Extended
abstracts. International Workshop on Improving Water Productivity and
Livelihood Resilience in Karkheh River Basin, ed. Ghafouri, M. Soil
Conservation and Watershed Management Research Institutes (SCWNRI),
Tehran, Iran. September 10-11, 2007.
Mitchell, T.D.; Carter, T.R.; Jones, P.D.; Hulme, M.; New, M. 2004. A
comprehensive set of high-resolution grids of monthly climate for Europe
and the globe: The observed record (1901–2000) and 16 scenarios (2001–
2100). Working Paper 55, Tyndall Centre.
Modallaldoust, S.; Bayat, F.; Soltani, B.; Soleimani, K. 2008. Applying digital
elevation model to interpolate precipitation. Journal of Applied Sciences
8(8): 1471-1478.
Modarres, R.; da Silva, V. de P.R. 2007. Rainfall trends in arid and semi-arid
regions of Iran. Journal of Arid Environments 70: 344-355.
Molden, D. 1997. Accounting for water use and productivity. SWIM Paper 1.
Colombo, Sri Lanka: International Water Management Institute.
Molden, D.; Sakthivadivel, R. 1999. Water accounting to assess use and productivity
of water. Water Resources Development 15: 55-71.
Molle, F. 2003. Development trajectories of river basins: A conceptual framework.
Research Report 72. Colombo, Sri Lanka: International Water Management
Institute.
Molle, F.; Mamanpoush, A.; Miranzadeh, M. 2004. Robbing Yadullah’s water to
irrigate Saeid’s garden: Hydrology and water rights in a village of central
Iran. Research Report 80. Colombo, Sri Lanka: International Water
Management Institute.
Molle, F.; Turral, H. 2004. Demand management in a basin perspective: Is the
potential for water saving overrated? Paper presented at the International
Conference on Water Demand Management, June 2004, Dead Sea, Jordan.
Molle, F. 2006. Planning and managing water resources at the river-basin scale:
Emergence and evolution of a concept. IWMI Comprehensive Assessment
Research Report 16. Colombo, Sri Lanka: International Water Management
Institute.
Mousavi, S.J.; Moghaddam, K.S.; Seifi, A. 2004. Application of an interior-point
algorithm for optimization of a large-scale reservoir system. Water
Resources Management 18: 519-540.
Mul, M.; Savenije, H.H.G.; Uhlenbrook, S. 2009. Spatial rainfall variability and
runoff response during an extreme event in a semi-arid, meso-scale
catchment in the South Pare mountains, Tanzania. Hydrology and Earth
System Sciences 13: 1659–1670.
References
167
____________________________________________________________________
Mul, M.L.; Mutiibwa, R.K.; Uhlenbrook, S.; Savenije, H.H.G. 2008. Hydrograph
separation using hydrochemical tracers in the Makanya catchment,
Tanzania. Physics and Chemistry of the Earth 33 (1): 151-156.
Muthuwatta, L.P.; Ahmad, M.D.; Bos, M.G.; Rientjes, T.H.M. 2010. assessment of
water availability and consumption in the Karkheh River Basin, Iran—
Using remote sensing and geo-statistics. Water Resources Management 24:
459-484, DOI 10.1007/s11269-009-9455-9.
Nash, J.E.; Sutcliffe, J.V. 1970. River flow forecasting through conceptual models:
I. A discussion of principles. Journal of Hydrology 10: 282-290.
Nash, L.L.; Gleick, P.H. 1991. Sensitivity of streamflow in the Colorado Basin to
climatic changes. Journal of Hydrology 125: 221-241.
Nathan, R.J.; McMahon, T.A. 1990a. Evaluation of automated techniques for base
flow and recession analyses. Water Resources Research 26:1465–1473.
Nathan, R.J.; McMahon, T.A. 1990b. Identification of homogeneous regions for the
purpose of regionalization. Journal of Hydrology 121: 217-238.
Nazemosadat, M.J.; Cordery, I. 2000. On the relationships between ENSO and
autumn rainfall in Iran. International Journal of Climatology 20: 47-61.
Neitsch, S.L.; Arnold, J.G.; Kiniry, J.R.; Williams, J.R. 2005. Soil and water
assessment tool theoretical documentation: Version 2005. Texas, USA:
USDA, Soil and Water Research Laboratory/Blackland Research Center.
New, M.; Hulme, M.; Jones, P.D. 2000. Representing twentieth century space-time
climate variability part 2. Development of 1901–1996 monthly grids of
terrestrial surface climate. Journal of Climate 13: 2217–2238.
Newman, B.D.; Wilcox, B.P.; Archer, S.R.; Breshears, D.D.; Dahm, C.N.; Duffy,
C.J.; McDowell, N.G.; Phillips, F.M.; Scanlon, B.R.; Vivoni, E.R. 2006.
Ecohydrology of water-limited environments: A scientific vision. Water
Resources Research 42, W06302, doi:10.1029/2005WR004141.
Niadas, I.A.; Mentzelopoulos, P.G. 2008. Probabilistic flow duration curves for
small hydro plant design and performance evaluation. Water Resources
Management 22: 509-523
Oudin, L.; Andreassian, V.; Perrin, C.; Michel, C.; Le Moine, N. 2008. Spatial
proximity, physical similarity, regression and ungauged catchments: A
comparison of regionalization approaches based on 913 French catchments.
Water Resources Research 44: W03413, doi: 10.1029/2007WR006240.
Oudin, L.; Perrin, C.; Mathevet, T.; Andreassian, V.; Michel, C. 2006. Impact of
biased and randomly corrupted inputs on the efficiency and the parameters
of watershed models. Journal of Hydrology 320 (1-2): 62-83.
Oweis, T.; Hachum, A. 2009. Optimizing supplemental irrigation: Tradeoffs
between profitability and sustainability. Agricultural Water Management
96: 511-516.
Palmer, M.A.; Bernhardt, E.S. 2006. Hydroecology and river restoration: Ripe for
research and synthesis. Water Resources Research 42: W03S07,
doi:10.1029/2005WR004354.
168
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Parajka, J.; Merz, R.; Blöschl, G. 2005. A comparison of regionalisation methods for
catchment model parameters. Hydrology and Earth System Sciences
Discussion 2: 509-542.
Parsons, A.J.; Abrahams, A. D. 1994. Geomorphology of desert environments. In:
Geomorphology of Desert Environments, ed. Abrahams, A.D.; Parsons,
A.J. Boca Raton, Fla.: CRC Press, pp. 1-12.
Poff, N.L.; Allan, J.D.; Bain, M.B.; Karr, J.R.; Prestegaard, K.L.; Richter, B.D.;
Sparks, R.E.; Stromberg, J.C. 1997. The natural flow regime. Bioscience
47(11): 769- 84.
Postel, S. 2000. Entering an era of water scarcity: The challenges ahead. Ecological
Applications 10 (4): 941-948.
Postel, S. 2003. Securing water for people, crops, and ecosystems: new mindset and
new priorities. Natural Resource Forum 47: 89-98.
Postel, S. 2005. Liquid assets: The critical need to safeguard freshwater ecosystems.
Washington, DC: Worldwatch Institute.
Postel, S.; Richter, B. 2003. Rivers for life: Managing water for people and nature.
Washington, DC: Island Press.
Quenouille, M.H. 1956. Notes on bias in estimation. Biometrika 43: 353-360.
Qureshi, A.S.; Ahmad, M.D.; Gichuki, F.; Clemet, F.; Masih, I. 2005. Diagnostic
tour of the Karkheh River Basin: General observations and suggestions for
basin focal project priorities. Interim Project Report, Basin Focal Project.
International Water Management Institute, Karaj, Iran. (Duplicated).
Richter, B.D.; Baumgartner, J.V.; Wigington, R.; Braun, D.P. 1997. How much
water does a river need? Freshwater Biology 37: 231-249.
Rijsberman, F.R. 2006. Water scarcity: Fact or fiction? Agricultural Water
Management 80: 5-22.
Rijsberman, F.R.; Molden, D.J. 2001. Balancing water uses: Water for food and
water for nature. Thematic background papers. International Conference on
Freshwater. Bonn, 3-7 December, pp. 43-56.
Ringeltaube, J. 2002. The European Water Framework Directive-An example for
water management in national and international river basins. In: FRIEND
2002- Regional hydrology: Bridging the Gap between research and
practices, Proceedings of the fourth international FRIEND conference held
at Cape Town, South Africa, March 2002. IAHS publ. no. 274. 2002. pp.
19-25.
Rockström, J.; Karlberg, L.; Wani, S.P.; Barron, J.; Hatibu, N,; Oweis, T.;
Bruggeman, A.; Farahani, J.; Qiang, Z. 2010. Managing water in rain-fed
agriculture-the need for a paradigm shift. Agricultural Water Management
97: 543-550.
Saghafian, B.; Davtalab, R. 2007. Mapping snow characteristics based on snow
observation probability. International Journal of Climatology 27: 12771286.
Santhi, C.; Kannan, N.; Arnold, J.G.; Di Luzio, M. 2008. Spatial calibration and
temporal validation of flow for regional scale hydrologic modeling. Journal
of the American Water Resources Association 44(4):829-846.
References
169
____________________________________________________________________
Savenije, H.H.G.; van der Zaag, P. 2008. Integrated water resources management:
Concepts and issues. Physics and Chemistry of the Earth 33: 290-297.
Seckler, D. 1996. The new era of water resources management: From “dry” to
“wet” water savings. Research Report 1. Colombo, Sri Lanka: International
Irrigation Management Institute.
Seckler, D.; Amarasinghe, U.; Molden, D.J.; de Silva, R.; Barker, R. 1998. World
water demand and supply, 1990 to 2025: scenarios and issues. IWMI
Research Report 19. Colombo, Sri Lanka: International Water Management
Institute.
Seibert, J. 1999. Regionalization of parameters for a conceptual rainfall-runoff
model. Agriculture and Forest Meteorology 98-99: 279-293.
Seibert, J. 2000. Multi-criteria calibration of a conceptual rainfall-runoff model
using a genetic algorithm. Hydrology and Earth System Sciences 4 (2): 215224.
Seibert, J. 2002. HBV light version 2, user's manual. Oregon, USA: Oregon State
University Department of Forest Engineering Corvallis. Uppsala: Uppsala
University, Dept. of Earth Science, Hydrology.
Servat, E.; Dezetter, A. 1993. Rainfall-runoff modelling and water resources
assessment in northwestern Ivory Coast. Tentative extension to ungauged
catchments. Journal of Hydrology 148: 231-248.
Shiklomanov, I.A. 1999. World water resources and water use: present assessment
and outlook for 2025. St. Petersburg, Russia: State Hydrological Institute.
Shiklomanov, I.A.; Rodda, J.C. 2003. World water resources at the beginning of the
21st century. Cambridge, UK: Cambridge University Press.
Schultze, G.A. 2001. Integrated water resources management: The requirements of
the European Union, the problem of environmental impact assessment, and
implementation of the sustainable development principle. In: Integrated
water resources management, Proceedings of a symposium held at Davis,
California, April 2000. IAHS Publ. no. 272. 2001, pp. 3-11.
Sivapalan, M.; Takeuchi, K.; Franks, S.W.; Gupta, V.K.; Karambiri, H.; Lakshmi,
V.; Liang, X.; McDonnell, J.J.; Mendiondo, E.M.; O’Connell, P.E.; Oki, T.;
Pomeroy, J.W.; Schertzer, D.; Uhlenbrook, S.; Zehe, E. 2003. IAHS decade
on predictions in ungauged basins (PUB), 2003-2012: Shaping an exciting
future for the hydrological sciences. Hydrological Sciences Journal 48(6):
857-880.
Smakhtin, V.U. 2001a. Low flow hydrology: A review. Journal of Hydrology 240:
147-186
Smakhtin, V.U. 2001b. Estimating continuous monthly baseflow time series and
their possible applications in the context of the ecological reserve. Water
SA 27 (2): 213-218.
Smakhtin, V.U.; Revenga, C.; Döll, P. 2004. Taking into account environmental
water requirements in global-scale water resources assessments.
Comprehensive Assessment of Water Management in Agriculture Research
Report 2. Colombo, Sri Lanka: International Water Management Institute.
170
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Snellen, W.B.; Schrevel, A. 2004. IWRM for sustainable use of water: 50 years of
international experience with the concept of integrated water resources
management. Background document to the FAO/Netherlands Conference
on Water for Food and Ecosystems. Wageningen, Alterra, Alterra-report
1143.
Soil Conservation Service Engineering Division. 1986. Urban hydrology for small
watersheds. Technical Release 55. Washington, DC: US Department of
Agriculture.
Srinivasan, R.; Ramanarayanan, T.S.; Arnold, J.G.; Bednarz, S.T. 1998. Large area
hydrologic modeling and assessment part II: Model application. Journal of
the American Water Resources Association 34(1): 91-101.
Starks, P.J.; Moriasi, D.N. 2009. Spatial resolution effect of precipitation data on
SWAT calibration and performance: Implications for CEAP. Transactions
of the ASABE 52(4): 1171-1180.
Sullivan, C.A.; Meigh, J.R.; Simon, S.; Lawrence, P.; Calow, R.; McKenzie, A.;
Acreman, M.C.; Moore, R.V. 2000. The development of a water poverty
index. Report to DFID. Wallingford, UK: Centre for Ecology and
Hydrology.
Sutcliffe, J.V. 2004. Hydrology: A question of balance. IAHS special publication 7.
Wallingford, UK: IAHS Press.
Sutcliffe, J.V.; Carpenter, T.G. 1968. The assessment of runoff from a mountainous
and semi-arid area in western Iran. In: Hydrological Aspects of the
Utilization of Water. IAHS General Assembly of Bern, IAHS Publ. 76.
Wallingford, UK: IAHS Press, pp. 383-394.
Tennant, D.L. 1976. Instream flow regimens for fish, wildlife, recreation and related
environmental resources. Fisheries 1: 6-10.
Tetzlaff, D.; Uhlenbrook, S. 2005. Effects of spatial variability of precipitation for
process-oriented hydrological modelling: Results from two nested
catchments. Hydrology and Earth System Sciences 9: 29-41.
Tharme, R. 2003. A global perspective on environmental flow assessment:
Emerging trends in the development and application of environmental flow
methodologies for rivers. River Research and Applications 19: 397-441.
Tizro, A.T.; Voudouris, K.S.; Eini, M. 2007. Groundwater balance, safe yield and
recharge feasibility in a semi-arid environment: A case study from western
part of Iran. Journal of Applied Sciences 7(20): 2967-2976.
Tobin, K.J.; Bennett, M.E. 2009. Using SWAT to model streamflow in two river
basins with ground and satellite precipitation data. Journal of the American
Water Resources Association 45(1): 253-271.
Tripathi, M.P.; Raghuwanshi, N.S.; Rao, G.P. 2006. Effect of watershed subdivision
on simulation of water balance components. Hydrological Processes 20:
1137-1156.
Tu, M. 2006. Assessment of the effects of climate variability and land use change on
the hydrology of the Meuse River Basin. PhD thesis. UNESCO-IHE
Institute for Water Education, Delft, the Netherlands and Vrije Universiteit,
Amsterdam, the Netherlands. London, UK: Taylor & Francis Group.
References
171
____________________________________________________________________
Uhlenbrook, S. 2006. Catchment hydrology with satellites, models and rubber boots.
Inaugural Address of Stefan Uhlenbrook, Professor of Hydrology, Head of
the Hydrology and Water Resources Core at the UNESCO-IHE Institute for
Water Education in Delft, The Netherlands
Uhlenbrook, S.; Leibundgut, C. 2002. Process-oriented catchment modelling and
multiple-response validation. Hydrological Processes 16: 423-440.
Uhlenbrook, S.; Frey, M.; Leibundgut, C.; Maloszewski, P. 2002. Hydrograph
separations in a meso scale mountainous basin at event and seasonal time
scales. Water Resources Research 38(6): 1-14.
Uhlenbrook, S.; Seibert, J.; Leibundgut, C.; Rodhe, A. 1999. Prediction uncertainty
of conceptual rainfall-runoff models caused by problems to identify model
parameters and structure. Hydrological Sciences Journal 44 (5): 779-797.
Uhlenbrook, S.; Franks, S.; Heal, K.; Hubbard, S.; Karambiri, H.; Oki, T.; Valeo, C.
2006. Key messages, recommendations and concluding remarks. Chapter 8.
Hydrology 2020: An integrating science to meet world water challenges,
ed. Oki, T.; Valeo, C.; Heal, K. IAHS Publication 300. Wallingford, UK:
IAHS Press.
UNEP. 2001. Partow, H. The Mesopotamian Marshlands: Demise of an Ecosystem.
Early warning and assessment technical report, UNEP/DEWA/TR.01-3
Rev.1, Division of Early Warning and Assessment, United Nations
Environment Program, Nairobi, Kenya.
Vakili, A.; Mousavi, S.F.; Karamooz, M., eds. 1995. An overview of water resources
development in Iran. Proceedings of Regional Conference on Water
Resources Management. Conference Secretariat; Isfahan University of
Technology, Isfahan, Iran.
van der Zaag, P. 2005. Integrated water resources management: Relevant concept or
irrelevant buzzword? A capacity building and research agenda for Southern
Africa. Physics and Chemistry of the Earth 30: 867-871.
van der Zaag, P.; Gupta, J. 2008. Scale issues in the governance of water storage
projects.
Water
Resources
Research
44,
W10417,
doi:10.1029/2007WR006364.
Vandenberghe, V.; van Griensven, A.; Bauwens, W.; Vanrolleghem, P.A. 2005.
Propagation of uncertainty in diffuse pollution into water quality
predictions: application to the River Dender in Flanders, Belgium. Water
Science & Technology 51(3-4): 347-354.
Vandewiele, G.L.; Xu, C.Y.; Huybrechts, W. 1991. Regionalisation of physicallybased water balance models in Belgium: application to ungauged
catchments. Water Resources Management 5: 199-208.
van Griensven, A.; Bauwens, W. 2003. Multi-objective auto-calibration of semidistributed water quality models. Water Resources Research 39 (12): 1348,
DOI 10.1029/2003WR002284.
van Griensven, A.; Meixner, T.; Grunwald, S.; Bishop, T.; Diluzio, M.; Srinivasan,
R. 2006. A Global sensitivity analysis tool for the parameters of multivariable catchment models. Journal of Hydrology 324: 10-23.
172
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Vogel, R.M.; Fennessey, N.M. 1995. Flow duration curves. II. a review of
application in water resource planning. Water Resources Bulletin 31 (6):
1029-1039.
von Storch; Navarra, A., eds. 1995. Analysis of climate variability. New York:
Springer.
Vörösmarty, C.J.; Green, P.; Salisbury, J.; Lammers, R.B. 2000. Global water
resources: Vulnerability from climate change and population growth.
Science 289:284-288.
Wagener, T.; Wheater, H.S.; Gupta, H.V. 2004. Rainfall-runoff modelling in gauged
and ungauged catchments. London, UK: Imperial College Press.
Wagener, T.; Wheater, H.S. 2006. Parameter estimation and regionalization for
continuous rainfall-runoff models including uncertainty. Journal of
Hydrology 320: 132-54.
Wagener, T.; Sivapalan, M.; Troch, P.; Woods, R. 2007. Catchment classification
and hydrologic similarity. Geography Compass 1, 10.1111/j.17498198.2007.00039.x
Wakindiki, I.I.C.; Ben-Hur, M. 2002. Indigenous soil and water conservation
techniques: Effects on runoff, erosion, and crop yields under semi-arid
conditions. Australian Journal of Soil Research 40: 367-379.
Wallace, J. S. 2000. Increasing agricultural water efficiency to meet future food
production. Agriculture, Ecosystems and Environment 82: 105-119.
Wang, Q.J. 1991. The genetic algorithm and its application to calibrating conceptual
rainfall-runoff models. Water Resources Research 27: 2467-2471.
Watson, B.M.; Srikanthan, R.; Selvalingam, S.; Ghafouri, M. 2005. Evaluation of
three daily rainfall generation models for SWAT. Transactions of the ASAE
48(5): 1697-1711.
Weber, A.; Fohrer, N.; Möller, D. 2001. Long-term land use changes in a mesoscale
watershed due to socio-economic factors — effects on landscape structures
and functions. Ecological Modelling 140: 125-140.
Williams, J.R. 1969. Flood routing with variable travel time or variable storage
coefficients. Transactions of the ASAE 12(1):100-103.
Welderufael, W.A.; Woyessa, Y.E. 2010. Stream flow analysis and comparison of
base flow separation methods: Case study of the Modder River Basin in
Central South Africa. European Water 31: 3-12.
Winchell, M.; Srinivasan, R.; Di. Luzio, M.; Arnold, J.G. 2008. ARCSWAT 2.0
Interface for SWAT2005-user’s guide. Blackland Research Center, Texsas
Agricultural Experiment Station and Grassland, Soil and Water Research
Laboratory, USDA Agricultural Research Service Temple, Texsas.
WHO (World Health Organization). 2000. Global water supply and sanitation
assessment
2000
report.
http://www.who.int/water_sanitation_health/Globassessment/GlobalTOC.ht
m. WWAP (World Water Assessment Program). (2006). Water: A Shared
Responsibility. The United Nations World Water Development Report 2.
References
173
____________________________________________________________________
Xiubin, H.; Zhanbin, L.; Mingde, H.; Keli, T.; Fengli, Z. 2003. Down-scale analysis
of water scarcity in response to soil-water conservation on Loess Plateau of
China. Agriculture, Ecosystems and Environment 94: 355-361.
Yang, H.; Reichert, P.; Abbaspour, K.; Zehnder, A.J.B. 2003. A water resources
threshold and its implications for food security. Environmental Science and
Technology 37:3048-3054.
Yang, J.; Reichert, P.; Abbaspour, K.C.; Xia, J.; Yang, H. 2008. Comparing
uncertainty analysis techniques for a SWAT application to the Chaohe
Basin in China. Journal of Hydrology 358: 1-23.
Yadav, M.; Wagener, T.; Gupta, H. 2007. Regionalization of constraints on the
expected watershed response behavior for improved predictions in
ungauged basins. Advances in Water Resources 30: 1756-1774.
Yilmaz, K.K.; Gupta, H.V.; Wagener, T. 2008. A process-based diagnostic approach
to model evaluation: Application to the NWS distributed hydrologic model.
Water Resources Research 44: W09417, doi: 10.1029/2007WR006716.
Yue, S.; Wang, C.Y. 2002. Applicability of prewhitening to eliminate the influence
of serial correlation on the Mann-Kendall test. Water Resour. Res. 38 (6),
1068, doi:10.1029/2001WR000861.
Yue, S.; Pilon, P.; Cavadias, G. 2002. Power of the Mann-Kendall and Sprearman’s
rho Tests for detecting monotonic trends in hydrological series. Journal of
Hydrolgy 59: 254-271.
Zahraie, B.; Kerachian, R.; Malekmohammadi, B. 2008. Reservoir operation
optimization using adaptive varying chromosome length genetic algorithm.
Water International 33(3): 380-391.
Zahraie, B.; Hosseini, S.M. 2009. Development of reservoir operation policies
considering variable agricultural water demands. Expert Systems with
Applications 36: 4980-4987.
Zangvil, A.; Karas, S.; Sasson, A. 2003. Connection between eastern Mediterranean
seasonal mean 500 hPa height and sea-level pressure patterns and the
spatial rainfall distribution over Israel. International Journal of
Climatology 23: 1567-1573.
Zehnder, A.J.B.; Yang, H.; Schertenleib, R. 2003. Water issues: The need for action
at different levels. Aquatic Sciences 65: 1-20.
Zhang, X.; Harvey, K.D.; Hogg, W.D.; Yuzyk, T.R. 2001. Trends in Canadian
streamflow. Water Resources Research 37: 987-998.
LIST OF FIGURES
Figure 1. Trends in the global water withdrawals by sector of economic activity. 2
Figure 2. Water withdrawals by sector in Iran. ..................................................... 8
Figure 3. Cereal area, production and import in Iran during 1961-2007. ............ 9
Figure 4. Overview of trends in per capita availability of renewable water
resources and population growth of Iran (1961-2050). ............................................. 9
Figure 5. Location of Karkheh Basin in Iran and its hydrological and
administration units. ................................................................................................. 11
Figure 6. Digital elevation map of the Karkheh Basin and the streamflow
monitoring network................................................................................................... 13
Figure 7. Mean monthly climate of the Karkheh Basin, illustrated by precipitation
(P), temperature (T) and potential evapotranspiration (ETP) at the three climatic
stations Kermanshah, Khorramabad and Ahwaz. Data source: Meteorological
Organization of Iran. ................................................................................................ 15
Figure 8. Spatial variability of soil and land use types in the Karkheh Basin...... 16
Figure 9. Methodological framework followed in this research study ................. 23
Figure 10.
Intra-annual variability of mean daily streamflows, illustrated for the
data of the hydrological year 1962-63. .................................................................... 37
Figure 11.
View of the inter-annual variability of mean daily streamflows,
illustrated for the streamflows at Jelogir station (1961-2001). ................................ 38
Figure 12.
The flow duration curves (FDCs) of selected gauging stations. ....... 39
Figure 13. Mean monthly discharge at selected locations in the Karkheh Basin,
Iran. .......................................................................................................................... 41
Figure 14. The variability of mean monthly streamflows, indicated by the
Coefficient of Variation (CV) at selected locations in the Karkheh Basin, Iran. ...... 42
Figure 15. The reliability of the mean monthly surface water availability, indicated
by the monthly FDCs at the Paye Pole station at the Karkheh River. ...................... 43
Figure 16.
Long-term variability in annual surface water availability across the
Karkheh Basin. ......................................................................................................... 45
Figure 17.
The reliability of the annual surface water availability, indicated by
annual FDCs at the selected gauging stations across the Karkheh River system. ... 47
Figure 18.
Finger diagram presentation of the basin level water accounts of the
Karkheh river basin for the hydrological year 1993-94. .......................................... 48
Figure 19. Location of the Karkheh Basin in Iran and some of its important
features. .................................................................................................................... 55
Figure 20.
Timing of the 1-day maximum streamflow, illustrated by the records
at Pole Dokhtar. ....................................................................................................... 59
Figure 21.
Declining trend in May streamflow at Ghore Baghestan and Holilan.
.......................................................................................................... 62
Figure 22.
Increasing trends observed at Pole Dokhtar, illustrated by December,
March, 1 and 7 days maximum streamflows............................................................. 62
176
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Figure 23. The linkages of trends in streamflow in May and precipitation in April
and May, illustrated by the case of Ghore Baghestan. ............................................. 68
Figure 24. Linkages of extreme floods with precipitation, indicated by the 7-days
maximum streamflows at Pole Dokhtar and precipitation in March and February at
Khorramabad. .......................................................................................................... 70
Figure 25.
Salient features of the study area and location of the study
catchments and used climatic stations. ..................................................................... 79
Figure 26. Naturalized and observed daily time series of streamflows of Aran
catchment.................................................................................................................. 83
Figure 27. Regionalization results of the four tested methods................................ 92
Figure 28. Comparison of FDCs for the similarity analysis. .................................. 93
Figure 29. Impact of parameter uncertainty on regionalization results, illustrated
by the exceeding percentiles of Nash-Sutcliffe efficiency (NSE) obtained from the 50
parameter sets used during regionalization based on similarity in the FDC. .......... 94
Figure 30. The Karkheh basin and location of the selected streamflow gauges (a);
and the location of studied subcatchments and used climatic data stations (b). .... 100
Figure 31. Mean annual precipitation for Case II (a) and percentage difference
between Case II and Case I (b)............................................................................... 109
Figure 32.
Comparison of daily, monthly and annual precipitation among Case I
and Case II, illustrated by three selected subcatchments. ...................................... 110
Figure 33. Comparison of daily precipitation for Cases I and II for a selected month
(March 1996). ......................................................................................................... 111
Figure 34. Scatter plots of NSE and R2, highlighting the comparative performance
under cases I and II. ............................................................................................... 115
Figure 35. difference in the daily NSE and R2 in Case II as compared to Case I for
the calibration and validation periods.................................................................... 116
Figure 36. Observed and simulated daily hydrographs for Cases I and II for a
selected period January to June 1996 at three stations: (a) Jelogir, (b) Sarab seidali,
and (c) Khers Abad. ................................................................................................ 117
Figure 37. Monthly summary of the calibration and uncertainty analysis results.
(The 95PPU band is shown by thin green bars) ..................................................... 126
Figure 38. Values used in the development of S1 for the monthly potential
evapotranspiration (Epot), actual evapotranspiration (Eact), and the difference
between Epot and Eact . ............................................................................................. 129
Figure 39.
Simulated streamflows for the baseline period and the three scenarios
(S1, S2 and S3) at the basin level (Paye Pole station). Also shown in the figure are
95PPU uncertainty bands. ...................................................................................... 132
Figure 40.
Impact on monthly streamflows (May-July) due to proportion of area
upgraded from rain-fed to irrigated agriculture at the basin level. ....................... 133
Figure 41. Assessment of the prediction uncertainty of the modeling results at
annual time resolution under the three tested scenarios. ....................................... 136
Figure 42. Assessment of the prediction uncertainty of the modeling results at
monthly time resolution at the basin level. ............................................................. 136
List of Figures
177
____________________________________________________________________
Figure 43. Comparison of the estimated reference evapotranspiration (ETo) by using
the Hargreaves method and the FAO Penman-Monteith method at the Kermanshah
climatic station in the Karkheh Basin, Iran. ........................................................... 180
APPENDIX
Appendix A. Short description of the Hargreaves method and its application in
the study basin
The Hargreaves equation is commonly used for estimating reference
evapotranspiration when limited amount of climatic data is available. This empirical
method requires only temperature data to estimate. The Hargreaves equation is as
follow (Hargreaves et al. 1985).
ETo = 0.0023(Tmean + 17.8)(Tmax − Tmin )0.5 Ra
(19)
Where ETo refers to reference evapotranspiration, expressed in mm/d, Tmean, Tmax
and Tmin are daily mean, maximum and minimum air temperatures, expressed in oC,
Ra is extraterrestrial ration, expressed here in mm/d.
The results of the Hargreaves methods were compared with the FAO PenmanMonteith method (Allen et al. 1998) using daily climatic data for the period January
1987 to December 2000 for the Kermanshah climatic station (Figure 43). The results
of the both methods were found in close agreement with each other. On average, the
Hargreaves method underestimated annual total ET0 by an amount of about 5%
compared to those of the FAO Penman-Monteith method. Considering these small
differences, the Hargreaves method was considered appropriate to use in the study
basin where limited climatic data was available.
180
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
ETo (mm/d) by FAO Penman-Monteith method
16.0
14.0
y = 1.04x + 0.11
R2 = 0.87
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
ETo (m m /d) by Hargreaves m ethod
Figure 43. Comparison of the estimated reference evapotranspiration (ETo) by using
the Hargreaves method and the FAO Penman-Monteith method at the Kermanshah
climatic station in the Karkheh Basin, Iran.
ABOUT THE AUTHOR
Ilyas Masih was born in 1975 in Baddomalhi, Narowal, Pakistan. He obtained his
BSc Agricultural Engineering degree in 1997 from University of Agriculture
Faisalabad, Pakistan and completed his MPhil degree in Water Resources
Management in 2000 from Centre of Excellence in Water Resources Engineering,
University of Engineering and Technology, Lahore, Pakistan.
Ilyas has over ten years of experience as a researcher in the field of hydrology
and water resources management. He has worked at IWMI from September 2001 to
March 2011. He has worked at IWMI offices in Pakistan, Iran and Sri Lanka. During
his professional career, he has worked on wide range of issues. Most of his
undertakings involved close interaction with diversified teams of individuals
representing various professional disciplines and different cultures. He has extensive
experience on issues related to rainfall-runoff modeling at catchment to basin scales,
analysis of long term variability and trends in the climate and streamflows, water
allocation analysis and trade-offs between upstream uses and water availability for
the downstream uses including environmental flow requirements, water balance and
water productivity assessments in different agro-ecosystems, groundwater
monitoring, evaluation and sustainable management, conjunctive use of surfacewater
and groundwater resources for irrigation, secondary salinization of soils, water
savings in rice-wheat cropping systems, scale considerations in up scaling water
management interventions, and participatory water resources management. The use
of rigorous scientific methods, collection of field and secondary data, application of
analytical tools and hydrological/water management models, and synthesis of results
for formulating meaningful conclusions for scientists, policy makers and other
stakeholders are prominent features of his professional work. IWMI awarded him a
PhD research fellowship in 2006 to undertake PhD studies as an IWMI research staff
member. He was enrolled for the PhD studies in 2006 at UNESCO-IHE, Institute of
Water Education, Delft, the Netherlands. His PhD research is on the issues of basin
scale hydrology and water resources management in the Karkheh Basin, Iran.
Ilyas has co-authored a number of scientifically important and practically
relevant papers (see selected publications below) and has presented his research at
various national and international workshops and conferences.
He has recently joined UNESCO-IHE as lecturer in water resources planning.
Ilyas is married to Huma and they have a daughter Sarah Ilyas.
Selected Publications
Masih, I.; Maskey, S.; Uhlenbrook, S.; Smakhtin, V. 2011. Assessing the impact of
areal precipitation input on streamflow simulations using the SWAT model.
Journal of the American Water Resources Association .(JAWRA)
47(1):179-195. DOI: 10.1111/j.1752-1688.2010.00502.x.
182
Understanding Hydrological Variability for Improved Water Management
____________________________________________________________________
Masih, I.; Uhlenbrook, S.; Maskey, S.; Smakhtin, V. 2011. Streamflow trends and
climate linkages in the Zagros Mountain, Iran. Climatic Change 104: 317338. DOI 10.1007/s10584-009-9793-x.
Masih, I.; Uhlenbrook, S.; Maskey, S.; Ahmad, M.D. 2010. Regionalization of a
conceptual rainfall-runoff model based on similarity of the flow duration
curve: A case study from the semi-arid Karkheh basin, Iran, Journal of
Hydrology 391: 188-201. DOI:10.1016/j.jhydrol.2010.07.018.
Masih, I.; Ahmad, M.D.; Turral, H.; Uhlenbrook, S.; Karimi, P. 2009. Analysing
streamflow variability and water allocation for sustainable management of
water resources in the semi-arid Karkheh River Basin, Iran. Physics and
Chemistry of the Earth 34 (4-5): 329–340.
Ahmad, M.D.; Islam, Md. A.; Masih, I.; Muthuwatta, L. P.; Karimi, P.; Turral, H.
2009. Mapping basin-level water productivity using remote sensing and
secondary data in the Karkheh River Basin, Iran. Water International
34(1):119-133.
Ahmad, M.D.; Giordano, M.; Turral, H.; Masih, I.; Masood, Z. 2007. At what scale
does water saving really save water? Journal of Soil and Water
Conservation 62(2):29A-35A.
Qureshi, A.S.; Masih, I.; Turral, H. 2006. Comparing land and water productivities
of transplanted and direct dry seeded rice for Pakistani Punjab. Journal of
Applied Irrigation Science 41(1): 47-60.
Humphreys, E.; Meisner, E.; Gupta, R.; Timsina, J.; Beecher, H.G.; Lu, T.Y.; Sing,
Y.; Gill, M.A.; Masih, I.; Guo, Z.J.; Thompson, J.A. 2005. Water savings in
rice-wheat systems. Plant Production Science 8(3): 242-258.
Ahmad, M.D.; Masih, I.; Turral, H. 2004. Diagnostic analysis of spatial and
temporal variations in crop water productivity: A field scale analysis of the
rice-wheat cropping system of Punjab, Pakistan. Journal of Applied
Irrigation Science 39(1):43-63.
Qureshi, A.S. Asghar, M.N.; Ahmed, S.; Masih, I. 2004. Sustaining crop production
in saline groundwater areas: A case study from Pakistani Punjab.
Australian Journal of Agricultural Research 55:421-431.
Ahmad, M.D.; Turral, H.; Masih, I.; Giordano, M.; Masood, Z. 2007. Water saving
technologies:myths and realities revealed in Pakistan’s rice-wheat systems.
Research Report 108. Colombo, Sri Lanka: International Water
Management Institute.
Jehangir, W.A.; Masih, I.; Ahmed, S.; Gill, M.A.; Ahmad, M.; Mann, R.A.;
Chaudhary, M.R.; Qureshi, A.S.; Turral, H. 2007. Sustaining crop water
productivity in rice-wheat systems of South Asia: A case study from Punjab
Pakistan. IWMI Working paper 115. Colombo, Sri Lanka: International
Water Management Institute.
Qureshi, A.S.; Turral, H.; Masih, I. 2004. Strategies for the management of
conjunctive use of surface water and groundwater resources in semi-arid
areas: A case study from Pakistan. IWMI research report 86. Colombo, Sri
Lanka: International Water Management Institute.
This study provides a hydrology based assessment of (surface) water resources
and its continuum of variability and change at different spatio-temporal scales in the
semi-arid Karkheh Basin, Iran, where water is scarce, competition among users is high
and massive water resources development is under way. The study reveals that the
ongoing allocation planning is not sustainable and essentially requires reformulation,
with consideration of spatio-temporal variability and observed trends in the
streamflows regarding flood intensification and decline in low flows.
The development of innovative methods for quantification of the hydrological fluxes
(i.e., regionalization of model parameters based on similarity of the flow duration
curve and the use of areal precipitation input in the hydrological modeling) helped
better understanding and modeling the basin hydrology. The investigation of scenarios
for upgrading rain-fed areas to irrigated agriculture, using SWAT, recommends the
promotion of in-situ soil and water conservation techniques. Conversion of rain-fed
areas to irrigation causes significant reduction in the downstream flows, and requires
additional considerations such as less development in the upper catchments, practicing
supplementary irrigation and developing water storage. The knowledge generated
is instructive for hydrological assessment and its use in water resources planning and
management in the river basin context.
Was this manual useful for you? yes no
Thank you for your participation!

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

Download PDF

advertisement