Transcript The Impacts of Climate Change on Hydrology and Water Resources

The Impacts of Climate Change on Hydrology and Water Resources in Zayandeh-Rood Basin – Iran
Abrishamchi, A.1*, Azaranfar, A.1, Ghasemi, S.S. 1 Tajrishy, M1, Marino, M.A.2, Abrishamchi, A2
1. Dept. of Civil Engr., Sharif University of Technology, Azadi Ave., Tehran, Iran.
2. Dept. of Civil and Environmental Engr., University of California, Davis, One Shields Avenue, Davis, CA 95616.
Abstract
Study Area
Rising concentration of greenhouse gases may have significant consequences on the global climate. Changes in
temperature and precipitation have a serious impact on hydrologic processes and water resources. Climate model
projections show an increase in the global mean air temperature by 1 to 3.5˚C over the next century. One of the most
important problems in the management and planning of water resources is forecasting of streamflow. This study focuses
on the impact of climate changes on streamflow in the Zayandeh-rood river basin in Isfahan, Iran. The Soil and Water
Assessment Tool (SWAT[1]) has been used for forecasting streamflow changes due to climate change. SWAT is a daily
time step model that allows for detailed simulation of the water balance. SUFI-2[2] model is used for calibrating SWAT
model’s parameters. SWAT model is calibrated and validated by utilizing observation data in 1990-1998. Next,
downscaling precipitation and temperature due to CGCM2 model are used. These data are derived from two scenarios,
A2 and B2 for two periods: 2021-2050 (immediate future) and 2071-2100 (far future). The SWAT model was run for all
cases. The hydrology of the climate change shows winter streamflow increases due to more precipitation and higher air
temperature in winter than observed. The results also show that less snowpack and an earlier snowpack melt resulting
in an earlier spring runoff that caused shift in peak flow for both periods.
Zayandeh-Rood watershed is located in the center of Iran. The
most important river in this watershed is Zayandeh-Rood River.
The river originates from the eastern Zagros’s mountainside
and ends up at the marsh Gavkhouny. Its length is 350 Km
from west to east. In this paper we focus on upper part of
Zayandeh-Rood watershed that has an area about 4,000 Km2.
it is less than 10% of whole watershed’s area. The study area
(Figure 1), called Zayandeh-Rood Dam basin.
Since its initial report in 1990, the Intergovernmental Panel on Climate Change has consistently noted that climate is
changing (IPCC, 1990; 1996; 2001). These reports, the studies upon which they are based, and other studies indicate that
water resources are particularly susceptible to the impacts of climate change [9].
Climate change may also impact water supplies on the watershed level. Watersheds that located at high elevations may
not be impacted by modest changes in temperature, as most precipitation will continue to fall as snow. Watersheds at low
elevation will likewise likely be unaffected, as precipitation will continue to fall as rain. Changes in spring and summer
precipitation may have significant impacts [6].
A significant portion of this project focused on streamflow simulation due to climate change scenarios (possible futures) at
the regional scale. Hydrologic simulation model, SWAT, have been used to evaluate the impacts of climate scenarios on
water resources. Downscaled precipitation and temperature from CGCM2 model are used as SWAT’s input.
Climate Change Scenarios
A General Circulation Model (GCM) is a computer model representing the atmosphere, oceans, land and ice caps. A GCM
simulates the behavior of the climate system. Different GCMs not only yield different globally averaged surface air temperature,
but also produce different special patterns of change in temperature, precipitation and other variables. The greenhouse gas
forcing scenarios used by all GCM experiments are expressed in terms of equivalent CO2 concentration [2].
Hydrological Model
In this study we chose SWAT (Soil and Water Assessment Tool) because of its availability in
handling input data. SWAT is a comprehensive model that requires a diversity of information in
order to run. It is a continuous time model, i.e. a long-term yield model. The model is not designed
to simulate detailed, signal-event flood routing. This model was developed to predict the impact of
land management practices on water, sediment and agricultural chemical yield in large complex
watersheds with varying soils, landuse and management conditions over long periods of time [8].
Data used
As shown in Figure 5 and Table 4, increase in air temperature in A2 scenario is more than B2 scenario,
therefore streamflow in A2 scenario has decreased more compare to the B2 scenario. In this period like the
immediate future (2021-2050) monthly streamflow shifts from April to March and February. Future air
temperature is more than observed air temperature during these months, therefore it can cause in early
snowmelt and shifts peak flow. Figure 5 shows decreases in average monthly streamflow during the seasons,
and earlier peakflow timing for the 2071-2100 period. Changes in the streamflow with increased winter time
warm precipitation (rain) suggest the increased likelihood of more extreme events, such as winter floods.
Simulation results for A2 and B2 scenarios based on climate change by SWAT model for both periods show
that the shift in peak flow. The shift is because of increase in air temperature and change in precipitation during
these months and early snowmelt. In general for both scenarios and both periods runoff volume shows
decrease, this decrease for first period is less than second period. It shows decreasing in streamflow gradually
by time that caused by climate change and increase greenhouse gases.
Table 3- Final calibration values (SWAT’s parameters)
2021-2050
100
SWAT needs variety of input data such as DEM map, Soil map and Land cover or landuse map
and also climate data. observed data is used for model calibration and verification process. At first
SWAT parameters were calibrated on historical streamflow data by using SUFI2 between 1990~
1995, then historical streamflow data from 1995 to 1997 is used for model validation.
Mean Average Streamflow
(m3/s)
The earth’s climate system (the atmosphere, ocean, biosphere, etc.) has been altered by human activities (e.g., fossil fuel
combustion and deforestation) in many ways. One of the consequences is a sharp rise in concentrations of greenhouse
gases (CO2, CH4, etc.) and aerosols since the late-nineteenth century. Because the atmospheric loading of greenhouse
gases is likely to rise in the twenty-first century as a result of the expected growth in the world economy and population,
meaningful predictions of climate changes in the next century have many practical applications (e.g., for impact
assessments and to guide future emission controls) [4].
Figure1- Zayandeh-Rood Dam basin
In the next step, downscaled data from two scenarios (A2, B2) based on CGCM2 model have
been used to perturb the time series of precipitation and temperature, for two periods; immediate
future (2021-2050) and far future (2071-2100). SWAT model was run for all cases for prediction
variation on streamflow in Zayandeh-Rood River in these periods. The required metrological data
(daily precipitation and temperature time series, average daily solar radiation, average wind
speed, etc.) were collected from 7 stations located in the study area.
Results and Discussion
Model was calibrated for observed monthly and yearly data between 1990 and 1995.
Determination coefficient for yearly and monthly calibration are R2=0.944 and R2=0.888 (Fig. 2).
Two other years, 1996 and 1997, were used for verification process. In this step determination
coefficient are; R2=0.899 for yearly validation and R2=0.89 for monthly validation (Fig. 3). Final
calibrated values (of SWAT’s parameters) are listed in Table 3. After calibration and validation
stages model has been run for two future periods due to climate change. That has been
downscaled from two scenarios (A2, B2) based on CGCM2 model. Results of simulation based on
A2 and B2 scenarios for two periods (2021- 2050 and 2071-2100) are shown in Figures 4 and 5
and also listed in Tables 3 and 4.
Current climate change models are at the global scale so the results are too broad to influence local policy. Downscaling the
model results helps local and regional planners to gain a sense of how climate change could impact their area, therefore,
provides them with the ability to consider climate change in their planning activities. GCMs contain significant inherent
uncertainty, so the results are best expressed in terms of a range of possible future climates. The IPCC (1994) recommends
that the results from all scenarios be considered when conducting climate change work. Some well-known GCMs are; Canadian
Global Coupled model (CGCM2), UK’s Hadley Centre model (HadCM3), and Australian Commonwealth Scientific and Industrial
Research. The special report on emission scenarios (IPCC 2001) has defined four future global greenhouse gas emission
scenarios; A1, B1, A2 and B2 [7,5].
Observed Streamflow
Simulated Streamflow
Streamflow (m3/s)
70
Observed Streamflow
250
Simulated Streamflow
200
60
50
40
30
20
150
100
50
0
10
1990
1991
1992
80
70
60
50
40
30
20
10
0
Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug
Month
Figure 4- Simulation for Immediate Future
Conclusion
For both scenarios and periods, early snowmelt and increase in precipitation in winter and also converting
snowfall to rainfall during these months may cause winter flood and dry summer. In general, each of these
simulations has suggested that Zayandeh-Rood snowmelt driven streamflows are likely to peak earlier in the
season under global warming due to increased atmospheric greenhouse gas concentrations. The peak
snowmelt month similarly shifts earlier the snow water equivalent is significantly reduced and the peak is earlier,
because of increase in temperature in winter and, results in earlier season snowmelt. The peak timing for each
simulation shifts towards earlier in the year as the snow-to-rain ratio decreases.
Monthly Calibration
Yearly Calibration
In this study we have used outputs of Canadian Global Coupled Model (CGCM2) with combination of two scenarios; A2 (high
emissions), B2 (low emissions). Increasing in CO2 concentration for these scenarios (A2, B2) is shown in Table1. Model results
were developed for two periods; 2021-2050 and 2071-2100. A statistical downscaling model has been used to establish
statistical relationships between the GCM output and local weather station variables. Details are provided by Azaranfar et al.
(2006), its results show increase in temperature by 2oC-5oC and increase in averaged precipitation in December and January
[3]. Outputs of this method are used as inputs of hydrological model.
90
A2
B2
Observed
Table 4- Mean Monthly Average Streamflow for A2 and B2 Scenarios for Far Future
Streamflow (m3/s)
Introduction
As shown in Figure 4 and Table 3, simulated discharge for immediate future period (2021-2050) based on both
A2 and B2 scenarios shows a little difference between observed in historical situation streamflow data and
simulated streamflow data in September, October and November and modest change in June, July and August.
Streamflow increases in December, January and February due to the increase in precipitation and also
increase in air temperature which converts snowfall to rainfall during these months. Monthly peak flow shifts
from April to March; early warmth in air temperature caused early snowmelt and it shifts spring peak flow.
1993
1994
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
1995
There are a number of aspects of future climate simulation analysis studies that need to be extended. There is a
need for more GCM ensemble members of the most recent simulations. Climate models will never predict the
future, but can provide projections with an uncertainty that can be bracketed. It is these bracketed solution sets
that may ultimately provide water resources decision-makers with the type of information needed to safeguard
one of our more important natural resources. By using these operational models for simulating and analyzing
hydrologic response to climate change, increased credibility and public acceptance is likely. Determining the
impacts of climate change on streamflow by evaluating the response of the outputs of scenarios is a valid
approach. The temperature shifts and precipitation ratios imposed on the historical time series constrain the
results to perturbations about the historical. However, this is the current impact assessment approach and this
study will be useful for applications of water demand and agro-economic assessments.
Year
Month
References
Figure 2- SWAT Model Calibration Process
Table1- Increasing in CO2 concentration for A2 and B2 scenarios [3] (ppmv)
Monthly Validation
Yearly Validation
180
140
Observed Streamflow
Simulated Streamflow
Streamflow (m3/s)
Streamflow (m3/s)
120
100
80
60
40
20
Observed Streamflow
Simulated Streamflow
160
140
120
100
80
60
40
20
0
0
Jan.
Feb.
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Month
Figure 3- SWAT Model Validation Process
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Month
1-Abbaspour, Karim C., Jing Yang, Ivan Maximov, Rosi Siber, 2006, Modelling Hydrology and Water Quality in the prealpine/alpine Thur Watershed Using SWAT, J. Hydrology, 10.1016/j.jhydrol.2006.09.014
2-An, Hyunhee and Wayland Eheart, 2001, Protecting Midwestern Streams from Climate Change Impacts,ASCE 2004,
World Water Congress.
3- Azaranfar, Arash, A. Abrishamchi and M. Tajrishy, 2007, The Impact Assessment of Climate Change on Precipitation and
Temperature in Zayanderoud Watershed by Downscaling Projection of GCM, 2nd Water Resources Management
Conference, Isfehan, Iran.
4-Dai, Aiguo, T. M. L. Wigley, B. A. Boville, J. T. Kieh, AND L. E. Buja, 2001, Climates of the Twentieth and Twenty-First
Centuries Simulated by the NCAR Climate System Model, J. of Climate, Vol. 14, 485-519.
5- DeWalle, David R. and Bryan R. Swistock, 2000, Potential Effects of Climate Change and Urbanization on Mean Annual
Streamflow in the United States, J. WATER RESOURCES RESEARCH, Vol. 36, No. 9, 2655–2664.
6- Knapp, H. Vernon, Jaswinder Singh, Hua Xie, Yanqing Lian, and Misganaw Demissie, 2005, Potential Effects of Climate
Change and Variability on the Surface Water Resources of the Upper Midwest,ASCE, EWIR.
7- Langsdale, Stacy , Stewart Cohen, Rachel Welbourn and James Tansey, 2004, Dialouge on Adaptation to Climate
Change in the Okanogan Basin, British Colombia, Canada, Water World Congress.
8- Neitsch, S.L, J.G. Arnold, J.R. Kiniry, J.R. Williams, 2001, Soil and Water Assessment Tool User’s Manual, Version 2000.
9- VanRheenen, Nathan T., and Richard N. Palmer, 2002, Evaluating Potential Climate Change Impacts on Water
Resources System Operation, UCOWR, 35-50.