Handbook of Climate Change Adaptation

DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

A Multi-model Framework for Climate Change Impact Assessment

Alireza Gohari*, Mohammad Javad Zareian and Saeid Eslamian
Department of Water Engineering, Isfahan University of Technology (IUT), Isfahan, Iran

Abstract
The chapter aims to estimate the climate change impacts within a probabilistic multi-model
framework. The suggested approach attempts to improve the reliability of the climate change impact
assessment approach by considering the three main sources of uncertainty.
Climate change impacts on the climate variables in Iran’s Zayandeh-Rud River Basin have been
evaluated. Multi-model ensemble scenarios are used to deal with the uncertainty in climate change
projection for the study period (2015–2044). The probabilistic multi-model ensemble scenarios,
which include the 15 GCMs, are used to project the temperature and precipitation for the near future
period (2015–2044) under 50 % risk level of climate change.
Downscaled climate variables suggest that generally temperature will rise in the Zayandeh-Rud
River Basin while the level of temperature increase varies between. The maximum monthly
precipitation reduction will occur in winter. This can be of considerable importance for the basin
having a semiarid Mediterranean climate in which winter precipitation is the main source of
renewable water supply.
In the proposed framework, the uncertainties of GCMs, emission scenarios, and climate variabil-
ity of daily time series are handled by the combination of change factors and a weather generator.
Covering the full range of potential climate change, such framework can provide the valuable
lessons to policy maker for adapting to climate change.
The Zayandeh-Rud River Basin has been constantly facing the water stress problem during the
past 60 years. The results of the climate change impacts on the basin’s climate variables can provide
the policy insights for regional water managers to address well the water scarcity in the near future.

Keywords
Climate change; Impact assessment; Climate variable; Zayandeh-Rud; Iran

Introduction
Climate change is a significant change in the weather conditions in long-time periods. It may be the
changes in weather parameters or in the distribution of weather events (i.e., more or fewer extreme
weather events). Climate change is considered as one of the major factors that will affect the water
availability and life in the future (Bates et al. 2008; Eslamian et al. 2011).
Humans have been the significant contributors to greenhouse gas emissions by different industrial
and household activities. The Intergovernmental Panel on Climate Change (IPCC) stated that in the

*Email: alirezagohari@gmail.com

Page 1 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Fourth Assessment Report of climate change (AR4), the air temperature increased by 0.74
C from
1906 to 2005 (IPCC 2001).
Increased evaporation (resulting mainly from the higher temperatures), combined with regional
changes in precipitation characteristics (e.g., total amount, variability, and frequency of extremes),
has the potential to affect mean runoff, frequency and intensity of floods and droughts, soil moisture,
and water supplies for irrigation and hydroelectric generation, which are the little effects of climate
change in the recent years (Ghosh and Mujumdar 2009; Zhang et al. 2010; Eslamian et al. 2011).
Projections of climatic variables globally can be performed with General Circulation Models
(GCMs), which provide the projections at large spatial scales. These models are considered as the
most credible tools for the projections of future global climate change (IPCC 2007).
An important limitation in the GCMs’ application is the some sources that prevent from finality in
the model outputs. These limits are often expressed as uncertainty (Moss and Schneider 2000).
These include:

1. Estimating the amount of greenhouse gas and aerosol emissions is hard and has a lot of
uncertainties (related to the emission scenarios) (Parry et al. 2004).
2. There are some uncertainties in GCMs (Sajjad Khan et al. 2006).
3. Global climate model sensitivities don’t have the finality (Elmahdi et al. 2008).

The uncertainty in the global model configuration has long been recognized as one of the most
important parts of the overall uncertainty, especially considering the first decades of the twenty-first
century when the different emission scenarios do not lead to dramatically different climate responses
(Benestad 2004).
To resolve these uncertainties, the different methods based on physical and mathematical relation-
ships are presented in climate change modeling (Schaefli et al. 2007).
There are two general approaches for using of the GCMs. The first method is using a single GCM
model for predicting future data (Jones and Thornton 2003; Guo et al. 2010). This may lead to an
incorrect estimate of the calculated parameters in climate change models (Lee et al. 2011).
In the second method, multiple climate change scenarios are produced by the different GCMs to
capture probable range of climate change impacts. This method helps to identify, better understand
and more realistic modeling of the climate change (Medellin-Azuara et al. 2008; Kloster et al. 2010;
Abrishamchi et al. 2012).
There are many researches in the literature for the use of multiple impact assessment models to
better represent and manage these uncertainties (Tao et al. 2009; Lizumi et al. 2009; Tao and Zhang
2010; Daccache et al. 2011; Ozdogan 2011). In the recent studies, probabilistic outputs from
ensembles of the GCMs models and emission scenarios have been used to achieve a better
representation of uncertainties and comprehensive impact assessment. Some of them suggested
that an average over the set of GCM outputs provides a dominant climate simulation related to any
individual model (Bader et al. 2008; Liu et al. 2010). Elmahdi et al. (2008) have used seven
AOGCMs from the IPCC third assessment report to project the future temperatures. They generated
1,000 samples of air temperature time series for uncertainty analysis and risk assessment of water
demand.
This chapter aims to assess a multi-model framework for climate change impact assessment. In
this framework, the uncertainties of GCMs and emission scenarios are handled by an innovated
downscaling method (combination of change factors and a weather generator). The suggested
approach attempts to improve the reliability of the climate change impact assessment approach by
considering the three main sources of uncertainty as discussed earlier.

Page 2 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Fig. 1 Detailed flowchart of methodology

Methodology Outlines
In this study, the multi-model ensemble scenarios within the risk framework are used in order to
manage uncertainty of the GCM outputs and emission scenarios. This probabilistic framework has
been constructed by the combination of change factor and LARS-WG methods in the downscaling
of ensemble GCMs. A detailed flowchart of this methodology has been shown in Fig. 1. According
to Fig. 1, the major stages of this study consist of three steps: (a) generation of temperature and
precipitation climate change scenarios by 15 GCMs for the case study, (b) generation of cumulative
distribution function for climate change scenarios and extraction of climate scenarios corresponding
to 50 % probabilities, and (c) generating temperature and precipitation daily time series by the
LARS-WG stochastic downscaling method.

Climate Change Scenarios Generation

Generating climate change scenarios is the first step to achieve probable patterns of the future
climate based on assumptions of future atmospheric concentrations. The AOGCM-derived scenar-
ios of climate change are the most common scenario type in the impact assessments. In this chapter,
15 GCMs outputs under two emission scenarios (A2 and B1) from the Fourth Assessment Report
(AR4) of the Intergovernmental Panel on Climate Change (IPCC) are used (IPCC 2007). The
detailed description about these models is shown in Table 1. A2 scenarios assume rapid population
growth coupled with slow economic and technological development until 2100. B1 scenarios are
characterized by very rapid global socioeconomic growth and population rising until 2050 and then
declining toward the end of the century.
In the first step, monthly temperature and precipitation variables for the baseline period
(1971–2000) and future period (2015–2044) have been extracted from the Data Distribution Center
(http://www.ipcc-data.org) (DDC) of IPCC. The values of difference for the temperature and relative
change for precipitation between the 30-year monthly average baseline period and future period are
calculated for each month. These values represent the climate change scenarios for the temperature
and precipitation’s 30-year monthly averages. Here, climate change scenarios in the future period to
baseline period are created separately for the different AOGCMs under two emission scenarios.

Page 3 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Table 1 Description about 15 GCMs from the Fourth Assessment Report (AR4) IPCC
Model Abbreviation Center
HadCm3 HADCM3 UKMO (UK)
ECHAM5-OM MPEH5 MPI-M (Germany)
CSIRO-MK3.0 CSMK3 ABM (Australia)
GFDL-CM2.1 GFCM21 NOAA/GFDL (USA)
MRI-CGCM2.3.2 MRCGCM MRI (Japan)
CCSM3 NCCCSM NCAR (USA)
CNRM-CM3 CNCM3 CNRM (France)
MIROC3.2 MIMR NIES (Japan)
IPSL-CM4 IPCM4 IPSL (France)
GISS-E-R GIER NASA/GISS (USA)
BCM 2.0 BCM Beijing Climate Center (China)
CGCM3 T47 CGCM Canadian Centre for Climate Modeling and Analysis
ECHO-G ECHO Meteorological Institute, University of Bonn Meteorological Research
INMCM 3.0 INMCM Russian Academy of Science, Institute of Numerical Mathematics
NCARPCM NCRPCM National Center for Atmospheric Research (NCAR), USA

Risk Assessment of Climate Change Impacts

There are the high levels of uncertainties in the results of the AOGCM climate change scenarios.
These uncertainties affect the choice of method and the confidence that can hurt the results in the
impact assessment studies. There are many methods for uncertainty management, for example,
expression of the results as a central prediction, central prediction with error bars, known probability
distribution function, a bounded range with no known probability distribution, and a bounded range
within a larger range of unknown possibilities (OECD 2003). Here, a bounded range with known
probability distribution (Gohari et al. 2013a) is used for managing uncertainties due to use of ten
AOGCMs, which is created from the weighting of the AOGCMs. This probabilistic multi-models
approach includes the three following steps.

Weighting the AOGCMs

The ranges of monthly climate change scenarios are not the same due to different abilities of the
AOGCMs to local climate simulation. So, each of the 15 GCMs used in this study separately has
been weighted based on the mean observed temperature-precipitation (MOTP) method. In this
method, each AOGCM has been weighted based on the difference between average of temperature
and precipitation simulated by AOGCM in the base period from corresponding observed value,
following the study by Massah Bavani and Morid (2005):

1  
D T ij
Wi ¼ X15 XN
  (1)
D T ij
1
j¼1 i¼1

where Wi is the weight of each model in month i and DTi is the temperature and precipitation change
field for GCM j in month i.

Page 4 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

a 0.15 b

Weights of GCM models

Weights of GCM models

0.08

0.10 0.06

0.04
0.05
0.02

0.00 0.00
0.6 0.7 0.8 0.9 1.0 1.1 1.2 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Relative precipitation change Temperature change (°C)

Fig. 2 Discrete PDFs, relating monthly temperature and precipitation changes to the weights of corresponding GCMs.
Graph (a) shows the developed PDF of relative precipitation changes in May and graph (b) shows the developed PDF of
temperature changes in February

Probability Distribution Function (PDF)

In this step, probability distribution function (PDF) is made, where the X and Y values represent the
temperature or precipitation monthly changes and the weight of corresponding AOGCM, respec-
tively (Fig. 2). In order to produce a time series of temperature and precipitation for the future period,
the temperature and precipitation long-term monthly averages are needed. Therefore, discrete
probability distributions of climate change scenarios must be converted to continuous probability
distribution. Due to a high correlation and having a low number of parameters, the two parameters
beta distribution function as one of the best function can be fitted on these discrete distributions:

ðx  aÞp1 ðb  xÞq1
f ðxÞ ¼ a  x  b; p, q > 0 (2)
Bðp,qÞðb  aÞpþq1

where x and p and q are the variable and shape parameters for beta distribution function respectively,
and B(p, q), the beta function.
The values of p and q are changed to get the best fit based on the maximum likelihood estimation
method. Here, the sum of squared error (Eq. 3) is used to show how well the beta function fits the
data:
Xn
SSE ¼ ðy  ¥ i Þ2
i¼1 i
(3)

where yi is the data point, ¥i is the estimation of beta function, and n is the number of data points (n ¼
15). The small size of sample data set can affect the goodness of fit. This is indeed a limitation of this
type of assessment caused by having access to a limited number of GCMs.

Cumulative Distribution Function (CDF)

In this step, the developed PDFs are converted to CDFs. Cumulative distribution function (CDF)
curve is made, where X and Yvalues represent the temperature or precipitation monthly changes and
corresponding exceed probability (Pi). Since there is no possibility to use each point of these CDF in
the impact assessment models, the monthly temperature and precipitation change scenarios have
been extracted corresponding to 50 % discrete probabilities as a moderate-risk level of climate
change (Fig. 3).

Page 5 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Fig. 3 Developed CDFs based on the presented PDFs in Fig. 2

Stochastic Downscaling
The direct use of the AOGCM-derived climate scenarios could be impossible in impact assessment
models due to large deviation of these model-derived downscale variables time series (i.e., temper-
ature and precipitation, etc.) with real data. But long-term monthly averages of these variables are
highly correlated to the real data (Semenov 2007). The downscaling techniques bridge the gap
between the AOGCM outputs and required inputs by the impact assessment models (Wilby and
Wigley 1997). One of the downscaling tools to generate daily climate scenarios is a stochastic
weather generator (WG) (Wilks and Wilby 1985; Semenov 2007). A WG is a model, which has an
ability to simulate synthetic time series of daily weather for the future by using predicted climate
change scenarios from GCMs.

LARS- Weather Generator

LARS-WG is one of the stochastic WGs that generate synthetic daily time series of minimum and
maximum temperatures, precipitation, and solar radiation (Semenov 2007). The LARS-WG can
generate the daily time series from monthly climate change scenarios. In the first step, the weather
generator parameters calculate based on the probability distributions of locally observed daily
weather variables (Semenov 2007). So, the semiempirical distributions of observed data such as
frequency distributions calculate for the wet and dry series duration. Fourier series are used for
describing the precipitation amount, solar radiation, and minimum and maximum temperatures. In
the next step, LARS-WG generates synthetic weather data by combining a climate change scenario
file for precipitation amount, wet and dry series duration, mean temperature, temperature variability
and solar radiation with the resulting parameter files. In the final step, the observed data statistical
characteristics are compared with those of synthetic data. A number of statistical tests (i.e., the
chi-squared test, Student’s t-test, and F-test) are used in this comparison to determine the differences
between the distributions, mean and standard deviation values of the synthetic and observed data set
(Semenov and Barrow 1997; Semenov et al. 1998).
In the present work, future daily time series of maximum and minimum temperatures and
precipitation are generated for 2015–2044 based on observed daily time series and climate change
scenarios for 50 % probabilities using LARS-WG. For generating 30 years daily time series,
300 years daily time series (10  30 years daily time series) for future period are generated by
this WG and then the average daily values of these ten time series are calculated.

Page 6 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Case Study
The Zayandeh-Rud River Basin with an area of about 26,917 km2 is located in central Iran (Fig. 4).
In recent decades, due to high agricultural and industrial development potential, the basin has
witnessed economic growth and increased population (Madani and Marino 2009). Currently,
more than 3.7 million people are living in the basin, making it the second most populated watershed
in Iran. As the major water consumer, agriculture sector uses more than 73 % of water supply
(Zayandab Consulting Engineering Co. Report 2008). Cultivation of high-water-demand crops (i.e.,
rice, corn, wheat, and barley) and low irrigation efficiency of 34–42 % contribute to the high
agricultural water demands (Gohari et al. 2013b).
The Zayandeh-Rud River with an average flow of 1,400 million cubic meters (MCM), including
650 MCM of natural flow and 750 MCM of the transferred flow, starts from the Zagros Mountains in
the west of the basin and flows into the Gav-Khooni Marsh in the east of the basin (Gohari
et al. 2013b). The Gav-Khooni Marsh is recognized as an international wetland under the Ramsar
Convention (1971). The river has been tapped for increasing water consumption within and outside
the basin. This makes the river the most important water resource of the basin for its residents and
their urban, industrial, and agricultural uses, as well as for the survival of the ecosystem of the
Gav-Khooni Marsh.
Two climatological stations and a rain gauge station are selected in the study area. Table 2 presents
the description of these stations.

Fig. 4 Zayandeh-Rud River Basin

Page 7 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Table 2 Brief description of the selected observation stations

Geographical position
Station name Longitude Latitude Elevation (m) Type of station

0
Damaneh-Fereydan 50 29 33
010 2,300 Climatological station

0
Chelgerd 50 05 32
270 2,300 Rain gauge station
Isfahan 51
680 32
630 1586 Climatological station

Table 3 Large-scale temperature changes for EHO-G model in Isfahan station

Year Jan Feb March Apr May Jun Jul Aug Sept Oct Nov Dec
2015 1.72 0.58 3.59 2.77 1.69 2.55 1.72 2.16 1.72 2.06 2.31 1.04
2016 0.16 2.66 1.71 0.03 0.59 0.32 1.21 1.17 2.41 2.73 1.33 0.88
2017 1 0.69 0.69 1.4 1.86 3.57 2.5 1.33 1.03 1.63 2.38 1.83
2018 1.34 0.19 0.75 1.66 2.66 1.35 1.7 1.38 2.61 1.36 1.86 1.28
2019 1.26 0.95 2.5 1.3 2.46 0.5 2.5 1.44 1.27 1.96 0.82 2.5
2020 1.18 3.01 0.73 0.48 1.68 1.97 0.73 2.62 2.41 2.59 1.06 0.89
2021 2.27 1.03 0.05 2.38 1.04 0.61 0.8 1.93 0.45 2.37 1.86 1.81
2022 1.7 2.27 3.55 3.13 1.35 1.99 1.67 1.11 0.4 1.61 1.05 2.02
2023 2.32 1.51 1.36 1.57 0.72 3.23 1.43 1.75 2.89 2.44 1.52 2.7
2024 1.41 2.19 0.56 1.99 1.44 3.28 1.51 2.5 1.63 2.17 1.69 1.08
2025 0.49 0.97 0.19 4.37 2.3 2.9 0.85 0.94 3.27 3.39 2.34 2.34
2026 0.09 0.19 0.25 0 0.84 1.55 2.07 1.56 2.37 3.97 0.59 0.27
2027 0.43 0.05 1.14 2.49 1.78 2.03 0.72 1.5 3.88 2.91 1.12 1.68
2028 0.78 1.69 1.66 1.91 0.98 1.6 1.86 1.03 2.53 1 0.74 0.67
2029 1.33 1.94 0.21 2.53 2.03 1.9 1.56 1.84 2.85 2.02 2.72 2.78
2030 0.94 2.85 1.35 0.57 0.87 2.56 0.55 2.22 1.52 2.6 2.42 1.96
2031 1.86 2.22 1.24 2.66 1.37 2.59 2.31 1.41 2.01 1.38 1.2 0.65
2032 1.09 0.51 1.47 0.64 1.27 3.13 2.54 2.07 0.5 3.1 2.85 3.48
2033 2.93 1.15 1.66 4.3 1.81 0.68 3.06 2.1 2.99 1.74 3.92 2.74
2034 1.8 0.1 2.35 1.98 1.41 3.05 2.52 1.69 2.59 1.76 1.66 0.43
2035 0.92 0.59 1.45 2.79 1.6 2.55 2.46 1.96 1.15 1.79 2.63 2.45
2036 1.8 1.39 3.25 0.4 1.97 1.13 1.75 3.07 1.82 0.52 2.19 3.2
2037 0.46 0.12 2.29 1.67 2.13 1.22 1.64 2.07 2.81 2.1 1.23 1.54
2038 0.56 0.31 1.39 2.06 1.36 2.31 1.71 2.02 2.01 1.57 3.24 0.5
2039 0.17 1.3 0.19 2 2.74 2.28 2.49 1.62 2.9 2.23 2.34 1.88
2040 2.18 0.74 2.28 2.36 2.35 1.14 1.29 2.59 3.13 2.76 0.96 0.72
2041 1.03 1.21 1.79 0.28 1.61 0.1 1.5 1.61 1.3 4 4.46 1.5
2042 1.58 1.38 1.61 4.23 0.83 1.69 1.85 1.64 1.02 1.2 0.13 2.83
2043 1.69 1.51 0.38 1.83 2.12 1.11 2.29 1.85 2.65 2.85 0.36 1.48
2044 0.43 1.45 2.1 1.5 2.03 2.06 2.33 2.27 2.06 3.15 2.43 0.35

Impact of Climate Change on Climate Variables

Large Scale Climate Change Scenario for Temperature and Precipitation
The results of temperature changes for Isfahan synoptic station projected from ECHO-G model are
presented in Table 3. Mean monthly temperature changes are generally expected to increase under

Page 8 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

climate change. Most GCMs suggest higher future temperatures in the study region while the range
of expected temperature changes varies between months.

Uncertainty of Climate Change Projection

Weighting of AOGCMs
The weights of 15 GCMs for Isfahan station are shown in Table 4. The values of AOGCM weights
vary between different months. The maximum weight is assigned to the CGCM3.0 model in July.
The comparison of average weights of different models shows that minimum weight is assigned to
the NCARPCM. The average monthly weights show that the maximum contribution of different
months is related to July.

Probability Distribution Function

The discrete PDFs of climate change scenarios are developed for each month. The parameter
estimation method is used to convert the discrete PDFs to continuous ones. Table 5 shows the
values of SSE and beta distribution parameters for continuous temperature. The low values of SSE
underline the suitability of beta distribution.

Regional Temperature and Precipitation Changes

Figures 5 and 6 respectively, show the estimated local (downscaled) temperature for Damaneh-
Fereydan and Isfahan stations under climate change for various risk levels. Generally, temperature
increases are expected for all months under climate change. But, the levels of increase vary between
the months. The maximum temperature changes in Damaneh-Fereydan are expected in spring and
summer months, respectively, under A2 and B1, while maximum temperature change fields in
Isfahan station are simulated in summer and fall months, respectively, under A2 and B1.
Unlike the results for temperature, the local precipitation changes do not show a general increas-
ing trend (Figs. 7 and 8). The results show that the maximum monthly precipitation decreases in
Isfahan and Chelgerd stations in January under A2 and B1.
The projected annual and seasonal changes of 30-year mean temperature and total precipitation
were calculated under A2 and B1 (Tables 6 and 7). With 5–15 % decrease in Chelgerd’s precipitation

Table 4 Estimated weights for 15 GCMs in Isfahan station for temperature

Model Jan Feb March Apr May Jun July Aug Spt Oct Nov Dec
BCM2.0 0.017 0.020 0.025 0.035 0.008 0.025 0.004 0.020 0.015 0.007 0.027 0.019
CGCM3.0 0.022 0.029 0.032 0.037 0.011 0.109 0.896 0.154 0.033 0.012 0.031 0.019
CGCM232 0.119 0.084 0.071 0.076 0.542 0.195 0.039 0.135 0.046 0.691 0.082 0.054
CNRCM3 0.017 0.029 0.045 0.048 0.016 0.038 0.008 0.033 0.030 0.009 0.027 0.017
CSIROMK3 0.014 0.023 0.028 0.027 0.006 0.016 0.002 0.010 0.008 0.004 0.017 0.013
ECHAM5ON 0.060 0.095 0.251 0.198 0.276 0.187 0.015 0.193 0.173 0.051 0.099 0.050
ECHO-G 0.051 0.067 0.092 0.091 0.025 0.069 0.009 0.055 0.071 0.067 0.133 0.066
GFDL21 0.022 0.031 0.030 0.029 0.007 0.025 0.004 0.020 0.015 0.007 0.023 0.019
GISS-ER 0.396 0.232 0.145 0.066 0.026 0.183 0.003 0.014 0.018 0.068 0.051 0.076
HADCM3 0.026 0.032 0.031 0.029 0.006 0.016 0.002 0.013 0.010 0.006 0.024 0.022
INMCM3.0 0.077 0.079 0.059 0.045 0.009 0.015 0.001 0.009 0.009 0.006 0.033 0.032
IPSLCM4 0.043 0.054 0.059 0.062 0.018 0.046 0.010 0.293 0.532 0.044 0.109 0.060
MICRO3.2 0.090 0.171 0.075 0.187 0.027 0.029 0.002 0.011 0.012 0.015 0.294 0.511
NCARCCSM3 0.029 0.033 0.035 0.042 0.016 0.034 0.004 0.031 0.019 0.009 0.032 0.028
NCARPCM 0.015 0.021 0.022 0.026 0.006 0.013 0.001 0.009 0.009 0.005 0.019 0.015

Page 9 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Table 5 The estimated values for beta distribution parameters in Isfahan station (temperature)
Month a b a B SSE
Jan 5.00 6.00 0.54 1.52 0.01
Feb 5.78 5.00 0.30 1.49 0.01
March 6.36 4.00 0.19 1.33 0.00
Apr 4.13 6.00 0.22 2.35 0.00
May 5.00 4.88 0.11 1.95 0.00
Jun 5.00 4.00 0.44 1.73 0.01
July 3.24 6.00 0.05 3.01 0.02
Aug 6.00 3.85 0.09 1.75 0.03
Sept 3.00 5.00 0.25 2.50 0.02
Oct 5.83 4.78 0.41 1.43 0.01
Nov 3.91 6.00 0.37 1.75 0.01
Dec 2.56 7.00 0.53 2.48 0.01

25
Damaneh-A2 2015-2044 Damaneh-B1
2015-2044
20 1971-2000 1971-2000
Average air temprature (∞C)

15

10

-5

-10
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

Fig. 5 Comparison of the baseline (1971–2000) and future (2015–2044) period 30-year average temperature under A2
and B1 emission scenarios in Damaneh-Fereydan station

35
Isfahan-A2 2015-2044 Isfahan-B1
2015-2044
30 1971-2000 1971-2000
Average air temprature (⬚C)

25

20

15

10

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

Fig. 6 Comparison of the baseline (1971–2000) and future (2015–2044) period 30-year average temperature under A2
and B1 emission scenarios in Isfahan station

Page 10 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

400
Chelgerd-A2 Chelgerd-B1
350 2015-2044 2015-2044
1971-2000 1971-2000
300

P recipitatio n (m m )
250

200

150

100

50

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

Fig. 7 Comparison of the baseline (1971–2000) and future (2015–2044) period 30-year mean monthly precipitation
under A2 and B1 emission scenarios in Chelgerd station

35
Isfahan-A2 Isfahan-B1
2015-2044 2015-2044
30
1971-2000 1971-2000
25
Precipitation (mm)

20

15

10

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

Fig. 8 Comparison of the baseline (1971–2000) and future (2015–2044) period 30-year monthly mean precipitation
under A2 and B1 emission scenarios in Isfahan station

and 0.58
C increase in Damaneh-Fereydan’s temperature at annual scale, the upper subbasin will
face warmer and drier conditions under climate change. The results of Isfahan station indicate that
the lower sub-basin will experience warmer (0.82–0.95
C) and dryer (6–17 % reduction in
precipitation) conditions than the upper subbasin in the future.

Conclusions
In this study, 15 GCMs are used under two emission scenarios to estimate climate change impacts on
climate variables in the Zayandeh-Rud River Basin. To deal with the high uncertainty in the
estimated temperature and precipitation changes under climate change, the weighted ensembles of
GCMs’ outputs were generated and climate change variables at 50 % risk levels (a medium-risk
level) were estimated. The results suggest that generally temperature will rise in the study area while
the level of temperature’s increase varies between the months. Monthly precipitation changes do not
show a general increasing or decreasing trend.

Page 11 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Table 6 Temperature changes (
C) in the study area under different emission scenarios
Isfahan Damaneh-Fereydan
Month A2 B1 A2 B1
Jan 0.13 0.10 0.10 0.03
Feb 0.05 0.02 0.00 0.11
Mar 0.95 0.71 1.05 0.91
Apr 0.77 0.95 1.00 0.65
May 1.03 0.93 0.78 0.89
Jun 0.70 0.31 0.46 0.53
Jul 1.45 0.54 0.61 0.94
Aug 1.81 1.27 0.62 0.75
Sep 1.01 1.34 0.69 0.82
Oct 1.05 1.11 0.39 0.62
Nov 1.12 1.28 0.72 0.55
Dec 1.31 1.22 0.55 0.36
Max 1.81 1.34 1.05 0.94
Min 0.05 0.02 0.00 0.11
Mean 0.95 0.82 0.58 0.58
Winter 0.38 0.28 0.38 0.28
Spring 0.83 0.73 0.75 0.69
Summer 1.42 1.05 0.64 0.84
Fall 1.16 1.20 0.55 0.51

Table 7 Precipitation changes (%) in the study area under different emission scenarios
Isfahan Chelgerd
Month A2 B1 A2 B1
Jan 0.57 0.52 0.62 0.27
Feb 0.39 0.28 0.52 0.17
Mar 0.07 0.01 0.37 0.31
Apr 0.66 0.43 0.27 0.24
May 0.44 0.15 0.19 0.01
Oct 0.29 0.34 0.23 0.63
Nov 0.39 0.42 0.44 0.52
Dec 0.05 0.10 0.03 0.07
Max 0.66 0.43 0.44 0.63
Min 0.57 0.52 0.62 0.27
Mean 0.17 0.06 0.05 0.15
Winter 0.30 0.26 0.26 0.05
Spring 0.11 0.14 0.23 0.13
Fall 0.24 0.00 0.23 0.36

Upper subbasin precipitation especially in winter is known as the main resource of surface water
in the Zayandeh-Rud River Basin. The results indicate that maximum monthly precipitation
reduction (5–26 %) will occur in winter and annual precipitation is expected to decrease by

Page 12 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

5–15 %. This can be of a considerable importance for the Zayandeh-Rud River Basin with semiarid
Mediterranean climate in which winter precipitation is the main source of renewable water supply.
Temperature rise will lead to more precipitation falling as rain, instead of snow, and the snowpack
will melt earlier in the spring. The reduced snowfall due to increasing temperature in winter months
will generally lead to more severe water shortages in such arid and semiarid regions under climate
changes.

References
Abrishamchi A, Jamali S, Madani K, Hadian S (2012) Climate change and hydropower in Iran’s
Karkheh River Basin. In: Proceeding of the 2012 world environmental and water resources
(EWRI) congress, ASCE, 20–25 May 2012, Albuquerque, pp 3341–3349
Bader DC, Covey C, Gutowski WJ, Held IM, Kunkel KE, Miller RL (2008) Climate models: an
assessment of strengths and limitations. U.S. climate change science program synthesis and
assessment product 3.1. Department of Energy, Office of Biological and Environmental Research,
Washington, DC, 124 pp
Bates BC, Kundzewicz ZW, Wu S, Palutikof JP (2008) Climate Change and Water. Technical Paper
of the Intergovernmental Panel on Climate Change, IPCC Secretariat, Geneva, 210 pp.
Convention of Wetlands (1971) The Ramsar Convention of Wetlands, Ramsar. http://www.ramsar.org
Daccache A, Weatherhead EK, Stalham MA, Knox JW (2011) Impacts of climate change on
irrigated potato production in a humid climate. Agric For Meteorol 151:1641–1653
Elmahdi A, El-Gafy I, Kheireldin K (2008) WBFS model: strategic water and food security planning
on national wide level. IGU-2008 Water Sustainability Commission, Tunis
Eslamian S, Khordadi MJ, Abedi-Koupaei J (2011) Effects of variations in climatic parameters on
evapotranspiration in the arid and semi-arid regions. Glob Planet Change 78:188–194
Eslamian SS, Gilroy KL, McCuen RH (2011) Climate Change Detection and Modeling in
Hydrology, Climate Change - Research and Technology for Adaptation and Mitigation, [(Ed.)
Blanco j.], ISBN: 978-953-307-621-8, InTech, Available from: http://www.intechopen.com/
books/climate-changeresearch-and-technology-for-adaptation-and-mitigation/climate-change-
detection-and-modeling-in-hydrology
Ghosh S, Mujumdar PP (2009) Climate change impact assessment: uncertainty modeling with
imprecise probability. J Geophys Res 114. doi:10.1029/2008JD011648
Gohari A, Eslamian S, Abedi-Koupaei J, Massah Bavani AR, Wang D, Madani K (2013a) Climate
change impacts on crop productivity in Iran’s Zayandeh-Rud River Basin. Sci Total Environ
442:405–419
Gohari A, Eslamian S, Mirchi A, Abedi-Koupaei J, Massah Bavani AR, Madani K (2013b) Water
transfer as a solution to water shortage: a fix that can backfire. J Hydrol 491:23–39
Guo R, Lin Z, Mo X, Yang C (2010) Responses of crop yield and water use efficiency to climate
change in the North China Plain. Agric Water Manag 97:1185–1194
IPCC (2001) Climate Change 2001: Impacts, Adaptation and Vulnerability. IPCC Third Assessment
Report, Cambridge University Press
IPCC (2007) General guidelines on the use of scenario data for climate impact and adaptation
assessment. Cambridge University Press, Cambridge/New York
Jones PG, Thornton PK (2003) The potential impacts of climate change on maize production in
Africa and Latin America in 2055. Glob Environ Change 13:51–59

Page 13 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Kloster S, Dentener F, Feichter J, Raes F, Lohmann U, Roeckner E, Fischer-Burns I (2010) A GCM

study of future climate response to aerosol pollution reductions. Clim Dyn 34:1177–1194
Lee J, Gryze SD, Six J (2011) Effect of climate change on field crop production in California’s
Central Valley. Clim Change 109:335–353
Liu S, Mo X, Lin Z, Xu Y, Ji J, Wen G, Richey J (2010) Crop yield responses to climate change in the
Huang-Huai-Hai Plain of China. Agric For Meteorol 97:1195–1209
Lizumi T, Yokozawa M, Nishimori M (2009) Parameter estimation and uncertainty analysis of
a large-scale crop model for paddy rice: application of a Bayesian approach. Agric For Meteorol
149:333–348
Madani K, Marino MA (2009) System dynamics analysis for managing Iran’s Zayandeh-Rud River
Basin. Water Resour Manag 23:2163–2187
Massah Bavani AR, Morid S (2005) The impacts of climate change on water resources and
agricultural production. J Water Resour Res 1:40–7 (In Persian)
Medellin-Azuara J, Harou JJ, Olivares MA, Madani K, Lund JR, Howitt RE, Tanaka SK, Jenkins
MW, Zhu T (2008) Adaptability and adaptations of California’s water supply system to dry
climate warming. Clim Change 87:75–90
Moss RH, Schneider SH (2000) Uncertainties in the IPCC TAR: recommendations to lead authors
for more consistent assessment and reporting. In: guidance papers on the cross cutting issues of
the third assessment report of the IPCC. Pachauri R, Taniguchi T, Tanaka K (Eds), World
Meteorological Organization. Geneva, p. 33–51
OECD (2003) Special issue on climate change: climate change policies: recent development and
long term issues. Organization for Economic Cooperation and Development (OECD), Paris
Ozdogan M (2011) Modeling the impacts of climate change on wheat yields in northwestern Turkey.
Agric Ecosyst Environ 141:1–12
Parry ML, Rosenzweig C, Iglesias A, Livermore M, Fischer G (2004) Effects of climate change on
global food production under SRES emissions and socio-economic scenarios. Glob Environ
Change 14:53–67
Sajjad Khan M, Coulibaly P, Dibike Y (2006) Uncertainty analysis of statistical downscaling
methods. J Hydrol 319:357–382
Schaefli B, Hingray B, Musy A (2007) Climate change and hydropower production in the Swiss
Alps: quantification of potential impacts and related modelling uncertainties. Hydrol Earth Syst
Sci 11:1191–1205
Semenov MA (2007) Development of high-resolution UKCIP02-based climate change scenarios in
the UK. Agric For Meteorol 144:127–138
Semenov MA, Barrow EM (1997) Use of a stochastic weather generator in the development of
climate change scenarios. Clim Change 35:397–414
Semenov MA, Brooks RJ, Barrow EM, Richardson CW (1998) Comparison of the WGEN and
LARSWG stochastic weather generators for diverse climates. Clim Res 10:95–107
Tao F, Zhang Z (2010) Adaptation of maize production to climate change in North China Plain:
quantify the relative contributions of adaptation options. Eur J Agron 33:103–116
Tao F, Yokozawa M, Zhang Z (2009) Modeling the impacts of weather and climate variability on
crop productivity over a large area: a new process-based model development, optimization, and
uncertainties analysis. Agric For Meteorol 149:831–850

Page 14 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Wilby RL, Wigley TML (1997) Downscaling general circulation model output: a review of methods
and limitations. Prog Phys Geogr 21:530–548
Wilks DS, Wilby RL (1985) The weather generation game: a review of stochastic weather models.
Prog Phys Geogr 23:329–357
Zayandab Consulting Engineering Co. Report (2008) Determination of resources and consumptions
of water in the Zayandeh-Rud River Basin, Iran (In Persian)
Zhang Q, Xu CY, Tao H, Jiang T, Chen YD (2010) Climate changes and their impacts on water
resources in the arid regions: a case study of the Tarim River basin, China. Stoch Environ Res Risk
Assess 24:349–358

Page 15 of 16
Handbook of Climate Change Adaptation
DOI 10.1007/978-3-642-40455-9_91-1
# Springer-Verlag Berlin Heidelberg 2014

Index Terms:
Climate change 1
Climate variable 8
Impact assessment 4
Iran 7
Zayandeh-Rud 7

Page 16 of 16