Investigation of the effects of climate change on hydrological drought and pattern detection for severe climatic conditions using CCT and SWAT models in the semi-arid region - Case study: Karkheh river basin

doi:10.21203/rs.3.rs-2220517/v1

Download PDF

Research Article

Investigation of the effects of climate change on hydrological drought and pattern detection for severe climatic conditions using CCT and SWAT models in the semi-arid region - Case study: Karkheh river basin

https://doi.org/10.21203/rs.3.rs-2220517/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Drought causes an imbalance in the hydrological condition of the area. Climate change is exacerbating this situation. In this study, the hydrological drought under the influence of climate change in Karkheh river basin was investigated using the SRI index. For this purpose, the Soil and Water Assessment Tool model was calibrated (1990 to 2009) and validated (2010 to 2018) using data from 17 meteorological stations and 11 hydrometric stations. Then, based on the runoff simulated by the model, the index in all sub-basins for the base period (1990 to 2018) is calculated. By introducing the microscale results of 5 climate models in the Climate Change Toolkit program under RCP 2.6 and RCP8.5 scenarios to the SWAT model, the SRI index was simulated for the near future (2043 to 2071) and the distant future (2072 to 2100); and its intensity, duration and frequency were compared with the baseline period. The results show that hydrological drought will decrease in the near future for both scenarios; while in the distant future this result will be reversed. The CCT model includes the Critical Consecutive Day Analyzer (CCDA), whose application in the region showed that at least once a wet period similar to the 2018 flood conditions will be observed for the near future.There will also be at least one similar drought in 2014 for the distant future in the region. However, in previous climate studies, future events have not been calculated based on identifying the pattern of those events in the past.

Climate change

Hydrological drought

Extreme events

CCT

SWAT

Drought is a natural hazard with adverse impacts on the environment, agricultural activities, and water resources. In the literature, it is defined as a long-term dry period affecting different components of the hydrological cycle. As a complex phenomenon, quantitative determination of drought has always been a matter of concern (Vicente-Serrano et al. 2012; Vidal et al. 2010; Dai 2011) owing mainly to its dependence on various components of the water cycle and its development over long-term periods. Climate change is likely to change drought patterns and exacerbate the severity of drought events in the foreseeable future. Therefore, a more comprehensive approach to drought should be considered simultaneously to incorporate: 1. Various components of the hydrological cycle and their interactions; 2- Drought spatial and temporal characteristics using integrated methods; And 3- Future changes in the components under climate change scenarios. The existing literature mainly looks at one of the mentioned dimensions. Despite the importance of this perspective for the effective management of regional drought, there are limited attempts to consider different aspects of drought using a standard method (Peters 2006).

Drought is classified into meteorological, agricultural, hydrological and socio-economic categories. The first three types of drought reflect the physical characteristics of a drought phenomenon (ie, physical drought). Economic drought is related to water scarcity, the impact of which is manifested through socio-economic systems. Although droughts are caused by a lack of rainfall, hydrological drought is usually followed by a meteorological drought because it takes some time for the lack of rainfall to appear in various subsurface components of the hydrological system, such as soil moisture, groundwater and streams (Wilhite and Glantz 2009).

The adverse effects of droughts can be mitigated by monitoring their spatio-temporal distribution. A common measurement tool used for this purpose is the application of drought indices, which are mostly used for rainfall or runoff variables to assess the spatio-temporal characteristics of drought. For example, the Standard Runoff Index (SRI) is used to monitor hydrological droughts. Liu et al. (2012) used standard precipitation index (SPI), index (SRI) and Palmer Drought Intensity Index (PDSI) to predict future Oklahoma River Basin drought patterns. Their results showed that drought events will increase in the future (2010–2099). They also recommended PDSI and SRI as the most functional indicators for assessing drought risk. Vicente-Serrano et al. (2020) in a study described the world's hydrological and meteorological droughts under future climate change. Their results show that climatic and hydrological droughts are likely to increase during the 21st century, with approximately 30% of the world's land under water emission scenarios experiencing water shortages. Kang and Sridhar (2021) briefly assessed drought using climate forecast in the Mekong River Basin. They used the Soil and Water Assessment Tool (SWAT) model to simulate soil moisture, runoff and evapotranspiration, and to calculate three drought indicators to assess historical drought (1953–2016) and seasonal drought forecast. They concluded that 76.1% of the basin areas will experience reduced rainfall and rising temperatures, resulting in increased drought. Rodrigues et al. (2019) investigated the effect of climate change under diffusion scenarios on the flow and drought of basins in the Brazilian bio Cerrado. Their results showed that the duration, severity and frequency of meteorological and hydrological droughts will increase in future periods. Ashraf Vaghef et al. (2017) examined California climate change using the CCT model from 2020 to 2050. Their results show that in the coastal areas of Northern California, the frequency of wet seasons may be at risk of flooding. Kamali et al. (2017) assessed the risk of drought under climate change scenarios in the Karkheh river basin, and their results showed that meteorological, hydrological, and especially agricultural droughts will increase in the future.

In this study, runoff was simulated in several basin hydrometric stations using SWAT and SWAT-CUP hydrological models. The simulated runoff was then compared with the observational runoff based on the Nash-Sutcliffe coefficient to determine the accuracy of the observational data. Also, precipitation data, minimum temperature, maximum temperature were simulated using CCT program for two periods of near future (2043 to 2071) and distant future (2072 to 2100). These data were then entered into the SWAT model to simulate future runoff. Finally, changes in the severity, duration and frequency of hydrological drought in the next two periods compared to the base period (1990 to 2018) were examined by the SRI Drought Index. Then, using the CCT model capability to calculate critical periods (CCDA), dry and wet periods in the future were obtained. Using the flood and drought conditions of previous periods as a critical limit, it is possible to identify the frequency and frequency of floods and drought for the future and apply the right management. In this case, the damage caused by floods and droughts can be reduced.

CCT is developed to provide an easy-to-use platform for climate change studies. As management of big databases, bias correction and downscaling, and interpolation of climate data to finer resolution are essential processes in climate change studies; Ashraf Vaghefi et al. (2017) developed the software to consider all these vital tasks in one package. In this work, the Climate Change Toolkit (CCT) was used, which was developed with several objectives in mind. These were i) handling of big data, as it is required by climate change analyses, especially at large scales and long time periods, ii) easy and seamless calculation of necessary steps in climate change studies, such as data reformatting, data interpolation, downscaling and bias correction, and iii) projection of historical extreme events into the future by pattern recognition of past events. Therefore, in climate studies of the world's major watersheds, such as the Karkheh Basin, the CCT model is a good option for performing micro-scaling processes.

2.1 Study area

Karkheh River is the third high-water river in Iran, whose water is controlled by the largest reservoir dam ever built in Iran and the Middle East (Faiznia 2008). The Karkheh river basin spans in the geographical range between 46 °, 6ʹ to 49°, 10ʹ East longitude and 30 °, 58ʹ to 34 °, 56ʹ North latitude. It is one of the main basins in the west of the country with an area of 51527 km2, of which about 33674 km2 are located in mountainous areas and 1785.19 km2 are plains and foothills (Fig. 1). The vast basin of the Karkheh River has a variety of climatic conditions. The plain of Khuzestan and the southern parts of the basin are semi-arid with mild winters and hot and long summers. The northern parts and mountainous areas have cold winters and mild summers. The basin temperature varies from − 25°C to a maximum of 50°C. The mean annual rainfall in the basin varies from 300 to 800 mm per year, occurring mostly in winter. Climatically, Karkheh watershed belongs to semi-arid areas (Iran Water and Power Resources Development Company 2004).

2.2 Data

In this study, in order to calculate the meteorological drought index, climatic data of 17 stations of the Meteorological Organization were used. Also, to calculate the hydrological drought index, runoff data of 11 hydrometric stations located in Karkheh watershed were used (Tables 1 and 2). Both types of data were on a daily time scale and related to the statistical period 1990–2018. Data required to implement and calibrate the SWAT model include digital elevation model (DEM) map, land use, soil, climatic data, and runoff data from the Karkheh watershed. To extract precipitation data, minimum temperature and maximum temperature related to the future, two categories of data were used, which include: 1- Data of previous climate models (1950 to 2005): This data has been converted from NetCDF to ASCII format for easier access in the model and includes precipitation, maximum and minimum temperatures. 2. Future Climatic Data (2006 to 2100): These data are similar to those of previous climate models with different time periods.

Table 1

Meteorological stations of Karkheh watershed
Station	North latitude	East longitude	Altitude above sea level (m)
Ahvaz	48.4	31.2	22.5
Aleshtar	48.15	33.49	1567.2
Bostan	48	31.43	7.8
Boroujerd	48.45	33.55	1629
Dezful	48.23	32.24	143
Hamedan	48.32	34.52	1741.5
Ilam	46.26	33.38	1337
Kangavar	47.59	34.3	1468
Kermanshah	47.9	34.21	1318.6
Khorramabad	48.17	33.26	1147.8
Kuhdasht	47.39	33.31	1197.8
Malayer	48.51	34.15	1777.8
Mazo	48.31	32.47	450
Noorabad	48	34.3	1859.1
Pol-e-Dokhtar	47.43	33.9	713.5
Ravansar	46.39	34.43	1379.7
Safiabad	48.25	32.16	82.9

Table 2

Hydrometric stations of Karkheh watershed
Station	Longitude	Latitude	Altitude (meters)
Hamidiyeh	48° 26′00″	31° 30′00″	20
Pai-e-Pol	48° 09′00″	32° 25′00″	90
Pol-e-Zal	48° 05′00″	32° 49′00″	300
Pol-e-Dokhtar	47° 43′50″	33° 09′44″	700
Afrineh	47° 53′41″	33° 19′53″	820
Cham Anjir	48° 14′56″	33° 26′44″	1140
Sarab Seyed Ali	48° 12′00″	33° 47′13″	1530
Noorabad	47° 58′00″	34° 05′00″	1800
Do Ab-e-Merk	46° 47′00″	33° 46′06″	1290
Kakarza	48° 15′06″	33° 46′06″	1530
Pulchehr	47° 26′00″	34° 21′00″	1280

2.3 SWAT model

The SWAT model is a continuous time concept model that can be implemented in hourly, daily or long time steps. The structure of this model is based on sets of different mathematical equations and experimental formulas. In the SWAT model, it is possible for the user to specify the different managements that exist in each Hydrological Response Unit (HRU) for the model. The user is also able to define the type of irrigation system and irrigation time, beginning and end of planting period, time and amount of fertilization, pesticides and plowing and furrow time for the model. In addition to conventional management information, information about livestock grazing, automatic fertilization and water use management systems can be entered into the model (Arnold, 2010). To achieve these goals, this model uses equations related to climate, soil characteristics, topography, vegetation and land management practices instead of regression equations in the watershed. Hence, the physical processes related to water movement, the model using this data directly simulates sediment movement, crop growth and nutrient cycle. The hydrological cycle simulated by the SWAT model is based on water balance continuity relationships:

$${\text{S}\text{W}}_{\text{t}}={\text{S}\text{W}}_{0}+{\sum }_{\text{i}=1}^{\text{t}}({\text{R}}_{\text{d}\text{a}\text{y}}-{\text{Q}}_{\text{s}\text{u}\text{r}\text{f}}-{\text{E}}_{\text{a}}-{\text{W}}_{\text{s}\text{e}\text{e}\text{p}}-{\text{Q}}_{\text{g}\text{w}})$$

In the above relation, $\text{t}$ time (days), ${\text{S}\text{W}}_{\text{t}}$ the final amount of water in the soil (mm), ${\text{S}\text{W}}_{0}$ initial amount of water in the soil (mm), ${\text{R}}_{\text{d}\text{a}\text{y}}$ amount of precipitation on the $\text{i}$ day (mm), ${\text{Q}}_{\text{g}\text{w}}$ amount of return flow on the $\text{i}$ day (mm), ${\text{E}}_{\text{a}}$ is the amount of evapotranspiration on the $\text{i}$ day (mm), ${\text{Q}}_{\text{s}\text{u}\text{r}\text{f}}$ is the amount of surface runoff on the $\text{i}$ day (mm) and ${\text{W}}_{\text{s}\text{e}\text{e}\text{p}}$ is the amount of water entering from the soil profile to the unsaturated area of the soil on the $\text{i}$ day (mm).

In the SWAT model, there are two methods for estimating surface runoff: the SCS Curve Number Method and the Green and Ampt method. In this research, the curve number method has been used according to Eq. 2:

$${\text{Q}}_{\text{s}\text{u}\text{r}\text{f}}=\frac{{({\text{R}}_{\text{d}\text{a}\text{y}}-0.2\text{S})}^{2}}{({\text{R}}_{\text{d}\text{a}\text{y}}+0.8\text{S})}$$

Parameter $\text{S}$ changes spatially with changes in soil, land use, management and slope of the area and temporarily with changes in water in the soil. $\text{S}$ is defined as Eq. 3:

$$\text{S}=25.4(\frac{1000}{\text{C}\text{N}}-10)$$

In the above relations, ${\text{Q}}_{\text{s}\text{u}\text{r}\text{f}}$ is the surface runoff (mm), ${\text{R}}_{\text{d}\text{a}\text{y}}$ is the daily rainfall depth (mm), $\text{S}$ is the moisture retention parameter (mm) and $\text{C}\text{N}$ is the permeability parameter of the basin in terms of permeability. This parameter is dimensionless and is a function of soil type, available vegetation and soil moisture.

The SWAT model uses three different methods to calculate potential evapotranspiration. These methods include Hargreaves-Samani, Priestley-Taylor, and the Penman-Mantis method. The choice and application of each of these methods depends on the available climatic data. In this study, the Hargreaves-Saman method has been used, which requires only the parameter of incoming temperature and solar radiation (depending on latitude) to calculate the potential evapotranspiration (Arnold, 2010).

2.4 SWAT model sensitivity analysis

In hydrological models, many physical parameters are interacting and accurate measurement is not possible in large basins or requires a lot of time and money. To achieve this goal, it is necessary to check the efficiency of the converter before using it through sensitivity analysis, calibration and validation. The first step in evaluating the efficiency of the model is to perform sensitivity analysis to determine the sensitive parameters in the implementation of the model for calibration. Through sensitivity analysis, the amount of change in the output values of the model is calculated in exchange for the change for inputs and the sensitive parameters in the implementation of the model are determined. The methods used to perform sensitivity analysis are generally classified into two groups of local and global analysis. Local analysis, also known as the one-factor method at a time, actually examines the model simulation response to the continuous changes of each parameter under constant other parameters (Morgan 2011). In the present study, sensitivity analysis was performed by local analysis.

In the present study, in order to calibrate the model, the SUFI2 algorithm (Sequential uncertainty fitting algorithm) was used in the framework of SWAT-CUP (SWAT Calibration and Uncertainty Programs) software. In this program, it is assumed that each parameter is evenly distributed in a domain with a certain uncertainty. The SUFI2 algorithm is executed in 9 steps. In this algorithm, the objective function is defined differently and the calibration of the model continues until the values of each objective function reach the optimal value. (Abbaspour, 2007). In this study, the Nash-Sutcliffe coefficient was used to optimize the objective function, which can be calculated from the following relation:

$$\text{N}\text{S}\text{E}=1-\frac{{\sum }_{\text{i}=1}^{\text{n}}{({\text{Q}}_{\text{m}.\text{i}}-{\text{Q}}_{\text{s}.\text{i}})}^{2}}{{\sum }_{\text{i}=1}^{\text{n}}{({\text{Q}}_{\text{m}.\text{i}}-\stackrel{-}{{\text{Q}}_{\text{m}}})}^{2}}$$

In the above relation, $\text{n}$ number of observations, ${\text{Q}}_{\text{m}.\text{i}}$ and ${\text{Q}}_{\text{s}.\text{i}}$ are measured and estimated runoff values (cubic meters per second) and $\stackrel{-}{{\text{Q}}_{\text{m}}}$ mean measured runoff values (cubic meters per second), respectively. The numerical value of the $\text{N}\text{S}\text{E}$ coefficient varies from infinite negative to 1 (optimal value) and the closer it is to one, indicates that the model has a better estimate. Generally, if the Nash-Sutcliffe index is more than 0.75, the efficiency of the excellent model is considered satisfactory, if it is between 0.5 and 0.75, and if it is less than 0.5, it is considered unacceptable (Nash and Sutcliffe, 1970).

After calibration, the validity of the model is measured using the parameters obtained in the calibration phase and the values of observations that were not used in the calibration phase. In case of acceptable simulation, the model will be ready for use.

In the present study, the measured data of 11 hydrometric stations of Karkheh watershed have been used to calibrate and validate the model simulation results. The simulation time was selected for calibration (1990–2009) and for validation (2010–2018) and the model was implemented in which the first three years of the time, namely 1990, 1991 and 1992, were considered as the balancing period (Warm up). Were determined in order to achieve optimal performance in model calibration.

2.5 Climate change and climate models

Climate is defined as the average, or more precisely, the statistical description of the surface variables of a climatic system, such as precipitation and temperature, over time. Climate change refers to the change in the climatic condition of a region, including change in the mean or other statistical features of surface variables over time (10 years or more). Drought is a recurring, temporary meteorological event that results from a lack of rainfall relative to normal or the expected climatic value that can occur in any climate, but its characteristics vary considerably from region to region. There are different types of droughts; Meteorology, agriculture, hydrology and socio-economic. Different indicators have been developed and applied for drought monitoring in each of these groups. Each drought phenomenon is mainly characterized by three characteristics of severity, duration and frequency of occurrence. The characteristics of drought may not change much over time or may change due to climate change.

The coupled Atmospheric-Ocean General Circulation Models (AOGCMs) are a group of climatic models proposed to promote these models with a more complex structure and show the chemical and biological interactions of the climatic system. On the other hand, the scientific basis, the quality of the observational data, the assessment of extreme events and the models used in the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC) have been significantly enhanced than previous reports, thus reducing uncertainty in some aspects of climate change (IPCC 2014).

The new climate models are based on the framework of the Stage 5 Coordinated Climate Models Coordination Working Group Phase 5 (CMIP5) and the Representative Concentration Trajectory (RCP) scenarios that simulate future climate change at regional and global scales. The CMIP5 GCM models and Earth System Models (ESMs) incorporate the interaction of the atmospheric component with land use and vegetation into modeling. The second-generation Canadian Land System Model (CanESM2) is a comprehensive model and the fourth generation of paired general circulation models (CGCM4) and is part of the CMIP5 model series, consisting of atmospheric, ocean and surface components. The historical and future projection periods of the CanESM2 are 1850–2005 and 2006–2100, respectively, and its scenarios include RCP2.6, RCP4.5, RCP6 and RCP8.5. The definitions of these scenarios are given below. Table 3 lists some specifications of CMIP5 models.

Table 3

Specifications of some models in the IPCC database (Rezazadeh et al. 2018)
Model name	Founding group (country)	Atmospheric resolution (degrees)
GFDL-ESM2G	NOAA GFDL(United States of America)	2˟2.5
IPSL-CM5A-LR	IPSL(France)	1.9˟75.3
NorESM-ME	NCC(Norway)	2.5˟9.1
BCC-CSM1.1	BCC (China)	2.8˟8.2
CESM1-BGC	NSF-DOE-NCAR(United States of America)	1.25˟9
CanESM2	CCCMA(Canada)	2.8˟8.2

2.6 The Climate Change Toolkit

The Climate Change Toolkit (CCT) is an effective tool for extraction, interpolation and bias-correction of data obtained from global General Circulation Models (GCM). This model is also used to analyze extreme events such as drought and flood. This model is connected to five global databases of ISI-MIP. It also uses four RCP scenarios. Table 4 lists the names of global databases and emission scenarios in the CCT model. In this study, all five models embedded in CCT were used to project maximum and minimum precipitation and temperature under two scenarios of RCP2.6 and RCP8.5. The required data in this software include: 1) Observational data of precipitation, maximum temperature and minimum temperature obtained from the statistics of 17 meteorological stations of the Karkheh river basin; 2) historical global data (1970 to 2005): These data are available on a scale of 0.5 degrees, which can be used in the absence of observational data; 3) Data from previous climate models (1950 to 2005): This data were reformatted from NetCDF to ASCII for easier access to the model. These data include precipitation, maximum and minimum temperature. 4) Future Climate Data (2006 to 2099): These data are similar to those of previous climate models but in different periods.

Table 4

Names of global databases and emission scenarios in the CCT model (Rezazadeh et al. 2018)
Database name in CCT	Model ISI-MIP	Scenario name in CCT	Scenario name in ISI-MIP
GCM1	GFDL-ESM2M	Scenario1	RCP2.6
GCM2	HadGEM2-ES	Scenario2	RCP4.5
GCM3	IPSL-CM5A-LR	Scenario3	RCP6
GCM4	MIROC	Scenario4	RCP8.5
GCM5	NorESM-M

2.7 Validation of CCT projected data

To validate the models, the coefficient of determination or R2 (Eq. 5) was first calculated to compare the temperature and precipitation simulated by the models based on both emission scenarios and the actual values recorded at the stations. Given that the R2 coefficient alone is not a suitable criterion for model evaluation, the mean absolute error or MAE and the root mean square error or RMSE (Equations 6 and 7) were also computed and presented.

${\text{R}}^{2}=\frac{{(\sum ({\text{P}}_{\text{i}}-\stackrel{-}{\text{P}})({\text{O}}_{\text{i}}-\stackrel{-}{\text{O}}))}^{2}}{\sum {({\text{P}}_{\text{i}}-\stackrel{-}{\text{P}})}^{2}\sum {({\text{O}}_{\text{i}}-\stackrel{-}{\text{O}})}^{2}}$	(5)
$\text{M}\text{A}\text{E}=\frac{\sum _{\text{i}=1}^{\text{n}}\left\|{\text{P}}_{\text{i}}-{\text{O}}_{\text{i}}\right\|}{\text{n}}$	(6)
$\text{R}\text{M}\text{S}\text{E}=\sqrt{\frac{\sum _{\text{i}=1}^{\text{n}}{\left({\text{P}}_{\text{i}}-{\text{O}}_{\text{i}}\right)}^{2}}{\text{n}}}$	(7)

Where ${\text{P}}_{\text{i}}$ is the predicted values, ${\text{O}}_{\text{i}}$ is the measured values, n is the number of samples used, $\stackrel{-}{\text{P}}$ is the average of the predicted values and $\stackrel{-}{\text{O}}$ is the average of the measured values. In situations where the estimated and observed values are equal, RSME and MAE values are idea, equal to zero and ${\text{R}}^{2}$ is 1 (Dawson et al. 2007).

2.8 The CCT uncertainty evaluation

There are several uncertainty sources associated with different stages of climate variables simulation by AOGCM models such as those related to the simulation of climate models at regional scales, application of various downscaling methods, and emission scenarios (Feng et al. 2011).

One method of analyzing the uncertainty of climate models and emission scenarios is the model parameter weighting in which the selected models are weighed based on the deviation of the simulated meteorological parameter in the base period from the mean observational data according to Eq. 8 (Harris and Wilby 2006). According to this method, the models obtaining a high weight in the past modeling are expected to achieve somehow the same weight in predicting the future and, therefore, are selected as the optimum model (Ekstrom and Fowler 2009)

$${\text{W}}_{\text{i}}=\frac{{\left(\frac{1}{\varDelta {\text{F}}_{\text{i}}}\right)}^{2}}{\sum _{\text{i}=1}^{\text{n}}{\left(\frac{1}{\varDelta {\text{F}}_{\text{i}}}\right)}^{2}}$$

Where ${\text{W}}_{\text{i}}$ is the weight of each model in the studied month, $\varDelta {\text{F}}_{\text{i}}$ is the long-term mean deviation of the simulated parameter by each model in the base period from the mean actual or observed data and n is the number of models.

Due to climate change, climatic fluctuations have increased and events such as tornadoes, floods, hail, and droughts are expected to be more intense and occur at shorter intervals. Current estimates indicate that the most significant potential environmental changes across the world are driven by climate change and include those influencing the components of the hydrological cycle such as floods, droughts and storms and challenge the future water resource management for human and ecosystem development. The increasing frequency and severity of floods and droughts have been confirmed by the latest IPCC report, which discusses the evidence relating to the occurrence and visible effects of climate change in the present (IPCC 2007).

As mentioned, in the CCT model, the length of wet and dry periods can be examined by defining the threshold for wet and dry days. Table 5 is used to define the threshold.

Table 5

Definition of dry and wet period length threshold
Region	Dry Period	Wet Period
Tropical Regions	Period Length > 60 days precipitation < 2 mm day^− 1.and. max temperature > 30° C	Period Length > 2 days precipitation > 50 mm day^− 1
Semi-Arid Regions	Period Length > 120 days precipitation < 2 mm day^− 1.and. max temperature > 35° C	Period Length > 1 day precipitation > 20 mm day^− 1

2.9 SRI index

The Standardized Runoff Index (SRI) was selected to characterize the hydrological drought. The methodology of this index is identical to that of the standardized precipitation index (SPI) method, except that the SRI applies runoff instead of precipitation (Wang et al. 2011; Vu et al. 2015). To calculate the SPI index as one of the most well-known drought indices, the cumulative probability function was first calculated by fitting the gamma distribution on the monthly rainfall data or the total rainfall in any desired period, followed by its transformation to the normal cumulative distribution. SPI values were calculated using Eq. 9 (Wu et al. 2007):

$$\text{S}\text{P}\text{I}=\frac{{\text{X}}_{\text{i}}-\stackrel{-}{\text{X}}}{{\text{S}}_{\text{x}}}$$

Where ${\text{X}}_{\text{i}}$is rainfall per month, $\stackrel{-}{\text{X}}$ is the average rainfall at a given time scale, ${\text{S}}_{\text{x}}$ is the standard deviation of rainfall at a given time scale. In this study, the runoff parameter for the base period and the next two periods was simulated by the SWAT model. Based on the results of numerous studies, drought indices are more suitable at long-term scales (12 or 24 months) for examining the effects on water resources and future periods (Buttafuoco et al. 2015). Therefore, the SRI index was calculated on a 12-month scale using Visual Basic programming in Excel software. The determination of different drought classes of this index was investigated based on the SPEI index of Edwards and McKee (1997) classification (Table 6).

Table 6

Drought Classes (Edwards and McKee 1997)
Drought class	Index value
Extremely wet	2 and more
Moderately wet	1.5 to 1.99
Slightly wet	1 to 1.49
Normal	-0.99 to 0.99
Mild drought	-1.49 to -1
Moderate drought	-1.99 to -1.5
Extreme drought	less than − 2

3.1 Results of implementation and calibration of SWAT model

Table 7 presents the values obtained from the model performance evaluation index in runoff simulation for all studied stations. As can be seen, the Nash-Sutcliffe coefficient and the coefficient of determination are close to one in all hydrometric stations except the five stations of Sarab Seyed Ali, Pulchehr, Noorabad, Hamidiyeh and Pai-e-Pol, in the calibration and validation periods. In other words, the coefficients obtained in the other 6 stations are more than 0.5, which indicates that the model has an acceptable ability to simulate runoff. In Sarab Seyed Ali, Pulchehr and Noorabad stations, the SWAT model could not simulate the runoff well, which could be due to the location of these stations in the high areas of the basin and tributaries, and as a result, it is snow-covered. Lack of proper distribution of meteorological stations in these areas makes the model unable to properly simulate snow runoff. In Hamidiyeh and Pai-e-Pol stations, the SWAT model could not establish an acceptable relationship between the observed and simulated runoff due to the impact of the construction of Karkheh Dam on the hydraulic process of the river flow. The results of calibration and validation of the SWAT model (observational data, simulated and 95% uncertainty estimation band) as a sample for Afrineh hydrometric station are shown in Figs. 2 and 3, respectively. These two figures are outputs of SWAT-CUP software.

Table 7

Statistical comparison of monthly simulated and observational discharges in the calibration period
Hydrometric station	Statistical period length (years)		Nash – Sutcliffe		The coefficient of determination
Hydrometric station	Validation	Calibration	Validation	Calibration	Validation	Calibration
Hamidiyeh	9	20	-0.04	-0.19	0.22	0.02
Pai-e-Pol	9	20	0.18	0.02	0.29	0.47
Pol-e-Zal	9	20	0.68	0.56	0.8	0.87
Pol-e-Dokhtar	9	20	0.75	0.73	0.75	0.86
Afrineh	9	20	0.7	0.74	0.73	0.85
Cham Anjir	9	20	0.7	0.76	0.75	0.88
Sarab Seyed Ali	9	20	0.38	0.25	0.52	0.42
Noorabad	9	20	0.4	0.22	0.4	0.45
Do Ab-e-Merk	9	20	0.61	0.68	0.68	0.84
Kakarza	9	20	0.65	0.73	0.68	0.86
Pulchehr	9	20	0.42	0.11	0.42	0.22

3.2 Results of validation of simulated data by CCT implementation

Table 8 shows the validation results of the models used to microscale the GCM models for precipitation data. Tables similar to this table were created for maximum temperature and minimum temperature. These tables are based on the average accuracy of the models used in each of the 17 meteorological stations. Based on the statistical indicators shown in these tables, the model with the highest R2 and the lowest MAE and RMSE was chosen as the best and most reliable model. Hence, as shown in Table 7, the GFDL-ESM2 had the best performance in simulating precipitation in the Karkheh river basin under both RCP8.5 and RCP 2.6 scenarios. Based on the maximum temperature table, the HadGEM2-ES model is more suitable for simulating this data than the other models in both scenarios. The minimum temperature table also shows that the MIROC model is the most efficient in both scenarios for simulating this data. In all three tables, the other models, although having acceptable values of R2, are not recommended due to the relatively high degree of calibration characteristics.

Table 8

Validation of the CCT simulated precipitation in Karkheh River Bains
Scenario	Model	RMSE	MAE	R²
RCP2.6 RCP8.5	GFDL-ESM2	3.2 4.1	1.1 1.5	0.9989 0.9971
RCP2.6 RCP8.5	HadGEM2-ES	4.5 5.8	2.3 3.5	0.9906 0.9908
RCP2.6 RCP8.5	IPSL-CM5A-LR	7.2 6.7	5.6 4.2	0.9901 0.9902
RCP2.6 RCP8.5	MIROC	5.6 8.1	3.2 5.4	0.8946 0.8934
RCP2.6 RCP8.5	NoerESM1-M	5.3 4.6	3.4 2.6	0.7936 0.8961

3.3 Results of the CCT uncertainty evaluation

The results of weighted GCMs in precipitation forecast show that except for three summer months, in other months both scenarios of GFDL-ESM2 model have the highest weight among other models. The results of weighted GCMs in predicting the maximum temperature show that in all months, especially in the three summer months, both scenarios of the HadGEM2-ES model have the highest weight in estimating this parameter. Regarding the minimum temperature, the MIROC model scenarios, with a slight difference compared to other GCM models, have the highest accuracy for estimating this parameter.

In general, the study of weight changes of GCM models shows that the weight of GCM models to estimate the maximum temperature and precipitation has more variability than their weight to determine the minimum temperature. In other words, GCM models have a closer accuracy when used to estimate the minimum temperature, as long as they are to be used to determine the maximum temperature and precipitation.

After determining the best model, precipitation data, maximum temperature and minimum temperature produced in the selected models and scenarios were analyzed to investigate the climate change situation of temperature and precipitation in the future.

The results of using the critical period calculation operator in the CCT model showed that in the near future, the average cumulative frequency of wet periods will show a large number of wet periods (between 35 and 42 times) in Karkheh watershed. While in the distant future, the average cumulative frequency of dry periods will show a large number of dry periods (between 28 and 47 times) in this watershed.

3.4 Changes in hydrological drought compared to the base period

Using the discharge parameter, the standardized runoff index (SRI) was calculated. This index was obtained for the base time in a 12-month period for 11 hydrometric stations. For the near and distant future, the SWAT model calculated the outflow runoff for 57 sub-basins. Using the Theissen method, the weighted average runoff output of the sub-basins affected by each hydrometric station was determined. Thus the future runoff for each station was calculated.

The SRI index was generated by the GFDL-ESM2, HadGEM2-ES, IPSL-CM5A-LR, MIROC and NoerESM1-M models in three basic periods (2018 − 1990), near future (2071 − 2043) and distant future (2100 − 2072) with two scenarios of RCP2.6 and RCP8.5.

In this study, the results of comparisons between SRI indices in different time periods are the average of these five climatic models; because for the SRI index, similar to the uncertainty and validation of five models for precipitation data, maximum temperature and minimum temperature, the results showed that all five models have similar and acceptable accuracy for future SRI data reconstruction.

To evaluate the hydrological drought situation in the whole study area, the weighted average weight of SRI index on an annual scale over 29 years for the base period and the next two periods was determined using Theissen method. The results of the study of the relationship between the SRI values of the base period and the SRI values of the next two near and far periods on a twelve-month scale are shown in Figs. 4 (A) and 4(B), respectively. As shown in Fig. 4 (A), the SRI index values for both RCP2.6 and RCP8.5 scenarios and especially for RCP8.5 in the near future will be more positive than the baseline period. The more positive value of the SRI index in the RCP8.5 scenario can be attributed to the greater increase in rainfall and the lower temperature increase in the near future for this scenario compared to RCP2.6. According to Fig. 4(B) for the distant future, the values of this index in both scenarios RCP2.6 and RCP8.5 and especially for RCP8.5 are more negative than the base period. The more negative value of the SRI index in the RCP8.5 scenario can be attributed to the smaller increase in rainfall and the greater temperature increase in the distant future for this scenario compared to RCP2.6. Inverse distance weighting (IDW) method was used in the ARC-GIS software to investigate the changes of hydrological drought in the base period and two periods in the near and distant future, as shown in Figs. 5 and 6, respectively. According to Fig. 5, as in Fig. 4 (A), the hydrological drought will decrease in the near future for both RCP2.6 and RCP8.5 scenarios. However, according to Fig. 6, as in Fig. 4(B), the conditions will be completely opposite, and in both scenarios, the conditions of the watershed will be drier than in the hydrological period. In Figs. 5 and 6, concentric circular spots may be the effect of hypersensitivity of the interpolation method to low-dispersion data.

Probability density functions (PDF) were used to investigate the changes in the intensity of hydrological drought in the near and distant future periods compared to the baseline period which can be seen in Figs. 7(A) and 7(B), respectively. Figure 7(A) shows that in the near future for RCP2.6, the severity of the drought will be the same as the baseline period, except that the probability of its occurrence is almost doubled. However, in this time period for RCP8.5, the probability density function is shifted to the right, ie the intensity of the drought will be less than the base period, but the probability of its occurrence is less than doubled. As shown in Fig. 7(B) in the distant future for RCP2.6, the severity of the drought is the same as the baseline period, except that the probability of occurrence is almost doubled. While in this time period for RCP8.5, the probability density function is shifted to the left, ie the intensity of the drought will be higher than the base period, but the probability of its occurrence is less than doubled.

According to Figs. 4 to 7, the intensity of hydrological drought is in the normal range in all three time periods. The normal range is the intensity of the drought between the very severe wet season and the very severe drought.

To assess the frequency and duration of droughts, when the SRI index is less than zero for at least two months, drought is said to have occurred (Kamali et al. 2017). Accordingly, the results of studying the frequency changes of hydrological drought in the near future and the distant future compared to the base period are shown in Figs. 8(A) and 8(B), respectively. Figure 8(A) shows that the average number of hydrological droughts in the base period is RCP2.6 in the near future and RCP8.5 in the near future are 9, 7 and 5, respectively; this means that the number of droughts has decreased in the near future. According to Fig. 8(B), the average number of hydrological droughts in the base period is RCP2.6 far future and RCP8.5 far future are 9, 13 and 16, respectively. In other words, the number of droughts will increase in the distant future. The results of the study of changes in the duration of hydrological drought in the near future and the distant future compared to the base period are shown in Figs. 9(A) and 9(B), respectively. According to Fig. 9(A), the average duration of the hydrological drought in the base period is RCP2.6 in the near future and RCP8.5 in the near future at 14, 11 and 10 months, respectively, which means a reduction in the duration of the drought in the near future. Figure 9(B) shows that the average duration of hydrological drought in the base period, RCP2.6 distant future and RCP8.5 distant future are 14, 16 and 23 months, respectively. In other words, the duration of droughts will increase in the distant future.

Drought is classified as a natural disaster that has major effects on parts of an ecosystem. Although it is not possible to prevent its occurrence, but its measures can reduce the negative effects. Since the study area has the largest reservoir dam in the Middle East on the Karkheh River, so the study of various droughts, including hydrological drought in the basin is very important.

In this study, runoff was simulated using SWAT model input data. These data include digital elevation model (DEM), FAO soil map, waterway network, land use and meteorological station data of Karkheh watershed (from 1990 to 2018) including precipitation, minimum temperature and maximum daily temperature. Then the results of runoff simulation were entered into SWAT-CUP software and calibrated and validated using SUFI-2 algorithm. Two-thirds of the data were used for calibration and one-third for validation. In order to evaluate the calibration and validation results, two coefficients of determination and Nash-Sutcliffe were used. The calibration results of the model show a good correlation between the simulated data and the observed values of the hydrometric stations of Karkheh watershed because the two coefficients of determination and Nash-Sutcliffe have obtained acceptable values in most of the hydrometric stations of the basin. Determination and Nash-Sutcliffe coefficients in Sarab Seyed Ali, Pulchehr, Noorabad, Hamidiyeh and Pai-e-Pol stations have been obtained less than acceptable values. It is natural that the SWAT model underestimated the Nash-Sutcliffe coefficients at these stations. In other words, in these stations, the amount of runoff is estimated to be less than the actual amount, which other researchers have pointed out this weakness of the model (Abbaspour et al., 2007). The reason for the reduction of these coefficients in Sarab Seyed Ali, Pulchehr and Noorabad stations is the weakness of the model in simulating the snowmelt process, which is more important in mountainous basins. Due to the mountainous nature of the basin at the location of these three hydrometric stations, the discrepancy in runoff estimation can be attributed to the weakness of the model in proper simulation of the snowmelt process. Another reason for such a problem could be the inaccuracy of the input data, such as the lack of meteorological stations and their lack of proper distribution, because as we know, the SWAT model converts the data of these stations into runoff at different time scales. Regarding human intervention in the environment and runoff changes in the downstream parts of Karkheh, especially stations such as Hamidiyeh and Pai-e-Pol, which are related to Karkheh Dam, it can be said that the construction of this hydraulic structure greatly reduces runoff in these stations (Mokhtari et al., 2020). In the next step, the output of GFDL-ESM2, HadGEM2-ES, IPSL-CM5A-LR, MIROC and NoerESM1-M models under RCP 2.6 and RCP8.5 scenarios were scaled by CCT program in 17 meteorological stations. Then, to check the uncertainty of the models and scenarios, the output of the models in the near future and the distant future was compared to the base period. Thus, the best model and scenario for generating precipitation and maximum temperature, minimum temperature and standard runoff index (SRI) data were selected. The overall results show that GCM models operate with different accuracy in estimating precipitation and future temperature, which is estimated based on the weight of the model at each station. The results of weighted GCMs in precipitation prediction showed that GFDL-ESM2 model under both scenarios has the highest weight among other models. In addition, the highest weight in estimating the maximum temperature was obtained by the HadGEM2-ES model under both scenarios. Finally, the MIROC model, with a slight difference compared to other GCM models, was recognized as the best model in estimating the minimum temperature. GCM models perform similarly in minimum temperature prediction, while their performance in precipitation and maximum temperature prediction is significantly different. Annual rainfall will increase in both future periods and especially in the near future. This result contradicts the results of Seidi et al. (2011) in the Karkheh Basin who used only the HADCM3 model. The maximum and minimum annual temperatures will increase in most cases in the distant future. These results were observed for the Karkheh Basin in the studies of Abrishamchi et al. (2012) as well as Seidi et al. (2011). This result is also consistent with the results of the Kang and Sridhar (2021) study in the Mekong River Basin. In addition, hydrological drought will be reduced in the near future for both RCP2.6 and RCP8.5 scenarios. In the distant future, for both scenarios, the basin will experience more hydrological drought. This result is consistent with the study of Kamali et al. (2017) in the Karkheh watershed. In other watersheds it is similar to the study of Zhao et al. (2019), Vicente-Serrano et al. (2020) and Rodrigues et al. (2019). The results of using the critical period calculation operator in the CCT model showed that in the near future, the average cumulative frequency of wet periods is higher than dry periods, which may increase the risk of flooding. This result is similar to the study by Vaghef et al. (2017) in California. In the distant future, the situation will be quite the opposite of the near future, and the average cumulative frequency of dry periods is higher than wet periods.

Therefore, the need to pay more attention to water resources management in Karkheh basin based on climate scenarios in the future should be on the agenda of managers and researchers. One of the limitations of the CCT program is the use of only 5 climate models. It is suggested that other types of droughts, climate models and scenarios be considered in future studies.

Acknowledgements

We thank USGS (The United States Geological Survey), FAO (Food and Agriculture Organization), ESRI (Environmental Systems Research Institute), Meteorological Organization of Iran[1], Ministry of Energy of Iran[2] and the Water Weather Energy Ecosystem Technology and Data Institute[3] for facilitating data collection. We also thank Shahrekord University.

Funding

No funds, grants, or other support were received during the preparation of this manuscript.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Fahimeh Mokhtari, Saeed Soltani and Afshin Honarbakhsh. The first draft of the manuscript was written by Fahimeh Mokhtari and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

[1] https://www.irimo.ir/eng/index.php

[2] https://moe.gov.ir/?lang=en-us

[3] www.2w2e.com

Abbaspour K, Yang J, Maximov I, Siber R, Bogner K, Mieleitner J, Zobrist J, Srinivasan R (2007) Modelling hydrology and water quality in the prealpine/alpine Thur watershed using SWAT. Journal of Hydrology 333:413– 430. https://doi.org/10.1016/j.jhydrol.2006.09.014
Abrishamchi A, Jamali, S, Madani K, Hadian S (2012) Climate change and hydropower in Iran’s Karkheh river basin. World Environmental and Water Resources Congress, 20-24 May, Albuquerque, New Mexico, United States. https://doi.org/10.1061/9780784412312.336
Arnold J, Srinivasan R, Muttiah S, Williams R (2010) Large area hydrologic modeling and assessment part I: Model development. Water Resources Association 34: 73-89. https://doi.org/10.1111/j.1752-1688.1998.tb05961.x
Ashraf Vaghefi S,Abbaspour N, Kamali B, Abbaspour K (2017) A toolkit for climate change analysis and pattern recognition for extreme weather conditions - Case study: California-Baja California Peninsula. Environmental Modelling & Software 96:181-198. https://doi.org/10.1016/j.envsoft.2017.06.033
Buttafuoco G, Caloiero T, Coscarelli R (2015) Analyses of drought events in calabria (southern Italy) Using Standardized Precipitation Index. Water Resources Management 29: 557-573. https://doi.org/10.1007/s11269-014-0842-5
Dai A (2011) Drought under global warming: A review. WIREs Climate Chang 2: 45–65. https://doi.org/10.1002/wcc.81
Dawson C, Abrahart R, See L (2007) Hydro Test: a web-based toolbox of evaluation metrics for the standardized assessment of hydrological forecasts. Environmental Modelling & Software 22: 1034-1052. https://doi.org/10.1016/j.envsoft.2006.06.008
Edwards D, McKee T (1997) Characteristics of 20th century drought in the United States at multiple time scales. climatology report number 97-2. http://hdl.handle.net/10217/170176
Faiznia S (2008) Applied sedimentology with emphasis on soil erosion and sediment production. Gorgan University of Agricultural Sciences and Natural Resources Publications, Iran
Feng J, Lee D, Fu C,Tang J, Sato Y, Kato H, Mcgregor J, Mabuchi K (2011) Comparison of four ensemble methods combining regional climate simulations over Asia. Meteorology and Atmospheric Physics 111(1):41-53. https://doi.org/10.1007/s00703-010-0115-7
Fowler H, Ekstrom M (2009) Multi-model ensemble estimates of climate change impacts on UK seasonal precipitation extremes. International Journal of Climatology 29: 385-416. https://doi.org/10.1002/joc.1827
IPCC (2014) Synthesis Report. Contribution of working groups I, II and III to the fifth assessment report of the intergovernmental panel on climate change [Core Writing Team, R.K. Pachauri and L.A. Meyer (Eds.)]. IPCC, Geneva, Switzerland.
IPCC-TGICA (2007) General guidelines on the use of scenario data for climate impact and adaptation assessment. Version 2. Prepared by T.R. Carter on behalf of the Intergovernmental Panel on Climate Change, Task Group on Data and Scenario Support for Impact and Climate Assessment. 66 pp.
Iran Water and Power Resources Development Company (2004) Report of basic studies in Karkheh area: hydrology.Iran.
Kamali B, Houshmand D, Yang H, Abbaspour K (2017) Multilevel drought hazard assessment under climate change scenarios in semi-arid regions—A case study of the Karkheh river basin in Iran. Water 9, 241. https://doi.org/10.3390/w9040241
Kang H, Sridhar V (2021) A near‐term drought assessment using hydrological and climate forecasting in the Mekong River Basin. International Journal of Climatology 41: E2497-E2516. https://doi.org/10.1002/joc.6860
Liu L, Hong Y, Bednarczyk C, Yong B, Shafer M, Riley R, Hocker J (2012) Hydro-climatological drought analyses and projections using meteorological and hydrological drought indices: A case study in Blue river basin, Oklahoma. Water Resources Management 26: 2761–2779. https://doi.org/10.1007/s11269-012-0044-y
Mokhtari F, Honarbakhsh A, Soltani S, Abdolahi K, Pajohesh M (2020) Evaluating hydrological SWAT model in runoff simulation of Karkheh Watershed. Desert Ecosystem Engineering 9(27): 47-58. http://dx.doi.org/10.22052/deej.2020.9.27.25
Morgan R, Nearing M (2011) Handbook of erosion modelling. Blackwell Publishing Ltd, United States of America.
Nash J, Sutcliffe J (1970) River flow forecasting through conceptual models: Part 1. A discussion of principles. Journal of Hydrology 10: 282‐290. https://doi.org/10.1016/0022-1694(70)90255-6
Peters E, Bier G, Lanen H, Torfs P (2006) Propagation and spatial distribution of drought in agroundwater catchment. Hydrology 321: 257–275. https://doi.org/10.1016/j.jhydrol.2005.08.004
Rezazadeh M, Tahami M, Kermani M (2018) Practical training and theory of SWAT and SWAT-CUP models including CCT climate change model. Kerman Cultural Services, Iran. pp. 237-256.
Rodrigues J, Viola M, Alvarenga L, Mello C, Chou S, Oliveira V, Uddameri V, Morais M (2019) Climate change impacts under representative concentration pathway scenarios on streamflow and droughts of basins in the Brazilian Cerrado biome. International Journal of Climatology 40: 2511-2526. https://doi.org/10.1002/joc.6347
Seidi A, Khoshakhlagh F, Shamsipour A (2011) Investigation of the effects of climate change on water resources of Karkheh watershed. Dissertation, University of Tehran
Vicente-Serrano S, Begueria S, Lorenzo-Lacruz J, Camarero J et al. (2012) Performance of drought indices for ecological,agricultural, and hydrological applications. Earth Interact 16: 1–27. https://doi.org/10.1175/2012EI000434.1
Vicente‐Serrano S, Dominguez‐Castro F, McVicar T, Tomas‐Burguera M, Pena‐Gallardo M, Noguera I, Lopez‐Moreno J, Peña D, Kenawy A (2020) Global characterization of hydrological and meteorological droughts under future climate change: The importance of timescales, vegetation‐CO2 feedbacks and changes to distribution functions. International Journal of Climatology 40: 2557-2567. https://doi.org/10.1002/joc.6350
Vidal J, Martin E,Franchisteguy L, Habets F, Soubeyroux J, Blanchard M, Baillon M (2010) Multileveland multiscale drought reanalysis over France with the Safran-Isba-Modcouhydrometeorological suite. Hydrology and Earth System Sciences 14: 459–478. https://doi.org/10.5194/hess-14-459-2010
Vu M, Raghavan V, Liong S (2015) Ensemble climate projection for hydro-meteorological drought over a river basin in Central Highland, Vietnam. KSCE Journal of Civil Engineering 19: 427-433. https://doi.org/10.1007/s12205-015-0506-x
Wang D, Hejazi M, Cai X, Valocchi A (2011) Climate change impact on meteorological agricultural and hydrological drought in centeral Illinois.Water Resources Research 47. https://doi.org/10.1029/2010WR009845
Wilby R. Harris I (2006)A framework for assessing uncertainties in climate change impacts: low flow scenarios for the River Thames, UK. Water Resource Research 42: 1-10. https://doi.org/10.1029/2005WR004065
Wilhite D, Glantz M (2009) Understanding the drought phenomenon: The role of definitions. Water International 10, 111–120. https://doi.org/10.1080/02508068508686328
Wu H, Svoboda M, Hayes M, Wilhite D, Wen F (2007) Appropriate application of the standardized precipitation index in arid locations and dry seasons. International Journal of Climatology 27: 65-79. https://doi.org/10.1002/joc.1371
Zhao P, Haishen L, Huicai Y, Wenchuan W, Guobin F (2019) Impacts of climate change on hydrological droughts at basin scale: A case study of the Weihe River Basin, China. Quaternary International 513: 37-46. https://doi.org/10.1016/j.quaint.2019.02.022

No competing interests reported.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Investigation of the effects of climate change on hydrological drought and pattern detection for severe climatic conditions using CCT and SWAT models in the semi-arid region - Case study: Karkheh river basin

Status:

Version 1

Abstract

Figures

1 Introduction

2 Material And Methods

2.1 Study area

2.2 Data

2.3 SWAT model

2.4 SWAT model sensitivity analysis

2.5 Climate change and climate models

2.6 The Climate Change Toolkit

2.7 Validation of CCT projected data

2.8 The CCT uncertainty evaluation

2.9 SRI index

3 Results

3.1 Results of implementation and calibration of SWAT model

3.2 Results of validation of simulated data by CCT implementation

3.3 Results of the CCT uncertainty evaluation

3.4 Changes in hydrological drought compared to the base period

4 Discussion And Conclusion

Declarations

References

Additional Declarations

Status:

Version 1

\({\text{R}}^{2}=\frac{{(\sum ({\text{P}}_{\text{i}}-\stackrel{-}{\text{P}})({\text{O}}_{\text{i}}-\stackrel{-}{\text{O}}))}^{2}}{\sum {({\text{P}}_{\text{i}}-\stackrel{-}{\text{P}})}^{2}\sum {({\text{O}}_{\text{i}}-\stackrel{-}{\text{O}})}^{2}}\)	(5)
\(\text{M}\text{A}\text{E}=\frac{\sum _{\text{i}=1}^{\text{n}}\left\|{\text{P}}_{\text{i}}-{\text{O}}_{\text{i}}\right\|}{\text{n}}\)	(6)
\(\text{R}\text{M}\text{S}\text{E}=\sqrt{\frac{\sum _{\text{i}=1}^{\text{n}}{\left({\text{P}}_{\text{i}}-{\text{O}}_{\text{i}}\right)}^{2}}{\text{n}}}\)	(7)