This study assesses climate change's impact on drought in Iran's Dez Basin. It introduces the Hydro-Meteorological Drought Index (HMDI), integrating the Standardized Precipitation Evapotranspiration Index (SPEI) and Standardized Runoff Index (SRI). Using Climatic Research Unit Time Series (CRU TS) data (1980-2012) and downscaling forecasted data from three CMIP6 models (2020-2052) for SSP1-2.6 and SSP5-8.5 scenarios, we employ the rainfall-runoff Hydrologiska Byråns Vattenbalansavdelning Hydrological Bureau's Water Balance Model (HBV)-Light model to predict future streamflow. Drought characteristics are analyzed. Under SSP5-8.5, CanEsm5 shows substantial temperature and runoff increases, notably in Bakhtiari and Borujerd sub-basins (63% and 56%). Future droughts are expected to intensify, particularly under SSP5-8.5. The most severe HMDI-derived drought (HMDI 12) in Borujerd station is projected to increase from -43.44 to -44.05. SSP5-8.5 is likelier to cause severe and prolonged HMDI-derived droughts than SSP1-2.6 or the historical period. The analysis suggests that normal drought levels will persist, while mild and severe drought levels will rise in the future.

  • Using CMIP6 to assess the impact of climate change.

  • Using compound drought index to monitor hydrological and meteorological drought.

Drought disasters are frequent under the effects of both human activity and global climate change, and they significantly affect the nation's economic, social production, and the natural environment over the long term (Scanlon et al. 2017; Hameed et al. 2020). Therefore, drought monitoring and assessment globally or locally are crucial to reduce its impact (Wang, et al. 2019). To achieve this purpose, several indices were developed using a single variable (for example, precipitation; SPI1; McKee et al. 1993, runoff; SRI2; Shukla & Wood 2008), multiple variables (for example, precipitation and temperature; SPEI3; Vicente-Serrano et al. 2010), and composite indices, such as the multivariate standardized drought index (MSDI)4; (Varol et al. 2023) for soil moisture, temperature, and precipitation (Svoboda & Fuchs 2016). Composite indices have been created (Yang et al. 2018) using linear and nonlinear methods (Beersma & Buishand 2004; Li et al. 2015), including copula functions, scalogram models, principal component analysis (PCA) (Fahimirad & Shahkarami 2021), entropy theory (Waseem et al. 2015; Naderi & Moghaddasi 2022), and Markov chain approaches.

Global climate models (GCMs) are essential tools for climate projection and attribution, predicting future climate change and reproducing historical climatic conditions. The IPCC assessment reports have been historically based on the coupled model intercomparison projects (CMIPs). CMIP's Phase 3 (CMIP3) and Phase 5 (CMIP5) provided the present and future climatic data analyzed in IPCC's 4th and 5th assessment reports, respectively. The sixth phase of the CMIP's new multi-model datasets is now available, with CMIP6 models significantly improved, resulting in higher spatial and vertical resolutions, updated microphysics parameterizations, more sophisticated deep convective schemes, and altered ocean ice models (Eyring et al. 2016; Carvalho et al. 2022). Instead of using representative concentration pathway (RCP) scenarios, CMIP6 should use shared socioeconomic pathway (SSP) scenarios for future projections (O'Neill et al. 2016), offering a more detailed description of future socioeconomic development and more precise climate projections. CMIP6 models achieve better results when simulating historical climate (Lun et al. 2021).

Recent studies investigate how climate change affects rainfall and runoff in meteorological and hydrological drought, which greatly decreases the amount of water supplied in all forms, including streamflow. In the Cheongmicheon watershed in South Korea, Abdulai & Chung (2019) investigated the meteorological and hydrological droughts caused by climate change under the RCP 4.5 scenario. Their conclusions were based on the standardized precipitation evapotranspiration index (SPEI) and streamflow drought index (SDI), which measure the frequency of short-term severe or extreme droughts. In another research, Atallah et al. 2023 investigated that how the Wadi Louza in northwest Algeria responded to conditions of hydrological drought using an HBV-light hydrological model and using the SRI. According to the result, the two driest hydrological years were 1991–1993 and 2005–2006, and a 12-month period was ideal for developing efficient drought mitigation plans. Extreme droughts were predicted by the HBV-light model for the basin.

In order to reduce the impact of hydro-meteorological drought under climate change and enhance its impact mechanisms, it is important to developing awareness of hydro-meteorological drought variations. Therefore, in this study, the hydrological and meteorological drought indices (SRI and SPEI) for the Dez Dam Basin in Iran are combined to form a new composite index for this study dubbed the hydro-meteorological drought index (HMDI). The necessary data were retrieved from three CMIP6 models and the Climatic Research Unit Time Series (CRU TS) models using the principal component analysis (PCA) method under the SSP1-2.6 and SSP5-8.5 scenarios, respectively. In addition to the HBV-light model, streamflow simulations under real and imagined circumstances are run. In addition to researching the past and present effects of climate change, the run theory was used to calculate drought characteristics.

Study area and evaluation datasets

The Dez River Catchment is located in Khuzestan, southwest Iran. It is a sub-basin of the Karun Basin with semi-arid climate(Adib et al. 2021) between 48° and 9 min and 15 s to 50° and 18 min and 37 s east and wide 31° and 35 min and 51 s to 34° and 7 min and 46 s located North. It is a significant break in the Persian Gulf Basin and the central portion of the Zagros Mountains (Figure 1). The Karkheh basin (from the west), the Qurichay and Zayandehroud basins (from the north), and the Karun basin (from the east and south), are the basins that surround the Dez Dam basin region which has four sub-basins of Tireh, Marbereh, Sezar, and Bakhtiari. This basin's flow is anticipated to provide water to a number of industries, including aquaculture, manufacturing, fishing, and hydropower generation. Snowmelt from the end of winter to the start of spring supplies the majority of the basin's yearly water production via precipitation (Gholami et al. 2022). The primary challenges of this research have been the lack of long-term data with sufficient distribution since certain sites in the distribution region were inadequate and some stations were closed. Regarding this limitation, CRU data sets were used. Sezar and Bakhtiari, the basin's two principal rivers, join to form the Dez River, which drains the entire basin, close to Tange Panj. The basin has an annual average temperature of 24.2 °C. Figure 1 also depicts the locations of meteorological and hydrometric stations for the four sub-basins. Note that a representative meteorological station from each sub-basin was chosen to represent the results (Table 1). As seen, the Bakhtiari and Tireh sub-basins have the highest and lowest average annual rainfall and runoff, respectively.
Table 1

The description sub-basin Dez basin and the selected meteorological stations

Sub-basinStationArea (km2)Q (mm/month)P (mm/month)
Tireh Borujerd 3,477 13.56 39.61 
Marbereh Dorud 2,553 15.56 60.12 
Sezar Sepiddasht Sezar 3,281 24.67 58.12 
Bakhtiari Bakhtiari 5,973 46.98 94.80 
Sub-basinStationArea (km2)Q (mm/month)P (mm/month)
Tireh Borujerd 3,477 13.56 39.61 
Marbereh Dorud 2,553 15.56 60.12 
Sezar Sepiddasht Sezar 3,281 24.67 58.12 
Bakhtiari Bakhtiari 5,973 46.98 94.80 
Figure 1

The case study.

Datasets

  • - Station-based observations

The Iranian Ministry of Energy provided the monthly temperature, precipitation, and runoff records for the chosen stations in the Dez River basin. These documents served as proof of the expected data.

  • - Climate research unit (CRU)

  • - The CRU created a time series of monthly climate data with a spatial resolution of 0.5° from 1901 to 2016 (New et al. 1999; Mitchell & Jones 2005). Utilizing monthly ground-based climatic variables over land, data were griddled. These data were interpolated using inverse distance weighted interpolation (IDW). The observational point-based data from the four sub-basin stations in the Dez Basin were compared to this dataset. A representative station from each sub-basin was validated using the normalized root mean square error (NRMSE), mean bias error (MBE), and coefficient of determination (r) standards. As can be observed, statistical analysis demonstrated the dependability and correctness of CRU data (Table 2). For instance, the amounts of r, NRMSE, and MBE in the Bakhtiari station are 0.96, 0.48, and 0.05, respectively. As a result, this dataset was utilized to obtain monthly temperature and precipitation data (https://data.ceda.ac.uk).

  • - Future climate data

Table 2

Assessing the performance of the CRU dataset over the case study

VariableStationrNRMSEMBE
Precipitation Borujerd 0.66 0.92 0.07 
Dorud 0.81 0.71 0.09 
Sepiddasht Sezar 0.68 0.93 0.11 
Bakhtiari 0.85 0.67 1.68 
Temperature Borujerd 0.8 0.14 −0.04 
Dorud 0.93 0.12 1.23 
Sepiddasht Sezar 0.89 0.52 −2.35 
Bakhtiari 0.96 0.48 0.05 
VariableStationrNRMSEMBE
Precipitation Borujerd 0.66 0.92 0.07 
Dorud 0.81 0.71 0.09 
Sepiddasht Sezar 0.68 0.93 0.11 
Bakhtiari 0.85 0.67 1.68 
Temperature Borujerd 0.8 0.14 −0.04 
Dorud 0.93 0.12 1.23 
Sepiddasht Sezar 0.89 0.52 −2.35 
Bakhtiari 0.96 0.48 0.05 

From https://esgf-node.llnl.gov/search/cmip6/, the outputs of the CMIP6 climate model were downloaded. RCPs and Shared Socioeconomic Pathways (SSPs) scenarios are combined in the IPCC's sixth assessment report's scenarios to study climate change (Eyring et al. 2016). The historical simulation period spans the time of temperature observation, making it ideal for comparison with the records of actual temperature measurements. The two SSP scenarios for the future – SSP1-2.6 as a low forcing scenario (sustainable development) and SSP5-8.5 as a large forcing scenario – were examined using the three CMIP6 models that were chosen for analysis due to their accurate simulating regional climate patterns, capturing complex interactions, and reproducing past trends, particularly within this study region. Table 3 provides thorough details on these models. It should be highlighted that the historical simulation is valid for comparison with the existing data because it spans the temperature observation period from 1980 to 2012.

Table 3.

List of the CMIP6 models that were used in this study

CMIP6SourceResolution
CANESM5 Canadian Center for Climate Modeling and Analysis, Canada 2.81*2.81° 
BCC-CSM2-MR Beijing Climate Center, China 1.125°*1.125° 
IPSL-CM6A-LR Institute Pierre-Simon Laplace, France 1.26°*2.5° 
CMIP6SourceResolution
CANESM5 Canadian Center for Climate Modeling and Analysis, Canada 2.81*2.81° 
BCC-CSM2-MR Beijing Climate Center, China 1.125°*1.125° 
IPSL-CM6A-LR Institute Pierre-Simon Laplace, France 1.26°*2.5° 

Methodology

Methodology framework

An overview of the methodology of the current research is as follows (Figure 2).
Figure 2

The research flowchart.

Figure 2

The research flowchart.

Close modal

This flowchart shows how the necessary information was taken from the CRU dataset, weather station, and AR6 models. Climate projection creates climate scenarios for both the historical and the future. Furthermore, the HBV-light model is calibrated and validated as part of the hydrological model, and monthly runoff simulation calculations are made under historical and projected climate scenarios. After the calculation of SPEI and SRI, combined drought index HMDI were estimated. The severity and duration of the drought were determined using various indices and run theory.

HBV-light model

The rainfall–runoff model called the hydrologiska byråns vattenbalansavdelning (HBV) is a semi-distributed one. It mimics streamflow by accounting for variables such as temperature, precipitation, and potential evapotranspiration (PET) (Bergström 1976, 1995; Seibert 1997). The model is composed of runoff generation, soil moisture, and snow and glacier routines (Figure 3). Five elevation zones, spaced roughly 500 m apart, had their snowfall and melted calculated using the degree-day approach. This temperature-based strategy is helpful since the HBV-light model does not require significant datasets for the high upstream elevations, in contrast to physically based energy balance models (containing climate, snow depth, or snow water equivalent data to distribute the climate variables spatially). Real evapotranspiration and groundwater recharge serve as the mechanisms for storing water in the soil moisture cycle. As the last step, the lumped response function converts recharge to discharge. An in-depth examination of the model is given in Seibert & Vis (2012). Several software programs have included the HBV concept. HBV-light was used in this instance (Seibert & Vis 2012). With parameter ranges similar to those previously proposed, the model was calibrated using a genetic algorithm (Seibert 2000). The 14 free parameter values, for instance, were chosen by Seibert & Vis (2012) after 3,500 model iterations. Based on the non-parametric variation of Kling–Gupta efficiency, Non-Precipitation Events (NPE), 100 different calibration attempts were made for each catchment, resulting in ensembles with 100 parameter sets (Pool et al. 2018).
Figure 3

The structure of the HBV-light model (Seibert 2000).

Figure 3

The structure of the HBV-light model (Seibert 2000).

Close modal

Downscaling

The change factor (CF) technique, which also produced precipitation and temperature records for subsequent intervals, was used to downscale the results of the general circulation model (GCM). The CF method is typically used to transform historical temperature and precipitation series, as well as climate change scenarios, into monthly ratios. Equations (1) and (2) calculate the ‘difference’ and ‘ratio’ for temperature and precipitation based on the long-term monthly average of the future period and historical period that the same GCM has simulated in each cell of the computational grid, respectively (IPCC-TGCIA 2007).
(1)
(2)
where and indicate the observed precipitation and temperature time series in the historical (baseline) period, and are the average time series of future precipitation and temperature, and and are the average time series of historical (baseline) precipitation and temperature.

Standardized precipitation evapotranspiration index

Based on a monthly climate water balance, which is the total of PET and precipitation (P), this index is computed using the standard values of the probability distribution function of the deficit or surplus accumulation of a climate water balance at various time scales (McKee et al. 1993; McKee 1995). The monthly mean temperature is a function of the Thorn–Thwaite equation, which is used to calculate PET (Thornthwaite 1948). For the month I, the difference between P and PET is determined as follows:
(3)
where the following relationship, applied over various time periods, is used to calculate D values:
(4)
where k (months) is the time frame being considered, and n is the month that the calculation considers. SPEI is calculated using the three-parameter log-logistic distribution. Estimating the parameters of the log-logistic distribution can be done using the L-Moment method (Ahmad et al. 1988). The following equation is the probability distribution function for the D series according to the log-logistic distribution
(5)

where α, β, and γ are scale, shape, and location parameters, respectively, for D values in the range (γ < x < ∞).

Standardized runoff index

Shukla & Wood (2008) created the standardized runoff index (SRI) to recognize hydrological droughts. Similar to the standardized precipitation index (SPI), SRI is calculated (Gao & Zhang 2016). The normalizing function was determined to be the two-parameter logarithmic function (Bąk & Kubiak-Wójcicka 2016). Cumulative runoff values for each month and n months must be gathered in order to calculate SRI values on various time scales. A total of five different time series – 1, 3, 6, 9, and 12 months – were examined. SRI values are used in the calculation (6) to express the departure from the median in standard deviation units:
(6)

In this case, f(X) is the transformed runoff total, μ is the mean value of the normalized X, and δ is the X's standard deviation.

Principal component analysis

Principal component analysis (PCA), a linear statistical method, uses an orthogonal linear transformation to transform observation data into a new dataset with filtered elements while maintaining the maximum variability of the original dataset. Principle component analysis (PCA) was used to extract the pertinent drought data from prospective hydro-agro-meteorological variables X1, X2, … , Xn, and principal components (PC) were constructed for index computation (Keyantash & Dracup 2004). Months were examined separately to account for how seasonality may impact drought. The following major components are produced when the original data are merged with the eigenvectors E for month j (j = 1, … , 12):
(7)
where X is a m × n matrix of normalized observational data and Z is a m × n matrix of PCs (each element is processed by subtracting the mean value and then dividing by the column standard deviation). E, sometimes referred to as loading, is a n × n unit eigenvector matrix produced by PCA that can illustrate the connections between PCs and the original dataset. It is clear that PCs are essentially uncorrelated vectors with orthogonal treatments; hence it is inappropriate to represent them all numerically as a single entity.

Hydro-meteorological drought index

This index is proposed as an integrated index for evaluating and monitoring droughts since it can accurately characterize the performance of hydro-meteorological drought. Equations (5) and (6) are used to determine the monthly values of SPEI and SRI for each time series in order to achieve this objective. Then, in the form of a matrix with predetermined dimensions, the value of the combined drought index is calculated for each month. It should be noted that the classification of HDMI is similar to the selected individual indices.

Evaluation criteria

In this research, evaluation criteria including r, NRMSE, and mean MBE were used for data analysis and model evaluation. Their relationships are presented in the following equations:
(8)
(9)
(10)
where n is the calculated average, standard deviation, and different numbering of the data, and X stands for simulated data. The data that are simulated and observed are indicated by the subscripts P and o, respectively. The linear relationship between the simulated and observed data is represented by the number r, which ranges from 0 to 1. The linear relationship between the simulated and observed data is stronger when r is close to 1. The linear relationship between the simulated and observed data is stronger when r is close to 1. Metrics like NRMSE and MBE are used by researchers to measure model biases and accuracy, enhancing the validity of climate projections and intercomparison studies.

Future climate projections

The precipitation and temperature data of three CMIP6 models were downscaled using the CF approach under two scenarios in order to generate future climatic data. Then, the evaluation criteria were calculated for them (Equations (9)–(11)). For instance, in Bakhtiari station, the amount of r was 0.98 and 0.32 for temperature and precipitation, respectively. Additionally, the least MBE of precipitation and temperature were seen at Borujerd station, equal to 1.41 and 0.66. The lowest NRMSE amounts were observed in Borujerd and Sezar stations with less precipitation compared to two other stations (Figure 4), because the model simulations were more accurate with less bias in regions with less amount of precipitation. In order to provide future climate data, the CANESM5 model was selected since it had the lowest NRMSE and MBE and the highest r. As a result, climatic information was predicted for the chosen stations for the years 2020–2052 using the SSP1-2.6 and SSP5-8.5 scenarios. As seen in Figure 5, there are no discernible patterns in the monthly precipitation at the chosen stations when compared to the historical period. November (3.1) and August (0.1) saw the most historical increases and decreases in average monthly precipitation, respectively. It should be observed that the majority of the months show a growing tendency for SSP5-8.5 scenarios as opposed to SSP1-2.6 situations. At the chosen stations, however, there is a trend toward higher monthly temperatures when compared to the historical period. Under all scenarios, the selected stations recorded an increase in August of the greatest magnitude (1.9 and 2.8 °C, respectively). Additionally, under SSP1-2.6 and SSP5-8.5, respectively, the amount of average monthly temperature above historical levels is rising by around 1.2 and 1.9 °C.
Figure 4

Result of evaluation criteria for the three CMIP6 models at the selected stations.

Figure 4

Result of evaluation criteria for the three CMIP6 models at the selected stations.

Close modal
Figure 5

Changes in monthly mean temperature and precipitation compared to the historical period for SSP1-2.6 and SSP5-8.5 scenarios: (a) Bakhtiari, (b) Borujerd, (c) Dorud, and (d) Sepiddasht Sezar station.

Figure 5

Changes in monthly mean temperature and precipitation compared to the historical period for SSP1-2.6 and SSP5-8.5 scenarios: (a) Bakhtiari, (b) Borujerd, (c) Dorud, and (d) Sepiddasht Sezar station.

Close modal

Runoff model performance

Using monthly data on precipitation, temperature, and evapotranspiration, the HBV model simulates streamflow at four sub-basins from 1987 to 2010. Note that the model calibration and validation periods at these basins were different. Table 4 shows the model performance for each sub-basin as well as calibration and validation period details. Streamflow data gathered over a minimum of 13 years was used to calibrate the model. The model's performance was evaluated after calibration using at least 6 years of observed flow data. In each station, the correlation coefficient of the models was greater than 0.8, and the NRMSE was less than 1. At the Bakhtiari station, for example, the correlation coefficients for the calibration and validation periods were 0.88 and 0.85, respectively. According to the statistical criteria, the calibrated model looks to perform adequately. Figure 6 depicts the streamflow in each of the four sub-basins throughout the calibration and validation periods. As can be seen, the model has occasionally forecasted less runoff than was actually witnessed. This is due to the model's conceptually basic and comparably uncomplicated structure, which results in the runoff with a groundwater layer (Abebe et al. 2010; Al-Safi & Sarukkalige 2018; Moghaddasi et al. 2022). Then, the future runoff was simulated based on it under two scenarios. Figure 7 displays that the average monthly runoff will increase than historical runoff over the case study except for Sepiddasht Sezar station (Sezar basin) under two scenarios. It should be noted that the highest increase occurred in Dorud (about 2.7) and Borujerd station (about 2). In addition to, the largest and lowest variability of runoff was observed at Bakhtiari and Borujerd stations. The trend of runoff values was bigger than the median, as can be shown (positive skewness).
Table 4

Evaluation criteria for the observed and simulated streamflow by HBV-Light model over the case study

StationsPeriodrNRMSEMBE
Borujerd Calibration 1987–2000 0.87 0.094 −1.96 
Validation 2001–2007 0.88 0.093 −1.70 
Dorud Calibration 1989–2001 0.85 0.09 −2.6 
Validation 2002–2008 0.82 0.10 −0.63 
Sepiddasht Sezar Calibration 1987–2000 0.93 0.08 −6.7 
Validation 2001–2010 0.87 0.09 5.09 
Bakhtiari Calibration 1989–2001 0.88 0.09 −2.42 
Validation 2002–2008 0.85 0.11 −1.9 
StationsPeriodrNRMSEMBE
Borujerd Calibration 1987–2000 0.87 0.094 −1.96 
Validation 2001–2007 0.88 0.093 −1.70 
Dorud Calibration 1989–2001 0.85 0.09 −2.6 
Validation 2002–2008 0.82 0.10 −0.63 
Sepiddasht Sezar Calibration 1987–2000 0.93 0.08 −6.7 
Validation 2001–2010 0.87 0.09 5.09 
Bakhtiari Calibration 1989–2001 0.88 0.09 −2.42 
Validation 2002–2008 0.85 0.11 −1.9 
Figure 6

The observed and simulated runoff during calibration and validation periods.

Figure 6

The observed and simulated runoff during calibration and validation periods.

Close modal
Figure 7

Changes of monthly mean runoff compared to the historical period for SSP1-2.6 and SSP5-8.5 scenarios: (a) (left) Borujerd, (right) Dorud, (b) (left) Sepiddasht Sezar, (right) Bakhtiari stations.

Figure 7

Changes of monthly mean runoff compared to the historical period for SSP1-2.6 and SSP5-8.5 scenarios: (a) (left) Borujerd, (right) Dorud, (b) (left) Sepiddasht Sezar, (right) Bakhtiari stations.

Close modal

HMDI calculation

Firstly, the existing drought indices including SPEI and SRI were computed at the 3- and 12-month time scale for historical and future periods. In addition, the HMDI was developed based on the mentioned individual indices. As an example of the future, 12-month time scale is mentioned in Figure 8. According to this figure, the HMDI indicated similar variation patterns with existing indices during the historical period for four sub-basins, such as the 1990–1994 and 2008–2012 dry periods. Then, the most important characteristics of drought including severity and duration were extracted based on Run theory (Yevjevich 1967) using HMDI. Note that a threshold of −0.5 was chosen, with smaller values indicating a drought event. As it can be seen, Table 5 shows the number, severity, and duration (month) of drought events for the historical (1980–2012) and future periods during the period) 2020–2052) under SSP1-2.6 and SSP5-8.5 scenarios for four sub-basins. In general, the frequency of dry periods decreases as the time scale lengthens, which causes the severity and duration of drought to increase. The most severe drought in recorded history for all sub-basins has severity, duration, and number ranging from −15.56 to −83.42, 15 to 61 months, and 6 to 24. For example, in Bakhtiari station, the number of droughts decreased from HMDI 3 to HMDI 12 from 18 to 8 in the historical period and from 20 to 9 in SSP5-8.5. Meanwhile, duration and severity increase from 32 to 57 and −44.76 to −83.42, respectively. The results revealed that the probability of severity and duration of the HMDI-derived drought generated by the SSP5-8.5 scenario is increased in comparison with SSP1-2.6 and during the historical period in Dez basin, so the watershed would face a modest increase in runoff in the future. According to the study's findings, the effects of climate change can differ depending on where it happens.
Table 5

The number, highest severity, and duration (month) of drought events under the historical and future periods

Historical
SSP1-2.6
SSP5-8.5
StationIndexNumberDurationSeverityNumberDurationSeverityNumberDurationSeverity
MaxMaxMaxMaxMaxMax
Bakhtiari HMDI 3 18 32 −1.44 24 −1.43 23 −1.45 
HMDI 12 57 −1.56 13 16 −1.46 12 17 −1.42 
Borujerd HMDI 3 24 15 −1.47 36 −1.37 30 −1.35 
HMDI 12 10 60 −1.33 21 −1.31 10 18 −1.40 
Dorud HMDI 3 17 29 −1.35 23 −1.41 24 −1.42 
HMDI 12 47 −1.43 11 19 −1.36 11 17 −1.39 
Sepiddasht Sezar HMDI 3 15 20 −1.43 22 −1.43 21 −1.42 
HMDI 12 61 −1.37 11 17 −1.44 11 19 −1.44 
Historical
SSP1-2.6
SSP5-8.5
StationIndexNumberDurationSeverityNumberDurationSeverityNumberDurationSeverity
MaxMaxMaxMaxMaxMax
Bakhtiari HMDI 3 18 32 −1.44 24 −1.43 23 −1.45 
HMDI 12 57 −1.56 13 16 −1.46 12 17 −1.42 
Borujerd HMDI 3 24 15 −1.47 36 −1.37 30 −1.35 
HMDI 12 10 60 −1.33 21 −1.31 10 18 −1.40 
Dorud HMDI 3 17 29 −1.35 23 −1.41 24 −1.42 
HMDI 12 47 −1.43 11 19 −1.36 11 17 −1.39 
Sepiddasht Sezar HMDI 3 15 20 −1.43 22 −1.43 21 −1.42 
HMDI 12 61 −1.37 11 17 −1.44 11 19 −1.44 
Figure 8

The monthly SPEI-12, SRI-12, HMDI-12 in historical and future periods: (a) Borujerd, (b) Dorud, (c) Sepiddasht Sezar, and (d) Bakhtiari stations.

Figure 8

The monthly SPEI-12, SRI-12, HMDI-12 in historical and future periods: (a) Borujerd, (b) Dorud, (c) Sepiddasht Sezar, and (d) Bakhtiari stations.

Close modal
As illustrated in Figure 9, the most and least incidence percentage of severe drought class (extreme) occurred at the Dorud (6.75) and Borujerd (3.09) stations during the historical period. The percentage total of mild and severe drought classes was increased under two scenarios than the historical period at the selected stations except for Sepidasht Sezar station. For instance, the incidence percentage of the mild and severe drought classes ranged from 11.95 to 16.62 and 3.09 to 5.19, respectively, at the Borujerd station. The extreme drought class rarely occurred in the case study, especially at Borujerd station. It can also be said that the region will probably not experience the extreme drought class in the future period.
Figure 9

The percentage change of drought classes under the historical and future periods at (a) Borujerd, (b) Bakhtiari, (c) Sepiddasht Sezar, and (d) Dorud and 12-month time scale.

Figure 9

The percentage change of drought classes under the historical and future periods at (a) Borujerd, (b) Bakhtiari, (c) Sepiddasht Sezar, and (d) Dorud and 12-month time scale.

Close modal

This study investigated the impact of climate change on drought conditions in Iran's Dez Basin, which is significantly important for hydropower production and water supply for drinking, industry, and agriculture. The positive aspect of the current research is the division of the basin area into four sections, which enables a separate examination of each region's climatic change. In this study, monthly data sets including precipitation and temperature for a period of 22 years (1980–2012) were considered. Moreover, three GCM models, CANESM5, BCC-CSM2-MR, and IPSL-CM6A-LR models were chosen for analysis due to their accurate simulating regional climate patterns. The hydrological response of the Dez basin to climate change was simulated using the conceptual HBV rainfall–runoff model. Then two meteorological (SPEI) and hydrological (SRI) indices were used to calculate streamflow historical and future drought. Finally, the newly combined drought index HMDI was built using the prior individual indexes on 3- and 12-month time scales for characterizing past and future drought periods.

The result showed that CANESM5 was chosen as the best model because it had the highest r and the lowest NRMSE and MBE. In order to extract data for future climate under the SSP1-2.6 and SSP5-8.5 scenarios, revealing a significant increasing trend in temperature and precipitation. The HBV-light model performed well during the calibration phase, demonstrating its suitability for predicting the hydrological state of the watershed in the future. The average yearly streamflow increase was measured under two scenarios, and the SPEI and SRI drought indices were derived. The HMDI index on 3- and 12-month time scales for past and future periods showed the most severe drought produced by the SSP5-8.5 scenario in Borujerd, 12-month scale, and the least one with SSP5-8.5 scenario in Borujerd station from 11.71 to −12.75. Therefore, the drought severity in SSP5-8.5 was predicted more than in SSP1-2.6. The findings suggest the need to examine how climate change affects local scales and predict a significant increase in runoff in the future. Additionally, the responses of the four sub-basins were found to vary in many situations. Finally, the drought classes were extracted, and under the historical period, the Dorud (6.75) and Borujerd (3.09) stations had the highest and lowest incidence percentages of the severe drought classes in the future. Under the two scenarios, the incidence percentage total of the mild and severe drought classes increased in the selected stations, with the exception of the Sepidasht Sezar station. These findings point to an impending drought in the Dez basin. The detrimental effects of climate change, according to this study, will cause substantial droughts to occur in the investigated region in the future. Consequently, it is essential to develop a management strategy to eradicate its inescapable adverse impacts. It is recommended that future research will integrate nonlinear approaches and utilize more variables, such as evapotranspiration and runoff in order to create new combination indices and compare their results.

All relevant data are available from an online repository or repositories. CRU Data: https://data.ceda.ac.uk/; CMIP 6 data: https://esgf-node.llnl.gov/search/cmip6/.

The authors declare there is no conflict.

1

Standardized Precipitation Index

2

Standardized Runoff Index

3

Standardized Precipitation Evapotranspiration Index

4

Multivariate Standardized Drought Index

Abebe
N. A.
,
Ogden
F. L.
&
Pradhan
N. R.
2010
Sensitivity and uncertainty analysis of the conceptual HBV rainfall–runoff model: Implications for parameter estimation
.
Journal of Hydrology
389
(
3–4
),
301
310
.
https://doi.org/10.1016/j.jhydrol.2010.06.007
.
Adib
A.
,
Kisi
O.
,
Khoramgah
S.
,
Gafouri
H. R.
,
Liaghat
A.
,
Lotfirad
M.
&
Moayyeri
N.
2021
A new approach for suspended sediment load calculation based on generated flow discharge considering climate change
.
Water Supply
21
(
5
),
2400
2413
.
https://doi.org/10.2166/ws.2021.069
.
Al-Safi
H. I. J.
&
Sarukkalige
P. R.
2018
Evaluation of the impacts of future hydrological changes on the sustainable water resources management of the Richmond River catchment
.
Journal of Water and Climate Change
9
(
1
),
137
155
.
https://doi.org/10.2166/wcc.2017.144
.
Atallah
M. H.
,
Djellouli
F.
,
Bouanani
A.
,
Baba-Hamed
K.
,
Faisal
A. A.
&
Hasan
K.
2023
Assessment of catchment behavior of the Wadi Louza in NW-Algeria under hydrological drought conditions
.
Earth Systems and Environment
7
(
1
),
297
306
.
Ahmad
M. I.
,
Sinclair
C. D.
&
Werritty
A.
1988
Log-logistic flood frequency analysis
.
Journal of Hydrology
98
(
3-4
),
205
224
.
Bąk
B.
&
Kubiak-Wójcicka
K.
2016
Assessment of meteorological and hydrological drought in Torun (central Poland town) in 1971–2010 based on standardized indicators
. In:
3rd International Conference Water Resources and Wetlands Conference Proceedings
, pp.
164
170
.
Bergström
S.
1976
Development and application of a conceptual runoff model for Scandinavian catchments, Bulletin Series A, No. 52 – SMHI report RH07, Norrköping: Swedish Meteorological and Hydrological Institute, 134
.
Bergström
S.
,
1995
The HBV model
. In:
Computer Models of Watershed Hydrology
(
Singh
V. P.
, ed.).
Water Resources Publication
,
Highlands Ranch
, pp
443
476
.
Beersma
J. J.
&
Buishand
T. A.
2004
Joint probability of precipitation and discharge deficits in the Netherlands
.
Water resources research
40
(
12
).
Carvalho
D.
,
Rafael
S.
,
Monteiro
A.
,
Rodrigues
V.
,
Lopes
M.
&
Rocha
A.
2022
How well have CMIP3, CMIP5 and CMIP6 future climate projections portrayed the recently observed warming
.
Scientific Reports
12
(
1
),
11983
.
Eyring
V.
,
Bony
S.
,
Meehl
G. A.
,
Senior
C. A.
,
Stevens
B.
,
Stouffer
R. J.
&
Taylor
K. E.
2016
Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization
.
Geoscientific Model Development
9
(
5
),
1937
1958
.
https://doi.org/10.5194/gmd-9-1937-2016
.
Fahimirad
Z.
&
Shahkarami
N.
2021
The impact of climate change on hydro-meteorological droughts using copula functions
.
Water Resources Management
35
(
12
),
3969
3993
.
https://doi.org/10.1007/s11269-021-02918-z
.
Gao
L.
&
Zhang
Y.
2016
Spatio-temporal variation of hydrological drought under climate change during the period 1960–2013 in the Hexi Corridor, China
.
Journal of Arid Land
8
(
2
),
157
171
.
https://doi.org/10.1007/s40333-015-0022-3
.
Gholami
H.
,
Moradi
Y.
,
Lotfirad
M.
,
Gandomi
M. A.
,
Bazgir
N.
&
Shokrian Hajibehzad
M.
2022
Detection of abrupt shift and non-parametric analyses of trends in runoff time series in the Dez river basin
.
Water Supply
22
(
2
),
1216
1230
.
https://doi.org/10.2166/ws.2021.357
.
Hameed
M.
,
Ahmadalipour
A.
&
Moradkhani
H.
2020
Drought and food security in the Middle East: An analytical framework
.
Agricultural and Forest Meteorology
281
,
107816
.
https://doi.org/10.1016/j.agrformet.2019.107816
.
Intergovernmental Panel on Climate Change, (IPCC): Climate Change 2007
2007
Synthesis Report of the Fourth Assessment Report
,
IPCC
.
Keyantash
J. A.
&
Dracup
J. A.
2004
An aggregate drought index: Assessing drought severity based on fluctuations in the hydrologic cycle and surface water storage
.
Water Resources Research
40
(
9
).
https://doi. org/10.1029/2003WR002610
.
Li
Q.
,
Li
P.
,
Li
H.
&
Yu
M.
2015
Drought assessment using a multivariate drought index in the Luanhe River basin of Northern China
.
Stochastic Environmental Research and Risk Assessment
29
(
6
),
1509
1520
.
https://doi.org/10.1007/s00477-014-0982-4
.
Lun
Y.
,
Liu
L.
,
Cheng
L.
,
Li
X.
,
Li
H.
&
Xu
Z.
2021
Assessment of GCMs simulation performance for precipitation and temperature from CMIP5 to CMIP6 over the Tibetan Plateau
.
International Journal of Climatology
41
(
7
),
3994
4018
.
https://doi.org/10.1002/joc.7055
.
McKee
T. B.
1995
Drought monitoring with multiple time scales. In: Proceedings of 9th Conference on Applied Climatology, Boston, 1995.McKee TB, Doesken NJ, Kleist J. 1995 Drought monitoring with multiple time scales. In: 482 9ths Conference on Applied Climatology, Am Meteor Soc 233–236
.
McKee
T. B.
,
Doesken
N. J.
&
Kleist
J.
1993
The relationship of drought frequency and duration to time scales
. In:
Proceedings of the 8th Conference on Applied Climatology
, Vol.
17
, No.
22
, pp.
179
183
.
Mitchell
T. D.
&
Jones
P. D.
2005
An improved method of constructing a database of monthly climate observations and associated high-resolution grids
.
International Journal of Climatology: A Journal of the Royal Meteorological Society
25
(
6
),
693
712
.
https://doi.org/10.1002/joc.1181
.
Moghaddasi
M.
,
Anvari
S.
&
Akhondi
N.
2022
A trade-off analysis of adaptive and non-adaptive future optimized rule curves based on simulation algorithm and hedging rules
.
Theoretical and Applied Climatology
148
(
1
),
65
78
.
https://doi.org/10.1007/s00704-022-03930-y
.
Naderi
K.
&
Moghaddasi
M.
2022
Drought occurrence probability analysis using multivariate standardized drought index and copula function under climate change
.
Water Resources Management,
1
24
.
https://doi.org/10.1007/s11269-022-03186-1
.
New
M.
,
Hulme
M.
&
Jones
P.
1999
Representing twentieth-century space–time climate variability. Part I: Development of a 1961–90 mean monthly terrestrial climatology
.
Journal of Climate
12
(
3
),
829
856
.
https://doi.org/10.1175/1520-0442(1999)012%3C0829:RTCSTC%3E2.0.CO;2
.
O'Neill
B. C.
,
Tebaldi
C.
,
van Vuuren
D. P.
,
Eyring
V.
,
Friedlingstein
P.
,
Hurtt
G.
,
Knutti
R.
,
Kriegler
E.
,
Lamarque
J.-F.
,
Lowe
J.
,
Meehl
G. A.
,
Moss
R.
,
Riahi
K.
&
Sanderson
B. M.
2016
The scenario model intercomparison project (ScenarioMIP) for CMIP6
.
Geoscientific Model Development
9
,
3461
3482
.
https://doi.org/10.5194/gmd-9-3461-2016
.
Pool
S.
,
Vis
M.
&
Seibert
J.
2018
Evaluating model performance: Towards a non-parametric variant of the Kling-Gupta efficiency
.
Hydrological Sciences Journal
63
(
13–14
),
1941
1953
.
https://doi.org/10.1080/02626667.2018.1552002
.
Scanlon
B. R.
,
Ruddell
B. L.
,
Reed
P. M.
,
Hook
R. I.
,
Zheng
C.
,
Tidwell
V. C.
&
Siebert
S.
2017
The food-energy-water nexus: Transforming science for society
.
Water Resources Research
53
(
5
),
3550
3556
.
https://doi.org/10.1002/2017WR020889
.
Seibert
J.
2000
Multi-criteria calibration of a conceptual runoff model using a genetic algorithm
.
Hydrology and Earth System Sciences
4
(
2
),
215
224
.
https://doi.org/10.5194/hess-4-215-2000
.
Seibert
J.
&
Vis
M. J.
2012
Teaching hydrological modeling with a user-friendly catchment-runoff-model software package
.
Hydrology and Earth System Sciences
16
(
9
),
3315
3325
.
https://doi.org/10.5194/hess-16-3315-2012
.
Shukla
S.
&
Wood
A. W.
2008
Use of a standardized runoff index for characterizing hydrologic drought
.
Geophysical research letters
35
(
2
).
Svoboda
M.
&
Fuchs
B.
2016
Handbook of drought indicators and indices
.
Drought and water crises: Integrating science, management, and policy
,
155
208
.
Thornthwaite
C. W.
1948
An approach toward a rational classification of climate
.
Geographical Review
38
(
1
),
55
94
.
https://doi.org/10.2307/210739
.
Varol
T.
,
Atesoglu
A.
,
Ozel
H. B.
&
Cetin
M.
2023
Copula-based multivariate standardized drought index (MSDI) and length, severity, and frequency of hydrological drought in the Upper Sakarya Basin, Turkey
.
Natural Hazards
116
(
3
),
3669
3683
.
https://doi.org/10.1007/s11069-023-05830-4
.
Vicente-Serrano
S. M.
,
Beguería
S.
&
López-Moreno
J. I.
2010
A multiscalar drought index sensitive to global warming: The standardized precipitation evapotranspiration index
.
Journal of Climate
23
(
7
),
1696
1718
.
https://doi.org/10.1175/2009JCLI2909.1
.
Waseem
M.
,
Ajmal
M.
&
Kim
T. W.
2015
Development of a new composite drought index for multivariate drought assessment
.
Journal of Hydrology
527
,
30
37
.
https://doi.org/10.1016/j.jhydrol.2015.04.044
.
Wang
Q.
,
Liu
Y. Y.
,
Zhang
Y. Z.
,
Tong
L. J.
,
Li
X.
,
Li
J. L.
&
Sun
Z.
2019
Assessment of spatial agglomeration of agricultural drought disaster in China from 1978 to 2016
.
Scientific reports
9
(
1
),
14393
.
Yang
J.
,
Chang
J.
,
Wang
Y.
,
Li
Y.
,
Hu
H.
,
Chen
Y.
, Huang, Q. &
Yao
J.
2018
Comprehensive drought characteristics analysis based on a nonlinear multivariate drought index
.
Journal of Hydrology
557
,
651
667
.
Yevjevich
V. M.
1967
Objective Approach to Definitions and Investigations of Continental Hydrologic Droughts
.
Doctoral Dissertation
,
Colorado State University. Libraries
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).