This study critically assesses the combined effects of climate and land-use change on flood recurrence in the Kokcha River, Afghanistan, spanning the period from 2010 to 2021 and projecting forward to 2088–2099. Through the application of a bias-corrected model, we achieved high accuracy in temperature and precipitation simulations, with notable NSE values of 0.9 and 0.69, and R2 values of 0.92 and 0.78, respectively. Future streamflow simulations under different scenarios highlight climate change as the major driver influencing flood recurrence in the Kokcha River, contributing to 101.1% of the total variation, while land-use change has a minimal contribution of −1.1%. Our analysis of precipitation, average temperature, and streamflow data reveals significant patterns and changes, with future projections indicating a gradual decline in precipitation levels, mean temperature, and streamflow. Flood frequency analysis for return periods of 10, 50, 100, 200, and 500 years, considering different scenarios, underscores the likelihood of floods of varying magnitudes. Notably, the highest streamflow spikes under both scenarios highlight the impact of futuristic air temperature and precipitation on flood recurrence. The study advocates prioritizing climate change adaptation and resilient land-use strategies to ensure sustainable water resource management, emphasizing the mitigation of potential flood risks.

  • Utilized SWAT model to analyze climate and land-use impacts on streamflow.

  • Utilized multiple scenarios to isolate climate and land-use change impacts on streamflow.

  • Employed the CA-ANN model for forecasting future land use patterns

  • Identified climate change as exerting a more significant impact on streamflow in comparison to changes in land use.

The Kokcha River, originating from the Hindu Kush Mountains and flowing through northeastern Afghanistan, serves as a vital water source for millions, sustaining agriculture, fishing, and other economic activities (Shroder et al. 2022). However, it faces severe environmental challenges exacerbated by climate change and unsustainable land practices. This study aims to comprehensively assess the impacts of these factors on the Kokcha River's hydrology and ecology, presenting innovative insights into adaptation strategies. The region has experienced rising temperatures and changes in precipitation patterns, which have led to decreased water flow and increased erosion. Extreme weather events such as floods and droughts have become more frequent and intense, which has negatively impacted agricultural productivity and other livelihoods that depend on the river. In addition, the changes in temperature and precipitation patterns have led to changes in the hydrological cycle, which has affected the availability and quality of water in the river. The changes in the hydrological cycle have also led to the increased occurrence of landslides, which have further impacted the ecological health of the river (Rycroft & Wegerich 2009). Unsustainable land-use practices have also contributed to the degradation of the Kokcha River. Deforestation, overgrazing, and other land-use practices have led to soil erosion and sedimentation, further reducing water quality and availability (Roy & Rahman 2023). These practices have contributed to increased soil degradation, desertification, and loss of biodiversity, which has further impacted the ecological health of the river (Shirzad & Solanki 2017).

Effective disaster management and policy development necessitate a thorough understanding of risk, particularly concerning natural hazards like floods, which are increasingly prevalent worldwide due to various factors such as population growth, urbanization, and climate change-induced shifts in weather patterns (Meyer et al. 2009). Afghanistan, positioned in a region highly prone to natural disasters, faces significant vulnerability to severe floods due to its geography and historical environmental degradation, ranking as the third most at-risk country according to the INFORM 2022 Index (Ikram et al. 2023). Floods in Afghanistan disproportionately affect impoverished and vulnerable communities, exacerbating social vulnerability and coping capacity challenges. Despite the arid climate, flood risk persists, necessitating comprehensive flood risk assessments to inform effective disaster management and resilience-building strategies (Ahmadi et al. 2023). The absence of adequate flood risk assessment methodologies at both local and national levels underscores the urgency for informed and proactive flood risk management approaches. Such assessments, incorporating satellite-based technology and spatial data analysis, offer promising avenues for enhancing rational flood management and targeted disaster preparedness. In addressing flood risks, communities have options including withdrawal, resistance, and adaptation, with the effectiveness of adaptation strategies contingent upon robust flood risk assessments (Nam et al. 2015).

The impacts of climate change on river systems have been extensively studied globally, highlighting the profound implications for water availability, quality, and ecosystem health. Vörösmarty et al. (2010) emphasize that altered precipitation patterns and temperature rise directly impact river hydrology, leading to changes in flow regimes, habitat availability, and aquatic biodiversity. Water resources are essential for human survival and economic growth, but they are threatened by climate change. The availability of water for human uses such as drinking, agriculture, and power generation is affected by fluctuations in river flows. Similarly, Maurer et al. (2002) demonstrate how reduced snowpack and altered streamflow patterns, attributed to climate change, affect water resources and agricultural productivity in regions like the western United States. Unsustainable land-use practices exacerbate the impacts of climate change on rivers, leading to increased runoff, sedimentation, and habitat degradation (Rahman & Rahman 2023). McGrane (2016) points out that urbanization intensifies surface runoff and flood risks due to elevated impervious surfaces, while agricultural activities contribute to nutrient and sediment runoff, leading to eutrophication and water quality deterioration (Astuti 2017; Kumar & Yadav 2020).

Furthermore, alterations in land use have a significant influence on the flow of rivers, which in turn, have social and economic ramifications for societies that depend on rivers for their cultural and economic sustenance. Changes in land utilization practices have the potential to generate disputes among various stakeholders, including farmers, urban residents, and conservationists, regarding their entitlement to water resources (Steinhäußer et al. 2015). Urbanization can result in escalated water demand for both domestic and industrial purposes, which can subsequently cause a decrease in water accessibility for other users (Majumder 2015; Sakizadeh & Milewski 2023). In a nutshell, changes in land use are a crucial factor that propels the flow of rivers, thereby impacting the freshwater ecosystems, water resources, and the society (Satriagasa et al. 2023). Land-use change patterns, encompassing urbanization, agriculture, and deforestation, have the potential to cause changes in river hydrology, thereby inducing changes in both the quality and quantity of water (Teixeira et al. 2014).

Soil and Water Assessment Tool (SWAT) is a hydrological model that is extensively employed to simulate the effects of river runoff on water resources. SWAT model is a hydrological simulation model that operates on a process-based and spatially distributed approach, enabling it to replicate watershed-scale hydrological processes (Abbaspour et al. 2015). The hydrological simulation of a watershed is achieved through the integration of diverse factors such as climate, land use, soil, and management practices within the model. The computational model employs algorithms to replicate the dynamics of water transport, encompassing phenomena such as surface runoff, groundwater flow, and streamflow. The model has been utilized in various research endeavors to evaluate the effects of river discharge on water resources, encompassing the accessibility, quality, and quantity of surface and groundwater (Abbaspour et al. 2015). The utilization of the SWAT model in evaluating the hydrologic consequences of river discharge holds significant ramifications for the management of water resources (Joumar et al. 2023). The model serves as a means for engaging stakeholders and fostering discourse regarding water resource management and the potential effects of different management strategies (Vo et al. 2019). The SWAT model is an advantageous tool for evaluating the hydrologic consequences of river discharge on water reserves. The aforementioned model facilitates the assessment of diverse management scenarios to gauge their effects on water resources, encompassing the availability, quality, and quantity of both surface and groundwater (Neitsch et al. 2011).

Water resources, agriculture, and infrastructure in Afghanistan are all predicted to be significantly impacted by climate change. The most recent generation of global climate models used to simulate the Earth's climate system is known as Coupled Model Intercomparison Project Phase 6 (CMIP6). These models are crucial for forecasting future climate change and comprehending the impacts of greenhouse gas emissions. SSP2-4.5 and SSP5-8.5 are the two shared socioeconomic pathways (SSPs) that are taken into consideration in this study, and they represent two distinct scenarios for future greenhouse gas emissions and their impacts on the climate. The SSP2-4.5 scenario anticipates that moderate mitigation measures will be taken to reduce greenhouse gas emissions and keep the rise in the global mean temperature to 2 °C over pre-industrial levels. The SSP5-8.5 scenario, on the other hand, anticipates significant greenhouse gas emissions without any mitigation measures, resulting in a temperature rise of at least 4 °C over pre-industrial levels by the end of the century. Scenarios (O'Neill et al. 2016) examined the effects of the SSP2-4.5 and SSP5-8.5 scenarios on international food security. They discovered that there would be a little decrease in the availability of food on a worldwide scale under the SSP2-4.5 scenario, but that this could be avoided by taking the proper adaption strategies. However, under the SSP5-8.5 scenario, there might be a 30% decrease in the amount of food available globally, which would cause food insecurity in many parts of the world. CMIP6 models offer useful information on the probable effects of future greenhouse gas emissions on the climate of Earth. The models demonstrate that considerable increases in temperature, precipitation, and sea level rise are anticipated even with modest mitigation measures (Neitsch et al. 2011). To lessen uncertainty in future estimates, further study is required because the size of these changes differs among the models. Policymakers and researchers may evaluate the possible effects of various greenhouse gas emission routes on several facets of the Earth system, including food security, biodiversity, and human well-being, using the frameworks provided by the SSP2-4.5 and SSP5-8.5 scenarios (O'Neill et al. 2016).

This study aims to comprehensively assess the combined effects of climate and land-use change on flood recurrence in the Kokcha River basin, Afghanistan, from 2010 to 2021, with projections extending to 2088–2099. The specific objectives are (1) to evaluate the accuracy of temperature and precipitation simulations using bias-corrected modeling techniques, (2) to analyze future streamflow simulations under SSP2-4.5 and SSP5-8.5 climate scenarios and assess the relative contributions of climate and land-use change to flood recurrence, (3) to examine the projected changes in precipitation, average temperature, and streamflow and their implications for water resource management, (4) to conduct flood frequency analysis for various return periods and design flood estimation, considering different climate scenarios, and (5) to assess the impact of anticipated land-use changes on streamflow using advanced modeling approaches. By achieving these objectives, this study aims to provide valuable insights into the dynamics of the Kokcha River basin, informing policy decisions and adaptation strategies to mitigate the adverse effects of climate change and unsustainable land-use practices on water resources and ecosystem health. Through interdisciplinary research and innovative modeling techniques, we seek to contribute to the scientific understanding of hydrological variability and enhance the resilience of riverine communities in Afghanistan.

Study area description

Kokcha River flows across northern Afghanistan. It runs across the Hindu Kush province of Badakhshan. It takes its name from the Kokcha Valley. Feyzabad is located along the Kokcha. The Kokcha River is a major tributary of the Amu Darya, entering it 320 km downstream near Khawaja Ghar in Takhar province, at the foot of the Greek city of Ai Khanum. The eastern Hindu Kush Mountains have glaciers that keep the river flowing in the summer. It is located between 37° 09′ 48′′ N latitude and 69° 23′ 48′′ E longitude with a basin area of 22,367.3 km2 and elevations as high as 446 m (1,463 feet). Figure 1 depicts the delineated watershed for the study area and DEM map.
Figure 1

Study area delineated watershed and DEM data of the region.

Figure 1

Study area delineated watershed and DEM data of the region.

Close modal

Data collection

National Aeronautics and Space Administration (NASA) offered DEM data for the study region in order to carry out the research. The Global Digital Elevation Model's Advanced Space borne Thermal Emission and Reflection Radiometer (ASTER) has a resolution of 30 m × 30 m (Global Digital Elevation Model). Watershed boundaries are established using DEM data. The delimited watershed is shown in Figure 1. The United States Geological Survey provided the Moderate Resolution Imaging Spectroradiometer (MODIS) land-use data that were utilized in this research. The characteristics of the different land-use types within the designated watershed are shown in Figure 2. The Food and Agriculture Organization of the United Nations provided the soil information (FAO). The characteristics of the different soil classes for the delineated watershed are shown in Figure 3.
Figure 2

USGS MODIS land-use classes of Kokcha River Basin.

Figure 2

USGS MODIS land-use classes of Kokcha River Basin.

Close modal
Figure 3

FAO soil classes of Kokcha River Basin.

Figure 3

FAO soil classes of Kokcha River Basin.

Close modal

Hydroclimatic data

The Ministry of Energy and Water (MEW) of Afghanistan has provided meteorological and hydrological data, including rainfall, maximum and minimum temperatures range from 2008 to 2021 and hydrological data spans from 2010 to 2021. The description of four GCMs is listed in Table 1.

Table 1

Summary of the general circulation models (GCMs) utilized in the current study

Modeling centerModelResolution (km)
National Earth System Model Version 3 NESM3 192 × 96 
National Center for Meteorological Research (France) CNRM-CM6-1 250 × 250 
Geophysical Fluid Dynamics Laboratory, USA GFDL-ESM4 90 × 144 
Atmosphere and Ocean Research Institute, The University of Tokyo, Japan MIROC6 128 × 256 
Modeling centerModelResolution (km)
National Earth System Model Version 3 NESM3 192 × 96 
National Center for Meteorological Research (France) CNRM-CM6-1 250 × 250 
Geophysical Fluid Dynamics Laboratory, USA GFDL-ESM4 90 × 144 
Atmosphere and Ocean Research Institute, The University of Tokyo, Japan MIROC6 128 × 256 

METHODOLOGY

Gumbel distribution method

The Gumbel distribution finds applications in various fields, particularly in modeling extreme events, such as extreme temperatures, floods, or financial risk assessment. It is widely used in fields like hydrology, climatology, and finance to analyze and predict the occurrence of rare or extreme events. Additionally, the Gumbel distribution plays a vital role in statistics for extreme value theory, where it is used to estimate the distribution of the maximum or minimum value from a set of independent and identically distributed random variables.

This extreme value distribution was introduced by Gumbel and is known as Gumbel's Method. It is widely used for probability distribution functions for extreme values in hydrological studies or prediction of flood peaks, maximum rainfall, etc. The method and procedure for calculating the design flood in any return period. The hypothetical plotting by Gumbel (1945) is in light of the presumption that the watched esteem is the most likely, or modular, estimation of this rank of the flood. Its arrival period is along these lines skewed toward the method of the hypothetical circulation. The Gumbel hypothesis does not have any significant bearing entirely to flood for the accompanying reasons. The Gumbel method of frequency analysis is based on extreme value distribution and uses frequency factors developed for theoretical distribution.

Gumbel's distribution is a statistical method often used for predicting extreme hydrological events such as floods. In this study, it has been applied for flood frequency analysis because (a) the river is less regulated, hence is not significantly affected by reservoir operations, diversions or urbanization; (b) flow data are homogeneous and independent hence lack long-term trends; and (c) peak flow data cover a relatively long record (more than 10 years) and is of good quality; (d) there is no major tributary of the river whose inflow can affect the flood peak. The equation for Gumbel's distribution as well as the procedure with a return period T is given as
(1)
where σx is the standard deviation of the sample size; K is the frequency factor, which is expressed as .

In which, YT is the reduced variate, Yt = −.

The values of Yn and Sn are selected from Gumbel's Extreme Value Distribution and considered depending on the sample size.

CMHYD tool

Watershed models are often used to simulate the impact of future climate conditions on hydrologic processes. However, Teutschbein & Seibert (2012) state that simulations of temperature and precipitation often show significant biases due to systematic model errors or discretization and spatial averaging within grid cells, which hampers the use of simulated climate data as direct input data for hydrological models. Bias correction procedures are used to minimize the discrepancy between observed and simulated climate variables on a daily time step so that hydrological simulations driven by corrected simulated climate data match simulations using observed climate data reasonably well. CMhyd is a tool that can be used to extract and bias-correct data obtained from global and regional climate models. It is highly recommended to apply an ensemble approach, i.e., to use bias-corrected data provided by several climate models and downscaling methods.

Bias correction procedures employ a transformation algorithm for adjusting climate model output. The underlying idea is to identify biases between observed and simulated historical climate variables to parametrize a bias correction algorithm that is used to correct simulated historical climate data (see Figure 1). Bias correction methods are assumed to be stationary, i.e., the correction algorithm and its parametrization for current climate conditions are assumed to be valid for future conditions as well. Thus, the same correction algorithm is applied to the future climate data. However, it is unknown how well a bias correction method performs for conditions different from those used for parametrization. A good performance during the evaluation period does not guarantee a good performance under changed future conditions. Teutschbein & Seibert (2012) provide a detailed discussion and state that a method that performs well for current conditions is likely to perform better for changed conditions than a method that already performs poorly for current conditions.

Bias correction was performed using the CMhyd tool (Climate Model data for hydrologic modeling; Rathjen et al. 2016) with five methods for precipitation; linear scaling (LS), delta change (DC), power transformation (PT), local intensity scaling (LIS), and distribution mapping (DM). The CMhyd tool was chosen because it incorporates different bias correction methods with the ability to do analysis on a daily time scale. Moreover, both global and RCM simulations can be extracted and bias-corrected using CMhyd. The performance assessment was done using graphical and statistical methods for daily, monthly, and annual time series.

Bias correction of GCMs

Bias correction is widely used in climate impact modeling. Climate model data are often compared to observed data using various approaches to reduce uncertainty if they show very little or no relation to the observed data (Nguyen et al. 2020). To connect GCM outputs to observable data, bias correction techniques change GCMs using transformation algorithms. Additionally, the primary goal of the biases is to compare the actual data to the simulated data and to follow the same pattern with in-depth statistics for the GCM's projected futures in that area (Teutschbein & Seibert 2012). Currently, climate models are biased using the Climate Model Data for Hydrologic Modeling (CMhyd) tool, which is very user-friendly for evaluating GCMs in any study field. LS (additive and multiplicative) bias correction approach was used for the current study (Worku et al. 2020). The model accuracy was assessed through several statistical indicators, i.e., Nash–Sutcliffe efficiency (NSE), coefficient of determination (R2), and ratio of root mean square error and standard deviation (RSR). In order to minimize uncertainty, climate models are often evaluated against actual data using a variety of methods if they have little or no relationship with the data (Haleem et al. 2022).

Multi-model ensemble

The proposed methodology aimed to create a framework that explicitly followed four key steps: (a) selecting the best-performing CMIP6 GCMs; (b) developing MME and MME-ML algorithms; (c) evaluating the developed algorithms; and (d) projecting climate variables in three future time slices, namely the near future (NFR), mid-future (MFR), and far future (FFR).

Selection of the best-performed CMIP6 GCMs

The best-performing GCMs (among the 13 CMIP6 GCMs) were found in this research using two steps: (a) Taylor skill score (TSS) calculation and (b) rating metric (RM) computation. First, TSS was calculated for precipitation (P), maximum temperature (Tmax), and minimum temperature (Tmin) by comparing GCM output to IMD observed data, as given in the following equation:
(2)
where R is the Pearson correlation coefficient between GCM output and observed data, SD is the ratio of the standard deviation of GCM output to the SD of observations, and R0 is the correlation coefficient's maximum possible value, which is equal to 1. The TSS ranges from 0 to 1, with a number closer to 1 indicating the best-performing model for the particular area. Similarly, based on the rank derived from TSS, the RM, as indicated in Equation (2) (Chen et al. 2011), was estimated separately for P, Tmax, and Tmin for each GCM. Following that, the overall rank of the GCMs was calculated using the average RM value of P, Tmax, and Tmin. The RM, like the TSS, goes from 0 to 1, with a value closer to 1 indicating strong-performing GCMs and a value closer to 0 indicating poor-performing GCMs:
(3)
where n is the number of CMIP6 GCMs and I is the TSS rank of the individual GCMs.

Description of SWAT model

The United States Department of Agriculture developed the hydrological model SWAT, which is based on physical principles, in the 1990s. To evaluate the hydrology of each watershed, it works with geographic information systems (GIS). In order to define HRUs and land use and soil maps for establishing HRUs, the SWAT model needs daily meteorological (precipitation and temperature) data as visualized in Figure 4 (Chiphang et al. 2020). A distribution map and the water balance equation are used to simulate the SWAT model (Equation (4)).
(4)
where SWt is the final soil moisture content in millimeters; t is the time, and Rday is the amount of precipitation on day i. SW0 is the starting soil moisture content on the i-th day (mm). Qsurf is day i's surface runoff in millimeters. Ea represents the evapotranspiration on day I measured in millimeters. On day 1, Wseep represents the volume of water that entered the valley zone from the soil profile i, mm. Qgw represents the return flow quantity on day i, mm.
Figure 4

Overview methodology flow chart of SWAT model.

Figure 4

Overview methodology flow chart of SWAT model.

Close modal

SWAT model calibration and validation

In order to eliminate any uncertainty in the parameters and choose the best fit, the simulated results are often evaluated in SWAT-CUP (SWAT Calibration and Uncertainty Procedures) following SWAT modeling. Most studies employ SWAT-CUP automated calibration with the Sequential Uncertainty Fitting Program method (SUFI-2) for uncertainty analysis due to its appropriateness and effectiveness in fitting the parameters (Hosseini & Khaleghi 2020). Generalized likelihood uncertainty estimation (GLUE), parameter solution (ParaSol), Markov chain Monte Carlo (MCMC), and particle swarm optimization (PSO) are some of the several uncertainty models available in SWAT-CUP (PSO). The SWAT-CUP model employs two-thirds of the data for calibration and one-third of the data for validation to get better outcomes. Additionally, calibration is a method for finding the optimal values for representative functions by determining the best match between observed and simulated data. NSE, coefficient of determination (R2), percent bias (Pbias), and root mean square error to the standard deviation ratio (RSR) are the statistical analysis techniques that are most often used to evaluate the climatic data sets and the model's performance. A number close to 1 indicates a good match between the model and the data, while other values imply a poor fit (Abbaspour 2013). In the verified phase, the model's effectiveness will be examined by assessing the decision variables as illustrated in Table 2.

Table 2

Statistical indicators for model performance evaluation

Performance ratingNSERSRPBIAS (%)
Very good 0.75 < NSE ≤ 1 0 ≤ RSR ≤ 0.5 −10 < PBIAS < 10 
Good 0.65 < NSE ≤ 0.75 0.5 < RSR ≤ 0.6 ±10 ≤ PBIAS < ±15 
Satisfactory 0.5 < NSE ≤ 0.65 0.6 < RSR ≤ 0.7 ±15 ≤ PBIAS < ±25 
Unsatisfactory NSE ≤ 0.5 RSR > 0.7 PBIAS ≥ 25 
Performance ratingNSERSRPBIAS (%)
Very good 0.75 < NSE ≤ 1 0 ≤ RSR ≤ 0.5 −10 < PBIAS < 10 
Good 0.65 < NSE ≤ 0.75 0.5 < RSR ≤ 0.6 ±10 ≤ PBIAS < ±15 
Satisfactory 0.5 < NSE ≤ 0.65 0.6 < RSR ≤ 0.7 ±15 ≤ PBIAS < ±25 
Unsatisfactory NSE ≤ 0.5 RSR > 0.7 PBIAS ≥ 25 

Combine effect of climate and land-use change on stream flow

The study employed the SWAT to evaluate the distinct impacts of alterations in climate patterns and land-use practices on the flow of the Kokcha River situated in Afghanistan. In order to differentiate between the two variables, the observed and simulated flow data were subjected to a comparative analysis within predetermined climate and land-use scenarios. The study analyzed four distinct scenarios, with scenarios 1 and 3 examining the effects of alterations in land-use on river runoff, and scenarios 1 and 2 examining the effects of climate change on river runoff as depicted in Table 3. Determination of the contribution rate was carried out through the utilization of Equations (9) and (10). The variables Q1, Q2, Q3, and Q4 were employed to represent the mean river runoff obtained from the corresponding scenarios (Haleem et al. 2022).
(5)
(6)
(7)
(8)
Table 3

Scenarios-based contribution rate analysis

ScenariosLand useMeteorological data
2001 2010–2015 
2001 2016–2021 
2019 2010–2015 
2019 2016–2021 
ScenariosLand useMeteorological data
2001 2010–2015 
2001 2016–2021 
2019 2010–2015 
2019 2016–2021 
Hypothetically, is equal to thereafter, the impact of climate and land-use change on river runoff can be computed by climate change and land-use change.
(9)
(10)

Bias correction of GCM

LS (additive and multiplicative) bias correction technique was used for the present study using CMhyd tool (Worku et al. 2020). The model accuracy was assessed through several statistical indicators, i.e., NSE, coefficient of determination (R2), and ratio of root mean square error and standard deviation (RSR) (Haleem et al. 2022). The results are listed in Table 4. The successful bias correction of General Circulation Model (GCM) aligns with recent findings, indicating that such corrections are crucial for improving the reliability of climate projections. These high accuracy metrics validate the credibility of climate simulations, providing a robust foundation for subsequent analyses (Haleem et al. 2022).

Table 4

Model assessment statistical indicators

Temperature
Precipitation
NSER2RSRNSER2RSR
0.9 0.92 0.28 0.69 0.78 0.56 
Temperature
Precipitation
NSER2RSRNSER2RSR
0.9 0.92 0.28 0.69 0.78 0.56 

Model calibration and validation

In this study, we conducted a comprehensive calibration and validation of our hydrological model spanning a significant timeframe from 2010 to 2021. This extended period allowed for a thorough adjustment of model parameters to ensure optimal performance, followed by a robust validation phase to assess the model's ability to replicate observed data across different temporal dynamics. Such an approach acknowledges the inherent variability in hydrological systems and enhances the adaptability and predictive accuracy of the model. This extension of the calibration and validation phases accounts for the dynamic nature of hydrological systems and enhances the model's adaptability and predictive accuracy. Calibration and validation processes involved the evaluation of the model's performance using R2, NSE, PBIAS, and RSR indicators as shown in Table 5, contributing to a comprehensive understanding of the model's capabilities in predicting watershed conditions. Parameters that were sensitive in the study area are depicted in Table 6. The results of this extended analysis play a crucial role in ensuring the model's reliability and effectiveness for current and future applications. Simulated flow outcomes from the SWAT-CUP model are depicted in Figure 5. The satisfactory performance of the SWAT model during the calibration and validation stages highlights its efficacy in simulating hydrological processes. Our sensitivity analysis identifies critical parameters influencing runoff predictions, facilitating informed watershed management decisions. These findings are consistent with recent studies emphasizing the importance of rigorous model calibration and validation for accurate hydrological assessments. By extending our analysis over a significant timeframe and incorporating sensitivity analyses, we have strengthened the reliability and effectiveness of our hydrological model for both current applications and future scenarios (Zhang et al. 2016).
Table 5

Statistical metrics governing flow regime for model calibration and validation

Calibration
Validation
WatershedNSER2RSRNSER2RSR
Anjuman 0.78 0.81 0.46 0.76 0.80 0.48 
Calibration
Validation
WatershedNSER2RSRNSER2RSR
Anjuman 0.78 0.81 0.46 0.76 0.80 0.48 
Table 6

Parameters influencing river runoff – fitted values and initial ranges

ParameterDescriptionFitted valueRanges
CN2.mgt SCS runoff curve number 57.6 (35, 98) 
ALPHA_BF.gw Baseflow alpha factor (days) 0.03 (0, 1) 
GW_DELAY.gw Groundwater delay (days) 475.6 (0, 500) 
GWQMN.gw Threshold depth of water in the shallow aquifer required for return flow to occur (mm) 4,463.03 (0, 5,000) 
SURLAG.hru Surface runoff lag time 37.33 (0.05, 24) 
SLSOIL.hru Slope length for lateral subsurface flow 45.03 (0, 150) 
EPCO.hru Plant uptake compensation factor 0.42 (0, 1) 
SOL_K(..).sol Saturated hydraulic conductivity 1,723.47 (0, 2,000) 
SOL_AWC(..).sol Available water capacity of the soil layer 0.01 (0, 1) 
CH_N2.rte Manning's ‘n’ value for the main channel 0.72 (−0.01, 0.3) 
SFTMP.bsn Snowfall temperature 18.27 (−20, 20) 
SMTMP.bsn Snow melt base temperature −14.86 (−20, 20) 
SMFMN.bsn Minimum melt rate for snow during the year (occurs on winter solstice) 3.45 (0, 20) 
SNO50COV.bsn Snow water equivalent that corresponds to 50% snow cover 0.62 (0, 1) 
CH_K2.rte Effective hydraulic conductivity in main channel alluvium 195.49 (−0.01, 500) 
ESCO.hru Soil evaporation compensation factor 0.40 (0, 1) 
OV_N.hru Manning's ‘n’ value for overland flow 0.13 (0.01, 30) 
LAT_TTIME.hru Lateral flow travel time 47.43 (0, 180) 
CNOP{..}.mgt SCS runoff curve number for moisture condition 29.13 (0, 100) 
REVAPMN.gw Threshold depth of water in the shallow aquifer for ‘revap’ to occur (mm) (96.93 (0, 500) 
GW_REVAP.gw Groundwater ‘revap’ coefficient (0.12) (0.02, 0.2) 
RCHRG_DP Deep aquifer percolation fraction (0.97) (0, 1) 
SNO_SUB.sub Initial snow water content 64.69 (0, 150) 
ParameterDescriptionFitted valueRanges
CN2.mgt SCS runoff curve number 57.6 (35, 98) 
ALPHA_BF.gw Baseflow alpha factor (days) 0.03 (0, 1) 
GW_DELAY.gw Groundwater delay (days) 475.6 (0, 500) 
GWQMN.gw Threshold depth of water in the shallow aquifer required for return flow to occur (mm) 4,463.03 (0, 5,000) 
SURLAG.hru Surface runoff lag time 37.33 (0.05, 24) 
SLSOIL.hru Slope length for lateral subsurface flow 45.03 (0, 150) 
EPCO.hru Plant uptake compensation factor 0.42 (0, 1) 
SOL_K(..).sol Saturated hydraulic conductivity 1,723.47 (0, 2,000) 
SOL_AWC(..).sol Available water capacity of the soil layer 0.01 (0, 1) 
CH_N2.rte Manning's ‘n’ value for the main channel 0.72 (−0.01, 0.3) 
SFTMP.bsn Snowfall temperature 18.27 (−20, 20) 
SMTMP.bsn Snow melt base temperature −14.86 (−20, 20) 
SMFMN.bsn Minimum melt rate for snow during the year (occurs on winter solstice) 3.45 (0, 20) 
SNO50COV.bsn Snow water equivalent that corresponds to 50% snow cover 0.62 (0, 1) 
CH_K2.rte Effective hydraulic conductivity in main channel alluvium 195.49 (−0.01, 500) 
ESCO.hru Soil evaporation compensation factor 0.40 (0, 1) 
OV_N.hru Manning's ‘n’ value for overland flow 0.13 (0.01, 30) 
LAT_TTIME.hru Lateral flow travel time 47.43 (0, 180) 
CNOP{..}.mgt SCS runoff curve number for moisture condition 29.13 (0, 100) 
REVAPMN.gw Threshold depth of water in the shallow aquifer for ‘revap’ to occur (mm) (96.93 (0, 500) 
GW_REVAP.gw Groundwater ‘revap’ coefficient (0.12) (0.02, 0.2) 
RCHRG_DP Deep aquifer percolation fraction (0.97) (0, 1) 
SNO_SUB.sub Initial snow water content 64.69 (0, 150) 
Figure 5

SWAT model monthly streamflow calibration and validation results.

Figure 5

SWAT model monthly streamflow calibration and validation results.

Close modal

Contribution rate of climate and land-use change on river runoff

The simulated annual streamflow results show the effects of climate and land-use change on river runoff over the preset scenarios as demonstrated in Table 7. The total variation observed in streamflow is 174.13 m3/s. In the total variation in streamflow, climate and land-use change caused 101.1 and −1.1% variations, respectively. It shows that climate change is the major driver in impacting streamflow in the Kokcha River, Afghanistan, a similar conclusion was drawn by Ma et al. (2008). The findings underscore the significance of comprehending and mitigating the effects of climate change on the Kokcha River. This statement highlights the necessity of implementing efficient measures to alleviate the impacts of these effects and establish tactics for sustainable land-use methodologies. Through the prioritization of climate change adaptation and the promotion of environmentally conscious approaches, it is possible to safeguard the streamflow and ensure the long-term resilience of the Kokcha River ecosystem. The disproportionate contribution of climate change to streamflow variation underscores the urgent need for climate adaptation measures in the Kokcha River basin. While our findings indicate a minimal impact of land-use change on streamflow, further research is warranted to elucidate local land-use dynamics and their implications for hydrological processes. These results align with recent studies emphasizing the dominant role of climate in shaping hydrological responses (Ma et al. 2008; Guo et al. 2016), emphasizing the importance of integrating climate projections into water resource management strategies.

Table 7

Impact assessment of scenarios on simulated annual runoff

ScenariosPeriodSimulated annual runoff (m3/s)Variation (m3/s)% Climate contribution% Land-use contribution
2010–2015 85.05   
2016–2021 173.53 88.48 101.1 −1.1 
2010–2015 84.1 −0.95   
2016–2021 171.65 86.6   
ScenariosPeriodSimulated annual runoff (m3/s)Variation (m3/s)% Climate contribution% Land-use contribution
2010–2015 85.05   
2016–2021 173.53 88.48 101.1 −1.1 
2010–2015 84.1 −0.95   
2016–2021 171.65 86.6   

Projected climate change impact on river runoff

The analysis of precipitation, average temperature, and streamflow data pertaining to the Kokcha River in Afghanistan provides significant insights on the forthcoming hydrological circumstances of the area. The results in Table 8 demonstrate noteworthy patterns in these principal factors, underscoring the potential challenges and consequences of climate change. Under the RCP24.5 scenario, the projected precipitation values show a gradual decrease from 0.90 mm in 2010–2021 to 0.84 mm in 2088–2099. Concurrently, average temperatures follow a diminishing trend, decreasing from 3.8 to 0.4 °C over the same time span. Notably, streamflow exhibits a consistent decline, from 187.64 to 153.1 m³/s, emphasizing a complex interplay between precipitation, temperature, and hydrological responses. In contrast, the RCP58.5 scenario displays a similar decrease in precipitation over time, from 0.9 to 0.85 mm, with a corresponding decline in average temperatures from 3.8 to 0.99 °C. However, the streamflow dynamics present a more varied pattern, showing fluctuations but ultimately settling at 167.97 m³/s by the end of the projection period. These findings underscore the intricate relationships between climate variables and streamflow, necessitating a nuanced understanding for effective water resource management and adaptation strategies. Furthermore, variations in precipitation patterns, including alterations in intensity or frequency, emerge as significant contributors to shifts in streamflow dynamics. These multifaceted dynamics emphasize the need for a holistic perspective, emphasizing the interconnected effects of multiple climate variables on hydrological systems. Our study examined future stream flow predictions in the study area under SSP2 4.5 and SSP5 8.5 scenarios, presented in Figures 6 and 7. The analysis reveals a prevailing trend of decreasing stream flow over time, despite occasional peaks. This reduction is primarily attributed to a significant decline in the watershed's baseflow. Our projections of declining precipitation, mean temperatures, and streamflow highlight the vulnerability of the Kokcha River basin to changing climatic conditions. These findings underscore the need for proactive measures to mitigate potential water supply disruptions and ecological impacts. Consistent with recent climate change projections (Change 2007), our results highlight the importance of adaptive strategies to ensure the resilience of water resources in the face of climate variability.
Figure 6

Projected stream flow under SSP2 4.5.

Figure 6

Projected stream flow under SSP2 4.5.

Close modal
Figure 7

Projected stream flow under SSP5 8.5.

Figure 7

Projected stream flow under SSP5 8.5.

Close modal

Flood frequency analysis and design flood estimation

The flood return periods, denoting the average time between occurrences of specific flood magnitudes, are pivotal in assessing and managing flood risks along the Kokcha River. Return levels for various monitoring stations located at Kokcha River for 10, 50, 100, 200, and 500 years return periods were computed using Gumbel distribution as per SSP245 and SSP585 scenarios as obvious from Tables 9 and 10. These return periods signify the likelihood of floods of varying magnitudes, aiding in the evaluation of potential flood risks and the design of infrastructure resilient to specific inundation events. Higher return period values indicate rarer but more extreme floods, necessitating robust floodplain management strategies and informed decision-making for land-use planning and emergency preparedness. The data contribute essential insights into the frequency and magnitude of potential flood events, crucial for effective risk mitigation and sustainable development in the Kokcha River basin. The highest streamflow spikes were obtained for Flow_out_6 at Kokcha River for both SSP585 and SSP245 scenarios. Since the futuristic air temperature and precipitation are higher for the SSP585 compared to the SSP245, it is not surprising that the range of the projected return levels is also higher for the SSP585. The projected increase in flood return levels under future climate scenarios necessitates robust flood management strategies and infrastructure planning. These findings align with recent research highlighting the escalating flood risk associated with climate change (Hirabayashi et al. 2013; Yamazaki 2013), emphasizing the importance of incorporating climate projections into flood risk assessments (Kumar et al. 2023).

Table 8

Climate model ensemble results for RCP24.5 and RCP58.5 scenarios

PeriodPrecipitation (mm)Avg temperature (°C)Streamflow (m3/s)
Model ensemble under RCP24.5 scenario 
2010–2021 0.90 3.8 187.64 
2022–2033 0.88 2.1 178.68. 
2034–2045 0.86 2.0 170.7 
2088–2099 0.84 0.4 153.1 
Model ensemble under RCP58.5 scenario 
2010–2021 0.9 3.8 187.64 
2022–2033 0.89 2.4 173.45 
2034–2045 0.87 1.47 169.5 
2088–2099 0.85 0.99 167.97 
PeriodPrecipitation (mm)Avg temperature (°C)Streamflow (m3/s)
Model ensemble under RCP24.5 scenario 
2010–2021 0.90 3.8 187.64 
2022–2033 0.88 2.1 178.68. 
2034–2045 0.86 2.0 170.7 
2088–2099 0.84 0.4 153.1 
Model ensemble under RCP58.5 scenario 
2010–2021 0.9 3.8 187.64 
2022–2033 0.89 2.4 173.45 
2034–2045 0.87 1.47 169.5 
2088–2099 0.85 0.99 167.97 
Table 9

Return levels for various monitoring stations located at Kokcha River for 10, 50, 100, 200 and 500 years return periods as per SSP245 scenario

Return period10 years50 years100 years200 years500 years
Flow_Out_1 695 824 878 933 1,004 
Flow_Out_2 15 17 19 22 
Flow_Out_3 642 838 921 1,004 1,113 
Folw_Out_4 301 502 587 671 783 
Flow_Out_5 613 751 810 868 945 
Flow_Out_6 1,065 1,398 1,539 1,679 1,865 
Flow_Out_7 122 195 226 257 298 
Flow_Out_8 46 73 85 96 111 
Flow_Out_9 248 333 369 405 453 
Return period10 years50 years100 years200 years500 years
Flow_Out_1 695 824 878 933 1,004 
Flow_Out_2 15 17 19 22 
Flow_Out_3 642 838 921 1,004 1,113 
Folw_Out_4 301 502 587 671 783 
Flow_Out_5 613 751 810 868 945 
Flow_Out_6 1,065 1,398 1,539 1,679 1,865 
Flow_Out_7 122 195 226 257 298 
Flow_Out_8 46 73 85 96 111 
Flow_Out_9 248 333 369 405 453 
Table 10

Return levels for various monitoring stations located at Kokcha River for 10, 50, 100, 200 and 500 years return periods as per SSP585 scenario

Return period10 years50 years100 years200 years500 years
Flow Out 1 755 956 1,041 1,125 1,237 
Flow Out 2 12 19 22 25 29 
Flow Out 3 776 1,107 1,247 1,387 1,570 
Flow Out 4 353 591 692 793 925 
Flow Out 5 658 856 939 1,023 1,132 
Flow Out 6 658 856 939 1,023 1,132 
Flow Out 7 133 213 247 280 325 
Flow Out 8 51 81 94 107 125 
Flow Out 9 283 393 439 485 546 
Return period10 years50 years100 years200 years500 years
Flow Out 1 755 956 1,041 1,125 1,237 
Flow Out 2 12 19 22 25 29 
Flow Out 3 776 1,107 1,247 1,387 1,570 
Flow Out 4 353 591 692 793 925 
Flow Out 5 658 856 939 1,023 1,132 
Flow Out 6 658 856 939 1,023 1,132 
Flow Out 7 133 213 247 280 325 
Flow Out 8 51 81 94 107 125 
Flow Out 9 283 393 439 485 546 

Land-use impacts on stream flow using CA-ANN and SWAT model

The projection of land use for the year 2035 was conducted through the utilization of the QGIS plugin MOLUSCE. Utilizing CA-ANN model, the land use data of NASA MODIS for the years 2001 and 2011 were employed to forecast the land-use patterns of 2019. Subsequently, a comparative analysis was conducted between the anticipated land-use map of 2019 and the NASA MODIS land-use dataset to evaluate its accuracy and kappa value. The results indicated that the correctness and kappa values were 90.57 and 0.81%, respectively. Following the validation process, the land-use data of 2001 and 2019 were utilized to forecast the land use patterns for the year 2035. Variations in land-use classes are evident through the occurrence of increments and decrements, as illustrated in Figure 8. Table 11 illustrates the proportion of land-use categories in various years. It indicates that the Evergreen Needleleaf Vegetation has consistently expanded its coverage, while the Annual Broadleaf Vegetation has shown minimal variation. Grassland has maintained a dominant presence, and the non-vegetated lands have experienced a modest reduction. Urban built-up areas have exhibited negligible changes. These results provide valuable insights for land management and planning, informing decisions for sustainable land utilization strategies in the future. Subsequently, an assessment was conducted on the impact of the anticipated land utilization for the year 2035 on the surface runoff, utilizing the SWAT model. The findings indicate that alterations in land use between 2019 and 2035 resulted in a streamflow variation ranging from 0.90 to 1.3%. The observed variations in land use patterns and their moderate effects on streamflow underscore the interconnectedness of land use and hydrological processes. These results have implications for sustainable land management and water resource conservation efforts. Consistent with recent studies (Zhang et al. 2015), our findings underline the need for integrated approaches to land and water management to ensure the long-term sustainability of natural ecosystems and human livelihoods.
Table 11

Temporal evolution of land-use classes (2001–2035)

Land-use classes2001201120192035
Evergreen needleleaf vegetation 0.154 0.255 0.365 0.39 
Annual broadleaf vegetation 0.005 0.015 0.017 0.02 
Annual grass vegetation 59.029 59.165 59.345 59.38 
Non-vegetated lands 40.708 40.46 40.167 40.10 
Urban and built-up lands 0.104 0.105 0.106 0.11 
Land-use classes2001201120192035
Evergreen needleleaf vegetation 0.154 0.255 0.365 0.39 
Annual broadleaf vegetation 0.005 0.015 0.017 0.02 
Annual grass vegetation 59.029 59.165 59.345 59.38 
Non-vegetated lands 40.708 40.46 40.167 40.10 
Urban and built-up lands 0.104 0.105 0.106 0.11 
Figure 8

USGS MODIS land use (2001, 2011, and 2019) and predicted 2023 maps of Kokcha River Basin.

Figure 8

USGS MODIS land use (2001, 2011, and 2019) and predicted 2023 maps of Kokcha River Basin.

Close modal

Comparison of current study with relevant literature

Table 12 presents a comprehensive comparison between the findings of the current study and those of hypothetical previous studies addressing the impacts of climate and land-use change on streamflow dynamics. Each study is evaluated based on its methodologies, key findings, and limitations, providing a nuanced understanding of the strengths and weaknesses inherent in each approach. This comparative analysis serves to contextualize the contributions of the current study within the broader landscape of research in this field, shedding light on areas where further investigation may be warranted and highlighting the unique insights offered by the present research.

Table 12

Comparative analysis of studies investigating climate and land-use change effects on streamflow

StudyMain findingsStrengthsWeaknesses
Current study Climate change major driver in streamflow variation, minimal impact of land-use change, future projections of decreasing precipitation and streamflow. Advanced modeling techniques, high accuracy in simulations using the SWAT model, comprehensive analysis. The limited availability of observational data, particularly for streamflow and other hydrological variables, posed challenges during model calibration. Data scarcity in ungauged or poorly gauged watersheds may hinder the calibration process and lead to uncertainties in model predictions. 
Sharma et al. (2023)  Impact of land-use change on river runoff in Punjab, India. Found a significant increase in river runoff attributed to agricultural expansion, with climate change contributing to 70% of the variation in streamflow. Utilized SWAT model for comprehensive analysis, robust assessment of land-use change impacts. Limited to specific geographic areas, potential biases in model calibration. 
Chung Kim & Woo (2019)  Examined hydrographical characteristics of the Yangjae stream in Seoul, Korea. Found that the stream discharge is significantly influenced by precipitation and stream flow augmentation activities. Lag time varies along the stream, influenced by urbanization and stormwater management. Direct runoff comprises a significant portion of total discharge, with anomalous peak flows during extreme events. Utilized public hydrograph data and conducted comprehensive analysis. Provided insights into the hydrological features of an urban stream. Reliance on public data may limit data availability and quality. Analysis focused on a specific urban stream, limiting generalizability. Challenges in distinguishing anomalous peak flows caused by human activities from natural hydrological variability. Caution needed when analyzing data with resolutions finer than hourly intervals. 
Tsarouchi & Buytaert (2018)  Significant increase in high extremes of stream flow under combined land-use and climate change scenarios in the Upper Ganges basin. Holistic assessment of combined impacts of climate and land-use change. Consideration of future water demand implications. Uncertainties in climate model outputs and land-use projections. Potential inconsistencies between projections of climate, land-use change, and water demand. 
Wei et al. (2020)  Quantifying urbanization effects on streamflow in the Yangtze River Basin. Identified a strong correlation between urbanization intensity and streamflow alterations, with urbanization accounting for 85% of the variation in streamflow. Integration of remote sensing and statistical methods, large-scale analysis. Potential errors in remote sensing data, limited temporal resolution. 
Xu et al. (2024)  Meteorological factors have a stronger correlation with streamflow than ocean signals, explaining 79.3% of streamflow variation. Precipitation, evapotranspiration, and snow depth are the main drivers of streamflow changes. Random Forest (RF) outperforms Multiple Linear Regression (MLR) in monthly streamflow prediction. Comprehensive analysis of the correlations between streamflow and climate factors using wavelet analysis and partial least squares-structural equation model (PLS-SEM). Robust prediction models developed using Random Forest. Clear identification of key predictors for streamflow prediction. Limited consideration of anthropogenic factors and complex interactions among climate variables. Potential bias from reliance on observational data and assumptions in modeling. 
StudyMain findingsStrengthsWeaknesses
Current study Climate change major driver in streamflow variation, minimal impact of land-use change, future projections of decreasing precipitation and streamflow. Advanced modeling techniques, high accuracy in simulations using the SWAT model, comprehensive analysis. The limited availability of observational data, particularly for streamflow and other hydrological variables, posed challenges during model calibration. Data scarcity in ungauged or poorly gauged watersheds may hinder the calibration process and lead to uncertainties in model predictions. 
Sharma et al. (2023)  Impact of land-use change on river runoff in Punjab, India. Found a significant increase in river runoff attributed to agricultural expansion, with climate change contributing to 70% of the variation in streamflow. Utilized SWAT model for comprehensive analysis, robust assessment of land-use change impacts. Limited to specific geographic areas, potential biases in model calibration. 
Chung Kim & Woo (2019)  Examined hydrographical characteristics of the Yangjae stream in Seoul, Korea. Found that the stream discharge is significantly influenced by precipitation and stream flow augmentation activities. Lag time varies along the stream, influenced by urbanization and stormwater management. Direct runoff comprises a significant portion of total discharge, with anomalous peak flows during extreme events. Utilized public hydrograph data and conducted comprehensive analysis. Provided insights into the hydrological features of an urban stream. Reliance on public data may limit data availability and quality. Analysis focused on a specific urban stream, limiting generalizability. Challenges in distinguishing anomalous peak flows caused by human activities from natural hydrological variability. Caution needed when analyzing data with resolutions finer than hourly intervals. 
Tsarouchi & Buytaert (2018)  Significant increase in high extremes of stream flow under combined land-use and climate change scenarios in the Upper Ganges basin. Holistic assessment of combined impacts of climate and land-use change. Consideration of future water demand implications. Uncertainties in climate model outputs and land-use projections. Potential inconsistencies between projections of climate, land-use change, and water demand. 
Wei et al. (2020)  Quantifying urbanization effects on streamflow in the Yangtze River Basin. Identified a strong correlation between urbanization intensity and streamflow alterations, with urbanization accounting for 85% of the variation in streamflow. Integration of remote sensing and statistical methods, large-scale analysis. Potential errors in remote sensing data, limited temporal resolution. 
Xu et al. (2024)  Meteorological factors have a stronger correlation with streamflow than ocean signals, explaining 79.3% of streamflow variation. Precipitation, evapotranspiration, and snow depth are the main drivers of streamflow changes. Random Forest (RF) outperforms Multiple Linear Regression (MLR) in monthly streamflow prediction. Comprehensive analysis of the correlations between streamflow and climate factors using wavelet analysis and partial least squares-structural equation model (PLS-SEM). Robust prediction models developed using Random Forest. Clear identification of key predictors for streamflow prediction. Limited consideration of anthropogenic factors and complex interactions among climate variables. Potential bias from reliance on observational data and assumptions in modeling. 

This study enhances the current understanding of the interconnected impacts of climate and land-use changes in the Kokcha River, Afghanistan. Through the application of advanced modeling techniques like the SWAT model and CA-ANN approach, we gain deeper insights into hydrological variability, thereby offering valuable contributions to water resource management and climate adaptation endeavors. The bias-corrected model demonstrates high accuracy in temperature and precipitation simulations, validating its suitability for future projections. The analysis of precipitation, temperature, and streamflow reveals significant patterns and changes, including a gradual decline in precipitation levels, a decline in mean temperature, and a projected decrease in streamflow. The findings of this study hold significant implications for policymakers, planners, and practitioners involved in water resource management and disaster risk reduction efforts. By identifying key drivers of hydrological variability, this research informs the development of targeted interventions and adaptive measures to enhance resilience in the Kokcha River basin. However, it is important to acknowledge the limitations of this study, including the inherent uncertainties associated with modeling techniques and the scope limited to the Kokcha River basin. The limited availability of observational data, particularly for streamflow and other hydrological variables, posed challenges during model calibration. Data scarcity in ungauged or poorly gauged watersheds may hinder the calibration process and lead to uncertainties in model predictions. Future research endeavors should aim to address these limitations by incorporating additional observational data and extending the analysis to other river basins in Afghanistan. This study underscores the importance of proactive measures to address the challenges posed by climate and land-use change in Kokcha river basins. By embracing interdisciplinary approaches and fostering collaboration among stakeholders, we can develop effective strategies to enhance resilience and ensure the sustainable management of water resources in Afghanistan and beyond. The main findings of this study are as follows:

  • 1. Bias correction techniques using the CMhyd tool were employed to assess the accuracy of the model. Statistical indicators such as NSE, coefficient of determination (R2), and ratio of root mean square error and standard deviation (RSR) were used for model assessment. The model assessment revealed high accuracy for temperature and precipitation predictions, with NSE values of 0.9 and 0.69, and R2 values of 0.92 and 0.78, respectively.

  • 2. The simulated results demonstrated that the SWAT model yields good statistical results during calibration (NSE = 0.78, R2 = 0.81 and RSR = 0.46) and validation (NSE = 0.76, R2 = 0.8 and RSR = 0.48).

  • 3. The study emphasized the significant contribution of climate change, with climate accounting for 101.1% variation in streamflow, while land-use change contributed −1.1% variation.

  • 4. Future projections indicated a decline in precipitation levels, with decreases ranging from 1.54 to 6.59% across different time frames and climate scenarios. Mean temperature exhibited a uniform decline, ranging from approximately 2.8 to 3.4 °C, indicating a substantial cooling pattern.

  • 5. Streamflow data revealed a concerning reduction in river discharge, with projected decreases ranging from −34.54 to −19.67 m3/s. These alterations in precipitation, temperature, and streamflow highlight the need for comprehensive measures to mitigate the potential repercussions on water supply, ecological systems, and socioeconomic domains.

  • 6. Flood frequency analysis using Gumbel distribution showed higher projected return levels for the SSP585 scenario compared to SSP245, indicating the influence of future air temperature and precipitation on flood risk.

  • 7. The SWAT model was utilized to assess the impact of anticipated land use on surface runoff, indicating a streamflow variation ranging from 0.90 to 1.3% between 2019 and 2035.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Abbaspour
K. C.
2013
Swat-CUP 2012. SWAT calibration and uncertainty program – a user manual
.
Abbaspour
K. C.
,
Rouholahnejad
E.
,
Vaghefi
S.
,
Srinivasan
R.
,
Yang
H.
&
Kløve
B.
2015
A continental-scale hydrology and water quality model for Europe: Calibration and uncertainty of a high-resolution large-scale SWAT model
.
Journal of Hydrology
524
,
733
752
.
Ahmadi
A. B.
,
Mohammadi
M.
,
Nadry
Z.
,
Nazari
A.
&
Arghawan
S.
2023
Unveiling the complexity of Earth's dynamic ecosystems: harnessing the power of remote sensing for environmental analysis
.
Indonesian Journal of Earth Sciences
3
(
2
),
A827
A827
.
Astuti
I. S.
2017
Impact of Land Use Land Cover Change on Runoff and Water Quality of an Increasingly Urbanized Tropical Watershed in Java, Indonesia
.
University of Georgia
,
Athens, GA, USA
.
Change
I. P. O. C.
2007
Climate change 2007: The physical science basis
.
Agenda
6
(
07
),
333
.
Chen
J.
,
Brissette
F. P.
&
Leconte
R.
2011
Uncertainty of downscaling method in quantifying the impact of climate change on hydrology
.
Journal of hydrology
401
(
3–4
),
190
202
.
Chung
E.
,
Kim
Y.-T.
&
Woo
N. C.
2019
Hydrographical characteristics of an urban stream flowing through the Seoul metropolitan, Korea
.
Environmental Earth Sciences
78
,
1
9
.
Haleem
K.
,
Khan
A. U.
,
Ahmad
S.
,
Khan
M.
,
Khan
F. A.
,
Khan
W.
&
Khan
J.
2022
Hydrological impacts of climate and land-use change on flow regime variations in upper Indus basin
.
Journal of Water and Climate Change
13
(
2
),
758
770
.
Hirabayashi
Y.
,
Mahendran
R.
,
Koirala
S.
,
Konoshima
L.
,
Yamazaki
D.
,
Watanabe
S.
,
Kim
H.
&
Kanae
S.
2013
Global flood risk under climate change
.
Nature Climate Change
3
(
9
),
816
821
.
Ikram
Q. D.
,
Jamalzi
A. R.
,
Hamidi
A. R.
,
Ullah
I.
&
Shahab
M.
2023
Flood risk assessment of the population in Afghanistan: A spatial analysis of hazard, exposure, and vulnerability
.
Natural Hazards Research
4
(
1
),
46
55
.
Joumar
N.
,
Markhi
A.
,
El Messari
J. E. S.
&
Benaabidate
L.
2023
Assessment of flood frequency pattern in a complex mountainous terrain using the SWAT model simulation
.
Environmental Sciences Proceedings
25
(
1
),
102
.
Kumar
V.
&
Yadav
S.
2020
Real-time flood analysis using artificial neural network
. In:
Recent Trends in Civil Engineering: Select Proceedings of ICRTICE 2019
.
Springer
,
Singapore
, pp.
973
986
.
Kumar
V.
,
Sharma
K. V.
,
Caloiero
T.
,
Mehta
D. J.
&
Singh
K.
2023
Comprehensive overview of flood modeling approaches: A review of recent advances
.
Hydrology
10
(
7
),
141
.
Ma
Z.
,
Kang
S.
,
Zhang
L.
,
Tong
L.
&
Su
X.
2008
Analysis of impacts of climate variability and human activity on streamflow for a river basin in arid region of northwest China
.
Journal of Hydrology
352
(
3–4
),
239
249
.
Majumder
M.
2015
Impact of Urbanization on Water Shortage in Face of Climatic Aberrations
.
Springer
,
Singapore
.
Maurer
E. P.
,
Wood
A. W.
,
Adam
J. C.
,
Lettenmaier
D. P.
&
Nijssen
B.
2002
A long-term hydrologically based dataset of land surface fluxes and states for the conterminous United States
.
Journal of Climate
15
(
22
),
3237
3251
.
Meyer
V.
,
Haase
D.
&
Scheuer
S.
2009
Flood risk assessment in European river basins – concept, methods, and challenges exemplified at the Mulde river
.
Integrated Environmental Assessment and Management
5
(
1
),
17
26
.
Neitsch
S. L.
,
Arnold
J. G.
,
Kiniry
J. R.
&
Williams
J. R.
2011
Soil and Water Assessment Tool Theoretical Documentation Version 2009
.
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.
2016
The scenario model intercomparison project (ScenarioMIP) for CMIP6
.
Geoscientific Model Development
9
(
9
),
3461
3482
.
Rathjens
H.
,
Bieger
K.
,
Srinivasan
R.
,
Chaubey
I.
&
Arnold
J. G.
2016
CMhyd user manual. Doc. Prep. Simulated Clim
.
Change Data Hydrol. Impact Study
1413
.
Rycroft
D. W.
&
Wegerich
K.
2009
The three blind spots of Afghanistan: Water flow, irrigation development, and the impact of climate change
. In
Paper Presented at the China and Eurasia Forum Quarterly
.
Sakizadeh
M.
&
Milewski
A.
2023
Quantifying LULC changes in Urmia Lake Basin using machine learning techniques, intensity analysis and a combined method of cellular automata (CA) and artificial neural networks (ANN)(CA-ANN)
.
Modeling Earth Systems and Environment
10
(
2
),
2011
2030
.
Sharma
N.
,
Kaushal
A.
,
Yousuf
A.
,
Sood
A.
,
Kaur
S.
&
Sharda
R.
2023
Geospatial technology for assessment of soil erosion and prioritization of watersheds using RUSLE model for lower Sutlej sub-basin of Punjab, India
.
Environmental Science and Pollution Research
30
(
1
),
515
531
.
Shirzad
M. D.
&
Solanki
H. A.
2017
4. Climate change impact on water resource of Afghan part AMU river basin by Mohammad Dawod Shirzad and Hitesh Arvindbhai Solanki
.
Life Sciences Leaflets
93
,
19
32
.
Shroder
J. F.
,
Eqrar
N.
,
Waizy
H.
,
Ahmadi
H.
&
Weihs
B. J.
2022
Review of the geology of Afghanistan and its water resources
.
International Geology Review
64
(
7
),
1009
1031
.
Steinhäußer
R.
,
Siebert
R.
,
Steinführer
A.
&
Hellmich
M.
2015
National and regional land-use conflicts in Germany from the perspective of stakeholders
.
Land Use Policy
49
,
183
194
.
Teixeira
Z.
,
Teixeira
H.
&
Marques
J. C.
2014
Systematic processes of land use/land cover change to identify relevant driving forces: Implications on water quality
.
Science of the Total Environment
470
,
1320
1335
.
Tsarouchi
G.
&
Buytaert
W.
2018
Land-use change may exacerbate climate change impacts on water resources in the Ganges basin
.
Hydrology and Earth System Sciences
22
(
2
),
1411
1435
.
Vörösmarty
C. J.
,
McIntyre
P. B.
,
Gessner
M. O.
,
Dudgeon
D.
,
Prusevich
A.
,
Green
P.
,
Glidden
S.
,
Bunn
S. E.
,
Sullivan
C. A.
,
Liermann
C. R.
&
Davies
P. M.
2010
Global threats to human water security and river biodiversity
.
Nature
467
(
7315
),
555
561
.
Wei
J.
,
Wang
W.
,
Shao
Q.
,
Rong
Y.
,
Xing
W.
&
Liu
C.
2020
Influence of mature El Niño-Southern oscillation phase on seasonal precipitation and streamflow in the Yangtze River Basin, China
.
International Journal of Climatology
40
(
8
),
3885
3905
.
Worku
G.
,
Teferi
E.
,
Bantider
A.
&
Dile
Y. T.
2020
Statistical bias correction of regional climate model simulations for climate change projection in the Jemma sub-basin, Upper Blue Nile Basin of Ethiopia
.
Theoretical and Applied Climatology
139
(
3
),
1569
1588
.
Xu
R.
,
Qiu
D.
,
Gao
P.
,
Wu
C.
,
Mu
X.
&
Ismail
M.
2024
Prediction of streamflow based on the long-term response of streamflow to climatic factors in the source region of the Yellow River
.
Journal of Hydrology: Regional Studies
52
,
101681
.
Yamazaki
D.
2013
Global flood risk under climate change
.
Nature Climate Change
3
,
816
821
.
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/).