Abstract This study evaluates the performance of physically-based and deep learning models in simulating runoff and estimating terrestrial water storage (TWS) in the Hablehroud River Basin, a semi-arid watershed in northern Iran with increasing irrigation demands. Two semi-distributed and physically-based models, including SWAT (Soil and Water Assessment Tool), VIC (Variable Infiltration Capacity), and lumped and semi-distributed configurations of Bidirectional Long Short-Term Memory (BLSTM-L and BLSTM-S), were applied using daily meteorological and hydrometric data. The GLEAM v4.2 (Global Land Evaporation Amsterdam Model) dataset was used to estimate evapotranspiration, and a water balance method was used to determine monthly TWS. The monthly TWS results from each model varied considerably, especially during the growing season, but the annual storage estimates from each model exhibited a similar bias. The BLSTM-S model showed excellent consistency in monthly TWS estimation and the highest accuracy in streamflow simulation (NSE = 0.87, KGE = 0.91). According to observational analysis, BLSTM-S best represented the seasonal pattern of water being withdrawn during the agricultural months and primarily stored in the winter and early spring (often as snow in mountainous regions). These results suggest that in areas affected by irrigation, monthly TWS is a more sensitive indicator of model performance. Although physically-based models offer process transparency, their higher monthly biases can reduce their reliability in short-term water allocation. The study highlights the added value of deep learning, particularly semi-distributed BLSTM, in improving both runoff simulation and seasonal water storage representation for operational water management.

ABSTRACT Anthropogenic activities, particularly ecological and water conservancy projects, have profoundly altered land surfaces. However, quantifying their basin‐scale hydrological impacts remains challenging. This study focused on the Qin River Basin (QRB), a major tributary in the ecologically sensitive middle reaches of the Yellow River where such projects are extensively implemented. Employing the Soil and Water Assessment Tool (SWAT) model, we quantified the impacts of the Returning Agricultural Land to Forest (RAF) and Returning Agricultural Land to River (RAR) projects on QRB hydrological processes from 2010 to 2018. Results indicated that the SWAT model performed robustly with Nash‐Sutcliffe efficiency coefficients ( NSE ) of 0.70 to 0.72, determination coefficients ( R 2 ) of 0.71 to 0.79 and percent bias ( PBIAS ) below 11%. RAF implementation reduced total basin runoff by 3.00%, driven by increased surface runoff up to 3.89%, decreased lateral flow by 11.46%, and significantly increased groundwater flow by 2366.67%. Spatially, surface runoff increases concentrated in the northern basin and eastern tributaries, while lateral flow reductions were most severe in central regions. For RAR projects, hydrological responses scaled positively with agricultural‐to‐waterbody conversion area, showing 2.5‐fold greater efficacy on slopes under 15° versus under 6°. Basin‐wide runoff components increased proportionally to buffer width expansion, though spatial tradeoffs emerged: upper reaches exhibited surface runoff reduction from 1.29% to 21.74%, while lower reaches experienced groundwater depletion from 1.99% to 36.96% decrease. These findings highlight that spatial planning informed by slope conditions is vital for effective water resource management in semi‐humid basins under intensive anthropogenic intervention.

Ministry of Forestry has designated Kreo Sub-watershed, part of Garang Watershed, a critical area due to high erosion rates contributing to flooding in Semarang. Rapid land use changes accelerate environmental degradation, increasing erosion and sedimentation risks. This study measures erosion and sedimentation rates in Kreo Sub-watershed using SWAT (Soil and Water Assessment Tool), determines Erosion Hazard Index, and proposes erosion control solutions based on Land Rehabilitation and Soil Conservation Analysis (ARLKT) with vegetative conservation. ARLKT approach includes simulating new land use scenarios to assess their impact on erosion reduction. To ensure SWAT modelling accurately represents field conditions and not overestimate, allowing conservation recommendations based on ARLKT applied appropriately, a field-based sedimentation analysis also conducted. The study utilizes rainfall, soil type, slope, and land use data in 2019 and 2024 from satellite imagery and validated using a confusion matrix. Results indicate a shift in Erosion Hazard Index from predominantly ‘Moderate’ in 2019 to ‘High’ in 2024, underscoring urgent need for sustainable watershed management. By integrating remote sensing, field validation, and hydrological modeling, this study offers a precise, data-driven approach to erosion control. The findings serve critical reference for policymakers in developing effective conservation strategies to enhance watershed resilience.

In response to the issues of low runoff prediction accuracy and difficulty in parameter determination in regions frequently experiencing extreme hydrological events, this study is based on data such as digital elevation, land use, soil type, and meteorology. The SWAT-LSTM (Long Short-Term Memory) model is coupled based on the Bayesian Model Averaging (BMA) method. The simulation accuracies of the optimized model are, respectively, compared with those of the SWAT (Soil and Water Assessment Tool) model and the SWAT-LSTM model. Taking the Zuli River Basin as an example, the optimal runoff prediction model for this basin is determined. Combining with future meteorological data, runoff predictions for the period from 2025 to 2030 are carried out. The findings indicate that the SWAT-LSTM-BMA coupled model is the optimal runoff prediction model for the Zuli River Basin. Compared with the SWAT model and the SWAT-LSTM model used alone, its simulation accuracy has been systematically improved. During the calibration period, R2 increased by 8–12%, NSE increased by 9–13%, and MSE decreased by 14–30%. During the validation period, R2 increased by 10–12%, NSE increased by 10–14%, and MSE decreased by 16–31%. Based on the model and the prediction of future climate data under multiple scenarios, the annual runoff of the basin will show a decreasing trend compared with the historical period between 2025 and 2030, with a decrease of 12–15%. The coupling framework proposed in this study effectively improves the accuracy of runoff prediction and provides a reliable theoretical foundation and technological assistance for revealing the evolution law of extreme hydrological events and the management of water resources in the basin.

Land use/cover (LULC) changes has a fundamental effect on the hydrological components in the Lake Tana Basin. The Lake Tana Basin, the source origin of the Blue Nile, has experienced notable LULC transitions over the past two decades. The present study evaluates the effect of land use/cover (LULC) changes on hydrological components in the Lake Tana basin using the soil and water assessment tool (SWAT). Two LULC maps, one from the International Livestock Research Institute (ILRI) for 2004 and another developed from Landsat 8 images for 2021, were used. Both models were calibrated and validated by SUFI-2 using observed discharge data, results showed strong performance (NSE > 0.79, R2 > 0.79 calibration; NSE > 0.90, R2 > 0.94 validation). Between 2004 and 2021, agricultural land decreased by 10.2% and forest cover declined by 33.1%, while wetlands and rangelands increased by 81.4 and 299.2%, respectively. Moreover, urban land was presented as a new class. These changes affected the basin’s hydrology as surface runoff increased from 111.6 to 118 mm/year (+ 5.8%), lateral flow decreased from 106.3 to 100.7 mm/year, and shallow aquifer evaporation declined by 10.2%. Evapotranspiration remained nearly constant at 1066 mm/year dominated by the lake evaporation. The results confirm the significant influence of LULC changes on the hydrological components of the Lake Tana Basin which highlight the need for sustainable land and water management.

Agricultural arsenic pollution poses increasing environmental and public health challenges. Making evidence-based conservation strategy is key for effective pollution control, but is challenged by data scarcity which is common in China. To address the scarcity of monitoring data, we developed an integrated methodology combining the Soil and Water Assessment Tool (SWAT) and the Load Estimator (LOADEST) to assess long-term variations in the arsenic load within the Zhangjiang River (ZR) watershed, China. Our findings suggest that approximately 1% of the urbanized area may contribute to up to 75% of the current stream arsenic load (a preliminary inference based on load differences between GTDK and upstream sites), though this conclusion is constrained by data limitations (e.g., stream flow parameters transferred from an adjacent watershed, limited arsenic monitoring scope, and low NSE at GTDK). This area could be a potential pollution hotspot, while diffuse arsenic pollution across the watershed is on the rise due to expanding agriculture, increased contaminated manure usage and the shifting hydroclimatic condition. Results showed that recycling arsenic-rich animal waste as manure could have the unintended consequence of building up an arsenic storage pool in farmland soils, turning croplands into pollution sources and increasing the risk of diffuse arsenic pollution, thus calling for adjustment in current agricultural management strategy. The proposed modeling method proves as a promising tool for investigating arsenic pollution in data-sparse region, supporting the assessment and optimization of agricultural management practices and policies for arsenic pollution control.

ABSTRACT Simulating hydrological and sediment conditions in data‐scarce regions is a key challenge in hydrology. This study focused on 14 data‐scarce tributaries in the Henan boundary area of the Sanmenxia Reservoir. Using methods like parameter transfer, coefficient of variation, and watershed similarity evaluation, we selected data‐rich reference watersheds and applied the Soil and Water Assessment Tool (SWAT) model to simulate runoff and sediment yield. Calibrated parameters were then transferred to the study watersheds. Results showed: (1) The Ba River and Hongnongjian River had a similarity of 0.89, while the Jian River and other small watersheds in the area had similarities between 0.75 and 0.95, indicating the Ba River and Jian River were suitable as reference watersheds. (2) The SWAT model simulated runoff and sediment yield for these watersheds from 1960 to 2022, with results calibrated and validated. (3) Transferred parameters were used to simulate runoff and sediment transport in each small watershed, yielding multi‐year averages with low errors (1.54% for runoff, 7.1% for sediment). These findings provide valuable references for managing small watersheds in the Sanmenxia Reservoir area.

Abstract The Baubau watershed, located in Southeast Sulawesi, Indonesia, serves as a critical source of clean water for Baubau City. However, the watershed is increasingly threatened by land conversion, vegetation degradation, and elevated surface runoff. This study aims to evaluate the spatial distribution of erosion and sedimentation risks using the Soil and Water Assessment Tool (SWAT), implemented within the ArcSWAT platform. The model incorporates land use data (2024), soil characteristics, topographic features, and daily climate parameters from 2014 to 2024 collected from six meteorological stations. Simulation results indicate that the majority of the watershed area experiences very low to low erosion rates, attributed mainly to the dominance of mixed forest and wetland vegetation. In contrast, regions with steep slopes and minimal vegetation, particularly open or built-up areas, exhibit significantly higher erosion and sediment yields. These findings underscore the importance of targeted land conservation efforts in upstream regions to maintain hydrological balance and ensure the sustainability of clean water resources. This study confirms the utility of the SWAT model as a robust tool for hydrological simulation and offers a scientific foundation for policy development and sustainable watershed management in Baubau City.

ABSTRACT Understanding future hydroclimatic changes is essential for managing water resources in rapidly developing basins. This study used projections from six CMIP6 global climate models to explore four Shared Socioeconomic Pathways (SSP126, SSP245, SSP370, and SSP585) for China’s Qinhuai River Basin. The models were bias-corrected by quantile mapping and coupled with the Soil and Water Assessment Tool (SWAT) to project climate and runoff/streamflow. By 2100, temperature increases by 1.6°C under SSP126 to 5.6°C under SSP585, with the fastest warming rate (0.5°C per decade) in SSP585. Annual precipitation increases by 8.3–10.4% under SSP126/SSP245 and by about 10% under SSP370/SSP585 for 2081–2100. Multi-model ensemble projects increases in both precipitation and runoff; SSP370 shows the largest trend, with precipitation rising 2.5 mm/year and runoff 0.96 mm/year. Probability density analyses indicate the highest mean annual discharge (82–83 m3/s) occurs in the 2080s across all scenarios, while the largest variance appears in the 2060s under SSP126 (79.2) and in the 2080s under SSP245 (79.0). These patterns underscore complex interactions between climate change and socioeconomic pathways. The ensemble approach captures overall responses and helps reduce uncertainty. Further validation with long-term observations is encouraged to strengthen pathway-specific impact assessments and support integrated water-resources management.

ABSTRACT Water is an essential resource to preserve, yet it faces numerous pressures, including nitrate pollution from nitrogen inputs in agriculture. Models serve as valuable tools for analysing nitrate transfer and regulation processes within watersheds, helping to identify pollution sources. The coupling of the Soil and Water Assessment Tool (SWAT) with the drainage network biogeochemical model RIVE provides a comprehensive modelling approach called SWAT‐RIVE, which was previously tested on a section of the Garonne River (France). This study evaluates the ability of SWAT‐RIVE to represent hydrological and biogeochemical dynamics in the Vienne catchment (France). The objective of this paper is to evaluate and simulate hydro‐biogeochemical dynamics from 1993 to 2017, focusing on nitrate transfer and regulation at the watershed scale, including wetlands and epilithic biofilm interfaces. As the nitrogen cycle is interconnected with other elements, such as organic carbon, phosphorus and silica, influencing processes like denitrification and plant or algal growth, the SWAT‐RIVE representation of these elements was also assessed. Daily water and nitrate dynamics were well simulated at the catchment scale, with average NSE values of 0.45 and 0.15, R 2 values of 0.52 and 0.62 and KGE values of 0.65 and 0.39, respectively. Some other variables were accurately simulated at the outlet, particularly dissolved oxygen (NSE = 0.96, R 2 = 0.96, KGE = 0.89), dissolved silica (NSE = 0.85, R 2 = 0.93, KGE = 0.72) and dissolved organic carbon (NSE = 0.52, R 2 = 0.82, KGE = 0.50), confirming the possibility of using SWAT‐RIVE outputs to evaluate nitrate dynamics at the catchment scale. Despite several limitations, the coupling of SWAT and RIVE leads to a more precise quantification of biogeochemical processes on hillslopes and in the watercourse, making it possible to consider the use of SWAT‐RIVE in other watersheds.

ABSTRACT Historically, severe drought events, coupled with land use and land cover changes, have significantly influenced streamflow behaviour This study enhances the understanding of these hydrological processes by assessing streamflow responses to future climate and land use and land cover change in a transitional Brazilian basin between semiarid and humid tropical forest biomes. Projections from 10 global climate models available through the Climate Change Dataset for Brazil (CLIMBra) were utilised incorporating bias correction via the Quantile Mapping method. Future land use and land cover changes were simulated using the land change modeller (LCM), while hydrological projections were generated through the soil and water assessment tool (SWAT), which was calibrated and validated with satisfactory performance, achieving coefficients of determination ( R 2 ) and Nash‐Sutcliffe (NSE) efficiencies in the ranges of 0.62–0.79 and 0.61–0.76 for calibration and 0.48–0.90 and 0.41–0.84 for validation, respectively. The results indicate a substantial expansion of agricultural and pasture areas, with a 280% increase over recent decades. Climate projections under the SSP2–4.5 and SSP5–8.5 scenarios show a progressive temperature rise and declining rainfall trends, with the SSP5–8.5 scenario exhibiting a steeper increase in temperature. Paradoxically, hydrological modelling suggests an intensification of streamflow extremes, with peak discharges ranging from 200 to 300 m 3 /s, particularly, in regions prone to extreme precipitation events. Notably, under SSP5–8.5, a more pronounced rise in flood peaks is observed, indicating elevated flood risks, even in moderate emissions scenarios. These findings underscore the necessity for adaptive water resource management strategies to mitigate future hydrological vulnerabilities in the basin.

ABSTRACT Analysing the hydrological regime of watersheds and driving factors is important for understanding hydrological processes. This study used hydrological data from 1986 to 2018 from the Luoyugou (LYG) watershed to quantitatively analyse the hydrological regime and contribution rates of driving factors using the range of variability approach (RVA) and the Soil and Water Assessment Tool (SWAT) model, Budyko and two empirical methods. The results show that: (1) The degree of hydrological alteration (DHA) for most ecologically relevant hydrologic indicators (ERHIs) was 68.16% and 69.13%, indicating a high degree of alteration. The hydrological regime and river ecosystem have undergone significant changes. (2) Under the influence of human activities, the flood sediment load decreased by 66.74% and the flood sediment‐carrying capacity increased significantly. (3) Large‐scale human activities, such as land use/cover change (LUCC) and soil and water conservation measures (SWCMs), are the dominant factors causing changes in the hydrological regime. The SWAT model and double mass curve (DMC) method have higher accuracy of the contribution rate than Budyko and the linear regression model (LRM) method, as the latter two have limitations that may lead to an underestimation of contributions. This study can serve as a reference for assessing hydrological regimes and the impacts of ecological restoration in other regions worldwide.

Abstract The Segeri Watershed, particularly the Mangguliling Sub-watershed spanning Barru and Pangkajene Kepulauan Regencies in South Sulawesi, faces intense ecological pressure from land use change, agricultural expansion, tourism development, and illegal mining. This study assesses erosion hazard levels and develops targeted land conservation strategies through integrated biophysical and socio economic analyses. A spatial modeling approach was applied using the Soil and Water Assessment Tool (SWAT) with an 8 m DEM (DEMNAS), 2021 Sentinel-2 land cover data (validated at ≥85% accuracy), RePPProT soil maps, slope classes, and climate data from NASA POWER and local stations. Field surveys and community interviews provided complementary socio economic insights. SWAT simulations indicate an annual rainfall of 2,040 mm, potential evapotranspiration 2,298 mm, and actual evapotranspiration 1,006 mm, with subsurface Xlow emerging as the dominant hydrological pathway. Erosion hazard mapping shows that 42.6% of the area falls in the heavy category (III-B) and 29.6% in the very heavy category (IV-SB), with an average soil loss of 535 Mg/ha/year. Simulated sediment yield reaches a maximum of 611.49 Mg/ha/year in steep, grasslandcovered HRUs, indicating severe erosion hotspots. Most agricultural lands lie in high-risk zones, while community awareness of watershed conservation is limited and no integrated management program is in place. Conservation zoning, developed in accordance with Perdirjen P.8/PDASHL/SET/KUM.1/8/2018, identiXies four priority zones ranging from strict protection to controlled cultivation, supported by mechanical, vegetative, and micro-water management measures. The Xindings provide a scientiXically grounded, spatially explicit framework for sustainable watershed management, offering guidance for land-use planning and conservation policymaking, and fostering collaboration among local communities and stakeholders to safeguard the Mangguliling Sub-watershed ecological functions.

Assessing climate change impact on watershed hydrological processes and stream temperature by considering CO2 emissions.

Applications of SWAT (soil and water assessment tool) model: A systematic review

The pollution of nitrogen (N) and phosphorus (P) is also an acute environmental problem at the global level, which contributes to the intensive process of eutrophication of freshwater, estuarine, and coastal aquatic environments. The overloading of nutrients, which is mainly caused by nonpoint sources such as agricultural runoff, wastewater discharge, and atmospheric deposition, initiates harmful algal growth, lowers the transparency of the water, and decays the dissolved oxygen (hypoxia) and biodiversity. To solve this ubiquitous issue, it is necessary to change the traditional single nutrient management approach to source-to-sink control. This study introduces a combined model where it is essential to consider innovative methodologies of nutrient reduction, such as improving watershed management methodologies and new ecotechnologies. The framework focuses on an amalgamation of Best Management Practices (BMPs), including high-level fertilizer optimization, and state-of-the-art ecological design, including constructed wetlands blended with specific adsorption media. The Soil and Water Assessment Tool (SWAT) algorithm is included in the prediction model based on the transport of nutrients, and it allows for the best location of mitigation actions. The general Discussion of the different ways of addressing the problem reveals the need to implement site-specific solutions. The most important one is that a holistic management model that aims at achieving N and P reduction and the recovery of resources is proposed. This will help reestablish the ecological balance, improve the quality of water, and assist in maintaining the health of the aquatic ecosystems.

ABSTRACT Study region: Densu River Basin (DRB) in Ghana. Study focus: Urbanization, agricultural expansion, and population growth are transforming land use across West Africa, impacting hydrological regimes and ecosystem resilience. This study assesses land use and land cover (LULC) changes in Ghana’s Densu River Basin (DRB) using the Soil and Water Assessment Tool (SWAT). The model, calibrated (2012–2019) and validated (1990–2011) with streamflow data, showed strong performance (NSE = 0.79; R² = 0.85). Three LULC scenarios – agriculture-, forest-, and urban-dominated – were simulated using satellite imagery from 1990 to 2019. The 2018 Landsat baseline map, validated with 91.5% accuracy, showed the urban scenario increased surface runoff (>50% of water yield) and sediment load (>83,000 metric tons/year), while the forest scenario enhanced percolation and groundwater recharge. Sedimentation in the Weija Reservoir, a key water source for Greater Accra, deposits ∼47,500 m³ annually, posing long-term risks to reservoir capacity and water security. These findings highlight tropical catchment's vulnerability to land use changes and support scenario-based hydrological modelling for river basin planning in rapidly urbanizing, data-scarce sub-Saharan Africa.

Understanding streamflow dynamics in watersheds affected by human activity and climate variability is important for sustainable water and environmental resource management. This study evaluates the vulnerability of Alabama watersheds to anthropogenic and climatic changes using an integrated framework combining GIS, remote sensing, hydrological modeling, and machine learning (ML). Three Soil and Water Assessment Tool (SWAT) models, differing in spatial resolution and soil inputs, were developed to simulate streamflow under baseline and land-use/land cover (LULC) scenarios from 1990 to 2023. The model, built with consistent 100 × 100 m rasters and fine-resolution SSURGO (Soil Survey Geographic Database) soil data, achieved the best calibration and was selected for detailed analysis. Streamflow trends were assessed over two periods (1993–2009 and 2010–2023) to help isolate climate variability (from LULC effects), while LULC changes were evaluated using 1992, 2011, and 2021 maps. A Long Short-Term Memory (LSTM) model further enhanced simulation accuracy by integrating partially calibrated SWAT outputs. Watershed vulnerability was ranked using a multi-criteria framework. Two watersheds were classified as highly vulnerable, nine as moderately vulnerable, and three as having low vulnerability. Basin-level contrasts revealed moderate climate impacts in the Tombigbee Basin, greater climate sensitivity in the Black Warrior Basin, and LULC-dominated impacts in the Alabama Basin. Overall, LULC change exerted stronger and more spatially variable effects on streamflow than climate variability. This study introduces a transferable SWAT–ML vulnerability ranking framework to guide watershed and environmental management in data-scarce, human-modified regions.

Assessing the impacts of land cover change (LCC) on hydrology and sediment load is essential for the sustainable management of urban watersheds. Modeling LCC using machine learning techniques enhances the ability to generate realistic future scenarios, providing a robust basis for informed watershed management decisions. This study projects future LCC and evaluates its effects on hydrological processes and sediment load in the Alibeyköy Watershed, one of the key water sources for Istanbul, Türkiye. Future LCC scenarios were generated using the Multilayer Perceptron-Markov Chain (MLP-MC) approach. The Soil and Water Assessment Tool (SWAT) was subsequently used to simulate the hydrological process and sediment transport in the watershed for the baseline year and projected future scenarios. The SWAT model was calibrated and validated for streamflow and sediment loads using continuous data. In addition, field-measured sediment load data collected were used for further validation. The main findings of this study are as follows: (1) built-up areas are projected to expand substantially over the coming decades, while natural land covers, including forests and rangelands, are expected to decline markedly; (2) this urban expansion is associated with increased surface runoff; and (3) a notable rise in dead storage volume. While the increase is notable, it represents only a small fraction of reservoir storage; however, its cumulative effect justifies continued monitoring and management attention. This study combines machine learning-based land cover change, hydrological modeling, and the use of field-based monitoring data to develop an integrated approach for watershed analysis. The resulting framework enables reliable prediction of hydrological components and sediment loads, supporting the design of effective and evidence-based watershed management strategies.

Amid Nepal’s expanding hydropower sector, the Sunkoshi Multipurpose Scheme stands as a pivotal inter-basin transfer project. As the country seeks to maximize its abundant water resources, strengthening hydropower resilience against the inevitable impacts of climate change is imperative for ensuring long-term energy sustainability. This study conducts a comprehensive assessment of climate change impacts on the Sunkoshi River Basin and proposes an adaptive management strategy through dynamic rule curve modifications, optimizing reservoir operations in response to projected shifts in water availability across different time horizons of the 21st century. Seven bias-corrected General Circulation Models (GCMs) (ACCESS-CM2, BCC-CSM2-MR, CanESM5, EC-Earth3, MPI-ESM1-2-HR, MPI-ESM1-2-LR, and NorESM2-MM) were adopted for the projection of climate variables under Shared Socio-Economic Pathways (SSP)245 and SSP585 scenarios which were further utilized for the projection of future discharge in the Soil and Water Assessment Tool (SWAT). The anticipated inflow data served as input to the Hydraulic Engineering Center- Reservoir System Simulation (HEC-ResSim) software to simulate the reservoir operation and propose modified rule curves for Sunkoshi No.1, Sunkoshi No.2, Sunkoshi No.3, and Dudhkoshi hydropower projects for the time frame of 2030s, 2060s, and 2080s. Six different rule curves were proposed and average yearly energy generations were maximized ranging from 25.5%, 61.07%, 71.26%, and 10.50% for Sunkoshi No.3, Sunkoshi No.2, Sunkoshi No.1, and Dudhkoshi power plants respectively. These results could be helpful for long-term planning, urging policymakers to integrate dynamic rule curve modifications in the broader context of sustainable energy production and climate change adaptation.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-19252-8.