Scholarlink

<p>Understanding the combined influence of land use/land cover (LULC) change and climate variability on watershed hydrology is critical for sustainable water resource management in rapidly urbanizing semi-arid regions. In the Kunderu watershed, India, historical LULC analysis indicates rapid urban expansion from approximately 25&#xa0;km² in 2013 to 113&#xa0;km² in 2022, accompanied by increases in agricultural land and concurrent reductions in forest cover and surface water bodies. An integrated SWAT–CA–Markov–CMIP6 modelling framework was employed to quantify the hydrological implications of these changes under historical and future conditions. The Soil and Water Assessment Tool (SWAT) model achieved good calibration and validation performance (Nash–Sutcliffe Efficiency (NSE) ≥ 0.70; Coefficient of determination (R²) ≥ 0.70), indicating reliable streamflow simulation and a runoff-limited hydrological regime dominated by infiltration and groundwater contributions. Future LULC projections for 2050, generated using machine-learning (ML) classifiers, integrated with a Cellular Automata-Markov (CA–Markov) approach, indicate continued urban expansion, with built-up area increasing to approximately 223&#xa0;km², alongside further losses of ecologically sensitive land classes. Climate projections from bias-corrected Coupled Model Intercomparison Project Phase 6 (CMIP6) models (Max Planck Institute Earth System Model version 1 (MPI-ESM1-2-HR) and Australian Community Climate and Earth System Simulator - Coupled Model 2 (ACCESS-CM2) under Shared Socioeconomic Pathway 1-2.6 (SSP1-2.6), SSP2-4.5, and SSP5-8.5 scenarios indicate increased rainfall variability and pronounced seasonal warming. Projected LULC and climate scenarios indicate higher surface runoff and evapotranspiration, along with lower groundwater recharge and return flow, reflecting a shift from infiltration-dominated to surface-driven hydrological behavior. These results emphasize the combined influence of land-use change and climate variability on watershed hydrology and the need for integrated adaptive management to support long-term resilience.</p> Graphical Abstract <p>Topographic and ancillary datasets (Digital Elevation Model (DEM), slope, soil, and road proximity) together with observed climate data were used to configure and calibrate the SWAT model for the Kunderu watershed, India. Multi-temporal LULC maps for 2013, 2018, and 2022 are analysed using ML classifiers (Support Vector Machine (SVM), Decision Forest (DF), and Multi-Layer Perceptron (MLP)) in combination with a CA–Markov framework to derive transition probabilities and generate LULC projections for 2050. Future climate forcings from CMIP6 models (MPI-ESM1-2-HR and ACCESS-CM2) under SSP1-2.6, SSP2-4.5, and SSP5-8.5 are bias-corrected using Quantile Mapping (QM) before integration with the projected LULC in the SWAT framework to simulate future hydrological components, including surface runoff, evapotranspiration, return flow, and groundwater recharge.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrated Assessment of Land use and Climate Change Impacts on Hydrology: A Scenario-Based SWAT Study of the Kunderu Basin, India

  • 期刊论文

  • 发表日期：

    2026年7月1日

查看全文
  • G M Sayanth,
  • Aneesh Mathew,
  • Chinthu Naresh

摘要

Understanding the combined influence of land use/land cover (LULC) change and climate variability on watershed hydrology is critical for sustainable water resource management in rapidly urbanizing semi-arid regions. In the Kunderu watershed, India, historical LULC analysis indicates rapid urban expansion from approximately 25 km² in 2013 to 113 km² in 2022, accompanied by increases in agricultural land and concurrent reductions in forest cover and surface water bodies. An integrated SWAT–CA–Markov–CMIP6 modelling framework was employed to quantify the hydrological implications of these changes under historical and future conditions. The Soil and Water Assessment Tool (SWAT) model achieved good calibration and validation performance (Nash–Sutcliffe Efficiency (NSE) ≥ 0.70; Coefficient of determination (R²) ≥ 0.70), indicating reliable streamflow simulation and a runoff-limited hydrological regime dominated by infiltration and groundwater contributions. Future LULC projections for 2050, generated using machine-learning (ML) classifiers, integrated with a Cellular Automata-Markov (CA–Markov) approach, indicate continued urban expansion, with built-up area increasing to approximately 223 km², alongside further losses of ecologically sensitive land classes. Climate projections from bias-corrected Coupled Model Intercomparison Project Phase 6 (CMIP6) models (Max Planck Institute Earth System Model version 1 (MPI-ESM1-2-HR) and Australian Community Climate and Earth System Simulator - Coupled Model 2 (ACCESS-CM2) under Shared Socioeconomic Pathway 1-2.6 (SSP1-2.6), SSP2-4.5, and SSP5-8.5 scenarios indicate increased rainfall variability and pronounced seasonal warming. Projected LULC and climate scenarios indicate higher surface runoff and evapotranspiration, along with lower groundwater recharge and return flow, reflecting a shift from infiltration-dominated to surface-driven hydrological behavior. These results emphasize the combined influence of land-use change and climate variability on watershed hydrology and the need for integrated adaptive management to support long-term resilience.

Graphical Abstract

Topographic and ancillary datasets (Digital Elevation Model (DEM), slope, soil, and road proximity) together with observed climate data were used to configure and calibrate the SWAT model for the Kunderu watershed, India. Multi-temporal LULC maps for 2013, 2018, and 2022 are analysed using ML classifiers (Support Vector Machine (SVM), Decision Forest (DF), and Multi-Layer Perceptron (MLP)) in combination with a CA–Markov framework to derive transition probabilities and generate LULC projections for 2050. Future climate forcings from CMIP6 models (MPI-ESM1-2-HR and ACCESS-CM2) under SSP1-2.6, SSP2-4.5, and SSP5-8.5 are bias-corrected using Quantile Mapping (QM) before integration with the projected LULC in the SWAT framework to simulate future hydrological components, including surface runoff, evapotranspiration, return flow, and groundwater recharge.