Touseef et al. (2025) Hydrological modeling of karst terrain in southwestern Pakistan using modified SWAT

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2025
• Date: 2025-12-17
• Authors: Muhammad Touseef, Lihua Chen
• DOI: 10.1016/j.ejrh.2025.103055

Research Groups

• College of Civil Engineering and Architecture, Guangxi University, Nanning, China
• Key Laboratory of Disaster Prevention and Structural Safety of Ministry of Education, Nanning, China
• Guangxi Key Laboratory of Disaster Prevention and Engineering Safety, Nanning, China

Short Summary

This study developed and evaluated a modified Soil and Water Assessment Tool (SWAT) model incorporating a novel three-zonal conceptualization (epikarst, vadose, and phreatic zones) to improve hydrological modeling in the data-scarce karst Nari River Basin, southwestern Pakistan. The enhanced model demonstrated approximately 10 % higher Nash-Sutcliffe Efficiency (NSE) during calibration and validation, better capturing flood peaks and low-flow events compared to the original SWAT model.

Objective

• To develop and evaluate a modified SWAT model that explicitly incorporates a three-zonal conceptualization—epikarst, vadose, and phreatic zones—to improve the simulation of recharge, storage, and discharge dynamics in karst terrains.
• To hypothesize that representing these hydrological compartments separately within the model structure will lead to significant improvements in the accuracy of streamflow and groundwater predictions.

Study Configuration

• Spatial Scale: Nari River Basin, Baluchistan, southwestern Pakistan, covering an area of 6.98 x 10^10 square meters.
• Temporal Scale:
  • Data analysis period: 1979–2023 (various datasets, e.g., precipitation anomalies 1981–2023, temperature anomalies 1980–2023, discharge simulation 1979–2021).
  • Model warm-up period: 1998–2000.
  • Model calibration period: 2000–2006 (dry years).
  • Model validation period: 2006–2010 (wet years).

Methodology and Data

• Models used:
  • Modified Soil and Water Assessment Tool (SWAT) model, a semi-distributed, physically-based basin-scale model.
  • The modification integrates a three-zonal conceptualization of the karst system: epikarst, vadose, and phreatic zones, replacing the original single linear reservoir groundwater model.
  • Numerical modeling approach uses additional storage and discharge equations, zone-specific recharge coefficients, recession constants, and new delay parameters (e.g., gwdelay, karstdelay).
• Data sources:
  • Digital Elevation Model (DEM): SRTM-DEM with 30 meter resolution from CGIAR-CSI.
  • Land Cover: USGS Land-Cover Institute (LCI) MCD12Q1 (15-arc-second precision) for 2001–2010.
  • Soil Data: FAO–UNESCO world soil map (5 kilometer resolution).
  • Weather Data (precipitation and temperature): Pakistan Meteorological Department (PMD) from 14 stations in Balochistan (1981–2023 for rainfall, 1980–2023 for temperature). Missing data filled using Inverse Distance Weighting (IDW).
  • Normalized Difference Vegetation Index (NDVI): Derived from Landsat 8 SR remote sensing data for 2015 and 2020.

Main Results

• A modified SWAT model was developed, integrating a novel three-zonal conceptualization of the karst system (epikarst, vadose, and phreatic zones) into its hydrological structure.
• The enhanced model demonstrated approximately 10 % improvement in Nash-Sutcliffe Efficiency (NSE) during calibration (NSE = 0.922) and validation (NSE = 0.720) compared to the original SWAT model.
• The modified model better captured flood peaks and low-flow events under arid conditions, showing a more pronounced response to changing conditions.
• Parameter sensitivity improved significantly, with the CN2 t-value increasing by 50 % and the p-value decreasing by 80 %.
• The modified model typically forecasts increased discharge values during flood occurrences (e.g., 1994 and 2016) compared to the original model, suggesting improved simulation of extreme hydrological events.

Contributions

• Development of a novel modified SWAT model that explicitly incorporates a three-zonal conceptualization (epikarst, vadose, and phreatic zones) for improved simulation of karst hydrology.
• Significant enhancement of hydrological modeling performance (e.g., ~10 % higher NSE, better capture of extreme events, improved parameter sensitivity) in data-scarce, semi-arid karst environments.
• Provides a more physically representative modeling framework for karst basins, addressing limitations of traditional models in complex karst systems.
• Offers a transferable methodology for improving groundwater recharge and discharge simulation under climate change scenarios, particularly in regions with limited hydrological data.
• Fills a key methodological gap in applying SWAT to complex karst terrains, specifically within the Nari River Basin context, thereby improving predictive reliability for water resource management.

Funding

• State Key Program of National Natural Science of China (52439002)
• Guangxi Science and Technology Development (2025GXNSFGDA02850009)
• National Natural Science Foundation of China (52179010, 52069002, 52209010)
• Major Science and Technology Projects of the Ministry of Water Resources of China (SKR-2022038)
• China Southern Power Grid Company Limited (GXKJXM20240127)
• Innovation Project of Guangxi Graduate Education (YCBZ2023002)

Citation

@article{Touseef2025Hydrological,
  author = {Touseef, Muhammad and Chen, Lihua},
  title = {Hydrological modeling of karst terrain in southwestern Pakistan using modified SWAT},
  journal = {Journal of Hydrology Regional Studies},
  year = {2025},
  doi = {10.1016/j.ejrh.2025.103055},
  url = {https://doi.org/10.1016/j.ejrh.2025.103055}
}