Subrahmanian et al. (2025) A Novel Approach To Integrate Low Impact Development (LID) Modules into the SWAT Model To Facilitate Sustainable Urban Drainage Planning at River Basin Scales

Identification

• Journal: Water Resources Management
• Year: 2025
• Date: 2025-12-26
• Authors: Sreethu Subrahmanian, Arun Rajasekaran Sankarbalaji, Elanchezhiyan Duraisekaran, S. Murty Bhallamudi, Balaji Narasimhan
• DOI: 10.1007/s11269-025-04363-8

Research Groups

• Indian Institute of Technology Madras, Chennai, Tamil Nadu, India

Short Summary

This study integrates novel physics-based Multi-Layer Green-Ampt (MLGA) infiltration subroutines for Low Impact Development (LID) measures into the Soil Water Assessment Tool (SWAT) model, enhancing its capability for sustainable urban drainage planning and flood mitigation at large river basin scales. The enhanced model demonstrates improved hydrological simulation accuracy and significant reductions in surface runoff and increases in aquifer recharge under various LID implementation scenarios.

Objective

• To incorporate a new infiltration subroutine based on the physics-based Multi-Layer Green-Ampt (MLGA) approach into the land phase simulation module of the Soil Water Assessment Tool (SWAT) model.
• To add new subroutines for bio-retention cells, rain gardens, roof gardens, permeable pavements, and infiltration trenches, based on the MLGA infiltration method, into the enhanced SWAT model.

Study Configuration

• Spatial Scale: Adyar River basin, Chennai, India (830 km²).
• Temporal Scale: Simulation period from 2001 to 2019 (19 years) with a 2-year warm-up period, using a 30-minute sub-hourly time step.

Methodology and Data

• Models used:
  • Soil Water Assessment Tool (SWAT)
  • Multi-Layer Green-Ampt (MLGA) infiltration method (integrated into SWAT)
  • Green-Ampt Mein Larson (GAML) method (for comparison)
  • SWAT-CUP with Sequential Uncertainty Fitting (SUFI 2) algorithm for calibration.
  • Extreme Value Type I probability distribution for flood frequency analysis.
  • Microsoft Visual Studio 2022 for integration and debugging.
• Data sources:
  • SRTM Digital Elevation Model (DEM) of 30 m x 30 m resolution.
  • Soil data (1:50,000 scale) from Tamil Nadu Agricultural University (TNAU 2016).
  • Land use/land cover data (56 m resolution) from the National Remote Sensing Centre (NRSC 2023).
  • Daily temperature data from the Indian Meteorological Department (IMD).
  • Sub-hourly precipitation data obtained by temporally disaggregating rainfall data from seven non-recording rain gauge stations using Global Precipitation Mission data.
  • Observed inflow data for Chembarambakkam reservoir (daily time step) for calibration and validation.
  • Waterbodies datasets (storage capacities and surface areas) from the Water Resources Department of Tamil Nadu.

Main Results

• The integrated MLGA-based SWAT model demonstrated good performance for the study area, achieving R² and NSE values of 0.77 and 0.69, respectively, for monthly inflow simulations to the Chembarambakkam reservoir. For daily simulations, R² was 0.595 and NSE was 0.585. These performance indices were higher than those obtained using the existing GAML method.
• Evaluation of LID implementation scenarios at the river basin level showed a significant reduction in surface runoff (up to 16.9% for the highest implementation scenario, SC4) and a corresponding increase in aquifer recharge and lateral flow.
• LID scenarios were most effective at reducing flood peaks for shorter return periods (e.g., 2 years) and in heavily urbanized subbasins, with peak runoff reduction reaching 27.2% for a 2-year return period flood in a heavily urbanized subbasin.
• The effectiveness of LIDs in reducing flood peaks decreased with increasing flood return period, showing reductions in the range of 2–4% for a 100-year return period at the basin level.
• Increasing the percentage of runoff routed to LIDs and expanding LID implementation areas led to greater overall runoff reduction and aquifer recharge.

Contributions

• Development and integration of novel physics-based Multi-Layer Green-Ampt (MLGA) infiltration subroutines for five specific LID measures (bio-retention cells, rain gardens, roof gardens, infiltration trenches, and permeable pavements) into the SWAT model.
• Enhancement of SWAT's capability to simulate vertical heterogeneity in soil profiles within LIDs and non-LID portions, addressing a critical limitation of existing single-layered infiltration models in SWAT.
• Demonstration of the applicability and effectiveness of the enhanced SWAT model for large-scale LID impact assessment and sustainable urban drainage planning at river basin scales, a domain previously limited by available hydrological tools.
• Provides a computationally efficient, physics-based tool for comprehensive LID impact assessment in spatially heterogeneous river basins, facilitating improved planning and design guidelines for sustainable urban water management.

Funding

• Prime Minister’s Research Fellowship (Project Number: SB22230248CEPMRF008310)
• SUTRAM for Easy Water (Project Number: DST/TM/WTI/WIC/2K17/82(G))

Citation

@article{Subrahmanian2025Novel,
  author = {Subrahmanian, Sreethu and Sankarbalaji, Arun Rajasekaran and Duraisekaran, Elanchezhiyan and Bhallamudi, S. Murty and Narasimhan, Balaji},
  title = {A Novel Approach To Integrate Low Impact Development (LID) Modules into the SWAT Model To Facilitate Sustainable Urban Drainage Planning at River Basin Scales},
  journal = {Water Resources Management},
  year = {2025},
  doi = {10.1007/s11269-025-04363-8},
  url = {https://doi.org/10.1007/s11269-025-04363-8}
}