Sebastian et al. (2025) Incorporating varying vegetation characteristics driven by Hydrometeorology in the land surface modeling by variable Infiltration Capacity model

Identification

• Journal: Journal of Hydrology
• Year: 2025
• Date: 2025-11-07
• Authors: Dawn Emil Sebastian, Subimal Ghosh
• DOI: 10.1016/j.jhydrol.2025.134580

Research Groups

• Department of Civil Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India
• School of Environment and Sustainability, Indian Institute for Human Settlements, Bengaluru, Karnataka 560080, India
• Interdisciplinary Program in Climate Studies, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India

Short Summary

This study demonstrates the critical role of dynamic vegetation in hydrological modeling, particularly for evapotranspiration in India, by integrating a machine learning model (LSTM) to simulate vegetation variability within the Variable Infiltration Capacity (VIC) model, revealing an 18% increase in annual evapotranspiration compared to static vegetation approaches.

Objective

• To demonstrate the pivotal role of vegetation variability in influencing evapotranspiration, particularly during the Indian summer monsoon and post-monsoon seasons.
• To develop and integrate an efficient grid-scale Machine Learning model (LSTM) into the Variable Infiltration Capacity (VIC) model to account for dynamic vegetation characteristics (fraction of vegetation cover, Leaf Area Index, and albedo) driven by hydrometeorological variables.
• To simulate future hydrometeorological scenarios in India using this novel approach and assess the impact of changing vegetation properties on future evapotranspiration fluxes.

Study Configuration

• Spatial Scale: India (grid-scale)
• Temporal Scale: Indian summer monsoon season, post-monsoon period, and future hydrometeorological scenarios.

Methodology and Data

• Models used:
  • Variable Infiltration Capacity (VIC) model (hydrological model)
  • Long Short-Term Memory (LSTM) model (machine learning model for vegetation dynamics)
  • MIROC6 General Circulation Model (GCM) (for future projections)
• Data sources:
  • Hydrometeorological variables (precipitation, temperature)
  • NEX-GDDP-CMIP6 dataset projections (downscaled from MIROC6 GCM)
  • Vegetation properties: fraction of vegetation cover, Leaf Area Index, and albedo (derived from LSTM)

Main Results

• Incorporating spatial and temporal vegetation variability into the VIC model resulted in an 18% increase in total annual evapotranspiration compared to conventional simulations using static vegetation parameters.
• Vegetation variability significantly influences evapotranspiration, particularly during the latter half of the Indian summer monsoon season and the subsequent post-monsoon period, which are characterized by high vegetation activity.
• The developed LSTM model successfully represents vegetation changes (fraction of vegetation cover, Leaf Area Index, and albedo) as a function of hydrometeorological variables.
• Changing vegetation properties have a significant impact on future evapotranspiration fluxes, a factor often overlooked when employing constant vegetation parameters in climate change impact assessments.

Contributions

• Demonstrated the critical importance of incorporating dynamic vegetation characteristics in hydrological models for accurate evapotranspiration estimation, especially in the context of climate change impact assessments.
• Developed a novel and efficient framework for integrating machine learning (LSTM) to simulate grid-scale vegetation variability (fraction of vegetation cover, Leaf Area Index, albedo) into a physics-based hydrological model (VIC).
• Addressed the limitations of prevailing methodologies that rely on static vegetation parameters, offering a computationally practical alternative to complex land surface models within earth system models.
• Provided an advanced framework for simulating future hydrometeorological scenarios that accounts for evolving vegetation properties, highlighting their significant impact on future evapotranspiration fluxes.

Funding

• Funding information is not explicitly provided in the abstract or introduction of the paper.

Citation

@article{Sebastian2025Incorporating,
  author = {Sebastian, Dawn Emil and Ghosh, Subimal},
  title = {Incorporating varying vegetation characteristics driven by Hydrometeorology in the land surface modeling by variable Infiltration Capacity model},
  journal = {Journal of Hydrology},
  year = {2025},
  doi = {10.1016/j.jhydrol.2025.134580},
  url = {https://doi.org/10.1016/j.jhydrol.2025.134580}
}

Generated by BiblioAssistant using gemini-2.5-flash (Google API)