Publications

@article{PATHAK2026181239,
title = {Integrated multi-hazard assessment for climate-resilient watershed management: A transferable prioritization framework from Nepal's Mid-Hills},
author = {Lalit Pathak and Kamana Joshi and Pradeep Ghimire and Rabin Malla},
url = {https://www.sciencedirect.com/science/article/pii/S0048969725028815},
doi = {https://doi.org/10.1016/j.scitotenv.2025.181239},
issn = {0048-9697},
year = {2026},
date = {2026-01-01},
journal = {Science of The Total Environment},
volume = {1012},
pages = {181239},
abstract = {Mountain watersheds face cascading multi-hazard risks that require integrated assessment frameworks to guide effective adaptation options, yet current approaches focus on single hazards or lack spatial precision for local-scale planning. Himalayan mid-hill watersheds exemplify this challenge, facing escalating hazards and soil erosion amid rapid land use transitions. This study integrated Maximum Entropy (MaxEnt) modeling, the Revised Universal Soil Loss Equation (RUSLE), and multi-temporal land use analysis within a spatially explicit Multi-Criteria Decision Analysis (MCDA) for watershed prioritization. Applied to the 703.36 km2 Marin watershed in central Nepal, this framework integrated multi-source datasets including 30 m SRTM DEM, ICIMOD LULC data (2000−2022), CHIRPS rainfall, ERA5 temperature, TerraClimate drought indices, and field-verified hazard inventories. MaxEnt models achieved robust performance for landslide (AUC = 0.828 ± 0.018) and fire susceptibility (AUC = 0.798 ± 0.025) mapping. Results revealed that 27.3 % of the watershed exhibits high/very high landslide susceptibility, 24.4 % faces elevated fire risk, 18.3 % lie in flood-prone zones, and 32.9 % severe erosion (>20 t/ha/yr).MCDA integration classified 24 % of the watershed requiring critical intervention, due to steep slopes (>30°), fragile lithology, and intensive land use transitions. Sensitivity analysis confirmed framework robustness with rank correlations r = 0.96–0.98 across ±20 % weight variations, while Monte Carlo simulation (n = 1000) provided uncertainty bounds for prioritization decisions (σ = 0.03–0.06). This transferable methodology addresses critical gaps in mountain watershed management by providing a scalable prioritization framework applicable to similar physiographic regions globally.},
keywords = {Climate adaptation, MaxEnt modeling, Mountain hydrology, Multi-hazard assessment, RUSLE, Watershed prioritization},
pubstate = {published},
tppubtype = {article}
}