We share datasets, trained models and analysis code to support open, reproducible hydrology.
Catchment Attributes and Meteorology for Large-sample Studies – India. Daily meteorological time series, observed and LSTM-predicted streamflow, and static catchment attributes for 472 catchments in Peninsular India — designed to foster large-sample hydrological studies and bring Indian catchments into global research.
Catchment Attributes and MEteorology for Large-sample Studies – India – Projected Streamflow. Provides meteorological forcings and projected daily streamflow for the same 472 Peninsular Indian catchments as CAMELS-IND, under future climate scenarios — extending large-sample hydrological research in India into a changing-climate context.
Uses the identical Peninsular India catchment set as CAMELS-IND, keeping the two datasets directly comparable.
Projected streamflow simulated across a diverse multi-model ensemble of General Circulation Models.
Covers SSP126, SSP245, SSP370 and SSP585 — a full range of shared socioeconomic pathways.
Daily projected streamflow (cumecs) for 2015–2100 under each scenario, plus historical-period simulations for each GCM.
Precipitation, maximum temperature and minimum temperature, provided for both historical and future periods under each GCM and SSP scenario.
All streamflow and forcing variables are provided at daily resolution, matching the CAMELS-IND standard.
Sharing our model codes, data, results and analysis notebooks.
A Novel Multi-Model Ensemble Framework for Fluvial Flood Inundation Mapping. This code contains machine learning-based multi-model ensemble framework for fluvial flood inundation mapping.
Differentiable Parameter Learning (dPL) + HBV Hydrologic Model with Reservoir Module. In this version, we have extended the model by incorporating a reservoir module, enabling simulation of regulated catchments. This enhancement allows the model to better represent flow dynamics influenced by reservoir operations.
PIDuaL is an extension of the Mass-Conserving Long Short-Term Memory (MC-LSTM) model for hydrological modeling. It enables the simultaneous prediction of streamflow and evapotranspiration through a modified water-balance structure.
This dataset accompanies the manuscript “Benchmarking and Selecting Optimal Hydrological Models for Large-Sample Applications Considering Complexity and Uncertainty” (Mangukiya et al., under review, Water Resources Research). It provides simulated streamflow outputs and evaluation metrics for 47 conceptual rainfall–runoff models applied across 159 watersheds in Peninsular India.
Monthly, Seasonal and Yearly Net Primary Productivity (NPP) data of India from 2003-2020 modelled using the CASA model.
