Dehradun: At a time when cloudbursts, flash floods and erratic rainfall are becoming increasingly frequent across the Himalayas, researchers at the Indian Institute of Technology (IIT) Roorkee have developed a high-resolution climate projection dataset that could improve how future climate risks are assessed across India, particularly in ecologically fragile mountain states like Uttarakhand.The dataset, named INDRA-CMIP6, has been developed by researchers from the institute’s department of hydrology to address a long-standing limitation in conventional global climate models — their inability to accurately capture highly localised weather patterns in complex terrains such as the Himalayas.The study, recently published in Nature, noted that most global climate models are designed to project broad regional trends and often smooth out sharp variations in rainfall and temperature that occur over short distances in mountainous regions. As a result, localised extreme events such as cloudbursts or intense rainfall in a single valley may not be adequately reflected in wider forecasts.The research was carried out by Joyjit Mandal, Divya Sardana, Akash Singh Raghuvanshi and Ankit Agarwal of IIT Roorkee. The dataset focuses on projections of daily rainfall as well as maximum and minimum temperatures across the Indian subcontinent.“The model attempts to bridge this critical gap,” said Ankit Agarwal, associate professor at IIT Roorkee and one of the researchers involved in the study. “India is already witnessing rising temperatures, shifting rainfall patterns, urban flooding, heat stress and increasing pressure on water resources. National-scale or coarse global projections are often insufficient for local-level planning, especially in disaster-prone Himalayan regions,” he added.Explaining the significance of the dataset, Agarwal said India is already witnessing rising temperatures, shifting rainfall patterns, more frequent heat stress, intense rainfall events, urban flooding, agricultural risks and growing pressure on water resources. “For decision-makers, however, national-scale or coarse global projections are often not enough. Adaptation planning requires climate information at scales relevant to districts, river basins, cities, farms, infrastructure corridors and ecosystems,” he added.According to the researchers, INDRA-CMIP6 provides rainfall and temperature projections at a much finer spatial scale than many existing global models. It draws on historical climate records from 1950 to 2014 and includes future projections up to 2100 under multiple greenhouse gas emission scenarios.Researchers said the improved resolution could help authorities better assess climate risks related to infrastructure, hydrology, water availability and disasters. “In Himalayan states like Uttarakhand, where weather conditions can change dramatically within a few kilometres, such local-scale projections are considered crucial for planning roads, dams, urban expansion and disaster mitigation strategies,” they added.The study also noted that existing coarse-resolution projections often struggle in mountainous terrain because valleys, ridges and elevation changes strongly influence local weather systems. Better representation of these features, the researchers said, could improve understanding of future extreme rainfall events and temperature shifts.The dataset has been developed using advanced downscaling and bias-correction techniques applied to CMIP6 global climate model outputs. “It could also support future studies on monsoon variability, glacier-fed river systems and climate adaptation planning across the Indian subcontinent,” researchers said.