IIT Bombay KrishiDrought Portal Forecasts Drought 4 Weeks Ahead

IIT Bombay KrishiDrought gives free weekly, district-level agricultural drought forecasts for India up to four weeks ahead, using satellite data and AI.

The Indian Institute of Technology Bombay (IIT Bombay) has launched KrishiDrought.in, a free public portal that forecasts agricultural drought across India up to four weeks in advance. Formally launched on 28 October 2026, the platform combines satellite data, hydroclimatic datasets and deep learning models to issue weekly, district-level soil-moisture forecasts at roughly 9 km resolution.

The portal was built by the Ecohydrology research group led by Prof. Karthikeyan Lanka at IIT Bombay’s Centre of Studies in Resources Engineering (CSRE), working with the institute’s Centre for Climate Studies. Prof. Lanka, whose agricultural drought research earned him the NASI Young Scientist Award 2026, has now turned that work into a tool any farmer, official or student can open in a browser.

How the IIT Bombay KrishiDrought portal works

Instead of describing a drought after crops have already suffered, KrishiDrought looks ahead. The system tracks how soil moisture is changing and estimates where it is likely to fall below normal over the coming weeks. According to IIT Bombay, it brings together four kinds of input:

Satellite observations of land surface and soil moisture
Hydroclimatic datasets on rainfall, temperature and water balance
Dynamical weather and climate forecasts
Deep learning models that combine these signals into a forward-looking forecast

Users see a colour-coded map of India. Red shades flag districts heading towards drier-than-normal soil, while green shades mark wetter-than-normal conditions. Forecasts can be viewed for one, two, three or four weeks ahead, filtered by state and district, and are refreshed every Monday.

A campus research group shipping a national public tool

The launch was timed with the Golden Jubilee celebrations of CSRE, and IIT Bombay credited the students and researchers in the group for building the platform. For engineering and science students, it is a working example of research that goes beyond a journal paper: the same skills in remote sensing, data engineering and machine learning are being used to run a weekly operational service.

It also fits a pattern of IIT Bombay taking AI out of the lab and into public use, from the BharatGen initiative on Indian-language generative AI to its e-PG Diploma in Computer Science and AI for working professionals. Elsewhere on Indian campuses, students are applying similar satellite and AI methods to other problems, such as the IIT Kharagpur concept that helps satellites avoid space debris.

Why early drought warning matters

Agricultural drought sets in when soil moisture cannot meet crop water needs, which can happen even when rainfall figures look close to normal. A warning of a few weeks gives farmers time to plan irrigation and protect the most vulnerable crops, and gives state agriculture departments and district administrations time to target extension services and water-conservation work.

The forecasts are district-level estimates, not field-by-field predictions, and they work best alongside local weather updates and advice from agricultural extension officers. IIT Bombay has not yet published validation statistics for the forecasts.

Key takeaways

IIT Bombay launched KrishiDrought.in on 28 October 2026 during CSRE’s Golden Jubilee celebrations.
The free portal gives weekly agricultural drought forecasts across India up to four weeks ahead, at about 9 km resolution and district level.
It was built by Prof. Karthikeyan Lanka’s Ecohydrology group with the Centre for Climate Studies, using satellite data and deep learning.

FAQ

What is KrishiDrought.in?

It is a public web portal from IIT Bombay that forecasts agricultural drought, measured through soil moisture, for districts across India up to four weeks in advance.