Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26071Proceed with cautionacceptance 2/5

AI/ML-Based Integrated heavy rainfall Early Warning and Inundation Prediction System using Satellite, Radar, observational Weather and numerical weather prediction model data.

Ministry of Earth Sciences (MoES) · Disaster Management · Software

Two research problems in one sentence, and the inundation half needs metre-scale terrain and drainage data that simply is not public in India — if you take it, pick one catchment where you can get real elevation data and hindcast a single documented event honestly.

What it actually is

Warning that heavy rain is coming is only half of what a city needs; what people actually want to know is which streets will go under water. The ask joins the two together — forecast the rainfall from satellite, radar and model data, then translate that rainfall into where the flooding will actually be.

What to build

Two chained components: a heavy rainfall warning stage blending satellite precipitation estimates, radar reflectivity where available and numerical model output into a probabilistic rainfall forecast over a defined area and lead time, calibrated against gauge observations; and an inundation stage translating that forecast rainfall into flood extent over a terrain model with a drainage network and infiltration assumptions, producing a map of which areas flood to what depth and when — the whole chain demonstrated by hindcasting a real past flood event and comparing your predicted extent against the flooding that was actually observed.

Smallest thing that wins the room

Hindcast a well-documented past urban flood: feed in the rainfall that actually fell, produce the inundation map, and overlay it against satellite-observed flood extent from that day.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Moderate140–330 teams expectedroughly 1 in 119–275 wins it

Quieter than 44% of the 226 · #126 of 226 by expected field

A normal-sized field. Your idea has to be good, not miraculous.

Why: central ministry statements sat below the average.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • GPM IMERG satellite precipitation blending
  • gridded gauge calibration of forecast rainfall
  • LISFLOOD-FP or HEC-RAS 2D hydraulic modelling
  • SRTM / Cartosat DEM hydrological conditioning
  • Sentinel-1 SAR flood extent for validation
  • probabilistic rainfall threshold warning logic
  • Hydrometeorology
  • Flood modelling
  • Disaster early warning

Prior art to read before you start

heavy rainfall forecasting and warning · rainfall to inundation extent modelling · hindcast validation against observed flood extent

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.