Satellite Imagery Is Now Tracking Flood Risk Across the Ganga Plains in Real Time

The single most useful fact about satellite flood monitoring is also the least obvious one: the satellites doing the most important work are not taking photographs. The European Space Agency's Sentinel-1 constellation uses Synthetic Aperture Radar, or SAR, a C-band microwave pulse that bounces off the Earth's surface and returns to the sensor regardless of cloud cover, rain, or darkness. During the Indian monsoon, when optical satellites are effectively blind for weeks at a stretch, SAR keeps imaging. Water absorbs the radar pulse and returns almost nothing to the sensor. Land, vegetation, and built structures bounce it back strongly. The result is a map where flooded areas appear as dark patches against a bright background, readable even when the sky above them is an unbroken wall of cloud.

Sentinel-1A, launched in 2014, and the broader Sentinel-1 constellation can achieve a revisit time of roughly six days over the same location, which matters enormously when a flood is moving downstream through a river system that stretches across three states.

The Ganga basin drains approximately 26 percent of India's total geographical area and supports one of the densest concentrations of rural population on the planet. The plains that run through Uttar Pradesh, Bihar, and West Bengal sit at low elevation, with a complex web of tributaries, the Ghaghra, the Gandak, the Kosi, the Yamuna, each capable of flooding independently before their waters join the main channel. A gauge on the Ganga at Patna tells you what is happening at Patna. It tells you almost nothing about what the Kosi did three days ago in Nepal, or whether the embankments along the Gandak in Gopalganj district are holding.

Ground-based monitoring networks, river gauges, rain gauges, manual observation posts, are sparse relative to the basin's size and frequently go offline precisely when floods make them hardest to reach. The 2019 floods in Bihar displaced over one crore people across 13 districts. The 2022 monsoon season saw inundation in parts of eastern UP and Bihar that lasted for weeks. In both cases, the gap between what ground sensors reported and what was actually submerged was significant. Satellite imagery filled parts of that gap in near real time.

India does not rely solely on ESA's Sentinel constellation. ISRO's National Remote Sensing Centre, based in Hyderabad, is the operational hub for satellite-based disaster monitoring in the country. NRSC processes SAR data from Sentinel-1 and from ISRO's own earth observation satellites, including the Resourcesat series, to generate flood inundation maps that are updated as new passes become available. These maps are shared through ISRO's Bhuvan geoportal, where state disaster management authorities, the National Disaster Management Authority, and the Central Water Commission can access them directly.

The Space Applications Centre in Ahmedabad runs MOSDAC, the Meteorological and Oceanographic Satellite Data Archival Centre, which integrates rainfall data from ISRO's INSAT and METSAT series with the inundation maps to produce a more complete picture of where water is accumulating and where it is likely to move next. The Central Water Commission uses this combined data stream to issue flood forecasts for major river stations, with a forecast horizon that has been extended in recent years partly because satellite inputs now arrive faster than manual field reports.

Real-time in satellite flood monitoring does not mean a live feed. It means a map that is updated every one to six days, processed within hours of the satellite pass, and distributed to decision-makers before the next wave of inundation arrives. For a river system where a flood pulse can travel hundreds of kilometres over several days, that cadence is operationally useful in a way that weekly or fortnightly updates are not.

The downstream effect is measurable. State governments in Bihar and UP have used NRSC flood maps to prioritise evacuation routes and pre-position relief material in districts that the maps flagged as high-risk before water reached them. The Copernicus Emergency Management Service, which activates globally for declared disasters, has been called on for Indian flood events and provides rapid mapping products that Indian agencies can cross-reference against NRSC outputs.

Commercial operators are also part of the picture. Companies like Planet Labs operate constellations of small optical satellites with daily revisit capability, useful during the brief clear-sky windows that occur even in monsoon season. Maxar's high-resolution imagery has been used for damage assessment after flood peaks recede. The monitoring architecture is not a single system, it is several overlapping ones, each with different strengths in resolution, revisit frequency, and spectral range.

The question of who receives the warning, and how fast, remains the harder problem. A flood inundation map shared on Bhuvan reaches a state emergency operations centre. Whether it reaches a gram panchayat in a low-lying block of Darbhanga in time to move cattle and stored grain is a question of administrative infrastructure, not satellite capability. The imagery can now see the water coming. The last kilometre of that information chain is still human.