How Satellites Measure India's Hidden Groundwater From Space

India extracts roughly 250 cubic kilometres of groundwater every year, more than the United States and China combined. Most of it feeds irrigation. Almost none of it is measured at the point it disappears. The country has monitoring wells, but they are scattered unevenly, many are read manually once a season, and vast stretches of the Indo-Gangetic Plain have no instrumented well at all. What cannot be measured cannot be managed. For decades, the gap between what India knew about its aquifers and what was actually happening inside them was wide enough to lose a civilisation in.

The instrument that changed this is not a drill or a sensor lowered into the earth. It is a pair of satellites flying in formation 220 kilometres apart, measuring the distance between themselves to within a micron, roughly one-fiftieth the width of a human hair. The mission is called GRACE: Gravity Recovery and Climate Experiment, launched by NASA and the German Aerospace Center in 2002. Its successor, GRACE-FO, has been operating since 2018. The principle is elegant. Water has mass. Mass bends gravity. When groundwater accumulates or drains from a region, the gravitational field above it shifts by a tiny, measurable amount. As the two satellites pass over a depleted aquifer, the lead satellite slows fractionally before the trailing one does, because there is less mass pulling it forward. That lag, measured continuously, is converted into a map of water storage change across a footprint of roughly 200,000 square kilometres. It cannot resolve a single farm or a single district. What it can do is tell you, with high confidence, that the aquifer system beneath the Punjab lost a certain volume of water between one month and the next.

A landmark analysis published in Nature in 2009, using GRACE data from 2002 to 2008, estimated that groundwater beneath northwestern India, covering Rajasthan, Punjab, and Haryana, was depleting at a rate of about 4 cubic kilometres per year above what rainfall and river recharge could replace. That figure was not a model projection. It was a direct measurement of gravitational change converted to water volume. Later studies refined the estimate and extended the time series. The picture that emerged was consistent: the breadbasket states that feed India's Green Revolution are drawing on water that accumulated over thousands of years and is not coming back on any human timescale. GRACE data also captured something ground wells missed entirely, the acceleration. Depletion was not steady. It sped up in drought years and slowed slightly in good monsoon years, but the long-run trend did not reverse. Recharge was not keeping pace with extraction anywhere in the northwest.

India's own space programme has added a second layer to this picture. Satellites from ISRO, including the Resourcesat series and data from the Cartosat family, map land use, crop type, and irrigation extent across agricultural zones. When that surface data is combined with GRACE gravity anomalies and rainfall records from the India Meteorological Department, researchers can begin to separate natural variability from human extraction. The question GRACE alone cannot answer is why the water is going. The ISRO surface layer helps answer it: paddy cultivation in Punjab expanded significantly after minimum support price guarantees made rice more profitable than less water-intensive crops, and the satellite record of crop area correlates tightly with the gravity signal of aquifer loss. India also contributes ground-truth data to global hydrological models through the Central Ground Water Board's monitoring network. The combination of space-based gravity measurement and surface observation is more powerful than either alone, and Indian institutions are increasingly part of the research teams producing it.

GRACE and GRACE-FO measure change in total water storage, groundwater, soil moisture, surface water, snow, all at once. Separating the groundwater signal from the rest requires models and assumptions, which means the numbers carry uncertainty ranges that matter when policy decisions rest on them. The footprint of roughly 200,000 square kilometres also means that a state like Haryana appears as part of a regional average, not as a standalone measurement. Finer resolution requires either more satellites or different physics. NASA and ISRO have both identified next-generation geodetic missions as priorities. India's planned NISAR satellite, a joint NASA-ISRO synthetic aperture radar mission, will measure ground deformation, including the subsidence that happens when aquifers compact after being drained, at a resolution GRACE cannot approach. Land that sinks by centimetres as water is removed carries a permanent record of what was taken. Read from orbit, that record tells you not just that water is gone, but where it went and roughly when. The satellites orbiting above India's fields are not a solution to groundwater depletion. They are the first honest account of it, numbers that do not depend on who controls the pump or who reads the well.