State of Global Water Resources 2025: Driest Rivers in 35 Years

The difference between water passing through and water staying behind

A river carries water through a landscape, while an aquifer stores water within permeable rock or sediment. Aquifers are not usually large underground caverns. Water occupies connected pores and cracks, through which it can move towards wells, springs and rivers. Rainfall can replenish this water, but only part of the rain reaches the saturated layers. The rest may run across the surface, remain in shallow soil or return to the atmosphere.

This creates an important distinction between flow and storage. A river's discharge measures the volume passing a point over time. Groundwater storage describes water retained below ground. Heavy rain can rapidly increase river discharge while deeper reserves recover much more slowly. Continued pumping can also offset that recharge. Therefore, above-normal flows in parts of South Asia can coexist with persistent groundwater deficits without either observation being contradictory.

What the global assessment found

The assessment identifies 2025 as one of the driest years for global river discharge within its 35-year comparison. Below-normal flows affected approximately 36% of global basin area in the modelling assessment. This is an area-based result, not a statement that 36% of individual rivers dried up. Conditions also differed sharply between regions. South and Southeast Asia generally had stronger flows, while parts of North America, the Middle East and Central Asia experienced deficits.

Slow-changing reserves showed a longer-running problem. WMO identifies a decline in terrestrial water storage since around 2014–2016 and repeated groundwater deficits in several regions. Terrestrial storage includes groundwater, surface water, soil moisture, snow and ice; it is broader than groundwater alone. The agency also estimated global glacier losses of about 408 gigatonnes during the 2025 hydrological year, with substantial uncertainty. A gigatonne is a billion tonnes. All major glacier regions again recorded net losses, although their rates differed.

How researchers measure water they cannot directly see

Ground measurements remain essential, but wells and river gauges cannot provide equally dense coverage everywhere. Satellites help fill part of that gap. The Gravity Recovery and Climate Experiment and its Follow-On mission detect changes in Earth's gravitational field. When large amounts of water accumulate or disappear, the distribution of mass changes slightly. Measurements between paired satellites allow researchers to track those changes across broad regions. They do not produce a direct photograph of each underground aquifer.

Researchers must separate the components of the combined signal. Snow, soil moisture, lakes and groundwater can all contribute to a measured storage change. Independent observations and models help estimate their individual shares. This is why a satellite storage map and a groundwater-well record answer related but different questions. The report's river-flow and storage comparisons also use different reference periods. Reading a coloured map without its baseline, unit and geographical coverage can therefore create a misleading impression.

Why repeated groundwater deficits become costly

Pumping initially draws down water around a well. Over time, that withdrawal can be supplied by increased recharge, reduced natural discharge or depletion of stored water. These adjustments are not necessarily harmless. Where an aquifer feeds a stream, pumping may reduce the groundwater reaching that stream. A well can therefore affect surface water even when it stands some distance from the riverbank. The size and timing of the effect depend on local geology and hydraulic connections.

As water levels fall, pumps may need to lift water farther, increasing energy requirements. Some shallow wells can become unreliable before deeper wells fail. In suitable geological settings, loss of groundwater pressure can also compact sediments and contribute to land subsidence. Near coasts, altered pressure can encourage saline water to move into freshwater-bearing formations. These are recognised consequences of depletion, not claims that every Indian district mentioned in the assessment has experienced all of them.

What the findings mean for water management

For India, the central issue is whether withdrawals and replenishment remain in balance over successive seasons. Recharge structures can help where the geology and water quality are suitable. However, adding recharge while allowing extraction to rise without limits may leave the underlying imbalance unchanged. Decisions about irrigation, crop choices and water supply need local aquifer information, not only national rainfall totals. Surface-water and groundwater planning also need to be considered together because the two systems exchange water.

The assessment highlights another practical limitation: the places most exposed to extremes do not always have the strongest monitoring networks. Better measurements help distinguish a temporary dry spell from a sustained decline. They also support cooperation across river basins that cross administrative or national boundaries. A regional warning is therefore a starting point for investigation and management. It cannot by itself identify the safe pumping rate of a particular village well.

Conclusion

The report's warning concerns the loss of reserves that help societies withstand dry periods, not simply a shortage of rainfall everywhere. River recovery, groundwater recovery and glacier recovery are different processes. Reliable water planning must track each one, explain their connections and respond before repeated withdrawals turn seasonal pressure into lasting depletion.