Fate of water pumped from underground and contributions to sea-level rise

The contributions from terrestrial water sources to sea-level rise, other than ice caps and glaciers, are highly uncertain and heavily debated1–5. Recent assessments indicate that groundwater depletion (GWD) may become the most important positive terrestrial contribution6–10 over thenext 50years, probably equal in magnitude to the current contributions from glaciers and ice caps6. However, the existing estimates assume that nearly 100% of groundwater extracted eventually ends up in the oceans. Owing to limited knowledge of the pathways and mechanisms governing the ultimate fate of pumped groundwater, the relative fraction of global GWD that contributes to sea-level rise remains unknown. Here, using a coupled climate–hydrological model11,12 simulation, we show that only 80%of GWD ends up in the ocean. An increase in runo to the ocean accounts for roughly two-thirds, whereas the remainder results from the enhanced net flux of precipitationminus evaporation over the ocean, due to increased atmospheric vapour transport from the land to the ocean. The contribution of GWD to global sea-level rise amounted to 0.02 (±0.004)mmyr−1 in 1900 and increased to 0.27 (±0.04)mmyr−1 in 2000. This indicates that existing studies have substantially overestimated the contribution of GWD to global sea-level rise by a cumulative amount of at least 10mm during the twentieth century and early twenty-first century. With other terrestrial water contributions included, we estimate the net terrestrial water contribution during the period 1993–2010 to be +0.12 (±0.04)mmyr−1, suggesting that thenet terrestrialwater contribution reported in the IPCC Fifth Assessment Report report is probably overestimated by a factor of three. Sea-level rise (SLR) is a direct effect of climate change, through the thermal expansion of ocean waters and the contribution of melt water from ice sheets, ice caps and glaciers. In an initial assessment13, the net contribution of sub-polar terrestrial water storage change to global sea-level variation was highly uncertain and heavily debated14–18. Terrestrial water contribution to sea-level variation includes the filling (due to impoundments) and sedimentation of dams, groundwater depletion, drainage of endorheic lakes and wetlands, deforestation, and changes in natural water stores (soil moisture, groundwater, permafrost and snow)1,2,19,20. In the IPCC Fourth Assessment Report (IPCC AR4)3, the contribution of nonfrozen terrestrial waters to sea-level variation is not included owing to limited knowledge and the assumption that the negative contributions, such as the filling of dams, would compensate the positive contributions, mainly from GWD. GWD, that is, the extraction of groundwater reserves at rates greater than its replenishment, can result in a positive contribution to SLR due to the net transfer of dormant fossil groundwater to the active hydrological cycle, and eventually to oceans. Lacking in situ groundwater-level observations increases the uncertainty in estimating GWD for many parts of the world. One of the earliest studies13 estimated a global GWD of 86.7 km3 yr−1, which contributes 0.24mmyr−1 to SLR. Another study17 indicated that global GWD can contribute 0.10–0.30mmyr−1 to SLR. A regional study21 of theMiddle East andNorthAfrica estimated a rate ofGWD of 26.8 km3 yr−1, equivalent to 0.075mmyr−1 of SLR. Although these studies have evaluated direct groundwater storage changes, they have covered only limited global regions and do not account for some of the large aquifer systems where intensive groundwater mining has been well known (for example, Indo-Gangetic Plain and North China Plain)22. More recently, using a global hydrological model, one study6 estimated the global GWD rate for 2000 to be 283 (±40) km3 yr−1 (0.8 ± 0.1mmyr−1), responsible for 25 (±3)% of recently observed SLR3. Later, the same authors revised their estimate7 by introducing a multiplicative correction factor to the original estimates6 for nonarid areas where the increased capture may be significant. The results showed that, during the 20th century, the GWD contribution to global sea level has increased from 0.035 (±0.009)mmyr−1 in 1900 to 0.1 (±0.03)mmyr−1 in 1950, reaching 0.57 (±0.09)mmyr−1 in 2000. The flux-basedmethod6,7, using only groundwater pumping and recharge fluxes, however, tends to overestimate GWD as it does not account for the increased capture due to decreased groundwater discharge and enhanced recharge from surface waters due to change in groundwater level. The method also ignores the compensating effects of pumping in other hydrologic fluxes, such as groundwater recharge and discharge, and as such does not represent a net contribution to SLR. Alternatively, a volumebased study8 estimated a smaller global GWDof 145 (±39) km3 yr−1 (0.41 ± 0.1mmyr−1) during 2000–2008. This method used direct evidence of groundwater storage changes from in situ groundwater-level observations, calibrated groundwater modelling andGRACE satellite-basedwater storage data23–25 to calculate GWD (101.6 km3 yr−1; 0.29mm yr−1) for the USA and five other major aquifer systems of the world (north India, North China Plain,