Where Has All The Water Gone?

11/03/2010 03:20 pm ET | Updated Nov 18, 2013

Crossposted with

Mankind is moving buckets and buckets of water from land to the ocean.

Sometimes science moves slowly and sometimes quickly. This is an instance of quick.

A couple of weeks ago TheGreenGrok covered a paper by Tajdarul Syed of the University of California, Irvine, et al who used hydrologic data to estimate the rates at which water flowed from the continents to the sea. They found that the rate rose over the study period from 1994 to 2006 and that the strongest component of that increase was an increase in evaporation over the ocean. The authors noted that such trends, if they continue into the future, would be evidence of an intensification of the hydrologic cycle, in which increased evaporation over the ocean leads to increased precipitation over the continents and subsequently more river discharge into the ocean.

While I had some reservations about the study -- recognizing, as the authors did, that the time period wasn't sufficiently long enough to draw conclusions about changes in the hydrologic cycle, and finding there were uncertainties in the numbers they derived -- I generally saw the paper as but another confirmation of the fact that our climate is changing. (To be clear, the findings of Syed et al were and are in no way central to the climate change issue but in line with it.)

But a new piece of the puzzle has been added with a report just out in the journal Geophysical Research Letters by Yoshihide Wada of Utrecht University in the Netherlands and colleagues. And that new piece calls into question the conclusions of Syed et al.

The Wada et al paper is about groundwater, but before we get to the specifics, a little background.

Tapping Water From Below

The most obvious place to get water if not from rainfall is from the Earth's surface -- rivers, streams, and lakes. But increasingly surface water is not adequate. For one, there's just not enough; more people mean more demand for drinking water -- and more demand for food and growing food, which often means irrigation and hence more water. Secondly, what is on the surface is often no longer safe to drink, as more people also mean more surface water pollution from human refuse, runoff from farms, and industrial effluent.


Groundwater, increasingly turned to as surface water becomes more unusable or unavailable, is not without its problems. Chief among them are pollution and depletion when withdrawals exceed recharge. (Source: USGS)

So, communities are increasingly looking downward for their water -- that is, to water stored below the ground surface. Make no mistake, groundwater has been a boon to civilization, literally allowing us to turn deserts into gardens. Tapping groundwater for irrigation, for example, has allowed us to expand agriculture into parts of the world like Nebraska and India where rainfall had formerly limited yields. It's also allowed large populations to live in regions that wouldn't otherwise meet their demands for fresh water. It's estimated that more than two billion people (35 percent of the world's population) live in areas where rainfall and surface water are not sufficient to meet their needs.

However, reliance on groundwater has had some unintended consequences. In some parts of India and Africa, people have switched from drinking polluted surface water to drinking groundwater only to find their new source of water tainted with high but natural levels of arsenic and fluoride.

In principle groundwater is a renewable resource. Water is withdrawn but is generally replaced or recharged when rainwater from the surface percolates downward. However, the rate of recharge can be quite small -- on the order of a few millimeters a year -- or even non-existent as in the case of fossil aquifers.

Problems for all aquifers arise when the rate of withdrawal exceeds the rate of recharge. When that occurs, the groundwater resource can be depleted much like a non-renewable resource. This, for example, is what is happening to parts of the Ogallala Aquifer, which underlies and irrigates a large portion of America's breadbasket. By all estimates the Ogallala is in danger of failing, but the exact timing of that will depend on pumping costs, crop prices, and advances in technology. (More on mining water and the Ogallala Aquifer.)

A Global View of Groundwater

How badly is groundwater being depleted globally? To help answer that question, Wada and colleagues used a global hydrological model with groundwater extraction estimates for arid and sub-humid parts of the world to get at the margin of groundwater depletion that's not being recharged. Their conclusions are striking. They estimate that globally about 280 cubic kilometers of water (the equivalent of 113 million Olympic pools) were depleted from groundwater aquifers in 2000; in 1960 that number was only about 130 cubic kilometers. Not exactly a sustainable practice.

A Couple of 'Where' Questions

  1. Where is the depletion occurring? Areas with the largest losses include:

  • parts of India and Pakistan,
  • the U.S. Central Plains (the Ogallala),
  • California's Central Valley,
  • parts of southern Europe,
  • northern Africa, and
  • the Middle East.
  • Where is all that extra water going? Wada and co-authors argue the bulk of it drains into the ocean. It's not entirely clear that that is correct, but If it is, then their findings have implications beyond the sustainability of our water-use practices.
  • This last finding would call into the question the conclusions of Syed et al noted at the top of the post. The Wada and Syed papers are consistent and yet inconsistent. Consistent in the sense that Wada et al's finding of increasing groundwater depletion is in line with Syed et al's finding that the rate of river runoff is increasing. But the latter's inference that this trend is evidence of an intensifying hydrologic cycle is not.

    The other implication relates to sea-level rise, placing groundwater depletion right smack dab in the climate change issue.

    Sea Level and Groundwater

    The rate of sea-level rise for the period 1993-2007 is observed to be about 3.3 millimeters per year, but has slowed to 2.5 millimeters per year for the period 2003-2007. Current thinking is that that rise is caused mostly by sea-water expansion from increasing temperature and by the runoff of water from the melting of glaciers and ice sheets. Until the latest period, however, there has been a problem matching up the estimates of the contributions from these two processes with the observed rate, as these two sources of sea-level rise are smaller than the observed rate. Many scientists have concluded that this discrepancy could be explained by uncertainties in the estimated rates or other processes related to changes in land-water storage that have been difficult to quantify.

    Wada et al suggest that groundwater depletion, assuming all other stores of water (surface and atmospheric) remain constant, contributes about 0.8 millimeters per year or about 25 percent of the current rate of sea-level rise. The clear implication: depletion of groundwater is making a significant contribution to sea-level rise. Two caveats: Like the Sayed et al paper there are uncertainties in Wada and colleagues' analysis and the authors did not consider a potential offsetting process -- the impoundments of water on the continents in reservoirs.

    The idea that groundwater depletion might be contributing to sea-level rise is not new. In fact, it was discussed in the latest report of the Intergovernmental Panel on Climate Change (IPCC). However, because of uncertainties in this contribution and the possibility that this positive contribution to sea-level rise is offset by the negative effect of water impoundments in reservoirs, the IPCC authors did not quantify this process in their assessment. Should they have? Maybe so. But even so, the groundwater depletion story of Wada et al doesn't change the big-picture story of climate change and sea-level rise. Should the Greenland or Antarctic ice sheets go, the contribution to sea level from groundwater depletion will be ... well, like a drop in the bucket.