THEGREENGROK    Planetary Watch

Gun Smoke Seen in River Discharge

by Bill Chameides | October 12th, 2010
posted by Erica Rowell (Editor)

Permalink | 5 comments


Circumstantial evidence shows the water cycle is intensifying because of global warming.

The globe is warming. OK, but so what? Who cares about a degree or two increase in global temperatures. It really starts to matter if the warming leads to other climate disruptions that we really care about. Like rainfall. More rainfall, less rainfall, more intense rainfall, any change along those lines tends to get folks’ attention.

Scientists are pretty certain that global warming will lead to an intensified water cycle: warmer temperatures evaporate more water from the Earth’s surface and thus more water vapor goes into the atmosphere. To balance things out, more rain must fall out of the atmosphere. Since most of the evaporation occurs over the ocean, it follows that this intensification should lead to more precipitation, on average, over the continents.

Figuring Out How to Tell If There’s More Rain

This all makes sense from the point of view of basic thermodynamics and the law of mass balance, but is it actually happening? Has the amount of precipitation over the continents increased?

One way to find out is to measure the amount of rainfall everywhere and sum it up. A lot easier said than done. For one, it is not possible to measure everywhere, and rainfall can vary a lot spatially; so rainfall measurements at sites meant to represent rainfall for a given area may not be an accurate representation of the area’s rainfall. For another, rainfall is very “noisy” — with lots of variations from month to month and year to year, and so it’s hard to discern any long-term trends from such measurements.

Another approach is to use a metric like the total amount of water flowing from rivers into the ocean, which represents the sum total of all rainfall minus evaporation (assuming the total amount of water on land does not change). Because this metric sums up over all rainfall, it smoothes out a lot of the spatial and temporal variability in rainfall. If the water cycle is intensifying, then one should see an increased discharge of water from the continents to the ocean. In principle it should be easy to do this using water-flow data from the world’s major rivers. However, it appears that this approach is also problematic most notably due to the absence of a comprehensive, high-quality global dataset and problems with the various methodologies used to compensate for these deficiencies (example here).

New Study Helps Estimate River Discharge

So, are we stuck? Not quite.

In a paper published last week in the Proceedings of the National Academy of Sciences, Tajdarul Syed of the University of California, Irvine, and co-authors report on a clever approach that may provide a path forward. They essentially tease out the rate of river discharge into the world’s oceans using a “mass balance” approach where:

The change in the total mass of water in the ocean must be equal to:

  • the rate of precipitation over the ocean
  • plus the rate of river discharge
  • minus the rate of evaporation.

The authors use satellite data collected between 1994 and 2006 to calculate the change in mass, the rate of evaporation, and the rate of precipitation.* Taking the difference from these, they are able to calculate the residual (i.e., the rate of river discharge).

The results show a strong signal from the El Niño-Southern Oscillation (ENSO) cycles with a large rise in river discharge between 1994 and 1999 (recall that one of the strongest El Niño’s ever recorded peaked in 1998). River discharge then declined until about 2002, rose and peaked in 2003-2004 during another El Niño, and then declined.

Superimposed on these up-and-down fluctuations appears to be a longer-term upward trend in total river discharge. Over the 1994-2006 study period, the authors estimate that river discharge increased by almost 20 percent (from about 7,200 cubic miles to about 8,400 cubic miles, or more than nine trillion gallons, of water) per year.

What does it mean? It’s another smoking gun. But only a smoking gun.

The analysis of Syed and co-authors does not establish a cause-and-effect relationship between warming temperatures and river discharge. The period of the analysis only covers 13 years — not quite long enough to establish a climatic trend, especially given the degree of interannual variation seen in river discharge especially as it relates to ENSO cycles. And, because the analysis depends upon satellite data, there is always a question of instrument drift. (Notably, the authors make no effort to quantify “instrument- and algorithm-specific errors,” which “though important,” are “beyond the scope of this study.”)

In spite of these caveats, this type of analysis offers a promising methodology that should circumvent most of the limitations of previous studies and should become more precise as satellite records improve and lengthen.

My takeaway: this study provides one more piece of evidence in the forensics of climate change. The question to consider: You’re the DA. How much evidence do you need before you go to trial?



*The change in ocean mass is computed from global mean sea-level, salinity and temperature data. Ocean precipitation estimates are derived from global datasets (CMAP and GPCP) that merge satellite, surface radar and rain-gauge data. Ocean evaporation estimates are calculated from satellite data that looks at latent heat fluxes. More info here.

filed under: climate change, El Nino-Southern Oscillation, faculty, global warming, Planetary Watch, rainfall, science, weather
and: , , ,


All comments are moderated and limited to 275 words. Your e-mail address is never displayed. Read our Comment Guidelines for more details.

  1. David L. Hagen
    Oct 14, 2010

    The article does not appear to mention water pumping: “New study finds groundwater pumping for irrigation contributes to one-quarter of the sea-level rise observed in today’s oceans.” Compare > 100 year record showing solar driven precipitation. Linkages between solar activity, climate predictability and water resource development* WJR Alexander et al. JOURNAL OF THE SOUTH AFRICAN INSTITUTION OF CIVIL ENGINEERING Vol 49 No 2, June 2007, Pages 32–44, Paper 659 Insufficient evidence presented. Ignored contrary evidence. Case “not proven”.

    • Bill Chameides
      Nov 10, 2010

      David: With respect to your first point, groundwater pumping is implicitly included in their analysis. In Syed et al’s words, “Trends in ΔM∕Δt [ocean mass change over time] actually represent acceleration in global-ocean water mass. Note that contributions from the Greenland and Antarctic ice sheet, alpine glaciers, and other land-management practices (e.g., reservoir storage and groundwater mining) are implicitly included in this trend.” For more discussion on this topic check out this more recent post: With respect to your second reference, the fact that precipitation or river-flow volume varies with solar activity is irrelevant to the post.

  2. CHunter
    Oct 14, 2010

    How does this method account for melting as a not-caused-by-the-water-cycle reason for ocean increase? Between the polar ice caps and retreating mountain snowpack, this seems like it might be significant, but I really have no way of the scale of either of those contributions. What do the scientists think?

    • Bill Chameides
      Nov 10, 2010

      CHunter: Contributions from melting ice sheets and glaciers were implicitly included in the authors’ analysis.

      • CHunter
        Nov 10, 2010

        It was hard to tell from what was reported here. Thank you.

©2015 Nicholas School of the Environment at Duke University | Box 90328 | Durham, NC 27708
how to contact us > | login to the site > | site disclaimers >

footer nav stuff