THEGREENGROK    Planetary Watch

More Raindrops Are Falling on Our Head

by Bill Chameides | February 23rd, 2011
posted by Erica Rowell (Editor)

Permalink | 7 comments

 


Rainfall has become more intense in the Northern Hemisphere. Why is much of the discussion swirling around models instead of data?

Data show extreme precipitation in the Northern Hemisphere is on the rise.

We all know that without the showers we get no flowers. But too much showers — not so great. A little more than average rainfall can be a bother, and a lot more can be life-threatening. (Some anecdotal evidence here, here, here and here.) And so the prospect that global warming will lead to more intense rainfall and snowfall events probably should not be taken lightly.*

All well and good (or not so good), but should we do anything about it? Scientists’ warning of a potential outcome is one thing; proof that heavier rainfall is actually occurring is another. Now into the breach comes a new paper published in the journal Nature by Seung-Ki Min of Environment Canada and colleagues that purports to do exactly that.

Synopsis of the Study

The study consisted of two components. First the authors used observations of precipitation (i.e., rain and snow) between 1951 and 1999 across 6,000 ground-based stations to determine if the frequency of extreme precipitation events has changed. They then used eight different climate models with close to 40 separate simulations to assess if (in the model world) emissions of greenhouse gases from human activities change the frequency of these same types of events. Both the observations and the model simulations consistently showed an increasing trend in intense precipitation events over the study period. The correspondence between observations and model simulations led the authors to conclude that:

“human-induced increases in greenhouse gases have contributed to the observed intensification of heavy precipitation events found over approximately two-thirds of date-covered parts of Northern Hemisphere land areas.”

The Debate: What’s It All About

Not surprisingly, the publication of the Min et al paper was greeted with a good deal of media attention (some examples here and here). Much of it and the discussions that followed focused on the second part of the study — the authors’ use of computer models to establish a linkage between the positive trend in storm intensity in the Northern Hemisphere and anthropogenic releases of greenhouse gases.

Questions have been raised in some places about the robustness of the Min et al’s claim that they “show” a cause-and-effect relationship between increased precipitation intensity and greenhouse gases. Even the authors put a caveat on their findings: that the modeling simulations used to establish the linkage systematically “underestimate the observed increase in heavy precipitation with warming.” This is actually a two-edged sword: on the one hand it may mean that there is another (perhaps natural) driver of intensity not accounted for by the models; on the other, it may mean that the models underestimate the link between precipitation and greenhouse gases and therefore underestimate the consequences of future emissions.

Not Seeing the Data for the Model

Establishing the role of human activities in climate change is clearly an important issue, and so the focus on whether Min et al actually established such a link is germane, especially in the scientific realm. But from a policy point of view, you’ve gotta keep in mind that using a model simulation to “prove” the role of human activities and climate change is and will always be problematic. If you have any doubts, take a look at some of the musings of Kurt Gödel. Or, if you’re looking for a somewhat less esoteric take on the subject, try this article [pdf] by Naomi Oreskes and co-authors from 1994 (“Verification, Validation, and Confirmation of Numerical Models in the Earth Sciences,” Science, Vol. 263, No. 5147, Feb. 1994, pp. 641-646).

In short: if we’re going to insist on slam-dunk proof that what we see happening with the climate is due to greenhouse gases, we are going to have to wait a very long time — especially if we’re not going to depend on climate model simulations.

So Weary All the Time

I find all the debate over whether a link between human activities and precipitation trends has been scientifically established by Min et al to be scientifically interesting, but it takes the focus away from another really important part of their study. And that part is their finding that there has been a significant increase in the probability of extreme precipitation over the Northern Hemisphere over the last half of the 20th century. The trend, on average of a few tenths of a percent per year, has been especially pronounced over the United States and was especially large in the 1990s — an increase on the order of perhaps four percent over the decade.

This trend is not from a model simulation — it’s from data, real-world, empirical data. It’s really happening, folks.

While much of the world seems content to argue about climate models and twiddle their collective thumbs waiting for proof positive, the climate is changing and changing, more or less, in the ways we would expect it to as a result of greenhouse gas emissions. We can wait, the climate won’t.

__________________

Note

* As the climate warms, more evaporation occurs transferring more moisture to the atmosphere. That increased atmospheric moisture means that when it does snow or rain, those events tend to be heavier. More details here and here.

Correction: March 9, 2011, This post has been corrected to fix the misspelling of the surname of Naomi Oreskes.

filed under: climate change, faculty, global warming, Planetary Watch, rainfall, science, weather
and: , , , , , , , , , ,

7 Comments

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

  1. Ken Towe
    Mar 13, 2011

    According to NCDC-NOAA, the 2008 U.S. average rainfall was 30.48 inches, 1.34 inches above the long-term 1901-2000 average. This makes the 1901-2000 average 29.14 inches. The 2009 U.S. average annual rainfall was 31.47 inches and was 2.33 inches above the long-term average. In 2010 the U.S. average annual rainfall was 30.16 inches, 1.02 inches above the long-term average. According to W.A. Mattice, US Weather Bureau, in 1941 the average annual rainfall in the US was 32.36 inches, “somewhat over 3 inches higher than the mean” for 1886-1940. The 1886-1940 long-term average rainfall of 29.36 inches is well above the 1901-2000 20th century long-term average. Source: http://www.ncdc.noaa.gov/sotc/national/2008/13 Source: http://www.ncdc.noaa.gov/sotc/national/2009/13#precip Source: http://www.ncdc.noaa.gov/sotc/national/2010/13 Source: http://docs.lib.noaa.gov/rescue/mwr/069/mwr-069-12-0360.pdf

    • Bill Chameides
      Mar 22, 2011

      Ken: Paper was about extreme precipitation not average precipitation.

      • Ken Towe
        Mar 22, 2011

        BILL… Extreme precipitation? I’m not sure you read the 1941 report: http://docs.lib.noaa.gov/rescue/mwr/069/mwr-069-12-0360.pdf. In it Mattice states: “For the country as a whole 1941 was the third wettest year of record, exceeded only in 1905 with 32.69 inches and 1915 with 32.74 inches.” “The general review of 1941 precipitation might be briefly summarized by stating that the year was outstanding for heavy falls, with the last half of the year especially wet nearly everywhere west of the Mississippi.” “It was the wettest year of record in North Dakota, Utah, and New Mexico, with the last state having about a third more precipitation than in any previous year.” Then… in NATURE, 2003: http://www.manfredmudelsee.com/publ/pdf/flood.pdf Mudelsee et al. wrote: Central Europe: “For the past 80 to 150 yr, we find a decrease in winter flood occurrence in both rivers, while summer floods show no trend, consistent with trends in extreme precipitation occurrence.”

        • Bill Chameides
          Mar 28, 2011

          Ken: As you well know, the fact that there were years in the past with extreme rainfall, even record extremes, in no way negates the premise that there is an increasing trend in extreme rainfall events.

  2. MattN
    Mar 8, 2011

    Overall: Someone needs to do some serious quality control on the data. • The use of the HADEX RX1day dataset should be suspended until the data is fixed. • The HADEX RX1day dataset also should not be used until gridcell averages can be properly recalculated without distance-weighting. • The use of a subset of models which are selected without any ex-ante criteria damages the credibility of the analysis • If a probability-based index is going to be used, it should be used on the raw data rather than on averaged data. Using it on grid-cell averages of raw data introduces spurious uncertainties. • If a probability-based index is going to be used, it needs to be applied uniformly across all gridcells rather than using different distributions a gridcell by gridcell basis. • No analysis is given to justify the use of “optimal fingerprinting” with non-Gaussian data. Out of the 731 US stations with rainfall data, including Alaska, Hawaii and Puerto Rico, 91% showed no significant change in the extreme rainfall events, either up or down. Of the 340 mainland US stations with 40 years or more of records, 92% showed no significant change in extreme rainfall in either direction. The conclusions of their paper are not supported by their arguments and methods, and are contradicted by the lack of any visible trend. I can’t belive this paper was actually published. Just absolutely dreadful “science”.

    • Bill Chameides
      Apr 13, 2011

      MattN: If you are going to cut and paste a WUWT post — http://wattsupwiththat.com/2011/02/20/nature-unleashes-a-flood-of-bad-science — it’s appropriate to cite that link, if for no other reason than that others can follow your argument back to its source. There are good statistical reasons for not using the “raw data.” Because we are trying to uncover a small, spatially diffuse trend out of a noisy and spatially heterogeneous signal, it is most effective to do so by looking at the large scale. The finding that many individual stations contain no significant trend does not contradict the overall conclusion.

  3. chrisd
    Mar 8, 2011

    …is Oreskes, not Orestes. Not a big deal, but it does mess up some search engines (though not the google!).

©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