The excellent “No Se Nada” weblog of Kevin Vranes alerted us to an important new paper on precipitation trends.
The new GEOPHYSICAL RESEARCH LETTERS paper by Thomas M. Smith, Xungang Yin and Arnold Gruber is entitled “Variations in annual global precipitation (1979–2004), based on the Global Precipitation Climatology Project 2.5° analysis”,
with the abstract,
“The Global Precipitation Climatology Project (GPCP) has produced a combined satellite and in situ global precipitation estimate, beginning 1979. The annual average GPCP estimates are here analyzed over 1979–2004 to evaluate the large-scale variability over the period. Data inhomogeneities are evaluated and found to not be responsible for the major variations, including systematic changes over the period. Most variations are associated with El Niño/Southern Oscillation (ENSO) episodes. There are also tropical trend-like changes over the period, correlated with interdecadal warming of the tropical SSTs and uncorrelated with ENSO. Trends have spatial variations with both positive and negative values, with a global-average near zero.”
The conclusion of their paper (subscription required) includes the text,
“The merged satellite and in situ GPCP global precipitation annual averages were examined for 1979–2004. Most variations are associated with ENSO and have no trend. A separate mode of variation shows a trend over the period. Testing indicates that this trend is significant and is not caused by data inhomogeneities. The trend mode is associated with simultaneous tropical SST variations over the period, with increased tropical precipitation over the Pacific and Indian Oceans associated with local warming of the SSTs. Increased precipitation in some regions is balanced by decreased precipitation in other regions, and the global average change is near zero. Although the trend mode is strong for this period, the record length is barely long enough to begin evaluation of interdecadal variations.”
This paper is interesting as one of the students in my class in the Spring of 2005 , Sheri Conner Gausepohl, explored this important issue. Her results are consistent with the conclusions in the new Smith et al paper. In our class, we concluded that the more uniform spatial coverage permitted by satellites provides a more accurate diagnosis of trends in a global context, in this case for precipitation, than do studies that focus primarily on in-situ surface data.
In addition, the Smith et al GRL study further documents the loss of information when there is a focus on a global average metric (in this case, they found that “the global average change is near zero”). The long term trends and interannual variability that matter to society are on the regional scale (e.g. droughts and floods) in which the Smith et al study did find important changes (i.e. “Increased precipitation in some regions is balanced by decreased precipitation in other regions…”).
The study also raises questions on the extent to which water vapor content has increased globally, if the global averaged precipitation has not changed significantly. A positive atmospheric water vapor feedback, in response to an increase in the radiative warming of the well-mixed greenhouse gases, is at the foundation of the IPCC perspective on global warming.