October 13, 2006

Where are the Droughts?

Filed under: Droughts, Precipitation

One of the pillars of the greenhouse apocalypse is the global warming will lead to a higher frequency, intensity, duration, and spatial extent of droughts in the future. This prediction is fairly easy to understand in terms of basic physical principles. Higher temperatures will lead to higher rates of potential evapotranspiration (PE), so even if rainfall stays the same or even increases slightly, the increase in PE will make droughts worse, make them last longer, make them more frequent, and make them expand their spatial extent. To make the matter even scarier, many climate models predict a decrease in precipitation in continental interiors, so with less rainfall, higher temperatures, and higher PE rates, drought frequency, intensity, spatial extent, and duration may substantially increase in places like the American heartland.

Literally dozens of articles have appeared in the scientific literature showing results that lead to the prediction of increased drought conditions in the central United States. The Intergovernmental Panel of Climate Change (IPCC) states in the Summary for Policymakers that “Increased summer continental drying and associated risk of drought” is “Likely, over most mid-latitude continental interiors” during the 21st century. In terms of seeing such a pattern in the observed climate record in the 20th century, the IPCC concludes it is “Likely, in a few areas.”

An important article appeared in the literature recently with some surprising results given the predictions of the climate models. Konstantinos Andreadis and Dennis Lettenmaier of the University of Washington have published a paper in Geophysical Research Letters entitled “Trends in 20th century drought over the continental United States,” and the results are peculiar—in light of climate model projections—to say the least. In the abstract, they write “Droughts have, for the most part, become shorter, less frequent, and cover a small portion of the country over the last century.”

Andreadis and Lettenmaier used a Variable Infiltration Capacity (VIC) model to simulate soil moisture and runoff over the continental United States; they ran the model at a spatial resolution of one-half degree and at a daily time step. The VIC solves energy and water balance equations over each grid cell, accounting for variability in soil (three layers of soil are considered), vegetation, precipitation, and topography. The model considers actual variations in precipitation, air temperature, and wind speed over the period 1925 to 2003 to calculate soil moisture and runoff. The team tested the VIC repeatedly and provided convincing evidence that the model very accurately simulates soil moisture and runoff.

Andreadis and Lettenmaier calculated trends in soil moisture and found that 1450 cells (43.6% of the domain) exhibit a statistically significant (p=0.05) upward trend, while far fewer (95 cells, or 2.9% of the domain) showed a statistically significant downward trend. They wrote “the wetting trends cover the majority of the country” and as seen in their figure below (Figure 1), the central United States is populated by “blue” grid cells indicating a significant increase in soil moisture. Furthermore, they found that “Annual trends in runoff are very similar to those found for soil moisture” and that “These results agree with previous studies that have suggested a general increase in streamflow over the conterminous U.S.”

Figure 1. Annual trends in soil moisture. Blue triangles show significant upward trends, red triangles show significant downward trends (from Andreadis and Lettenmaier, 2006).

Andreadis and Lettenmaier next turned their attention to trends in drought duration, and they found that 47 cells (1.4% of the domain) had a significant upward trend indicating increased duration, while more than twice that many (102 or 3.1% of the domain) showed a significant decrease in duration. As seen in the map below (Figure 2), there is no widespread evidence of any increase in drought duration in the central United States. They computed trends in drought severity and found 121 cells (3.6% of the domain) with a significant upward trend but 198 cells (6.0% of the domain) had a significant downward trend. Next up was drought frequency, and Andreadis and Lettenmaier concluded “There is a predominant reduction in drought frequency for the eastern U.S. and Midwest”. Finally, they looked at spatial extent of drought and found “Soil moisture (and runoff) drought spatial extent showed a downward trend which however was insignificant for all thresholds.”

Figure 2. Annual trends in drought duration. Blue triangles show significant downward trends, red triangles show significant upward trends (from Andreadis and Lettenmaier, 2006).

So in a world in which numerical models of climate are predicting increased drought frequency, intensity, duration, and spatial extent for the central United States given the ongoing buildup of greenhouse gases, Andreadis and Lettenmaier examine trends in drought over the period 1925 to 2003 and found that “droughts have, for the most part, become shorter, less frequent, less severe, and cover a smaller portion of the country.”


Andreadis, K. and D. Lettenmaier. 2006. Trends in 20th century drought over the continental United States.” Geophysical Research Letters, 33, 2006GL025711.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress