In his recent press blitz, NASA’s James Hansen tries to tie extreme weather events, such the current drought affecting much of the central U.S., to anthropogenic global warming. But the real world argues otherwise.
Hansen is quite adept at timing global warming pronouncements with extreme weather events. Recall that it was during a similar hot, dry period back in the summer of 1988 that Hansen first testified to Congress that global warming from human greenhouse gas emissions was impacting current weather events—testimony which many credit as giving rise to the global-warming-is-going-to-be-bad movement. But then, as now, the tie-in between weather events and human changes to the atmospheric greenhouse effect is tenuous at best, and tie-ins to specific events are ill-supported and ill-advised. In the best case, the anthropogenic emissions-driven rise in global temperatures has a small ancillary impact on a specific extreme weather event, but in the vast majority of the cases, its role is nugatory and undetectable.
Such is the case with the impact of global warming on U.S. drought.
In his recent op-ed and article in the Proceedings of the National Academy of Sciences (recall that the peer review process at PNAS is more like watered down “pal review”) Hansen clearly states that U.S. droughts are being enhanced by global warming.
Hansen has a lot of weight in his corner on this one, because the standard measure of drought—the Palmer Drought Severity Index (PDSI)—has temperature incorporated into its equation. The higher the temperature, the greater the potential evaporation term, and thus the tendency towards drier conditions. But (obviously) precipitation amount is also included in the calculation of the PDSI. More precipitation pushes the index away from drought conditions. So just because temperature is on the rise does not mean that droughts must become more frequent—changes in precipitation could intervene and counteract the influence of a temperature increase.
Across the U.S., over the past century or so, both temperature and precipitation have been on the upswing. Figure 1 shows that U.S. annual PDSI from 1895 through 2011. The more positive the PDSI values, the wetter conditions are, the more negative the PDSI values, the drier things are. There is no statistically significant trend over the 117 period of record, although the non-significant tendency is upwards (towards wetter conditions), evidence that the increase in precipitation is more than offsetting the impact from rising temperatures in the U.S.
Figure 1. U.S. annual average palmer Drought Severity Index value, 1895-2011 (data source: National Climatic Data Center).
But what we are really after is not the impact of U.S. temperatures on U.S. drought, but, from Hansen’s assertion, the impact of global temperatures on U.S. drought.
We can examine this by first assessing the influence of global temperatures on U.S. temperatures. By using a simple linear regression, we find that about one-third of the variability in U.S. temperatures is explained by global temperature variations. Figure 2 shows the observed annual U.S. temperatures from 1895 through 2011 and that part of them which is explained by global temperatures. Notice that global temperatures can capture the trend in U.S. temperatures, but little of the year to year variation. But hey, c’est la vie—as we have stressed for years, local and regional conditions largely march to a tune that is different than the one being drummed by global average conditions.
Figure 2. The observed annual U.S. temperatures from 1895 through 2011 (light gray) and that part of them which is explained by global temperatures (black). Global temperatures explain about 33% of the variance in U.S. temperaures (temperature data source: Goddard Institute for Space Studies).
In order to see how much of the global temperature signal that is present in the U.S. PDSI history, we can regress that part of the U.S. temperature record that is related to global temperatures (the inner (black) curve in Figure 2) with the U.S. PDSI history (Figure 1). No significant relationship emerges from this analysis (the explained variance between the global signal in the U.S. temperature record and the U.S. PDSI history is less than 1%) (Figure 3, left).
In other words, global warming does not influence U.S. drought.
Now, let’s take a look at how much of the U.S. drought record is related to the non-global warming part of the U.S. temperature record. Doing this analysis, we find the influence to be a statistically significant 20% (Figure 3, right). The other 80% likely has to do with various and sundry influences on precipitation variability.
Figure 3. (left) Scattergram of U.S. annual PDSI values and the global warming influence on U.S. average annual temperatures, 1895-2011. (right) Scattergram of U.S. annual PDSI values and the non-global warming influence on U.S. average annual temperatures, 1895-2011.
The bottom line result here is that the influence of temperature on the U.S. PDSI (drought) history is one that is being driven by natural variability not global warming.
In other words, the situation is as it always has been. And the 2012 drought conditions, and every other drought that has come before, is the result of natural processes, not human greenhouse gases emissions.
Simply put, Jim Hansen is wrong.