Over the last several months, the idyllic weather we’ve had in western Washington has contrasted sharply with reports of heat waves in the Northeast, drought in the Midwest, and wildfires in the Western U.S. (including eastern Washington). As I listened to news reports, I often wondered whether changes in climate have contributed to these events. It turns out I’m not the only one wondering. Climatologists have been working to answer that very question. Using newly developed methods, they have been providing more insight into the connections between increased greenhouse gas concentrations and the likelihood of extreme weather events.
In the introduction to a recent report from the American Meteorological Society and the National Oceanic and Atmospheric Administration, editors Thomas Peterson, Peter Stott and Stephanie Herring point out that it’s analogous to analyzing the home runs hit by a baseball player who has been using steroids. The player would almost certainly have hit some number of home runs without steroids. Therefore, it is not possible to attribute any particular home run to steroid use. But it is possible to calculate the increased probability of home runs that has resulted from his use of the drugs.
Researchers in our region are taking the lead in some of the research featured in the report above. In one chapter a team led by David Rupp and Phil Mote of Oregon State University’s Oregon Climate Change Research Institute (OCCRI) examined whether changes in the levels of greenhouse gases altered the likelihood of the 2011 heat wave and drought in Texas. That year was the hottest and driest growing season experienced since record-keeping began in 1895. Altogether, the 2011 event caused estimated direct economic impacts of $5.2 billion on the state’s agricultural sector.
To carry out the analysis, Rupp’s team generated a very large set of global climate model simulations. In essence, they produced a lot of different versions of weather that could have happened under the present climate cycles and greenhouse gas conditions so that they could calculate the likelihood of having drought conditions similar to those that did occur in 2011. Then they compared this to an equally large number of simulations of years in the 1960s that had similar sea surface temperature patterns to the present. Because the 1960s years chosen were similar to present day in terms of the cycles in climate such as El Niño that are not related to human-induced changes in greenhouse gas levels, they could attribute any changes in the frequency to changes in greenhouse gas concentrations.
The team concluded that precipitation conditions similar to those seen in 2011 are now slightly more likely than they were in the 1960s because of increased greenhouse gas concentrations, and that the heat conditions are roughly 20 times more likely. While this work is ongoing and additional analysis is planned, the results are none-the-less intriguing.
The weather events analyzed in the report all occurred in 2011. Participating scientists utilized recent analytical and computational advances to complete analyses much more quickly than previously would have been possible. Rupp et al.’s work was facilitated by OCCRI’s participation in www.climateprediction.net, which uses computing power donated by the public to carry out work that would otherwise have been prohibitively expensive (and slow). The authors are continuing to develop the work and take it through the scientific peer review process. But in the meantime, is this kind of analytical speed helpful?
I think that in many situations, this speed has the potential to make the scientific analysis more relevant to those making management and policy decisions. Information provided while the extreme weather event is still fresh in the public’s mind is more likely to contribute to plans to reduce future harm. For example, knowing that extreme heat will be much more likely to occur in the future in Texas, while extreme drought is not increasing nearly as much in frequency, could be very helpful as managers invest limited resources to minimize economic damage from future heat waves and droughts.
Of course, this benefit has to be balanced against the possibility that ongoing work may not support the initial conclusions. After all, the careful (and time consuming) process of scientific review is designed to enhance the confidence in the scientific conclusions, and may be particularly important for studies like these that represent the “state of the science”. This makes it particularly important that the level of scientific confidence is clearly communicated to resource managers along with the conclusions of a particular study. And that any ongoing studies that confirm or contradict initial conclusions are also communicated to resource managers as part of an ongoing conversation.