There is still much that climate models do not get right, including in regions where climate change is especially important to understand.
SAN FRANCISCO, CA., DEC. 10, 2013 — I came across this poster (PDF) yesterday, and it’s a great example of what you can find in the poster sessions, where the researcher is usually standing right there and is more than willing to answer questions and talk to you about his or her work.
This poster is by Walter Robinson and colleagues at North Carolina State University, discussing something I hadn’t heard of before: there is a warming “hole” near the center of the U.S. where warming has been surprisingly absent.
This region — shown by the box in the poster’s first figure — covers roughly Illinois, Wisconsin, Iowa, and some parts west. Warming projections for daily high temperatures in this region vary with climate models, but (bar chart in center) the ensemble of models peaks at about 0.2 degrees C per decade from 1959 to 2007.
But this isn’t close to the observed trend, which is only about 0.02 degrees C per decade. That’s a difference of almost a degree Celsius (2 degrees Fahrenheit) over the last 50 years, in a region that is one of the agriculturally productive areas in the world.
Global climate models do not replicate this hole, and Robinson and his colleagues found that aerosols do not explain it either. Nor does an atmosphere-only model with real sea surface temperature data.
Models predict the right trend for daily low temperatures, but not for daily high temperatures — something farmers, who already tend towards skepticism, are obviously going to care about.
For now it’s still a puzzle. And a good example of how good scientists are ever probing into the details of the data, testing the performance of climate models and making clearer what models can do and what they can’t — yet, at least.