Californians used to seaside breezes got a rude shock this weekend, as even cool Palo Alto soared into the triple digits. While this is just one data point in the long term studies of global warming, it contains a useful lesson in the difference between mean temperature and temperature distribution.
We often hear the quip “Hey, two degrees is no big deal” as a defensive reaction to the overwhelming evidence of climate change. Well, two degrees Celsius doesn’t seem big, but the resulting shift in the overall distribution of temperatures can lead to extreme impacts. As the IPCCC graph shows, with a normal or even Gaussian temperature distribution, shifting the mean to the right raises the likelihood of extreme weather events dramatically. Put simply, what used to be a once-in-a-hundred-years dog day of summer becomes a once-a-year heat wave like the one California is suffering through now.
In fact, the last scenario of both increasing mean and variance may be the most troubling. A January 2004 study in Nature (pdf) showed that a shift in mean alone was unlikely to explain an extremely improbable 2003 European heat wave. The paper gives some evidence that a local climate model with increased climate variability does a much better job explaining things.
While we are hesitant to draw conclusions from one particular weather event, the hot weekend got us both sweating and thinking about the future.