Cross-posted from In the Field
It’s a warm and sunny spring day today in Vienna – can you guess then whether the coming summer will be colder or warmer than usual? Well, if you think that you hardly have more – though certainly not less – than a 50% chance of getting it right you’re, uhm, right. But guess what: Supercomputer-powered seasonal climate forecasts don’t do much better.
Seasonal climate predictions work relatively well only in the tropics. In Europe and North America their predictive skills are still pretty poor, meaning that forecast and observed climatology are often two very different things. And in some regions seasonal forecasts are actually worse than plain guessing.
This means that seasonal climate forecasts don’t yet provide reliable, if any, guidance for farmers, tourism managers, forest fire fighters, or for me and you. The idea that slowly varying boundary conditions, such as sea surface temperature distribution, snow cover and soil moisture, push the climate in a certain direction is well-established. But statistical climatology is one thing, daily wheather is another.
Andreas Weigel of the Swiss Weather Service, a rising star in the seasonal forecast community, made a few suggestions here at the EGU as to how predictive skills could be improved. Using more than one climate model is one promising possibility, statistical post-processing an re-calibrating forecasts is another, he explained in his well-received medal award lecture today
In the same session, Marie Boisserie of Florida Stae University in Tallahassee reported that when she included realistic initial soil moisture conditions to a climate model it greatly improved its predictive skills. Two-month forecasts of summer temperatures and precipitation in the US were more than twice as accurate than without the precipitation-derived soil moisture data.
Problem is that as yet there exists no reliable global observational database of soil moisture.
All eyes are now on the European Space Agency’s Soil Moisture and Ocean Salinity (SMOS) mission, designed to observe soil moisture over the Earth’s landmasses and salinity over the oceans, to be launched in June.