Post by Mark Duffett on Apr 23, 2012 22:39:28 GMT 9.5
Are any other readers members of climateprediction.net? The project (in which climate model parameter space is explored by running multi-member ensembles on thousands of participants' PCs) recently had a notable publication (www.nature.com/ngeo/journal/vaop/ncurrent/full/ngeo1430.html) in Nature. The upshot was summarised as follows:
Another, more pessimistic way of looking at the results is that there are some plausible models, which reproduce 20th century climate successfully, that go on to result in warming in excess of 4C by 2050 on a mid-range emission scenario.
The experiment, first launched in 2006, represents the first multi thousand member ensemble of simulations using a complex coupled atmosphere-ocean climate model, and addresses some of the uncertainties that previous forecasts, using simpler models or only a few dozen simulations, may have over-looked. Results from the experiment suggest that a global warming of 3 degrees Celsius by 2050 is as equally plausible as a rise of 1.4 degrees (relative to the 1961-1990 average). This range is derived from the range of simulations in the ensemble that accurately reproduce observed temperature changes over the last 50 years.
The results suggest that the world is very likely to cross the '2 degrees barrier' at some point this century if emissions continue unabated, and that those planning for the impacts of climate change need to consider the possibility of warming of up to 3 degrees (above the 1961-1990 average) by 2050 even on a mid-range emission scenario.
Another, more pessimistic way of looking at the results is that there are some plausible models, which reproduce 20th century climate successfully, that go on to result in warming in excess of 4C by 2050 on a mid-range emission scenario.