Numerical Ensemble Based SST Predictions

In our 2007 model suite, we also included another way to estimate future changes, based on output from global climate models. The multi-model ensemble mean of model runs assembled for the IPCC AR4 report (available at were found to have skill in predicting past values of both MDR SST and Northern Hemisphere temperature in a paper by Laepple et al. (2007). These ensemble means provide a non-linear estimate of future changes that lies between the flat line estimates and the linear trend estimates described above. For this model, our predictions are based on an estimate of the current state and an estimate of the change from that state. The ensemble mean, which effectively averages out the natural variability of the model runs, predicts the future non-linear trend. This is then adjusted to the current state using a bias correction. The current state is estimated using the mean of the last 8 years, since 8 years is the optimal window length calculated by minimizing the out-of-sample RMSE of hindcasts.

Our simple technique of using the IPCC ensemble mean, bias-corrected to the current climate, as a prediction for future temperatures, compares favorably with both our purely statistical predictions and the predictions from a complex initial-condition driven forecast model by Smith et al. (2007).

Was this article helpful?

0 0
Survival Treasure

Survival Treasure

This is a collection of 3 guides all about survival. Within this collection you find the following titles: Outdoor Survival Skills, Survival Basics and The Wilderness Survival Guide.

Get My Free Ebook

Post a comment