10/18/2010
Improved Estimation of Prospective Climate Change
Summary
In a recent study co-authored by a DOE-funded scientist from LLNL, future climate predictions are made based on the combined use of many climate models weighted according to their observational predictive skill. This approach provides greater reliability and statistically significant predictions than is possible using a single climate model. The authors considered two future greenhouse scenarios: one in which greenhouse gas emissions are stabilized around the year 2050 and another in which greenhouse gases continue to increase. Seventeen coupled ocean-atmosphere model simulations were used, weighted according to their ability to predict observations of 20th century climate. For both scenarios, statistically significant temperature increases occur globally. Precipitation changes were more variable with statistically significant changes only in some regions and only for the scenario in which greenhouse gases continued to increase. These new predictive methods should lead to improvement of model consensus on climate change.
References
Q. Duan and T.J. Phillips. 2010. “Bayesian estimation of local signal and noise in multimodel simulations of climate change,” Journal of Geophysical Research: Atmospheres, 115, D18123, doi:10.1029/2009JD013654.