Toward the Seasonal Prediction of the Timing of Extremes and Hydroclimate

Abstract:

Results with a new high-resolution (50km atmospheric/land resolution) global coupled climate model, targetted to the understanding, intraseasonal-to-decadal prediction and near-term projection of regional and extreme climate, are explored. Focus is placed on extending the understanding and prediction from seasonally-aggregated quantities, to the seasonal timing of regional hydroclimate changes and extremes. We build on skillful seasonal predictions of the regional activity of tropical cyclones as well regional hydroclimate. It is shown that mean-state errors are a key constraint on the simulation and prediction of variations of regional climate and extremes, and methodologies for overcoming model biases are explored. Improvements in predictions of regional climate are due both to improved representation of local processes, and to improvements in the large-scale climate and variability from improved process representation. The seasonal phase-locking of interannual variability, and long with predictions of intraseasonal variabilit, provide potential sources of predictability of the timing of regional hydroclimate and extremes.