Statistical Methods in Climate and Seasonal Prediction (Invited)

Statistical methods have been intertwined with climate prediction for decades. Recognizing the uncertainty of climate, making probabilistic climate forecasts is standard practice. One could say that a useful working definition of climate is the statistics of weather. This talk will present a brief discussion of the statistical methods used in climate prediction including seasonal forecasting, interannual variability and climate change. Some of the areas touched on will be determining the current climate state and climate change, prediction of modes of climate variability and impacts, dynamical versus statistical climate modeling, use of ensembles and multiple model ensembles, and post-processing of dynamical models to correct for bias and determine uncertainty in climate predictions. In addition to describing current use of statistical methods in climate science, I will pose some questions about future frontiers in climate prediction and the role of statistical methods.