Evaluation Of A Mesoscale Short Range Ensemble Forecasting System Over The Northeast United States PDF Download

This study developed and evaluated a short-range ensemble forecasting (SREF) system with the goal of producing useful forecast probability (FP). Real- time, 0 to 48-h forecasts from four different SREF systems were compared for 129 forecast cases over the Pacific Northwest. Eight analyses from different operational forecast centers were used as initial conditions (ICs) for running the Fifth-Generation Pennsylvania State University-National Center of Atmospheric Research Mesoscale Model (MM5). Additional ICs were generated through linear combinations of the original 8 analyses, but this did not result in an increase in FP skill commensurate with the increase in ensemble size. It was also found that an ensemble made up of unequally likely members can be skillful as long as all members at least occasionally perform well. Model error is a large source of forecast uncertainty and must be accounted for to maximize SREF utility, particularly for mesoscale, sensible weather phenomena. Inclusion of model perturbations in a SREF increased dispersion toward statistical consistency, but low dispersion remained problematic. Additionally, model perturbations notably improved FP skill (both reliability and resolution), revealing the significant influence of model uncertainty. Systematic model errors (i.e., biases) should always be removed from a SREF since they are a large part of forecast error but do not contribute to forecast uncertainty. A grid-based, 2-week, running-mean bias correction was shown to improve FP skill through: 1) better reliability by adjusting the ensemble mean toward the verification's mean, and 2) better resolution by reducing unrealistic ensemble variance.