Ensemble-based prediction and diagnostics during the PREDICT field experiment

During the NSF-sponsored PRE-Depression Investigation of Cloud-systems in the Tropics (PREDICT) field experiment, which took place between August 15th and September 30th 2010, a variety of synoptic-scale and mesoscale products based on ensemble forecasts were disseminated. Ensembles comprised around 200 simultaneous global model forecasts from 9 operational centers, as part of the THORPEX Interactive Grand Global Ensemble (TIGGE), and a higher resolution, data-assimilative Weather Research and Forecasting ensemble initialized using the Ensemble Kalman Filter (WRF/EnKF). With the primary focus of PREDICT being to test the 'marsupial' theory on how clusters of thunderstorms organize into tropical depressions, ensemble mean and probabilistic measures of low-level circulation and thickness were provided to yield quantitative estimates of genesis. Other relevant ensemble-based products included mean fields and probability distributions of vertical wind shear, relative humidity, low-level vorticity, Okubo-Weiss parameter, and upper-level divergence. Additionally, a variety of ensemble sensitivity guidance products was produced to diagnose the influence of initial environmental perturbations on the developing tropical disturbance. The utility of all these ensemble products during the field experiment will be presented, together with preliminary statistical evaluations of the respective ensembles' abilities to predict probabilistic metrics relevant to genesis. Refined hypotheses based on the findings to date will be proposed.