A51P-0304
Assessing Tropical Cyclone Forecast Skill in the GEFS Reforecast Version 2

Friday, 18 December 2015
Poster Hall (Moscone South)
Weiwei LI1, Zhuo Wang1 and Melinda Peng2, (1)University of Illinois at Urbana Champaign, Urbana, IL, United States, (2)Naval Research Lab Monterey, Marine Meteorology, Monterey, CA, United States
Abstract:
The prediction skill of tropical cyclone (TC) activity in the Global Ensemble Forecasting System (GEFS) Reforecast Version-2 is evaluated. The GEFS captures the seasonality of TC genesis (TCG) reasonably well over the Atlantic. The performance of the model varies over different basins, and quantitative errors are found in the spatial distribution of TCGs. The biases in TC tracks are largely attributed to the biases in the TCG and the the large-scale steering flows. Analyzing the genesis potential index (GPI) shows that the biases in the monsoon trough and subtropical high induce the TC biases in the western North Pacific, while the southward displacement of the ITCZ leads to errors in TCG over the eastern North Pacific. The over-prediction of TCGs near the West African coast is associated with the hyperactive tropical easterly waves over the African continent. The diabatic heating rate (Q1) field in the GEFS forecasts suggests a much deeper and stronger convection than in the ERA-Interim, which can be attributed to the deficiency of model cumulus scheme. The study suggests that the better prediction of the large-scale circulation and synoptic-scale waves can help improve the TC forecasts.