A significant challenge faced by tropical cyclone (TC) modelers and the forecasting community is the prediction of intensity change. The National Hurricane Center (NHC)’s forecast verification shows large improvement in TC track forecasts over the last decade. However, intensity forecasts remain a challenge. While some improvements in TC intensity forecasting have been demonstrated, predicting large changes in TC intensity remains very problematic. This is particularly true for identifying episodes of rapid intensification (RI) or rapid weakening (RW), which remains the highest operational forecasting priority of the NHC. The multi-scale nature of the problem with environmental, oceanic, and inner-core processes all likely playing important roles in determining if and when a TC will undergo RI increases the difficulty. The deficiencies of models to forecast these large changes leads to large intensity errors and contributes to the problem of intensity forecasting. This study establishes benchmarks for model performance for large changes in intensity, addressing both RI and RW. Additional analyses are included to document the characteristics of the forecasting errors. The primary focus is the HWRF model, though forecasts from COAMPS-TC and NHC are compared and contrasted.

For this study forecasts from HWRF, COAMPS-TC and NHC were evaluated for the Atlantic (AL) and eastern North Pacific (EP) basins. The HWRF operational forecasts were also evaluated for the western North Pacific (WP) basin.

The RI/RW capabilities within the Model Evaluation Tools - Tropical Cyclone (MET-TC) package have been upgraded to include the more advanced evaluation methods used in this study. These capabilities will be available in METv5.1, coming in October 2015.

Contact information:
Mrinal Biswas (biswas@ucar.edu)