Improving Tropical Cyclone Rapid Intensity Change Forecasts

The Statistical Hurricane Intensity Prediction Scheme (SHIPS) first became available in real time in 1991 for Atlantic tropical cyclones. Numerous improvements have been made to the SHIPS model over the last two decades, and the longer range forecasts have improved by about 20% over that time period. In the past several years, the SHIPS model and the related Logistic Growth Equation Model (LGEM) have been among the most accurate operational intensity models in terms of mean error statistics. One of the limitations of both SHIPS and LGEM, however, is that neither captures rapid intensification (RI) events very well. For this reason, the probabilistic Rapid Intensification Index (RII) was developed to supplement the deterministic SHIPS and LGEM forecasts. The RII has shown skill in the ability to identify tropical cyclones that rapidly intensified in the following 24 h.

Two methods are being developed to improve the performance of SHIPS and LGEM for RI cases. These include the utilization of the new NCEP reanalysis fields for SHIPS/LGEM model development and a modification of the growth rate parameter in LGEM based on the probabilities from the RII.

The new reanalysis fields will provide higher spatial resolution (0.5 deg. lat/lon) for evaluation of predictors and more consistent time series of thermodynamic variables. The current SHIPS database utilizes the older NCEP reanalysis fields (2.5 deg. lat/lon) from 1982-1999 and operational NCEP global model analyses for 2000-present. The bias characteristics of the relative humidity from the operational and old reanalysis fields are very different, which limits the influence of moisture in the SHIPS/LGEM prediction. The higher quality and more consistent input from the new reanalysis should improve the prediction of SHIPS and LGEM.

The LGEM is a time stepping model where the maximum wind grows at a rate that is estimated from environmental variable, but where the maximum wind is restricted to be between zero and an estimate of the maximum possible intensity. Given the relative success of the probabilistic RII, the growth rate in LGEM will be modified depending on the RII probability to better account for rapidly intensifying cases. The modified growth rate as a function of the RII probability will be determined empirically as part of the LGEM fitting procedure. Tests will be performed to determine if the modified versions of SHIPS and LGEM provide increased skill relative to the operational versions.

The views, opinions, and findings contained in this manuscript are those of the authors and should not be construed as an official National Oceanic and Atmospheric Administration or U.S. government position, policy, or decision.