Large, about 100 knots or 50 m/s, official National Hurricane Center (NHC) advisory intensity forecast errors during eastern North Pacific (ENP) Hurricane Linda (1997) clearly indicate an important tropical cyclone (TC) research need. Satellite sensors are almost exclusively used to specify and predict TC intensity in the otherwise data-sparse ENP. Therefore, data from satellite sensors is often the only observational information available to improve the understanding of and the specification and prediction of TC intensity. Rainfall measurements of TCs from passive-microwave satellite sensors are routinely available to NHC forecasters in near real time. The major diabatic heating source in TCs is rainfall and its concomitant latent heat release (LHR). Previous NWP and satellite observational studies have noted a lag time between changes in LHR and future TC intensity changes. Therefore, satellite rainfall measurements ought to be correlated to future intensity changes. If so, statistically significant correlations can be exploited to develop linear regression models to predict future intensity changes.
Using rainfall parameters from 251 observations of ENP TCs by Special Sensor Microwave/Imager (SSM/I) passive-microwave radiometers onboard the Defense Meteorological Satellite Program (DMSP) constellation of the National Polar-orbiting Operational Environmental Satellite System (NPOESS), low (i.e., r ~ 0.3-0.4), but significant, correlations between rainfall parameters and future intensity changes were found. By combining these significantly correlated rainfall parameters with climatic, weekly SST, persistence, and global NWP model data, multiple linear regression models to predict future intensity changes were developed. The resulting models explained about 60 percent of the variance of intensity change at 12-h intervals between 12 and 72 hours into the future. A homogeneous comparison of up to 46 cases at the 12- and 24-h forecast intervals found that a combination of persistence, climatic, weekly SST, and rainfall-parameter data outperformed: official NHC forecasts; Statistical Hurricane Intensity FORcasting (SHIFOR) guidance; guidance from the climatic and persistence method of Hobgood; and, guidance from the climatic, persistence, weekly SST, and global NWP model method of Petty. Mean absolute forecast errors were 2.8 and 5.0 m/s for the 12- and 24-h forecast intervals, respectively. The corresponding relative absolute forecast errors were three and 14 percent less than official NHC forecasts. This research demonstrated that satellite microwave rainfall information could improve ENP TC intensity forecast guidance products for the 12- and 24-h forecast intervals