Frequency, intensity, and sensitivity to sea surface temperature of North Atlantic tropical cyclones in best-track and simulated data

Abstract

Synthetic hurricane track data generated from a downscaling approach are compared to best-track (observed) data to analyze differences in regional frequency, intensity, and sensitivity of limiting intensity to sea surface temperature (SST). Overall, the spatial distributions of observed and simulated hurricane counts match well, although there are relatively fewer synthetic storms in the eastern quarter of the basin. Additionally, regions of intense synthetic hurricanes tend to coincide with regions of intense observed hurricanes. The sensitivity of limiting hurricane intensity to SST computed from synthetic data is slightly lower than sensitivity computed from observed data (5.5 ± 1.31 m s[superscript −1] (standard error, SE) and 8.6 ± 1.57 m s[superscript −1] (SE), respectively); however, the synthetic data produce sensitivity values that are much closer to the observed values than those obtained from two global climate models (GCMs) in a previous study. Despite a close match in the magnitude of basin wide sensitivities, the spatial variability of sensitivities do not match. These values tend to be highest in the western portion of the basin for the observed data, while the opposite is true for the synthetic data.