Another Look at the Use of Maximum Potential Intensity of Tropical Cyclones

Several approaches have been taken to understanding and quantifying an upper bound of tropical cyclone (TC) intensity, or maximum potential intensity (MPI). Empirical studies have shown a direct relationship between tropical cyclone maximum intensity and sea surface temperature (SST). Theoretical approaches (e.g., Emanuel 1987) have found that additional thermodynamic factors, such as outflow temperature and atmospheric relative humidity, play an important role in defining the upper bound of intensity. Both the empirical and theoretical models have been compared to observations and provide reasonable upper limits to TC intensity. Ideally, theoretical models of MPI are preferred for use in most applications because they provide analytical solutions that consider both atmospheric and oceanic thermodynamic conditions, and are applicable to future climate scenarios. However, recent numerical studies have found that simulated tropical cyclones can exceed theoretical MPI by as much as 50%. Understanding this infrequent yet undesirable model behavior provides motivation for this work.

This study revisits two types of MPI models in the context of hurricane intensity forecasting. The Emanuel analytical model MPI is calculated for Atlantic and E. Pacific TC cases from 1982-2010 and compared to the empirically-predicted (i.e., based on SST) values using the inputs for the Statistical Hurricane Intensity Predictions Scheme (SHIPS). Approximately 7% (13%) of the observed intensities of the Atlantic (E. Pacific) cases tested were found to exceed the theoretical MPI as calculated, as compared to less than 1% for the empirically-derived MPI. This result is not surprising, however, since some of the cases in the sample set likely violate one or more of the assumptions used to derive the analytical model. The storm properties and environmental conditions of these cases will be discussed and a set of criteria for objectively identifying cases likely to exceed theoretical MPI will be proposed. Based on these findings, a new combined empirical-theoretical MPI parameter is developed. This parameter, along with an evaluation of its performance in SHIPS, will be presented. Recommendations for future use will also be offered.

DISCLAIMER: The views, opinions, and findings in this report are those of the authors and should not be construed as an official NOAA and/or U.S. Government position, policy, or decision.