The growing inadequacy of an open-ended Saffir-Simpson hurricane wind scale in a warming world

Abstract

Global warming increases available sensible and latent heat energy, increasing the thermodynamic potential wind intensity of tropical cyclones (TCs). Supported by theory, observations, and modeling, this causes a shift in mean TC intensity, which tends to manifest most clearly at the greatest intensities. The Saffir-Simpson scale for categorizing damage based on the wind intensity of TCs was introduced in the early 1970s and remains the most commonly used metric for public communication of the level of wind hazard that a TC poses. Because the scale is open-ended and does not extend beyond category 5 (70 m/s windspeed or greater), the level of wind hazard conveyed by the scale remains constant regardless of how far the intensity extends beyond 70 m/s. This may be considered a weakness of the scale, particularly considering that the destructive potential of the wind increases exponentially. Here, we consider how this weakness becomes amplified in a warming world by elucidating the past and future increases of peak wind speeds in the most intense TCs. A simple extrapolation of the Saffir-Simpson scale is used to define a hypothetical category 6, and we describe the frequency of TCs, both past and projected under global warming, that would fall under this category. We find that a number of recent storms have already achieved this hypothetical category 6 intensity and based on multiple independent lines of evidence examining the highest simulated and potential peak wind speeds, more such storms are projected as the climate continues to warm.

Keywords: Saffir-Simpson hurricane intensity scale; climate change; tropical cyclone.

Fig. 1.

(A) The five recent storms that reached our hypothetical category 6 wind intensity. The triangles denote which storms reached these intensities. Since windspeeds are recorded in discrete 5 kn bins, the triangles overlap at those shared LMIs. (B) Time series of the annual-maximum LMI in the western North Pacific historical (IBTracs) record of intensity (blue), and the bias-corrected data prior to 1973 (red). The black dashed line identifies our hypothetical nominal category 6 intensity.

Fig. 2.

(A) ERA5 1979 to 2019 average annual exceedance of category 6 wind speed threshold. Units: days. (B) Linear rate of change in ERA5 average annual exceedance of category 6 wind speed threshold from 1979 to 2019. Units: days per year. (C) Linear trends in average annual exceedance of the category 6 wind speed threshold from 1979 to 2019 averaged over 40S to 40N excluding 10S to 10N. blue: CMIP6 historical, red: CMIP6 piControl, black: ERA5 reanalysis. (D) Increase in the tail of the PI distribution indicating a 2.6× increase in the chances of PI exceeding category 6 from the first half of the ERA5 reanalysis to its second half over the region 40S to 40N excluding 10S to 10N.

Fig. 3.

Change in annual PI exceedance of the category 6 threshold since present day at various future global warming levels from perturbed ERA5. Upper Left: 1.5 °C above preindustrial. Upper Right: 2.0 °C above preindustrial. Lower Left: 3.0 °C above preindustrial. Lower Right: 4.0 °C above preindustrial. Units: Days.