Citation

Hondula DM, Meltzer S, Balling RC, Iñiguez P. Bull. Am. Meterol. Soc. 2022; 103(9): E2017-E2031.

Copyright

(Copyright © 2022, American Meteorological Society, Publisher Allen Press)

DOI

10.1175/BAMS-D-21-0069.1

PMID

unavailable

Abstract

Public heat alerts are important risk communication tools, but there has been no systematic analysis of how frequently they are issued or how patterns in alert frequency relate to regional climatology or heat-health impacts. We compiled and analyzed all excessive heat warnings and heat advisories (collectively, heat alerts) issued by the U.S. National Weather Service for 2010-19. Heat alert frequency was correlated to climatological indicators derived from reanalysis data aggregated to Weather Forecast Office (WFO) polygons and to estimates of heat-attributable mortality for 134 metropolitan areas. The type of heat alerts used and the frequency with which they were issued were highly variable. Across 77% of the country, heat advisories were the primary product issued. The median location experienced 2.3 heat alert days per year. Regions with the highest frequency (approaching 25 heat alert days per year) included the southern Midwest and Great Plains, as well as the desert Southwest. The 95th-percentile daily maximum heat index was the climatological indicator most strongly correlated with heat alert frequency across all WFOs (r = 0.71). Locations that issued heat alerts more frequently than would be expected based on climatology were primarily located along the Pacific coast; those that issued heat alerts less frequently than expected were in southern Texas and southern Florida, the latter of which includes multiple cities with high rates of heat-attributable mortality. Our results suggest that the public may be receiving mixed signals about the severity of the heat hazard, with some hotter locations particularly underserved by heat risk messaging.

Language: en