Has Climate Change Affected US Hurricane Damage Yet?

by Dr. Hugh Willoughby

Abstract

From 1900 through 2021, US nominal hurricane damage (cost when incurred, without adjustment) increased 9% yr−1 as US Gross Domestic Product (GDP) increased 6.45% yr−1. Published analyses contend that “normalized” past hurricane losses, adjusted to the present for inflation, local population, and individual wealth yield zero trend (e.g., Weinkle et al. 2018). Fitting an exponential to the normalized data binned by pentads yields a growth rate of 0.94% yr−1 (p = 0.0932), and the same procedure applied to damage adjusted with the GDP growth rate yields growth rate at 2.44% yr−1 (p < 0.001). The key difference between these approaches is that normalization responds to local population changes, but GDP detrending responds to national ones. These results suggest that ~1.5% yr−1 of the change in the GDP adjusted damage resulted from more rapid development on hurricane-prone coasts and 1% yr−1 resulted from climate change. Growth at 1% yr−1 is equivalent to a factor > 3 over 121 yrs. The statistical distributions of damage by landfall are approximately log normal. The log mean of the normalized distribution corresponds to $0.9B ($0.9 x 109) in 2018 currency, and range between plus and minus one standard deviation is $0.07B to $12B. The corresponding values for GDP detrended damage are $0.7B and $0.06B to $8B. The figures for annually aggregated damage are four or five times larger. The annual distributions tend to be platykurtic and negatively skewed because they represent a superposition of more and less active seasons. Pareto distributions fitted to the tails of the annual data indicate that seasonal impacts > U$ 400-500 billion with return periods of two centuries are a disturbing possibility.