"Hurricane Matthew has left at least 1000 Haitian dead, tens of thousands homeless, and millions in dire need of assistance. In the U.S., it has led to 19 deaths, left millions without power, and caused massive floods in North Carolina.
We are reminded once again how powerful and destructive natural hazards can be. People in Haiti will face a long recovery. Billions of economic damages will be inflicted by Hurricane Matthew in southeastern U.S. The vivid memory of this hurricane may stay with many people who have witnessed its monstrous presence at first hand. This once-in-a-life experience may also determine many people's judgments on the changing pattern of hurricane strength.
My coauthors and I in our paper on perceptions of changing hurricane strength found that maximum wind speed from the last landfall is the most powerful predictor of an individual's perception of changes in hurricane strength among the characteristics associated with the last hurricane landfall (Shao et al. 2017). Coastal residents who experienced higher maximum wind speeds are more likely to think hurricanes are becoming stronger. It appears that maximum wind speed tends to leave the deepest impression in coastal residents' memories, which includes impressive surf along the beachfront, as well as swaying and felled trees, and flying debris. Winds can be very destructive. The Saffir-Simpson categorization of hurricanes is therefore based on wind speeds. However, the biggest threat to life and property is actually water, not wind. Storm surge is a bigger contributor to deaths compared to winds. Super Storm Sandy is a perfect example. It was identified as a Category 1 storm. However, its enormous size and high storm surge had incurred tremendous amounts of property damages in the densely populated east coast.
So, once again, there is a mismatch between perceived and actual risks.
Shao, W., Xian, S., Keim, B. D., Goidel, K., and Lin, N. 2017 “Understanding perceptions of changing hurricane strength along the U.S. Gulf Coast” International Journal of Climatology. DOI:10.1002/joc.4805