| Tropical cyclone losses in the United States have shown an increasing trend since the beginning of the 20th century. This is mainly due to increased exposure along America's coast. The amount of coastal property at risk persistently increases due to inflation, wealth increase, and population growth. When researchers have normalized the loss record to remove the influence of exposure and vulnerability change, no trend can be discerned in the damage record. This has been used to refute the claim that tropical cyclones are becoming more potentially destructive, and to keep the locus of explanation firmly in socio-demographic trends. But physical variation, in storm size, intensity and location, still make a significant difference in the impact of any individual storm event. This fact occasionally induces calls for renewed efforts at hurricane modification and routinely evokes a sense of either relief or alarm at "close calls" that, except for a difference of a few miles in landfall location or a modest weakening of peak winds, separate hurricane disasters from catastrophes. This project examined the effect of landfall location on storm damage using the Federal Emergency Management Agency's (FEMA) risk assessment model, HAZUS. Thirty-mile track shifts were prescribed for the top 10 most damaging storms in the normalized record since 1988. The alternate storms yielded drastically different damage estimates from the original storms, indicating large spatial variations in exposure. Each landfall shift resulted in a rank change in the overall normalized record. The damage record is dominated by individual extreme events like those used in this analysis, and although random, differences in landfall location would presumably average out in a long record. The fact that a few storms account for a large majority of losses, and that small differences in their landfall yield large differences in impact, points to a very large noise to signal ratio that would make it difficult to discern a climate-induced trend, and may also obscure some dimensions of socio-economic exposure and vulnerability trends. |
