How Many Lives Have Been Saved By Improved Hurricane Forecasts?

Trends In The Ratio Of Damage To Deaths Caused By United States Land-Falling Hurricanes

24th AMS Conference on Hurricanes and Tropical Meteorology (2000)

Chris C. Robbins and James Gross

National Hurricane Center, Miami, Florida

We have estimated that, if the pre-reconnaissance trend continued at the time of Hurricane Andrew, approximately 4,354 deaths may have occurred instead of “only” 23.

1. Introduction

Hurricane forecasting appears to have its origins during the middle nineteenth century, as early as 1847 (Sheets 1990). During the twentieth century, hurricane forecasting has evolved into a true science thanks to the establishment of various observational, numerical modeling and communications systems. Trends in the number of deaths attributed to land-falling hurricanes suggest that these systems have significantly enhanced our ability to forecast hurricanes. Specifically, as the coastal populations have increased during the past century, deaths have actually decreased. The sections that follow discuss relationships between observed trends in the hurricane damage-to-death data and technological advances made in the science of hurricane forecasting.

2. Damage And Deaths Data

Hurricane damage estimates for the period 1900-1989 are shown in Figure 1. These figures are damages adjusted to 1996 dollars (Hebert, et al. 1997). This can be compared to Figure 2, which shows total population in 10-year increments for all coastal counties from Texas to Maine (Jarrell, et al. 1992). Since damage is a function of not only population, but also wealth (Pielke, Jr. and Landsea 1998) and hurricane frequency, there is not necessarily a direct correlation between the two figures. However, it is assumed for this study that as coastal population increases, damage from land-falling hurricanes also increases. Further, as coastal population increases, the number of casualties would also be expected to increase. However, trends in hurricane-related deaths suggest that the opposite is true; as damage increased during the twentieth century, deaths decreased (Figure 3).

Figure 1. Total hurricane damage in 10-year increments — **Figure 1.** Total hurricane damage in 10-year increments

Figure 2. Total population of all coastal counties from Texas to Maine. — **Figure 2.** Total population of all coastal counties from Texas to Maine.

3. Damage-to-Death Ratios

Of particular interest is the trend noted in the damage-to-death ratios computed for each year during the twentieth century. The ratio is nearly constant between 1900 and 1942. However, in the early 1940’s an upward trend begins as deaths begin to decline while damage continued to increase. For clarity, these same data were converted to a logarithmic scale with an exponential trend line superimposed (Figure 5).

This follows the methodology of Doswell et al. (1999) who used tornado damage-to-death ratio trends to justify advancements in tornado forecasting. The upward trend in the hurricane damage-to-death ratios during the early 1940’s is consistent with the establishment of Air Reconnaissance effort, which was formally established in 1944 (Sheets 1990) and continues today.

**Figure 3.** Total deaths in 10-year increments resulting from land-falling hurricanes.

4. Conclusions

We have estimated that, if the pre-reconnaissance trend continued at the time of Hurricane Andrew, approximately 4,354 deaths may have occurred instead of 23. This is entirely speculative, of course, and should not be taken literally; however, a rough estimate on the order of 1,000 lives spared is not unreasonable. Therefore, it is strongly suggested that trends in the damage-to-death ratios are linked to the vast technological advances made during the twentieth century. The establishment of air reconnaissance flights into hurricanes may have been the turning point in hurricane forecasting since these give precise location and condition information to hurricane forecasters who use the information to make forecasts. The continued upward trend in the damage-to-death ratios is likely a result of all advancements (e.g., television, radio, satellite, numerical modeling) subsequent to the establishment of air reconnaissance.

Figure 4. Hurricane damage-to-death ratios computed for each year 1900 to 1996. Linear regression lines are superimposed for the period 1900-1942 and 1943-1996. The break was made to correspond to the establishment of air reconnaissance in 1943. — **Figure 4.** Hurricane damage-to-death ratios computed for each year 1900 to 1996. Linear regression lines are superimposed for the period 1900-1942 and 1943-1996. The break was made to correspond to the establishment of air reconnaissance in 1943.

Figure 5. Same as Figure 4 except converted to logarithmic scale with exponential regression lines superimposed. — **Figure 5.** Same as Figure 4 except converted to logarithmic scale with exponential regression lines superimposed.

Hurricane Andrew on August 23, 1992 (1230 UTC) *(Not included in original publication)*

5. References

Doswell, C. A., A. R. Moller, and H. E. Brooks, 1999: Storm spotting and public awareness since the first tornado forecasts of 1948. Wea. Forecasting, 14: 544-557.

Hebert, P., J. Jarrell, and M. Mayfield, 1997: The deadliest, costliest, and most intense United States hurricanes of this century (and other frequently-requested hurricane facts). NOAA, Technical Memorandum NWS TPC-1, 30pp.

Jarrell, J., P. Hebert, and M. Mayfield, 1992: Hurricane experience levels of coastal county populations from Texas to Maine. NOAA, Technical Memorandum NWS NHC-46, 152pp.

Pielke Jr., R. A., and C. W. Landsea, 1998: Normalized hurricane damages in the United States: 1925-1995. Wea. Forecasting, 13: 621-631.

Sheets, R. C., 1990: The National Hurricane Center-past, present, and future. Wea. Forecasting, 5: 186-232.