Pulwarty, R. S., 1999: Hurricane impacts in the context of climate variability, climate change and coastal management policy on the eastern U.S. sea-board. In Climate, Change and Risk, T. Downing, A. Olsthoorn, and R. Tol (Eds.), Routledge, 173-204.


The power of hurricanes has been amply demonstrated in recent years, as major storms have pounded Caribbean island nations and the US mainland. Hurricane Andrew, which struck south Florida on 24 August 1992, ranks as the most costly storm disaster in US history, with long-term estimates of losses exceeding $30 billion. Estimates of losses if Miami, just 30 km north, had been struck are in the range of $40-60 billion. A continuation of the same track would have put New Orleans under 5 to 6 m of water. The meteorological conditions that would create the difference in the two courses of movement (the actual course and a direct hit on Miami) are essentially undetectable with present observing systems (Sheets, 1994). In a careful analysis, Pielke and Landsea (1998) show that a repeat of the 1926 hurricane impacting S.E. Florida and Alabama would result in losses exceeding $70 billion (1995 $).

The exposure of human communities to natural phenomena has been increasing as a result of population growth and movements into areas of risk (White, 1945; Burton and Kates, 1964; Pielke and Pielke, 1997; Pulwarty and Riebsame, 1997, among others). The result has been an increase in economic losses in recent decades. It remains crucial to identify the relative contributions of physical, social, economic and political factors to these losses. Glantz and Price (1994) point out that regional and national institutions that develop strategic plans for disaster preparedness often do not take into consideration the possible implications of climate change, or at least climate variability on greater than year-to-year time scales. This chapter seeks to (1) evaluate some of the changes in hurricane frequency that have occurred over the past century, (2) assess the state of knowledge of the effect of climate change on hurricane intensity, frequency, and damage potential, and (3) place these changes in the context of important socioeconomic trends and the evolution of coastal policy on the eastern seaboard of the USA. In particular, the chapter highlights not only the changing level or risks related to the tremendous growth in coastal population throughout this region, but the importance of understanding the nature of such changes. In addition to hurricanes and tropical storms, the east coast is subject to northeasters, which track down the coast from the north-east. While these storms can be devastating because of their slow movements and long lifetimes, our focus here is strictly toward hurricanes.

Three factors are usually identified for study in assessments of the hurricane hazard: (1) weather generators specifying the frequency, magnitude, and other characteristics such as storm surge and wind speed; (2) the population-at-risk (exposure), specifying the number and geographic distribution of people and buildings in areas prone to hurricane impact; and (3) vulnerability, commonly specified as the susceptibility of the population-at-risk to injury or damage when an event of a given severity occurs (e.g. Friedman, 1977). The number of people in a particular place indicates the degree of exposure, and the likelihood of a particular event in a particular location represents the probability of occurrence. Exposure, as a component of vulnerability, was identified as early as 1964 (Burton and Kates, 1964) and in numerous publications by Gilbert White and others (see Berke et al., 1993; Jarrell et al., 1992).

Adaptation, in varying degrees, is key to the human ecology of vulnerability to natural disasters. Four forms of adaptation have been described in the hazards literature (Alexander, 1991):

  1. persistent occupation of the hazard zone despite the risk involved: (a) with comprehensive measures for risk mitigation and hazard abatement, (b) with only warning and evacuation measures, or (c) without any protection measures (the state of maximum vulnerability);
  2. co-habitation with the damage caused by past disasters;
  3. abandoning damaged or destroyed structures but relocating within the risk zone;
  4. migration to safer zones (a) planned or (b) unplanned.
While these have been identified for a number of years (see Burton et al., 1993), we are still bound by the question raised by White (1945) in the context of floodplain management: "Why are certain adjustments to the risk of floods preferred over others, and why, despite investment in those adjustments, are societal losses increasing?" (see also Mileti et al., 1995).

Because of the great potential for hurricane-caused destruction of property and loss of life, the science of hurricane forecasting has always been intimately connected with preparedness for natural disasters, and hurricane forecasters have a long history of working closely with emergency management authorities, usually on an event-by-event basis (Pielke and Pielke, 1997). The trend towards greater urbanisation in hurricane-prone areas, growth of ageing populations, the necessity for environmental protection, and other changes in land-use practices suggest that the threats posed by conceivable variations in hurricane activity should be continually reappraised. It is noted, however, that methods for collecting data on social vulnerability and for merging that information with data on physical risks and environmental vulnerability are at an early and evolving stage of development (Mather and Sdasyuk, 1991; Pulwarty and Riebsame, 1997).

This review draws heavily on Diaz and Pulwarty (1997) and studies therein, Platt (1994); Beatley et al. (1995), Clark (1992), Popkin (1990) and the disaster mitigation literature in general. A review of this literature in the context of vulnerability assessment and sustainable development is provided by Pulwarty and Riebsame (1997) and Anderson (1995).

The major questions addressed in this study are:

  • What features of climate control the frequency and intensity of extremes, i.e. in this case Atlantic hurricanes?
  • How have these features varied in the past and how are they expected to change under climate change scenarios?
  • Can we expect increased hurricane activity in the coming decades to levels that were more prevalent during earlier periods?
  • What is the damage potential of past hurricanes under present-day conditions?
  • Do the regional and national level changes support the assumptions concerning event frequency and population density that have become the basis for strategies of preparedness and preventative planning?