A New Kind of Science

nationwide data set of losses from 1975 to 1998 was compiled to assess the trends. Temporal patterns of deaths and injuries, monetary damages, and—in some cases—the number of events are systematically examined by year in chapter 5, and the authors undertake a systematic spatial assessment of the statewide totals in chapter 6. Explanations for some of the patterns are offered, particularly for the most significant disasters and for the states with most events or the greatest losses. Further refinement and evaluation of patterns of economic losses and death are undertaken by normalizing losses by population, land area, and gross domestic product (GDP). The authors advance the discussion from simple descriptions of loss patterns to explanations of the patterns of disaster-loss burden, and some surprises emerge from the arithmetic. For instance, North Dakota, Iowa, andMississippi not only suffered the greatest monetary losses per capita during the period, but also suffered the greatest losses of property and crops compared to their state GDP!For afinal analysis, the authors created an overall hazard score (averaged proportion of the states’ contributions to the national totals of events, deaths, and damages) and used it to rank the states. Using this ranking, states were assigned to categories of ‘‘proneness,’’ from highest (Florida, Texas, andCalifornia) to lowest (Rhode Island, Delaware, Alaska and other small or lightly populated states). The conclusion we are to draw is that the amount of loss a state has experienced indicates its disaster proneness. Finally, ‘‘Charting a Course for theNext Two Decades’’ by Cutter describes what is needed to produce the models and data appropriate for mitigation and planning assessments. In order for an effective assessment of events and losses to occur, progress is required in several areas: development of vulnerability science, the creation of a national hazard events and losses database, and the establishment of a national loss inventory and events clearinghouse. To do so, Cutter argues, we need to rethink thewaywe monitor, assess, andmanage our vulnerabilities. She briefly describes the shifts needed in data gathering and provision, sustainability and distributive justice, strategic planning, research funding, and societal awareness of issues that influence the prospects for disaster. While American Hazardscapes is intended to provide a broadunderstanding of the geography of loss due to hazards in the United States, it suffers from its openly acknowledged limitations. Though criticizing the quality of currently available data, the authors use those data to indicate the prospects for future disasters. The elimination of extreme events is no longer believed tobe the key loss reduction. Instead,we must identify and avoid places too dynamic for permanent occupation and adjust to the inevitable events in ways that limit prospects for loss. Mitigation must address the vulnerabilities that cause greater exposure and profound upset of our social systems and create more complex catastrophes. The data employed in this assessment describe (however imperfectly) the losses suffered over two and a half decades. The largest disasters overwhelm the patterns of loss in their analysis. The authors imply, based on proneness rankings, that those who lost the most are the most prone to loss. But in reality, losses are byproducts of the interplay of two dynamic geographies: the pattern of extreme events and the pattern of human use of the landscape. The former is often poorly understood, may not behave consistently, andmay operate on greater than twenty-five-year cycles. The latter may change so rapidly that it surpasses our capacity to measure it and map it, and postdisaster land use and human perception may be radically changed. These geographies were outside the scope of this book, however, and given new homeland security efforts and reorganization of the Federal Emergency Management Agency, the past is an even poorer indicator of the future.