Current global and national databases that monitor losses from natural hazards suffer from a number of limitations, which in turn lead to misinterpretation and fallacies concerning the “truthfulness” of hazard loss data. These biases often go undetected by end users and are generally a product of the type of information stored in loss databases and how they are constructed. This paper highlights some common shortcomings and root causes for data misinterpretation by asking what biases are present in existing databases and how these then manifest themselves in actual loss figures. For illustrative purposes, four widely used, nonproprietary, Web-based hazard databases are examined: the international Emergency Events Database (EM-DAT), the international Natural Hazards Assessment Network (NATHAN), the Spatial Hazard Events and Losses Database for the United States (SHELDUS), and the National Weather Service's Storm Events. We identify six general biases: hazard bias, temporal bias, threshold bias, ac... Abstract Current global and national databases that monitor losses from natural hazards suffer from a number of limitations, which in turn lead to misinterpretation and fallacies concerning the “truthfulness” of hazard loss data. These biases often go undetected by end users and are generally a product of the type of information stored in loss databases and how they are constructed. This paper highlights some common shortcomings and root causes for data misinterpretation by asking what biases are present in existing databases and how these then manifest themselves in actual loss figures. For illustrative purposes, four widely used, nonproprietary, Web-based hazard databases are examined: the international Emergency Events Database (EM-DAT), the international Natural Hazards Assessment Network (NATHAN), the Spatial Hazard Events and Losses Database for the United States (SHELDUS), and the National Weather Service's Storm Events. We identify six general biases: hazard bias, temporal bias, threshold bias, ac...

(2006). The Long Road Home: Race, Class, and Recovery from Hurricane Katrina. Environment: Science and Policy for Sustainable Development: Vol. 48, No. 2, pp. 8-20. doi: 10.3200/ENVT.48.2.8-20

In the United States, direct losses from natural hazards are on the rise with hurricanes, flooding, and severe storms contributing about three quarters of the total damages. While losses from severe storms have been stable over the past fifty years, hurricane and flood losses have tripled. Per capita losses are also increasing showing that impacts outpace population growth with high per capita losses occurring largely in the Southeast and Midwest. If the loss escalation of the past two decades continues into the future, then direct losses of $300 to $400 billion within a single decade are possible. In order to reverse this trend, sustainable development, vulnerability reduction, and hazard mitigation must become priorities and current loss reduction efforts need to be evaluated and re-assessed in terms of their effectiveness. These conclusions are drawn from the analysis of spatial and temporal trends in direct losses from natural hazards using SHELDUS^TM data from 1960 through 2009. Loss data are adjusted for inflation, population, and wealth to capture both trends in total losses and per capita losses. The loss data are then compared to disaster-related federal government and private insurance expenditures.

Emergency planning and response increasingly involve close interactions between a diverse array of actors across fields (emergency management, public health, law enforcement, etc.); sectors (government, non-profit and for-profit); and levels of government (local, state and federal). This article assesses the temporal dynamics of emergency management networks in two moderately sized communities that have served as large-scale disaster evacuation hosting sites in the past decade. The paper uses two strategies for tracking the evolution of these networks across time. First, we develop a network roster using newspaper and newswire data sources across a decade. Second, we develop a view of the evolution of the networks by analysing emergency operations plans for each community. Analysis of data reveals a contrast between a core set of consistent (mostly governmental) actors and a peripheral set of rapidly turning over (mostly non-governmental) actors – though the account depends on the mode of data on which one focuses. The article concludes with a discussion of the advantage presented by having a two-tier network for evacuation hosting that mixes core and periphery across multiple sectors.