The purpose of the work is to produce and refine a predictive model of floodplain land value changes over time and space by testing several theoretical models presently at different levels of validation, and develop an heuristic framework for the relationships between residential property values and the incidence of flooding. The literature from both natural hazards research and urban economics will be integrated and incorporated into the conceptual model. This model will be tested with data from a limited number of study sites, and the analysis will use multiple regression techniques, the hedonic approach and the "repeat sales" method (Palmquist). The results of the research will create a data base and an analytical framework useful to other researchers, decision-makers and floodplain managers.
