Observational Scale and Modeled Potential Residential Loss from a Storm Surge

Two geographically related questions with regard to hurricane-induced storm-surge impacts were investigated: (1) What observational scale of analysis is appropriate? (2) Is the effect of observational scale on model results predictable? These two research questions were investigated in the context of storm surge-induced impacts to single-family residential structures in Florida. The study was conducted for 21 coastal counties in Florida at five spatial scales of analysis: parcel, block, block group, tract, and county. The research findings reveal a monotonically decreasing relationship between predicted standardized residential loss (the ratio of predicted loss at scale X and the predicted loss at parcel scale) and the observational scale of analysis. This monotonic relationship was consistent for most Florida counties, primarily due to the notable spatial distribution of housing units and proximity to the coastline.