Using enhanced dasymetric mapping techniques to improve the spatial accuracy of sea level rise vulnerability assessments

摘要

This article reports on the application of dasymetric mapping techniques to interpolate and disaggregate block group population counts to smaller areal units (i.e., tax parcels) and derive surface population models with more realistic representations of population distributions in our residential study area in Miami-Dade, Florida. Three methods of dasymetric interpolation were tested: (i) binary, (ii) three-class, and (iii) limiting variable. Our enhanced limiting variable approach introduced an adjustment factor for parcel vacancy rates in the dasymetric calculations, and applied dasymetric mapping techniques to disaggregate future population projections to the tax lot level of analysis. The limiting variable interpolation generated the lowest coefficient of variation (0.188), followed by the three-class interpolation (0.645). We also found that population densities vary substantially within land use classes of single family, medium density and high density classes, and these variations also highlighted the importance of incorporating vacancy rates when interpolating population counts to categorical land use data. Overall, the enhanced dasymetric mapping technique is particularly useful for examining the impact of sea-level rise as its derivatives are compatible with high resolution LiDAR and orthoimagery data. Coastal counties can also benefit from such high resolution surface population models to enhance the accuracy of hazard-related vulnerability assessments and to guide the development of relevant shore zone conservation and adaptation strategies.