Applying Bayesian Maximum Entropy to extrapolating local‐scale water consumption in Maricopa County, Arizona

Abstract

Understanding water use in the context of urban growth and climate variability requires an accurate representation of regional water use. It is challenging, however, because water use data are often unavailable, and when they are available, they are geographically aggregated to protect the identity of individuals. The present paper aims to map local‐scale estimates of water use in Maricopa County, Arizona, on the basis of data aggregated to census tracts and measured only in the City of Phoenix. To complete our research goals we describe two types of data uncertainty sources (i.e., extrapolation and downscaling processes) and then generate data that account for the uncertainty sources (i.e., soft data). Our results ascertain that the Bayesian Maximum Entropy (BME) mapping method of modern geostatistics is a theoretically sound approach for assimilating the soft data into mapping processes. Our results lead to increased mapping accuracy over classical geostatistics, which does not account for the soft data. The confirmed BME maps therefore provide useful knowledge on local water use variability in the whole county that is further applied to the understanding of causal factors of urban water demand.