Spatial and temporal modeling of parcel-level land dynamics

Neighborhood and historical conditions are important factors in land dynamics. However, models that explicitly incorporate spatial and temporal dependencies face challenges in data availability, methodology and computation. In this research, parcel-level dynamics are investigated using the geocoded Auditor's tax database for Delaware County, Ohio, including 73,560 parcels over the period 1990-2012. A binary spatio-temporal autologistic model (STARM), incorporating space and time and their interactions, is used to investigate parcel-level dynamics. The results show that the model is able capture the impacts of contemporaneous and historical neighborhood conditions around parcels, as well as the effects of other variables such as distances to various facilities and infrastructures, agricultural and residential land-use shares within a half mile radius circle, and population density and growth expectation at the census tract level.