Determinants of residential land-use conversion and sprawl at the rural-urban fringe

We estimate a probit model of residential land conversion using parcel-level data from a rural-urban county in Ohio. Spatial landscape pattern metrics are used to quantify land-use patterns and to link patterns of residential sprawl with factors estimated to influence parcel-level land conversion. Findings indicate that the location of new residential development is influenced by preferences for lower density areas that nonetheless are close to existing urban development. Combined, these forces have generated a moderated pattern of residential sprawl. Spatial error autocorrelation is controlled using a spatial sampling technique that enables consistent estimation of the probit model.