摘要：Rural-urban fringe counties (i.e. counties adjacent to other counties with large and growing urban centers) often experience intense development pressure as a result of urban growth and expansion. While growth-initiated development can take many forms, the majority of the development that occurs in these exurban counties is in the form of single-family residential dwellings. Moreover, it is in the form of subdivision developments that range in size from very small two and three lot minor subdivisions to massive multi-phase major subdivisions with hundreds of lots and numerous amenities. In this paper, we focus on the land development patterns in Carroll County, MD, an urban-fringe county in the Baltimore-Washington D.C. metro region. The purpose of this paper is to explore whether a basic set of factors, both constant and time-dependent and acting at different spatial scales, can explain the timing and location of major versus minor subdivision developments in Carroll County. Using a micro-level panel of land parcel conversion, historical land records for subdivision development and ArcGIS software a new dataset was created that traces the entire history of the subdivision process in the county. Datasets were also created that trace the history of land preservation so that we could control for official open space and its interaction with the decision to subdivide through time. Using these data and a number of land use variables created from them from 1993-2007, we apply a competing risks duration model to analyze which factors affect major versus minor subdivision development. Visual inspection as well as a descriptive analysis of a series of landscape metrics based on distance from the metropolitan center reveals a different pattern outcome for small versus large subdivisions with larger developments following more closely to the predictions of the urban economic model. Empirically, we find further evidence that the factors affecting the timing of minor versus major subdivision developments are indeed different. Distance and access to road networks have less of an effect on minor over major developments, while surrounding preservation and the option to preserve have less of an effect on major subdivisions. To make the risk comparison relevant and to focus on areas that have experienced the most fragmentation as a result of residential land conversion as well as the most policy attention, we restrict our analysis to parcels located in minimum density zoning districts.
