ABSTRACT
Introduction The ability to understand and predict changes in land-use patterns as the result of individuals’ economic decisions is necessary for the effective design of environmental, public finance, and growth management policies. Changes in landuse pattern are the cumulative result of numerous individual decisions regarding the use of lands. Accordingly, the study of land-use change at a micro or individual scale provides for novel opportunities to understand the human behavior underlying these decisions and assess the effects of environmental, public finance, and growth management policies on these decisions. As many local and state governments in the U.S. grapple with increasing growth pressures, the need to understand the economic factors that influence individual choices regarding the use of land has taken on added urgency in recent years. In this chapter, we describe an empirical study of residential land-use change in Maryland at the rural-urban fringe. Our description focuses on the estimation of a micro-economic model of land-use change, giving particular attention to the spatial aspects of the model as well as its predictive capabilities. The rural-urban fringe begins where suburbs end, and extends into rural areas. At the rural-urban fringe, changes in land use often coincide with transitions from traditional, rural communities to more developed, urban communities. Nelson (1992) and Daniels (1999) refer to these communities in transition as ‘exurban’ areas. Between 1960 and 1990, population in exurban counties increased by 60 million people, which accounted for over 25 per cent of the population growth within this time period (Nelson 1992). Coupled with this growth are evolving patterns of low-density development that have resulted in an increasingly low-density and sprawled land-use pattern. The pattern of land-use change in Calvert County, Maryland, one of the fastest growing exurban counties in Maryland, is typical of this growth. Between 1981 and 1997, this county experienced a 94 per cent increase in population and a 191 per cent increase in the
number of acres in low-density residential use. Because exurban areas have outpaced urban and suburban areas in population growth for the last several decades, growth pressures are commonly observed at the rural-urban fringe. The rate and extent of development in these areas has raised public concerns regarding issues such as loss of open space and agricultural lands, traffic congestion, and crowding in schools. Models such as the one developed here provide a means to inform the development of policies aimed at managing these pressures by providing a tool to assess the effects of alternative policies and to ascertain the influence of different factors on land-use choices at the individual parcel level. In short, these models enable communities to consider ‘what if’ types of questions and to better anticipate the location and timing of future development. In this chapter, we develop a spatially explicit economic model of land-use change suitable for addressing changes in land use at the rural-urban fringe. As their name suggests, spatially-explicit models emphasize spatial relationships. Hence, a spatially explicit model of land-use change devotes specific attention to the absolute and relative locations of land-use changes, variation in characteristics of the landscape over space, and potential interdependencies between decisions over space. Spatially explicit economic models were introduced in Chapter 7, which provides an overview of different empirical methods used to model land-use change. Recent advances in data and computing, notably Geographic Information System (GIS) data and modeling tools, have fostered the evolution of spatially explicit models in a variety of disciplines, and the evolution of spatially explicit models of land-use change will certainly be influenced by future advances in spatial computing and data availability. We use a duration modeling framework, as reviewed in Chapter 7, which allows us to better capture how the cumulative effects of changes in variables over time influence future land-use decisions and the timing of land conversion. Because the timing of land conversion is often of great interest to communities, the duration modeling framework offers an appealing perspective from which to model land-use change. Our ability to address the spatial heterogeneity of the landscape over time is the direct result of having access to parcel-level land use data. These data enable us to track the characteristics of land parcels over time and space and permit a richer categorization of rural land-use change than that afforded by publicly available macro-scale land use data. We take a spatially disaggregated approach to modeling residential land-use change that accounts for the spatial heterogeneity of policies (e.g., zoning) and landscape features (e.g., slope, soil type, locational amenities) that influence individual land-use decisions. The residential land-use change model described here focuses on the conversion of undeveloped lands such as agricultural, forest, and open-space lands to residential use. This subset of landuse changes is of greatest concern for local and state policymakers interested in changes at the rural-urban fringe. We do not explicitly deal with commercial development, which makes up a relatively small proportion of developed uses in most exurban areas and typically follows residential development into these areas. However, we recognize that land-use changes associated with commercial development are also of interest to policymakers and citizens.
