Exurban development is a continuing trend in America. The dilemma for open space conservation in exurbia is that, while people move to exurbia for open space experiences and to be closer to environments that they perceive as natural, exurban development occupies open space and impacts the structure, function and dynamics of ecological systems. The aim of this dissertation is to provide information for policymakers and designers about salient issues relating to homeowner preferences for open spaces in home-buying choice so that future exurban development would maintain the long term availability of open space experiences, as well as the quality of ecosystem services provided by that open spaces.

Three issues are discussed in three Chapters. Chapter II considers how nearby open space, and differences among types of open spaces, may be related to exurban residents' home-buying choices. It suggests that open space plays an important role in exurban homeowners' home-buying choices. And exurbanites' have varied preferences for different types of open spaces, which should be designed/planed with different strategies.

Chapter III is a theoretical exploration of potential causal explanations for relationships between preference for having open space near one's home and actually having open space landuse/landcover (LULC) near one's home. It suggests that people's choice of living environment, may cause them to value nearby open space, rather than nearby open space inculcating human preference. And, maintaining forests and wetlands as a part of exurban landscapes can be part of strategic planning efforts that satisfy respondents' preference and simultaneously preserve open space as natural habitat.

Chapter IV investigates potentials and challenges of using landscape metrics to infer cultural values of landscapes. Without confirming the potentials, the study suggests that landscape metrics may not validly measure landcover characteristics that are related to some landscape cultural values for many possible reasons including: inherent limitations of metrics tested, construct differences between planimetric data and landscape experience, data resolution, etc.