Date of Completion

5-11-2013

Embargo Period

11-16-2013

Keywords

Environmental Kuznets Curve, Open Space, Northeastern United States, Geographically Weighted Regression, Quantile Regression

Major Advisor

Dean M. Hanink

Associate Advisor

Robert G. Cromley

Associate Advisor

Chuanrong Zhang

Associate Advisor

Daniel L. Civco

Field of Study

Geography

Degree

Doctor of Philosophy

Open Access

Campus Access

Abstract

The environmental Kuznets curve (EKC) is a derivation of the curve developed by Simon Kuznets in the 1950s. The original Kuznets curve describes the relationship between inequality and per capita income in countries while the environmental Kuznets curve is used to describe the relationship between environmental characteristics and per capita income. Since the early 1990s the EKC has become a widely used empirical model for exploring the relationship between economic growth and environmental characteristics, including land use/cover. This dissertation extends EKC-related research theoretically in the form of a conceptual model linking the non-regenerative land cover types of agriculture and open space to per capita income when considered in an urban, exurban, and rural framework. The non-regenerative land cover model concerns a class of land covers best analyzed in a spatial cross-section.

A study region within the northeastern United States using the county as the spatial unit of analysis, a finer spatial scale than used in most EKC studies, is used in the empirical investigation. The relationships between per capita income and a variety of regenerative and non-regenerative land cover types, including forest cover, wetland cover, agricultural cover, and open space, are examined within the study region Empirical tests are conducted using a variety of regression analyses to determine if the tested relationship conforms to the appropriate EKC income polynomial. Global regressions in the form of ordinary least squares and spatial error models are used to provide baselines for local test comparisons using geographically weighted regression and quantile regression. Global and local regression analyses are used in determining the functional form, linear or polynomial, of the relationship between land cover and per capita income. In general the results of this research support an EKC across the variety of land covers examined.Archival abstract submitted

Download
COinS