UVM Theses and Dissertations
Format:
Print
Author:
Howard, Ruth Hollister
Dept./Program:
Natural Resources
Year:
2012
Degree:
MS
Abstract:
Urbanization has created a complex transition zone between the urban core and rural landscapes. This zone faces environmental challenges that are distinct from those in urban and rural areas, yet no agreement exists on where this zone begins or ends. Existing classification systems of the urban-rural gradient fail to account for the transition zone in an ecologically meaningful way.
This project aims to: (1) define key indicators of transition zones within urban-rural gradients with an emphasis on environmental health; (2) develop a new method using cluster analysis and data reduction techniques to spatially define urban, rural, and transition zones, known as Urban-Rural Transition Zones (URTZ); and (3) implement this approach in the Baltimore-DC Corridor and demonstrate how it builds and differs from existing classification systems.
A suite of ecosystem health indicator variables were summarized by 1911 Census Block Groups for the Baltimore to Washington, DC corridor. A two-step cluster analysis was used to differentiate the zones of the urban -rural gradient in the study area. This analysis was repeated using many different combinations of variables, including both high and medium resolution land cover data.
Of the many resulting outputs, the preferred one was determined based on three best fit criteria. This map's five zones included urban, urban fringe, urban. transition, rural transition and rural. The indicator variables used in differentiating these zones included coverages of total vegetation, total impervious surface density, a geographically weighted regression coefficient quantifying variation in the relationship between trees, and impervious surface cover, and a principal component coefficient summarizing a number of distance-based variables.
The final output was compared to output from the commonly-used "Wildland Urban Interface" (WUI) method to determine which better predicts differences in population density and presence of buried streams, both established indicators of urbanization. Neither was used in the initial analysis. An ANOVA showed the five URTZ zones were significantly different from each other (P<.001) based on both population density and buried stream density using the high resolution land cover data. The medium resolution data showed each zone significantly different (P<.001) except the two transition zones. The ANDVA of the WUI zones and buried streams show no significant difference between the transition zone and rural areas. The results suggest that the URTZ method captures a significant amount of ecologically meaningful landscape variation, while the WUI method may not have enough specificity for determining transition zones at this scale. Further, high resolution imagery may only be required if several levels within a transition zone are needed for the project goals.
This project aims to: (1) define key indicators of transition zones within urban-rural gradients with an emphasis on environmental health; (2) develop a new method using cluster analysis and data reduction techniques to spatially define urban, rural, and transition zones, known as Urban-Rural Transition Zones (URTZ); and (3) implement this approach in the Baltimore-DC Corridor and demonstrate how it builds and differs from existing classification systems.
A suite of ecosystem health indicator variables were summarized by 1911 Census Block Groups for the Baltimore to Washington, DC corridor. A two-step cluster analysis was used to differentiate the zones of the urban -rural gradient in the study area. This analysis was repeated using many different combinations of variables, including both high and medium resolution land cover data.
Of the many resulting outputs, the preferred one was determined based on three best fit criteria. This map's five zones included urban, urban fringe, urban. transition, rural transition and rural. The indicator variables used in differentiating these zones included coverages of total vegetation, total impervious surface density, a geographically weighted regression coefficient quantifying variation in the relationship between trees, and impervious surface cover, and a principal component coefficient summarizing a number of distance-based variables.
The final output was compared to output from the commonly-used "Wildland Urban Interface" (WUI) method to determine which better predicts differences in population density and presence of buried streams, both established indicators of urbanization. Neither was used in the initial analysis. An ANOVA showed the five URTZ zones were significantly different from each other (P<.001) based on both population density and buried stream density using the high resolution land cover data. The medium resolution data showed each zone significantly different (P<.001) except the two transition zones. The ANDVA of the WUI zones and buried streams show no significant difference between the transition zone and rural areas. The results suggest that the URTZ method captures a significant amount of ecologically meaningful landscape variation, while the WUI method may not have enough specificity for determining transition zones at this scale. Further, high resolution imagery may only be required if several levels within a transition zone are needed for the project goals.