Graduation Semester and Year
2023
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Urban Planning and Public Policy
Department
Urban and Public Affairs
First Advisor
Ardeshir Anjomani
Abstract
In this dissertation research, we explored land use change for Activity and Residential land uses in the Dallas-Fort Worth Metropolitan Area (DFW) from 1990 to 2020 in association with spatial and socioeconomic factors. It aims to study three aspects of land use change: dynamics, drivers, and impacts. The research started with three hypotheses, all proven and accepted by the research findings. We reviewed the relevant literature from five aspects of land use change: theories, models, methods, drivers, and impacts. Based on the reviewed literature, the land use change model we discussed is an economics-based spatiotemporal modeling approach in which the basic concept of land use change is a historical spatial development pattern. For land use change dynamics, we used spatial statistics methods of Global Moran’s I for spatial autocorrelation, Getis-Ord Gi* for hot spot analysis, and Anselin Local Moran’s I for cluster/outlier analysis. For drivers of land use change, on the other hand, we used Spatial Panel Data Regression (SPDR) to model the factors that drive land use change. Finally, for the impacts of land use change, we used the results from the spatial statistics and the SPDR to discover the impacts of land use change on the urban structure. The results of dynamics of land use change show that Activity land uses gradually leave central regions of the DFW Metropolitan Area in favor of the peripheral regions. On the other hand, Residential land uses, while expanding toward suburbs, tend to fill the vacancies in the CBDs created by Activity land uses. Also, the Z-score (as a measure of compactness) is decreasing while the P-value (the probability of randomness of the distribution pattern) is increasing for both Activity and Residential land uses over time, indicating the scatteredness of hot spots of these land uses. The new hot spots for Activity land uses are primarily located in the north, east, and south of the DFW Metropolitan Area. In contrast, the inner parts (including downtown Dallas, downtown Fort Worth, and the corridor between them) are losing Activity land uses. Therefore, while the null hypothesis of Complete Spatial Randomness (CSR) is rejected, economies of scale (agglomeration economies) and spatial dependency theories of land use distribution are weakening in explaining the distribution pattern of Activity and Residential land uses in the DFW Metropolitan Area. We tested spatial and non-spatial regression models to model the drivers of land use change. Spatial regression models outperform non-spatial models; therefore, we used the Spatial Panel Data Regression (SPDR) model. Several theoretically and empirically essential variables were statistically significant, with the right signs explaining land use changes. Model results show that land use change is autoregressive, meaning the ratio of Activity and Residential land uses in a block group at a previous time (t-1) is the most significant driver of land use change. Also, vacant land is the second most significant driver of land use change, indicating ample land in the region is helping the sprawl. Finally, the second law of geography, the phenomenon external to a geographic area of interest that affects what goes on inside it, is applicable to land use change. By combining the results of dynamics and drivers of land use change, we analyzed how land use change may have impacted the urban structure of the DFW Metropolitan Area. To do so, we cross-examined the Z-score and distribution of AADT using the QQ Plot. The results of the analysis show that land use change resulted in urban sprawl and the spread of traffic congestion in the region. Therefore, the DFW Metropolitan Area has evolved into a sprawled multi-centric Metropolitan Area; it is a region of dynamism and growth, where several activity and residential centers have transformed, and new activity and residential centers have emerged. Finally, we discussed and recommended modeling, development, land use, transportation, socioeconomic, and environmental implications of land use change model results in the DFW Metropolitan Area. Understanding the causes and effects of land use changes helps planners, policymakers, and local legislators observe the impacts of such development policies and regulations. Furthermore, via urban and regional policies, they can promote policies to improve the attractiveness of urban or suburban areas as locations for investment and intervene with policies to mitigate the negative consequences of these changes.
Keywords
Land use change modeling, Spatial statistics, Spatial panel data regression, DFW metropolitan area
Disciplines
Public Affairs | Public Affairs, Public Policy and Public Administration | Social and Behavioral Sciences
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Behseresht, Ali, "Modeling Land Use Change using Spatial Statistics and Spatial Panel Data Regression in Dallas-Fort Worth Metropolitan Area (1990 - 2020)" (2023). Public Affairs Dissertations. 231.
https://mavmatrix.uta.edu/publicaffairs_dissertations/231
Comments
Degree granted by The University of Texas at Arlington