Semester Award Granted

Summer 2025

Submission Date

August 2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Thesis/Dissertation Advisor [Chair]

Tobin Hindle

Thesis/Dissertation Co-Chair

Diana Mitsova

Abstract

Surface Urban Heat Islands (SUHI) and Land Surface Temperature (LST) are important to study because they can have large impacts on human and environmental health and well-being. The research described in this dissertation includes three separate but related studies of SUHI and LST. The first study evaluates historic LST trends in a multiscalar view and examines the impact of environmental and socioeconomic explanatory variables on LST. Results show that NDVI and impervious surfaces are significant explanatory variables for the dependent variable LST. The rate of change (RoC) of each of the regression models ranges between 0.29oC/decade and 1.56oC/decade. The second study predicts LST trends using data at multiple scales from 2025 to 2055 using an Artificial Neural Network (ANN). Potential policies involving managed changes in Normalized Difference Vegetation Index (NDVI) are examined to assess the impacts on LST of adding or reducing NDVI. The second study also evaluates the urban simulation of land use land cover change (LULC) from 2025 until 2055 using the CA-Markov model. The results show that increasing NDVI can be an effective policy tool to regulate LST and to reduce SUHI impacts. Conversely, decreasing NDVI can exacerbate increases in LST. The LULC results show that built-up areas are growing rapidly and encroaching on rural wetland areas and 147 km2 in built-up area land cover were added during the past 20 years. The third and last study investigates preterm births (PTB) using environmental and socio-economic variables to determine the correlation and expected location of PTB. Ordinary Least Squares Regression and Geographically Weighted Regression are applied to determine how the explanatory variables relate to PTB. The results show that LST and Below Poverty are the most significant variables and have positive correlations with PTB. Understanding historical and future LST trends and their relationship with human health and well-being, is important for urban planners, emergency responders, and residents. The research described in the dissertation pioneers a multi-scale approach to assess the correlation between LST and SUHI and demonstrates the importance of greenery as measured by NDVI as a policy tool in regulating LST and SUHI.

Download

Share

COinS