Spatial Characteristics of the Remotely-Sensed Surface Urban Heat Island in Baton Rouge, LA: 1988-2003

• Main Author: Hardegree, Lynn Copeland
• Other Authors: Shih Hsu
• Format: Others
• Language: en
• Published: LSU 2006
• Subjects: Geography & Anthropology
• Online Access: http://etd.lsu.edu/docs/available/etd-04052006-105847/

Our understanding of urban effects on local climate remains unsatisfactory due to several difficulties: 1) the inherent complexity of the city-atmosphere system, 2) lack of a clear conceptual theoretical framework for inquiry, and 3) the high expense and enormous difficulties of acquiring a sufficient quantity of high-quality, high-resolution (both spatially and temporally) observations in cities. Using remotely-sensed data, this study analyzes urban heat islands (UHI) that are manifested through an elevation in the surface thermal emissions within urban regions known as surface heat islands (SHI). The study area for this research endeavor is Baton Rouge, Louisiana. Whereas the surface air temperature-derived UHI did not portray an accurate representation of distinct changes in surface temperature across the study area, the remotely-sensed surface temperature-derived SHI proved to reveal microscale differences that the surface air temperature-derived UHI was unable to depict. This study also provided verification that altering amounts of vegetation within a given land cover over time can reveal changes in surface temperature values, thus providing a means to reconstruct and predict future SHIs. This was achieved through regression equations predicting surface temperatures from known NDVI values. Finally, the moist static energy parameter was evaluated to test for a better indicator of the UHI over time throughout the study area. A decreasing temporal trend in MSE was identified throughout the study period (1988 - 2003) whereas no significant linear trend occurred in air temperature. This is supported by change detection rates generated from a comparison of the 1988 and 2003 LANDSAT data sets, as well as the range in 1988 and 2003 predicted surface temperatures (as a function of land cover).