Improved urban extreme weather simulation by capturing urban heterogeneity

Simple item page
dc.contributor.advisorYang, Zong-liang
dc.contributor.advisorNiyogi, Dev
dc.contributor.committeeMemberBanner, Jay L.
dc.contributor.committeeMemberJackson, Charles S.
dc.contributor.committeeMemberVizy, Edward K.
dc.creatorFung, Kwun Yip
dc.creator.orcid0000-0003-1213-165X
dc.date.accessioned2023-09-18T22:32:36Z
dc.date.available2023-09-18T22:32:36Z
dc.date.created2023-08
dc.date.issued2023-08-11
dc.date.submittedAugust 2023
dc.date.updated2023-09-18T22:32:37Z
dc.description.abstractThe current observations clearly indicate a world undergoing climate change, marked by increased frequency and intensity of heat waves and heavy rainfall events since 1950. The repercussions of extreme weather in urban settings are highly variable due to the diverse environments present. These risks will disproportionately affect vulnerable populations, making it essential to study the effects of urban heterogeneity on extreme weather. This dissertation is structured into three main sections. 1. Methodology Development: The first section delves into various machine learning techniques and urban datasets to produce a precise Local Climate Zone (LCZ) map. The LCZ classification, which divides urban areas into 10 classes, aids in capturing urban heterogeneity. The random forest classifier is highlighted as producing an accurate and high-resolution output in a short timeframe. Key factors governing classification accuracy are building height and imperviousness. Building height improves the accuracy of high-rise classes, while the imperviousness dataset enhances the accuracy of low-rise classes. The refined LCZ map contributes to more accurate computational simulations of urban heterogeneity. 2. Enhancing Tropical Cyclone Simulations: The second section focuses on evaluating improvements in tropical cyclone simulations by integrating LCZ to capture urban heterogeneity. While LCZ has proven beneficial for simulating temperature, wind, humidity, and non-hurricane rainfall, its impact during tropical cyclones has been underexplored. Simulation experiments, when compared with observations, demonstrate that incorporating LCZ in urban areas enhances the accuracy of 10-m wind, 2-m temperature, land surface temperature, and rainfall hotspot locations. This underscores the significance of considering urban heterogeneity when predicting and preparing for tropical cyclones. 3. Equitable Urban Overheating Mitigation: The third section investigates the efficacy of diverse urban overheating mitigation strategies, while also addressing equity concerns. Simulations of cool roofs, green roofs, and urban trees during five heatwave events reveal that the urban trees strategy is most effective in achieving equity by cooling down vulnerable neighborhoods in Houston. This study emphasizes the necessity of factoring in city layouts and demographics to develop equitable solutions. Overall, this dissertation addresses the impacts of urban heterogeneity on extreme weather comprehensively, outlining methodologies for accurate representation, improvements in tropical cyclone simulations, and the importance of equitable urban overheating mitigation strategies.
dc.description.departmentEarth and Planetary Sciences
dc.format.mimetypeapplication/pdf
dc.identifier.urihttps://hdl.handle.net/2152/121732
dc.identifier.urihttp://dx.doi.org/10.26153/tsw/48558
dc.language.isoen
dc.subjectUrban heterogeneity
dc.subjectExtreme weather
dc.subjectWRF
dc.subjectMachine learning
dc.subjectTropical cyclones
dc.subjectUrban overheating mitigation
dc.subjectEquity
dc.titleImproved urban extreme weather simulation by capturing urban heterogeneity
dc.typeThesis
dc.type.materialtext
thesis.degree.departmentGeological Sciences
thesis.degree.disciplineGeological Sciences
thesis.degree.grantorThe University of Texas at Austin
thesis.degree.levelDoctoral
thesis.degree.nameDoctor of Philosophy

Access full-text files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
FUNG-DISSERTATION-2023.pdf
Size:
49.25 MB
Format:
Adobe Portable Document Format
Download

License bundle

Now showing 1 - 2 of 2
No Thumbnail Available
Name:
PROQUEST_LICENSE.txt
Size:
4.45 KB
Format:
Plain Text
Description:
Download
No Thumbnail Available
Name:
LICENSE.txt
Size:
1.84 KB
Format:
Plain Text
Description:
Download

Collections

UT Electronic Theses and Dissertations