Author

Date of Award

2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Atmospheric and Earth Science

Committee Chair

Udaysankar Nair

Committee Member

John R. Mecikalski

Committee Member

Xiaomin Chen

Committee Member

Rahul Ramachandran

Committee Member

Sundar A. Christopher

Research Advisor

Udaysankar Nair

Subject(s)

Hurricanes, Cities and towns--Growth--Environmental aspects, Weather forecasting, Artificial intelligence

Abstract

Hurricanes remain among the most destructive natural hazards, producing widespread impacts through extreme rainfall, flooding, and damaging winds. Rapid urbanization in coastal regions further amplifies these risks by modifying land surface characteristics and increasing population, while accurately forecasting hurricane evolution remains a major challenge. This dissertation addresses these challenges through three interconnected research objectives focusing on urbanization-induced hurricane modification, hydrological response, and AI-based hurricane forecasting. The first objective investigates how incremental multi-decadal urban growth modifies hurricane rainfall distribution and rainband evolution during Hurricane Harvey (2017). Convection-permitting WRF simulations demonstrate that realistic urban growth alters boundary-layer structure, storm-relative inflow, convective organization, and rainband propagation, resulting in nonlocal redistribution of rainfall and downstream precipitation enhancement. The urbanized simulation delayed the inland propagation of a major outer-core rainband by approximately 3 h 20 min relative to the cropland simulation, allowing organized convection to persist longer near the coastal and urban-periphery environment. The second objective examines how urbanization and precipitation variability influence hydrological response using a factor separation framework coupled with WRF–Hydro simulations. Results show that precipitation variability dominates runoff magnitude and timing, while urbanization primarily enhances surface runoff and suppresses infiltration. The interaction between urbanization and precipitation further produces nonlinear amplification of runoff during extreme rainfall periods. The third objective evaluates AI-based forecasting systems for hurricane track and intensity prediction and investigates methods for reducing systematic forecast biases. Results indicate that higher-resolution AI models improve hurricane prediction skill; however, substantial intensity biases remain during rapid intensification events. To address these limitations, deep learning-based Hurricane Intensity Estimation (HIE) models were developed as post-processing frameworks, reducing wind speed prediction errors from approximately 17 m s−1 to nearly 3.5 m s−1 relative to baseline AI forecasts. Overall, this dissertation demonstrates that urban growth can substantially modify hurricane rainfall and hydrological response through coupled storm-scale processes, while AI-based forecasting systems provide promising pathways for improving hurricane prediction.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.