Author ORCID Identifier:
Date of Graduation
7-2026
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Environmental Dynamics (PhD)
Degree Level
Graduate
Department
Environmental Dynamics
Advisor/Mentor
Aly, Mohamed
Committee Member
Cothren, Jackson
Second Committee Member
Tullis, Jason
Third Committee Member
Ruhl-Whittle, Laura
Keywords
ENSO-NAO Teleconnections; Hydroclimatic Variability; Severe Convective Hazards; South-Central United States; Watershed Scale
Abstract
Large-scale climate oscillations influence regional hydroclimate and severe convective hazards, but their effects vary by phase combination, hazard type, and watershed moisture conditions. This dissertation examines the relationships among the El Niño-Southern Oscillation (ENSO), the North Atlantic Oscillation (NAO), spring hydroclimatic variability, and severe convective hazards across the Illinois River, Robert S. Kerr Reservoir, and Upper White River watersheds in Arkansas and Oklahoma. The study integrates three components: understanding hydroclimatic controls, quantifying hazard variability, and predicting seasonal hazard occurrence. First, PRISM precipitation data from 1950 to 2025 were used to evaluate spring precipitation and calculate the Standardized Precipitation Index (SPI). Hydroclimatic variability was analyzed using linear mixed-effects models. Neutral ENSO years were generally the wettest and most variable, whereas El Niño years were significantly drier. A significant interaction between La Niña and NAO showed that their effects cannot be interpreted independently. Spring precipitation also increased by approximately 1.1 mm per year during the study period. Second, NOAA Storm Events data and regression models were used to assess the occurrence and intensity of hail, severe thunderstorm wind, and tornadoes. ENSO-NAO combinations were evaluated for hazard occurrence, whereas ENSO and NAO were analyzed separately for intensity. Neutral ENSO with Negative NAO was associated with approximately 89% fewer hail events and 91% fewer wind events than Neutral ENSO with Neutral NAO. Tornado occurrence did not respond significantly to ENSO-NAO combinations. Spring SPI was significantly associated with wind and tornado occurrence. Negative NAO was also associated with larger hail and greater odds of stronger tornadoes. Finally, spatiotemporal analyses and Random Forest models were applied to hazard records from 1975 to 2025. All three hazards showed significant increasing trends. Tornadoes exhibited the most persistent spatial clustering, with 45 cells classified as consecutive hot spots. The models explained 24.8% to 27.5% of the variance in seasonal hazard counts. Spring SPI was the strongest predictor, followed by winter precipitation, while NAO provided little predictive information. La Niña produced 57% more predicted hail events and 38% more wind events than El Niño. Tornado responses were weaker and nonmonotonic. These findings show that ENSO and NAO effects are state-dependent and differ by hazard type and response measure. Although these oscillations provide useful climatic context, watershed moisture conditions offer more direct information for seasonal hazard prediction and regional preparedness.
Citation
Abrego-Bonilla, J. (2026). ENSO-NAO Effects on Spring Hydroclimatic Variability and Severe Convective Hazards across Watersheds in South-central United States (1950-2025). Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/6401
