Semester of Graduation

Spring 2024


Master of Oceanography and Coastal Sciences (SOCS)


Oceanography and Coastal Sciences

Document Type



The Caribbean Sea is home to numerous small, yet beautiful and ecologically diverse, island countries. Unfortunately, due to their small size and rocky soils, they lack access to large water reservoirs, lakes, rivers, and groundwater aquifers as in many continental settings. Consequently, when these islands experience a drought, the negative effects can develop rapidly, persist, and require extensive recovery time. The research region for investigating hydrologic features in this project is specifically Puerto Rico, which along with the rest of the Caribbean, is an underrepresented region in this specific area of research. Additionally, Puerto Rico is a climate-data-rich region relative to the other Caribbean Islands. It consists of two dominant rainfall seasons: The early rainfall season (ERS; April – July) and the late rainfall season (LRS; August – November). Earlier research shows that the ERS is responsible for approximately half of the precipitation variability in Puerto Rico, the largest of any season. This project expanded the frontier of drought research into a tropical environment, where it has been largely overlooked in meteorology and hydrology research.

The Saharan Air Layer (SAL) is primarily known for transporting vast amounts of dust and is a forcing mechanism for Caribbean drought. Data from the Climate Forecast System, version 2 (CFSv2), model was used to train a self-organizing map (SOM) to categorize different spatial and temporal patterns of the SAL and its effects on Puerto Rican drought. Although research has studied dust transport in other regions, the Caribbean is both literally and figuratively “getting left in the dust.” Findings show that the apparent SAL nodes in the SOM were more frequent during drought years and less frequent during pluvial years. The mean Gálvez-Davison Index (GDI) value over Puerto Rico proved to be the most beneficial variable in forecasting drought, reaching R2 correlative values of up to 0.53. Findings from this project will help earth system scientists understand how dust transport on broad scales can lead to precipitation deficits in distant regions.



Committee Chair

Miller, Paul

Available for download on Friday, April 02, 2027