Search for: All records

These data represent a self-organizing map (SOM) classification of all trans-Atlantic integrated dust fluxes (IDT) between June-July 1981-2020 as presented in: Miller, P. W., and C. Ramseyer, In press: The relationship between the Saharan Air Layer, convective environmental conditions, and precipitation in Puerto Rico. Journal of Geophysical Research: Atmospheres.  Each daily IDT field is paired to one of 12 discrete pathways in idt_bmus_junjul.csv. The mean composite IDT over the tropical North Atlantic for each of these 12 patterns, as well as the mean composite Galvez-Davison Index (ERS_idt_node_gdi_1981_2020_junjul.nc) and mean composite precipitation over Puerto Rico (ERS_idt_node_prcp_1981_2020_junjul.nc) for the same node-date pairings are also provided. See the above-referenced manuscript for more details. 

Abstract In groundwater-limited settings, such as Puerto Rico and other Caribbean islands, societal, ecological, and agricultural water needs depend on regular rainfall. Though long-range numerical weather predication models explicitly predict precipitation, such quantitative precipitation forecasts (QPF) critically failed to detect the historic 2015 Caribbean drought. Consequently, this work examines the feasibility of developing a drought early warning tool using the Gálvez–Davison index (GDI), a tropical convective potential index, derived from the Climate Forecast System, version 2 (CFSv2). Drought forecasts are focused on Puerto Rico’s early rainfall season (ERS; April–July), which is susceptible to intrusions of strongly stable Saharan air and represents the largest source of hydroclimatic variability for the island. A fully coupled atmosphere–ocean–land model, the CFSv2 can plausibly detect the transatlantic advection of low-GDI Saharan air with multimonth lead times. The mean ERS GDI is calculated from semidaily CFSv2 forecasts beginning 1 January of each year between 2012 and 2018 and monitored as the initialization approaches 1 April. The CFSv2 demonstrates a broad region of statistically significant correlations with observed GDI across the eastern Caribbean up to 30 days prior to the ERS. During 2015, the CFSv2 forecast a low-GDI tongue extending across the Atlantic toward the Caribbean with 60–90 days lead time and placed Puerto Rico’s 2015 ERS beneath the 15th percentile of all 1982–2018 ERS forecasts with up to 30 days lead time. A preliminary GDI-based QPF tool tested herein is a statistically significant improvement over climatology for the driest years.