Stratified statistical models of North Atlantic basin-wide and regional tropical cyclone counts

摘要: [1] Using the historical Atlantic tropical cyclone record, this study examines empirical relationships between climate state variables and counts. The considered as predictors include indices of El Nino/Southern Oscillation Northern Oscillation, both “local” “relative” measures Main Development Region sea surface temperature. Other measuring Meridional Mode West African monsoon. all potential in a forward stepwise Poisson regression, we examine counts variables. As further extension on past studies, basin-wide named storm cluster analysis time series representing distinct flavors cyclones, are modeled. A wide variety cross validation metrics reveal that or sums over appropriately chosen clusters may be more skillfully modeled than individual series. Ultimately, most skillful models typically share three predictors: for main development region temperatures, North Oscillation.