Preconditioning of the precipitation interannual variability in southern Mexico and Central America by oceanic and atmospheric anomalies

Abstract

AbstractThe interannual variability of precipitation in southern Mexico, northwestern Central America, and part of the northeastern tropical Pacific and its relation with previous oceanic and atmospheric anomalies are studied using the National Centers for Environmental Prediction Climate Forecast System Reanalysis data for 1979–2015. The rainy season over the study region was divided into three intraseasonal phases: before, during, and after the mid‐summer drought. The interannual precipitation anomalies between the three phases are significantly correlated with each other, although there are years in which the precipitation anomalies for different phases differ in their sign. The interannual variability of the precipitation in this region is closely related with previous oceanic and atmospheric anomalies that occurred in the 60–31 days before the beginning of each phase. Multilinear regression models, between precipitation anomalies and previous anomalies of oceanic and atmospheric variables, explain 49, 60, and 55% of the variance of the precipitation interannual variability for the respective phase. Within a phase, in more than 76% of the cases, the estimated precipitation anomalies had the correct sign but in most cases their amplitude, irrespective of sign, were underestimated, as is expected from a regression model which aims to estimate the conditional mean. The more relevant processes associated with the interannual variability of precipitation in the study area are the westward displacement and strengthening of the North Atlantic Subtropical High, the meridional displacement of the Intertropical Convergence Zone and the strength of the Caribbean Low‐Level Jet. In general, the mechanisms that explain interannual variability of precipitation are the same through the phases. However, the magnitude of the anomalies and the size of the regions evolve in time in a particular way for each variable.