The prediction of the May-June (MJ) precipitation as the first peak of the rainy season is important in the Central American isthmus because wetter (drier) MJ seasons tend to be associated with early (late) onsets of the rainy season on the Pacific slope. Having a late start of the rains, followed by a drier season in MJ in conjunction with a deep Mid-Summer Drought (MSD), would affect significantly key socioeconomic sectors in the isthmus like hydropower generation, water supply for human consumption (as main cities in the isthmus are located on the Pacific slope) and agriculture. Using 162 gauge stations, we built skillful Canonical Correlation Analysis (CCA) prediction models for MJ season as the first peak of the rainy season, using as predictands monthly rainfall accumulations and Standardized Precipitation Index (SPI) values over Central America. Sea Surface Temperature anomalies (SSTA) were used as predictors handling a domain bounded by 63° N-10° S and 152° E-15° W, along with the Palmer Drought Severity Index (PDSI) values covering the isthmus. Leading times from December to April were explored in the predictor fields. CCA models, using February´s SSTA and April´s PDSI showed significant skill values for the prediction of MJ accumulations and the SPI over an important portion of Central America. Models´ loadings showed that warmer (cooler) Eastern equatorial SSTAs in the Pacific along with cooler (warmer) SSTAs in the Tropical North Atlantic (TNA) during February, tend to be related with drier (wetter) conditions in almost all the isthmus during the next MJ season. It is suggested that Sea Surface Temperature (SST) mode could modulate MJ precipitation in Central America influencing the position of the Intertropical Convergence Zone (ITCZ) and the strength of the trade winds. Additionally, it was observed that drier (wetter) soil moisture (PDSI) in April tends to be related with drier (wetter) precipitation conditions in almost all the isthmus during next MJ season.
