Seasonal Precipitation Predictability for the Northern Hemisphere Using Concurrent and Preseason Atmospheric Water Vapor Transport and Sea Surface Temperature

Abstract Integrated atmospheric water vapor transport (IVT) is a determinant of global precipitation. In this paper, using the CERA-20C climate reanalysis dataset, we explore three questions in Northern Hemisphere precipitation for four seasons: 1) What covariability between leading spatiotemporal modes seasonal sea surface temperature (SST), IVT, and Hemisphere? 2) How well can spatial be reconstructed from IVT SST same season? 3) next season predicted current season’s IVT? Wavelet analyses identify covariation those 2–8-yr band, with highest amplitude March–May (MAM) season, provide firm physical explanation potential predictability. We find that subset 10 principal components fields has significant trends connections modes, provides an accurate statistical concurrent reconstruction one-season-ahead forecast thus providing pathway to improving understanding prediction extremes context change attribution, longer prediction, scenarios. The same-season model explain 76% variance, next-season 58% variance hemispheric on average.