The land climate predictability at seasonal and interannual time scales is largely due to the influence of the ocean. The connections between global sea surface temperature anomaly (SSTA) and precipitation anomaly over land as a whole are assessed using observations and Atmospheric Model Intercomparison Project (AMIP) simulations for 1957-2018 in this work. With a novel bulk connectivity matrix, the regions of SSTA having the most significant connections with global land precipitation anomaly are identified and the seasonal evolution is evaluated. The similarities and differences between the observations and AMIP simulations are examined.
In both observations and AMIP simulations, SSTA in the tropical central and eastern Pacific connects strongly with the global land precipitation anomaly. Compared with that in the tropical Pacific, the connections with SSTA along the equatorial Indian and Atlantic Oceans are weaker. However, the seasonal evolution of the connection shows distinguished patterns between the observations and the AMIP simulations with the strongest (weakest) connections in October (June) in the observations, in March and October (June) in a single-member of the AMIP simulation, and in February (June) in the 17-member ensemble mean of the AMIP simulations. The ensemble averaging enhances the strength of the connectivity and improves its seasonality. The results of the bulk connectivity matrix in this work can serve as a benchmark to evaluate the connection of SSTA with global land precipitation variation in climate models.