The Role of the Ocean in Midlatitude, Interannual-to-Decadal-Timescale Climate Variability of a Coupled Model

Abstract

Three 1000-yr climate simulations with an atmospheric general circulation model (AGCM) coupled to, respectively, a slab mixed layer model, an ocean GCM, and responding to yearly repeating daily sea surface temperature (SST) and sea-ice coverage climatology derived from the fully coupled run were analyzed and compared. When coupled to a slab mixed layer, surface air temperature (SAT) and SST are strongly coupled and the reddening is significantly larger than in the case of coupling to a dynamically active ocean. A simple one-dimensional stochastic model is developed to explain the different spectra of SAT above land and ocean. It is argued that ocean advection generating SST variability that does not match the principal modes of SAT above the ocean is the main factor in damping SAT variability. The variability of SAT and 800-hPa geopotential height (GEO) and covariability of SST?SAT and SST?GEO have been analyzed, and it is found that coupling does not change the dominant patterns of atmospheric variability, but it affects the spectra. The relative importance of the dominant patterns of variability is not affected by coupling, nor do significant peaks arise in the spectra. Coupling does give rise to preferred modes of covariability between SST and SAT or GEO. A dynamically active ocean affects the spectra of these modes and occasionally gives rise to a significant spectral peak on decadal to interdecadal timescales. Also, a dynamical ocean affects SAT spectra above sea by a systematic deviation from the fitted AR(1) process.