A Regional Ocean–Atmosphere Model for Eastern Pacific Climate: Toward Reducing Tropical Biases*

Abstract

AbstractThe tropical Pacific Ocean is a climatically important region, home to El Niño and the Southern Oscillation. The simulation of its climate remains a challenge for global coupled ocean–atmosphere models, which suffer large biases especially in reproducing the observed meridional asymmetry across the equator in sea surface temperature (SST) and rainfall. A basin ocean general circulation model is coupled with a full-physics regional atmospheric model to study eastern Pacific climate processes. The regional ocean–atmosphere model (ROAM) reproduces salient features of eastern Pacific climate, including a northward-displaced intertropical convergence zone (ITCZ) collocated with a zonal band of high SST, a low-cloud deck in the southeastern tropical Pacific, the equatorial cold tongue, and its annual cycle. The simulated low-cloud deck experiences significant seasonal variations in vertical structure and cloudiness; cloud becomes decoupled and separated from the surface mixed layer by a stable layer in March when the ocean warms up, leading to a reduction in cloudiness. The interaction of low cloud and SST is an important internal feedback for the climatic asymmetry between the Northern and Southern Hemispheres. In an experiment where the cloud radiative effect is turned off, this climatic asymmetry weakens substantially, with the ITCZ migrating back and forth across the equator following the sun. In another experiment where tropical North Atlantic SST is lowered by 2°C—say, in response to a slow-down of the Atlantic thermohaline circulation as during the Younger Dryas—the equatorial Pacific SST decreases by up to 3°C in January–April but changes much less in other seasons, resulting in a weakened equatorial annual cycle. The relatively high resolution (0.5°) of the ROAM enables it to capture mesoscale features, such as tropical instability waves, Central American gap winds, and a thermocline dome off Costa Rica. The implications for tropical biases and paleoclimate research are discussed.