Improved Simulations of Tropical Pacific Annual-Mean Climate in the GFDL FLOR and HiFLOR Coupled GCMs

Abstract

The National Oceanic and Atmospheric Administration's Geophysical Fluid Dynamics Laboratory has recently developed two global coupled general circulation models, the Forecast‐oriented Low Ocean Resolution (FLOR) model and the High atmospheric resolution Forecast‐oriented Low Ocean Resolution (HiFLOR) model, which are now being utilized for climate research and seasonal predictions. Compared to their predecessor Coupled Model version 2.1 (CM2.1), the new versions have improved ocean/atmosphere physics and numerics and refinement of the atmospheric horizontal grid from 220 km (CM2.1) to 55 km (FLOR) and 26 km (HiFLOR). Both FLOR and HiFLOR demonstrate greatly improved simulations of the tropical Pacific annual‐mean climatology, with FLOR practically eliminating any equatorial cold bias in sea surface temperature. An additional model experiment (Low Ocean Atmosphere Resolution version 1) using FLOR's ocean/atmosphere physics, but with the atmospheric grid coarsened toward that of CM2.1, is used to further isolate the impacts of the refined atmospheric grid versus the improved physics and numerics. The improved ocean/atmosphere formulations are found to produce more realistic tropical Pacific patterns of sea surface temperature and rainfall, surface heat fluxes, ocean mixed layer depths, surface currents, and tropical instability wave activity; enhance the near‐surface equatorial upwelling; and reduce the intercentennial warm drift of the tropical Pacific upper ocean. The atmospheric grid refinement further improves these features and also improves the tropical Pacific surface wind stress, implied Ekman and Sverdrup transports, subsurface temperature and salinity structure, and heat advection in the equatorial upper ocean. The results highlight the importance of nonlocal air‐sea interactions in the tropical Pacific climate system, including the influence of off‐equatorial surface fluxes on the equatorial annual‐mean state. Implications are discussed for improving future simulations, observations, and predictions of tropical Pacific climate.