Abstract/Summary

The effect of air-sea coupling on the simulated boreal summer intraseasonal oscillation (BSISO) is examined using atmosphere--ocean-mixed-layer coupled (SPCAM3-KPP) and uncoupled configurations of the superparameterized (SP) Community Atmospheric Model, version 3 (SPCAM3). The coupled configuration is constrained to either observed ocean mean state or the mean state from the SP coupled configuration with a dynamic ocean (SPCCSM), to understand the effect of mean-state biases on the BSISO. All configurations overestimate summer mean subtropical rainfall and its intraseasonal variance. All configurations simulate realistic BSISO northward propagation over the Indian Ocean and western Pacific, in common with other SP configurations. Prescribing the 31-day smoothed sea-surface temperature (SST) from the SPCAM3-KPP simulation in SPCAm3 worsens the overestimated BSISO variance. In both coupled models, the phase relationship between intraseasonal rainfall and SST is well captured. This suggests that air-sea coupling improves the amplitude of the simulated BSISO and contributes to the propagation of convection. Constraining SPCAM3-KPP to the SPCCSM3 mean state also reduces the overestimated BSISO variability, but weakens BSISO propagation. Using the SPCCSM3 mean state also introduces a one-month delay to the BSISO seasonal cycle compared to SPCAM3-KPP with the observed ocean mean state, which matches well with observations. Based on a Taylor diagram, both air-sea coupling and SPCCSM3 mean state SST biases generally lead to higher simulated BSISO fidelity, largely due to their abilities to suppress the overestimated subtropical BSISO variance.