Precipitation in Boreal Summer Simulated by a GCM with Two Convective Parameterization Schemes: Implications of the Intraseasonal Oscillation for Dynamic Seasonal Prediction
Journal Article
Abstract; In this paper, the intraseasonal oscillation (ISO) and its possible link to dynamical seasonal predictability within a general circulation model framework is investigated. Two experiments with different convection scheme algorithms, namely, the simplified Arakawa–Schubert (SAS) and the relaxed Arakawa–Schubert (RAS) convection algorithms, were designed to compare seasonal simulations from 1979 to 2002 on a seasonal model intercomparison project (SMIP)-type simulation test bed. Furthermore, the wave characteristics (wave intensity, period, and propagation) of the simulated ISO signal provided by the model with two different convection schemes for extended boreal summers from 1997 to 2004 were compared to the observational ISO signal. Precipitation in the boreal summer was fairly well simulated by the model irrespective of the convection scheme used, but the RAS run outperformed the SAS run with respect to tabulated skill scores. Decomposition of the interannual variability of boreal summer precipitation based on observations and model results demonstrates that the seasonal predictability of precipitation is dominated by the intraseasonal component over the warm pool area and the SST-forced signal over the equatorial Pacific Ocean, implying that the seasonal mean anomalies are more predictable under active ISO conditions as well as strong ENSO conditions. Comparison of the ISO simulations with the observations revealed that the main features, such as the intensity of precipitation variance in the intraseasonal time scale and the evolution of propagating ISOs, were reproduced fairly well by the model; however, the wave characteristics associated with the ISO signals were better captured by the experiment with the RAS scheme than the SAS scheme. This study further suggests that accurate simulation of the ISO can improve the seasonal predictability of dynamical seasonal prediction systems.
