Predictable and Unpredictable Components of Wave-Like Pattern along the Asian Jet in JMA/MRI-CPS3 Seasonal Forecasts

Abstract The seasonal prediction skill of wave-like pattern along the Asian jet in July–August (JA) is assessed using monthly hindcast data from Japan Meteorological Agency/Meteorological Research Institute-Coupled Prediction System version 3 (JMA/MRI-CPS3), the operational seasonal prediction system of the Japan Meteorological Agency (CPS3). The first three principal components of Asian jet variability, derived from empirical orthogonal function (EOF) analysis of upper-tropospheric meridional wind anomalies over southern Eurasia during JA in the reanalysis, exhibit a zonal wavy anomaly pattern characteristic of the Silk Road pattern. In CPS3, the reproducibility of interannual variability in the first and second EOF scores is limited, with their amplitudes reaching approximately 70% of those observed in the reanalysis. The underestimated amplitude is likely linked to the bias of slightly northward-shifted Rossby waveguide along the Asian jet and the significantly underestimated upstream blocking frequency. The weak correlation between the two EOF scores and El Niño–Southern Oscillation (ENSO) suggests that these patterns are mainly influenced by internal atmospheric variability, contributing to significant uncertainty in seasonal forecasting. Conversely, the interannual variability of the third EOF score is moderately replicated, particularly with lead times of a few months in CPS3. This score exhibits a significant correlation with ENSO, suggesting that the third EOF pattern associated with ENSO is more reliably forecasted compared to the first and second EOF patterns associated with atmospheric internal variability. These findings indicate that variability in the Asian summer jet comprises both unpredictable and predictable components, highlighting it as a crucial signal in seasonal predictions of the Asian summer climate.