The two interannual variability modes of the Western North Pacific Subtropical High simulated by 28 CMIP5–AMIP models

Abstract

Using the output of the Atmospheric Model Intercomparison Project (AMIP) experiments of 28 models from the Fifth Phase of the Coupled Model Intercomparison Project (CMIP5), the models’ performances in the simulation of the two dominant interannual variability modes of the Western North Pacific Subtropical High (WNPSH) are investigated. In the observation, the positive phases of these two modes feature an anomalous anticyclone over the western North Pacific (WNP), but the first mode (M1) is closely connected with the sea surface temperature (SST) anomalies over the tropical Indian Ocean (TIO), the maritime continent (MC) and the equatorial central Pacific (CP), while the second mode (M2) is closely connected with the SST anomalies over the WNP. The M1 is well captured by the CMIP5–AMIP models forced by the historical SST, suggesting the M1 is an SST-forced mode. The CMIP5–AMIP models capture the close relationship of the M1 with the SST anomalies over the TIO, the MC and the CP. The forcing mechanisms of M1 in the CMIP5–AMIP models are consistent with the observation, including a Kelvin wave emanating from the TIO and a local Hadley circulation originating from the MC. Different from the high reproducibility of the M1, the M2 is only moderately reproduced by the multi-model ensemble (MME) mean of the CMIP5–AMIP models. The simulated anomalous WNPSH of the M2 is weaker and shifts southwestward in the MME and many individual models compared to the observation. Among the five anomalous WNPSH years associated with the M2, the MME captures the anomalous WNPSH only in 1993 and 1994 but not in 1980, 1981 and 1987. The partial reproducibility of the M2 by the CMIP5–AMIP models suggests the M2 is neither a pure atmospheric internal mode nor a pure SST-forced mode. The observed close relationship between the anomalous WNPSH and the WNP SST anomalies is underestimated by the CMIP5–AMIP models, suggesting the local SST–WNPSH relationship may depend on the air–sea interaction over the WNP.