Remote versus local controls of east Pacific intraseasonal variability.

摘要

The Madden-Julian Oscillation (MJO) is the dominant mode of tropical intraseasonal variability and propagates eastward at 5 m/s with primary signals in wind and precipitation. During boreal summer, interactions between intraseasonal variability in the eastern Hemisphere and the east Pacific warm pool are often described as a local amplification of the propagating MJO. However, the precise mechanism by and degree to which intraseasonal variability in the eastern Hemisphere affects the east Pacific warm pool are not well understood. One school of thought holds that the MJO initiates a dry intraseasonal Kelvin wave response in the west Pacific that rapidly propagates into the Western Hemisphere and initiates intraseasonal convective variability there.;To quantify the relationship between the source (Eastern Hemisphere) and amplification region (east Pacific warm pool), sensitivity tests in two separate models are used to determine the importance of local versus remote controls of east Pacific warm pool intraseasonal variability. The two models include the National Center for Atmospheric Research Community Atmosphere Model 3 (CAM3) and the International Pacific Research Center Regional Atmosphere Model (IRAM). The two models use different schemes to isolate the east Pacific from eastward-propagating intraseasonal variability that impinges from the west. Removing the influence of the MJO on the east Pacific warm pool in these two models reveals different insights into local versus remote control of intraseasonal variability in the east Pacific. The CAM3 produces comparable intraseasonal variability in winds and precipitation in the east Pacific when Kelvin wave signals from the west are removed, suggesting that the Eastern Hemisphere MJO helps to pace east Pacific intraseasonal variability, although east Pacific variability can exist in isolation from the MJO. Thus, the CAM3 supports independent intraseasonal variability in the east Pacific warm pool that may be phase locked to intraseasonal variability in the Eastern Hemisphere in observations. However, the IRAM has very small east Pacific intraseasonal variability when isolated from global MJO signals. The weak intraseasonal variability in IRAM may be a result of mean low-level wind biases that cause 30 to 90 day surface flux anomalies to be out of phase with 30 to 90 day precipitation and low level wind anomalies. As a result, the IRAM model does not support an independent local mode of intraseasonal variability in the east Pacific.