Shifting seasonal cycle, changing monsoons, and the role of sea ice

CMIP3 models predict a delay in the phase of the seasonal cycles of both temperature and precipitation. Regionally, this implies that surface temperature peaks later in mid-to-high latitudes and the rainy season is delayed in monsoon regions. Since the models all include different parameterizations of complex physical processes, agreement on any aspect of future climate is noteworthy and suggests that a simple physical mechanism is responsible for the changes. Robust responses in global climate models are important to understand since they either indicate a real change in the Earth's climate or a mistake that all of the models are making. Here we present results diagnosing the structure of the seasonal shifts and preliminary results as to the mechanism responsible for the delay.

The cause of the temperature delay is most likely due to sea ice loss at high latitudes. As open ocean replaces sea ice in the polar regions, the high latitude ocean increases its effective heat capacity and responds slower to seasonal insolation. This hypothesis, originally suggested by Mann and Park is consistent with the model prediction of the maximum temperature delay occurring at high latitudes. It is also a simple explanation, which is desirable because the phenomenon manifests itself in all of the CMIP3 models. We show support of this hypothesis from a simple energy balance model, CMIP3 correlations, and an atmospheric global climate model.

The precipitation delay is largest in the tropics. We believe that it is being forced by the temperature delay in the high latitudes, though the mechanism for this forcing is not entirely clear. Previous modeling studies have indicated that high latitude sea surface temperature changes can shift the Intertropical Convergence Zone and monsoon regions through moisture/energy fluxes on long time scales. We will show evidence of whether or not the precipitation delay is directly due to the temperature delay and if so, describe how the moisture/energy flux changes can communicate the high latitude changes equatorward on short time scales.

While total precipitation changes in monsoon regions do not show a robust response, the change in timing is robust. The delay to the start and end dates of the rainy season in monsoon regions is asymmetric with a net effect of shortening the length of the rainy season. We will discuss whether the cause of this asymmetry in rainfall anomalies is reduced surface moisture at the start of the rainy season due to a drier dry season, or asymmetries in the high latitude forcing in the northern and southern hemispheres.