Assessing atmospheric sensitivity to sea surface temperature variations using the Maximum Entropy Production Principle

Abstract

Paltridge (1975,1978) proposed that the atmosphere seeks to maximize entropy production, as a non-equilibrium thermodynamic system. He constructed a simple box model of the atmosphere which yields surprisingly realistic predictions for the latitudinally-averaged surface temperature, fractional cloud cover and meridional heat fluxes.

The poleward shift of the mid-latitude storm tracks is one of the most robust predicted features under global warming. Graff and LaCasce (2012) examined the relation between perturbed sea surface temperature (SST) and the intensity and position of the storm tracks using an Atmospheric General Climate Model (CAM 3.0). In line with several other studies, their findings suggest that both the intensity and position change in response to altered SST.

Heat transport in the atmosphere is intensified in the storm tracks. Thus, a change in the position of the storm tracks should correspond to a change in the maximum heat transport. The purpose of this study is to examine atmospheric sensitivity, and the sensitivity of atmospheric heat transport in particular, to changes in SST using a simplified energy balance model based on the principle of Maximum Entropy Production (MEP), similar to Paltridge's MEP model (1975; 1978).

A hierarchy of box models are developed and investigated, including Paltridge's original model. An Atmospheric MEP (AMEP) model is developed to make a model comparison with CAM 3.0 possible. The results from the AMEP model are compared to data from Graff and LaCasce (2012), to explore if we can capture the SST dependence. Despite having feedback mechanisms, the AMEP model is able to predict the main tendencies of the changes in temperature, convective heat flux and meridional heat transport. However,it fails to capture the shift in the meridional heat transport.