ClimaTea Journal Club

This week Joe Fitzgerald will present on Shin et. al. 2010. Here are his thoughts on the paper: In previous ClimaTea meetings we have discussed the related subjects of drought, heat waves, and the relationship between SST anomalies and anomalous conditions over North America. In this week’s ClimaTea, we will return to this collection of subjects, focusing on the relationship between drought in North America and tropical SSTs. Shin et al. (2010) ask the following question: For a given SST anomaly pattern, can we quantify the expected effect on North American drought? They attempt to provide an answer using a technique closely related to the concepts of Green’s functions, Linear Response Functions, and Linear Inverse Modeling. Their framework captures much of the nonlinearity of the system, while benefiting from the analytical advantages of mathematical linearity. By conducting a large set of climate model simulations driven by climatological SSTs, except with small “patches” of SST that are perturbed from their climatological values in each simulation, they construct the linear response of the (simulated) North American climate to small changes in the boundary conditions. They compare the results of their method to similar approaches based on direct regression of the observations and find some improvement using their method. More importantly, beyond the technical aspects, I think that the approach taken in this paper - to work toward an understanding of the sensitivity of the climate based on the linear dynamics of climate statistics - provides a powerful and clear way to think about climate problems more generally.