On the assessment of nonlocal climate feedback. Part I: The generalized equilibrium feedback assessment

A statistical method is developed to assess the full climate feedback of nonlocal climate feedbacks. The method is a multivariate generalization of the univariate equilibrium feedback assessment (EFA) method of Frankignoul et al. As a pilot study here, the generalized EFA (GEFA) is applied to the assessment of the feedback response of sea surface temperature (SST) on surface heat flux in a simple ocean-atmosphere model that includes atmospheric advection. It is shown that GEFA can capture major features of nonlocal climate feedback and sheds light on the dynamics of the atmospheric response, as long as the spatial resolution (or spatial degree of freedom) is not very high. Given a sample size, sampling error tends to increase significantly with the spatial resolution of the data. As a result, useful estimates of the feedback can only be obtained at sufficiently low resolution. The sampling error is also found to increase significantly with the spatial scale of the atmospheric forcing and, in turn, the SST variability. This implies the potential difficulty in distinguishing the nonlocal feedbacks arising from small-scale SST variability. These deficiencies call for further improvements on the assessment methods for nonlocal climate feedbacks.