Environmental Science and Engineering Seminar
Recent research has highlighted that radiative feedbacks — and thus climate sensitivity — are not constant in time but depend sensitively on sea surface temperature and ocean heat uptake patterns ("pattern effect"). This realization has lead to an explosion of feedback definitions and methods to estimate equilibrium climate sensitivity. I will contrast three flavors of radiative feedbacks—equilibrium, effective, and differential feedbacks—and discuss their physical interpretation and applications. I show that their values can differ by more than 1Wm-2K-1 and their implied climate sensitivity can differ by several degrees. As a climate model equilibrates, radiative feedbacks change continuously on decadal to millennial timescales, in most regions reducing their tendency to stabilize the climate. In the equatorial Pacific, however, feedbacks become more stabilizing with time. The global feedback evolution is initially dominated by the tropics and later by the mid-latitudes.
Further, I will discuss to which degree coupled climate models can be expected to reproduce observed changes in sea surface temperature patterns. With seven climate model large ensembles we find that internal variability on the local scale is too large to differentiate between model biases and a forced response yet. This implies that predicting the evolution of global and local surface temperatures is currently limited by understanding internal decadal variability and the relevance of different regions' surface temperature and ocean heat uptake on radiative feedbacks.