Re-visiting our understanding of surface temperature response to climate forcing

Abstract:

We know that the surface temperature response, dT, to a perturbation in the climate forcing is determined by (1) the magnitude of the forcing, (2) any feedback effects and (3) the effective heat capacity of the system. These three components can be related through an energy budget model of the form dT = dQ/C, where dQ is the net heat perturbation from the combination of the forcing perturbation and the feedback processes, and C is the effective heat capacity of the system. On long, multi-decadal to centennial timescales, C is defined by the ocean. But on shorter timescales, the effective heat capacity in the atmosphere can determine the pattern of the surface temperature change. It has been demonstrated that the effective heat capacity of the atmosphere is defined by the volume of air through which that heat is distributed i.e. defined by the planetary boundary layer depth, h. The boundary-layer depth can vary from ~100 m to a few km across different locations, and during the seasonal and diurnal cycles. So even under a uniform forcing we can expect a strongly asymmetrical warming (or cooling) based on the climatology of h. We demonstrate this relationship in the recent warming period using a combination of surface observations and reanalysis products and find that it is the climatology of h which is the strongest predictor of the pattern of warming during the satellite era (1979-present).

This has important implications for the detection of climate forcing and feedback signals through the surface temperature. In many current detection and attribution studies of climate feedback processes we assume a linear relationship between a perturbation in the forcing, dQ, and the corresponding change in temperature, dT. However, the inverse relationship between h and the strength of dT means that we can get strongly amplified temperature responses in conditions with shallow h, and this linear relationship breaks down. We demonstrate that these conditions where h strongly affects dT occur more than 40% of the time. Here we present a re-evaluation of the effect of selected climate feedback processes (clouds, soil moisture and precipitation) on the surface climate by assessing the co-variability of the forcing, dQ, with the effective heat capacity of the atmosphere, h, which allows us to compare feedback processes on a like-for-like basis.