AbstractThe 11-year solar cycle is associated with a roughly 1Wm-2 trough-to-peak variation in total solar irradiance and is expected to produce a global temperature response. The sensitivity of this response is, however, contentious. Empirical best estimates of global surface temperature sensitivity to solar forcing range from 0.08 to 0.18 K [W m-2 ]-1. In comparison, best estimates from general circulation models forced by solar variability range between 0.03-0.07 K [W m-2]-1, prompting speculation that physical mechanisms not included in general circulation models may amplify responses to solar variability. Using a lagged multiple linear regression method, we find a sensitivity of globalaverage surface temperature ranging between 0.02-0.09 K [W m-2]-1, depending on which predictor and temperature datasets are used. On the basis of likelihood maximization, we give a best estimate of the sensitivity to solar variability of 0.05 K [W m-2]-1 (0.03-0.09 K, 95% c.i.). Furthermore, through updating a widely-used compositing approach to incorporate recent observations, we revise prior global temperature sensitivity best estimates of 0.12 to 0.18 K [W m-2]-1 downwards to 0.07 to 0.10 K [W m-2]-1. The finding of a most-likely global temperature response of 0.05 K [W m-2]-1 supports a relatively modest role for solar cycle variability in driving global surface temperature variations over the 20th century and removes the need to invoke processes that amplify the response relative to that exhibited in general circulation models.
Review of Statistical estimation of global surface temperature response to forcing under the assumption of temporal scaling
