On the influence of solar cycle lengths and carbon dioxide on global temperatures

Abstract By combining Solar Cycle Lengths (SCL) and CO 2 this paper predicts a global average surface temperature (GAST) anomaly of 1.5K in the year 2100 compared to 0.42K in 1996–2006. This assumes a continuing CO 2 increase of 2 ppm per year and our derived form of Transient Climate Response (TCR), whose value 1.93 ± 0.26 K (K) per CO 2 doubling would be 1.23 times higher if the Sun were ignored. After the CO 2 effect has been subtracted out, the SCL explains a healthy 55% of the remaining variance. It also estimates that 37% of the recent warming from 1980 to 2001 was due to solar effects. We then compare with models created from Scafetta (2010, 2013) (the first of which has the best fit of all) and from radiative forcings estimated by Myhre et al. (2001) and Skeie et al. (2011). The latter confirms the solar contribution to 1980–2001 warming as 33%, in contrast to the negligible value given by Benestad & Schmidt (2009). It also gives a TCR of 1.3K if only CO 2 continues to rise, and 2.0K if CH 4 and NO 2 also rise proportionately. Likewise this model estimates the ratio between the sensitivities of forcings from the Sun and greenhouse gases as 2.9 (versus 1.0 for Benestad & Schmidt (2009)). We develop a negative exponential model for post-forced warming to derive a ratio between Equilibrium Climate Sensitivity (ECS) and TCR, estimated to be 1.15. Two statistical novelties of the paper are the computation of the exact left tail probability of the Durbin-Watson statistic, and the demonstration of an approximate relationship between the Akaike Information Criterion and the tail probability of the F-statistic.