Evidence for a link between century-scale temperature variations in the Northern Hemisphere and long-term changes in volcanic activity

Five Northern Hemisphere temperature reconstructions covering time intervals with duration of 1210–2015 years and three volcanic activity records covering time intervals with duration of 900–1100 years were analyzed. Joint analysis of these paleorecords performed by means of both Fourier and wavelet approaches showed that both Northern Hemisphere temperature and volcanic activity exhibit century-scale (50–130 year) variability. Volcanic and temperature century-scale variations were found to be negatively correlated, with the correlation coefficient being highest when volcanic changes lead temperature changes by 10–20 years. The same effect was found for global temperature. In many cases, the correlation between century-long changes in volcanic activity and corresponding variations in individual temperature reconstructions is significant with a p -value less than 10 −3 . The combined probability of the null hypothesis—the assumption that there is no real relationship between a century-scale change in volcanic activity and temperature in the Northern Hemisphere, and that any correlations found are due to chance alone—was less than 10 −5 . No evidence was found for a significant linear correlation between the Gleissberg century-long solar activity cycle and temperature. The effect of possible overlapping of proxy indices used in temperature reconstructions is discussed, as well as possible physical mechanisms underlying the identified long-term ocean–atmosphere correlations and observed time lags. Recommendations for further research are proposed.