Solar Influences On Eurasian Winter Atmospheric Circulation And Temperature Extremes

Solar activity is one of the most important external forcing factors in the climate system.However,up to now,the role of the 11-year solar cycle(SC)in regional decadal climate variability remains uncertain and requires further investigations.In this study,solar influences on spatial patterns of Eurasian winter atmospheric circulation and temperature extremes are investigated based on a multiple linear regression(MLR)method,multi-sources observational and reanalysis data,with particular focus on North Atlantic-Eurasia surface climate responses.In addition,by using model simulations from phase 5 of the Coupled Model Inter-comparison Project(CMIP5),the abilities of 7 models in reproducing the surface SC signals are also evaluated.The main conclusions are summarized as follows:(1)Robust and significant solar signals in surface air temperature(SAT)are detected in mid-to-high latitudes of the Eurasia.The spatial pattern of sea level pressure(SLP)responses to solar activity is similar but not identical to that of the North Atlantic Oscillation(NAO).Compared to the NAO,geographic distribution of solar-induced SLP anomalies shifts eastward,with significantly enhanced influences over Eurasia.These circulation and temperature anomalies are also present 1-3 years after solar maximum,and peaks at 1-2 years lags.The surface responses can be largely attributed to the "top-down" mechanism.During high solar activity winters,an enhanced stratospheric zonal wind anomaly propagates poleward and downward through wave-mean flow interactions,causing NAO-like circulation anomalies in the troposphere.However,the "bottom-up" mechanisms may also contribute to the SC signals.Inhomogeneous surface heating caused by anomalous solar activity modifies the land-ocean thermal contrast(LOTC),which probably enhances the solar-induced circulation patterns.Such a positive feedback may potentially strengthen the solar influences.(2)SC also substantially influences the wintertime temperature extremes over Eurasia.In Eastern Europe and Siberia,the frequency of wintertime cold(warm)temperature extremes decreases(increases)significantly during high solar activity years,mainly reflecting a result of a shift in the sub-seasonal SAT probability distribution function(PDF)associated with increased winter mean SAT.Although the SC signal in wintertime SAT is rather weak over the East Asia,the occurrence of both cold and warm temperature extremes decreases substantially during high solar activity winters.The responses in temperature extremes over East Asia are resulted from changes in the shape of the SAT PDF,which primarily reflects as a decrease in sub-seasonal SAT variance.Further analysis suggests the SC influences the temperature extremes over East Asia through the following processes:on the one hand,SC signal on winter mean SAT features robust warming in high-latitude Siberia that decay southward,which weakens the meridional temperature gradient over East Asia;on the other hand,the SC also weakens transient eddy activities over East Asia.Under the combined effects of the above two processes,sub-seasonal SAT variance over East Asia is weakened,resulting in a reduced occurrence of both cold and warm temperature extremes.(3)The changing response of the North Atlantic/European winter circulation to the SC is also investigated.The synchronous correlation between SC and winter climate over the North Atlantic/European region is non-stationary.Analyses of only the most recent decades suggest a significant NAO-like response pattern that is synchronous to the SC.However,analyses of long-term climate data sets dating back to the late 19th century suggest a SLP response that lags the SC by 2-4 years in the southern node of the NAO(i.e.Azores region).Results from a lead/lag MLR analysis with a sliding window of 44-years over the period 1751-2016 confirm previous analyses,in which the average response for the whole period features a statistically significant 2-4-year lagged SLP response centered over the Azores region.Overall,the lagged nature of Azores SLP response is generally consistent in time.Stronger and statistically significant SC signals tend to appear in the periods when the SC forcing amplitudes are relatively larger.Individual month analysis indicates the consistent lagged response in December-January-February average arises primarily from early winter months(i.e.December and January),which has been associated with ocean feedback processes that involve reinforcement by anomalies from the previous winter.Additional analysis suggests that the synchronous NAO-like response in the recent decades arises primarily from late winter(February),while the variations in February responses are found to be modulated by the phase of the multi-decadal NAO variability.(4)The multi-model mean(MMM)of the 7 CMIP5 models can reproduce the lagged nature of the Azores SLP response to some degree,but with weaker amplitude and statistical significance compared to that of the observations.However,the MMM failed to reproduce the sub-seasonal evolution of the SC signal.MMM can also reproduce the positive NAO-like circulation anomalies synchronous to the SC during the recent decades,as well as the significant warming over mid-to-high latitudes of the Eurasia.These circulation and temperature responses still exist after solar maximum and peaks at 2 years lag.But the model failed to reproduce the non-stationary feature of the synchronous SC signals.Besides,the ozone implementation of the model has substantial impact on the simulated surface SC signals.The models with prescribed ozone failed to simulate the lagged NAO-like response,while in those models the ozone is calculated interactively the lagged NAO-like response is well reproduced.