Arctic Sea Ice Motion Of Multiple Time Scale Effect And Its Mechanism

The Arctic is a "new frontier" of international cooperation and competition,and an important front for countries to compete for strategic resources and expand development space.The change of Arctic sea ice motion not only directly affects the Arctic climate change and the safety of shipping routes,but also affects the material exchange and energy balance between the Arctic and the global ocean.In particular,the multi-time scale change of Arctic sea ice motion will bring great uncertainty and nonlinear characteristics to the local ecological environment,material exchange and climate change.Therefore,the multi-time scale evolution characteristics of Arctic sea ice motion and the nonlinear relationship between the sea ice motion and environmental factors have become a hot topic in the field of polar sea ice motion research in recent years.To solve the above problems,this paper has carried out research in the following aspects:First of all,an Empirical Orthogonal Function(EOF)method was used to separate spatiotemporal characteristics of Arctic sea ice motion and key environmental factors in a long time series,including 10 m sea surface wind field,sea level pressure,sea ice concentration and sea surface temperature.Analyze the physical significance of the dominant temporal and spatial patterns.Secondly,based on Ensemble Empirical Mode Decomposition(EEMD)method,the multi-time scale feature analysis is conducted.The original sequences of temporal patterns of sea ice motion and key environmental factors were decomposed into Intrinsic Mode functions(IMF)components ranging from high frequency to low frequency,and the multi-time scale variation characteristics were analyzed to explore the nonlinear evolution process.Furthermore,based on the Hilbert-Huang Transform(HHT)method,the statistical characteristic quantity of each IMF component was calculated,and the multi-time scale pattern of Arctic sea ice motion was constructed to obtain the simulation and prediction results of its linear superposition,and the degree of nonlinearity in the evolution of the temporal pattern of sea ice motion was calculated.The influence degree of nonlinear feedback on low frequency oscillation process is analyzed.Finally,by studying the nonlinear coupling process between Arctic sea ice motion and key environmental factors,and their development process on high and low frequency time scales,the influence and feedback mechanism between them on each time scale is clarified.The main conclusions are as follows:(1)Under the background of global warming,the main patterns of Arctic sea ice motion showed a trend of strengthening over time,and the velocity changes showed significant seasonal characteristics.The velocity of sea ice in winter months was higher than that in summer months,and the maximum velocity was more than 12cm/s.The maximum velocity was mainly distributed in the Beaufort Sea,the basins along the east and west coasts of Greenland,the northern Barents Sea and the Bering Sea.Based on the analysis of EOF temporal and spatial patterns,the spatial distribution characteristics of the three dominant patterns of Arctic sea ice motion are respectively similar to the basin-scale anticyclonic circulation pattern(rotation direction is consistent with Beaufort vortex),Trans Polar Drift(TPD)and reverse transpolar drift pattern.The spatial distribution of winter and summer patterns is similar.(2)In terms of time-frequency characteristics,the correlation between the dominant patterns of Arctic sea ice motion and atmospheric factors is greater than that of oceanic factors.However,on the low-frequency time scale with an average period of more than 4 years,the influence of oceanic factors gradually becomes prominent,and even exceeds that of atmospheric factors on individual time scale components.For example,the correlation between the anticyclonic circulation pattern of summer sea ice motion and the Arctic Ocean Oscillation(AOO)pattern of the wind field reaches 0.78(passes the 99% confidence test),and the highest correlation with the dominant pattern of ocean factors is only 0.45.However,the correlation analysis results of the separation time scale show that the correlation between EOF2 pattern of winter sea ice motion and EOF3 pattern of sea surface temperature on the interdecadal low frequency time scale(IMF6 component)even reaches 0.92(through 99%confidence test),which is much higher than the correlation of atmospheric factors at the same scale.Therefore,by separating time scales,we can better understand the low-frequency information contributed by ocean factors,and more accurately outline the variation characteristics of sea ice motion patterns on multiple time scales.(3)In general,the variance contribution rate of Arctic sea ice motion in the high frequency component is greater than that in the low frequency component,but there are obvious differences in the specific dominant patterns.For example,the contribution rate of the high frequency component of the large-scale anticyclone circulation pattern is the highest,especially the cumulative variance contribution rate of the first two time scales in summer(IMF2-IMF3)is close to 40%,which is much higher than that of other patterns.There are components with prominent contribution rates in both high and low frequency time scales of TPD pattern,and the modulation cycle changes of this pattern at each time scale are complex.Among them,the IMF5 component in summer has a significant cyclic change with an average period of about11.4 years,and its variance contribution rate is as high as 15.7%(the other two patterns on the same scale are only about 2%-5%).The inverse TPD pattern has the highest contribution rate in the low and middle frequency time scales,and the variance contribution rate of the lowest time scale(IMF6)component in summer is much larger than the other two patterns at the same time scale.(4)The nonlinear degree of the Arctic sea ice motion pattern is mainly affected by the atmospheric factors,and the oceanic factors will also react to the nonlinear degree to a certain extent.For example,the nonlinear evolution process of large-scale anticyclonic circulation pattern of sea ice motion is mainly caused by the nonlinear forcing of atmospheric environmental factors on high frequency time scales,and the nonlinear degree of this pattern is the strongest.The TPD pattern has the most significant coupling effect with the atmosphere and ocean environment,and its change process usually has both lead and lag correlation with ocean factors for 4-5months.Therefore,the pattern can be well correlated with atmospheric and ocean environmental factors at both high and low frequency time scales through nonlinear feedback and energy exchange processes at high and low frequency time scales,respectively,and its degree of nonlinearity is in the middle.The inverse TPD pattern is mainly affected by the sea ice concentration distribution pattern regulated by the Atlantic warm water invasion on the low frequency time scale,and can interact with the high frequency process through its energy conversion process on the low frequency time scale,so as to affect the change process of the inverse TPD pattern on the high frequency time scale.Therefore,its high frequency component shows low correlation with atmospheric factors,and the nonlinear degree of this pattern is the weakest.