Analysis On The Variation Of The Characteristics Of Arctic Sea Ice Motion

In this paper, we use the Arctic sea level pressure (SLP)(1979-2006) and sea icemotion data (SIM)(1979-2007) obtained from the International Arctic Buoy Programme(IABP), the corresponding NCEP/DOE daily10m wind speed data, monthly Arctic sea iceconcentration data obtained from NSIDC, the NOAA monthly AO Index and the yearly seaice area outflow through Fram Strait provided by Dr. Kwok (1978/79-2006/07). Weanalyze the characteristics of Arctic SIM based on different time scale, the characteristicsof SIM in the Fram Strait and its affiliate sea, and the correlation between SIM and windspeed using some climate diagnostic methods, sucn as, correlation analysis, trend analysis,wavelet analysis and so on.Firstly, we find that there is a close relation between the characteristic of SIM and SLP.Two obvious features, the anticyclonic Beaufort gyre (BG) and transpolar drift, arereflected whose strength is different on different time scale. BG moves with Beauty High(BH) on monthly scale. There is a low-pressure system in the Arctic from June toSeptember. And because of the low-pressure center, sea ice takes on cyclonic motion whichfollowing the movement of the low-pressure. The transpolar drift is weaker than that inother months. Although the strength of transpolar drift is different on different time scale,sea ice always moves along the isobar up to the Arctic center, and then moves across theisobar to outflow Arctic in most months.Secondly,1979-2006is divided into three periods relative to their SLP and AOindices:1979-1987(Period I),1988-1994(Period II), and1995-2006(Period III). Thedifference between the period II and other periods is large but there is only a littledifference between the other periods, which may be result from the different SLP. The transpolar drift is stronger when Beaufort gyre is weaker, and vice versa. On seasonal scale,the most significant feature in the period II is that the characteristics of SIM in winter andsummer are similar to that in high AO period.Thirdly, the x-component of sea ice velocity in the Fram Strait and its affiliate seaincreased as the rate of11.6%per decade from1979to2007and has a about16-year cycle.The seasonal cycle of the x-component of the sea ice velocity appears to be a V-shapedcurve and its minimum value appears in August. The sea ice velocity change is close to theoutput sea ice area outflow through Fram Strait and the x-component of the wind velocity.There is a significant positive relationship between the seasonal cycle of the x-componentof sea ice velocity and that of the x-component of wind velocity. Overall, the increase ofthe x-component of wind velocity could lead to the increase of the x-component of sea icevelocity, however, it is not the only reason. Sea ice changes of itself (for example, sea icethinning) maybe another reason. Besides, sea ice velocity significantly increase does notbring the increase output of sea ice area outflow through Fram Strait, which may be due tothe decrease of the sea ice concentration in Fram Strait.Finally, the correlation between sea ice velocity and wind velocity is analyzed usingSIM data and10m wind velocity (2000-2007), which has spatial and temporal variability.The correlation coefficient in winter is high, while that in summer is low. The correlationcoefficient near coast is small but they are relatively high in the center of the Arctic andCanada basin. And the correlation coefficient of the eastern coast of the Arctic is small.The result which is not consistent with previous studies using data before2000is that thecorrelation value between sea ice velocity and wind velocity is high in north of CanadianArctic Archipelago.