Journal of Oceanography, Vol. 63 (No. 3), pp. 505-515, 2007
Yunhe Zhao1* and Antony K. Liu2
1Caelum Research Corporation, Rockville, MD 20850, U.S.A.
2NASA Goddard Space Flight Center, Greenbelt, MD 20771, U.S.A.
(Received 1 October 2005; in revised form 28 August 2006; accepted 28 August 2006)
Abstract: Daily Arctic sea-ice motion maps during the winter seasons (December-March) from December 1988 to March 2003 derived from NSCAT, QuikSCAT, SSM/I, and AMSR-E data by a wavelet analysis method have been merged with those derived from buoy data. These merged sea-ice motion data have been used to study the circulation regimes and winter-to-winter variability of Arctic sea-ice motion. The relation between sea-ice motion and the pressure field in the Arctic Ocean was also studied by applying Principal Component Analysis (PCA) to the monthly merged sea-ice motion data and the monthly pressure field data from IABP. The mean Arctic sea-ice motion map of the 15 winter seasons has two distinct features: the Beaufort Gyre and a cyclonic circulation system in the Eurasian Basin, which moves ice from the Laptev Sea to Fram Strait. The strengths and sizes of the two features change from one winter season to another. Seasons with a strong or normal Beaufort Gyre alternate with seasons with a weak or no Beaufort Gyre every one to three seasons. The principal components of the first two modes of PCA of the monthly sea-ice motion are closely correlated with their counterparts of the monthly pressure field in the Arctic Ocean. The mode-one components of these two anomalies alternate between anticyclonic and cyclonic circulation systems. The correlation between Arctic Oscillation indexes and the principal components of the first mode of PCA of the monthly Arctic sea-ice motion is low but statistically significant.