Journal of Oceanography, Vol. 65 (No. 4), pp. 487-498, 2009
Ayumi Fujisaki1*, Hajime Yamaguchi1, Takenobu Toyota2, Akio Futatsudera1 and Masaru Miyanaga1
1Applied Fluids Engineering Laboratory, Department of Environmental and Ocean Engineering, School of Engineering, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
2Institute of Low Temperature Science, Hokkaido University, Kita-ku, Sapporo 060-0819, Japan
(Received 17 April 2008; in revised form 7 October 2008; accepted 13 March 2009)
Abstract: The air-ice drag coefficient under neutral stratification CDN was measured with the eddy correlation method in the southern Sea of Okhotsk. The disturbance of the wind field caused by the ship's structure was evaluated by computational fluid dynamics (CFD), and two types of correction methods were applied to estimate the error span of CDN: one is based on the results of CFD, and the other is based on the parameterization of CDN over open water suggested by Taylor and Yelland (2001). The CDN × 103 values finally obtained ranged from 1.9 to 5.4 with a mean value of 2.7 by the CFD correction and from 1.5 to 5.0 with a mean value of 3.1 by the other method. This is somewhat larger than the value of 2.5 suggested by Shirasawa (1981), and in the same range as 2.2-4.0 over rough ice and 3.1-5.0 over very rough ice, values which were complied by Guest and Davidson (1991) for first year ice. Most of the ice conditions were characterized by broken floes with a diameter less than 100 m and raised rims, which made the surface rougher than flat, level ice. The relation between CDN and ice concentration was not clear, mainly because the contribution of the form drag caused at the freeboard was undetectable due to the great variation of ice surface condition. The roughness length zM was also evaluated using the model developed for snow covered ice in a previous study.