transect made along the Siberian coast. The alongshore flow, v, varied primarily over length scales of 100-200 km and suggested regions of alongshore convergence and divergence. The cross-shore velocity, u, varied on scales ranging from πR0 (R0, the baroclinic radius of deformation, 〜6km) to 2πR0. In some regions the relative vorticity, ξ, (ξ 〜 -uy) was 0.4f (f = 1.4 × 10-4s-1 is the Coriolis parameter), indicating that nonlinear effects were important in the momentum balance. We computed the offshore perturbation buoyancy flux (u'p') from the deviations of u and p from the alongshore average of these variables. This flux was 〜2 × 10-2 kg m-2s-1, where the sign indicates an offshore flux of fresh water. This flux, the kinematic properties reflected in Figure 2, and the satellite imagery imply that the coastal flow was unstable. The most likely source of the eddy energy is available potential energy of the front between the SCC and Bering Sea Water. The transfer of this energy into the eddies was likely accomplished by cross-shore perturbation buoyancy fluxes working on the mean density field.. We estimate this energy transfer rate to be 〜10-4J m-3s-1. By contrast, the rate of working by the wind on the water column is 〜5 × 10-6J m-3s-1 [This estimate is given by (pamC10U310)/h where m (an efficiency factor) and C10 (drag coefficient) are both 〜10-3 and Pa is the density of air (Denman and Miyake, 1973)]. Typical September wind speeds, U10, are 〜 5ms-1, and the depth of the surface mixed layer, h, is 〜 25m. Our comparison suggests that within the SCC horizontal mixing is much more effective than vertical mixing in modeling water mass properties. The instabilities weaken the cross-shore density gradient and therefore the cross-shore baroclinic pressure gradient, Pbc1. We found that the latter diminished by a factor of 2 between the western (Long Strait) and eastern (Cape VanKarem) endpoints of the transect shown in Figure 2.
Instabilities are not the only means by which the SCC is weakened and its waters dispersed offshore however. The inflow through Being Strait is geostrophically controlled and forced by the sea level difference between the Pacific and Arctic oceans (Overland and Roach, 1987. This pressure gradient, PPA, diminishes in magnitude proceeding northward from Bering Strait as the coasts of Siberia and Alaska diverge. Its sense is also opposite the Pbc., of the SCC. Thus, at some point, Y01 along the coast the two cancel and the SCC is deflected offshore. The alongshore position, s, of Y0 also depends upon the cross-shore barotropic pressure gradient, Pbt, established by the winds which vary primarily in the alongshore direction. Thus, Y 0(s,t) marks the termination point of the SCC and it is where Pbt(s,t) + Pbc(s,t) + PPA(s)=0. (The variable t refers to time.) We show evidence that offshore flow at Y0,(s,t) is vigorous, associated with squirts, and that it involves vigorous mixing between Bering Sea Water and SCC water. We argue that Pbt(s,t) + Pbc(s,t) will vary substantially on seasonal and synoptic time scales implying that Y0(s,t) will do so also. An extreme example of this variability is shown by the data from 1995 (Figure 1; right panels) when the SCC was completely absent from the Chukchi Sea in September (and August also). (The dilute coastal water in this figure was limited to a restricted alongshore domain; the source was local and not fresh water adveded from the East Siberian Sea). Persistent, albeit weak, southeasterly winds through summer and ear1y fall prevented the SCC from developing and entering the Chukchi Sea in 1995. These winds promoted coastal divergence and upwelling which effectively displaced the location of Y0(s,t) in the East Siberian Sea.
3. REFERENCES
Aagaard, K. and E. C. Carmack, 1989: The role of sea ice and other fresh water in the Arctic circulation. J. Geophys. Res. 94:14485-14498
Coachman, L. K., K. Aagaard, and R. Tripp,1975: Bering Strait: The Regional Physical
Oceanography, 172 pp., University of Washington Press, Seattle.
Denman, K. and M. Miyake, 1973: Upper layer modification at Ocean Station Papa:
observations and simulation. J. Phys. Oceanogr. 3: 185-196.
Melling, H. 1993: The formation of a haline shelf front in wintertime in an ice-covered arctic sea. Cont. Shelf Res., 13: 1123-1147.
Overland, J. and A. T. Roach, 1987: Northward flow in the Bering and Chukchi sea, J. Geophys. Res., 92:7097-7105
Weingartner, T.J., D. Cavalieri, K. Aagaard, and Y. Sasaki: (in press, J, Geophys. Res,) Circulation, dense water formation and outflow on the northeast Chukchi shelf.
