HR: 14:55h
AN: OS53A-04    [Abstracts]
TI: What Must be Observed to Diagnose & Model the Circulation of the Gulf of Maine on Monthly and Seasonal Timescales?
AU: * Pringle, J M
EM: jpringle@cisunix.unh.edu
AF: University of New Hampshire, 142 Morse Hall, 39 College Rd., Durham, nh 03824 United States
AB: The Gulf of Maine is, at depth, a semi-enclosed sea off the coast of New England and the Canadian Maritimes. Its circulation is driven by large-scale density gradients formed by winter-time water mass transformations, buoyancy forcing from estuarine and from the deep ocean, and the winds. An analysis of the sources of variability in the six-month mean transport across the central Gulf of Maine finds that that most of variability is caused by large scale hydrographic variability (approximately 50%), with smaller variations forced by changes in transport into the basin from the Scotian Shelf (about 35%) and by winds (15%). Most numerical models of the region are not able to capture the dominant source of variability in the model, changes in the density field. This failure is not due to flaws in the model, but to inadequate observations of the hydrographic structure of the gulf, and of its inflows. Because this data is not routinely obtained, the effects of large scale climate patterns such as the North-Atlantic Oscillation are not reflected in the models. Assimilation of surface temperature does not solve this problem; due to the importance of salinity in the Gulf of Maine and the increase of temperature with depth, surface temperature do not uniquely specify conditions at depth, even in the winter. A surprisingly modest observational program would be adequate to constrain the models, and allow them to capture the dominant source of circulation variability in the Gulf of Maine. The minimum necessary requirements of such an observational program are defined. It is suggested that a similar analysis should be prepared for other coasts where the internal density field is an important factor in driving the circulations on timescales of months or longer, as has been found to be true on the Mid-Atlantic bight and elsewhere.
DE: 4215 Climate and interannual variability (3309)
DE: 4219 Continental shelf processes
DE: 4255 Numerical modeling
SC: Ocean Sciences [OS]
MN: 2004 AGU Fall Meeting