Summary

The summary will review the strengths and weaknesses of a MIT GCM regional study of the Labrador Sea.

• Convective Cycle
  Mixed layer depths gradually deepen throughout the winter until convection and rapid restratification occurs. The model convects in the southwest corner of the interior Labrador Sea within 10 days of the actual onset of convection. Eddy kinetic energies jump immediately after convection. The Labrador Sea Water formed has correct salinity within .005 salinity and temperature within $.02 ^{\circ} Celsius$. A pool of anomalously fresh, cold water appears over the convective region in late winter and is inconsistent with observations but does not affect the convection. Convection modifies the density structure to at least 1500 meters, which is consistent with the 1000-1500 meters of convection actually observed. Lateral mixing from the boundaries rehomogenizes the convective region immediately. The description of convection in the model appears excellent.
  

• Mean Circulation
  The cyclonic circulation of the mean circulation is captured for the surface Labrador Current waters, the deeper Irminger Current, and the deep western boundary current. Anomalies can be transported around the rim in less than a year at the surface, which is necessary for the connection of the Labrador Sea with surrounding basins.
  

• Space and Time Variability
  As seen in a sea surface height frequency spectrum, the model produces nearly enough variability at periods between 100 days and one year, but is too low in energy at high frequencies. The spatial pattern of temporal variability is highly correlated with observations, but the background energy should be higher. Model fields are too smooth in space and SSH slope variability is not produced to the same extent as simple SSH variability. Wavenumber spectra are uniformly low in energy at all lengthscales. However, this model generally produces more variability than a comparable study 4 years ago.
  

• Vertical Structure of Kinetic Energy
  The shape of the model kinetic energy profile and the partition of energy in vertical modes is reasonable with the present data of current meters. The model accurately calls attention to the increasingly barotropic motions at high latitudes. The model kinetic energy profile at site 91 also calls attention to the model's inability to produce adequate variability in the interior Labrador Sea.
  

• Missing Physics
  A thin, atmospherically-sensitive surface layer in the model is probably symptomatic of a lack of mixed layer dynamics. This can be remedied by adding the KPP (Large et. al) parameterization, or could possibly be improved by using a depth dependent vertical diffusivity with very large values at the surface. The role of freshwater in the model is uncertain because both the Davis Strait inflow and sea ice do not exist. The salinity restoring force should be examined closely. A shelf sponge layer could be used to input the correct freshwater forcing because ``E-P'' is small in this region and properties are quickly entrained into the interior Labrador Sea.
  

• Computational Efficiency
  The high resolution version of the model is extremely expensive. Increased model resolution seems to increase the model's ability to produce variability at all space and time scales. A breakthrough performance, however, would require horizontal resolution of 4 km and seems computationally unreasonable at this time. A lower resolution version could be used to test various remedies for the major model deficiencies and may be useful for the actual assimilation of data in the future. A model with 20 vertical levels and one degree horizontal resolution would be extremely efficient. Such a model is now implemented on the new MIT GCM code. A reduction of the model domain size is also a possibility. Improved boundary conditions, such as using relative velocities from hydrography and mean transport from a global model run, would then be implemented if the boundaries were closer to the area of interest. The addition of explicit eddy parameterization would have to be used if computational efficiency is needed.
  

The model shows the ability to predict and describe first order processes in the Labrador Sea. The fidelity of the model in a multi-year run is unknown but promising. Finally, it is also unknown if the increased turbulent nature of an eddy-resolving model will make the adjoint code and subsequent state estimation effort increasingly complex and intuitively unwieldy.

Acknowledgments: I would like to thank my advisor, Carl Wunsch, and Detlef Stammer for guidance in choosing this project. I am also deeply indebted to Julio Sheinbaum for adapting and implemented this model run in the Labrador Sea. Special thanks are also due to John Marshall, Christophe Herbaut and Chris Hill, who have been instrumental in the development of the Labrador Sea regional model. Xingwen Li, Peter Huybers, Xiaoyun Zang, and W. Greg Lawson were capable troubleshooters.

REFERENCES

Boning, C.W., and R.G. Budich, 1992: Eddy Dynamics in a Primitive Equation Model: Sensitivity to Horizontal Resolution and Friction. JPO, 22, 361-380.

Fu, L.-L., T. Keffer, P. Niiler, and C. Wunsch, 1982: Observations of mesoscale variability in the werstern North Atlantic: A comparative study. J. Mar. Res., 100, 24955-24963.

Jamous, D., C. Hill, A. Adcroft, and J. Marshall, 1997: A Parallel Navier-Stokes Ocean Model: Formulation and Documentation. In press.

Jones, H., and J. Marshall, 1993: Convection with rotation in a neutral ocean: a study of open-ocean deep convection. J. Phys. Oceanogr., 23, 1009-1039.

Jones, H., and J. Marshall, 1997: Restratification after deep convection. J. Phys. Oceanogr., 27, 2276-2287.

Khatiwala, S., and M. Visbeck, 2000: An estimate of the eddy-induced circulation in the Labrador Sea. Submitted.

Lab Sea Group, 1998: The Labrador Sea Deep Convection Experiment. Bulletin of the American Meteorologincal Society, 79.

Large, W., J.C. McWilliams, S.C. Doney, 1994: Oceanic vertical mixing: a review and model with nonlocal boundary layer parameterization. Reviews of Geophysics, 32, 363-403.

Lavender, K.L., R.E. Davis, and W.B. Owens, 2000: Direct velocity measurements describe a new circulation regime in the Labrador and Irminger Seas. Submitted to Nature.

Li, X., and C. Wunsch, 1999: Relationship of TOPEX/POSEIDON Altimetric Height to Steric Height in the North Atlantic. MIT-WHOI Joint Program Research Report.

Lilly, J.M, P.B. Rhines, M. Visbeck, R. Davis, J.R.N. Lazier, F. Schott, and D. Farmer, 1999: Observing Deep Convection in the Labrador Sea during Winter 1994/95. Journ. Phys. Oc., 29, 2065-2098.

Marshall, J., and F. Schott, 1998: Open Ocean Convection: Observations, Theory, and Models. Center for Global Change Science, Report No.52.

Marshall, J., C. Hill, L. Perelman, and A. Adcroft, 1997: Hydrostatic , quasi-hydrostatic, and nonhydrostatic ocean modeling. JGR,102-C3, 5733-5752.

Pickart, R., 1992: Water mass components of the North Atlantic deep western boundary current. DSR, 39 (9), 1553-1572.

Rhines, P.B., and J.R.N. Lazier, 1995: A 13-year record of convection and climate change in the deep Labrador Sea. Atlantic Climate Change Program: Proceedings of the PI's Meeting.

Spall, M.A., and R.S. Pickart, 2000: Where does dense water sink? A subpolar gyre example. Submitted.

Stammer, D., 1997: Global Characteristcs of Ocean Variability Estimated From Regional Topex/Poseidon Altimeter Measurements. JPO, 27, 1743-1769.

Stammer, D., and C. Boning, 1992: Mesoscale variability in the Atlantic Ocean from Geosat altimetry and WOCE high-resolution numerical modeling. JPO, 22, 732-752.

Stammer, D., and C. Wunsch, 1998: Temporal Changes in Eddy Energy of the Oceans. DSR, In press.

Stammer, D., R. Tokmakian, A. Semtner, and C. Wunsch, 1996: How well does a 1/4 degree global circulation model simulate large-scale oceanic observations? JGR, 101, 25779-25812.

Wunsch, C., 1997: The Vertical Partition of Horizontal Kinetic Energy. JPO, 27, 1770-1793.

Wunsch, C., and D. Stammer, 1998: Satellite Altimetry, the Marine Geoid, and the Oceanic General Circulation. Annu. Rev. Earth Planet. Sci., 26, 219-53.