MODELING WORKGROUP MEETING-2

Meeting held at GLERL, May 27, 1998

Participated: K.Bedford, D.Beletsky, C.Chen, P.Chu, B. Eadie, D.Endicott, G.Lang, M. Lansing, J. Mingshun, S.O Neil, J.Niebauer, P.Roebber, D.Schwab, M. Settles, A.Winkelman. Chair : D.Schwab

Contact for this report: Dima Beletsky, dima.beletsky@noaa.gov


All PIs presented status reports and plans. C.Chen also made a presentation on the modeling program in the the KITES project. We began to create a linked hydrodynamic-sediment-transport-biological model (there is probably a need for an acronym). The foundation of that model is an existing Great Lakes version of the Princeton Ocean Model. Ice transport, sediment transport and biological models are being built as extentions of the existing code. Ice transport model will be a Great Lakes version of oceanic ice model developed by L. Kantha, G. Mellor and S. Hakkinen. Sediment transport model is being developed by K.Bedford's group. We also discussed the possibility of employing a somewhat simpler version of J.Lou's (NRC postdoc at GLERL) sediment model. Biological model will be more complex than initially suggested NPZ model and will include 8 components. Currently, experiments are underway with a 1D version of that model. We will add an option to use a different advection scheme (currently we are planning to use Smolarkiewicz advection scheme) to the existing central difference scheme which is known to produce excessive diffusion in areas of large spatial gradients. It was also confirmed that a graduate student at OSU will be working on data assimilation problem.

We will start with using a new 2 km rectangular bathymetric grid developed at GLERL. The hydrodynamic model will employ 40 sigma levels. Having a one year simulation as a goal, in the beginning of the program we will be running seasonal simulations with the hydrodynamic and possibly meteorological model. Other models will be run only for selected episodes during 1992-97 and program field years. Hydrodynamic model will also be run for two pre- EEGLE years: 1996 and 1997. This will not only cover plume events that occured after the Lake Michigan Mass Balance Project (LMMBP) field years (1994-95), but will also provide more accurate initialization for 2 km model runs beginning 1998. Experience gained during LMMBP will also be useful in archiving model results and observations. The meteorological model MM5 will be run in two modes: real time and hindcast mode. Besides MM5, forcing functions will be derived from MAROBS, CCIW towers, and J.Vesecky buoy.

Lake Michigan will be added to the Great Lakes Forecasting System at OSU by summer of 1998 which will be beneficial for all EEGLE participants. In addition, several new products like transport stream functions or bottom shear stress could be added to the existing GLFS products. Initially, a hydrodynamic model and a wave model will be runnning on the 2 km grid, and providing nowcasts and 36 hr forecasts every 6 hours. The 2-dimensional sediment transport model will be added to the system later.


June 24, 1998