USERS MEETING'99
Edited by Dr. Huijie Xue
School of Marine Sciences
University of Maine
Orono, ME 04469-5741
Table of Contents
I. Program
II. Abstracts
III. Conference Attendees
I. Program for the Sigma Coordinate Ocean Model Users Meeting'99
Monday/20 SEP 99
07:30 - Breakfast (Holiday Inn)
Morning Session (Chair: Huijie Xue)
08:15 - Registration08:55 - Huijie Xue: Welcome
09:00 - Chris Mooers: Opening Remarks
09:20 - George Mellor: An Overview of the Progress with POM
09:45 - Dale Haidvogel: An Overview of the Progress with ROMS
10:10 - Coffee Break and Group Photo
10:40 - Kate S Hedstrom, Dale B Haidvogel and Jennifer Francis: A Coupled Ice-Ocean Model of the Arctic Ocean Using Satellite-Derived Forcing Fields
11:00 - A. Beckmann, H.H. Hellmer, and R. Timmermann: Designing a Coupled Ice-Ocean Model of the Weddell Sea. Part I: Ocean
11:20 - R. Timmermann, A. Beckmann, and H.H. Hellmer: Designing a Coupled Ice-Ocean Model of the Weddell Sea. Part II: Sea Ice
11:40 - Christian B. Rodehacke, Aike Beckmann, Hartmut H. Hellmer, and Wolfgang Roether: CFC Simulation in the Weddell Sea
12:20 - Lunch (Holiday Inn)
Afternoon Session (Chair: David Brooks)
13:30 - Nicolas Perenne, Dale B. Haidvogel, and Don L. Boyer: Laboratory/Numerical Model Comparisons of Flow Over a Coastal Canyon13:50 - John M. Klinck and Michael S. Dinniman: Model Boundary Condition and Coastal Dynamics
14:10 - Changming Dong, Hsien-Wang Ou and Dake Chen: Tide-Induced Mean Cross-Frontal Flux
14:30 - Tal Ezer: Simulations of the Seasonal Mixed-Layer with the Mellor-Yamada Turbulence Scheme
14:50 - Akinari Kaneko: Remedies for diapycnal Mixing Occurred due to Horizontal Diffusion in the Application of Sigma-Coordinate Model to Tokyo Bay
15:10 - Coffee Break
15:30 - Anne P. de Miranda (Darbon): On the Dynamics of the Zapiola Anticyclone15:50 - Julie D. Pietrzak and Nicolai Kliem: On the Pressure Gradient Error in Sigma Coordinate Ocean Models
16:10 - Robin Robertson: A POM Variant for Modeling Internal Tides near the Critical Latitude
16:30 - Emanuele Di Lorenzo, Arthur J. Miller, Bruce Cornuelle, Terry Chereskin, and John Moisan: Fitting Hydrographic and Biological Data in the CALCOFI Domain Using ROMS (Greens Function and Adjoint Development)
16:50 - Aijun Zhang, Eugene Wei, and Bruce Parker: Subtidal Water Level Assimilation into East Coast Ocean Model
18:00 - Reception (Vanderbilt Suite, Holiday Inn)
Tuesday/21 SEP 99
07:30 - Breakfast (Holiday Inn)
Morning Session (Chair: Tal Ezer)
08:30 - George Mellor, Sirpa Hakkinen, Tal Ezer and Richard Patchen: A Generalization of a Sigma Coordinate Ocean Model and an Intercomparison of Model Vertical Grids08:50 - Y.Tony Song and Yi Chao: An Adaptive Vertical Coordinate Formulation for Ocean Models
09:10 - Charles R. Denham and Richard P. Signell: "Seagrid" MATLAB Oceanographic Grid Generator
09:30 - J. Hermann, D. B. Haidvogel, E. L. Dobbins, P. J. Stabeno, D. Musgrave: Simulation of the Southeastern Bering Sea and Gulf of Alaska Using Coupled Regional/Basin-Scale Models
09:50 - Quamrul Ahsan and Alan F. Blumberg: Wind-Induced Stratification and Destratification Mechanisms in an Estuarine Environment
10:10 - Coffee Break
10:30 - Christopher N. K. Mooers and Lianmei Gao: The Sensitivity of IAS-POM's Response in the Northern Gulf of Mexico to Various Synoptic Wind Fields Characterizing the Passage of Tropical Storm Josephine
10:50 - Chul-hoon Hong and Kyu-Dae Cho: A Three-Dimensional Numerical Study of Effects of Typhoon on Oceanographic Conditions in the Korea Strait
11:10 - K.A.Korotenko: A 3D Flow/Particle Model for Prediction Oil Transport in Coastal Waters
11:30 - Fei Chai, Huijie Xue, Mingshun Jiang, and Andrew Thomas: Coupled Circulation/Ecosystem Model with Coastal Applications
11:50 - Julie D. Pietrzak and Rich Signell: Advection Confusion
12:10 - Lunch (Holiday Inn)
Afternoon Session (Chair: Rich Signell)
13:30 - 17:10
Discussion on Community Sigma Coordinate Ocean Models
- Philosophical Issues:
Could the "best" characteristics of the many different models be combined into just a few? How many community sigma-coordinate models do we need?
Where should community models be housed? Government, academia? Does it matter?
- Technical Issues:
What are the characteristics of the ideal community model? (e.g. documentation, test cases, web pages, support staff, modular extensible structure, revision control, machine independence, i/o, analysis and graphics tools)
- Practical Issues:
How is long-term funding and stability maintained for support, writing manuals, etc.? Is what the "hubs" of the NOPP "hub"-"node" model framework should do?
15:10 - Coffee Break
Wednesday/22 SEP 99
07:30 - Breakfast (Holiday Inn)
Morning Session (Chair: )
08:30 - D. Ko, R. Preller, M. Carnes, C Barron and P. Posey: An Experimental Real-Time North Pacific Ocean Nowcast/Forecast System08:50 - Hernan G. Arango, Pan Hai, Scott M. Glenn, Dale B. Haidvogel and Roni Avissar: Coastal Predictive Experiments at LEO-15
09:10 - Manchun Chen, Lie-Yauw Oey, and Eugene Wei: Nested-Grid POM: An Application for the Port of New York/New Jersey Water Level and Current Nowcast/Forecast Model System
09:30 - Richard A. Schmalz, Jr.: The Galveston Bay/Houston Ship Channel Nowcast/Forecast System and Its Relationship to the Navigation Channel Problem
09:50 - Bernard Barnier: On the Seasonal Variability and Eddies in the North Brazil Current: Insights from Model Intercomparison Experiments
10:10 - Coffee Break
10:30 - S.Piacsek, W.Oberpril, M.Okeefe, and M.Young: Mediterranean Simulations Using Massively Parallel Versions of the POM Code10:50 - Takashi Kagimoto, Yukio Masumoto, and Toshio Yamagata: Application of the POM to the World Oceans on the Numerical Wind Tunnel
11:10 - Huijie Xue, Yu Xu, and Dave Brooks: Simulation of Penobscot Bay Circulation, April - September 1998
Meeting Adjourn
II. Abstracts
WIND-INDUCED STRATIFICATION AND DESTRATIFICATION MECHANISMS IN AN ESTUARINE ENVIRONMENT
Quamrul Ahsan and Alan F. Blumberg
HydroQual, Inc, 1 Lethbridge Plaza, Mahwah, NJ 07430
e-mail : qahsan@hydroqual.com, phone : 201-529-5151 x7135
It has been observed that the proper timing, duration and direction of wind events interacting with the geometry of an estuarine system can control the intensity of stratification and destratification of the water column. A three-dimensional, time-dependent hydrodynamic model is used to examine these processes under low-flow conditions from September through October 1997 in Escambia Bay. The model is used to examine the mechanisms of wind mixing and the distribution of vertical mixing within the estuary. Intense mixing mechanisms may be closely tied with wind-generated internal velocity shear, especially during northward wind events. A northward wind generates surface currents through frictional coupling and produces an up-estuary water surface slope that upon interaction with the geometry, accelerates the bottom current in a down-estuary direction. A vertical instability occurs due to the generation of shear across the pycnocline, causing sufficient mixing to destratify water column. On the other hand, strong stratification occurs when a southward wind advects fresher upper bay surface water into the lower bay, generating down-estuary surface slopes, which, in turn drive bottom saline water in the up-estuary direction. The reversal of bottom currents is accentuated by the Gulf Breeze Peninsula separating the estuary from the Gulf of Mexico.
COASTAL PREDICTIVE EXPERIMENTS AT LEO-15
Hernan G. Arango, Pan Hai, Scott M. Glenn, Dale B. Haidvogel and Roni Avissar
IMCS, Rutgers University
A NOPP-sponsored, real-time, atmosphere-ocean nowcast and forecast experiments were carried out during July 1999 along the Southern New Jersey Coast. A coupled atmophere-ocean (RAMS/ROMS) regional modeling system was used to predict the 3D coastal circulation associated with recurrent summer upwelling events. The various data gathered by the observational network at the Long-Term Ecosystem Observatory (LEO-15) were used to initialize, update, and validate the coupled coastal model. The updating was done via data assimilation scheme, as data became available. Additional experiments were carried using surface forcing from the Navy Operational Global and Regional Atmospheric Prediction System (NOGAPS and COAMPS) forecasts which extended out to six days. The ocean forecasting schedule was tuned to our data sampling strategy which required a three-day forecast twice a week.
ON THE SEASONAL VARIABILITY AND EDDIES IN THE NORTH BRAZIL CURRENT: INSIGHTS FROM MODEL INTERCOMPARISON EXPERIMENTS
Bernard Barnier
LEGI-IMG, BP53, 38041 Grenoble France
Tel: 33-4-76-82-5066, Fax: 33-4-76-82-5271
The time dependent circulation of the North Brazil Current is studied with three numerical ocean circulation models, which differ by the vertical coordinate used to formulate the primitive equations. The models (the Kiel version of the GFDL model, MICOM, and SPEM) are driven with the same surface boundary conditions and their horizontal grid-resolution (isotropic, 1/3° at the equator) is in principle fine enough to permit the generation of mesoscale eddies. Our analysis of the mean seasonal currents concludes that the volume transport of the North Brazil Current (NBC) at the equator is principally determined by the strength of the meridional overturning, and suggests that the return path of the global thermohaline circulation is concentrated in the NBC. Models which simulate a realistic overturning at 24°N of the order of 16 to 18 Sv also simulate a realistic NBC transport of nearly 35 Sv comparable to estimates deduced from the most recent observations. In all models, the major part of this inflow of warm waters from the South Atlantic recirculates in the zonal equatorial current system, but the models also agree on the existence of a permanent coastal mean flow to the north-west, from the equator into the Carribean Sea, in the form of a continuous current or a succession of eddies. Important differences are found between models in their representation of the eddy field. The reasons invoked are the use of different subgrid-scale parameterisations, and differences in stability of the NBC retroflection loop due to differences in the representation of the effect of bottom friction according to the vertical coordinate that is used. Finally, even if differences noticed between models in the details of the seasonal mean circulation and water mass properties could be explained by differences in the eddy field, nonetheless the major characteristics (mean seasonal currents, volume and heat transports) appears to be at first order driven by the strength of the thermohaline circulation.
DESIGNING A COUPLED ICE-OCEAN MODEL OF THE WEDDELL SEA. PART I: OCEAN
A. Beckmann, H.H. Hellmer, R. Timmermann
Alfred-Wegener-Institute for Polar and Marine Research, Bremerhaven, Germany
BRIOS [Bremerhaven Regional Ice Ocean Simulations] is a long term modeling project that currently focuses on the Weddell Sea system and includes the ocean, sea ice and ice shelves in that region. The central model component is an extended version of SPEM, with nonlinear vertical s-coordinates and the inclusion of sub-ice cavities. Main research goals are the investigation of climate variability of deep and bottom water formation and spreading (e.g., the role of freshwater input), the determination of integrated quantities (e.g., meridional overturning) and a quantitative comparison with recent observations. In this talk the overall model configuration, the experimental strategies, as well as general algorithmic and parameterizational schemes will be discussed.
COUPLED CIRCULATION AND ECOSYSTEM MODEL WITH COASTAL APPLICATIONS
Fei Chai, Huijie Xue, Mingshun Jiang, and Andrew Thomas
School of Marine Sciences, University of Maine, Orono, ME 04469-5741, USA
A general ecosystem model structure was embedded into the Princeton Ocean Model (POM). The coupled circulation-ecosystem model structure and preliminary results of two coastal applications (the Gulf of Maine and the Pearl River Estuary) will be presented. Difficulties in developing such complex models will also be discussed.
For the Gulf of Maine ecosystem model, we include nitrate, silicate, two-sized phytoplankton and zooplankton, ammonium, and detritus nitrogen and silicate. This complex ecosystem model allows us to address multi-nutrient limitation on spring phytoplankton bloom dynamics and nutrient budget in the Gulf. Preliminary model results are compared with the satellite ocean color observations, but there are still some detailed issues need to be resolved. We are continuing to improve the circulation-ecosystem model by combining the results from the recent field programs and the satellite derived ocean color products. The second example of using the circulation-ecosystem model is for the Pearl River Estuary. The ecosystem model for the Pearl River Estuary is a phosphate limited model with two-sized phytoplankton, zooplankton, and detritus. The responses of the modeled circulation in the Pearl River Estuary to winter and summer monsoons will be presented. The modeled temperature and salinity fields during the two seasons have been compared with the limited observations.
NESTED-GRID POM: AN APPLICATION FOR THE PORT OF NEW YORK/NEW JERSEY WATER LEVEL AND CURRENT NOWCAST/FORECAST MODEL SYSTEM
Manchun Chen1, Lie-Yauw Oey2, and Eugene Wei1
1. Coast Survey Development Laboratory, NOAA/NOS, N/CS13, 1315 East West Highway, Silver Spring, MD 20910
e-mail: manchun.chen@noaa.gov, phone: (301) 713-2809 x 109
2. Program in Atmospheric and Oceanic Sciences, Princeton University
Grid nesting modeling can provide detailed circulation information in the geographical area of particular interest using less computation time than a model with high resolution everywhere; it can also be used to provide open boundary conditions to regional models from larger-scale models. In this paper, we present two coupling methods that exchange fluxes across the grid interface: one-way and two-way couplings. In the one-way coupling, the fine grid is driven by fluxes from the coarse grid, and the fine grid does not feed back to the coarse grid. In the two-way coupling, the fine-grid fluxes also feed back onto the coarse grid.
The coupling techniques are tested in a series of experiments on rectangular and semi-circular basins with different bottom topography that ranges from constant-depth type to random, using the homogeneous three-dimensional POM. Averaged water level and current difference between the fine grid and the coarse grid over the fine grid area are calculated for assessing the nesting accuracy. The results indicate that two-way coupling is more accurate, and flux-coupling is more superior than simpler coupling that uses only the field variables.
The nesting method is then applied to a nowcast/forecast model that computes water levels and currents in the New York Harbor. The fine grid model cell size and the external time step are one half of the coarse grid. The nested-grid model gives more detailed eddy and shear-flow patterns that are otherwise absent in the coarse grid especially in the Newark Bay and the Kill van Kull.
"SEAGRID" MATLAB OCEANOGRAPHIC GRID GENERATOR
Charles R. Denham and Richard P. Signell
Woods Hole Field Center, United States Geological Survey
384 Woods Hole Road, Woods Hole, MA 02543
e-mail: cdenham@usgs.gov, rsignell@usgs.gov, phone: (508) 457-2318
"Seagrid" is an unusually flexible application for generating oceanographic model grids in Matlab. It uses Ives-Zacharias conformal mapping and a Poisson solver to fill the interior of a closed boundary with a manually-adjustable orthogonal grid. This object-oriented program incorporates bathymetry and coastline-masking, and it is able to read its own output, so that existing grids can be viewed and updated. "SeaGrid" interactivity is managed by a GUI event-handler, an "on-the-fly" dialog-maker, and a comprehensive "help" system.
TIDE-INDUCED MEAN CROSS-FRONTAL FLUX
Changming Dong, Hsien-Wang Ou and Dake Chen
Lamont-Doherty Earth Observatory, Columbia University
The 2-D POM is applied to study the tide-induced cross-shelf processes. The cross-shelf circulations are simulated for the cases with fronts and without fronts over different topographies. The mean cross-frontal heat flux is calculated and its direction is found to be counter to the temperature gradient in the bottom boundary layer. The potential factors affecting the strength of the counter-gradient heat flux include the tidal amplitude, the mean thermal gradient, the topography. The mechanism for generating the counter-gradient heat flux is discussed, as well as its effect on the mean cross-shelf circulation. The study of cross-frontal flux will be extended to other scalar properties.
SIMULATIONS OF THE SEASONAL MIXED-LAYER WITH THE MELLOR-YAMADA TURBULENCE SCHEME
Tal Ezer
Princeton University
The seasonal changes in the upper ocean thermal structure and mixing are simulated and compared with observations in order to study the source of shortcomings in surface layer modeling. In particular, insufficient surface mixing and a too shallow summertime thermocline are common problems in ocean models that use the Mellor-Yamada turbulence scheme.
Sensitivity experiments explore the impact of surface forcing and turbulence parameterization on simulations of the oceanic mixed layer in a three-dimensional basin-scale model of the North Atlantic. The results quantify the improvement in the model upper ocean thermal structure, as surface forcing becomes more realistic from one experiment to another, for example, when 6h wind stress fields replace monthly mean climatological data, or when short wave radiation penetration is added to the surface heat flux forcing.
Recent improvement in the parameterization of dissipation in the Mellor-Yamada turbulence scheme by Mellor (1999), which has been tested before only with a one-dimensional model, has been evaluated now with a three-dimensional ocean model.
A COUPLED ICE-OCEAN MODEL OF THE ARCTIC OCEAN USING SATELLITE-DERIVED FORCING FIELDS
Kate S Hedstrom, Dale B Haidvogel and Jennifer Francis
Institute of Marine and Coastal Sciences, Rutgers University
In 1995, we completed a project with a coupled ice-ocean model of the Beaufort and Chukchi Seas. We have now extended these simulations in a number of respects. First, improved atmospheric forcing fields have been obtained by reanalysis of the TOVS satellite data in the Arctic. Preparation of a basin-wide set of surface forcing fields for the years 1980 through 1996 has been completed. We have also made substantial improvements to the coupled ice-ocean model. The ocean model is now finite-difference in all three dimensions, has a free sea surface, and utilizes a new advection scheme and time-stepping contributed by our colleagues from UCLA. The sea-ice momentum equations have also been rewritten by Paul Budgell to use the same gridding as the ocean (Arakawa-C), to use an improved elliptic solver, and to include a ridging scheme (Gray and Killworth). For the ice thermodynamics, we are now using Mellor-Kantha. In the future, we will also investigate one by Doug Martinson at LDEO.
A number of parameter studies have been performed on a coarse-resolution grid. We will present results from a multi-year simulation covering the Arctic Ocean at an average resolution of 35 km.
SIMULATION OF THE SOUTHEASTERN BERING SEA AND GULF OF ALASKA USING COUPLED REGIONAL/BASIN-SCALE MODELS
J. Hermann, D. B. Haidvogel, E. L. Dobbins, P. J. Stabeno, and D. Musgrave
IMCS, Rutgers University
We describe interannual comparisons of circulation fields in the Southeastern Bering Sea and the Gulf of Alaska, as deduced from two eddy-resolving regional circulation models based on the S-Coordinate Rutgers University Model (SCRUM). These models include both tidal and subtidal dynamics. Boundary forcing includes separate nudging of subtidal and tidal currents (M2, S2, N2, K1, O1), obtained from current meter records and global implementations of the Spectral Element Ocean Model (SEOM). Surface forcing includes daily wind and heat fluxes from NCEP reanalyses, ice melt in the Bering Sea, and monthly coastal freshwater runoff in the Gulf of Alaska. The spatial scales of model-generated eddies are comparable to scales observed in satellite imagery for the two regions. Results suggest preferred sites of eddy formation in the Gulf, and interannual differences in eddy activity within the Southeastern Bering Sea.
A THREE-DIMENSIONAL NUMERICAL STUDY OF EFFECTS OF TYPHOON ON OCEANOGRAPHIC CONDITIONS IN THE KOREA STRAIT
Chul-hoon Hong and Kyu-Dae Cho
Research Center for Ocean Industrial Development, Pukyong National University, Pusan 608-737, Korea
When typhoons passed around the Korea Strait during summer, some observation in this strait carried out by Mizuno et al. (1986) gives us the following oceanographic features; 1) the direction of the observed current was opposite to the northeasterly wind, 2) temperature rapidly increased having a time lag as the depth deepens after then decreased with oscillation. A primitive equation ocean model that makes use of a sigma-coordinate system and incorporates a typhoon model was used to examine the mechanism to generate these phenomena. The model region covers the East China Sea, the Yellow Sea, and a portion of the East Sea (Japan Sea). The model well reproduces the observed features, especially in temperature field, and clearly manifests how the above observed features happened. From early time when the typhoon was located in low latitude, an alongshore northward current in the west of Kyushu (hereafter West Kyushu Current) is generated by an alongshore wind in the typhoon. This current flows into the eastern channel, as a coastal jet, basically regardless to the wind field within the Korea Strait during this period. The above observed phenomena are generated by this current. The model results indicate that when typhoons pass around the Korea Strait, the West Kyushu Current is generated, and the oceanographic condition in the strait should be greatly influenced by this current.
APPLICATION OF THE POM TO THE WORLD OCEANS ON THE NUMERICAL WIND TUNNEL
Takashi Kagimoto, Yukio Masumoto, Toshio Yamagata
Institute for Global Change Research, Frontier Research System for Global Change, Seavans N 7F, 1-2-1, Shibaura, Minato-ku, Tokyo 105-6791, Japan
Masahiro Yoshida, Masahiro Fukuda, and Naoki Hirose
National Aerospace Laboratory, 7-44-1, Jindaiji-higashi, Chofu, Tokyo 182-0012, Japan
The Princeton Ocean Model has been configured for world ocean circulation, except for the Arctic Ocean, with 1/6 degrees horizontal grid resolution and 21 vertical levels. This model has been running on the "Numerical Wind Tunnel" at the National Aerospace Laboratory in Japan, using 64 processing elements. Because of the large memory requirement and the long calculation time, the model is parallelized using VPP Fortran, which looks like High Performance Fortran. So far we have carried out 15-year climatological calculation and have obtained the realistic ocean state such as currents and tracer fields. In the present study, not only the parallelization strategy of the code but also some intriguing examples of the simulated results are introduced.
REMEDIES FOR DIAPYCNAL MIXING OCCURRED DUE TO HORIZONTAL DIFFUSION IN THE APPLICATION OF SIGMA COORDINATE MODEL TO TOKYO BAY
Akinari Kaneko
Frontier Research System for Global Change, Sumitomo Hamamatsucho Bldg. 1-18-16, Hamamatsucho Monato-Ku, Tokyo, 105-0013, Japan
email: kaneko@frontier.esto.or.jp, phone: +81-3-5404-7850
Diapycnal mixing often occurs due to horizontal diffusion in the simulation of heat, salinity or water quality in estuaries by using à coordinated ocean model. The modified POM (Princeton Ocean Model) has been applied to Tokyo bay. Because spurious diapycnal mixing is focused on and true diapycnal mixing due to tides and winds might obscure this, only river flows are considered as external forces in the numerical experiments. In the cross sectional view of the results, the vertical velocities are magnified to meet the ratio of the horizontal length scale to the vertical one. Two remedies recommended in POM, Huang•s new diffusion formula1), Stelling's horizontal diffusion approximation2), and Song and Haidvogel• S coordinates3) have been examined for suppressing the spurious diapycnal mixing. It has been shown that the last two remedies have large effects on reducing the diapycnal mixing when they are used together with one of the remedies recommended in POM.
REFERENCES
1)W.Huang and M.Spaulding: J.Hydraul.Eng. 122(1996)349
2)G.S.Stelling and J.A.TH.M. Van Kester: Intl.J.Num.Methods Fluid 18(1994)915
3)Y.Song and D.Haidvogel: J.Comput,Phys. 115(1994)228
MODEL BOUNDARY CONDITIONS AND COASTAL DYNAMICS
John M. Klinck and Michael S. Dinniman
CCPO, Old Dominion University
Simulations using SCRUM consider wind driven flow over an idealized continental shelf with a submarine canyon. Simulations with upwelling or downwelling winds using open or periodic alongshore boundary conditions are analyzed. The offshore (oceanic boundary) is open in all cases. All the model runs were initialized with the flow at rest and run for 20 days. No nudging was used at any boundary since the true (or desired) solution there is not known.
Simulations with open boundaries develop an undercurrent along the upper continental slope, which is not present in the corresponding periodic cases. A net alongshore pressure gradient, allowing geostrophic cross-shore flow in the interior, develops in open cases. In contrast, the periodic simulations do not develop a net alongshore pressure gradient. Cases with upwelling winds have bigger differences due to boundary conditions than do downwelling cases. The periodic cases develop much stronger alongshore flow (a factor of two or more) which becomes dynamically unstable after two weeks of simulations. Coastal trapping retains energy in over the shelf in periodic simulations; open conditions allow export of transients. During startup, alongshore (upwelling) flow interacts with the canyon creating a compact cyclone on the shelf which advects into the downstream boundary. Local open boundary conditions reflect this eddy causing the flow to be increasingly perturbed. Different formulations in boundary conditions lead to minimal improvement. Various attempts at dissipative sponges were unsuccessful since the space and time variable boundary solution was not known and could not be calculated from the numerical solution. Weaker forcing or slower application of forcing decreased the size of the eddy, but did not eliminate the problem.
AN EXPERIMENTAL REAL-TIME NORTH PACIFIC OCEAN NOWCAST/FORECAST SYSTEM
D. Ko, R. Preller, M. Carnes, C Barron and P. Posey
Naval Research Laboratory
The ultimate goal of model development from applicational point of view is to apply the model to do nowcast/forecast. Applying the Princeton Ocean Model (POM) we have developed and have been running an experimental real-time nowcast/forecast system for the North Pacific Ocean. The North Pacific Ocean Nowcast/Forecast System or NPACNFS consists of a data-assimilating 1/4 degree North Pacific ocean model (NPACnf) based on the POM, a statistical ocean temperature/salinity analysis model called the Modular Ocean Data Analysis System (MODAS) and a real-time data stream from the NRL/NAVO satellite data fusion center and from NOGAPS (Navy Operational Global Atmospheric Prediction System) at the FNMOC (Fleet Numerical Meteorology and Oceanography Center). The NPACnf model is restarted everyday from previous nowcast fields. Once the model is restarted, it is continuously assimilating 3D synthetic temperature/salinity profiles generated by MODAS that are based on the satellite (GFO,TOPEX/Poseidon and ERS-2) measured sea surface height anomalies and the sea surface temperature, and forced by the surface heat fluxes and wind stresses from NOGAPS to produce the nowcast. Forecasts up to 72-hr are produced with available NOGAPS forecasts. In the meeting we will present the NPACNFS nowcast/forecast results and discuss the issues related to the real-time nowcast/forecast applying the ocean model.
A 3D FLOW/PARTICLE MODEL FOR PREDICTION OF OIL TRANSPORT IN COASTAL WATERS
K.A.Korotenko
P.P.Shirshov Institute of Oceanology, 36 Nakhimovskiy Pr., 117851 Moscow, Russia
e-mail: koroten@aha.ru, kkorotenko@rsmas.miami.edu
A hybrid model to predict processes of the pollution transport in sea coastal waters is developed on a base of the Princeton Ocean Model and random particle technique. In the model to be discussed, the "particles" represent the natural particles in a very direct sense. The number of model particles used in an application is determined by statistical considerations. Various algorithms to be implemented in the model to describe different mechanisms of particle interactions between each other and between the particles and ambient water allow to use the model for a wide range of purposes and various inconservative admixtures and pollutants.
Oil is modeled as a mixture of particles with certain properties representing behaviour of different fractions of the substance. Implementation of the hybrid model in numerical experiments with oil plumes and jets in the coastal waters of the Caspian Sea and their results are discussed.
FITTING HYDROGRAPHIC AND BIOLOGICAL DATA IN THE CALCOFI DOMAIN USING ROMS (GREENS FUNCTION AND ADJOINT DEVELOPMENT)
Emanuele Di Lorenzo, Arthur J. Miller, Bruce Cornuelle, Terry Chereskin, John Moisan
Scripps Institute of Oceanography
ROMS has been setup to study the physics and biology of the California Current in the CalCOFI region. Vertical resolution ranges from 5 meters in shallow water to 50 meter in deeper water. The horizontal grid is 120x80 grid point with a resolution of about 11 km. In the dynamical model five additional advection-diffusion equations have been added to include Nitrate, Ammonia, Detritus, Phytoplankton and Zooplankton.
At each time step a separate module computes the source sink term of the bio-equation. The module used to compute the interactions between the bio-tracer (source-sink term) comes from an original 1-D ecosystem model (Fasham et al., JMR, 48, 1990).
The model is able to capture the dynamics of eddies and the effect of topography. Climatological runs show the presence of a surface meandering current located in the same region of the observed California Current and strong mesoscale variability.
An inverse method (Green's function) is used to find the optimal initial condition for the model in order to fit the observations (ideally within observational errors) of a single CalCOFI survey in space and time. CalCOFI data from surveys taken during the 1997-98 El Nino, drifters and altimeter data are used to initialize the model. During the integration no furhter assimilation is used to improve the model fit. This allows us to investigate the dynamical and ecosystem balances that hold during the survey.
The same inverse problem can be approached by developing an adjoint code of the forward model. In order to do this we have tested the TAMC (Tangent linear Adjoint Model Compiler) on the ROMS code. Some preliminary results suggest that the structure of the ROMS code is not suitable as it is now to use with the TAMC.
A GENERALIZATION OF A SIGMA COORDINATE OCEAN MODEL AND AN INTERCOMPARISON OF MODEL VERTICAL GRIDS
George Mellor1, Sirpa Hakkinen2, Tal Ezer1 and Richard Patchen3
1. Atmospheric and Oceanic Sciences Program, Princeton University
2. NASA Goddard Space Flight Center
3. Dynalysis of Princeton
The Princeton Ocean Model is generalized so that the distribution of numerical levels need not ne constrained as in the conventional sigma coordinate system wherein the spacing of models are in the same proportion independent of bottom depth. The new formulation eliminates this proportionality constraint; also the spacing may vary temporally. The free surface capability is retained. The bottom level can follow the bottom topography or the same model can easily convert to a z - level coordinate system with a stepped bottom topography. A great variety of vertical coordinate schemes are future possibilities.
In this paper, a conventional z - level coordinate system, a conventional s - coordinate system and a generalized sigma scheme are compared for the case of a simplified, wind driven, northern hemisphere basin. Therefore, these are model comparisons with identical algorithms except for the different vertical grids. For smaller viscosity, the z - level calculations are noisy whereas the sigma schemes are not. The full paper has been submitted to: Ocean Forecasting: Theory and Practice, Springer Pub., N. Pinardi (Ed.).
Work in progress extends the above results to comparisons of the same vertical coordinate systems, but with surface forcing mimicking equator to polar heat fluxes. One objective, other than a general comparison of model results, is to understand the role of the bottom boundary layer in the deep water formation process.
ON THE DYNAMICS OF THE ZAPIOLA ANTICYCLONE
Anne P. de Miranda (Darbon),
LEGI-IMG, B.P. 53, 38041 Grenoble Cedex 09, FRANCE
e-mail: Anne.de-Miranda@hmg.inpg.fr, phone: (33 or 0) 476 825065
Recent observations obtained in the South Atlantic suggest the existence of a strong anticyclonic flow positioned over a major bottom topographic feature known as the Zapiola Drift. Here a numerical simulation of the South Atlantic is described in which this anticyclone is reproduced, and the model is used to diagnose the dynamics maintaining this flow. With a mean barotropic transport of 140 Sv and bottom velocities of 10 cm/s, the simulated Zapiola Anticyclone compares well with in situ observations. Furthermore, the model surface eddy kinetic energy shows a local minimum over the drift, in agreement with observations from TOPEX/POSEIDON. Numerical experiments show that the circulation feature is sensitive to the intensity of the eddy field and to the particular value of the bottom friction. Both of these tendencies are in agreement with a theoretical explanation of the Zapiola Anticyclone that has recently appeared elsewhere. Thus we argue that the anticyclone is maintained by eddy-driven potential vorticity fluxes accelerating flow within topographically closed, ambient potential vorticity contours. As far as we know, this is the first South Atlantic simulation to reproduce the Zapiola Anticyclone in a realistic fashion. The quantitative success of this experiment is attributed to the use of a topography following (sigma) coordinate in a spatially well resolved model. A comparison with a z coordinate model will be made which show no evidence of the Zapiola Anticyclone.
THE SENSITIVITY OF IAS-POM'S RESPONSE IN THE NORTHERN GULF OF MEXICO TO VARIOUS SYNOPTIC WIND FIELDS CHARACTERIZING THE PASSAGE OF TROPICAL STORM JOSEPHINE
Christopher N. K. Mooers and Lianmei Gao
Ocean Prediction Experimental Laboratory(OPEL), RSMAS, University of Miami
e-mail: cmooers@rsmas.miami.edu, lgao@rsmas.miami.edu
Tropical Storm Josephine formed in the southwestern Gulf of Mexico in early OCT 96, intensified and nearly covered the Gulf as it advanced, and exited through the northeastern Gulf after approximately a week. The response of the northern Gulf shelf waters to Josephine was impulsive and intense as documented in a few dozen surface drifters off the Florida Panhandle and a few moored current meters off Texas. The observed transient flow was westward.
Numerical simulations of the transient response with IAS-POM were driven by several synoptic wind datasets: NSCAT, NCEP Global Analysis, NCEP Eta-30 km, and CARDONE Hindcast. The wind datasets were validated versus available NDBC buoy wind data. Furthermore, it was possible to explore the sensitivity of IAS-POM's response to different wind datasets, and their space-time resolution, versus the observed response. (IAS: Intra-Americas Sea; ca. 6 to 30N, 55W west to the Americas.) Overall, the simulated response to the downwelling-favorable winds was dominated by a tranisent, topographically trapped flow of order 0.5 m/s stretching over 1,000 km along the northern and western Gulf that propagated like a coastally-trapped wave. Because several tropical and extratropical cyclones traverse the Gulf Mexico per year, the Gulf 's response, as occurred for Josephine, may be a significant mechanism for cyclonic alongshore displacement/exchange of water volumes.
In summary, though the validation of model winds versus observed winds was not perfect, nor was the validation of simulated versus observed currents and trajectories, it appears that mesoscale-resolution NSCAT and Eta winds, together with POM, are approaching useful accuracy in simulating the coastal ocean's current response to storm systems.
LABORATORY/NUMERICAL MODEL COMPARISONS OF FLOW OVER A COASTAL CANYON
Nicolas Perenne, Dale B. Haidvogel, Don L. Boyer
IMCS, Rutgers University
We explore the ability of laboratory models to provide useful benchmarks for ocean circulation models. The test case considered here-that of the flow driven by an oscillatory forcing over a submarine canyon-involves background rotation, density stratification and steep topography. Velocity fields measured by Particle Tracking Velocimetry and time series of density fluctuations are quantitatively compared to those obtained from a terrain-following, high-order, finite element ocean model. Comparison of the two models shows good overall agreement in the structure and magnitude of the strongest residual currents which occur at the level of the shelf break in our configuration. The associated residual vorticity field at the shelf break is also structurally consistent between the two models, although the residual divergence is not. Residual currents higher up and lower down in the water column are weaker than at the shelf break, and the agreement between the laboratory and numerical models is less good at these levels, possibly indicative of the controlling influence of the surface and bottom boundary layers.
MEDITERRANEAN SIMULATIONS USING MASSIVELY PARALLEL VERSIONS OF THE POM CODE
S.Piacsek+, W.Oberpril*, M.Okeefe*, M.Young$
+Code 7322,Naval Research Laboratory,Stennis Space Center, MS 39529
$Code 5593,Naval Research Laboratory,Washington,D.C. 20375
*Laboratory for Computer Science and Engineering,Univ.of Minnesota,MN
Simulations of the Mediterranean using MPI (message passing) and HPF (high performance fortran) versions of the POM code will be presented. The scalability of these codes and performance on the SGI O2K (Origin2000) and the Cray T3E platforms will be discussed, as well some preliminary results of simulations at 3.5 km resolution.
ON THE PRESSURE GRADIENT ERROR IN SIGMA COORDINATE OCEAN MODELS
Julie D. Pietrzak and Nicolai Kliem*
Civil Engineering, Delft University of Technology, Stevinweg 1, Delft, The Netherlands
*Danish Meteorological Institute, Lyngbyvej 100, Copenhagen, Denmark
The sigma co-ordinate transformation is frequently used in three dimensional primitive equation models. Unfortunately it is well known that this transformation can produce an error in the calculation of the horizontal pressure gradient terms. Understanding the cause and significance of this error for coastal simulations is important as these regions typically have steep topographic gradients and large density differences. However, the importance of this error in simulations of coastal currents is difficult to test in reality. The competing effects of the models dynamics, its sensitivity for example, to the chosen grid resolution or numerical techniques are hard to separate. In this talk we identify the pressure gradient error and assess its magnitude in relation to two analytical reference solutions. A number of commonly used techniques to reduce the error are compared. Of these techniques, z-level based pressure gradient calculations are shown to improve the simulation results. A step in the direction of a more realistic simulation is then taken and in this case a laboratory experiment is used as our reference solution. The results demonstrate that the flow in the Skagerrak, a region with steep topography and large density gradients, can be simulated using a sigma co-ordinate model. However, errors do exist which need to be taken into account when interpreting the model results. This is important as the Skagerrak dominates the North Sea and Baltic Sea System.
ADVECTION CONFUSION
Julie D. Pietrzak1and Rich Signell2
1Civil Engineering, Delft University of Technology, Stevinweg 1, Delft, The Netherlands
2U.S. Geological Survey, 384 Woods Hole Road, Woods Hole, MA 02543-1598
This talk addresses the issue of higher order advection schemes in ocean models. While particular attention is paid to the role of forward in time advection schemes in ocean modelling studies, this talk attempts to show the relationship between a number of commonly used advection schemes. The emphasis is on the reduction of confusion when selecting an appropriate advection scheme for the study of a given physical process. The competing effects of numerical diffusion versus numerical dispersion (or wiggles) are addressed. In so doing the underlying and simple relationships between apparently complicated schemes are described. Simple flow tests are discussed in order to highlight these relationships. Finally more realistic coastal flow problems are addressed and results from an idealised river plume study are presented.
A POM VARIANT FOR MODELING INTERNAL TIDES NEAR THE CRITICAL LATITUDE
Robin Robertson
Lamont-Doherty Earth Observatory, Rt. 9W, Palisades, NY 10964
e-mail: rroberts@ldeo.columbia.edu, phone: (914) 365-8576
While attempting to model the M2 internal tide near its critical latitude in the Weddell Sea using POM, an unrealistic oscillation developed in the velocity field. This oscillation developed whether or not stratification was present whenever a sloping bottom was near the critical latitude. Although the oscillation had a frequency near the M2 tidal frequency and many of the characteristics of an internal wave, it was believed to be unrealistic, since it was not present in any of the observations for this region, and since it developed in a homogeneous ocean, which violates linear internal wave theory. Further investigation revealed it to be depth-variable inertial oscillations, which resulted from the baroclinic pressure gradient term. The pressure correction to density results in depth-dependence in the baroclinic pressure gradient term over a sloping bottom. When the domain is near the critical latitude, this depth-dependent baroclinic pressure gradient term initiates and drives depth-variable inertial oscillations over the slope even with a homogeneous ocean. Since the baroclinic pressure gradient term should be zero for a homogeneous ocean, both the term itself and this response are in error. An alternative method for dealing with density and the determination of the baroclinic pressure gradient term was developed in order to alleviate this problem. Using the new method, POM was able to generate realistic internal tidal velocity fields. The erroneous depth-variable inertial oscillations were no longer present. Additionally, the new velocity fields both agreed with observations and with the predictions of linear internal wave theory.
CFC SIMULATIONS IN THE WEDDELL SEA
Christian B. Rodehacke, Aike Beckmann, Hartmut H. Hellmer, and Wolfgang Roether
Institute of Environmental Physics, University of Bremen
The Weddell Sea is one of the major source regions of AABW, which water mass provides much of the ventilation of the deep world ocean. Tracer (CFC) simulations have been carried out as a part of the BRIOS (Bremerhaven Regional Ice-Ocean Simulations) project. This project uses a stand alone ocean model (SPEM-type) focused on the Weddell Sea sector with the aim to provide information on circulation and property transport in the area. The purpose of the tracer simulations is to help to validate rates of formation and interior pathways of deep water masses in the model. The model has a horizontal resolution of 1.5 degrees and is initialized with hydrographic data from the Southern Ocean Atlas. The surface forcing data are averaged monthly means from a stand alone sea ice-mixed layer model, based on 6-hourly ECMWF(1985-1993) re-analysis data. A number of tracer simulations for CFC-11 and CFC-12 have been sucessfully completed. They illustrate the major areas of deep water formation, the spreading patterns of the freshly ventilated deep/bottom water in the Weddell Gyre, and can be used to derive ventilation timescales. The results are discussed comparing simulated tracer distributions with observations.
DEVELOPING A MOVABLE NESTED-MESH SIGMA-COORDINATE OCEAN MODEL FOR AIR-SEA INTERACTION STUDIES (withdrawn)
Clark Rowley
Graduate School of Oceanography, URI
A mesh movement scheme is implemented in a multiply-nested sigma-coordinate ocean model. Mesh movement can be specified, or determined in the course of the model run so as to follow an evolving oceanic feature, such as a wave front or propagating eddy, or atmospheric forcing, such as a tropical cyclone. Mass, heat, and momentum are conserved during the movement. The mesh movement scheme is tested in idealized and realistic configurations of the model. The tests demonstrate that the solutions in the fine-mesh region of the nested meshes reproduce well the equivalent solutions from uniform fine-mesh models. Because the representations of bottom topography and coastlines change as nested fine-resolution meshes move across the model domain, special techniques are implemented to treat these changes. An external/internal mode-splitting technique in the time integration scheme is used. Idealized and realistic tests of the treatment of the bottom topography and coastline in the presence of moving meshes are described. The model is applied for simulations of the ocean response to tropical cyclones, in which the moving meshes maintain high resolution near the cyclone center. Simulations of the ocean response to several landfalling western Pacific tropical cyclones are performed.
GLOBAL SEASONAL DISTRIBUTIONS OF OCEAN CHLOROPHYLL: COUPLING A BIOLOGICAL MODEL WITH POM (withdrawn)
Richard Scheper and Watson Gregg
NASA/GSFC
The Princeton Ocean Model (POM) was coupled with and biological model to produce a representation of seasonal chlorophyll distributions globally. The biological model includes the dominant source and loss terms associated with phytoplankton, and POM provides the circulation fields and turbulent mixing processes to determine the distribution of phytoplankton and nutrients. The biological model contains 3 nutrients (nitrate, ammonium, and silicate) and three functional phytoplankton groups (diatoms, chlorophytes, and picoplankton).
The coupled model is run for three years and the distributions evaulated with respect to ocean color images (CZCS, SeaWiFS), and nutrient archives from NOAA/NODC.
THE GALVESTON BAY/HOUSTON SHIP CHANNEL NOWCAST/FORECAST SYSTEM AND ITS RELATION TO THE NAVIGATION CHANNEL PROBLEM
Richard A. Schmalz, Jr.
NOAA/NOS, Coast Survey Development Laboratory, 1315 East-West Highway, Rm 7824, Silver Spring, Maryland 20910
e-mail: Richard.Schmalz@noaa.gov
The National Oceanic and Atmospheric Administration installed a Physical
Oceanographic Real Time System (PORTS) in June 1996 in Galveston Bay. Observations of water surface elevation, currents at prediction depth (4.6m) as well as near-surface and near-bottom temperature and salinity, and meteorological information are available at six-minute intervals. To enhance nowcast/forecast capability, the Mellor-Blumberg (1987) three-dimensional hydrodynamic model has been extended to include drying/wetting and a flux corrected transport scheme to treat the large horizontal salinity gradients. A fine resolution Houston Ship Channel model (Schmalz, 1997) has also been incorporated into the nowcast/forecast system driven by the Bay model using a one-way coupling scheme. The nowcast component works directly from the PORTS universal flat files.
During the forecast, the NWS Aviation and Extratropical Storm Surge Models are used to provide the meteorological and Gulf of Mexico subtidal water level residual
forcings, respectively. The nowcast/forecast system has been used to provide a daily 36-hour forecast, in itiated from a nowcast based on the previous 24 hours, using both bay and channel models in a pseudo-operational setting since April 1999. To seek improvements in the prediction of the current and density structure, a joint NOS-Sea Grant sponsored ADCP/CTD survey of a four nautical mile segment of the Houston Ship Channel is planned for early September 1999. The paper summarizes the present nowcast/ forecast system status and results achieved to date. A second focus is on its relationship to the navigation channel problem in general. Items to be considered include: 1) navigation channel geometry and grid generation, 2) the suitability of the hydrostatic assumption, 3) an alternate vertical coordinate, and 4) the need for improved spatial measurement density in conjunction with potential PORTS expansion activities.
AN ADAPTIVE VERTICAL COORDINATE FORMULATION FOR OCEAN MODELS
Y.Tony Song and Yi Chao
JPL, California Institute of Technology, Pasadena, California 91109
An adaptive method used in numerical ocean models is introduced for optimally simulating both coastal and deep-ocean circulation features. The method is based on the combined techniques of the generalized vertical coordinate system of Kasahara (1974), the level nudging scheme of Bleck and Benjamin (1993), and the general pressure gradient formulation of Song (1998), and is implemented into the S-Coordinate Rutgers University Model (SCRUM). The model is allowed to use desired vertical grid structure, such as the hybrid grid of z-coordinate and sigma-coordinate, for a specific problem.
Two different problems: a coastal upwelling and a global 1x1 degree model, are used to demonstrate the feasibility of the adaptive method. Our preliminary results show that the new method is capable of handling both shallow- and deep-ocean processes.
STUDIES OF TRACER DISPERSION IN COASTAL AREAS AND ESTUARIES (withdrawn)
Redoine Tahiri
LIMNOCEANE-University of Neuchatel Address: Rue Emile argand, 11. CH 2000 Neuchatel-Switzerland.
In this paper we present a three-dimensional model describing the physical processes which influence the distribution of dissolved tracer in shallow water. The model is based on the advection-diffusion-reaction equation. A description of the model numerical algorithm is presented. The iteration method inside a time step to determine the new concentration is discussed in detail. Criteria for stability are derived. The scheme developed in this paper is tested for the transport of Gaussian disturbance in a three-dimensional rotating flow field. The numerical solutions obtained are compared with the analytical solution. The model is then applied for a transport of diffusive tracer in the Gulf of Lion (France).
A DOMAIN DECOMPOSITION METHOD FOR THE ADVECTION-DIFFUSION PROBLEM (withdrawn)
Redoine Tahiri
LIMNOCEANE-University of Neuchatel Address: Rue Emile argand, 11. CH 2000 Neuchatel-Switzerland.
The main goal of this paper is to discuss the numerical solution of initial boundary value problems for the advection-diffusion equation, describing the transport of pollutant in shallow water, by domain decomposition methods. We formulate the advection-diffusion problem whose solution is to be addressed, and introduce an equivalent formulation which the basis of the domain decomposition method discussed here.
We discuss the combined effect of domain decomposition and time discretization. Finally, we analyze the result of numerical experiments which show that domain decomposition methods have a good potential for the solution of problems related the advection-diffusion equation.
DESIGNING A COUPLED ICE-OCEAN MODEL OF THE WEDDELL SEA. PART II: SEA ICE
R. Timmermann, A. Beckmann, H.H. Hellmer
Alfred-Wegener-Institute for Polar and Marine Research. Bremerhaven, Germany
As part of the BRIOS (Bremerhaven Regional Ice-Ocean Simulations) project a coupled ice-ocean model for the Weddell Sea has been developed. It is based on a modified version of the sigma-coordinate primitive equation ocean model SPEM and a dynamic-thermodynamic sea ice-model with viscous-plastic rheology (based on Hibler/Lemke) which also provides the thermohaline forcing within the cavities beneath the Antarctic ice shelves. Model runs are initialized with data from the Hydrographic Atlas of the Southern Ocean and forced with wind, cloudiness and temperature fields of the 6h-reanalyses of the European Centre for Medium Range Weather Forecasts (ECMWF). Besides the presentation of model results, the talk will shed lights on the model configuration, coupling strategy and parametrizations in the coupled model.
SIMULATION OF PENOBSCOT BAY CIRCULATION, APRIL - SEPTEMBER 1998
Huijie Xue1, Yu Xu1, and David Brooks2
1. School of Marine Sciences, University of Maine
2. Department of Oceanography, Texas A& M University
Penobscot Bay is the largest estuarine embayment along the Maine coast. It can be characterized by two deep channels on its eastern and western sides, which are separated by several islands and a shoal region in the middle of the Bay. Subtidal circulation in Penobscot Bay is influenced by winds, fresh water discharge from the Penobscot River, and the southwestward Maine Coastal Current flowing pass the mouth of the Bay.
The Princeton Ocean Model is forced by the observed winds and river discharges for the period between April and September 1998. Open boundary condition is specified using the results from a Gulf of Maine climatological model. The rate of change in temperature and salinity and the seasonal development of stratification are well simulated. In the model, both channels in the eastern and the western bay act like classic two-layer estuaries with outflow at the surface and inflow at the bottom, which agreed with the observed winter regime. Summer observations show that the two-layer structure remained in the eastern bay, while in the western bay inflow appears in the upper water column on the western side of the channel and outflow on the eastern side of the channel and in the lower water column on the western side. Seasonal transition seems to associate with density changes of the coastal water. We are incorporating the data from the June, July and August hydrographic surveys into the model and hope to improve model simulation of the summer circulation pattern.
APPLICATION OF A PHYSICALLY AND NUMERICALLY ROBUST OCEAN MODEL TO A STUDY OF GULF STREAM DYNAMICS (withdrawn)
S.-S. Yuk and G. A. Zarillo
Florida Institute of Technology, Melbourne, FL
The overall goal of this research is to develop and apply a physically and numerically robust three-dimensional model to simulate ocean circulation and transport. Over the past three decades a focus has been on the complicated dynamics of meso-scale eddies and meandering motions of major ocean circulation systems, as well as prediction of related ocean phenomena that may occur due to global climatic change. The widely applied eddy-resolving models using primitive equations of motion produce simulations over a range of spatial and temporal scales of ocean dynamics. The sigma coordinate system has been used to follow complex topography and can be used to resolve benthic boundary dynamics. However, persistent and serious problems with truncation errors arise when abrupt changes in topography are included in the model domain. Thus, ocean variables are often predicted in an unrealistic manner due to the need for smoothing of model topography to prevent excessive numerical error. In conventional models over high relief topography, the truncation error in the pressure gradient and the horizontal diffusion terms results in numerical instability and model "crashes". In the mixing processes of temperature and salinity associated with abrupt topography, diapycnal phenomena in the s -coordinate model can also give a rise to unrealistic eddy motions, nonsensical strong upwelling phenomena, and coastal jet flow during long numerical experiments. The overestimation of the horizontal diffusion breaks down the continuity of the momentum, and results in unrealistic prediction of temperature and salinity.
The conventional model equations have been re-formulated to exclude depth gradients in the momentum terms and without multiplication of depth in pressure force term. This form can be considered more primitive than conventional formulations. The numerical truncation errors are significantly reduced in the reformulation and similar in magnitude in the pressure gradient force and horizontal advective terms.
To fully develop the model and test its integrity, the model is applied to the western boundary region of the Gulf Stream where important dynamics such as meandering motions and meso-scale eddy motions occur. Results to date demonstrate that the reformulated model can be applied over sharp, high relief topography without smoothing in the model domain. Continuous model runs longer than one year have been completed. Model results and analysis of data indicate that Gulf Stream dynamics and energetics can be realistically simulated. Results demonstrate that the re-formulated model has the potential to be applied as a tool for studying meso- to macro-scale features of major ocean circulation systems during very long-term simulations without serious numerical error.
The specific objectives of this study include: 1) to reformulate the governing equation in order to minimize numerical truncation errors that are evenly distributed between advective terms and pressure gradients, 2) to accurately state the equations for solutions in finite difference form, 3) to apply the appropriate open boundary conditions to guarantee the long-term simulations and to attain the quasi-steady state motion in the prognostic mode, 4) to simulate meandering motions and meso-scale eddy motions at realistic time and spatial scales, and 5) to quantify the dynamic scale and energetics at various scales of motions associated with the Gulf Stream.
SUBTIDAL WATER LEVEL ASSIMILATION INTO EAST COAST OCEAN MODEL
Aijun Zhang, Eugene Wei, and Bruce Parker
Coast Survey Development Laboratory, NOAA/NOS, N/CS13, 1315 East West Highway, Silver Spring, MD 20910
e-mail: aijun.zhang@noaa.gov, phone: (301) 713-2809 x 113
In this study, the two dimensional Princeton Ocean Model (POM) has been implemented to simulate the wind-driven subtidal water level along the U.S. East Coast using National Centers for Environmental Prediction’s (NCEP) meso-scale Eta data Assimilation System (EDAS) analyzed winds as surface forcing. An optimal data assimilation (ODA) technique has been used to assimilate observed subtidal water levels at 15 coastal stations into the model. In the ODA procedure, the cost function is defined as the subtidal water level misfit, and its gradient is determined by the multi-perturbation finite difference method. The combination of the Davies, Swann, and Compey (DSC) and Powell’s algorithm is implemented to minimize the cost function. The method has been tested using identical twin experiments in which the "observations" are generated by numerical model with predefined control variables. Real observed subtidal water levels are then used to determine the optimal wind drag coefficient. Sensitivity experiment results are compared with observations and show that the data assimilation process results in an appreciable reduction in the Root-Mean-Square (RMS) difference between observed and simulated subtidal water levels as well as an increase in the correlation coefficients.
III. Sigma Coordinate Ocean Model Users Meeting'99 Attendees
Dr. Quamrul Ahsan
HydroQual, Inc.
1 Lethbridge Plaza
Mahwah, NJ 07430
Email: qahsan@hydroqual.com
Phone: (201) 529-5151 x7135
Hernan G. Arango
IMCS, Rutgers University
71 Dudley Road
New Brunswick, NJ 08901-8521
Email: arango@imcs.rutgers.edu
Phone: (732) 932-6555 x266
Dr. Bernard Barnier
CNRS
LEGI, BP53, 38041 Grenoble Cdex 9, France
Email: bernard.barnier@hmg.inpg.fr
Phone: (33) 4 76 82 50 66
Dr. Aike Beckmann
Alfred-Wegener-Institute for Polar and Marine Research
D-27515 Bremerhaven, Germany
Email: beckmann@awi-bremerhaven.de
Phone: +49-471-4831-793
Dr. David Brooks
College of Geoscience
Texas A&M University
College Station, TX 77843-3148
Email: dbrooks@cobscook.tamu.edu
Phone: (409)845-3651
Dr. Fei Chai
School of Marine Sciences, University of Maine
223 Libby Hall
Orono, ME 04469-5741
Email: fchai@maine.edu
Phone: (207)581-4317
Manchun Chen
NOAA/NOS
1315 East West Highway, N/CS13
Silver Spring, MD 20910
Email: manchun.chen@noaa.gov
Phone: (301) 713-2809 x 109
Chuck Denham
U.S.Geological Survey
384 Woods Hole Road
Woods Hole, MA 02543
Email: cdenham@usgs.gov
Phone: 508-457-2318
Changming Dong
Lamont-Dohert Earth Observatory, Columbia University
205 Oceano. BLD.
P.O. Box 1000
Route 9W
Palisades, New York 10964-8000
Email: cdong@ldeo.columbia.edu
Phone: 914-365-8529
Shawn Donohue
Royal Military College of Canada (RMC)
P.O. 17000 STN FORCES
KIKGNSTON, ONT. CANADA
K7K-7B4
C/O DEPT OF PHYSICS
Email: donohue-s@rmc.ca
Phone: (613) 544-8857
Dr. Tal Ezer
AOS, Princeton University
P.O.Box CN710
Sayre Hall
Princeton, NJ 08544-0710
Email: ezer@splash.princeton.edu
Phone: (609) 258-1318
Prof. Dale Haidvogel
IMCS, Rutgers University
71 Dudley Road, New Brunswick, NJ 08901
Email: dale@ahab.rutgers.edu
Phone: (732)-932-6555 x256
Dr. Ruoying He
Univ. of South Florida
Dept. of Marine Science
140 Seventh Ave. South
St. Petersburg FL 33701
Email: ruoying@ocgmodel.marine.usf.edu
Phone: (727) 553-1627
Kate Hedstrom
IMCS, Rutgers University
71 Dudley Road, New Brunswick, NJ 08901
Email: kate@ahab.rutgers.edu
Phone: 732-932-6555 x258
Dr. Rob Hetland
WHOI/USGS
384 Woods Hole Road
Woods Hole, MA 02543
Email: rob@re.ocean.fsu.edu
Phone: 508-457-2229
Dr. Robin Hewitt
DERA
Underwater Sensors and Oceanography Department,
Defence Evaluation and Research Agency, DERA Winfrith,
Dorchester DT2 8XJ, Dorset,
United Kingdom
Email: RHEWITT@MAIL.DERA.GOV.UK
Phone: +44 (0) 1305 212309
Dr. Chul-hoon Hong
Pukyong National University
599-1 Daeyeondong, Namgu Pusan 608-736, Korea
Email: hongch@dolphin.pknu.ac.kr
Phone: 82-51-620-6891
Dr. Minshun Jiang
School of Marine Sciences, University of Maine
Orono, ME 04469-5741
Email: jiang@athena.umeoce.maine.edu
Phone: (207)581-4349
Dr. Takashi Kagimoto
Institute for Global Change Research/Frontier Research System for Global Change
SEAVANS North BLDG. 7F 1-2-1 Shibaura,
Minato-Ku, Tokyo 105-6791, Japan
Email: kagimoto@frontier.esto.or.jp
Phone: +81-3-5765-7127
Dr. Akinari Kaneko
Frontier Research System for Global Change
SUMITOMO HAMAMATSUCHO BLDG. 1-18-16
HAMAMATSUCHO MINATO-KU TOKYO 105-0013, JAPAN
Email: kaneko@frontier.esto.or.jp
Prof. John Klinck
CCPO, Old Dominion University
Crittenton Hall
Norfolk, VA 23529
Email: klinck@ccpo.odu.edu
Phone: 757-683-6005
Dr. Dong S. Ko
Naval Research Laboratory
Code 7320 NRL, Stennis Space Center, MS 39529
Email: ko@nrlssc.navy.mil
Phone: (228) 688-5448
Dr. Konstantin Korotenko
Marine Turbulence Lab
P.P.Shirshov Institute of Oceanology
36 Nakhimovskiy pr.
117851 Moscow, Russia
E-mail: koroten@aha.ru
Phone: (095) 2182114
Honghai Li
HydroQual, Inc.
1 Lethbridge Plaza, Mahwah, NJ 07430
Email: hli@hydroqual.com
Phone: 201-529-5151 X 7116
Emanuele Di Lorenzo
Scripps Institution of Oceanography
9344-F Redwood Drive, La Jolla, 92037 CA
Email: edl@ucsd.edu
Phone: 619 534-6397
Dr. Carlos Lozano
Harvard University
29 Oxford Street
Cambridge, MA 02138
Email: lozano@pacific.harvard.edu
Andrea C. Mask
Center for Ocean-Atmosphere Prediction Studies/Florida State University
Suite 200, Johnson Building, 2035 E. Paul Dirac Drive,
Tallahassee, FL 32306-2840
Email: amask@coaps.fsu.edu
Phone: (850) 644-4174
Prof. George Mellor
AOS, Princeton University
P.O.Box CN710
Sayre Hall
Princeton, NJ 08544-0710
Email: glm@splash.princeton.edu
Phone: (609) 258-6570
Dr. Anne P. de Miranda
LEGI-IMG
BP 53, F-38041 GRENOBLE
Email: Anne.de-Miranda@hmg.inpg.fr
Phone: (33) 476 825 065
Prof. Christopher N. K. Mooers
OPEL/AMP/RSMAS/U. MIAMI
4600 Rickenbacker Cswy.
Miami, FL33149-1098
Email: cmooers@rsmas.miami.edu
Phone: 305-361-4088
Dr. Steve Piacsek
Naval Research Laboratory
Code 7322, NRL-SSC
Stennis Space Center, MS, 39529
Email: piacsek@new-jersey.nrlssc.navy.mil
Phone: 228-688-5316
Dr. Julie Pietrzak
Civil Engineering, Delft University of Technology
Stevinweg 1
Delft, The Netherlands
Email: j.pietrzak@ct.tudelft.nl
Dr. Terri Paluszkiewicz
ONR, Ocean Modeling and Prediction
800 North Quincy Street
Ballston, Tower one
Arlington, VA 22217-5660
Email: paluszt@onr.navy.mil
Phone: 703-696-4721
Dr. Robin Robertson
Lamont-Doherty Earth Observatory, Columbia University
Rt. 9W, Palisades, NY 10964
Email: rroberts@ldeo.columbia.edu
Phone: (914) 365-8576
Dr. Christian Rodehacke
University of Bremen, Institute of Enviromental Physics
post box 33 04 40, D-28334 Bremen, Germany
Email: c04m@zfn.uni-bremen.de
Phone: +49 421 218-4562
Dr. Richard A. Scheper
SAIC-GSC, NASA Goddard Space Flight Center (GSFC)
6603 Alta Avenue, Baltimore MD 21206
E-mail: scheper@salmo.gsfc.nasa.gov
Phone: 301-614-5507, 301-286-5624, 410-254-5973
Dr. Richard A. Schmalz, Jr.
NOAA/NOS CSDL N/CS13
SSMC-III Rm 7824
1315 East-West Highway
Silver Spring, MD 20910-3281
Email: Richard.Schmalz@noaa.gov
Phone: (301) 713-2809 x104
Dr. Rich Signell
U.S.Geological Survey
384 Woods Hole Road
Woods Hole, MA 02543
Email: rsignell@usgs.gov
Phone: 508-457-2229
Dr. Y. Tony Song
Jet Propulsion Laboratory, California Institute of Technology
JPL, MS 300-323, 4800 Oak Grove Dr. Pasadena, CA 91009
Email: song@pacific.jpl.nasa.gov
Phone: (626)393-4876
Dr. Michael W. Stacey
Department of Physics
Royal Military College of Canada
P.O. Box 17000 Stn Forces
Kingston, Ontario K7K 7B4
Canada
Email: stacey-m@rmc.ca
Tel: (613) 541-6000 (ext 6414)
Dr. Ralph Timmermann
AWI
27568 Bremerhaven, Germany
Email: rtimmerm@awi-bremerhaven.de
Phone: 0049/471/4831-785
Dr. Houjun Wang
Penn State University
603 Walker Building, University Park, PA 16802
Email: hxw7@psu.edu
Phone: 814-865-1678
Dr. Eugene Wei
NOAA/NOS
1315 East West Highway, N/CS13
Silver Spring, MD 20910
Email: eugene.wei@noaa.gov
Phone: (301) 713-2809 x 102
Danya Xu
School of Marine Sciences, University of Maine
208 Libby Hall
Orono, ME 04469-5741
Email: xudanya@163.net
Phone: (207)581-4319
Yu Xu
School of Marine Sciences, University of Maine
208 Libby Hall
Orono, ME 04469-5741
Email: yu_xu@umit.maine.edu
Phone: (207)581-4319
Dr. Huijie Xue
School of Marine Sciences, University of Maine
206 Libby Hall
Orono, ME 04469-5741
Email: hxue@maine.edu
Phone: (207)581-4318
Aihun Zhang
NOAA/NOS
1315 East West Highway, N/CS13
Silver Spring, MD 20910
Email: eijun.zhang@noaa.gov
Phone: (301) 713-2809 x 113