HOME | ABOUT US | CONTACT | REGIONAL MANAGERS
 
  Research & Development  
 
  Experimental Products / Prototypes  
     •   BookletChart™ (Experimental)  
     •   Water Level and Current Forecast Guidance  
 
  Hydrographic Support  
     •   Autonomous Underwater Vehicles (AUV)  
     •   Remotely Operated Vehicles (ROV)  
     •   Phase Differencing Bathymetric Sonar  
     •   Sound Velocity Profiler  
     •   Tidal Corrector Application (TCARI)  
 
  Products and Database Development  
     •   Bathymetric Warehouse  
     •   Coast Pilot Relational Database  
     •   VDatum Enhancements  
     •   Coastal Digital Elevation Models  
     •   Operational Data Acquisition & Archiving System  
     •   nowCOAST  
 
  Coastal & Estuarine Ocean Modeling  
     •   Forecast System Modeling  
     •   Inundation Modeling  
     •   Ecological Modeling Applications  
     •   Community Modeling & Standards  
     •   Model Evaluation Environment  
 
  Learn About  
     •   Coast Survey Development Laboratory  
     •   Hydrographic Survey Equipment  
     •   Datums and Transformations  
     •   How Hydrodynamic Models Are Used  
 
  Resources  
     •   Technical Reports & Publications  
     •   Coast Survey Partners  
     •   Standards and Requirements  
 
  Data Portals  
     •   nowCOAST: Real-Time Coastal Data Map Portal  
     •   Tides and Currents (General)  
     •   Physical Oceanographic Real-Time System (PORTS)  
     •   Operational Forecast Systems  
     •   Sea Nettle Forecast Guidance  
     •   North American Horizontal Datum Conversion Utility  
     •   University of New Hampshire, Joint Hydrographic Center  
 

Great Lakes Operational Forecast System (GLOFS)
The Marine Modeling and Analysis Programs (MMAP) branch of the Coast Survey Development Laboratory develops and evaluates nowcast/forecast modeling systems for U.S. seaports, estuaries, Great Lakes, and the coastal ocean. These nowcast/forecast systems are then implemented operationally at NOS’ Center for Operational Oceanographic Products and Services (CO-OPS).

During 2004-2006, MMAP in collaboration with CO-OPS, Ohio State University (OSU), and NOAA's Great Lakes Environmental Research Laboratory (GLERL), transitioned the Great Lakes Coastal Forecast System (GLCFS) developed by The Ohio State University and GLERL to operations at NOS. GLCFS was modified to fit into NOS’ Coastal Ocean Modeling Framework and renamed the Great Lakes Operational Forecast System (GLOFS).

The purpose of GLOFS is to provide short-term forecast guidance of water levels, currents, and water temperatures for the 5 Great Lakes for use by commercial shipping, recreational boating and emergency response communities.  GLOFS generates hourly nowcasts (analyses) and four times daily forecast guidance (out to 30 hours) using surface meteorological forcing derived from observations and forecast guidance from a National Weather Service operational weather prediction model. 

Model Description
GLOFS uses for its core hydrodynamic circulation model a version of the Princeton Ocean Model (POM).  POM was developed by Blumberg and Mellor and is a fully three-dimensional, non-linear primitive equation coastal ocean circulation model, with a second order Mellor-Yamada turbulence closure scheme to provide parameterization of vertical mixing processes.  The model solves the continuity equation, momentum equation, and conservation equation for temperature simultaneously in an iterative fashion. The resulting predicted variables are free surface elevation and three-dimensional velocity, salinity and temperature fields.

POM was modified by researchers at OSU and GLERL for use in the Great Lakes (Schwab and Bedford 1994, O’Connor and Schwab 1993).  For the rest of this summary, the modified version of the POM for the Great Lakes is referred to as POMGL. The Great Lakes are treated as an enclosed basin in POMGL. Therefore, there are no inflow/outflow boundary conditions: no fluid exchange between the lake and its tributarie
s, between the lake and ground water sources, or between the lake and anthropogenic influences.  Thus the model simulations do not include seasonal changes in lake-wide mean water level due to precipitation and evaporation. 
Separate lake forecast systems are run at CO-OPS for the 5 Great Lakes using the POMGL model but with unique grid dimensions, grid cell size, and sigma layers (see table below).

Lake

Grid Dimensions (x,y)

Cell Size

(km)

Sigma layers

Erie

81 x 24

5

11

Michigan

53 x 102

5

20

Ontario

61 x 25

5

20

Huron

81 x 75

5

20

Superior

61 x 30

10

20












Nowcast Cycle
The nowcast cycle for each lake forecast system is run hourly at CO-OPS to generate updated nowcasts of the lakes, including 3-D water temperatures and currents and 2-D water levels. The initial conditions for the nowcast cycle are provided by the previous hour’s nowcast cycle. The nowcast cycle is forced by gridded surface analyses of wind velocity and total heat flux. These analyses are created by the spatial and temporal interpolation of surface meteorological observations from platforms on and around the Great Lakes. On land, these networks include Automated Surface Observing System (ASOS), Coastal-Marine Automated Network (C-MAN), NOS National Water Level Observing Network (NWLON), and NOAA GLERL’s Real-Time Meteorological Observation Network.  Over water, the networks include the fixed buoys operated by the NWS/NDBC and Environment Canada. These observations are obtained by NOS’ Operational Data Acquisition and Archive System (ODAAS) from the NWS/NCEP/NCO observational ‘data tanks’ located on NCEP’s Central Computer Systems (CCS) twice per hour at approximately 25 and 48 minutes past the top of the hour.

The nowcast cycle is forced by gridded surface analyses of u- and v-wind components and total heat flux at two times: one hour prior to the time of the nowcast and the current hour of the nowcast.  All wind observations are adjusted to the standard 10 m (33ft) height and observations from land stations are adjusted to be more representative of overwater conditions.  The heat flux analyses are generated by the scheme of McCormick and Meadows (1988) which uses adjusted surface meteorological observations as well as the previous day’s lake surface average water temperature from GLERL’s Great Lakes Surface Environmental Analysis (GLSEA). The GLSEA temperature analysis is generated using sea surface temperature (SST) retrievals derived from the Advanced High Resolution Radiometer data obtained from NOAA’s polar-orbiter satellites. The nowcast cycle does not use surface pressure in forcing POMGL or any river inflows.

Map of Lake Erie showing sample surface current nowcast  from POMGL for Lake Erie. 

 Sample surface current nowcast from POMGL for Lake Erie.

Schedule of Nowcast and Forecast Cycles
The separate lake prediction systems comprising GLOFS are run operationally at NOS/CO-OPS in Silver Spring, MD. Each lake forecast system uses a separate configuration of the POMGL model. The lake forecast systems are run independently at slightly staggered times (see table below). The nowcast cycles are run hourly for each lake prediction system. The forecast guidance cycles are run 4 times a day (every 6 hours), with initial conditions at 0000, 0600, 1200, and 1800 UTC.   The later launch times for 4 of the 5 lakes are done in order to include late arriving observations from Canadian fixed buoys.

 

Lake

Hourly Nowcast Launch Time
(minutes past top of the hour)

Four/Day Forecast Guidance Launch Time
(minutes past 00,06,12, or18 UTC)

Erie

50

50

Michigan

33

33

Ontario

56

56

Huron

62

62

Superior

67

67


Skill Assessment of Predictions
MMAP personnel in collaboration with other team personnel are conducting a skill assessment of the nowcasts and forecast guidance for each lake prediction system. The skill assessment is being done in accordance with NOS’ standards as outlined by Hess et al. (2003).  In evaluating GLOFS predictions, NOS sought to take advantage of previous evaluations done by researchers at OSU and GLERL to fulfill NOS’ hindcast scenario requirements. Technical reports describing the results of the skill assessment are being prepared for all the lake prediction systems and will be available in late 2006 and early 2007.

Products
GLOFS predictions included in the nowcast and forecast guidance include water levels, water currents, and water temperature. The GLOFS nowcast and forecast guidance is available to users in several forms including graphical products and NetCDF files from CO-OPS. The graphical products include 1) time series that depict the most recent nowcast and forecast guidance at a specific station along with the latest observations and 2) map animations that depict an aerial view of a particular variable across the lakes. The water level output from the numerical model used by GLOFS is adjusted to be relative to the low water datums of the International Great Lakes Datum, 1985. In addition, in order to take into account seasonal changes in water level, a 7 day mean is calculated for each lake from observed water levels from NOS’ National Water Level Observation Network (NWLON). This mean value is added along with the IGLD to the model's water level predictions to obtain the predicted total water level which is displayed on the time series plots. Graphical displays of the GLOFS nowcasts and forecasts can be found the following pages at NOS/CO-OPS:

Lake Erie Operational Forecast System

Lake Michigan Operational Forecast System

Lake Superior Operational Forecast System

Lake Huron Operational Forecast System

Lake Ontario Operational Forecast System 

 

 

 Example of water level nowcast and forecast guidance for western Lake Erie from NOS’ LEOFS vs. observations from NOS water level gage available from the CO-OPS web page.

 Example of water level nowcast and forecast guidance for western Lake Erie from NOS’ LEOFS vs. observations
from NOS water level gauge available from the CO-OPS web page.


Files of gridded nowcasts and forecast guidance from GLOFS are available in netCDF via a  OPenDAP server at CO-OPS.    All CO-OPS official real-time products including forecast guidance from the nowcast/forecast systems for the Great Lakes and estuaries are monitored 24 x 7 by the CO-OPS’s Continuous Operational Real-Time Monitoring System (CORMS).  CORMS provides monitoring of NOS’ observing networks, data communications, and forecast guidance modeling systems.  The CORMS is currently monitored by a team of technicians in the Silver Spring, MD offices of NOAA.

Related Links:
GLERL’s Experimental GLCFS
NOAA/CoastWatch GLSEA SST Analysis
Ohio State Dept. of Civil & Environmental Engineering & Geodetic Science

Acknowledgments
The transition of the Great Lakes lake prediction system to operations at NOAA’ Ocean Service was a joint effort between NOAA’s GLERL, CO-OPS, Coast Survey Development Lab, and OSU.

Access to critical near-real-time observational data and forecast guidance inputs from the NWS has been made possible by Mr. David Michaud, Mr. Luke Lin, Ms. Chris Caruso Magee, and others at NOAA’s National Centers for Environmental Prediction (NCEP).

The skill assessment is made possible by the cooperation of Kurt Hess, Eugene Wei, and AJ Zhang in MMAP and Greg Lang at GLERL.

History and Future Work
The Great Lakes Forecasting System (GLFS) was developed by The Ohio State University (OSU) and NOAA's Great Lakes Environmental Research Laboratory (GLERL) in the late 1980s and 1990s to provide nowcasts and short‑range forecasts of the physical conditions (temperature, currents, water level, and waves) of the five Great Lakes. The development of GLFS was directed by Drs. Keith Bedford (OSU) and David Schwab (GLERL) and involved over a dozen OSU graduate students, research assistants and post doctoral researchers at GLERL and OSU, and other OSU faculty members. The development of GLFS was funded by over 36 contracts from 25 different sources. From the start, GLERL and OSU were interested in working cooperatively with NOAA (i.e. NWS) in “assessing the potential benefits [of GLFS] to NOAA’s scientific and operational programs in the coastal ocean”. In April 1991, Drs. Bedford and Schwab met with NWS and National Coastal Ocean Program (NCOP) representatives in Silver Spring, MD to discuss how they could work with NOAA line offices (NWS, NOS, etc…) to have GLFS products carefully evaluated through a demonstration program prior to NWS adopting the products as ‘guidance tools’ and which products might be distributed directly to end users.

GLFS used the Princeton Ocean Model and GLERL-Donelan wave model. The first 3-D nowcast for the Great Lakes was made for Lake Erie in 1992 at the Ohio Supercomputer Center on the OSU Columbus campus (Yen et al. 1994; Schwab and Bedford 1994). Starting in July 1995, twice per day forecasts were made for Lake Erie. GLFS was recognized with an award in 2001 by the American Meteorological Society as the first U.S. coastal forecasting system to make routine real‑time predictions of currents, temperatures, and key trace constituents.

In 1996, GLFS was ported to GLERL workstation in Ann Arbor, MI. GLERL’s workstation version of GLFS called The Great Lakes Coastal Forecast System (GLCFS) has been running in semi-operational mode at GLERL since 2001.

In 2004, the hydrodynamic model code of GLCFS for all five Great Lakes was ported to NOS/CO-OPS in Silver Spring, MD. GLCFS was reconfigured to run in the NOS Common Modeling Framework (COMFS) and to use surface meteorological observations from NOS Operational Data Acquisition and Archive System (ODAAS). The NOS version of GLCFS was renamed the Great Lakes Operational Forecast System (GLOFS). LMOFS and LEOFS began making operational nowcasts and forecasts for Lake Michigan on September 30, 2005 at CO-OPS. LSOFS, LHOFS, and LOOFS started generating operational predictions for the other three Great Lakes on March 1, 2006. LMOFS and LEOFS represent the first nowcast/forecast systems to be implemented by NOS for non-tidal water bodies.

Presently, MMAP and CO-OPS is working with personnel in NCEP Central Operations to demonstrate the capability to run LEOFS on NCEP’s Central Computer System. In addition, GLERL personnel are presently evaluating the impact of using climatological estimates of river discharge, higher model grid resolution, and the use of an ice module on POMGL nowcasts and forecasts.

References
Chu, P., J. G.W. Kelley, A-j. Zhang, G. Lang, and K. W. Bedford, 2006: Skill Assessment of NOS Lake Erie Operational Forecast System (LEOFS). NOAA Technical Report NOS CS (in preparation).

Hess, K. W., T. F. Gross, R. A. Schmalz, J. G. W. Kelley, F. Aikman, III, E. Wei, and M. S. Vincent, 2003: NOS standards for evaluating operational nowcast and forecast hydrodynamic model systems, NOAA Technical Report NOS CS 17, 47 pp.

Kelley, J.G. W., P. Chu, A-J Zhang, and G. Lang, 2006: Skill Assessment of NOS Lake Superior Operational Forecast System (LSOFS). NOAA Technical Report NOS CS (in final review).

Kelley, J.G. W., P. Chu, A-J Zhang, G. Lang, and D. J. Schwab, 2006: Skill Assessment of NOS Lake Michigan Operational Forecast System (LMOFS). NOAA Technical Report NOS CS (in review).

O’Connor, W. P. and D. J. Schwab, 1993: Sensitivity of Great Lakes Forecasting System nowcasts to meteorological fields and model parameters. Proc. ASCE Third Int. Conf. on Estuarine and Coastal Modeling, Oak Brook, IL, Amer. Soc. Civil Eng., 149-157.

McCormick, M. J. and G. A. Meadows, 1988: An intercomparison of four mixed layer models in a shallow inland sea. J. Geophys. Res., 93, 6774-6788.

Schwab, D. J. and K. W. Bedford, 1994: Initial implementation of the Great Lakes Forecasting System: A real-time system for predicting lake circulation and thermal structure. Water Pollution Res. J. of Canada, 29, 203-220.

Yen, C.-C., J. Kelley and K. Bedford, 1994. Daily procedure for GLFS nowcasts. Proceedings, National Conference on Hydraulic Engineering, Buffalo, NY, 202-206.

PARTNERS
John G.W. Kelley, Ph.D.
CSDL/MMAP

Mark Vincient, Ph.D.
Greg Mott
CO-OPS/Forecast Model Operations

David Schwab, Ph.D.
Greg Lang
GLERL

Keith Bedford, Ph.D
Philip Chu, Ph.D.
Ohio State University

User Survey  | Privacy Policy  |  Disclaimer  |  NOAA's National Ocean Service  |  NOAA  |  U.S. Department of Commerce 
Web site owner: NOAA Office of Coast Survey