Great Lakes Environmental Research Laboratory
Water Resources Research Program 1996/1997

Current Water Resources Research Program

Great Lakes water is used for drinking, power generation, commercial shipping, and recreation, and an extensive commercial and sport fishery. Both natural (evaporation) and anthropogenic (diversions, consumption) influences threaten this valuable resource. The goals of this program are to develop improved prediction, climatology, statistics for decision­making, and process studies, and to develop interfaces with policy and decision­makers.

The Task Leader for both of these tasks is Thomas Croley,734-741-2238; Tom.Croley@noaa.gov

Note: the organization of this program was revised and modified between FY96 and FY97. Therefore, there is not a one-to-one correspondence between the structure of the FY95-96 Accomplishments and Plans for this Program and the new FY96-97 Program structure.

1996/97 Accomplishments and Plans Cover Page

Research Overview page

• Coupled Hydrosphere-Atmosphere Research Model
  
  


• Next-Generation Runoff Models
  
  


• Great Lakes Hydrologic Data Base Development
  
  


• Understanding Hydraulic Processes
  
  

• Probabilistic Outlooks
  
  


• Hydrology Forecast Improvements
  
  


• On-going WaRFS Demo
  
  


• Integrated Watershed - Lake Erie Forecast System

ERL Research Task: GLERL 04 - Hydrologic Processes

Largely as a result of its climate impact research, GLERL saw a need for developing two-dimensional parameterizations of land and lake surfaces to replace its lumped-parameter models, and for integrating surface models with atmospheric models at the mesoscale level. Existing models are being integrated with atmospheric models into a first-version Coupled Hydrologic-Atmospheric Research Model (CHARM). As development of new distributed-parameter models for the atmosphere, land surface, lake thermodynamics, and lake ice progresses, they will be integrated into a second version of CHARM. Two-dimensional distributed parameterizations of land-surface hydrology are under construction as Next Generation Runoff Models. Extension of lake thermodynamics and heat storage models in two-dimensions, and development of improved models for groundwater and connecting channels hydraulics, are being performed under Understanding Hydrologic Processes, while new and improved hydrometeorological databases to support the model development and use are being constructed under Great Lakes Hydrologic Database Development.

Principal Investigator: Brent Lofgren (734-741-2383; Brent.Lofgren@noaa.gov)

Collaborating Scientists: Thomas Croley (GLERL); Peter Sousounis (Michigan State University)

Understanding how the Great Lakes affect the weather and understanding how the weather affects the Great Lakes will improve decision making concerning potential impacts of altered climate and anomalous (wet) seasonal weather patterns, such as that experienced in parts of the Midwest during the summer of 1993.

The following are anticipated to be steps toward a greater understanding of the Great Lakes region's hydrologic­atmospheric system under 'standard' and perturbed conditions:

1. develop a Coupled Hydrosphere­Atmosphere Research Model (CHARM) from existing atmospheric and hydrologic models by using two­way dynamic interactions,


2. enhance the model with second­generation surface parameterizations for lake thermal flux and runoff, and


3. refine earlier climate change estimates and estimates from other mesoscale modeling efforts by developing one­way linkages between them and the Great Lakes hydrology models.

1996 Progress and Accomplishments

The CHARM project has advanced over the past year primarily in terms of additional debugging and tuning of the model, both in its atmospheric and land components. This debugging process has involved a constant and iterative validation of model output against observed values of precipitation, air temperature, and humidity. CHARM is based on RAMS, which, along with all other mesoscale atmospheric models, has previously found its primary utility in the area of short­term forecasting, pollution dispersion modeling, and case­based simulations of clouds and thermal and orographic forcing effects. This means that when it is run for periods greater than a few days, "stealth bugs" can arise, as discussed in the following paragraphs, which also show that improvements have been achieved in the validity of the precipitation and air temperatures. With regard to validation of lake temperature and runoff, development has not reached the point of running simulations of long enough duration to make such comparisons worthwhile.

As an adjunct to the CHARM model, the sensitivity of large­scale atmospheric circulation and local energy fluxes to the presence versus absence of the Great Lakes was tested using the Princeton/NOAA Geophysical Fluid Dynamics Laboratory General Circulation Model (GFDL-GCM). The results of this investigation are presently in draft form, intended for submission to the Journal of Climate. They show that the presence of the Great Lakes, in contrast to a land surface, can effect an enormous change in the phase of the seasonal cycle of sensible and latent heat flux from the surface to the atmosphere. Also, the warming effect of the Great Lakes on the overlying atmosphere can produce an intensification and northward shift in the mean jet stream core during the autumn and winter. This study is helping to increase our understanding of the effect of the Great Lakes on atmospheric phenomena at spatial scales of at least ~1000 km and climatic time scales. It also serves as an introductory study which should assist in focusing later studies using the regional model of CHARM.

The land surface component of CHARM is based on a calibrated hydrological model with rather different available inputs (only daily precipitation and high and low temperature, instead of a full daily cycle of precipitation, air temperature, humidity, wind speed, and radiative fluxes) and required outputs (only runoff, not sensible and latent heat fluxes). As a result, the model has been extensively revised. At this point, the parameters of the land surface component are being re­calibrated to reflect the model's new configuration. Preliminary results show that this new calibration will yield greater and more persistent evaporation from the land surface, which may further help to increase the precipitation.

A lack of resources has prevented the development, validation, and incorporation into CHARM of the next­generation land surface scheme and two­dimensional lake thermodynamics scheme. The re­calibration of parameters described in the previous paragraph is intended as a stop­gap in the absence of the next­generation land surface scheme.

CHARM has not advanced far enough to make comparison between with­lake and without­lake conditions within CHARM a feasible undertaking. This will remain a priority. However, a similar task was done using a general circulation model, as described previously.

1997 Plans

• Continued testing at monthly to multi­year time scales, to ensure validity of CHARM for hydrologic and atmospheric simulation


• Procurement and further development of a version of CHARM suitable for a parallel­processing computer, making long­term climate simulations more feasible


• Incorporation of input data produced by a general circulation model (the NCAR Community Climate Model), also necessary for long­term climate scenarios


• The GFDL General Circulation Model will be run in a doubled carbon dioxide scenario with and without idealized Great Lakes to make a comparison regarding how the Great Lakes influence greenhouse warming, as a prelude to using the more computationally expensive CHARM.

Principal Investigator: Deborah Lee

Collaborating Scientists: Thomas Croley, Brent Lofgren (GLERL)

GLERL has developed conceptual­model­based techniques for simulating moisture storages and runoff from the 121 watersheds draining into the Laurentian Great Lakes. These refined runoff models will be integrated with atmospheric process models in another task and will incorporate recent advancements in measurements of hydrometeorological data. Linking surface hydrology process models with atmospheric process models will allow feedback between climate and land surfaces and result in more accurate estimates of regional and local impacts of climate change. These tools will be coupled with the GLERL GIS to develop products for resource managers and forecasters.

1996 Progress and Accomplishments

Oracle databases of daily minimum and maximum temperature, dew point temperature, precipitation, wind speed, evaporation, total sky cover, and snow fall for 1990-1995, observed at stations within the Great Lakes region (Canadian and American), were designed and linked to Arc/Info databases of station location. The station location databases also contain attribute information such as station name, agency, period-of-record begin and end dates. The databases were linked to produce Thiessen weighted maps of the variables and spatial averages over Great Lakes watersheds. Disk space limitations prevented storage of additional years of data. This limitation has been recently overcome and additional years of data will be added to the database in the upcoming year.

1997 Plans

• Transform US and Canadian Great Lakes basin soil profile information into gridded, standardized soil layers with attributes of dominant soil texture and layer thickness to use in land surface hydrologic parameterizations.
  
  


• Derive hydrologically consistent digital elevation models and river networks for Great Lakes sub­basins from existing US and Canadian digital elevation models to use in land surface hydrologic parameterizations.
  
  


• Link, via a graphical user interface, a geographic information system, a relational database management system and spatial statistics software, to create a framework for the development of the next generation runoff models.
  
  


• Evaluate and enhance existing land surface hydrologic parameterizations to incorporate into the next generation runoff models

Principal Investigator: Frank Quinn

Collaborating Scientists: Ray Assel, Deborah Lee, Thomas Croley (GLERL)

The goal of this project is to develop and provide new or improved historical hydrometeorological databases for Great Lakes climatological, water resource and water supply forecasting studies. This task will also include GLERL support to the International Coordinating Committee on Great Lakes Basin Hydraulic and Hydrologic Data and the U.S.-Canada Great Lakes St. Lawrence River Ice Information Working Group.

1996 Progress and Accomplishments

Significant progress was made during the year to analyze secular changes in annual and seasonal Great Lakes precipitation (this study will update and expand upon an earlier analysis of secular changes in Great Lakes precipitation published in 1981 by:

1. including an analysis of changes in the precipitation regime for each of the five separate Great Lakes Basins,


2. by including recently digitized historical monthly United States precipitation data prior to 1948 not included in that earlier study, and


3. by including contemporary data for an additional 11 year period (1980-1990) also not included in that earlier study):

Using the NOAA National Environmental Satellite, Data, and Information Service (NESDIS), Environmental Services Data and Information Management Program (ESDIM), monthly precipitation data that we digitized for each of the Great Lakes states, a nonparametric statistical test (Kurskal-Wallis) was used on individual stations to identify years in which there was a discontinuity in station data. If a discontinuity appeared, a second statistical test (Multiple Comparison) was then used to identify where the discontinuity occurred (i.e. what year). If station meta data showed a reason for the discontinuity, that station was preclude from further analysis; otherwise the change was attributed to a potential change in precipitation regime, if other stations also showed a similar pattern. If discontinuities could not be attributed to information given in the station metadata, e.g. changes in elevation or changes in station location, the station was used to calculate Great Lakes sub-basins area precipitation.

Area-weighted monthly precipitation was calculated for 121 sub-basins of the Great Lakes using Thiessen weighting procedure. Then annual and seasonal precipitation were calculated for the Great Lakes Basin from 1890 to 1990 from the 121 subbasin monthly precipitation data. These data are currently being used to identify changes in the precipitation regime using graphical techniques and statistical tests.

Preliminary results of this study were presented at the 39th Annual Meeting of the International Association of Great Lakes Research, Toronto, Ontario, Canada. May 26-31, 1996.

A study undertaken for the International Coordinating Committee on Great Lakes Basic Hydraulic and Hydrologic Data of the ratio of annual area-weighted total precipitation for the U.S. and Canadian land areas of the Lake Superior hydrologic basin from the late 1800s to 1990 shows there is a discontinuity or change in this ratio near 1947. A Students-t test indicates there is a significant difference between the mean ratio for the years before and the mean ratio for the years after 1947. We suspect that the low density of precipitation stations in the earlier years of the record prior to 1947 may be a contributing factor to the observed discontinuity.

We are developing a method to provide better estimates of the earlier year area-weighted precipitation value by developing a ratio of station precipitation with area-weighted precipitation during contemporary years, contouring the ratio field and using these ratios with available station data in the earlier years to estimate areally weighted precipitation in the earlier years of the record and then recalculate the US/Canadian ratio to see if we still get a significant difference in the mean ratio around 1947. Results of this study should provide us with better estimates of area average precipitation for Lake Superior in the earlier years of the record and the method used may be applied to Lake Huron and Georgian Bay to get better estimates of annual precipitation in the earlier years, prior to 1947 as well.

A major accomplishment was the completion of the Data Temperature Rescue Task. We completed digitizing monthly air temperatures from beginning of record to 1930 (for the states bordering the Great Lakes) and combined these data with National Climate Data Center (NCDC) digital temperature data from 1931 to 1990. Data reduction, verification, quality control, and analysis methods along with station metadata (station name, latitude, longitude, elevation, and period of record) were documented in a NOAA Technical Memorandum. A copy of these data was made available to the National Climate Data Center for archiving in April 1996. These data were also made available to the Midwestern Climate Center for verification of daily temperature data being digitized by state climatologists for the MCC. In future studies GLERL will use these data to develop improved understanding of temperature climatology and temperature regimes from the late 1800s to the 1930s. A final project report was also written and sent to the NOAA/NESDIS ESDIM Program Coordinator.

FY97 Plans

• An assessment is being undertaken to understand and document the changes to the Niagara River flow regime and the changes in methodology for computing the monthly flows between 1943 and 1974 to prevent unnecessary misunderstandings and to include the proper Niagara River hydraulics in Great Lakes water resource studies. This project is now planned for completion in FY98.
  
  


• Produce, and make widely accessible, a medium­resolution Great Lakes shoreline in the Topological Vector Profile of the National Spatial Data Transfer Standard to use in coastal management and shoreline process studies
  
  


• Code a modular, object­oriented, middle Great Lakes hydrologic routing model, coordinated between the US and Canada, to use in binational Great Lakes studies and operational regulation and forecasting
  
  


• Improve estimates of Lake Superior overland precipitation to use in Great Lakes water resources and climatology studies.
  
  


• Analyze secular changes in Great Lakes Basin annual precipitation to identify climatic changes and variability in precipitation over the past century.
  
  


• Procure and process meteorological data from U.S. and Canadian agencies to update through 1996 for use in GLERL's runoff and evaporation modeling.

Principal Investigator: Frank Quinn

Collaborating Scientist(s): Cynthia Sellinger (GLERL)

The hydraulic processes in groundwater and riverine flow are important for assessing the hydrologic water balance components of the Great Lakes basin and for developing improved models for water level forecasting and simulation. The primary objective of this research is to develop, test, and document groundwater and connecting channel hydraulics models for use in Great Lakes water resource studies.

FY96 Progress and Accomplishments

Accurate forecasts of the Great Lakes water levels and chemical composition requires data on the entire water budget including groundwater interaction. Groundwater that flows directly into lakes contributes not only to the lake's water volume but also to its chemical balance. A project has been underway for two years to study this interaction. Funded by the State Department, this project's overall challenge is to study groundwater flux to Lake Michigan and Lakes Sniardrwy and Zarnowiec in POLAND. Accomplishments during the past year include the completion of databases that contains hydrological parameters and the development of computer programs and methods to analyze these parameters.

An Acoustic Doppler Current Profiler was successfully deployed in the Detroit River to provide data for assessing the impact of increased aquatic growth due to the zebra mussel on Detroit River flows.

FY97 Plans

• Vertical velocity data will continue to be collected from the Acoustic Doppler Current Profiler at the Fort Wayne Section in the Detroit River to assess the impact of increased aquatic growth due to the zebra mussel on Detroit River flows.
  
  


• The vertical velocity data collected during 1986 - 1988 prior to the zebra mussel infestation will be analyzed to determine the pre-zebra mussel river roughness and flow retardation for comparison with the current data.
  
  


• The Modflow groundwater model, with and without the streamflow component, will be used to assess, from a modeling perspective, groundwater pathways in the Lake Michigan test basin.

ERL Research Task: GLERL 05 - Water Resources Forecasting

High and low Great Lakes levels cause extensive flooding, erosion, and damage to shorelines, shipping, and hydropower. The International Joint Commission, at the request of the US and Canadian governments, recommended improving forecast methodologies, hydrological models, data collection, and communication of hydrological forecast information. While forecasts of meteorology, riverine flooding, and water level fluctuations are available for several hours to several days, the Great Lakes community requires water resource forecasts over large areas and time periods. Products must include nowcasts and 1-day to 3-month probabilistic outlooks of lake supplies, lake levels, and connecting channel flows. These require careful tracking of moisture storage variables and heat storage variables. The products must be relevant to users and delivered in a clear and understandable manner that aids in planning and decision making. They must make maximum use of all available information and be based on efficient and true hydrological process models.

We combined models of large-basin rainfall-runoff and large-lake thermodynamics and heat storage into the Great Lakes Water Resources Forecasting System (WaRFS), most recently in cooperation with the National Weather Service's Office of Hydrology, Hydrologic Research Laboratory. Great Lakes WaRFS products are being used now by the US Army Corps of Engineers in three offices, the National Weather Service, Ontario Hydro, the New York Power Authority, and NOAA's Midwestern Climate Center.

Return to Project Index

Principal Investigator: Thomas Croley (734-741-2238; Tom.Croley@noaa.gov)

Collaborating Scientists: Jim Argel (Midwestern Climate Center); Jon Hoopingarner (NOAA/Climate Prediction Center)

WaRFS provides probabilistic outlooks of many hydrological variables at weekly, seasonal, and inter-annual time scales throughout the Great Lakes by using the new long-lead extended-climate outlooks provided by NOAA's Climate Prediction Center, in order of user priority . The system includes a recently developed innovative interface (WindowsÔ application) to facilitate WaRFS outlooks from NOAA's extended climate outlooks. (Click Here  to go to GLERL's Probabilistic Outlook Page).

1996 Progress and Accomplishments

Great progress was made in considering probabilistic meteorology outlooks in operational hydrology. There are now several kinds of probabilistic meteorology outlooks available to the water resource engineer or hydrologist. The National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center provides a monthly climate outlook at mid-month, consisting of a 1-month outlook for the next (full) month and thirteen 3-month outlooks, going into the future in overlapping fashion in 1-month steps . Each outlook estimates probabilities of average air temperature and total precipitation falling within the lower, middle, and upper thirds of observations from 1961-1990. The Climate Prediction Center also produces a 6-10 day outlook, covering the 5-day period beginning six days hence. It predicts which of five intervals of 5-day average air temperature or total precipitation are expected: less than the 10% quantile, between the 10% and 30% quantiles, between the 30% and 70% quantiles, between the 70% and 90% quantiles, or greater than the 90% quantile. The Climate and Water Information Branch of Environment Canada (EC) produces both a 1-month outlook at beginning- and mid-month and a 3-month outlook each quarter of average air temperature. Each outlook predicts which of three intervals (lower, middle, or upper thirds of observations from 1961-1990) of 1-month and 3-month average air temperature are expected. All of these outlooks differ in several important respects. They are defined over different time periods (5 d, 1 mo., 3 mo.) at different lag times (0 mo., 6 d, ½ mo., 1½, 2½, B, 12½ mo. from when they are issued; real lags depend on when they are actually used), and they specify either a probability of falling within an interval (event probabilities) or only the most-probable interval (most-probable event).

Users of probabilistic meteorology outlooks can interpret them through an operational hydrology approach that considers historical meteorology as possibilities for the future. The approach segments the historical record and uses each segment with models to simulate a hydrological possibility for the future. Each segment of the historical record then has associated time series of meteorological and hydrological variables, representing a possible "scenario" for the future. The approach then considers the resulting set of possible future scenarios as a statistical sample and infers probabilities and other parameters associated with both meteorology and hydrology through statistical estimation from this sample. However, the relative frequencies of selected events are fixed at historical values that are incompatible (generally) with those specified in probabilistic meteorology outlooks. Only by restructuring the set of possible future scenarios can we obtain relative frequencies of selected events that match probabilistic meteorology outlooks. There are many methods for restructuring the set of possible future scenarios; last year's accomplishment restructured to match forecast event probabilities as given in NOAA's monthly climate outlooks (1-month and thirteen 3-month outlooks of probabilities of average air temperature and total precipitation falling within three intervals). However, this method does not address matching most-probable event forecasts such as the NOAA 6-10 day outlook (most-probable of five intervals for 5-day average air temperature and total precipitation) or the EC 1-month and 3-month outlooks (most-probable of three intervals for 1-month and 3-month average air temperature). This approach was extended this year to mix all of these probabilistic meteorology outlooks (both event probabilities and most-probable events) to make hydrological outlooks.

Boundary condition equations for the weights are constructed, corresponding to forecast event probabilities, and boundary condition inequalities are constructed corresponding to forecast most-probable events. The inequalities are converted to equivalent equations through introduction of additional variables. The resulting set of all boundary condition equations is solved for physically-relevant values. Their solution becomes an optimization problem for the general case, similar to earlier consideration of only forecast event probabilities.

We switched from OBJECT PASCAL to DELPHI, which enabled us to upgrade our interface from WindowsÔ 3.1 to Windows95Ô. This included expanding our batch processing facility in running multiple hydrological outlooks. We expanded our capabilities to accommodate more PC displays and we made many minor updates reflecting usability issues to address requests of users. The major change, which is still under development, is the incorporation of new probabilistic meteorology outlooks into the interface and the expansion of the priority-structure determination routines to handle 100 constraints on probabilistic meteorology forecasts. We built improved algorithms for estimating weights to apply in making the hydrological outlooks and they default to the old methods as a special case when only "old" constraints are used (only NOAA's 1 and 3 month climatic outlooks).

Regarding a "back-end" interface, it is close to being finished, but was put aside while other issues are addressed. A stripped-down version was assembled for making monthly (only) outlooks to produce only graphic products (to support installations in the field). Several versions were built as we evaluated different software components commercially available. While we have a version now in use in the field, it cannot run in the background, which impedes its frequent use. It also has graphics limitations, many of which we've worked around; it appears of limited use and is only being used now temporarily. The second version is much more flexible but it is too slow currently because of large number of graphic objects required to make up a single graphical outlook product (about 700 objects). We are working with the product's technical support to try to resolve this problem.

1997 Plans

• Program the theory to mix probabilistic meteorology outlooks as boundary conditions for hydrologic outlooks to enable the use of mixed climate outlooks in making probabilistic hydrology outlooks.
  
  


• Update the front- and back-end graphical user interfaces to allow intuitive mixing of probabilistic meteorology outlooks, expanded graphical interpretations, and enhanced speed.

Principal Investigator: Thomas Croley (734-741-2238; Tom.Croley@noaa.gov)

Collaborating Scientist: Deborah Lee, Brent Lofgren (GLERL)

The present large-basin runoff models and large-lake thermodynamics and heat storage models provide an initial WaRFS development. It continues with parallel efforts in process model development, data stream incorporation, and integrated data management. Our existing lumped-parameter models will be integrated with atmospheric models to build a distributed-parameter Coupled Hydrologic-Atmospheric Research Model (CHARM I). As our new distributed-parameter models for the atmosphere, lake thermodynamics, and land surface progress, they will be integrated into a second version (CHARM II). All will be available for implementation into WARFS. Thus, we will continue to serve a variety of present users, whose needs are satisfied with basin-wide outlooks based on GLERL's lumped-parameter models, while also servicing some of these and others with WARFS outlooks based on GLERL's developing distributed-parameter process models.

1996 Progress and Accomplishments

No significant activities during FY96 due to lack of funds.

1997 Plans

• Incorporate CHARM I products (Coupled Hydrologic-Atmospheric Research Model, level I, using existing semi-distributed models of runoff, basin moisture, lake thermodynamics, and evaporation) to improve our hydrologic modeling in making hydrological outlooks.
  
  


• Incorporate both GIS & RDBMS uses for including data in near real-time now and to aid in hydrologic modeling later.

Principal Investigator:Thomas Croley (734-741-2238; Tom.Croley@noaa.gov)

There will be a lake-wide and regional demonstration of the WaRFS package that will be ongoing as WaRFS is developed. This will clarify product development and use, allow for validation and testing, and acquaint prospective users with the product. We will establish a Center for Water Resources Outlooks that will operate the WaRFS package and disseminate results. GLERL will aid in technology transfer, demonstrate product use, and help users to integrate products into management, operations, and policy making activities. Finally, there will be water supply simulation packages available to allow user agencies to make their own assessments (including operational agencies, research agencies, and academia); this can enable interactive investigation of "what if" scenarios by prospective users.

1996 Progress and Accomplishments

This year saw the installation and commencement of delivery of graphical hydrological outlook products for the Great Lakes on the world wide web! The Midwestern Climate Center began issuing probabilistic outlooks for 23 different hydrological variables over 7 water bodies and 121 watersheds within the Great Lakes basin. They are using our water resource forecasting system with their data assemblage and with probabilistic meteorology forecasts from NOAA's NCEP and Environment Canada. They update the outlooks weekly and are making only monthly products available for people to look at and download. We installed a web page link to their web site, to NCEP and EC, and to a short explanation of the method of incorporating probabilistic meteorology outlooks into hydrologic outlooks.

We built software to download all meteorological data daily from the Midwestern Climate Center over the prior 30 days, prepared daily. The data is then reduced and used in a hydrological outlook daily for all 23 hydrological variables, 7 water bodies and 121 watersheds of the Great Lakes basin. The data transfers involve from 75 to 250 stations and the entire procedure has been completely automated. While the forecasts are then available each day, they are not archived.

1997 Plans

• Install a viewport (hallway monitor) in our continuing demonstration of our (graphical) outlook products both to acquaint others with the forecast package and to monitor and evaluate product upgrades.
  
  


• Implement near-real-time data acquisition and use in the demo and in the field to enable timely estimates of current conditions and hydrologic outlooks.
  
  


• Implement spatial representation products for outlooks of many relevant hydrologic variables to enable easy understanding and interpretation of forecast products by the user community.

Principal Investigator: Thomas Croley (734-741-2238; Tom.Croley@noaa.gov)

Collaborating Scientists: Keith Bedford (Ohio State University)

The confluence between the Great Lakes and its watersheds is an extremely complex physical, biological, and chemical mixing zone. The different harbor/tributary needs cannot be addressed by simple data collection or the existing Great Lakes Forecast System (a short-term atmospheric and hydrodynamic forecast model for Lake Erie). Only an integrated data/model system can provide the required accurate information in a timely fashion. GLERL and The Ohio State University plan a fully integrated calibrated system which can, on an hourly basis, make accurate forecasts, nowcasts, and hindcasts of the watershed runoff volume, heat, and sediment loads to Lake Erie and their subsequent redistribution and fate in the lake. Implicit in this goal is the coupling of the NOAA GLERL Water Resources Forecasting System (WaRFS) with the Great Lakes Forecasting System (GLFS). These forecasts will be achieved with models sufficiently robust to incorporate land-use patterns within the basin and respond to high resolution precipitation forecasts. The coupled GLFS/WaRFS will be based upon existing operationally collected data and will be verified with existing data sets.

1997 Plans

• Apply existing lumped-parameter WaRFS to the Maumee River Watershed as a test basin to prepare for integration of WaRFS with GLFS.
  
  


• Integrate the existing lumped-parameter WaRFS with GLFS by creating a high-resolution tributary component for the test basin/tributary to prepare for later distributed-modeling refinements in hydrology modeling within GLFS.

Return to Project Index

1996/97 Accomplishments and Plans Cover Page
Research Overview page

Last updated: August 9, 2002 mbl

Home | Research | Data | Products & Services | Outreach | About GLERL | News & Events
Privacy Policy | Disclaimer | Accessibility Statement | GLERL only | Contact Us

GLERL 4840 S. State Rd., Ann Arbor, MI 48108-9719     (734) 741-2235
Lake Michigan Field Station, 1431 Beach St., Muskegon, MI 49441-1098     (231) 759-1560