Near-Shore Transport: Modeling, Observations, and Beach Closure Forecasting
Under the Clean water Act, the recent BEACH Act, and the TMDL rule, there has been a demonstrated need for improved monitoring for sources of human waste and for specific pathogens of concern. Human health can be explicitly tied to water quality, and despite major advances in the last several decades, fresh water systems remain at risk.
Our long term objectives are:
- Develop a modeling system based upon a fully three-dimensional hydrodynamic model (GLFS) for forecasting E. coli and Enterococci concentrations along Great Lakes coasts impacted by a specific plume (ultimately pathogens).
- Design the modeling system and nested grid for ease of application to other sites in the Great Lakes.
- Test model adequacy with extensive comparisons to data obtained from moored current meters, dye studies, and in situ water quality sampling.
- Determine the extent of ecological consequences from model simulations under various weather and loading conditions and if a well-constrained set of ecological outcomes exists.
- If fully successful, develop a training program for potential users.
The Great Lakes Forecasting System (GLFS, Bedford and Schwab, 1994; Schwab and Bedford, 1994) has been developed to provide short-range operational (regularly scheduled) predictions of such conditions for the open waters of the Great Lakes. Predictions include every-six-hour nowcasts and twice-a-day short-range (48 hr) forecasts (See www.glerl.noaa.gov/res/glcfs/gh/ and http://www.glerl.noaa.gov/res/glcfs/ghf). Variables predicted include the three dimensional velocity field, the three-dimensional temperature field, the water level distribution and the wind wave height, length, period, and direction, and resuspension, transport, and deposition of bottom sediments based on wave and current conditions (Lou et al., 2000). Predictions are made on a five kilometer horizontal grid with twenty vertical slices comprising the vertical grid. The Princeton Ocean Model (Blumberg and Mellor, 1987) serves as the base model for the forecast system. Weather data are acquired from the National Weather Service through the NOAAPORT satellite dissemination network and objectively analyzed in near real-time to drive the nowcasts. The system has undergone extensive testing. Day-to-day evaluations are performed with the NOS water level, buoy temperature, and wave data. More extensive evaluations have occurred in hindcast comparisons with field experiment data, including the full three-dimensional current and temperature comparisons in Kuan et al. (1995) and Kuan and Bedford (1995), the surface temperature comparisons in Schwab et al. (1992), and the forecast comparisons in Kelley et al., (1996) and Kelley (1995). As a result of these testing activities, we concluded that the whole lake circulation features could be forecast with reasonable accuracy 12-24 hours in advance.
When contrasting the information needs of water quality managers with the forecasting experience to date, three issues remain. First, the information requirements all occur with regard to activities in, near, and around the near-shore/inshore zone. It is well known that the greatest demand for lake/coastal resources is in the near-shore zone and accurate information is required in this zone. Second, the information needs of the managers are for water quality data; data not yet predicted or available in forecast form. Third, the water quality forecasts require knowledge of both point and non-point sources. Our research program will focus on point source loadings of E. coli (EC) into coastal environments from particular rivers and its impact on beach closures.
Traditional beach monitoring for E. coli typically requires
a 24 hour incubation period, resulting in people unintentionally swimming in contaminated
water, or conversely loss of local economic revenues and beach time. Results from stakeholder
needs assessment workshops have indicated the need for a better beach monitoring
program and more specifically the need for a beach forecasting system. In order
to address these needs, the GLERL-based Center of Excellence for Great Lakes and
Human Health is working to understand the influence of wind, waves, surface temperature,
and water currents on pathogen transport by conducting tests in the Grand River, a
major tributary of Lake Michigan, to track contaminant flow downstream to Lake Michigan
and its adjacent beaches.
The experiments began during the summer of 2006 when a conventional
dye, Rhodamine-WT, as well as a bacteriophage, PRD-1, were released into the Grand River
in Grand Rapids, Michigan and the tracer gas, Sulfur Hexafluoride, was released and tracked
in the Grand Haven area of Michigan. The data from these results helped to create and test
a model that will be able to predict the transport and distribution of contaminants from
the Grand River into Lake Michigan. Researcher Michael McCormick explains that a simple
model comparison against data from the sulfur hexafluoride sampling “highlights the complex
interaction between a buoyant river plume and coastal circulation.” It also highlighted
the need for higher resolution mapping.
The field experiments continued in the summer of 2007 with a similar tracking of the
non-toxic red dye, Rhodamine-WT, but this year the study is focused closer to Lake
Michigan with the intent of modeling the plume at the end of the Grand River. Acoustic
Doppler current profilers were used to measure currents and a towed fluorometer was
used to track the dye concentrations at various depths to better understand the movement
of the plume. Reflecting on the first summer of experiments, McCormick noted that there
was a need for a higher resolution in data gathering. “In 2006 researchers used an
adaptive grid to determine sampling locations, now they are testing every second”,
McCormick explains. In addition, researchers conducted bacteria transport research
with E.coli and total coliforms in the water and used aerial photography to track the
movement of dye in the River. Researcher David Schwab describes, “We released the
Rhodamine dye in the river so that we could accurately measure the dilution as the
river mixed with the lake.”
This data were employed to test a model that uses
real-time data about wind speed and direction, waves, water currents, and surface
temperature to predict plume movement. “We would eventually like to apply the methods
we develop for forecasting pathogen transport near the Grand River to other beaches
in the Great Lakes region,” added Schwab. The long term goal of this project is to
develop tools that will be useful in forecasting potentially harmful conditions in
the coastal waters and beaches in order to assist decision-makers with informing the
public in a timely manner to reduce water-related human health risks. There is a
potential extension of this study to combine the models used to determine contaminant
threats at beaches with watershed models so that we could model and predict the effects
of land use on beach closures.
Projects
| Project Title | Researcher(s)/Affiliation |
| Near Shore Observations | Michael McCormick (NOAA/GLERL) |
| Near Shore Modeling | David Schwab (NOAA/GLERL), Dimitry Beletsky (University of Michigan) |
| Near Shore Beach Closure Forecasting | |
| Project Safe | Richard Whitman (USGS) |
| Virtual Beach | Walter Frick (USEPA), Mantha Phanikumar (MSU) |
| Development of Beaches Database | Joan Rose (MSU) |
