Virtual oceans tackle real-world issues: DFO's Centre for Ocean Model Development for Applications

Every day, billions of tonnes of seawater flow in and out of the Bay of Fundy between Nova Scotia and New Brunswick, creating the highest tides in the world. To explore the potential for harnessing green energy from this incredible force of nature, in 2008 the Province of Nova Scotia approved a $12-million tidal power pilot project in the Minas Passage of the Upper Bay of Fundy. Three test turbines will be installed in the area to study the technical, environmental and economic practicalities of building a full-scale tidal farm of underwater sea turbines.

The potential impacts of such a project on the Bay of Fundy are the focus of a study led by DFO research scientist Dr. David Greenberg of the Bedford Institute of Oceanography (BIO). Greenberg is developing an ocean model to explore how a tidal farm could affect currents, high water levels and the formation of silt or mud in the area. Past research indicates that barriers in the Upper Bay of Fundy increase tidal high water levels in some places and decrease it in others.

National leadership and coordination in ocean modelling

The research is being carried out under the umbrella of COMDA - DFO's virtual Centre for Ocean Model Development for Applications. Hosted by BIO, the centre provides national leadership, coordination and advice in ocean model development and applications of priority to Fisheries and Oceans Canada. In the short term, the centre is focussing on climate and Arctic research, regional ecosystems-oriented research, and integrated and site-specific studies.

Ocean models are computer programs that represent the ocean in three dimensions and its evolution over time. They are based on ocean data collected by Earth observation satellites and direct measurement of conditions such as temperature, salinity and currents. Since the atmosphere also influences the state of the ocean, models also include atmospheric data including wind stress, precipitation, evaporation and cloud coverage. Modellers call this "atmospheric forcing." Based on this information, the model can be run to predict various aspects of ocean conditions and behaviour into the future.

"One of COMDA's main goals is to develop a common set of modelling tools that can be shared and built upon over time," says COMDA director Dr. Brenda Topliss. For example, the centre's near-shore working group is investigating the finite volume coastal ocean model (FVCOM) for the tidal power project as well as aquaculture, oil dispersion, circulation and risk assessment.

COMDA involves participants from DFO research labs across Canada and collaborative projects with other federal departments, agencies and universities, particularly Dalhousie University and the French organisation, Mercator Ocean. "Having a modelling community is important because we're working with other scientific communities on coupled models and the code gets very big and sophisticated," says Dr. Topliss.

Small-scale modelling for near-shore applications

Dr. Greenberg's model, a modified version of FVCOM, includes water depth, numerical representations of the coastline, and mathematical equations of motion that govern the flow of water including tides, rotation of the earth, bottom friction, viscosity and gravitational forces. The model's code was also modified to simulate water meeting a turbine instead of simply flowing back and forth with the tides.

"Turbines take energy out of the water and slow it down which can lead to the formation of silt (mud) in new places. This could have an impact on birds and aquatic life such as clams that live in or feed on the bottom," he says.

Current and tide data, both recent and historical, are included to "prove" the model. "We set the model's clock back then run it to see if it can reproduce what real-world observations have shown up from the past," says Dr. Greenberg. "Once the model is complete we'll be producing results over the next year or so and exploring long-scale studies related to a scaled-up turbine farm."

Large-scale modelling: building on NEMO

For global scale, basin scale (i.e. the entire north Atlantic or north Pacific) and most regional scale modelling projects, the centre has adopted the NEMO ocean model, developed in France by Mercator Ocean. It is designed to explore a wide variety of issues, with time scales ranging from hours to thousand of years, and spatial scales ranging from a meter to the entire world.

In collaboration with Environment Canada and National Defence, COMDA researchers are refining NEMO for Canadian applications through a program called CONCEPTS - Canadian Operational Network of Coupled Environmental PredicTion Systems. This program involves the development of basin- (Atlantic and Pacific) and global-scale versions of the coupled atmosphere-ocean-ice model.

"Once we tweak NEMO for Canada and prove it's consistent and going in the right direction, then Environment Canada will couple it with its atmospheric model to improve weather forecasts," says Dr. Topliss. Coupling enables the atmospheric model to influence the ocean model and the ocean to impact atmospheric conditions, as would happen in nature. This collaboration with the atmospheric community will also help DFO improve its ocean models for a wide range of applications.

"Large-scale models provide insight into oceanic climate variability - natural variations in the temperature, current and other oceanic conditions from day to day, year to year and decade to decade. This also improves our understanding of currents and temperature at a regional level, such as on the continental shelf." says Dr. Topliss.

C-NOOFS

In collaboration with both COMDA and CONCEPTS, Dr. Fraser Davidson of DFO's Northwest Atlantic Fisheries Centre in St. John's, Nfld., is leading a three-year pilot project to forecast the state of northwest Atlantic in near real time, focussing on waters off the east coast of Canada. The Canada-Newfoundland Operational Ocean Forecasting System (C-NOOFS) produces forecasts and analysis of oceanic conditions including currents, temperatures, salinity and waves. These forecasts have already been used to improve the Canadian Coast Guard's search and rescue planning software.

Regional-scale modelling to predict ocean currents and dispersion

At the regional scale, research scientist Dr. David Brickman of BIO is developing NEMO as an ocean circulation model of the Gulf of St. Lawrence, Scotian Shelf, Gulf of Maine and Bay of Fundy to determine the best locations for ships travelling intracoastal routes to exchange ballast water. The goal is to reduce the risk of introducing potentially invasive species from elsewhere into Canadian waters.

Ballast tanks of ships are filled with water to provide stability when not carrying a load. When the ship takes on cargo, the ballast water is released along with any ocean life it might contain.

The International Maritime Organization recommends that ships exchange ballast water in mid-ocean to lower the chance of species from one port finding their way to another. Ships travelling intracoastal routes don't reach mid-ocean, however. To address this issue, Dr. Brickman is using his model to estimate ocean conditions for the current day and the next five days and make daily prediction of the lowest risk place for ballast water exchange along 11 of the most common east coast shipping routes. The predictions are based on factors such as ocean currents, wind and properties of dispersion.

The model is currently at the demonstration phase; however once it's developed further Transport Canada is considering including the model's outputs with other advice it provides to ship captains.

"Models are based on our current understanding of nature and the mathematical laws which govern the behaviour of the ocean, atmosphere and ice," says Dr. Topliss. "As we increase our understanding of these natural processes, our ability to model past, current and future environments will improve."

To learn more about COMDA, click here.