October 2002

(Left) Final check on instruments before the main buoy is deployed.The town of Lunenburg is approximately one kilometre in the background. (Right) Satlantic employees make final adjustments to the solar panels during deployment.

One inevitable—and unfortunate—result of population growth is the increase in human inputs into the atmosphere, land and water.

Sewage outfalls and agricultural runoff are causing harmful algal blooms in increasing numbers. According to Safeguarding the Health of the Oceans (Worldwatch), the number of harmful algal blooms in the West Central Atlantic rose from 10 in 1970 to 330 in 1997. Explosive blooms, also known as “red tides”, have forced the closure of fisheries, caused illness, and cost hundreds of millions of dollars in lost fisheries revenue.

At the same time, scientists need to understand how the oceans respond to climate change, and a host of other complex environmental issues involving multidisciplinary oceanography and atmospheric science. In response, research scientists are driven to develop more accurate predictive models to assist in the stewardship of coastal areas.

In order to understand and predict coastal ocean behavior, a multidisciplinary approach is needed—one that measures and tracks a full range of physical, chemical and biological ocean processes, in addition to atmospheric dynamics. Such an approach pushes the limits of existing science and requires new technical solutions to measure a wide range of variables, and to handle the tremendous volumes of data collected. The Lunenburg Bay prototype observatory employs a unique combination of technologies that enable investigators to more effectively collect, manage and analyze very large data sets from the deployed moorings. As a result, information will be more readily available for multidisciplinary scientific models, and for commercial and recreational users of the bay.

Dr. John Cullen, Professor of Oceanography at Dalhousie University in Halifax, Nova Scotia, Canada, is one of a number of research scientists around the world who are working to develop new technologies for observing and predicting changes in the ocean such as sea-level rise, coastal eutrophication, and other critical ocean events that occur in the nearshore. He is Acting Director of the Dalhousie University-based Centre for Marine Environmental Prediction which is heading up the project, and Project Leader of the Marine Environmental Prediction System (MEPS). MEPS encompasses: 1) modeling seasonal conditions in the North Atlantic Basin, 2) furthering the development of a storm surge prediction system, and 3) creating a prototype ocean observatory whose data is input into a forecast model for a coastal embayment. This article will focus on how the ocean observatory project was designed and deployed to ultimately reach a holistic understanding of a typical tidal inlet on the south shore of Nova Scotia.

Diagram of the main buoy. The installation includes a second stationary buoy and a mobile buoy, similarly configured.

Ocean Observatory Project Goals
The immediate objectives are to deploy an array of instruments that can acquire oceanic and atmospheric data in near-real time, and use the data to improve existing predictive oceanographic and atmospheric models of Lunenburg Bay. (The Town of Lunenburg is a UNESCO World Heritage Site.) The project was designed to provide data to two broad user groups: research scientists, and end-users such as aquaculture farms, the fishing community, recreational boaters, tour boats, and recreational divers. Information about the project, as well as time series data generated by the observatory, will be posted on the Web.

The overarching purpose of the project, however, is to demonstrate that: (1) such an ocean observatory—and the scientists associated with it—can collect and process all of the information required to improve models that can forecast the critical events in a coastal area; and (2) similar observatories can be set up anywhere in the world in order to model oceanographic processes specific to the area, and to provide accurate forecasting capabilities to inform local decision making. As Dr. Cullen points out, “You could take the same observation and modeling approach, and the communication systems, and put them somewhere else—in Spain, for example—because the laws of physics don’t change.” Whereas Lunenburg Bay has little freshwater input, the observatory could be easily modified to operate in an inlet dominated by freshwater, observes Dr. Cullen. “In that case, the modelers would decide what modifications need to be made. After all, they change the tires at the race track when it rains, but it’s still the same car.”

Funding for the three-year infrastructure research project is being provided by the Canada Foundation for Innovation under the Innovation Fund, and the Canada / Nova Scotia COOPERATION Agreement on Economic Diversification. The Agreement is managed by the Atlantic Canada Opportunities Agency and Nova Scotia Economic Development. To assemble the team and develop the necessary infrastructure, Dalhousie University has partnered with Environment Canada (Meteorological Service of Canada—Atlantic), the Department of Fisheries and Oceans (at the Bedford Institute of Oceanography), the Institute for Catastrophic Loss Reduction, Satlantic Inc., SonTek, IBM, and Sun / OSS, all of whom made contributions.

The project will be using a network of instruments to measure the meteorological, physical, optical, and acoustical properties of the Bay. The near-real time measurements will be used to validate and improve existing forecasting models, and to provide important information to users of the bay.

Data Acquisition System Needed
To generate the data that will be required to feed the numerical models, a robust and highly flexible data acquisition and control system was required. For this, the Centre for Marine Environmental Prediction looked to Satlantic Inc., (a Halifax-based developer of advanced marine sensors and data acquisition systems. Dr. Cullen and Satlantic have advised one another on the development and use of new scientific instruments for seven years. “I’m able to use cutting-edge technology in my research without having to wait until it’s on a shelf,” observes Dr. Cullen, “and they can interact freely with a primary researcher who can suggest new instruments and test what they’re doing. It’s a win-win situation.”

Satlantic had already recognized the need for a flexible and highly efficient data acquisition and control system, and collaborated with Dr. Cullen to ensure that Satlantic’s next data acquisition system, DACNet, would meet the observatory project’s requirements.

“What we needed was an integrated data stream,” explains Dr. Cullen, “that would enable us to get information quickly—an efficient system for translating data from the sensors to us. Because we’re not a cabled system, there were power and wireless data communication issues to overcome, so we have a broadband wireless network solution and solar power.

“We wanted the capability to manage instruments remotely—to turn them on and off, and to change the sampling schedules. So we needed the ability to interact with the instruments whenever we feel like it—to do trouble shooting, or if an event occurs, to increase the sampling. And we needed information from a lot of different instruments, shot through the Internet to us, in near-real time.”

DACNet is a flexible, off-the-shelf system that can be customized to meet the requirements of any observatory. Its power management system enables operation without the power constraints that have traditionally limited the deployment of multiple, varied instruments. The system manages high-bandwidth communications, and provides dynamic instrument control from remote user stations. DACNet’s plug-and-play feature makes it easy to add instruments to an existing deployment.

“We are extremely pleased to see DACNet prove itself on this scale,” says Cyril Dempsey, Satlantic’s lead manager on the project. “This system has successfully overcome the problems that have traditionally been associated with limited power, constrained communications bandwidth, and complex data formats.”

Dr. Cullen describes DACNet’s role: “The sensors / instruments are the muscle of the observatory, DACNet is the nervous system, and the researchers and models represent the brain.”

Buoy and Instrument Deployment
The observatory project took one and a half years to plan before deploying three buoys, DACNet, and the first set of instruments from boats in Lunenburg Bay between mid-June and mid-July, 2002.

With the technical assistance of Satlantic, a team from Dalhousie University installed two stationary buoys, one mobile buoy, a shore-based station, and three wireless networks connecting the buoys to the station. The buoys are all within one kilometer of the shore. The DACNet computer and software was installed on each of the three buoys.

All three of the buoys are outfitted with arrays of Satlantic optical sensors which measure sunlight penetration in, and reflectivity from, the water in order to detect changes in the abundance of phytoplankton and the concentration of other colored constituents of the water. The instruments include: (1) an irradiance sensor that sits above the water and measures downwelling irradiance (light coming directly from the sun and sky); (2) a radiance sensor below the surface that measures upwelling radiance (sunlight reflected back up through the water); and (3) a “K-chain” of four irradiance sensors at varying depths which measure the penetration of downwelling irradiance. (1) and (2) above are hyperspectral sensors which measure 120 wave bands, and (3) is a chain of multispectral sensors that measure four wave bands: one ultra-violet, two blue, and one green. At the bottom of each chain, two metres from the bottom, is a Seabird conductivity, temperature and depth (CTD) sensor.

One acoustic doppler profiler (ADP) was mounted just below the surface of the water on one of the moorings. This instrument measures vertical profiles of the ocean current. After the initial deployment, two more acoustic instrument suites were added on August 7. Cyril Dempsey assisted another team of Dalhousie University technicians in installing an acoustic doppler velocimeter (ADV) and an acoustic doppler profiler (ADP) on a bottom pod approximately 50 meters from the stationary buoys. The ADV measures near-bed turbulence within 20 centimeters of the bottom, whereas the ADP records vertical profiles of the ocean current from its position approximately 70 centimeters from the ocean bottom.

The Functioning of the System
All of the sensors are networked, providing a single integrated data stream to the shore-based station. The acoustic sensors are programmed to collect data for 20 minutes every hour, and the optical sensors sample for 10 minutes an hour. The schedule changes at night, with the acoustic sampling continuing through the night during which time the optical sampling is reduced to two minutes per hour. The system’s power requirement peaks at 42 watts. To conserve power (which is supplied by four 85-watt solar panels) the power manager puts the buoys to “sleep” for 40 minutes every hour when sampling or data transmission is not required.

At the designated time, the onboard power supervisor sends a signal that powers up the computer which runs the DACNet program; this, then, turns on the instruments. When the last instrument has collected its data, a relay is turned on to power the wireless Ethernet bridge which transmits all of the time-stamped data to the base station. Once the transmission to shore is confirmed, those files are deleted on-board the buoys to make room for data from the next sampling cycle. All of the data are transmitted to shore at up to11 megabits per second. The raw data is then forwarded in near-real time to Dalhousie University where it is processed.

Should one of the researchers want to receive data from a particular instrument directly, or if he / she wants to change an instrument’s sampling frequency, these modifications can be made from Dalhousie University via a Web interface which uploads the changes to the base station.

Dr. Alex Hay, Professor of Oceanography and Chair in Ocean Acoustics Technology at Dalhousie University, is responsible for the acoustics instruments on the project. Commenting on the system’s ability to accommodate additional instruments following the initial deployment, Dr. Hay said, “This makes it possible for other people to come in and put their instruments in. It’s not easy for non-real-time systems to make allowances for that. This is a very important feature.”
For Dr. Cullen, this capability is key to the project. “Flexibility is what every research scientist wants,” he says. “I want to be able to call a colleague in California and say, ‘Do you want to write a proposal to deploy your instrument on our system for six months? We can provide this to you.’ We want to be able to do this without spending a lot of time and money arranging it.”

Interdisciplinary Science
Dr. Will Perrie, a Research Scientist with the Department of Fisheries and Oceans, and his team at the Bedford Institute of Oceanography plan to add another instrument to the Lunenburg Bay observatory in October. The acoustic doppler current profiler (ADCP), with a waves package which will be placed on a separate mooring, will measure directional waves. They also plan to add one bottom-mounted sensor that looks upward and measures the current profile.

Dr. Perrie’s group will be measuring surface waves and currents in order to validate detailed fine-resolution wave models of the area, which will then be embedded within coarser-resolution large-scale models of the entire Atlantic Ocean through a series of nested grids. They will be using the real-time measurements to “reality test” and further refine state-of-the-art wave models.

In addition, the Meteorological Service of Canada (MSC)—Atlantic (Environment Canada) has installed an array of instruments, both on the buoys and on shore to take a variety of atmospheric measurements. This data will test and validate MSC’s weather forecast models which form the basis for the wind fields that drive MSC’s operational wave models.

“When predicting the future state of the atmosphere,” said Garry Pearson, MSC’s Senior Researcher, “one key element that requires additional research is the effect of the oceans—the heat and moisture flux coming from the oceans and how these interact with the atmosphere.” In addition, the MSC will be enhancing their atmospheric model with sea surface temperature information from the ocean circulation model, developed by Dr. Jinyu Sheng, Assistant Professor of Oceanography at Dalhousie University and leader of the circulation modeling project for Lunenburg Bay. According to Pearson, coupling these models has never been done before in Canada. The information exchange will be reciprocal, with MSC providing wind data to the circulation model.

Data from the ocean circulation model will also feed into the sediment
transport and wave models.

“What we have in this observatory,” Pearson added, “is an attempt at a complete atmospheric and oceanographic description of what’s going on. To do that you have to have a highly instrumented site.

“In addition, the models we’re running are much higher resolution than what we normally use so that we can hopefully pick out more detail in all of the fields: temperature, wind and pressure. This takes more computing power, so this work is computationally intensive.”

The MSC will also be measuring the way fog moves in and out of the Bay, via a visibility sensor attached to the foghorn at the lighthouse, to enhance fog prediction capabilities.

The first practical application of the Lunenburg Bay observatory was MSC’s local weather forecasting for the week-long Volvo Youth Sailing Championship yacht race in Lunenburg in mid-July, 2002. Garry Pearson’s group provided participants with on-site meteorological support and Web-based forecasts (www.atl.gc.ca/sailing_e.html) each morning during the race.

In terms of Dr. Cullen’s own research, he is looking forward to using the light attenuation measurements to distinguish phytoplankton from other materials in the ocean, and also to possibly differentiate species groups of phytoplankton. He is also interested in using the hyperspectral sensor data to learn more about fluorescence as an indicator of species composition and the physiological condition of phytoplankton, which would significantly enhance current scientific knowledge about harmful algal blooms.

Looking beyond the project’s internal objectives, Dr. Cullen sees the Lunenburg Bay observatory as “very well suited for inclusion in GOOS (the Global Ocean Observing System) as an index site”.

GOOS is a worldwide initiative to provide accurate descriptions of the state of the oceans, including living resources, and to provide continuous forecasts of the future conditions of the sea over a variety of time and space scales via a globally coordinated data network. (http://ioc.unesco.org/goos )

Dr. Cullen reports that his team is currently processing data “in major loads”. He adds that, to date, the system is operating smoothly. It will be taken out of the water during the winter months, and he expects that when it is redeployed in the spring, the team will begin to input meteorological and physical oceanographic data into the numerical models.

In time, it is envisioned that an in-depth understanding of Lunenburg Bay will prove invaluable to scientists and marine users alike, and that the project will prove the efficacy of this observatory, which can then be replicated or adapted elsewhere in the world.

Andrew Safer is a freelance writer based in Halifax. He writes articles, case studies, white papers and other technical documents focused on the business case for new technologies. Safer’s previous Sea Technology articles include: “Canada’s Multibeam Platform: Advantages & Applications” (March, 1997) and “Spatial Visualization of the Marine Environment” (February 2001). Both articles can be viewed on his website at http://www.andrewsafer.com/trademag.html.

Media Relations | Success Stories | Internet
CV | Other Services | Contact | Home
Copyright ©2000 Andrew Safer