Remote Sensing: Nearshore Oceanography Gets a Technological Boost

A bloom of coccolithophores, an algae with a calcium carbonate shell, is clearly visible as a whitish cloud of water in a satellite image taken over the Scotian Shelf off the coast of Nova Scotia. Access to data from MERIS – one of the main instruments aboard the Earth observation satellite Envisat – is providing Fisheries and Oceans Canada scientists with higher resolution data suitable for nearshore oceanography research. Photo courtesy of the European Space Agency (ESA), the Canadian Space Agency (CSA), the Canada Centre for Remote Sensing (CCRS), and Fisheries and Oceans Canada's Ocean Colour Unit at the Bedford Institute of Oceanography (BIO)

In a satellite image of the Scotian Shelf off the coast of Nova Scotia, a whitish cloud of water surrounds Sable Island in swirls and filaments. The colour of the ocean is caused by a bloom of coccoliths, an algae with a calcium carbonate shell that gives it its white color. The bird's eye view of the bloom, which covered approximately 8,000 square kilometres, highlights the advantage of satellite imagery for oceanographic research – the ability to cover a large area in a short period of time.

"We could never determine the shape and size of that bloom by collecting water samples from a boat and analyzing it in a lab for coccolithophores...you would need thousands of buckets to determine the shape and it would be impossible to get that much detail," says Fisheries and Oceans Canada research scientist Dr. Ed Horne of the Bedford Institute of Oceanography. "This is the first time we've seen a coccolithophore bloom that large around Sable Island. We don't know what that means yet, but we'll use satellite imagery to keep an eye on it from year to year, given concerns that ocean acidification may dissolve the shells of organisms with calcium carbonate shells."

In addition to tracking phytoplankton blooms, satellite imagery is also a valuable tool for sediment research. A MERIS image taken on September 6, 2010, after hurricane Earl swept along Canada's east coast, reveals a cloud of suspended sediment along the coast of Cape Breton Island, Nova Scotia, caused by heavy runoff and high surf. Photos courtesy of the European Space Agency (ESA), the Canadian Space Agency (CSA), the Canada Centre for Remote Sensing (CCRS), and Fisheries and Oceans Canada's Ocean Colour Unit at the Bedford Institute of Oceanography (BIO)

Nearshore remote sensing gets a boost

Satellite imagery has been used for offshore oceanography over the last 10 years, however its application to the nearshore is fairly new.

Since the spring of 2009, an agreement between the Canadian Space Agency and the European Space Agency (ESA) has given scientists at Fisheries and Oceans Canada and universities full access to data from MERIS (Medium Resolution Imaging Spectrometer), one of the main instruments aboard Envisat, ESA's Earth observing satellite

With a resolution of 300 metres, three times higher than earlier satellite instruments, MERIS provides sufficient detail for imaging harbours, bays and lakes. This has opened the door to using satellite imagery for nearshore applications. Among its capabilities, MERIS can determine the exact "colour" of oceans and coastal zones, which reflect biological activity and other processes.

"There are two receiving stations in Canada that continually download all of the full resolution images from MERIS at no cost to us. This has revolutionized our access to this data," says Dr. Horne.

How satellites "see" ocean colour

Satellite instruments or sensors "see" variations in the colour of the ocean by detecting different wavelengths or bands of reflected light. The colours are recorded as numerical values that can be downloaded and converted back into images. Scientists can then compare the colour in satellite images at specific times and locations to actual measurements of chlorophyll or other suspended matter in samples of ocean water collected at the same time and location. The comparison, called ground truthing, enables scientists to develop mathematical formulas or algorithms for analyzing the same parameters in other satellite images. This technique has enabled scientists to track the timing and spread of offshore phytoplankton blooms around the oceans, which has helped explain the timing and success of year classes – fish in a stock that hatched in the same year.

MERIS helps solve Lake Ainslie mystery

Fisheries and Oceans Canada scientists have already used the Medium Resolution Imaging Spectrometer (MERIS) data to determine what caused discoloration of Lake Ainslie on Cape Breton Island, Nova Scotia, in 2009.

"Local residents didn't know what was causing discoloration of the lake. We went back into the MERIS archive and pulled out the data for the same area and time period. The images showed us the development of a cyanobacteria bloom as the lake changed colour," says Dr. Horne. "Ongoing monitoring of MERIS imagery over Lake Ainslie will enable us to answer questions such as does the bloom occur every year or only in warm years or cold years? We can also compare MERIS imagery to environmental data, including sea surface temperature, gathered by other satellites to see if there are any correlations."

Sept 2 2009

Sept 11 2009

Sept 5 2009

Sept 21 2009

Four true color satellite images of Cape Breton Island, N.S., taken by MERIS in September 2009, reveal the rapid growth of algae in Lake Ainslie as it turns from black to green between September 2 and September 5. Ongoing monitoring of the lake using MERIS imagery and on-the-ground sampling will enable researchers to answer questions such as whether the bloom occurs every year or only in warm or cold years. Photos courtesy of the European Space Agency (ESA), the Canadian Space Agency (CSA), the Canada Centre for Remote Sensing (CCRS), and Fisheries and Oceans Canada's Ocean Colour Unit at the Bedford Institute of Oceanography (BIO)

Satellite imagery and ground truthing

Collecting ground truth data is essential to interpreting what satellite instruments "see" in the ocean. Ground truthing involves comparing pixels on a satellite image with direct observations and measurements on the ground or, in this case, the ocean. This enables scientists to verify what they are seeing in the image and to convert the colours to tangible quantities, such as biomass of plankton or sediment concentration. Satellite and ground truth data must be collected at the same time and location.

"For example if we're studying sediment or chlorophyll (used to measure abundance of phytoplankton), we measure their concentration in ocean water samples and correlate that with features in MERIS images," says Dr. Horne.

"It is easier to interpret offshore satellite images because the only thing that changes the color of the ocean, other than the water itself, is the presence of phytoplankton," he says. "Interpreting nearshore satellite imagery is more complicated because in addition to phytoplankton, other suspended matter including sediment and coloured dissolved organic matter (CDOM) from terrestrial matter such as decomposing leaves can change the ocean colour and we don't necessarily know which factor has a more significant effect."

Since the main source of CDOM is from the land, concentrations vary with river outflow and rainfall. "This means that an algorithm for the nearshore that is valid for today may not be valid tomorrow and we'll probably have to develop separate ones for different seasons, conditions and locations," says Dr. Horne.

Sediment monitoring

One practical application for Medium Resolution Imaging Spectrometer (MERIS) data is determining how sediment may potentially affect turbines for the proposed tidal power project in the Bay of Fundy. Fisheries and Oceans Canada researchers have used sediment algorithms from the European Space Agency to analyse MERIS imagery from the area of the bay where the turbines are going to be set up.

"The algorithms worked quite well and suggest that there is variability in sediment concentrations over the year with a marked increase in March," says Dr. Horne. "Consulting companies working on this project are very interested in our sediment data, and this past March 2011, we were able to gather ground truth data to confirm that what we saw in the satellite images really is sediment."

A MERIS image (LEFT)of Nova Scotia (grey denotes land) reveals sediment concentrations (shown in colour) in the Minas Basin and Cape Split (top of image). At the bottom right of the image is Halifax Harbour, which narrows into Bedford Basin at the northern end. The Bedford Institute of Oceanography has been sampling water at the station labelled Pin 1 for more than 20 years. Dalhousie University recently installed a data buoy at this site to gather oceanographic data around the clock. A close-up (RIGHT) of the Bedford Basin reveals variations in chlorophyll concentrations over the basin. Image courtesy of the European Space Agency (ESA), the Canadian Space Agency (CSA), the Canada Centre for Remote Sensing (CCRS), and Fisheries and Oceans Canada's Ocean Colour Unit at the Bedford Institute of Oceanography (BIO)

Bedford Basin Ocean Monitoring Buoy

One of the challenges of ground truthing is being able to gather on-the-ground data at the exact time and location as the satellite. Adding to that challenge is the inability of MERIS to see through fog or clouds, limiting the number of clear images from a particular area. To maximize the number of matchups, Fisheries and Oceans Canada scientists are also using data from Dalhousie University's Bedford Basin Ocean Monitoring Buoy which became operational in 2009. Equipped with a variety of optical instruments for gathering oceanographic and atmospheric data, the buoy operates every day around the clock. The data it collects helps to fill in gaps in weekly sampling that Fisheries and Oceans Canada has carried out in Bedford Basin for the past two decades.

Bedford Basin Sampling

Bedford Basin will serve as a testing ground for the development of nearshore algorithms that will benefit a variety of studies including those related to climate change.

"With climate change, scientists are looking for shifts in temperature and salinity on the continental shelf," says Dr. Horne. "We expect that those same shifts will appear in Bedford Basin because over the years we've shown that what happens in Bedford Basin is indicative of what happens on the Scotian Shelf including phytoplankton blooms, upwelling and climatological shifts."

Dr. Horne and his colleagues are just beginning to develop algorithms for the Bedford Basin. "It's going to take a lot of ground truthing so we can sort out which variables are affecting the satellite signal, but it will be well worth it since we could never get this type of geographical coverage or temporal resolution from ship-based measurements," he says.