Scientists in South Africa are using ocean robots to study the unexpected and complex interplay between the Southern Ocean and the Earth’s changing climate.
The researchers have been inspired by the fact that this is the least studied part of the planet and yet one of the most dynamic and difficult to understand. It’s also an area of high species abundance and diversity, where common and highly specialized species both thrive.
The Southern Ocean Carbon and Climate Observatory (SOCCO) research team under the South Africa’s Council for Scientific and Industrial Research (CSIR) has been using robotics to observe the upper ocean’s physical processes and study their impacts on carbon dioxide exchange between the atmosphere and ocean, as well as on the growth of phytoplankton – microscopic algae that live in the surface ocean.
Local experts say these cutting-edge robotic gliders will shape the future of marine research and environmental monitoring of the Southern Ocean and South African waters. They hope that the data they gather will reveal how human activity is changing the planet.
What is known, SOCCO researchers say, is that the Southern Ocean accounts for fifty percent of all carbon dioxide absorbed by the oceans. This part of the world’s ocean is particularly sensitive to climate change because the creatures living here are susceptible to minor environmental shifts, which can lead to large changes in global ocean currents or ecosystem health.
“Our one research gap that we hope to address with these gliders, in conjunction with satellites and computer models, is better understanding the role of small scale features (such as ocean eddies and currents) and short term events (storms) on the overall state of the ocean,” says Dr. Sebastian Swart, CSIR’s principal scientist.
“Before glider technology was available, we were unable to measure these forms of features and variability – research ships and satellite data alone were not enough to investigate this,” he says.
Diving new depths
The first Southern Ocean glider deployments began in September 2012 near the remote Gough Island, about 2000 kilometres south west of Cape Town. For six months, gliders collected and sent data back from one of the world’s harshest environments directly to scientists at CSIR.
Operators based in South Africa used satellite communications to remotely control the location and depth of the robots. Since then, additional long-endurance robotics experiments have taken place in the Southern Ocean on an annual basis.
“Enhancing our understanding of how the Southern Ocean works and interacts with climate today will leave us in a position to better mitigate against the effects of climate change tomorrow,” says Swart.
The robotic glider fleet currently has nine units in total – a number that is expected to increase with the growing need of the global research community for reliable data collecting platforms that can withstand these inhospitable ocean environments. Four are surface wave gliders, which ride the ocean surface measuring CO₂, acidity, and other physical variables of the surface ocean. The other five are profiling gliders that can dive to a maximum depth of one kilometre below the ocean surface.
During each deployment, the gliders “tweet” data via satellite to scientists in Cape Town where it is stored, analysed and eventually made available to the global research community.
According to Swart, CSIR researchers are able to combine the analysis of data from surface wave and profiling gliders by pairing them together while deployed in the ocean to give a complete picture of changes in the water.
The gliders are maintained in Cape Town at the CSIR-led South Africa Marine Engineering and Robotics Centre (SAMERC) in partnership with a marine technology company called Sea Technology Services.
Manufactured in the US, the gliders are modified to suit local needs when they get to South Africa.
“This includes adding new sensors, building new sensors for gliders and ships, refurbishing and maintaining the gliders,” says Swart who has been deeply involved in building the robotic ocean glider program in South Africa.
Swart says a team of South African marine engineers and engineering students at SAMERC have been trained to use the ocean gliders – a unique and advanced skill that can be expanded across the rest of the continent. A string of development programs have already been lined up. Each year young black South African engineers are trained on this advanced ocean technology, particularly the use of robots and scientific sensors. More importantly, Swart says, young marine scientists are graduating from the country’s universities.
Under the guidance of Swart’s current team, the gliders have been instrumental in measuring ocean hydrography, chemistry and optics.
“More specifically they measure CO₂ exchange between the ocean and atmosphere as well as temperature, salinity, dissolved oxygen, light levels, bio-optical properties of the water and weather such as wind, air temperature and humidity,” Swart says.
The gliders are sent out to collect more data from the severe Southern Ocean environment each year.
“We are just beginning our fourth significant Southern Ocean robotics experiment,” says Swart revealing that the research is designed to observe the full seasonal cycle of the upper ocean and its physical-biological processes.
Four gliders were recently deployed from the SA Agulhas II, South Africa’s Antarctic research vessel, in the frigid and stormy winter sea.
“We hope these gliders will continue making measurements until late summer time in 2016 or approximately a year in total,” he says. “They provide measurements that can be used to improve models and forecasts of the oceans and weather.”
Swart’s team has also conducted experiments offshore of Port Elizabeth – the site of a strong current along South Africa’s east coast. Led by CSIR’s Dr. Marjolaine Krug in April 2015, the six-week project showed that it was possible, but challenging, for the gliders to navigate and collect data in areas dominated by strong currents.
Boosting ocean research
At the University of Cape Town’s Nansen-Tutu Centre for Marine Environmental Research (MARE), research and development in ocean prediction, and using measurements of the oceans from satellites, is done.
Dr. Björn Backeberg, oceanographic researcher and co-director of the Nansen-Tutu Centre says that in order to predict ocean currents, ocean models must be combined with ocean observations in order to limit model errors at the nowcast stage (i.e. today), providing a better estimate of the ocean state.
“Ocean prediction is an initial condition problem, so if we can improve our ocean state estimate at the nowcast stage, then we improve our forecast skill for tomorrow, the day after and next week,” he says.
The data collected from satellites controls ocean models in a process called data assimilation, Backeberg says, satellites only measure the ocean surface, thus information about the ocean interior is missing.
He says that to obtain measurements of the ocean interior, researchers have historically relied on measurements gathered by ships, but this form of data collection is incredibly expensive due to the large running costs of the ships.
To hire the research vessel, SA Agulhas II, it costs approximately 300,000 rand (over 23,000 USD) a day. The daily rate for a deployed glider instead is far less – about 3,000 rand (approximately 232 USD).
“Robots, or autonomous platforms, such as gliders and Argo profiling floats are a much more cost effective way of collecting data, especially in remote regions that have severe weather and ocean conditions, such as the Southern Ocean,” he says.
Backeberg says robots are able to collect high resolution information for periods up to six months, depending on battery power, and can provide consistent data throughout the year. Keeping a ship at sea for three months is comparatively much more expensive and observations were generally more sporadic in both space and time.
“Assimilating these types of observations in ocean models as well as using them to validate our ocean model will help improve our ability to predict the oceans more accurately,” he adds.
The future of marine conservation?
Wider applications for these cost-effective robots are already being seen in the fields of ocean governance and resource management, and gliders are being used to monitor fish stocks and patrol the ocean for any signs of illegal activity.
Backeberg adds that the gliders are also beginning to assist in fish stock assessments. The information collected by these robots is providing a vital component to our understanding of fish stock health, fish migration and distribution patterns.
But there are limitations.
“[The gliders] rely on acoustic measurements to estimate fish stock sizes. But acoustic measurements are unable to tell you what type of fish is being monitored. To do that you would still rely on ship measurements and stock assessment trawls,” he says.
Despite these limitations, Backeberg backs the use of intelligent technology in conjunction with existing methods to monitor the ocean.
“Unmanned robots may be an important component of future marine conservation, but they will need to be used in combination with ship surveys and satellite measurements.”
Huge economic benefits are also in store.
Swart says efforts to improve ocean governance and resource management through technology will be able to feed directly into South Africa’s initiative to unlock the potential of its “blue” economy, dubbed Operation Phakisa.
The Phakisa programme was launched in October 2014 by South African president Jacob Zuma to maximise the enormous economic potential of oceans while preserving them.
Ocean activity contributed 54 billion rand to South Africa’s economy and accounted for approximately 316,000 jobs in 2010, according to the government figures.
Swart sees the success of glider research missions potentially leading to economic growth.
“Gliders are a key part of unlocking the wealth of our ocean blue economy,” he says. “They can assist in understanding the ocean, protecting and managing its resources, and manage disasters and pollution.”
This piece originally appeared in SciDev.net.