Article body copy
One after another, the sensors went dark. In normal times, technicians tasked with maintaining the small network of meteorological instruments scattered off the coasts of Great Britain and Ireland could have traveled to fix or replace the defunct devices. But the ongoing COVID-19 pandemic meant they could only watch in vain as the technology failed, leaving weather forecasters without a handful of important data, including atmospheric pressure measurements. At the start of 2020, this regional network had 12 locations providing data. Because of kaput sensors, that number now stands at just seven.
Emma Steventon, marine networks manager at the United Kingdom’s Meteorological Office in Exeter, knew she had to come up with a plan. In June, she and her team sent eight drifting buoys to the port of Liverpool where they were loaded onto a ship and subsequently dropped into the Atlantic Ocean off Ireland’s southwest coast. The spherical buoys, encased in cardboard packaging that breaks down in seawater, soon separated and drifted off into the distance. “This was something new that we’ve not done before,” she says. The buoys, she anticipated, would provide a short-term fix, filling the data gap left by failing sensors. “We were expecting them to be picked up by the currents and be washed ashore within a few months.”
But the team was in luck. To date, all bar one of the buoys continue to supply data from the desired region. Though not a like-for-like replacement for the failed sensors, the buoys are picking up some of the shortfall in the flow of meteorological information to Steventon’s office.
This gaggle of instrumentation sloshing around in the North Atlantic represents just a tiny slice of a gigantic system. Globally, many thousands of buoys, floats, ship-based sensors, and human observers supply weather forecasters with precious data about the conditions at sea.
This vast data-gathering operation is a collection of several networks. The Data Buoy Cooperation Panel, for example, looks after drifting buoys such as those used by Steventon and her colleagues. Then there’s the Voluntary Observing Ship (VOS) program, in which either human officers or automated weather stations on ships record and transmit data for variables such as temperature and wind speed.
In the Ship of Opportunity Programme (SOOP), scientists travel on a commercial vessel such as a container ship and take atmospheric and oceanographic measurements as they traverse the ocean. Yet another network is Argo, a sensor array that uses thousands of floating high-tech devices able to automatically submerge themselves in the water to retrieve measurements at various depths—from the surface to thousands of meters below.
The data flowing from these and various other networks amounts to many millions of observations every year. Much of the data is continually transmitted to the World Meteorological Organization (WMO), and ultimately, it influences the models used by national forecasters to predict the weather.
The COVID-19 pandemic is playing havoc with several of these networks. Some ships fitted with instruments that automatically record and transmit temperature and pressure are stuck in port, unable to make their usual observations from wide open parts of the ocean where monitoring is less frequent and, therefore, most needed. Instruments on ships must also be calibrated or replaced with precalibrated devices to ensure their readings remain accurate. But in many cases, the technicians who visit ships to do this work have been barred from doing so.
The number of Argo devices in operation has fallen by 10 percent, too. “Ultimately, their power gives up,” explains Emma Heslop, an oceanographer with the Global Ocean Observing System (GOOS) who has watched as floats’ batteries have steadily died. The sharp drop in the number of research vessels at sea means scientists have not been replacing the floats at the usual rate of about 60 per month, she says.
Worse still, some data-gathering efforts have ceased almost entirely. Justine Parks manages a program to measure subsurface ocean temperatures for Scripps Institution of Oceanography at the University of California San Diego. As contributors to SOOP, Parks and fellow scientists travel aboard commercial ships traversing the Pacific Ocean, deploying hundreds of disposable sensors as they go. As the probes sink on their one-way journey to the seafloor, they transmit temperature readings along a thin, hairlike wire.
Typically, Parks and her colleagues would complete 20 crossings every year. But they’ve only managed one since last February, because shipping companies have restricted who can board their vessels. “It’s catastrophic,” says Parks. “I didn’t grasp that it was going to be so long in the beginning. You know, we were really holding out hope for summer and resuming a lot of our activities.” That never happened.
It’s not just marine observations that have dwindled. Aircraft are a further key source of meteorological data. But with international air travel sharply curtailed, data from flights has fallen significantly. One analysis published in July suggested that this caused temperature forecasts between March and May 2020 to be off by 0.5 to 1 °C in some regions.
So far, there is no hard evidence that COVID-19 disruptions to VOS, SOOP, Argo, and all the other observing programs have adversely affected weather forecasts, says Darin Figurskey, the operations branch chief at the Ocean Prediction Center in College Park, Maryland, part of the US National Oceanic and Atmospheric Administration. But with the pandemic likely to drag on for many more months, the worry for some is that noticeable errors will start to appear.
“The longer that we’re forced to go without being able to maintain these networks and sensors likely to fail, then yes, I’d imagine that we would start to see wider impacts coming in to the numerical weather prediction and impacting forecasts,” says Steventon.
It is not possible to reliably infer surface pressure from satellites, for example, which is why the thousands of sensors across the globe that gather this information are so important. A lack of data about surface pressure over a relatively small area can affect the accuracy of forecasts that matter to mariners, says Figurskey—from the development of storms, to times when the wind is scarce. “Having good pressure measurements certainly helps forecast hazards better,” he says.
Lars Peter Riishojgaard, director of the earth system branch at the WMO, agrees. “We cannot live without surface pressure measurements over the ocean, that I can say unequivocally,” he says. “How many can we lose and still carry on and do our job? I really hope that we don’t have to find out.”
There could be other, perhaps less obvious, consequences, too, argues Kevin Kloesel, a meteorologist at the University of Oklahoma: less robust meteorological observations might make people less likely to trust scientists, notably on subjects such as climate change. “The last thing we want to do as scientists is to allow this discussion to become about belief—Do you believe the data?” Kloesel says.
Currently, a race is on to patch as many gaps in the networks as possible.
Launching from Wellington, New Zealand, the crew of the research vessel Kaharoa, operated by the National Institute of Water and Atmospheric Research, endured a 75-day voyage rocked by storms to deploy more than 160 Argo floats in the Indian Ocean. The floats are now spread in a thin, meandering line from Australia to the west coast of South America. During their travels, the crew members were prevented from disembarking at ports in Australia and Mauritius because of border closures caused by the pandemic. Heslop, of GOOS, says another cruise could also go ahead soon in the Atlantic Ocean that would deploy 80 Argo floats off the coasts of Europe, the eastern United States, and South Africa.
Meanwhile, in the United Kingdom, besides the drifting buoys project, Steventon and her colleagues are also working with ships remotely—sending recalibrated instruments by mail to vessels in port, and emailing their instructions so crews can install the devices themselves. Helping ships’ officers do this technical work from afar, such as explaining how to connect the devices to their ships’ IT systems, has been tricky, says Steventon. “If that becomes the norm, it might become more challenging, because their time is precious,” she says.
In California, Parks says a shipping firm finally took one of her technicians aboard in November, and officers on two other vessels have agreed to take measurements on her team’s behalf. These band-aid fixes will yield only a fraction of the usual data, since the officers have other duties and are not trained to correct any measurement problems.
“The COVID-19 pandemic has been a shock to the system,” says Heslop, who points out that, as a whole, the scientists and engineers involved in the global observing system have shown resilience and have helped to ensure data continues to flow for the most part. But “there is a need to remain vigilant,” she adds.
So far, efforts like these have kept the meteorological insights ticking over. Time will tell exactly how serious the pandemic’s disruption of weather-data gathering and forecasting will be. For now, many researchers are doing their best to deploy whatever instruments they can, wherever they can.