This article was originally featured on Hakai Magazine, an online publication about science and society in coastal ecosystems. Read more stories like this at hakaimagazine.com.
Jake Spink fished British Columbia’s craggy coast for four decades. Now, as the president of the British Columbia Coast Pilots—an association of highly trained captains that guide thousands of tankers, cruise ships, and other large vessels into the province’s ports each year—Spink can draw detailed maps of the shoreline from memory. Even so, navigating the region’s fickle and sometimes perilous weather can be a challenge. There have been many times when Spink has motored up a flat calm channel, coffee in hand, and turned between two islands to suddenly face blinding spray and waves breaking over the bow. Twice, while piloting through storms near Haida Gwaii, off the province’s north coast, he fought 100-kilometer-per-hour wind—strong enough to uproot a tree—blowing from the southeast only to have it flip within ten minutes to howling just as hard from the northwest.
“For wind to change direction that fast, it’s shocking,” says Spink. “That’s what catches people unaware.”
Checking the weather forecast doesn’t always help. In British Columbia, as in many coastal regions, predicting the weather is notoriously difficult. High mountains and narrow channels funnel wind, trap fog, and cause moist air to dump torrents of rain, while extreme tides and fast currents build imposing waves. The result is a patchwork of hazardous meteorological microclimates, where conditions can differ wildly between neighboring fjords and across small islands.
The problem, explains Melissa Westland, a graduate student with the Weather Forecast Research Team at the University of British Columbia, is that weather models are based on regional weather patterns and don’t account for the dramatic influence of such diverse terrain. “Precipitation and wind are particularly hard to forecast,” Westland says. High winds stir up waves, making boating much more treacherous for recreational and commercial vessels.
The Canadian government operates roughly 60 land-based weather stations and buoys along the BC coast, but they are spread too far apart to capture conditions everywhere. To decide if it’s safe to cast off, then, ship captains are left shuffling between subscription weather services and free apps, says Scott Beatty, the CEO of Victoria-based start-up MarineLabs. Even ferry captains, responsible for safely shuttling hundreds of people between ports, have been known to consult windsurfing message boards to get information.
As artificial intelligence becomes more mainstream, Beatty’s company is building a new machine learning–based weather forecasting system to make shipping safer. Beatty says that most users don’t realize that existing weather apps, such as Windy and Windfinder, almost all depict the same government forecasts. But since 2017, MarineLabs has deployed more than 60 new weather buoys and sensor arrays to collect additional data about the conditions along crucial shipping routes near Vancouver and Prince Rupert, Canada’s largest West Coast ports, around major East Coast harbors, and in one test location in the United States.
Every 15 minutes, MarineLabs’ devices upload real-time wind and wave data to a digital dashboard; some also take 360-degree photos. With as little as nine months of sensor data, Beatty says, MarineLabs can use machine learning to pick out patterns in how the weather around the sensor differs from the wider regional forecast. For instance, a shallow reef might be extra windy or have larger waves than expected. Armed with these patterns, the start-up can generate improved hyperlocal forecasts—a paid service it started offering in July 2024. “It’s quite amazing,” Beatty says. “It works a lot better than we were expecting.”
Twenty years ago, Spink says, mariners watched the barometer and listened to marine radio reports to monitor storms. Now, equipped with hyperlocal weather forecasts, coastal pilots and port authorities have more precise information to decide whether a ship should hurry into harbor before a storm hits or ride it out at sea. “When we look at these buoys and we see [visibility] is restricted, then we can call it off before it’s too late,” he says.
Early attempts at hyperlocal forecasting proved the technique’s worth. In 2014, just a year after coast pilots on Canada’s east coast installed a weather buoy in the approach to the harbor in Halifax, Nova Scotia, the Australian Spirit—a tanker capable of carrying 700,000 barrels of oil—lost its rudder in bad weather. Using a forecast based on buoy data, the pilots identified a 48-hour window when conditions would be calm enough for tugs to guide the wounded ship into port. “It was just seamless,” says Andrew Rae, the retired captain who led the push to install the buoy. MarineLabs now has multiple sensor arrays dotting the Nova Scotia coast to monitor conditions in even greater detail—and act as a substitute when the original buoy goes offline.
Rae, who piloted some of the world’s largest container ships into the Port of Halifax for 35 years, says that better local forecasts are also helping the shipping industry’s push to decarbonize. If ship captains have advanced notice that a storm bars their path into port, he says, most will slow down to avoid the bad weather, reducing their fuel usage and emitting less planet-warming carbon dioxide.
MarineLabs is just one of many companies looking to help captains make these decisions. Others, including Sofar Ocean, located in San Francisco, California, are deploying drifting buoys that capture data from some of the vast voids in marine weather monitoring. Tomorrow.io, based in Boston, Massachusetts, launched its first monitoring satellite in 2023 to feed even more information into its customizable AI-powered forecasts. And on a larger scale, technology giants Google, Microsoft, Huawei, and others are developing AI algorithms that model and predict global weather, in some cases more accurately than standard government forecasts.
Westland wishes MarineLabs would release its raw sensor data to scientists studying how climate change is upending weather rather than keeping the information behind a paywall. But, she says, “it’s great to have more sensors in the ocean.”
Spink agrees. When he’s aboard a massive cruise ship bound for downtown Vancouver and the summer wind is gusting from the southeast, he’s grateful for MarineLabs’ system. “[There’s] more and more reliable data coming from these buoys,” he says. “We’d be happy having an extra five or six of them on the coast.”
This article first appeared in Hakai Magazine and is republished here with permission.