{"id":23644,"date":"2026-10-20T17:00:28","date_gmt":"2026-10-20T21:00:28","guid":{"rendered":"https:\/\/www.autonocion.com\/us\/?p=23644"},"modified":"2026-10-20T05:32:21","modified_gmt":"2026-10-20T09:32:21","slug":"23-foot-drone-sailboat-around-antarctica-200-days-alone","status":"publish","type":"post","link":"https:\/\/www.autonocion.com\/us\/23-foot-drone-sailboat-around-antarctica-200-days-alone\/","title":{"rendered":"A 23-foot drone sailboat left Tasmania to circle Antarctica alone for 200 days, met waves over 26 feet within days of leaving, and will log more than 13,000 miles measuring how much carbon the Southern Ocean swallows before coming back to swap its sensors"},"content":{"rendered":"

Last Friday we ran a story about a 1,500-pound drone sailboat that sailed into the eye of a Category 3 hurricane<\/a> off Pensacola and came out the other side still floating. At the time, that felt like the hardest thing you could reasonably ask a 23-foot boat to do. Then I read what one of its sister boats is doing right now, and I’m not so sure anymore.<\/p>\n

On September 20, a tow boat let go of Saildrone Explorer SD-1092 in Storm Bay, just south of Hobart, Tasmania, and the boat turned south on its own. It’s going to keep going until it has sailed all the way around Antarctica, which its operator says should take about 200 days and cover more than 13,000 miles. Nobody’s on board. There isn’t an engine to run out of fuel, either. It moves on a rigid wing, the electronics run off solar panels, and the whole point of the trip is a box of sensors that measures how much carbon dioxide is moving between the air and the water.<\/p>\n

Within days it was sailing through winds of up to 50 knots (about 58 mph) and waves over 26 feet, according to Saildrone<\/a>, the California company that builds and pilots these boats. One of the engineers who deployed it posted a frame from the onboard camera taken on September 21, with 35-knot gusts and a 22-foot significant wave height, and from what I can tell that counts as a fairly ordinary day in the water it’s headed for.<\/p>\n

So why sail a robot around Antarctica at all?<\/p>\n

What the boat is actually measuring<\/h2>\n

The ocean soaks up about a third of the CO2 people put into the air every year, according to NOAA, and scientists estimate the Southern Ocean alone handles about 40 percent of that. Do the math and that one ring of cold water around Antarctica is taking in roughly an eighth of everything we emit. I’d stress the “roughly,” because the entire reason this mission exists is that nobody is especially confident in those numbers.<\/p>\n

Measuring it isn’t complicated in principle. The key instrument on the boat is a NOAA-built sensor that reads how hard CO2 is pushing on the air side and how hard it’s pushing on the water side, which chemists call partial pressure. If the water side is higher, the ocean is giving CO2 back to the atmosphere. If the air side is higher, the ocean is drinking it in. That’s basically it.<\/p>\n

The hard part’s getting a sensor to the right spot. The Southern Ocean isn’t fenced in by continents the way the Atlantic or the Pacific are. It’s defined by a current, the Antarctic Circumpolar Current, which runs clockwise around the whole planet with nothing to stop it, and that current drags deep water up to the surface and pushes surface water back down. Research ships go there in summer. In winter, according to NOAA’s Adrienne Sutton, who co-leads the project, the traditional tools simply haven’t been able to make those observations, which is a polite way of saying nobody wants to be out there in July.<\/p>\n

\n
\n
Hull length<\/div>\n
23 ft<\/div>\n
Saildrone Explorer, 15-foot wing, about 1,500 pounds, 1 to 3 knots.<\/div>\n<\/div>\n
\n
One lap<\/div>\n
200 days<\/div>\n
More than 13,000 miles around Antarctica, Hobart to Hobart.<\/div>\n<\/div>\n
\n
First week<\/div>\n
26 ft waves<\/div>\n
Winds up to 50 knots within days of leaving Storm Bay.<\/div>\n<\/div>\n
\n
ACTIVE<\/div>\n
The fleet<\/div>\n
4 boats<\/div>\n
One launched, three more at three-month intervals during 2027, five years total.<\/div>\n<\/div>\n<\/div>\n

The last time one of these went around Antarctica<\/h2>\n

This isn’t the first attempt. In 2019 an earlier Explorer, SD-1020, left Bluff, New Zealand, on January 19 and came back on August 3, after 196 days and 11,879 nautical miles. On the way it took 50-foot waves, 80-mph winds and, according to Saildrone, collisions with icebergs. Saildrone had actually sent out a pair of boats, and the other one’s conventional wing got wrecked by the first storms. The one that made it had a square-rigged wing built specifically for the trip.<\/p>\n

Two things came out of that voyage. The first is that the boat survived, which apparently surprised a few people. The second one is what the people paying for the new fleet actually care about: in winter, the deep water that the Circumpolar Current churns to the surface is loaded with old CO2, and when it comes up it can release CO2 into the air instead of absorbing it. Robotic floats had been hinting at this, but as far as I can tell the 2019 boat was the first thing to measure the exchange directly, at the surface, through a Southern Ocean winter.<\/p>\n

Which is a problem if your climate models assume the Southern Ocean is a steady, reliable sponge. One boat doing one lap in one year gives you a hint. It doesn’t give you a dataset, and a dataset is what the people building climate models actually want.<\/p>\n

What’s different this time<\/h2>\n

Scale, mostly. Instead of one boat and one lap, the plan is four boats cycling continuously for five years. SD-1092’s the first one out. NOAA says the next three go out at three-month intervals during 2027, and once a boat finishes a lap it comes back into Hobart, gets its sensors swapped for fresh ones, and heads out again. If that works, there should be a Saildrone somewhere around Antarctica at every point of the year, including the months when there’s nothing else out there taking measurements.<\/p>\n

The routes aren’t fixed, either. Researchers at Columbia University’s Lamont-Doherty Earth Observatory are running the data analytics, and they’ll use machine-learning models to work out where the gaps in the carbon data hurt the most, then steer the boats through those regions. Columbia’s Galen McKinley, who co-leads the project with Sutton, says the team expects to adjust the plan over the course of the five years. NOAA calls it “AI-directed,” which sounds a bit grander than what it is, but the idea is sound: you send the boat where the model is least sure of itself.<\/p>\n

The boat changed too, though not by much. Saildrone kept the square-rig wing from 2019 but made it heavier and tougher, and added a camera. A square rig is worth explaining for a second. The standard Saildrone wing works like an airplane wing stood on end and can sail toward the wind. The square version gives that up and mostly runs with the wind behind it. In the Southern Ocean, where the wind blows west to east around the entire planet more or less without stopping, that’s probably the right trade. You don’t need to go upwind if you only ever need to go one direction.<\/p>\n

You may also be wondering who’s paying for all this. The money comes from Schmidt Sciences, the philanthropy set up by Eric and Wendy Schmidt, through a program called the Virtual Institute for the Carbon Cycle. NOAA, Columbia, the University of Washington, Australia’s CSIRO and South Africa’s CSIR are all involved. Nobody’s put a dollar figure on the program that I could find, and Saildrone hasn’t published a public tracker for SD-1092, so I can’t tell you where it is this morning. CSIRO’s Elizabeth Shadwick says each lap should cover more than 24,000 kilometers, and that’s where the 13,000-plus miles comes from. NOAA’s own release rounds it to roughly 12,000 miles, so take the exact number with a grain of salt.<\/p>\n

A sailboat at walking pace, for 200 days<\/h2>\n

Here’s a number I keep coming back to. An Explorer averages somewhere between 1 and 3 knots depending on the wind, call it 2 or 3 mph. That’s a brisk walk. SD-1092 is going to keep up a brisk walk for roughly 200 days through the roughest water on the planet, with nobody aboard to fix anything that breaks, and the people flying it are sitting at desks in California.<\/p>\n

Compare that with the alternative. A crewed research vessel costs a fortune to charter, needs a full crew of sailors and scientists, and loses days to weather whenever the Southern Ocean decides it’s had enough visitors. It also goes home in winter, which is exactly when the data is most interesting. A 1,500-pound boat that just keeps going strikes me as a pretty good deal, even if it has to limp back into Hobart with half its sensors dead. We’ve covered a lot of drone boats lately, most of them parked offshore watching for drug runners<\/a>, and frankly this is the more interesting use of the hardware.<\/p>\n

Where it all ends up is the Global Carbon Budget, the yearly accounting of what people emit and how much the land and the oceans take back. CSIRO says<\/a> the 2026 edition is due in November, and NOAA’s announcement of the full fleet<\/a> went out on October 13. SD-1092 is due back in Hobart roughly 200 days after it left, which puts it in port around early April 2027.<\/p>\n","protected":false},"excerpt":{"rendered":"

Last Friday we ran a story about a 1,500-pound drone sailboat that sailed into the eye of a Category 3 … <\/p>\n

Read more<\/a><\/p>\n","protected":false},"author":8,"featured_media":23704,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[121],"tags":[],"class_list":["post-23644","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","resize-featured-image"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/posts\/23644","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/comments?post=23644"}],"version-history":[{"count":2,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/posts\/23644\/revisions"}],"predecessor-version":[{"id":23705,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/posts\/23644\/revisions\/23705"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/media\/23704"}],"wp:attachment":[{"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/media?parent=23644"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/categories?post=23644"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.autonocion.com\/us\/wp-json\/wp\/v2\/tags?post=23644"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}