The Repository Abyss Report: Into the Dark

The Blue Goo: Filmed Five Times, Never Collected

October 25, 2026 · Runtime 23:31 · Watch on YouTube

In August 2022 a NOAA robot filmed a small, lumpy blue animal on the sand off St. Croix. The scientists watching live could not say if it was a soft coral, a sponge or a sea squirt, and it was never collected. We counted every appearance in the dive video, measured it against the robot's lasers, and read the DNA in the only samples that dive brought up.

Chapters

  1. 0:00 The blue goo
  2. 0:40 Dive 08, southwest of St. Croix
  3. 1:33 A slope of sponges
  4. 2:38 17:56, the first one
  5. 3:31 18:11, the famous one
  6. 3:53 Measured against the lasers
  7. 4:31 The third, fourth and fifth
  8. 7:32 Five, not several
  9. 7:54 The room names it
  10. 10:10 Three guesses
  11. 10:30 1. Soft coral
  12. 11:18 2. Sponge
  13. 12:57 3. Sea squirt
  14. 14:00 What the camera cannot settle
  15. 14:48 Why nobody picked it up
  16. 16:11 Six miles from shore
  17. 17:00 Every dive's samples
  18. 17:46 Photo number seven
  19. 18:11 The only sample: water
  20. 18:50 1.1 million DNA reads
  21. 19:05 What water cannot say
  22. 19:50 Still unnamed
  23. 20:35 Exactly what is missing
  24. 21:52 The urchins in hats
  25. 23:05 Next: the sea squirt that hunts

Transcript

The blue goo 0:00

This is the blue goo, filmed five times and never collected. Four hundred metres down off St. Croix, the scientists watching live could not tell if it was a coral, a sponge or a sea squirt. The robot's arm was right there. Why did nobody pick it up?

We went back to the dive itself. We counted every appearance in NOAA's video, measured the animal against the robot's lasers, and read the DNA in the only samples that dive brought up. Every moving picture in this episode is NOAA footage from that dive. And at the end, we will say exactly what is still missing, and why it was missing on the day.

Dive 08, southwest of St. Croix 0:40

The date is the thirtieth of August, twenty twenty-two. The ship is Okeanos Explorer, and the robot is Deep Discoverer, working with its companion vehicle, Seirios, which watches it from above. This was the third leg of an expedition called Voyage to the Ridge. It started in the Azores, on the Mid-Atlantic Ridge, and ended in the Caribbean, off Puerto Rico and the U.S. Virgin Islands. Dive eight was the second to last.

The target was the southwest corner of St. Croix, where part of the island's slope once slid downhill. The dive plan says the goal was to explore fish habitat on that landslide deposit. The robot touched bottom at twelve fifty-eight, universal time, six hundred and nine metres down, and spent the next seven hours driving uphill.

A slope of sponges 1:33

Most of the slope is sand, with ledges of pale limestone, and the first hours are fish. Fish hovering over the sediment, fish resting on the bottom, letting the robot come close. Then, on the rock, sponges. A lot of sponges. And the biologist leading the science that day, Joana Xavier, is a deep-sea sponge researcher.

That matters for what comes next. On the live feed, the team talks the whole time, from the ship and from shore. Here they are at half past five, on a sponge that hides most of its body in the sand.

17:56, the first one 2:38

At seventeen fifty-six, the robot is following a sea star across the sand, four hundred and thirty-two metres down. Then, just past it, a small blue cushion. Lumpy, covered in dark blue bumps, sitting on the sediment like it was dropped there. The camera looks for half a minute. And the room notices.

18:11, the famous one 3:31

Fifteen minutes later, at eighteen eleven, a second one, seven metres shallower. This is the one that became famous. Rounder this time, standing up off the sand, with a pale, see-through rim around its base, like a skirt. NOAA's photograph of this exact animal went around the world within a week.

Measured against the lasers 3:53

On the close-up, it looks big. It is not. A few seconds later, the camera pulls back, and two red dots appear on the sand, one on each side of it. Those are Deep Discoverer's lasers, and they are ten centimetres apart. So we measured it, frame by frame.

In fourteen frames in a row, while the camera drifted and the gap between the dots changed from sixty to a hundred and twenty pixels, the blue body stayed between three and three and a half centimetres across. That is smaller than a golf ball. The whole mystery fits in a closed hand.

The third, fourth and fifth 4:31

Eight minutes after that, at eighteen nineteen, the third one. Lower and wider, more like a cap than a ball, with bits of sediment stuck to its top and something white and curled caught at its edge. The camera stays longer this time. And the sponge specialist has just walked back into the room.

A minute later, someone on the call says the line this whole episode turns on. It comes at the very end, after the guesses run out. Listen for it.

Then nothing for half an hour. Sand, whip corals, a crustacean in a burrow. At eighteen fifty-one, at four hundred and fifteen metres, the fourth. And this one does not look like a ball at all. It is a flat, lobed sheet, spread across the sediment, with the same dark blue bumps all over it.

Two minutes later, the fifth. Round again, with the same clear rim, and a pale slit along one side. This one is not among the dive's still photographs. We found it only in the video.

Are these five the same thing? We cannot say that either. A ball, a cap, a flat sheet. What they share is the colour, the darker bumps, and on three of them a clear rim at the base. And all five sit on open sand. Not one of them on the rock ledges where this dive found its sponges.

Five, not several 7:32

So here is the count, from the dive itself. Five appearances in fifty-seven minutes, all of them on open sand, between four hundred and thirty-two and four hundred and fourteen metres, along about a hundred metres of seafloor. NOAA's own description says several times. Five is what the video shows.

The room names it 7:54

And over the fourth one, the room stopped trying to name it on the spot. This is the live feed at eighteen fifty-one.

Two minutes later, over the fifth, the options get listed one more time. This is where the name comes from. Listen for the one that stuck.

Three guesses 10:10

Soft coral, sponge, or tunicate, which is the formal name for a sea squirt. Those three are the room's guesses, and NOAA's caption repeats them word for word. Each one predicts something different that a camera might see. So let's take them one at a time, against the five animals.

1. Soft coral 10:30

A soft coral is a colony of tiny animals called polyps, and NOAA's own description gives the test. Each polyp has eight tentacles, and they are feathery, fringed with little side branches. The polyps open them to catch food drifting by. There were corals on this same dive, and on one of them, the room went looking for live polyps, branch by branch.

On the blue animals, across all five, we did not see a single open tentacle. That is not a verdict. Coral polyps can pull their tentacles in, and a closed colony can look like a bumpy skin. Those dark blue bumps could be closed polyps. They could also be something else entirely. The camera cannot tell a closed polyp from a wart.

2. Sponge 11:18

A sponge has no organs at all. Water comes in through many small pores and leaves through larger openings, and the body is held up by tiny mineral needles called spicules. A sponge does not move and does not close. So what the camera can look for is the exits, the larger holes the water leaves through.

On the five blue animals, we could not find a clear opening of that kind. At this distance, a small one would not show. And the real test of a sponge is the spicules, which are inside it. You need a piece of the animal for that. On the feed, the case against a sponge went further, and the reason was the colour.

3. Sea squirt 12:57

A sea squirt is the most surprising of the three, because it is not a simple animal. It is a distant relative of ours, with a body wrapped in a tough coat called a tunic. Water goes in through one siphon and out through another. And sea squirts can live alone, or as colonies.

That would fit the flat one at eighteen fifty-one. And the fifth one has that pale slit along its side. A siphon? A fold? A tear? At this distance, the video does not say. For a while, the sea squirt was the room's best guess. Here is how it was put, after the second sighting.

What the camera cannot settle 14:00

So the score after five encounters is not one, two, or three. It is zero. Every feature that would decide between them is either something the animal does when it feels like it, or something inside it. The camera recorded shape, size and colour. None of the three answers depends on those. Between sightings, the room had already summed it up.

Why nobody picked it up 14:48

Which brings us back to the question, and to the line from the feed. A pity we cannot poke it. Deep Discoverer collects animals on most dives. On this same expedition, it brought up fifty-two biological samples. The blue goo was three and a half centimetres across, on open sand, five times. Here is the reason it stayed there, in the dive report's own words.

None except water samples, as we were within the twelve nautical mile boundary of St. Croix's territorial waters and did not have a permit to sample. That twelve-mile line has existed since December nineteen eighty-eight, when a presidential proclamation extended the territorial sea of the United States, and of the U.S. Virgin Islands, to twelve nautical miles from shore. On the feed, it was said out loud fifteen minutes before the first blue animal appeared.

Six miles from shore 16:11

So we measured how far inside the line the robot was. From the dive track to the nearest point of St. Croix's coastline, on a standard public coastline map, it is about eleven and a half kilometres. Six nautical miles. Halfway to the boundary. Close enough to the island to need a permission the ship did not have.

And St. Croix was not in the plan. The expedition's written plan, dated July twenty twenty-two, lists the waters where it had asked for clearance: Portugal, Puerto Rico, the British Virgin Islands and Anguilla. The U.S. Virgin Islands are not on that list. The day before, on dive seven, on Lang Bank, off the same island, the report has the same sentence, word for word.

Every dive's samples 17:00

Here is the whole expedition, dive by dive, counted from NOAA's specimen list. Six dives brought up animals, and four of them rocks as well. The shipwreck dive took nothing, by design. And the two dives around St. Croix brought up no rock and no animal at all. Only water. On the feed, they said what the robot did instead.

Photo number seven 17:46

There is one more thing in that report. Its list of notable observations for dive eight mentions the sea urchins, a stingray, a squid and the layered limestone. Not the blue goo. The animal that would carry this expedition into the New York Times and onto NPR appears in the report exactly once, as the caption of photo number seven. Blue encrusting organism.

The only sample: water 18:11

So the only physical thing that came up from dive eight is water. Three bottles, closed near the bottom and kept for environmental DNA, the traces that animals shed into the water around them. This year, NOAA published the results, and two of those three bottles are in the public data. We downloaded it.

The timing is better than you might think. One bottle was closed at four ninety-three metres, more than an hour before the first sighting. The other was closed at nineteen oh six, four hundred and twelve metres down, about thirty metres from the fourth blue animal, fifteen minutes after the robot left it.

1.1 million DNA reads 18:50

Together, the two bottles hold one million, one hundred and twelve thousand DNA reads. Of those, thirty-five are sponge. Seven are soft coral. And sea squirt: zero. Not one read.

What water cannot say 19:05

That sounds like an answer. It is not. Water carries DNA from everything nearby, and this slope was covered in sponges we filmed all day. And across the whole expedition, the method barely sees sea squirts at all. Of twenty-five water samples, sponge DNA turned up in twenty. Soft coral in twelve. Sea squirt in only seven.

So zero sea squirt reads beside the blue goo does not rule out a sea squirt. And thirty-five sponge reads do not make it a sponge. A bottle of water can say what was around. It cannot point at one animal and give it a name. The only way to do that is a piece of the animal itself.

Still unnamed 19:50

A week after the dive, NOAA posted the clip, and the blue goo went everywhere. Newspapers, radio, the big news sites. In twenty twenty-five, NOAA's own summary of the whole campaign still calls it a taxonomic mystery. We searched the papers and the press since then, and found no identification.

So this is what can be said, with numbers. Five of them, three to three and a half centimetres for the one we could measure, on open sand, between four hundred and fourteen and four hundred and thirty-two metres, southwest of St. Croix. Blue, with darker bumps, sometimes a clear rim. No tentacles seen. Not collected.

Exactly what is missing 20:35

And this is exactly what is missing. Not a new robot, and not a deeper dive. A piece of tissue the size of a fingernail, in a jar. Under a microscope it would show spicules or not. Its DNA would sit next to sponges, corals or sea squirts in the reference libraries. And to take it, a permit for the waters around St. Croix. The room said it on the feed, in the middle of the hour.

The urchins in hats 21:52

The dive did not end on the mystery. At nineteen hundred, at four hundred and eleven metres, the robot drove into a gathering of sea urchins, at least thirty-five of them, each one wearing a hat made of debris. That is the headline the dive report chose, and you can hear why.

Next: the sea squirt that hunts 23:05

Next time, the other strange animal from this dive, the one everyone could name. A sea squirt is built to filter water. It pumps and sieves, all its life. But in the deep sea, one family of them lost the sieve. Their mouth became a pair of lips that close on whatever drifts in. That is the next Abyss Report. The sea squirt that hunts.

Description and sources

EX2206 dive 08, southwest of St. Croix, U.S. Virgin Islands, 30 August 2022 (Okeanos Explorer, ROV Deep Discoverer). In the dive video we count five appearances of the blue organism, 17:56 to 18:53 UTC, at 432 to 414 m, along about 100 m of sand. Against the robot's 10 cm lasers, the one NOAA photographed is 3 to 3.5 cm across. No animal or rock was sampled because the dive was inside St. Croix's 12 nautical mile territorial sea without a permit; by our measurement the track was about 6 nautical miles from shore. In NOAA's public eDNA data for two of the dive's water bottles, 1,112,572 reads: 35 sponge, 7 soft coral, 0 tunicate. NOAA's 2025 summary still calls it a taxonomic mystery.

PRINT-READY, FROM THIS CHANNEL

Read This Dive, Vol. 01 — the method, and five dives read with it

https://therepository.gumroad.com/l/read-this-dive

PRIMARY SOURCES

ARCHIVE PLATES — public-domain documents, with author, date, archive and licence.

  • | NOAA Ocean Exploration | public-domain

Abyss Report reads the deep ocean through the record that resolved it -- instrument, date, position and the primary sources above.

About the material. Every moving frame is NOAA Ocean Exploration video from EX2206 dive 08, 30 August 2022, public domain, credited on screen; depths come from the vehicle's 1 Hz track, which matches the dive report (the archive's image metadata reads about 8 m shallower). Several cuts play the original control room audio; each window was re-transcribed in isolation twice and used only where both passes matched. The count, the laser measurement, the distance to shore (Natural Earth 1:10m coastline) and the DNA read counts (OBIS, EX2206 samples D08_02W and 03W, CC0) are ours. Every diagram is drawn in code.

#DeepSea #NOAA #OceanExploration #BlueGoo #Okeanos