In late January 2020, the robotic submarine SuBastian descended into the steep-walled canyons off southwest Australia and recorded something few humans have ever witnessed: a giant phantom jelly, its meter-wide bell trailing ghostly arms through the floodlights, drifting in absolute silence. The encounter alone would have been remarkable. But when scientists filtered the water samples collected from those same canyon walls, they found something they were not expecting at all: environmental DNA belonging to Architeuthis dux, the giant squid.
The results, published in the journal Environmental DNA in early 2026, tie together two of the deep ocean’s most elusive animals in a single canyon system off Western Australia. The researchers describe the giant squid detection as a putative range extension for the species in the eastern Indian Ocean, meaning no prior genetic or physical evidence had placed it in these waters.
The expedition and what it found
The cruise, designated FK200126, ran from 26 January to 25 February 2020 aboard the research vessel Falkor, operated by the Schmidt Ocean Institute. Departing Albany and arriving in Fremantle, the team targeted three submarine canyons along the southwest Australian margin: Bremer, Leeuwin, and Perth. ROV SuBastian captured high-definition video and collected biological and geological specimens from the canyon walls, while the ship gathered multibeam bathymetry data to map the seafloor in fine detail.
Alongside the ROV dives, the team collected 178 ten-liter water samples at five discrete depths: surface, 200 meters, 500 meters, 1,000 meters, and bottom depths ranging from 1,750 to 4,540 meters. Two metabarcoding assays (COI Leray and 16S Fish) were applied to the filtered samples. The result: at least 226 taxa detected across the three canyons, including fish, cephalopods, and other marine fauna distributed in distinct depth bands.
Perth Canyon, the deepest of the three, sits within the Perth Canyon Marine Park, part of the Australian Government’s South-west Marine Parks Network. Nutrient-rich water upwells along the canyon’s steep walls, concentrating marine life from surface-dwelling fish to deep-water predators. That upwelling dynamic helps explain why both a giant phantom jelly and giant squid DNA could turn up in the same narrow geographic zone, even if the animals occupy different depth layers or pass through the canyon at different times.
The phantom jelly footage adds a rare visual anchor to the genetic data. Giant phantom jellies (Stygiomedusa gigantea), with bells that can exceed a meter across and trailing arms stretching several meters, have been observed fewer than about 130 times in over a century of deep-sea exploration, according to the Monterey Bay Aquarium Research Institute. Seeing one glide through SuBastian’s lights underscores just how little is known about the midwater fauna in these canyons, a three-dimensional habitat that extends far above the canyon floor where most traditional surveys have focused.
What remains uncertain
The pairing of the jelly footage with the giant squid DNA is striking, but several gaps separate the confirmed data from the full story the combination implies.
No publicly available dive log directly links the exact timestamp and coordinates of the phantom jelly video to the specific water sample that returned giant squid eDNA. Both observations occurred during the same expedition and within the same canyon system, but whether they happened on the same dive, the same day, or at the same depth is not specified in the published record.
The giant squid detection itself carries an important caveat. Metabarcoding amplifies and sequences short genetic fragments from filtered water. While the method is well established for broad-scale species detection, the researchers did not report follow-up Sanger sequencing or the collection of voucher specimens to confirm the species-level identification. Short fragments can occasionally produce false positives through contamination, database mismatches, or degraded DNA that resembles a closely related species. The authors’ use of the word “putative” signals confidence in the detection but stops short of definitive confirmation. Readers should treat it as a credible lead, not a confirmed sighting.
There is also no published hydrodynamic model from the cruise explaining how canyon currents might transport DNA. Genetic material shed by a giant squid could travel considerable distances in deep-water flows before landing on a filter. Without current-velocity data tied to the sampling locations, it is impossible to say whether the squid was physically present near the canyon wall or whether its DNA drifted in from kilometers away.
A broader question concerns timing. The cruise took place in early 2020, and the eDNA paper reached full publication roughly six years later. Long processing and review timelines are common in environmental genomics, where reference databases evolve and bioinformatic pipelines require repeated validation. But the gap also means the data represent a single temporal snapshot. Whether seasonal upwelling, episodic eddies, or other oceanographic cycles influenced which species appeared at depth during that specific window has not been tested.
How the evidence fits together
The strongest pillar here is the peer-reviewed eDNA paper: 178 samples, five depth strata, two assays, and hundreds of taxa. Those figures are independently checkable. The NOAA-archived cruise record for FK200126 independently confirms the expedition dates, vessel, route, and multibeam data collection, anchoring the timeline and geography without relying solely on the researchers’ narrative.
The ROV video sits in a different evidentiary category. High-definition footage from SuBastian is a direct visual record, but species identification from video depends on expert interpretation, including frame-by-frame review and comparison with museum specimens. Earlier ROV work in Perth Canyon established protocols for such identifications, and the canyon’s rich deep-sea community is well documented. Still, each new visual identification requires taxonomic review before it can be treated as definitive.
The giant squid eDNA detection occupies the most tentative position. Environmental DNA is a powerful screening tool, especially for animals that rarely encounter nets or cameras. But a single metabarcoding hit without confirmatory sequencing is best viewed as a signal, not proof. For comparison, a 2021 study in the Gulf of Mexico used similar eDNA methods to detect giant squid genetic traces in deep water, helping validate the approach but also illustrating how rarely such detections occur and how carefully they must be interpreted.
What makes the combined evidence compelling is the overlap. A protected marine park with documented upwelling, a robotic submarine filming rarely seen gelatinous megafauna, and genetic traces of one of the ocean’s most famous deep predators all converge in the same canyon system. Together, they sketch Perth Canyon as a biological hotspot where surface productivity, steep topography, and deep currents intersect to support a layered community from plankton to apex invertebrates.
What comes next for Perth Canyon
For conservation managers, the convergence underscores the value of existing protections while highlighting how much remains unknown. Perth Canyon Marine Park was designated on the basis of geomorphology, productivity, and known whale aggregations. The new eDNA and ROV observations suggest its significance may extend further into the deep pelagic realm than previously documented.
For scientists, the findings provide a clear roadmap: pair repeated eDNA transects with current measurements, deploy baited cameras and midwater ROV surveys, and seek physical specimens that could confirm the presence of giant squid and other elusive taxa in these waters. The Schmidt Ocean Institute’s ongoing deep-sea mapping campaigns in the Indian Ocean offer a plausible vehicle for follow-up work.
And for anyone who has ever stared at the ocean and wondered what moves through its deepest corridors, the image holds: a ghostly jelly drifting past a robot’s lights, and somewhere in the same dark water, invisible genetic traces of a creature that has haunted human imagination for centuries. The canyon walls that funnel nutrients and guide whales may also shelter animals that have never been filmed or formally described. One water sample and one unexpected encounter at a time, the boundary between legend and documented biodiversity keeps shifting.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
