The communities announced in July 2025 were not discovered during a July 2025 dive. The observations came from an expedition between 8 July and 17 August 2024, when the Chinese crewed submersible Fendouzhe crossed sections of the Kuril-Kamchatka Trench, the Kamchatka-Aleutian transition and the western Aleutian Trench.
What appeared in Nature on 30 July 2025 was the analysis. Xiaotong Peng, Mengran Du and a large international team reported chemosynthesis-based animal communities over a roughly 2,500-kilometer region, at depths from 5,800 to 9,533 meters.
This is one study, not settled consensus. It documents the deepest known animal community associated with chemical-rich seeps and an unusually extensive set of habitats. It does not show that every hadal trench contains the same ecosystem or that the entire surveyed distance was carpeted with animals.
The distinction I find most useful is between a depth record and a regional description. The first community at 9,533 meters was dominated by tubeworms. Clam beds and other bivalve-rich communities were found at different, generally shallower sites. “Tubeworms and clams” describes the expedition’s connected discoveries, not necessarily one mixed colony at the maximum depth.
At 9,533 meters, the pressure was about 98 megapascals
The deepest site sat on black mud in the Kuril-Kamchatka Trench, at the boundary between the trench bottom and the basement of an accretionary prism. The paper calls it The Deepest. Conditions there were about 2.25 degrees Celsius and 98 megapascals, close to 970 times atmospheric pressure at sea level.
Fendouzhe’s lights revealed dense colonies of frenulate siboglinids, a group of slender marine tubeworms. Their tubes reached roughly 20 to 30 centimeters in length and about one millimeter across. White free-moving bristleworms up to 6.5 centimeters long threaded among them.
Earlier work had found small chemosynthetic animal communities in the Japan Trench, including clams at 7,326 to 7,434 meters. Possible microbial mats were photographed deeper, at 10,677 meters in the Mariana Trench, in a 2020 Deep Sea Research study. Those structures were not accompanied by a documented seep animal community.
The new record is therefore not the deepest suggestion of chemosynthetic microbes. It is the deepest direct observation of a substantial animal community powered by seep chemistry. That narrower wording is what the evidence supports.
Fendouzhe turned isolated samples into a mapped view
Fendouzhe, or Striver, is a three-person, full-ocean-depth submersible operated by the Chinese Academy of Sciences’ Institute of Deep-sea Science and Engineering. It reached 10,909 meters during trials in Challenger Deep in 2020 and is rated to operate at 11,000 meters.
During the 2024 cruise, the vehicle made 23 dives. A University of Southern Denmark account says observers encountered extensive animal communities on 19 of them. Manipulator arms collected animals, sediment cores and geological samples while high-definition cameras recorded their positions and surroundings.
Direct observation changed the scale of the finding. Earlier trawls had recovered some animals associated with chemosynthetic habitats, hinting that seeps existed. A trawl can show what was caught, but it cannot preserve a clear picture of patch boundaries, animal density or the fault running through the site.
The communities were not continuous. The paper emphasizes patchiness among trenches, depths and individual seeps. At Wintersweet Valley, around 9,120 meters down, Fendouzhe followed a field for roughly two kilometers during one dive. Thousands of tubeworms occupied part of that field. Other sites were dominated by clams, tube-dwelling polychaetes or white microbial mats.
The animals outsource primary production to microbes
Photosynthesis turns light energy into chemical energy and fixes inorganic carbon into organic molecules. In complete darkness, chemosynthetic microbes perform the carbon-fixing step using energy released by reactions involving reduced chemicals such as methane or hydrogen sulfide.
Many seep animals make that microbial metabolism part of their own bodies. Siboglinid tubeworms host bacteria in specialized tissue and can lack a conventional mouth and gut. Bivalves can house chemosynthetic bacteria in their gills. The microbes build organic matter; the animal supplies access to chemicals, oxygen and a protected habitat.
A NOAA overview of cold-seep communities explains the division in broad terms: methane-oxidizing partners commonly nourish mussels or clams, while sulfide-oxidizing partners support tubeworms. Real communities contain multiple host species and microbial pathways, so this is not one metabolism copied into every animal.
The headline’s methane and hydrogen sulfide belong to the chemical system as a whole. The Nature paper measured methane-rich and sulfide-rich fluids and identified different animal assemblages across the sites. It did not demonstrate that every tubeworm and every clam simultaneously consumed both compounds through the same symbiont.
The fuel rose through faults created by a bending plate
The Kuril-Kamchatka and Aleutian trenches mark subduction boundaries where the Pacific Plate bends and descends beneath neighboring plates. ScienceBlog’s earlier explanation of why the Pacific Ocean is slowly closing describes the larger tectonic process. At trench scale, bending opens normal faults through the incoming plate and its sediment cover.
The deepest community occurred where one such fault reached the seafloor. Seismic data linked the fault to flexure of the incoming plate. Across the survey, black sediments, methane, sulfide, carbonate minerals and animals clustered near routes along which fluid could migrate upward from deeply buried layers.
Gas and isotope measurements point to a microbial origin for the methane. Methane made up all hydrocarbons detected in the analyzed headspace samples. Its carbon and hydrogen isotope signatures were consistent with microbes reducing carbon dioxide derived from sedimentary organic matter, rather than heat cracking buried material into thermogenic gas.
This makes “life without sunlight” true locally but incomplete globally. No sunlight reaches the trench floor, and the seep animals do not depend on local photosynthesis. Yet some of the organic carbon buried in those sediments originally fell from productive surface waters. The system reroutes old surface production through deep microbes, methane, sulfate reduction, sulfide and animal symbiosis.
Thousands of worms does not mean trillions across the trench
At its densest photographed patches, the study counted as many as 5,813 ± 1,335 siboglinids per square meter and 293 ± 69 bivalves per square meter. Those are striking local densities. They are not mean values for the 2,500-kilometer study region.
The team estimated density from video. Two laser points ten centimeters apart supplied a scale, a standardized quadrat was drawn on each selected frame, and visible animals were counted manually. For each dive, researchers chose three to ten representative screenshots showing the densest seep communities.
That method answers how crowded the conspicuous patches could become. It cannot count animals buried below the visible surface, and it was not designed to calculate a total population for an entire trench. “Thousands” is directly supported for colonies along large seep fields. A basin-wide biomass total would require systematic area mapping beyond these dive tracks.
The same caution applies to the 2,500-kilometer span. It is the geographic distance across which sites were recorded, not the length of one unbroken biological corridor. A review of northwest Pacific hadal zones shows how varied these connected trenches are in depth, topography and ecological setting.
The carbon-cycle claim is a hypothesis with a new foundation
Hadal food webs are often described as depending heavily on particles and carcasses sinking from above. The new sites add another route. A deep sediment biosphere can turn buried carbon into methane, faults can deliver methane-rich fluids upward, and chemosynthetic microbes can convert that chemical energy into animal biomass.
The authors suggest that methane could be stored in trench sediments for long periods rather than immediately subducted into deeper Earth. They also propose that geologically similar trenches may contain more chemosynthetic communities than direct observation has revealed.
Both ideas are plausible extensions of the measurements, not global inventories. The expedition sampled one interconnected northwest Pacific region. It did not measure every hadal trench, determine the worldwide mass of methane hydrate or calculate the fraction of the ocean carbon cycle supported by deep seeps.
What the dives establish is more concrete. At depths beyond nine kilometers, fault-fed fluids support microbes capable of sustaining dense, varied animal communities. The record is not just about survival under pressure. It shows that geology can deliver a dependable energy supply to large animals in one of the darkest and least observed habitats on Earth.