Krill fishing closed on 12 August: the 620,000-tonne threshold no longer serves the purpose it was created for

The Antarctic krill fishery closed on 12 August 2026, having reached the 620,000-tonne threshold set for Area 48 by the Commission for the Conservation of Antarctic Marine Living Resources. This is the second consecutive year the threshold has been hit before the end of the season. The real issue is neither the tonnage nor the date: it is that this figure was devised to trigger a spatial allocation of catches, that the allocation expired at the end of 2024, and that the first season without it produced exactly what it existed to prevent.

What the official figures say

2024 catch	498,350 t, 12 vessels
2025 catch	624,918 t, an all-time high, closure on 1 August
2026 catch	threshold reached, closure on 12 August, about 11 vessels
Trigger level	620,000 t for Subareas 48.1 to 48.4 (Conservation Measure 51-01)
Precautionary catch limit	5.61 million tonnes, nine times the threshold
Allocation among subareas	none since the 2025 season

Five countries fished in 2025: Norway for 326,307 tonnes, China for 231,083, South Korea for 28,622, Chile for 20,035 and Ukraine for 18,871. Over the past decade, Norway accounts for 62.6% of catches and China for 20.8%.

Where the 620,000-tonne figure comes from

This figure is almost always presented as a cautious ceiling derived from krill biology. The Commission itself explains that it is nothing of the sort, and even takes the trouble to warn against the misreading.

The official report states that the 620,000-tonne limit is based on historical catches and represents the sum of the maximum catches taken in each of the subareas over the whole history of the fishery. It adds that it should not be confused, as sometimes happens, with the highest catch ever taken across the whole of Area 48 in a single year. It is therefore an accounting figure, not a biological one.

Its role is more telling still. In 1991, Conservation Measure 32/X capped Area 48 at 1.5 million tonnes and specified that if the catch in Subareas 48.1, 48.2 and 48.3 exceeded 620,000 tonnes, limits by subarea would have to apply. The report spells out the reason: to prevent the whole of the cap being taken from a small part of the area.

In 2000, after an acoustic survey put the biomass of Area 48 at 56 million tonnes, Conservation Measure 32/XIX raised the cap to 4 million tonnes and kept the same mechanism, this time requiring limits for units smaller than the subareas. In 2007, Conservation Measure 51-01 introduced the term “trigger level” and maintained it until such time as the Commission defines an allocation of this total among smaller management units.

In other words, 620,000 tonnes has never been a harvest limit. It is the threshold beyond which fishing was to stop being free to concentrate wherever it chose.

What the end of Conservation Measure 51-07 produced

The awaited allocation arrived in 2009 with Conservation Measure 51-07: at most 25% of the threshold in Subarea 48.1, 45% in Subareas 48.2 and 48.3, 15% in 48.4. The report notes that these percentages deliberately add up to more than 100%, so as to leave the fleet some flexibility while spreading the effort.

The measure was interim and had to be renewed each year. In 2024, for the fourth consecutive year, the Commission failed to reach the necessary consensus, and it lapsed at the end of the season. China, backed by Russia, blocked the renewal at the Hobart meeting. The report concludes: from the 2025 season onwards, the combined catch of Subareas 48.1 to 48.4 is limited to 620,000 tonnes with no further spatial distribution of the threshold.

The 2025 season is therefore the first without an allocation, and the figures published by the Commission are unequivocal.

Sector	2024	2025
Subarea 48.1, Antarctic Peninsula	164,641 t	359,266 t
Share of the 620,000 t threshold taken in 48.1	26.6%	57.9%
Bransfield Strait, west and east	126,296 t	258,289 t
West of the Antarctic Peninsula	38,111 t	100,674 t

The peninsula subarea went from 26.6% to 57.9% of the threshold, where the lapsed measure had capped it at 25%. Catches there more than doubled in a single season. In the Bransfield Strait alone they also doubled, and west of the peninsula they rose by a factor of 2.6. Over the same period Subarea 48.3, around South Georgia, fell from 156,920 tonnes to 5,022. The fleet is not fishing more: it is fishing in the same place.

What a blue whale is actually looking for

Why does concentration matter more than tonnage? Antarctic krill, Euphausia superba, is very nearly the only link between phytoplankton and almost everything else: whales, seals, penguins, birds and fish. Among mysticetes the dependence is extreme. The blue whale eats several tonnes of it a day during the feeding season, and so do the Antarctic minke whale and the southern fin whale, as our page on diets sets out. These animals fast for the rest of the year.

But the argument would remain an intuition without a study published in 2019 in Scientific Reports, which measured what an Antarctic blue whale looks for in a swarm. It is the centrepiece of the case, and it is rarely cited in the debate on quotas.

The set-up is unusual. During a 42-day voyage in the Ross Sea aboard the Tangaroa, the team deployed 310 sonobuoys which recorded 42,489 vocalisations from blue whales, and obtained 7,437 positions by triangulation, listening to two buoys at a time. These animals can hear one another more than two hundred kilometres away. While passive acoustics located them, a scientific echosounder mapped the size, depth and density of krill swarms beneath the hull. In all, 1,129 swarms were characterised using eleven variables, then compared according to whether a whale was detected nearby or demonstrably far away.

The result is clear. Three characteristics predict the presence of a blue whale: the density of the krill, its depth and the height of the swarm. When a swarm is dense, above about 300 grams per cubic metre, shallow, less than thirty metres down, and some fifteen metres high, a blue whale has a strong chance of being present within a radius of forty kilometres. The model reaches an area under the curve of 0.91 at that scale, which is high.

The reason is mechanical. A rorqual feeds by engulfing a considerable volume of water in one go, a manoeuvre that is costly, and the more so the larger the animal. For the largest of them all, only swarms that are both dense and close to the surface return more than they cost. A diffuse layer does not feed it.

And the authors note the decisive point: this particular form of krill is neither widely nor uniformly distributed. The swarms a blue whale can exploit occupy a fraction of the ocean, structured geographically. Yet that is also exactly what a trawler is looking for: nobody fishes a diffuse layer. The two predators are not competing for an average biomass, they are competing for the same minority of swarms. This is why a catch of 620,000 tonnes, which amounts to 11% of the precautionary catch limit, may count for much more than that ratio suggests.

The authors are cautious about causality: they cannot rule out that the presence of the whales itself influences the shape of the swarms. But they write explicitly that this predator-prey relationship matters for the future management of krill fisheries.

Dead whales in the trawls

One impact, by contrast, is documented and quantified, and it is almost absent from public debate. The Commission’s report keeps a count of the incidental marine mammal mortalities in the fishery.

In 2021, three humpback whales were caught in krill trawls in Subareas 48.1 and 48.2. These were the first cetacean mortalities ever observed in this fishery. The report notes in passing that many of the bird deaths that year are linked to the same tows, through entanglement in the net. One humpback whale was killed in 2022, two in 2024 along with one individual released alive but injured, and one in 2025.

The detail of the 2025 case is worth reading twice. The vessel involved was operating without a cetacean exclusion device, with a single seal exclusion device. It fitted one for the rest of the season. Seal exclusion devices have been mandatory since 2008, following episodes that killed dozens of fur seals; their equivalent for the large cetaceans is not.

These numbers are small and that has to be said: seven humpback whales killed in five years, plus one injured, threaten no population. They count for two other reasons. First, because they all date from after 2021, in a fishery that has been active since the 1970s: this is a new phenomenon, one that accompanies the concentration of effort near the peninsula. Second, because they are observed only thanks to observer coverage raised to 100% of fishing days since 2021. Before that, nobody was counting.

East Antarctica, and what is actually known about it

The most widely discussed element of the summer of 2026 is an intention, not a fact. According to Sea Shepherd Global, which made the information public on 10 August, almost all the large krill trawlers have notified their intention to fish in Divisions 58.4.1 and 58.4.2, in East Antarctica, several thousand nautical miles from the usual grounds. These divisions fall under Conservation Measures 51-02 and 51-03 and together allow up to 892,000 tonnes, more than the whole of Area 48.

Three clarifications are called for, because this story is circulating in an approximate form.

The sector has not been untouched since 1991. The Commission’s report is precise: there was no commercial krill fishing in these two divisions between 1991 and 2016, then low catches from 2017 onwards. The description of a refuge intact for thirty-five years is therefore inaccurate; that of a sector almost entirely unexploited remains correct.

The claim about blue whales comes from an NGO. Sea Shepherd describes this sector as one of the major feeding grounds of the species. The source is an opinion piece signed by the organisation’s campaign director, which cites no study. What the literature establishes is more modest: a synthesis published in 2024 in Frontiers in Marine Science, based on 952 sonobuoys deployed between 2006 and 2021 across about 330 degrees of longitude, places the Indian sector, approximately 30 to 150 degrees east, among those where the acoustic presence of the Antarctic blue whale has varied most. The species is therefore indeed present there. But the 2015 Ross Sea study, the only one to have measured in detail what these animals exploit, did not cover this sector. The characterisation of East Antarctic swarms remains to be done.

Notifying is not fishing. Notification is an annual formality, to be filed before 1 June, which preserves an option. For the 2026 season, six countries had notified fifteen vessels. These declarations are subject to withdrawals and revisions, and the Commission publishes the current position.

It remains that the population concerned is the hardest hit of all the large cetaceans by the industrial whaling of the twentieth century. Reduced to less than 1% of its pre-war numbers, estimated at 239,000 individuals, it is rebuilding very slowly and is still listed as Endangered, as our page on conservation status recalls. Opening an industrial fishery in a sector that has almost none would amount to removing a de facto refuge. No rule prohibits it, precisely because that refuge has never been formalised.

Penguins are protected, whales are not

One detail of the season illuminates the logic of the arrangement. According to Aker BioMarine, nearly 90% of the catches taken near the peninsula were taken after 1 April, outside the penguin breeding season. Since 2019 the industry has also applied voluntarily closed areas around the large colonies of Subarea 48.1, covering 74,000 square kilometres, from 1 October to 15 March.

These precautions are real. They show that the fleet knows how to adjust its activity to limit an identified impact. But they are calibrated on the penguin cycle, whose colonies are visible, fixed and mapped on land. The Commission’s ecosystem monitoring programme, set up in 1985, also rests largely on land-based predators.

Nothing equivalent exists for the large cetaceans, whose feeding grounds are mobile and offer no fixed point to protect. The Commission has acknowledged this: in 2024 it set up a temporary working group devoted specifically to cetacean monitoring, and in 2025 its scientific committee recommended incorporating a strengthened monitoring programme into the new management approach for the fishery. That is a useful admission. Our pages on marine protected areas and the international legal framework set out this asymmetry: it is easier to protect what does not move.

Krill also plays a climate role

Krill makes a vertical migration every day: it rises to feed on phytoplankton near the surface at night, and descends several hundred metres during the day to escape predators. In doing so it carries carbon captured at the surface down to depth, as faecal pellets and moults, which settles out of reach of the atmosphere.

The large cetaceans amplify the transfer in the other direction. They feed at depth and defecate at the surface, returning iron and nitrogen to the lit layer where phytoplankton lack them. This mechanism, known as the whale pump, supports the production that feeds the krill itself. The system is a loop: fewer whales means less fertilisation, therefore less phytoplankton, therefore potentially less krill. Our page on blue carbon sets out these estimates and their uncertainties, which remain wide. Taking krill is therefore not only removing food from predators, and this dimension enters into no quota calculation.

The limits of what can be asserted

1. No population-level effect has been demonstrated. The direct mortalities in trawls are documented, but nobody has shown that krill fishing measurably affects the size of a cetacean population. The concern rests on the spatial overlap and on ecological reasoning, not on a data series.

2. Climate acts in the same direction. Krill depends on sea ice for its larval cycle, and that ice is retreating. Separating the effect of fishing from that of warming is precisely one of the two objectives assigned to the Commission’s monitoring programme since 1985, and it has not been met. Our page on climate change returns to this point.

3. The biomass remains poorly known. The estimate of 56 million tonnes used by the Commission dates from the synoptic survey of 2000. The industry cites a figure of 63 million tonnes from a 2019 survey. Neither of these numbers says how that biomass is distributed, which is the only question that counts for a predator.

4. Some of the figures come from the industry. Catches, vessels and incidental mortalities are published by the Commission. The share of catches after 1 April and the extent of the voluntary areas, on the other hand, come from the companies or from their association. They are plausible and nobody disputes them, but they are not independent.

What is at stake in October

The Commission works by consensus, which gives every State a de facto veto. It has already failed once on this precise question. The official report recounts the 2025 meeting: after the unprecedented closure of the fishery, the scientific committee and the Commission attempted to bring together revised catch limits, strengthened monitoring and elements of the marine protected area proposed for the peninsula, in stages, with the explicit aim of reducing the spatial concentration of the fishery. The Commission did not reach consensus.

The same two texts return in October 2026: a marine protected area of about 455,000 square kilometres around the Antarctic Peninsula, and an overhaul of the management of the fishery. The refusal to renew Conservation Measure 51-07 occurred four years running before it lapsed, and the failure of 2025 is the fifth. The question is therefore not whether a blockage is possible, but whether it will stop.

Sources: CCAMLR, Fishery Report 2025: Euphausia superba in Area 48, CCAMLR Secretariat, 2 March 2026, for the history of Conservation Measures 32/X, 32/XIX, 51-01 and 51-07, the origin of the 620,000-tonne threshold, the catches by subarea and by management unit, the incidental marine mammal mortalities, observer coverage and the reports of the CCAMLR-43 (paragraph 4.45) and CCAMLR-44 (paragraph 4.56) meetings. Miller E. J., Potts J. M., Cox M. J., Miller B. S., Calderan S., Leaper R., Olson P. A., O’Driscoll R. L. and Double M. C., “The characteristics of krill swarms in relation to aggregating Antarctic blue whales”, Scientific Reports 9, article 16487, 2019, DOI 10.1038/s41598-019-52792-4, New Zealand and Australian voyage of 2015 in the Ross Sea, IWC-SORP blue whale project. Miller B. S. et al., “Antarctic sonobuoy surveys for blue whales from 2006-2021”, Frontiers in Marine Science, 26 April 2024, DOI 10.3389/fmars.2024.1324816. Sea Shepherd Global, “Krill Fleet Eyes Expansion Into Blue Whale Feeding Grounds”, 10 August 2026. Aker BioMarine, press release of 13 August 2026, and Association of Responsible Krill harvesting companies (ARK), for the 2026 season and the voluntary measures.