Marine mammals

Dark whales breaking the surface of the sea, near the coast. — Southern right whales near Hermanus, South Africa (photograph by Olga Ernst, 2008). Olga Ernst, CC BY-SA 4.0, source

Figures revised in autumn 2026. Covers all the groups, cultural transmission, acoustic threats and demographic analysis.

In 1830, there were about twenty-seven thousand of them in the western South Atlantic. By the mid-1950s, four hundred and fifty remained.¹

In 2019, there were once again about twenty-five thousand.

This page tells the story of how we know this — and why the figure has changed twice, once without the whales having anything to do with it. It also explains why protection sometimes succeeds spectacularly, and why it fails when the nature of the threat has changed.

The range of life: several returns to the water

Before turning to hunting and figures, we need to widen our view. When we say “marine mammals”, we think first of whales. This is a distorted perspective: adaptation to marine life occurred at least four times independently in the course of evolution.

Marine mammals comprise four distinct major groups:

Cetaceans (whales, dolphins, porpoises, sperm whales): fully aquatic, they never leave the water and have lost their hind limbs over tens of millions of years.
Pinnipeds (seals, sea lions, walruses): semi-aquatic, they keep their four limbs, transformed into swimming paddles, and return to land or ice to give birth and moult.
Sirenians (dugongs and manatees): the only strictly herbivorous marine mammals. They graze tropical and coastal seagrass beds, acting as gardeners of the seabed.
Recent marine carnivores (sea otters and polar bears): the sea otter has the densest fur of any mammal (up to 140,000 hairs per cm²) to make up for the lack of an insulating layer of fat.²

Ecosystem engineers

Each of these groups shapes its environment. The sea otter is the textbook example: by eating sea urchins, it prevents kelp forests (giant algae) from being grazed bare. Where otters disappear, sea urchins proliferate, the kelp collapses, and with it the entire coastal fish nursery.³

For their part, the great whales act as a “nutrient pump”: they feed at depth and defecate at the surface, supplying iron and nitrogen essential to phytoplankton, which produces a major share of the planet’s oxygen.⁴

The collapse and the industrial machine

Modern industrial whaling arrived in the South Atlantic in 1904, when the first whalers set up in South Georgia. Alexandre Zerbini, who led the reconstruction of this history, reports what the whalers found at the time: the waters were full of them.⁵

They went first for humpback whales, because they live close to the coast and are therefore easy to reach.

By the end of the 1920s, the ships were finding only a few dozen a year.

What happened next, and what explains everything else

The whaling stations did not close.

They switched species. After the humpbacks, the blue whales. Then the fin whales. Then the sei whales. Finally, the minke whales.

This point goes far beyond the case of whales:

Exhausting a species does not stop an industry. It moves it.

The very same mechanism appears in our page on fishing. This is no coincidence: it is the same logic of exploitation in successive cascades.

Protection, and the result

International protection measures appeared in the 1960s. Then, in 1982, the International Whaling Commission (IWC) decreed a moratorium on commercial whaling, which came into force during the 1985-1986 season.⁶

In 2019, the western South Atlantic humpback whale population was estimated at 93% of its pre-exploitation level.⁷

This, incidentally, is how an honest researcher gives a figure: not “93%”, but “93% (95% PI = 73–100%)”. The range matters as much as the number.⁸

It is a major success for conservation biology. It has been measured, dated and published.

The fall in detail, year by year

The model allows the collapse to be followed closely, and it was brutal.

In 1904, when the whalers arrived, the population numbered about 24,700 individuals. Twenty-two years later, in 1926, about 700 remained. In that interval, more than twenty-five thousand humpback whales were killed.⁹

The low point was reached around 1958: about 440 animals, or 1.6% of the original population.¹⁰

Finally, one detail shows how fragile a recovery is: an upturn began in the early 1960s, then 190 whales were killed in 1967 by illegally operating Soviet fleets — and the population fell again. It was not until 1972 that all whaling ceased.¹¹

The heart of this story

Only a few years ago, an official assessment concluded that this population had recovered only 30% of its original level.¹²

The 2019 reassessment gives 93%.

It was not the whales that multiplied their rate of reproduction. It was the counting that became more refined.

The researchers built a model that brings together three kinds of data: catch records, abundance estimates, and information on the genetics and biology of the species.¹³

They did something more, which deserves mention: they ran the model with different assumptions — other data, other computational assumptions — to see how much the result moved. This is what is called testing the sensitivity of a model. We ask not only what result we obtain, but how much that result depends on our choices.¹⁴

Among these uncertainties, one in particular: the “struck and lost” whales.

These are the harpooned whales that escaped mortally wounded and sank or died further away. They appeared in no official catch record, since they had never been brought ashore or aboard the factory ships. By incorporating these estimates into the historical record, two things became clear:¹⁵

The starting population was larger than had been thought.
Its capacity to rebuild once the pressure was removed was markedly greater than forecast.

How do you put a figure on whales that, by definition, were never counted? By reading the whalers’ logbooks. A study published in 2021 went through nearly twenty thousand days of voyages in the southwestern Atlantic, recording every harpoon strike that missed or was lost.¹⁶

What the official records left out, the logbooks noted. The authors conclude that the scale of historical whaling may have been seriously underestimated.¹⁷

Those that did not come back

The story of the blue whale is readily told: hunted to the brink of extinction, protected, then recovered from about two thousand individuals to somewhere between ten thousand and twenty-five thousand.

That is true. If this page stopped there, it would be dishonest.

The North Atlantic right whale

It has been protected since 1935. Nine decades.¹⁸

It never came back.¹⁹

The latest published estimate (for the year 2024) gives about 384 individuals — after a low of about 360 in 2020. A slight rise, then, but from numbers so low that it does not change the diagnosis.²⁰

Only about seventy females capable of breeding remain.²¹

Why legal protection is not enough

Because hunting is no longer the main threat.

Every year, monitoring networks record new cases of entanglement in fishing gear (lobster traps, gillnets) and collisions with vessels.²²

In a population of barely 380 animals, every individual lost counts double. We stopped what used to kill them (the harpoon). We have not stopped what kills them now (shipping traffic and fishing gear).²³

The most worrying demographic detail

Females are declining faster than males.

Repeated entanglements do not always kill outright: they cause chronic stress, deep wounds and a colossal expenditure of energy spent dragging lengths of rope over thousands of kilometres. For biologists, this is one of the reasons why the interval between two births for a female has increased from 3 years to 7 to 10 years on average.²⁴

This means that reproductive potential collapses even before the total population figure drops sharply. The number of living individuals can give the illusion of relative stability while the species’ capacity to renew itself is already broken.

Counting the animals is therefore not enough. We have to count who they are.

A shared phenomenon

The North Pacific right whale (Eubalaena japonica) now numbers only about thirty individuals in its eastern stock.
The Mediterranean monk seal (Monachus monachus) numbers only about 450 to 600 adults, taking refuge in sea caves because of tourist pressure and urbanised coastlines.
Of the fourteen large whale species, seven remain classified as endangered or vulnerable.²⁵

The new threats: noise, plastic and invisible traps

To understand the contemporary marine world, we have to realise that the sea is no longer silent or free of obstacles. Current threats fall into three broad categories:

1. Underwater noise pollution

Marine mammals live in a world of sound waves. Visibility underwater rarely exceeds a few tens of metres; sound, on the other hand, travels more than four times faster there than in air, and over thousands of kilometres.

Cargo ship engines, seismic oil exploration (air guns) and low-frequency military sonar saturate the ocean. This permanent background noise:²⁶

Masks communication calls between mothers and calves or between breeding partners.
Disorients the acoustic navigation of toothed whales (dolphins, sperm whales), causing mass strandings during military exercises.
Induces chronic acoustic stress, impairing their immune system.

2. Bycatch and ghost nets

Industrial fishing deploys kilometres of lines and nets. Dolphins and seals get caught in trawls (such as common dolphins in the Bay of Biscay). Unable to surface to breathe, they drown within minutes.

“Ghost nets” — lost or abandoned at sea — go on fishing blindly for decades, trapping pinnipeds, turtles and cetaceans indiscriminately.

3. Chemical bioaccumulation

Because they are at the top of the food chain and live a long time, marine mammals concentrate organochlorine pollutants (PCBs), flame retardants and heavy metals (mercury) in their fat. In some populations of resident killer whales in the North Pacific or the Mediterranean, the mother’s milk is so loaded with toxins that the first-born receives a massive dose during nursing.²⁷

Songs and cultural transmission

It was long believed that culture — the transmission of learned behaviours from generation to generation without a genetic basis — was unique to humans. The study of marine mammals has swept away that certainty.

The humpback whales’ “top 50”

Among humpback whales, all the males of a given population sing the same complex song at any one time. But this song is constantly evolving.

Researchers have discovered a fascinating phenomenon of cultural spread from west to east in the Pacific Ocean: a new melody starting in the western South Pacific spreads rapidly eastward, from population to population, as far as the central South Pacific. The whales abandon their old song to adopt the new acoustic “fashion” within a few months.²⁸

Dialects among sperm whales and killer whales

Female sperm whales express themselves through series of clicks called codas. Each social clan has its own repertoire of codas, forming a true acoustic dialect. Two clans living in the same geographical area keep their distinct dialects and do not mix.
Killer whales probably pass on through social learning not only their vocalisations but also, according to most researchers, their highly specialised hunting techniques: deliberate stranding on the beaches of Patagonia to catch sea lions, or creating synchronised waves to tip seals off ice floes in Antarctica.²⁹

Consequence for conservation: When a population of killer whales or whales disappears, it is not only a genetic heritage that dies out; it is also accumulated cultural knowledge, a memory of migration routes and a language that vanish for ever.

What the researchers add: the global ecosystem

The scientific article announcing the recovery of the humpback whale in the South Atlantic also contains caveats. Here they are:

This growing population eats krill (Euphausia superba) — it is estimated to consume between 1.5 and 2.6 million tonnes of it per season. It therefore competes directly with penguins (Adélie, chinstrap) and seals (crabeater seal), which depend on the same resource.

Krill itself is weakened by warming waters and the melting of winter sea ice, beneath which grow the algae that krill larvae feed on.

The authors write that continued monitoring remains essential.³⁰

A success is not an end. It is a state measured on a given date, which needs to be measured again.

This is exactly what our page on the ozone layer says. Two different cases, the same lesson.

A word on the evolutionary trajectory

Whales, seals, sea otters and dugongs are all mammals. They breathe air with lungs, suckle their young, are warm-blooded, and their common ancestors lived on dry land.

They therefore returned to the water.

This fact deserves emphasis, because it contradicts a very widespread misconception: evolution does not climb a ladder. A lineage that had left the aquatic environment much earlier gradually returned to it 50 million years ago for cetaceans and sirenians, and more recently for pinnipeds. This made them neither more nor less “evolved” than their land-dwelling cousins.³¹

Evolution has no favoured direction. It has only opportunities and circumstances to adapt to.

Note 1

They declined from nearly 27 000 […] individuals in 1830 to only 450 […] whales in the mid-1950s.

Zerbini, Alexandre N. et al., “Assessing the Recovery of an Antarctic Predator from Historical Exploitation”, Royal Society Open Science, vol. 6, 190368, 2019