Science1 distinct publisher3 min readPublished
Industrial whaling stopped on a known date, which is why this counts among the cleanest natural experiments conservation has. But the figure now in circulation arrives with a gap: no baseline, no stated survey method, no error bar.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
For this population, whaling came close to a single-variable experiment nobody designed. The mortality was industrial and recorded, and it was switched off by law on a known date [4]. Conservation rarely gets a shape like that to work with. Most interventions act on habitat or bycatch inside systems where several pressures were moving at once, and the counterfactual has to be modelled rather than watched.
What the sequence cannot resolve is which prohibition did the work. The international moratorium, Brazil's own statute and the later protection of the Abrolhos breeding grounds all landed within a few years of each other [4][5], and a single recovery series has to carry all three. Abrolhos is the largest breeding ground for the South Atlantic population [3], which means the same water supplies both the trend and the explanation offered for it.
The number itself is thinner than the story resting on it. Read the "almost half" backwards and the implied world total sits somewhat above 70,000 animals [1]. AFP's account does not carry the survey method, an uncertainty range, or the 1980s figure the recovery is being measured from [16], and "driven to near extinction" [2] describes an outcome without giving a number to measure it against. What is on offer is an endpoint and a recollection: Enrico Marcovaldi, who founded the Humpback Whale Institute, says that four decades ago nobody knew there were whales in Brazil, and that he took his first one for a rock [7].
The identification work is the firmest part of the reporting, and it needs one caution. The 8,300 catalogued animals amount to roughly a quarter of the South Atlantic estimate [2], but that fraction counts matches, not coverage. A catalogue lists individuals ever photographed, an abundance estimate counts animals alive now, and the source does not say across how many seasons the identifications accumulated [16]. The value sits in the matches rather than the total. One whale logged in Brazil was found more than 15,000 kilometres away off Australia's east coast in research published in May [9], and skin and blubber taken by crossbow have linked an Abrolhos-born whale to a relative off South Georgia, some 4,000 kilometres south [10][11]. Milton Marcondes of the institute told AFP that Southern Hemisphere populations look far more interconnected than researchers had assumed [12].
The 2014 delisting is the part I would hold onto [6]. It was a decision taken on a trend, and the instrument that produced the trend is the same one that would register a reversal, so the counting has to outlast the political reason for having started it. The Hope Spot designation Mission Blue gave Abrolhos in October [15] marks recognition, not protection, and forty years on from the moratorium [3] the arithmetic remains the only thing that will show whether the recovery holds.
Ranked by verification strength, evidence, and original report placement.
Happy Whale, an artificial intelligence platform that matches the unique pattern of white markings on a humpback's tail, has identified 8,300 Brazilian humpbacks; the comparison once required many pairs of eyes and hours of work.
One humpback identified in Brazil was spotted more than 15,000 kilometres away off Australia's east coast, according to research published in May.
The whales now face threats from maritime traffic, fishing nets and warming seas caused by climate change, which threatens the krill they survive on.
The IBJ has received financial support from Brazil's state oil company Petrobras for 30 years; Petrobras drills for oil off Brazil, recently made new discoveries off the coast of the Amazon, and supports several species recovery projects in the country.
The warm waters off the coast of Brazil are the largest breeding ground for South Atlantic humpbacks, where males compete to mate and females give birth and nurse calves; researchers work each year at the Abrolhos archipelago off Bahia.
Humpback populations have thrived since a worldwide moratorium on commercial whaling in 1986.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
A named ship and a published timetable turn the Arctic into a bookable lane1 distinct publisher
science
Honduras puts 234 of 298 municipalities on drought alert, and names its own end date1 distinct publisher
science
Belgium's peat fire took seven days and four countries' aircraft to contain1 distinct publisher
science
Namibia signs off on Sandpiper, and seabed phosphate dredging stops being hypothetical1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Firsthand scenes, orphaned headline number
AFP's reporters were on the catamaran, and it shows: the crossbow biopsy, the fluke photography, the Abrolhos field season are all observed rather than briefed. The 35,000 figure is the exception. It appears once, in the third paragraph, with no institution behind it, no method, no error bar and no 1980s count to measure the rebound against, and nothing in the remaining thousand words comes back to it.
Photo-ID as routine field kit
The giveaway is not a launch announcement — it is that nobody in the story treats the software as new. Matching that used to need many pairs of eyes now runs in seconds, 8,300 Brazilian animals sit in the catalogue, and the Brazil-to-Australia record emerged from ordinary use rather than a set-piece experiment. What holds the score down: the report never says across how many seasons those identifications accumulated, so the depth of use stays a guess.
Recovery real, precision borrowed
'Comeback' is fair and the piece earns it — the 1987 Brazilian ban, the protected breeding ground, the 2014 delisting, a founder who mistook his first whale for a rock. What it cannot earn is the arithmetic it invites. 'About 35,000, almost half the world's total' implies a global stock a little over 70,000, and that conclusion rests entirely on a round number with no survey attached. The direction of travel is solid; the decimal places are on loan.
An oil company funds the good news
Two facts sit lower in the story than they deserve: Petrobras has bankrolled the institute for 30 years while making fresh discoveries off the Amazon coast, and every scientist quoted works for that same institute. Nobody is caught out here — AFP discloses the funding plainly — but a recovery story reported from a grantee's boat, with no outside biologist to test the count, is a structure worth naming.
One wire, no second look
A single AFP dispatch, carried by phys.org, is the entire record behind our coverage, and it is fresh and internally consistent. The field observation is trustworthy on sight; the population number deserves to be held loosely until a count with a stated method turns up somewhere it can be checked.