Science1 distinct publisher3 min readPublished
A decade of public posts from Australia's east coast shows humpbacks lunging at krill and baitfish between August and October, which makes the capital-breeder model a poor guide to where migratory protection belongs.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
Two measurements are in tension here, and they are not measuring the same quantity. Whaling and stranding records from mid-latitude Australian waters found empty stomachs in the vast majority of humpbacks taken on migration [9], which is a statement about gut contents at one moment. The new work counts behaviours instead: open mouths, ventral pleats expanded on a lunge, bubble nets, whales surfacing through dense baitfish schools [3]. Both results can hold at once. A stomach that is usually near empty is exactly what you'd expect from opportunistic snacking that leaves little residue, and that snacking is hard to square with treating the corridor as pure transit, which is what the capital-breeder framing invites [8].
The sampling is the part to hold at arm's length. Records came from public Facebook and Instagram posts by whale-watching operators, drone pilots and enthusiasts [3], so the resulting map is partly a map of where cameras and paying passengers happen to be. The NSW south coast around Eden, Merimbula and Bermagui supplied 45% of verified events, eastern and southern Tasmania 28%, the Victorian coast 12% [4], which puts 85% of the record in three stretches of coast [15]. That concentration would look much the same if feeding were widespread and observers were not. The authors' own note that Phillip Island and northern Tasmania had been underrepresented [5] reads as easily as coverage improving as it does behaviour spreading.
The counts have a shape worth naming. Divide 1,417 individuals by 199 observations and you get about seven whales per event [16]. But 16 of those events were super-groups of 20 to more than 150 animals [7]. That is 8% of the records, and even at the minimum group size they account for at least 320 whales, roughly 23% of the total [17]. The average is doing very little work; a few large aggregations over upwelling driven by the East Australian Current [12] carry a disproportionate share of the individuals.
What this finding leaves unanswered is how many calories any of it delivers. No stomach was weighed, so the energetic correction to the capital-breeder budget stays unmeasured. What the study revises is where whales are and how long they linger, not the accounting of their fat reserves. There is also no non-feeding denominator, so "common" here is a presence claim rather than a rate.
The management read survives that weakness, with a condition attached. Presence data can justify protection where feeding is documented, and it cannot justify leaving anywhere else out. The seasonal speed limits and gear changes the authors propose [11] are defensible at Eden or Bermagui on this evidence, about 90 events at that one stretch of coast [18], while the Phillip Island records argue against drawing a tight line around the known sites. Since surveying thousands of kilometres of coast every season is not affordable [13], opportunistic footage is the evidence base regulators actually have.
Ranked by verification strength, evidence, and original report placement.
A new study published in Marine Mammal Science investigated humpback whale feeding behaviour on the east coast of Australia using citizen science and social media information, and found mid-latitude feeding to be a common and widespread strategy among humpback whales.
The search produced 199 distinct feeding observations, representing an estimated 1,417 individual humpback whales documented feeding in Australian waters between 2013 and 2024.
Researchers searched Facebook and Instagram for public posts by whale-watching companies, drone pilots and whale enthusiasts, including only posts with clear visual evidence of specific feeding behaviours: whales opening their mouths, lunging with expanded ventral pleats, bubble-net feeding, or surfacing through dense schools of baitfish.
The NSW south coast, particularly near Eden, Merimbula and Bermagui, accounted for 45% of all verified feeding events; eastern and southern Tasmania accounted for 28% and the Victorian coast 12% of records.
The study highlighted feeding regions previously underrepresented in records, including coastal areas near Phillip Island in Victoria and across northern Tasmania.
Observations increased between August and October each year, coinciding with the southbound migration of humpback whales, a time when their energy reserves are at their lowest.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · August 27, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
invest
The remedy New Mexico won at trial is the one Meta's $18 billion settlement does not contain1 distinct publisher
invest
Meta's Pay Structure, Not Its Policy Page, Is What the States Put on the Stand1 distinct publisher
science
The 35C problem is a rostering problem: card data says afternoon trade does not come back until dark1 distinct publisher
invest
Meta's $1.4 trillion number is not the expensive part of the four-state trial1 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.
Peer-reviewed and quantified, but single-source and self-reported with no bias correction shown
The findings come from a named, DOI-identified paper in Marine Mammal Science and are reported with specific counts, regional shares, seasonality and group sizes, all of which hang together arithmetically. Against that, the cluster contains exactly one publisher carrying the authors' own summary; the data are opportunistic and presence-only, and the supplied text gives no observer-effort correction, de-duplication method or uncertainty range for the 1,417-individual estimate, which is the weakest link in inferring that mid-latitude feeding is 'common and widespread' rather than commonly photographed.
Method published and used at scale; none of the proposed management changes shown adopted
Adoption is real on the research side -- the citizen-science approach has been applied across a decade of records and cleared peer review -- but zero on the side that would matter operationally. The supplied source shows no regulator, fisheries authority, port or shipping body implementing seasonal or mobile management areas, speed reductions or gear rules; those remain author recommendations. Scored low to reflect method uptake only.
Modestly overstated: strong framing rests on opportunistic, effort-uncorrected records
The core observations -- verified feeding events, named hotspots, an August-October peak, 16 super-groups -- are solidly documented, and the headline conclusion is plausible. The overstatement is in the inferential leap: presence-only social-media records cannot by themselves establish that mid-latitude feeding is 'common and widespread' population-wide, and the source volunteers no effort correction while also acknowledging that historical whaling and stranding records pointed the other way. Combined with the absence of any adopted management change, the claim runs somewhat ahead of the demonstrated evidence base rather than far ahead.
Author-written advocacy piece, syndicated: visible promotional and policy interest, no commercial stake disclosed
The only source is the researchers' own first-person account of their paper, published through The Conversation under Creative Commons and republished by phys.org, which has a syndication interest in free science copy. The authors have a clear professional incentive to amplify their study and an explicit policy objective -- persuading managers to adopt seasonal areas, speed limits and gear rules -- and the piece closes by arguing for more citizen-science observation, which sustains their own data pipeline. No financial conflict, funder or commercial sponsor is disclosed in the supplied text, so this is ordinary academic and advocacy incentive rather than evidence of a hidden stake.
Moderate: internally consistent and DOI-anchored, but one publisher and one interested voice
Confidence is capped by structure rather than content quality. Every claim traces to a single article authored by the study team, so corroboration, methodological critique and stakeholder response are all absent; the arithmetic derivations are reliable only because the underlying totals are taken on trust. The specificity of places, dates, percentages and the journal DOI supports a mid-range rather than low score.