The researchers developed the method around a specific problem in aerial photogrammetry: a body seen from directly above may look shorter when it bends or dips beneath the surface. Conventional drone measurements generally screen out visibly curved animals because their reliability cannot be established with a single vertical view. Such exclusions can be especially costly for low-density, difficult-to-approach populations, and some marine-mammal studies have retained only 38% to 94% of their images after visual quality control. The new approach combines one drone filming vertically from above with a second filming from an oblique angle of about 45 degrees. By tracing the animals’ dorsal body lines in both views and calculating the intersection of their viewing rays in three-dimensional space, the researchers reconstructed the path of the back rather than simply measuring its projection onto the overhead image.

Technical accuracy depended on the geometry and calibration of the two-camera system. The DJI Mavic 3E drones flew at an altitude of 40 m, using 7× optical zoom with a 35-mm-equivalent focal length of 162 mm and producing an empirical ground sampling distance of 2.4 mm per pixel. Their real-time positions were corrected through a fixed GNSS reference station, while a vessel duckboard or buoy of known length was filmed at the water surface before and after each observation to establish scale. Videos and flight logs were synchronized by tapping chopsticks in front of both cameras before takeoff and after landing. Lens distortion was corrected for every frame, and researchers annotated the dorsal line, the boundary between exposed and submerged tissue, and, where visible, the nostrils and peduncle. For body sections underwater, the apparent location was projected onto the water surface and its ray was refracted using Snell’s law; body roll was not modeled because no apparent roll was observed.

Accuracy was assessed by comparing drone estimates with tape-measure measurements taken on 25 apparently healthy dugongs after they were temporarily removed from the water for medical examinations. The animals were marked at 10 evenly spaced points along the dorsal line, and multi-angle video was used to construct three-dimensional reference models, scaled from accurately measured fluke width. A separate validation of the modeling procedure using a stationary crate produced a mean absolute error of 0.59 ± 0.43 cm over six tests, although some peduncle-to-fluke measurements differed by several centimeters between deck measurements and the dugong models. For each animal, three surfacing events were sampled, with a “best” frame chosen when the body was most parallel to the surface and both endpoints were most visible. Frames one and two seconds on either side of that moment were examined to test performance under less favorable posture and visibility.

For the clearest frames, the two methods were close: the dual-drone estimate had an absolute error of 3.3 ± 2.8 cm, or 1.3% ± 1.1%, while the overhead-only estimate had an error of 5.0 ± 4.2 cm, or 1.5% ± 1.2%. The gap widened markedly in the less favorable frames, when average errors reached 4.8% ± 3.8% for the overhead method and 2.0% ± 1.6% for the reconstructed method. Neither the position of the oblique camera relative to the animal’s body axis nor the degree of curvature produced a systematic rise in the magnitude of error in the three-dimensional estimates. Repeated measurements across three breaths and best and non-best frames yielded a mean interquartile range of 4.4 ± 1.7 cm, or 1.7% ± 0.7% of body length. The method therefore produced measurements with accuracy and precision comparable to some earlier measurements of stationary animals and objects, despite free-swimming dugongs showing varied postures and disturbing the water surface.

The investigators also tested whether redefining the body-length endpoints could reduce the effects of curvature. Candidate measures ran from the snout or nostrils to either the fluke notch or peduncle. The overhead method showed a significant overall effect of endpoint choice, while the dual-drone method did not. In curved images, overhead measurements taken from the nostrils were consistently too short, with mean errors of −17.9 ± 6.6 cm for nostril-to-fluke-notch length and −13.2 ± 6.8 cm for nostril-to-peduncle length. By contrast, the dual-drone estimates showed no systematic directional change when endpoints were altered. The researchers did not recommend the peduncle as an endpoint because ripples often obscured its position, but the results did not support replacing either the snout or fluke notch with it.

Water movement remained a separate source of uncertainty. When fluke tips were clearly visible in extremely calm conditions, apparent fluke width still varied by approximately 1 ± 1 cm, equivalent to 0.4% ± 0.4% of body length, as currents and the animal’s own swimming disturbed the surface. The variation tracked apparent pixel length rather than drone-altitude readings, indicating that water distortion rather than vertical instability was the likely cause. Across all ground-truthed animals, the absolute error in fluke width averaged 2.3 ± 1.5 cm, or 0.86% ± 0.60% of body length. Errors decreased significantly as visibility improved: root-mean-square errors were 3.3 cm for poor frames, 3.0 cm for fair frames, 2.56 cm for good frames and 1.9 cm for excellent frames. Altitude fluctuations were small relative to the 40-m flight height, and a mixed-effects analysis found that camera yaw had no significant effect on body-length error.

The field survey showed why retaining curved images could change the size of usable ecological datasets. Among 188 free-swimming dugongs, the mean body curve ratio—a comparison between the overhead dorsal-line length and the reconstructed three-dimensional length—was 0.926 ± 0.049, with a range of 0.751 to 0.992. The researchers defined a threshold of 0.919 at which the predicted magnitude of overhead error exceeded the lower limit of the dual-drone method’s 95% confidence interval, which extended from −3.7% to 3.2% of body length. Only 31.9% of the sampled animals fell below that threshold, indicating that about 68% were more strongly curved than the point at which the conventional method was expected to perform less reliably. The finding does not mean the single-drone estimate was unusable for every such animal, but it provides an objective basis for identifying frames likely to benefit from three-dimensional reconstruction rather than relying only on visual screening.

The method also comes with clear constraints. Both body endpoints must remain visible in the oblique and vertical views, and turbidity, glare, sea state, water depth, surfacing behavior and surface disturbance can make this difficult. Turning animals sometimes prevented the snout from appearing in the oblique footage. Operating the second drone required an experienced pilot to maintain horizontal movement and yaw beside the animal, and the two complete drone systems, together with the GNSS receiver, cost approximately US$8,800 at 2025 prices. The flights were conducted at 40 m under protocols considered to have minimal impact, but 15 of the 188 randomly sampled dugongs surfaced without breathing in some events, and one animal made a sudden dive; the study did not establish whether these behaviors were responses to the drones. The reconstructed method may nevertheless be useful for other air-breathing marine megafauna, particularly species without dorsal fins or other prominent midline landmarks, provided suitable visibility, calibration and flight coordination can be maintained.

Subject of Research: Three-dimensional drone photogrammetry for measuring body length and posture in free-swimming dugongs

Article Title: Estimating Three‐Dimensional Posture to Correct Drone‐Based Measurements of Free‐Swimming Marine Mammals: An Application to Dugongs in Shallow Waters

Article References: Yamato, C., Sneath, H. L., & Lanyon, J. M. (2026). Estimating Three‐Dimensional Posture to Correct Drone‐Based Measurements of Free‐Swimming Marine Mammals: An Application to Dugongs in Shallow Waters. Ecology and Evolution, 16(10), Article e74446. https://doi.org/10.1002/ece3.74446

Image Credits: AI Generated

DOI: 10.1002/ece3.74446

Keywords: dugongs, drone photogrammetry, three-dimensional reconstruction, marine mammals, body curvature, wildlife monitoring, Moreton Bay, morphometrics, animal posture, biomass, Estimating, Three

The study documented 205 cetacean stranding events involving 281 animals during 2023. This was more than 15 times the average of approximately 18.5 stranded cetaceans recorded annually in Chile between 2014 and 2022, when two unusually large events involving 373 sei whales in 2015 and 124 pilot whales in 2016 were excluded. The animals represented at least 19 species, with long-beaked common dolphins the most frequently recorded at 56, followed by Burmeister’s porpoises at 40, dusky dolphins at 33, Chilean dolphins at 18, bottlenose dolphins at 17 and sei whales at 16. Most events involved one animal: 171 were single strandings, while 34 involved between two and nine animals.

Of the 281 stranded animals, 271 were found dead and 10 were alive when reported. Eight of the live animals subsequently died, while two were successfully released from entanglement. Most carcasses, 248, were described as being in normal body condition, whereas 33 showed clear emaciation. Sixty-six were in an advanced state of decay and could not be classified to species level. Cases occurred in 14 of Chile’s 15 administrative coastal regions, with Arica, the northernmost region, accounting for 152 cases. The study’s spatial analysis identified two high-risk clusters in northern Chile, where cetacean strandings and HPAIV-positive wild birds overlapped in time and geography, but neither cluster was statistically significant at p greater than 0.05.

Investigators sampled 27 stranded cetaceans from 13 species at 26 locations across 11 administrative regions. Selection was constrained by carcass freshness, location, logistical capacity and, in some cases, concerns about human health risk; it was not a random or population-representative sample. Swabs were collected from the spiracle, genital area and/or rectum, pooled for each animal in viral transport medium and initially tested for generic influenza A virus with a commercial diagnostic kit. Investigated results were then examined through real-time reverse-transcription PCR assays designed to detect H5 viruses of clade 2.3.4.4, determine their pathogenic classification and identify the N1 subtype. The sample represented less than one-tenth of all 281 animals recorded during the year.

Four of the 27 sampled cetaceans, or 18.5%, were positive for HPAIV H5N1 in swabs. They were two of five Burmeister’s porpoises sampled in northern Chile and two of five Chilean dolphins sampled in south-central Chile, with each positive animal coming from a different location. This amounted to a positivity rate of approximately 40% within each of those two species groups. The other swabbed animals were negative. Cycle-threshold values for generic influenza A detection ranged from 17.9 to 25, while the clade-specific, pathotyping and N1-subtyping assays produced values in the ranges of 18–27, 21–24 and 17–30, respectively. These findings represented the first reported detections of HPAIV H5N1 in Burmeister’s porpoises and Chilean dolphins.

The fifth infected cetacean was a long-beaked common dolphin initially found alive but which died shortly after stranding. Its carcass was recovered two days later because of logistical limitations and was moderately decomposed. Gross examination showed trauma associated with the stranding, including subcutaneous hemorrhages, as well as lung congestion and alveolar edema. No parasites or other major abnormalities were found in the thorax or abdomen, and no viral RNA was detected in the available swabs. After H5N1 was detected in tissue, investigators examined formalin-fixed, paraffin-embedded sections by histology and immunohistochemistry and used PCR and sequencing to examine archived tissue. The animal’s exact cause of death was not established.

PCR detected HPAIV H5N1 in two of eight tissue samples from the common dolphin: one lung section, with a cycle threshold of 37, and a brain section, with a value of 29. Sequencing of an approximately 150-base-pair amplicon identified the H5 subtype of clade 2.3.4.4b. The resulting partial consensus sequence showed 96.05% identity with avian sequences from Chile, Peru, Brazil and Texas, and 95.39% identity with avian sequences and South American sea lion sequences from Chile and Brazil. The researchers emphasized that this limited sequence was obtained to confirm viral presence and was not a comprehensive genetic analysis of the virus circulating in cetaceans. The available evidence therefore did not resolve how the virus reached individual animals or whether cetaceans played a broader role in transmission.

Histology revealed mild, multifocal leptomeningitis in the dolphin’s brain, with lymphoplasmacytic and histiocytic inflammation, occasional neutrophils and leukocytosis. The pattern was considered compatible with a leptomeningoencephalitis, but common alternative infectious causes—including cetacean morbillivirus, Toxoplasma gondii and Brucella species—were excluded by molecular testing. The observed infiltrates were therefore judged more likely to be related to HPAIV H5N1 in conjunction with the positive tissue PCR results, but they were nonspecific. The authors also noted the absence of perivascular cuffing and neuronal necrosis in the gray matter of the cerebral hemispheres, two pathological features commonly associated with avian influenza infection of the central nervous system. More severe lesions might also have been present in regions that were not sampled.

Immunohistochemistry found no influenza A viral antigen in the examined tissues. The investigators said this could reflect the sensitivity limits of the method in formalin-fixed material, a viral load below its detection threshold, severe lesions in unsampled regions or differences between molecular and antigen-based testing. The brain inflammation might be compatible with neurological impairment that contributed to the dolphin’s live stranding, and video evidence suggested possible neurological involvement, but the pathological findings were too mild to establish the cause of death. The authors therefore retained an inconclusive interpretation for both the stranding and death, emphasizing that the dolphin also had trauma and that histopathology could not directly connect H5N1 infection with neurological signs.

Spatial analysis found two nonsignificant high-risk clusters in northern Chile where cetacean strandings overlapped geographically and temporally with HPAIV-positive wild birds. Because both clusters had p values greater than 0.05, the analysis did not demonstrate a causal relationship or a statistically robust concentration of mortality. The correspondence nevertheless supports the possibility that the virus circulated in coastal marine ecosystems while cetaceans were stranding at an elevated rate. Wild birds are considered a plausible infection source because large numbers of seabirds tested positive during the outbreak, while surface feeding and shared coastal habitats could provide opportunities for environmental exposure. The available evidence did not directly establish transmission from birds to any of the cetaceans. HPAIV H5N1 had previously been reported in a bottlenose dolphin in Florida, a long-beaked common dolphin in Peru and a harbor porpoise in Sweden, but the Chilean findings extended detection to two additional species and provided the first evidence in this outbreak across several cetacean taxa.

The investigation had important constraints. Restrictions on access to stranding sites, concerns about the virus’s zoonotic potential, shortages of personnel and the scale of the seabird mortality emergency sharply curtailed sampling. In addition, necropsies were prohibited during the outbreak except when performed through public services, leaving only one cetacean available for detailed pathological investigation. The 18.5% positivity rate among the constrained sample cannot represent infection prevalence among all stranded cetaceans, nor does it show that H5N1 was responsible for the overall increase in strandings. Environmental contamination, unusual mortality events associated with harmful algal blooms, vessel collisions, fishing or aquaculture gear and other known causes of cetacean mortality also remained relevant. The study was therefore an epidemiological field report rather than a controlled test of viral pathogenicity, and its findings document an emerging cross-species exposure while leaving the causes of Chile’s unusual 2023 stranding increase unresolved.

Subject of Research: Epidemiological detection of HPAIV H5N1 in stranded cetaceans during Chile’s 2023 outbreak

Article Title: Highly pathogenic avian influenza virus H5N1 infection in stranded porpoises and dolphins in Chile

Article References: Ulloa, M., Neira, V., Sierra, E., Colom-Rivero, A., Muñoz, G., Johow, M., Lillo, F., Rojas, D., Ortega, R., Alegría-Morán, R., & Fernández, A. (2026). Highly pathogenic avian influenza virus H5N1 infection in stranded porpoises and dolphins in Chile. Frontiers in Veterinary Science, 13, 1969182. https://doi.org/10.3389/fvets.2026.1969182

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1969182

Keywords: HPAIV H5N1, Chile, cetacean strandings, Burmeister’s porpoises, Chilean dolphins, common dolphins, avian influenza, wildlife surveillance, marine mammals, viral pathogenesis, Highly, pathogenic

The research, published in the journal Metabarcoding and Metagenomics, focused on the short-finned pilot whale (Globicephala macrorhynchus), a highly social species with a global distribution and a well-documented propensity for mass strandings. Respiratory tract infections rank among the leading causes of death in cetaceans, making the microbial community of the airways a potentially powerful biomarker of respiratory health. Yet most existing data on cetacean-associated microbes come from captive, stranded, sick or injured individuals, which cannot be considered representative of healthy free-ranging populations. The new study set out to fill that gap by sampling wild whales in their natural environment.

Between September and October 2018, researchers conducted at-sea monitoring campaigns in the southern waters of Madeira Island, Portugal. Using a PERFORMAgene PG-100 swab collection kit attached to a five-metre extendable aluminium pole, the team positioned the sampling device roughly 40 to 50 centimetres above the blowhole of surfacing whales to capture droplets from the exhaled plume. The animals were approached slowly from an eight-metre rigid inflatable boat under permits from the Instituto de Florestas e Conservação da Natureza IP-RAM. In total, twelve blow samples were collected across six sampling events, nine from individual animals and three from pools of individuals within the same group. Photographic identification against Madeira’s catalogue of the species confirmed that the nine individual samples corresponded to seven distinct animals, two of which were sampled on separate occasions.

A notable methodological first accompanied the sampling: the PERFORMAgene kit, previously used for livestock and pets, had never been reported in the literature for collecting blow from free-ranging cetaceans. The kit offers practical advantages for fieldwork, stabilising DNA samples for up to a year at room temperature, which provides safe flexibility during transportation and storage. Although DNA concentrations extracted from the swabs were low, the method proved suitable for downstream metabarcoding analysis. The researchers also observed no evident changes in the whales’ swimming patterns or behaviour during sampling, suggesting the procedure causes minimal distress, though they recommend future studies formally monitor behavioural responses alongside collection.

The core of the study was a rigorous comparison of two primer sets targeting different hypervariable regions of the 16S rRNA gene, the standard molecular marker for characterising bacterial and archaeal communities. The V3-V4 region, amplified with primers 341F and 806R, had been used in previous cetacean blow studies, while the V4-V5 region, amplified with primers 515F-Y/926R, is currently recommended by the Earth Microbiome Project for targeting marine microbes. By applying both primer sets to the same twelve samples, using identical DNA extraction, library preparation, sequencing on an Illumina MiSeq platform and bioinformatic processing through the DADA2 pipeline, the researchers eliminated technical biases that typically confound such comparisons.

Sequencing yielded a total of 685,692 reads for the V3-V4 dataset and 862,763 reads for the V4-V5 dataset. After quality filtering, the final output comprised 415,340 sequences assigned to 2,764 Amplicon Sequence Variants (ASVs) in the V3-V4 dataset and 417,733 sequences assigned to 3,665 ASVs in the V4-V5 dataset. ASVs resolve sequences to the level of a single nucleotide difference, offering finer taxonomic resolution than traditional operational taxonomic units. Taxonomic classification was performed with a Naïve Bayes classifier against the curated SILVA database, and undesirable lineages such as chloroplast, eukaryote and mitochondrial sequences were excluded from further analysis.

The comparison revealed striking differences between the two primer sets. Alpha diversity metrics varied significantly: the V4-V5 dataset captured roughly 32 percent more bacterial ASVs, with higher values for the Observed and Chao1 richness measures, while the V3-V4 dataset produced higher Shannon and Inverse Simpson diversity values. A statistical test using the Wilcoxon Signed-Rank procedure confirmed significant differences between primer sets for most measures. Beta diversity analysis, based on Bray-Curtis dissimilarity and visualised through non-metric multidimensional scaling, showed that samples clustered according to the primer set used. A PERMANOVA test found that the primer choice explained approximately 25 percent of the variation in prokaryotic community distribution, a substantial technical effect that researchers designing future cetacean microbiome studies cannot afford to ignore.

Despite these differences, all blow samples shared a core microbiota composed of four dominant phyla: Actinobacteria, Bacteroidetes, Firmicutes and Proteobacteria. This aligns with previous findings in bottlenose dolphins, humpback whales and other cetacean species, where Proteobacteria is consistently highly abundant. At the genus level, the V3-V4 dataset was dominated by Pseudomonas, while the V4-V5 dataset was dominated by Cutibacterium, a genus typical of nasal microbiota. The oceanic bacterium Prochlorococcus, one of the most abundant bacteria in the sea, was detected in both datasets and is probably a seawater-associated organism within the blow. The researchers noted that control water samples were not taken in this study but consider them extremely important for future work to disentangle the influence of ambient seawater on the detected microbial composition.

Perhaps the most intriguing finding concerns sociality. In the V4-V5 dataset, individuals that were travelling together when sampled, such as the pairs Gma_2 and Gma_3 or Gma_4 and Gma_5, clustered together in the community analyses, a pattern not observed with the V3-V4 primers. This supports the hypothesis that social behaviour shapes the respiratory microbiome in cetaceans. Short-finned pilot whales live in matrilineal groups with long-lasting relationships, and behaviours such as surfacing and breathing near each other or feeding cooperatively could facilitate microbial transfer between individuals. Previous work on whale species has detected a positive correlation between sociality and microbial diversity in blow, and the new results strengthen the idea that social complexity influences microbiome composition in the respiratory tract of these marine mammals.

The study also carries implications for disease surveillance. Four of the most abundant genera detected in both datasets, Pseudomonas, Flavobacterium, Rhodococcus and Sphingomonas, are potential pathogenic genera commonly related to pulmonary infections, although their presence does not necessarily indicate disease since not all members of these taxa are pathogenic. The V4-V5 primers proved particularly valuable for detecting the rare microbial biosphere, the low-abundance taxa that could include organisms with pathogenic potential, and for capturing the core microbiome more consistently. The authors conclude that primer selection should depend on the analytical goal: V3-V4 performs better for capturing diversity present at higher relative abundance, while V4-V5 is preferable for identifying unique ASVs and rare taxa. A combination of both primer sets may ultimately provide the most robust characterisation of the blow microbiome. With further optimisation of sampling protocols and integration with complementary health measures such as photogrammetry-based body condition assessment, whale breath could become a routine biomarker for monitoring the physiological state of airways in free-ranging cetacean populations worldwide.

Subject of Research: Respiratory microbiome characterisation of free-ranging short-finned pilot whales using blow samples and 16S rRNA metabarcoding

Article Title: Hidden in the blow – a matrix to characterise cetaceans’ respiratory microbiome: short-finned pilot whale as case study

Article References: Santos, B., Afonso, L., Alves, F., Dinis, A., Ferreira, R., Correia, A. M., Valente, R., Gil, Á., Castro, L. F. C., Sousa-Pinto, I., Rosso, M., Centelleghe, C., Mazzariol, S., Magalhães, C., & Tomasino, M. P. (2024). Hidden in the blow – a matrix to characterise cetaceans’ respiratory microbiome: short-finned pilot whale as case study. Metabarcoding and Metagenomics, 8, Article e121060. https://doi.org/10.3897/mbmg.8.121060

Image Credits: AI Generated

DOI: 10.3897/mbmg.8.121060

Keywords: cetaceans, short-finned pilot whale, blow microbiome, respiratory health, 16S rRNA metabarcoding, Madeira Island, exhaled breath condensate, marine mammals, microbiome, non-invasive sampling, rare biosphere, sociality

The Indo-Pacific humpback dolphin, Sousa chinensis, ranges along the southeastern coast of China, through Southeast Asia, and along the coastal margins of the Bay of Bengal to the coast of Odisha in India. Yet some of its most important populations, including those in Xiamen Bay, the eastern Taiwan Strait, and the waters around Hong Kong, face intense anthropogenic pressure. Xiamen Bay is a particularly striking case: the region combines large-scale cargo transportation, extensive fishing operations, and routine channel dredging, all overlaid on a dynamic estuarine environment where freshwater from the Jiulongjiang River system meets strong tidal currents from the Taiwan Strait. Ferry traffic along the route to Gulangyu Island alone averages roughly five ships every twenty minutes, making the bay a natural laboratory for studying how coastal cetaceans cope with chronic disturbance.

Between 2017 and 2019, researchers completed 181 survey days across Xiamen Bay, Zhangzhou, Quanzhou, and Kinmen, accumulating 52,092 minutes of effective observation effort. Working from a survey vessel that followed predetermined transect lines, pairs of observers scanned the water with the naked eye or with 10×50 binoculars, rotating roles every thirty minutes to minimize fatigue. When dolphins were encountered, the team photographed individuals with a Canon EOS 1Dx Mark II and recorded behavior on video for later laboratory analysis. Dive duration was defined precisely as the time elapsed from the end of one surfacing event to the start of the next, measured with a handheld stopwatch and supplemented by video review. In total, the dolphins were tracked for 5,319 minutes, yielding 4,408 recorded dives, of which 893 occurred during traveling and 1,843 during foraging.

The headline finding from the diving data is that humpback dolphins are, by cetacean standards, shallow and short divers. Across all recorded dives, the mean duration was 20.41 seconds, and the overwhelming majority of dives, roughly 75 percent of both individual and synchronous diving events, lasted less than twenty seconds. Individual dives averaged 20.36 seconds while synchronous breathing events averaged 20.56 seconds, a statistically indistinguishable difference. Dive durations during traveling averaged 24.96 seconds compared with 21.31 seconds during foraging, but this difference also failed to reach statistical significance, suggesting that within this population the basic rhythm of surfacing remains remarkably stable regardless of what the animals are doing beneath the surface.

Age, however, told a more interesting story. Because Indo-Pacific humpback dolphins change color dramatically as they mature, from dark gray or nearly black calves to whitish adults with few remaining dark markings, the researchers were able to classify individuals into juveniles, subadults, and adults based on body size and pigmentation. Juveniles averaged just 15.43 seconds per dive, while subadults averaged 20.54 seconds and adults 20.64 seconds. When diving was analyzed by both age and behavior, juveniles were significantly shorter divers than subadults during travel, a difference that was statistically robust. The pattern fits a well-established principle in marine mammal biology: diving capacity depends on the maturation of oxygen stores in blood and muscle, which develop gradually with age and body size. Bottlenose dolphins, for example, reach near-adult levels of blood oxygen reserves at around three years of age, and humpback dolphins, which wean at roughly four years, may attain near-adult oxygen capacity by the subadult stage, consistent with the near-identical dive durations recorded for Xiamen’s subadults and adults.

The study also uncovered a curious interaction between age, behavior, and dive time. Among subadults, foraging dives were significantly shorter than traveling dives, averaging 19.89 seconds against 24.48 seconds. The authors speculate that this may reflect an age-specific trade-off between attention and foraging efficiency, with younger dolphins navigating the demands of hunting while still refining their skills. Juvenile dolphins in Xiamen showed shorter mean foraging dives than either subadults or adults, echoing observations from Shark Bay, Australia, where bottlenose dolphin calves begin foraging in shallow water within months of birth and only graduate to deeper, more demanding dives as their physiological capacity matures. Intriguingly, the mean dive duration of Xiamen’s dolphins, 20.41 seconds, closely matches the 20.89 seconds recorded for the Hong Kong population through land-based observations, hinting at common underlying mechanisms that the authors say warrant further investigation of environmental covariates.

The most consequential results concern vessels. Over the study period, researchers documented 197 instances of dolphins responding to boats, and statistical analysis revealed significant associations between dolphin responses and both vessel type and vessel distance. Speedboats proved the most alarming, eliciting negative responses, defined as active avoidance or fleeing, in 68.18 percent of encounters, typically involving abrupt directional changes or deep-diving behavior. Positive responses, in which dolphins actively approached or chased boats, were rare across all vessel types. Distance mattered just as much: negative responses occurred only when vessels came within 200 meters, and avoidance behavior rose sharply inside 100 meters, while positive responses were recorded only within 200 meters. Notably, unlike dolphins in Hong Kong waters, which frequently associate with double-trawl fishing vessels, the Xiamen animals showed no attraction to fishing boats, possibly because fishery resources in the bay are limited and the local fleet consists exclusively of small-scale gillnetters that hold little appeal.

Perhaps the most worrying pattern is what the dolphins did not do. Negative reactions accounted for only 22 percent of all documented responses, and even when vessels closed to within 100 meters, only 27 percent of individuals showed clear avoidance. The authors caution strongly against reading this apparent tolerance as adaptation. Neutral responses may mask hidden costs, including interference with acoustic communication, resting, and other vital behaviors that underpin individual health, reproductive success, and long-term population viability. Comparable studies of bottlenose dolphins have shown substantial reductions in feeding and resting when dolphin-watching boats approached within 50 meters, and dolphins that fail to flee a fast-approaching speedboat face elevated risks of collision and injury. In other words, a dolphin that seems unbothered may simply be unaware of the danger accumulating around it.

From these findings flow a set of practical conservation recommendations. The study’s authors urge that vessels maintain a minimum buffer zone of 100 meters from dolphins, comply with a strict speed limit of 10 knots in dolphin habitats, and adopt proactive avoidance practices when transiting areas where the animals are present. Because Xiamen’s humpback dolphins move across waters administered by multiple jurisdictions, including Xiamen, Zhangzhou, Quanzhou, and Kinmen, the researchers stress that coordinated regional management and stronger enforcement of existing protective regulations will be essential. The recommendations are deliberately concrete, translating behavioral thresholds measured in the field directly into operational rules for the shipping, ferry, and fishing industries that share the bay.

The authors are candid about the study’s limits. Behavioral recording of such a versatile species inevitably carries some observational error, and the research did not quantify how vessel noise or speed specifically influenced dolphin responses, leaving open questions about the mechanisms behind the avoidance patterns. They also emphasize that the long-term consequences of neutral responses remain uncertain and deserve focused study, particularly regarding acoustic masking and disruption of resting. Even so, the work represents a significant advance for a species about which knowledge of diving behavior and vessel sensitivity has, outside Hong Kong, remained scarce. For a population clinging to the edges of one of the world’s most industrialized coastlines, the message of the study is clear: the dolphins’ short, frequent surfacing makes them perpetually visible at the surface, and it falls to the operators of the bay’s thousand daily transits to give them the space and slow speeds they need to survive.

Subject of Research: Diving behavior and vessel-traffic responses of Indo-Pacific humpback dolphins in Xiamen Bay

Article Title: Diving pattern and behavioral response to vessel traffic of Indo-Pacific humpback dolphins ( Sousa chinensis ) in Xiamen, China: implications for conservation

Article References: Lu, Y., Zhuo, H., Li, H., Xu, X., & Yang, G. (2025). Diving pattern and behavioral response to vessel traffic of Indo-Pacific humpback dolphins (Sousa chinensis) in Xiamen, China: implications for conservation. Nature Conservation, 60, 125-138. https://doi.org/10.3897/natureconservation.60.162175

Image Credits: AI Generated

DOI: 10.3897/natureconservation.60.162175

Keywords: Indo-Pacific humpback dolphin, Sousa chinensis, Xiamen Bay, diving behavior, vessel traffic, speedboats, age classes, foraging, conservation, marine mammals, China, boat disturbance

The observations come from a nine-year study of the Pacific Coast Feeding Group (PCFG) of gray whales, a small subgroup of roughly 213 individuals within the Eastern North Pacific population of about 17,750. Unlike the rest of the population, which migrates between Arctic feeding grounds and Baja California, PCFG whales stay put, foraging in shallow coastal waters less than 20 meters deep between Northern California and Southern British Columbia. Their accessibility makes them ideal subjects for drone-based behavioral research, and their diverse repertoire of foraging tactics makes them ideal for studying how behaviors might pass from one generation to the next.

PCFG gray whales employ an unusually varied set of feeding strategies that target prey throughout the water column. Some whales feed at the surface, snapping their jaws while frequently turning. Others swim forward on their side, upside down, or belly-up below the surface. Still others perform stationary side-swims or dramatic headstands, pushing their heads into the seafloor through powerful fluke beats and pectoral fin action. Individual whales do not use these tactics uniformly; tactic choice is linked to body length, a proxy for age, and to the type of seafloor habitat. Because many calves born to PCFG mothers later join the feeding group themselves, researchers suspected that calves might be learning these tactics, and where to use them, directly from their mothers during the roughly seven months between birth and weaning.

Testing that idea required seeing what happens underwater, which is where drones transformed the field. Between May and October of 2016 through 2025, survey teams from a small rigid hull inflatable boat launched five different consumer drones to film whales from above. The boat kept at least 75 meters from the animals to avoid disturbing them, and pilots recorded roughly 15 minutes of continuous video per flight, tracking whales both at the surface and while submerged. A single experienced analyst then annotated every clip of a mother-calf pair in open-source behavioral coding software, using an ethogram of ten distinct behaviors and reviewing each clip at least twice per individual.

The resulting dataset included 15 mother-calf pairs from 13 unique mothers. In nine of those pairs, the mother was observed performing a foraging tactic with her calf nearby, and those pairs were filmed for significantly longer periods, with a median observation time of nearly 14 minutes, than the six pairs in which no foraging was seen, which were observed for a median of just over three minutes and spent their time at the surface or nursing. Crucially, of the nine calves whose mothers were seen foraging, six calves began foraging themselves. Every one of those six calves started its foraging tactic only after its mother had begun hers, on average about 13 seconds later, though the delay ranged from less than a second to nearly 49 seconds.

When the researchers compared the specific tactics calves used across 21 foraging dives, a striking pattern emerged. In nine dives, or 42.9 percent, the calf performed exactly the same tactic as its mother. In three other dives, the calf used a tactic its mother had performed earlier in the same observation period, hinting at delayed imitation of the kind documented in captive bottlenose dolphins. In four more dives, the calf used a tactic belonging to the same functional group as its mother’s, meaning the behaviors share similarities in form. And in five dives, the calf performed a body roll while its mother was headstanding, a behavior the researchers interpret as a failed but telling attempt to imitate a physically demanding maneuver.

The body rolls are the study’s most compelling evidence for imitation. Because the calves showed no visible feeding during the rolls and were nowhere near where their mothers’ heads were digging into the seafloor, the behavior cannot be explained as goal emulation or simple enhancement, in which an observer learns only what to feed on or where to feed. Instead, the researchers argue, the rolls represent an effort to reproduce the opaque body movements of the headstand itself. Headstanding is a steep dive combined with a roll, often held upside down and pitched for long durations, with a mean of 55 seconds, and it is used primarily by long, mature whales thought to be physically capable of the feat. The calves did not roll continuously; they paused mid-roll in the upside-down position, as though attempting to match the posture of the headstand but lacking the strength or coordination to complete it. Because the rolls always followed a mother’s headstand and have rarely been seen in any other context, they appear to be imperfect imitation attempts by young whales reaching beyond their physical abilities.

The study also found evidence that lessons learned at a mother’s side persist. Five of the 15 calves were later observed foraging independently in subsequent years. Four, seen as one- or two-year-olds, used surface feeding or forward-swimming tactics, and both individuals observed forward-swimming had watched their mothers forward-swim as calves, with one having performed the tactic alongside her. The fifth calf, which had watched its mother headstand and even attempted headstanding as a calf, was observed headstanding at six years old, consistent with a known developmental shift in PCFG whales from forward-swimming tactics toward headstanding as they mature. The researchers also noted two calves performing bubble blasts, underwater releases of air that rise to the surface in a circle. Although the mothers were not seen blasting at the time, both had been observed bubble blasting in previous years, and the calves’ behavior may reflect local enhancement, learning from experience in which contexts the buoyancy-regulating behavior is useful.

What makes these findings scientifically significant is how they were obtained. Most evidence of social learning and culture in cetaceans comes from diffusion analyses, which track how a behavior spreads through a population over time, such as the vertical transmission of sponging in Indo-Pacific bottlenose dolphins or the horizontal spread of lobtail feeding in humpback whales. Those approaches can reveal that a behavior is transmitted but not the specific learning mechanism behind it. Direct observation of mother-calf pairs, made possible by drones, allowed the researchers to match the timing and form of behaviors moment by moment, distinguishing among imitation, enhancement, and simpler social influence, in which a calf merely performs a tactic because its mother happens to be doing so. The calves’ later independent use of the tactics suggests genuine social learning rather than fleeting social contagion.

If gray whales are indeed capable of motor imitation, the implications extend well beyond one species. Imitation has long been considered a window into advanced cognition, and most documented evidence outside primates has come from animals in human care, including toothed whales such as bottlenose dolphins, orcas, and belugas. This study provides the strongest evidence yet that such processes underlie social learning in baleen whales, and the first suggestion of foraging culture in gray whales. Because culture is increasingly recognized as a factor in conservation, informing how populations are delineated and managed, understanding that these whales may inherit not just genes but feeding traditions could shape how scientists protect the small PCFG subgroup. For now, the image of a calf rolling upside down in the shallows, trying and failing to stand on its head like its mother, offers a rare and vivid glimpse of learning in progress in one of the ocean’s most iconic travelers.

Subject of Research: Social learning and imitation in gray whale mother-calf pairs observed via drone footage

Article Title: Whale See, Whale Do: Drone Observations Reveal Social Learning Processes in Mom‐Calf Pairs

Article References: Bird, C. N., & Torres, L. G. (2026). Whale See, Whale Do: Drone Observations Reveal Social Learning Processes in Mom‐Calf Pairs. Ecology and Evolution, 16(10), Article e74517. https://doi.org/10.1002/ece3.74517

Image Credits: AI Generated

DOI: 10.1002/ece3.74517

Keywords: gray whales, social learning, imitation, drones, foraging behavior, animal culture, cetaceans, mother-calf pairs, behavioral ecology, Pacific Coast Feeding Group, conservation, marine mammals

The research, led by Megan R. Shaffer of the University of Washington and colleagues including Ryan P. Kelly, Andrew O. Shelton, and Kim M. Parsons, focused on Hood Canal, a fjord-like arm of Puget Sound in Washington State, USA. The team deployed a surface-stationed autonomous sampler that collected water at high temporal resolution over a 48-hour period. Their goal was to track fish communities using a metabarcoding primer set designed to amplify DNA from fish species. But as is often the case in molecular ecology, the primers did not discriminate perfectly. They also amplified DNA from marine mammals, albeit with lower efficiency.

This unintended amplification of non-target organisms, which the authors term molecular bycatch, is a widespread but underappreciated phenomenon in eDNA studies. Just as fishing nets catch species beyond those they are designed to target, molecular tools capture genetic material from organisms the researchers never intended to survey. The presence of this bycatch can obscure patterns in the primary targets, making it difficult to distinguish real ecological trends from artifacts of differential amplification.

To address this problem, the researchers developed a quantitative framework that transforms raw sequence read proportions into absolute DNA concentrations. The approach relies on two key innovations. First, they used a mock community-based bias correction to account for the fact that different species amplify with different efficiencies during the polymerase chain reaction, or PCR. Second, they employed droplet digital PCR, or ddPCR, to measure the absolute concentration of DNA from individual species, providing an independent benchmark against which the metabarcoding predictions could be validated.

The results were striking. Non-target marine mammals were substantially under-represented in the raw metabarcoding data relative to their bias-corrected proportions. In other words, the primers amplified mammalian DNA far less efficiently than fish DNA, meaning that the true abundance of marine mammals in the water was much higher than the sequence reads suggested. Without correcting for this bias, researchers would have dramatically underestimated the presence of these non-target species, potentially missing important ecological signals entirely.

Cross-validation of the quantitative approach showed that predicted DNA concentrations closely matched independent ddPCR measurements for total 12S ribosomal DNA and for two single-species concentrations: Atlantic bottlenose dolphin and Pacific herring. This agreement held in both mock communities, where the composition was known in advance, and in environmental samples from Hood Canal. The consistency of these results demonstrates that the bias correction method is robust and can produce reliable absolute concentration estimates even in complex natural settings.

Once the researchers estimated absolute species concentrations for all detected organisms, they uncovered species-specific temporal patterns that were not apparent in the observed read counts alone. Some species showed peaks and troughs in DNA concentration that had been masked or distorted by the relative nature of metabarcoding data. Because sequence reads are inherently proportional, an increase in one species can appear to cause a decrease in another, even when both are rising in absolute terms. Converting reads to concentrations eliminates this artifact and reveals the true dynamics of each species.

The implications of this work extend far beyond Hood Canal. Metabarcoding is now used worldwide to monitor biodiversity, track invasive species, assess the impacts of climate change, and inform fisheries management. In many of these applications, non-target detections are either discarded as noise or ignored altogether. The new study suggests that doing so may lead to fundamentally incorrect conclusions about community composition and ecological change. Molecular bycatch is not merely a nuisance; it carries information that, when properly quantified, can enrich our understanding of ecosystems.

The authors emphasize that correcting for species-specific amplification bias is essential when evaluating community composition from metabarcoding data, particularly in the context of non-target detections. Their quantitative approach offers a practical path forward, combining mock community calibration with ddPCR validation to produce biologically meaningful estimates of species abundance. As eDNA surveys continue to expand in scope and scale, accounting for molecular bycatch will be critical to ensuring that the ecological stories told by these powerful molecular tools reflect reality rather than the quirks of laboratory amplification.

For ecologists and managers who rely on eDNA to make decisions about conservation and resource management, the message is clear. The DNA that shows up unexpectedly in a survey is not just contamination to be filtered out. It is a signal that, once corrected for amplification bias and converted to absolute concentrations, can reveal hidden dimensions of the community and change the interpretation of the entire time series. Molecular bycatch, once seen as a problem, may prove to be one of the most valuable byproducts of the eDNA revolution.

Subject of Research: Quantitative correction of non-target DNA detections in environmental DNA metabarcoding time series

Article Title: Non-target detections change the interpretation of environmental DNA communities

Article References: Shaffer, M. R., Allan, E. A., Kelly, R. P., Adams, N. G., Moore, S. K., Brasseale, E., Brandão‐Dias, P. F. P., Scott, O. M., Shelton, A. O., & Parsons, K. M. (2026). Non-target detections change the interpretation of environmental DNA communities. Metabarcoding and Metagenomics, 10, Article e174467. https://doi.org/10.3897/mbmg.10.174467

Image Credits: AI Generated

DOI: 10.3897/mbmg.10.174467

Keywords: environmental DNA, eDNA, metabarcoding, molecular bycatch, non-target detections, droplet digital PCR, amplification bias, Hood Canal, marine mammals, fish communities, 12S rRNA, biodiversity monitoring

Fisheries interactions are widely recognised as the leading human-caused source of marine mammal mortality worldwide. Static nets, hard-wearing monofilament gill, tangle and drift nets fixed in place or left to drift, account for the majority of deaths, and have been implicated in the declines of at least 11 critically endangered small cetaceans. But mobile gear is not benign: mid-water trawls, towed fast and large, target the same schooling squid and herring that dolphins hunt, and purse seines and ring nets can trap animals at depth. Abandoned, lost or discarded gear keeps killing long after crews leave, because fishing equipment is engineered to survive the ocean and carries enormous breaking strength. Crucially, the review notes, death in these events typically takes minutes, and can take an hour or more depending on the species, a timeframe that sits at the heart of the welfare concern.

Why marine mammals end up in nets in the first place turns out to be a story about behaviour as much as geography. Harbour porpoises, one of the most frequently bycaught species, do not usually approach gill nets deliberately. Studies suggest individuals are caught accidentally when attention shifts, perhaps during predator escape or play, or when auditory masking compromises their ability to detect thin monofilament using echolocation, the primary sensory tool they rely on to map their surroundings. Echolocation is also reduced during sleep, a factor the authors flag as an unexplored influence on entanglement risk. Juvenile porpoises appear disproportionately likely to be bycaught, possibly through inexperience or curiosity, though the true reasons remain unknown. In contrast, common and bottlenose dolphins are often actively feeding around trawls, exploiting the fishery as a foraging opportunity, and may even recognise characteristic engine noises associated with gear deployment or hauling. One witnessed study found common dolphins near a boat during or within 15 minutes of net deployment, and one dolphin hauled up alive, indicating entrapment during or just before hauling.

Because so few PUE events are directly observed, necropsy, the post-mortem examination of retrieved carcasses, has become the primary window into what happens between entrapment and death. The accumulated pathological record is extensive. External findings include net impression marks, cuts and lacerations, subcutaneous haemorrhage, tooth fracture and loss. Internal injuries span muscular haemorrhage, acute peritoneal hernia, fractures of the mandible, ribs, skull, fin bones and vertebrae, ocular hyphema, cervical spinal cord compression and systemic hypoxia. Histology adds a characteristic suite of changes: pulmonary congestion and oedema, intra-alveolar haemorrhage, fragments of striated muscle lodged in the air spaces, alveolar hyperinflation and emphysema, acute degenerative change in the heart muscle, and congestion of major organs. Diagnosis remains harder in pinnipeds than cetaceans because fur rarely records net marks, but the combined pattern, alongside circumstantial evidence such as overlap with known fisheries, now supports increasingly confident diagnosis.

The injuries tell a story about the animal’s own behaviour during entrapment. Most small marine mammals caught in nets asphyxiate, but many also sustain severe trauma, indicating vigorous struggling rather than passive submission. A striking example comes from UK stranding data: common dolphins caught in pair trawls, banned within 12 nautical miles of part of the English Channel coastline in 2004, showed a higher rate of broken beaks than other species, reflecting both the gear type and the violence of their escape attempts. Some dolphins reportedly enter a catatonic state, thought to reflect an accentuated diving bradycardia driven by the stress response, yet the lesion record shows physical activity after entanglement is common. Age matters too: calves and juveniles of minke, North Atlantic right and humpback whales, particularly animals under 11 metres, lack the body weight and strength to break ropes and surface to breathe, and in baleen whales, gear fixed to five or more body parts sharply reduces mobility and the chance of breaking free.

The physiological cascade that follows entrapment is well characterised in outline, even if measuring it in a dead animal is extraordinarily difficult. Fear and anxiety stimulate the limbic system, activating the sympathetic nervous system and the adrenal medulla within milliseconds, releasing adrenaline and noradrenaline; the hypothalamic-pituitary-adrenal axis follows within seconds, with cortisol measurable minutes later. In a marine mammal, this intersects with the dive response, which itself releases catecholamines and slows the heart, complicating interpretation. Catecholamines are released in pulses and cleared rapidly, and their breakdown products are unstable in necropsy material, limiting forensic use. Cardiac histology can reveal excessive catecholamine stimulation, but morphological changes may only appear after six to eight hours, longer than most PUE events last. The authors propose newer tools, including plasma troponin I assessment and immunohistochemistry of cardiac muscle, as promising avenues for quantifying the acute stress response and, with it, welfare impact.

What makes this review resonate far beyond pathology labs is the emerging evidence that entrapment is a social event, not just an individual tragedy. In the English North Sea, a passive acoustic monitor recorded two harbour porpoises making foraging buzzes near a static net before one became entangled. The trapped animal immediately switched from feeding buzzes to communication calls, displaced the net upward by about seven metres over four minutes while struggling, then went silent. The surviving porpoise, possibly a mother given the sub-adult size of the victim, produced only communication calls for a period and then continued clicking and calling toward the net for at least 59 minutes, possibly for hours. The researchers suggest this may be the first acoustic recording of a porpoise staying at the scene and calling repeatedly for a lost companion.

Similar patterns appear elsewhere. Bottlenose dolphins entrapped in an Adriatic mid-water pair trawl showed novel changes in whistle parameters, effectively broadcasting distress to pod mates. In the Ionian Sea, after a sub-adult male bottlenose dolphin died entangled in a static net, members of his pod produced an unusual number of vocalisations, apparently in response to his distress calls. Off southern Brazil, intense vocalisations with a high proportion of burst-pulses, suggestive of social communication and care-giving behaviour, continued for almost two hours around a 3,000-metre bottom-set gill net before a dead, entangled Franciscana dolphin was hauled up. Direct video observation in Western Australia captured two bottlenose dolphins in a trawl net showing a startle response and repeatedly pushing against the net’s upper surface for several minutes before becoming motionless. In the eastern tropical Pacific tuna purse-seine fishery, spinner dolphins dived deeper into nets while spotted dolphins showed agitation or a passive ‘sleeping’ behaviour, both interpreted as stress responses involving social separation, restraint and isolation.

The consequences ripple outward. Despite bycatch of spotted and spinner dolphins in the tuna fishery falling from more than six million per year to fewer than 1,000, those populations show no signs of recovery, and calf production has declined since the 1980s. Permanently separated dependent calves may starve after orphaning, unweaned and uncounted, and there is evidence that setting nets on dolphins can trigger miscarriage in pregnant females. Because many odontocetes depend on social cohesion for survival and reproductive success, the review argues that bycatch mortality may disrupt populations in ways that individual removal counts cannot capture, impeding recovery even when numbers of deaths fall. For the survivors, escape is not a clean ending: a Canary Islands study found 19% of sampled stranded cetaceans had been released alive but died from the interaction, and post-mortem work reveals capture myopathies and cardiac damage that can compromise animals long after they leave the net.

Case studies from UK waters ground the science in individual deaths. In January 2016, inspectors in Cornwall recovered two four-month-old grey seal pups from a gill net off Falmouth, freshly dead, recently fed and in good condition, with abrasions around their eyes, muzzles and flippers where they had pushed hopelessly against a 13.6-centimetre mesh. In December 2020, an adult male common dolphin live-stranded in the Helford River and died; necropsy showed clear monofilament net wounds on its rostrum, dorsal fin and fluke, with histology dating the injuries at least six to eight hours before death, a rare documented case of short-lived post-bycatch survival. UK seal bycatch is estimated at 400 to 600 animals annually, concentrated in southwest tangle and trammel nets, and bycatch accounts for 39% of necropsied stranded common dolphins in Cornwall between 2018 and 2024. The authors argue the evidence warrants bycatch prevention on welfare as well as conservation grounds, and call for wider deployment of hydrophones, net cameras, remote data loggers and a proposed pathological ‘fingerprint’ of asphyxia, alongside a hypoxic index and trauma scoring, so that the invisible minutes inside a net can finally be measured, and ultimately prevented.

Subject of Research: The welfare, behaviour and pathophysiology of marine mammals experiencing peracute underwater entrapment in fishing gear

Article Title: The impacts of peracute underwater entrapment in fishing nets on marine mammal behaviour and welfare

Article References: Dolman, S., Barnett, J., & Wessels, M. (2026). The impacts of peracute underwater entrapment in fishing nets on marine mammal behaviour and welfare. Frontiers in Ethology, 5, Article 1882616. https://doi.org/10.3389/fetho.2026.1882616

Image Credits: AI Generated

DOI: 10.3389/fetho.2026.1882616

Keywords: marine mammals, bycatch, peracute underwater entrapment, cetaceans, pinnipeds, fishing gear, asphyxiation, stress response, necropsy, animal welfare, harbour porpoise, common dolphin

Mercury is a non-essential, globally distributed toxic metal with no biological function, and its behavior in marine food webs makes it a particular concern for top predators. In aquatic ecosystems, inorganic mercury released from industrial sources, coal combustion, and artisanal gold mining is converted by microbes into methylmercury, an organic form that is readily absorbed by living tissue. Unlike many contaminants that dilute as they move up the food chain, methylmercury bioaccumulates in individual organisms and biomagnifies across trophic levels, meaning that animals at the top of the web, such as marine mammals, can carry body burdens orders of magnitude higher than the water they swim in. Because methylmercury crosses the placenta and the blood-brain barrier, developing fetuses are especially vulnerable to its neurotoxic effects.

The research team, led by Anaclara Roselli-Laclau of the Universidad de la República in Uruguay together with colleagues from Uruguay and Brazil, focused on two sympatric otariid species that breed side by side at Isla de Lobos but occupy different ecological niches: the South American sea lion (Otaria flavescens) and the South American fur seal (Arctocephalus australis). The researchers quantified total mercury, or THg, in two complementary matrices: hair from adult females and lanugo, the fine natal fur that pups grow in utero and shed in the weeks after birth. Sampling covered 11 mother-pup pairs of sea lions and 46 mother-pup pairs of fur seals, providing an unusually direct way to examine maternal transfer because both members of each pair were analyzed.

The results revealed a striking contrast between the two species. Adult female sea lions averaged 30.5 micrograms of mercury per gram of hair on a dry weight basis, with a standard deviation of 9.3, while female fur seals averaged just 5.7 micrograms per gram, with a standard deviation of 3.4. The pattern was mirrored in their offspring: sea lion pups averaged 25.3 micrograms per gram of lanugo, compared with only 4.4 micrograms per gram in fur seal pups. In other words, sea lions at this colony carried roughly five to six times more mercury than their fur seal neighbors, and the concentrations measured in sea lion pups rank among the highest ever recorded for pinnipeds anywhere in the world.

This interspecies gap is best explained by differences in diet and foraging ecology. Sea lions from Uruguayan colonies are known to feed at higher trophic positions and in waters closer to the Río de la Plata estuary, a system influenced by both natural and anthropogenic metal inputs, while fur seals tend to forage further offshore on prey lower in the food web. Because methylmercury biomagnifies with each step in the food chain, a predator eating larger, older, more predatory fish accumulates substantially more of the metal than one feeding on small pelagic species. Previous work by the same group had already linked trophic habits to bioaccumulation in these two otariids, and the new mother-pup data extend that picture to the earliest stage of life.

Perhaps the most consequential finding is that mercury was detectable in lanugo at all. Lanugo is formed entirely before birth, so any mercury it contains must have been transferred across the placenta from the mother during gestation. The presence of THg in this prenatal fur therefore provides direct evidence of prenatal mercury exposure in both species, confirming that placental transfer, and not only lactation, is a significant route by which these animals hand their contaminant burden to the next generation. This aligns with earlier studies on Baltic grey seals and northern elephant seals, which documented transplacental mercury transfer, but the Uruguayan data add a South Atlantic baseline for two species that had been comparatively understudied.

Statistical modeling showed that pup mercury concentrations were positively associated with maternal concentrations in both species, which is exactly what one would expect if the pup’s prenatal fur faithfully reflects the mother’s body burden during pregnancy. However, the strength of that relationship differed dramatically between the species. In sea lions, the final model explained 75 percent of the variance in pup THg, a remarkably tight fit that suggests maternal levels are the dominant driver of pup exposure in this species. In fur seals, by contrast, the model explained only 20 percent of the variance, indicating that other factors, such as individual variation in foraging behavior, gestation timing, or physiological processes governing mercury mobilization, play a much larger role in shaping how much of the metal reaches each fetus.

The models also uncovered a sex effect that was present in one species but absent in the other. Among sea lions, male pups had lower mercury concentrations than female pups for a given maternal mercury level, a pattern that could reflect sex-specific differences in prenatal growth, body composition, or maternal investment, since male otariid pups are typically larger and may dilute their contaminant burden across more tissue mass. No such sex effect was detected in fur seals. The authors note that this kind of species-specific nuance matters for anyone hoping to use pup fur as a shortcut for monitoring adult contamination, because the relationship between mother and pup is not uniform across species or even between the sexes within a species.

That leads to the study’s central practical message: pup lanugo is not a reliable quantitative proxy for the mercury concentration of an individual mother. The variability in the mother-pup relationship, especially in fur seals, is too great to back-calculate a specific female’s burden from her pup’s fur. But the researchers argue that lanugo remains valuable in a different way. Because it can be collected from pups with minimal disturbance, without immobilizing or even closely approaching adult females, it offers a minimally invasive matrix for characterizing prenatal mercury exposure and for tracking broader species-level and population-level patterns over time and across geography. For colonies that are difficult to sample intensively, this could transform monitoring programs that would otherwise require invasive procedures on protected animals.

The broader implications extend beyond the South Atlantic. Marine mammals are widely regarded as sentinel species for ocean and human health, and elevated mercury in newborn sea lions signals that contamination in the southwestern Atlantic food web is reaching the most sensitive life stages of long-lived apex predators. Mercury exposure in young mammals has been linked in laboratory and field studies to impaired immune function, altered lymphocyte proliferation, and neurodevelopmental deficits, and similar concerns have been raised for harbor seal pups and Steller sea lion calves elsewhere. As mercury continues to enter the ocean from atmospheric deposition and from gold mining in the Amazon basin and beyond, studies like this one provide the baseline data needed to detect change, identify the species and populations most at risk, and inform international efforts, such as the Minamata Convention, to curb the flow of this persistent neurotoxin into marine ecosystems.

Subject of Research: Maternal-to-pup transfer of total mercury in South American sea lions and fur seals measured through maternal hair and pup lanugo

Article Title: Maternal–Pup Mercury Relationship in South American Sea Lions and Fur Seals

Article References: Roselli-Laclau, A., García-Alonso, J., Valdés, A., Freitas-Souza, M., Almeida, M. G., de Rezende, C. E., & Franco-Trecu, V. (2026). Maternal–Pup Mercury Relationship in South American Sea Lions and Fur Seals. Archives of Environmental Contamination and Toxicology, 91(3), Article 23. https://doi.org/10.1007/s00244-026-01227-z

Image Credits: AI Generated

DOI: 10.1007/s00244-026-01227-z

Keywords: mercury, South American sea lion, South American fur seal, pinnipeds, lanugo, prenatal exposure, biomagnification, ecotoxicology, marine mammals, Uruguay, maternal transfer, heavy metals

The review, led by Suelen Goulart and colleagues at the Federal University of Santa Catarina in Brazil, followed the PRISMA protocol for systematic reviews, screening 255 articles from the CAPES Periodicals Portal and ultimately including 31 that met strict criteria: original peer-reviewed studies, published in English, using biopsy sampling on live marine mammals to assess environmental contaminants. Because the animals are sampled at sea without capture, restraint, or sedation, biopsies offer something necropsies of stranded animals cannot: a real-time snapshot of contamination in healthy, active populations. Stranded animals, the authors note, are often debilitated or decomposed, underrepresent robust populations, and introduce biases that impede continuous health monitoring. Biopsies also allow effective metabolic quenching, making the samples reliable for metabolomics, and they can be paired with ecological data such as satellite telemetry and stable isotope analysis to build a fuller picture of exposure.

The tissue most often sampled is blubber, which appeared in roughly 74 percent of the studies, followed by skin at about 39 percent. The preference is no accident. Blubber is the energy store and thermal regulator of marine mammals, and it preferentially accumulates lipophilic contaminants, making it the ideal matrix for persistent organic pollutants. Skin, by contrast, is rich in keratin and other proteins with sulfhydryl groups, and it turns out to be the superior target for non-lipophilic inorganic contaminants such as mercury, which bind proteins rather than fats. The choice of tissue, in other words, is dictated by the chemistry of the pollutant being hunted.

Persistent organic pollutants dominated the field: 23 of the 31 studies, or 74.19 percent, focused on them, with polychlorinated biphenyls (PCBs) and dichlorodiphenyltrichloroethanes (DDTs) the most frequently quantified, followed by hexachlorobenzene, polybrominated diphenyl ethers, and chlordanes. Both PCBs and DDTs were banned in the 1970s and 1980s, yet they persist in marine food webs with alarming intensity. In false killer whales from Hawaii, 84 percent of individuals exceeded safety thresholds for total PCBs, with average concentrations reaching 30,000 nanograms per gram lipid weight. In the Gulf of Mexico, bottlenose dolphins showed PCB levels as high as 82,500 nanograms per gram lipid weight and DDTs up to 164,000. Adriatic dolphins carried a DDT metabolite comprising 90 percent of their total DDT burden. The Mediterranean Sea emerged as the single most studied region, accounting for 39.1 percent of the POPs papers, a reflection of its history of industrial contamination, restricted water circulation, and dense coastal urbanization.

The review also reveals a striking biological pattern: in 82.6 percent of the POPs studies, males accumulated higher contaminant loads than females. The explanation lies in maternal offloading. During gestation, pollutants such as PCBs, DDTs, and flame retardants cross the placenta into fetal tissue, and after birth, calves receive further doses through milk, often reaching concentrations that exceed maternal levels. Females effectively shed their toxic burdens with each reproductive cycle, while males have no such outlet. Juveniles, too, showed elevated contamination, with humpback whale calves in the Gulf of Maine reaching 12,000 nanograms per gram lipid weight of PCBs, suggesting that accumulation begins well before sexual maturity and that developing bodies are especially vulnerable.

Trophic position proved to be a powerful predictor of exposure. Apex predators carried 10 to 40 times higher concentrations than species feeding lower in the food web. Killer whales in the Canadian Arctic accumulated total POPs of 216,000 nanograms per gram lipid weight, and sperm whales in the Mediterranean reached 22,849 nanograms per gram of PCBs, while krill-feeding fin whales in the same basin carried only 572. The reason is biomagnification: killer whales prey on contaminated seals and cetaceans, sperm whales consume lipid-rich squid, and each step up the chain concentrates the chemicals further. Filter-feeding humpback whales showed moderate contamination except in pollution hotspots such as the Mediterranean, where they reached 10,477 nanograms per gram of DDTs.

The health consequences documented in the review are severe. Concentrations above 9,000 nanograms per gram lipid weight are considered a critical threshold for immunosuppression in aquatic mammals, and 65.2 percent of the POPs studies reported levels exceeding known toxicity thresholds. Populations of long-finned pilot whales and Cuvier’s beaked whales in the Mediterranean, bottlenose dolphins in the Atlantic, Gulf of Mexico, and Indian Ocean, and false killer whales in Hawaii all surpassed the 17,000 nanogram threshold associated with reproductive impairment. At the molecular level, more than half of the POPs studies documented disrupted detoxification, endocrine, and neurological systems. In one innovative study, researchers cultured skin fibroblasts from biopsy samples of Norwegian killer whales and exposed them to a mixture of the ten most abundant pollutants at environmentally relevant concentrations, observing significant downregulation of the detoxification gene CYP1A1 and deregulation of lipid metabolism pathways. In Florida dolphins, high DDT levels were associated with reduced testosterone in males and suppressed cortisol in females, impairing reproduction and stress responses.

Inorganic contaminants received far less attention, with only eight studies, yet the findings were no less concerning. Mercury appeared in six of the eight papers, and the Mediterranean pilot whales carried the highest concentrations recorded, at 33,028 nanograms per gram. Resident bottlenose dolphins in Ecuador’s Gulf of Guayaquil all exceeded immunotoxic mercury thresholds. In Australia, stranded Burrunan dolphins from industrialized regions carried blubber mercury levels nearly three times higher than live individuals from less impacted waters, with hepatic concentrations reaching 840,000 nanograms per gram, evidence that long-term dietary accumulation eventually overwhelms the body’s detoxification capacity. The review also highlights chromium-induced genotoxicity in fin whale skin cells, which showed dose-dependent cytotoxicity and DNA damage in laboratory exposures. Perhaps most striking is how little is known about emerging contaminants: only two studies examined phthalates, yet the plasticizer DEHP was detected in every sample from three cetacean species, reaching an extreme outlier of 4,697 nanograms per gram in a bottlenose dolphin near Madeira.

Geographically, the science is strikingly lopsided. North America accounted for 38.1 percent of the continental study occurrences and Europe 28.6 percent, while Africa and South America contributed just 7.1 percent each and Antarctica and Australia 4.8 percent each. Cetaceans, particularly bottlenose dolphins, which appeared in 11 of the 31 studies, dominated the literature, while pinnipeds were nearly absent, largely because sampling seals requires capture, restraint, or sedation, making the procedure far more stressful than a remote dart. The authors argue that this bias leaves regions facing intense plastic disposal and chemical contamination, including Brazil’s so-called Blue Amazon and the coasts of Uruguay and Argentina, dangerously under-monitored, a concern amplified by recent regulatory shifts such as Brazil’s new pesticide law, which replaces stringent danger criteria with more flexible risk assessments.

The review closes with a call to action that extends well beyond marine mammalogy. The authors urge standardization of collection, storage, and analytical protocols, noting that differences in chromatographic methods have produced discrepancies in PCB measurements of up to 40 percent between studies, and they propose global biobanks and certified reference materials modeled on the United States National Institute of Standards and Technology. They advocate integrating biopsies with multi-omics approaches, stable isotope analysis, satellite telemetry, and machine learning to map pollution hotspots and identify resilient populations. Framed within the One Health paradigm, the message is that the contaminants accumulating in whale blubber and dolphin skin are the same ones moving through the seafood that coastal and Arctic communities depend on. In a warming ocean where climate change may remobilize legacy pollutants, the small dart sample retrieved from the wake of a diving whale may prove to be one of the most important early-warning systems humanity has for the health of the sea, and of itself.

Subject of Research: Biopsy-based contaminant monitoring in free-ranging marine mammals as a measure of ocean health

Article Title: Assessing oceans’ health through biopsy sampling in free-ranging marine mammals: a systematic review (2014–2024)

Article References: Goulart, S., Galani, G., Lima, L. F. D., Piccinin, I. N. L., Valente, F. D. S., Nunes, A., Renault-Braga, E. P., & Maraschin, M. (2026). Assessing oceans’ health through biopsy sampling in free-ranging marine mammals: a systematic review (2014–2024). Environmental Science and Pollution Research, 33(28), 14055-14085. https://doi.org/10.1007/s11356-026-38163-3

Image Credits: AI Generated

DOI: 10.1007/s11356-026-38163-3

Keywords: marine mammals, biopsy sampling, persistent organic pollutants, PCBs, DDTs, mercury, phthalates, cetaceans, ocean health, ecotoxicology, One Health, pollution monitoring

The methodology behind the study is as elegant as it is demanding. Rather than relying on anesthesia and invasive electrodes, the researchers used behavioral psychophysics, the gold standard for measuring perception in animals. Marine keepers trained each animal to voluntarily participate in the sessions, using positive reinforcement in the form of fish rewards. When a tone was played, the animal pressed a paddle; when no tone was present, it withheld the response. By systematically varying the frequency and intensity of the tones and tracking the animals’ detection performance, the team could plot a hearing threshold curve, known as an audiogram, for each individual. This approach, described by Taronga behavioural biologist Dr Benjamin Pitcher as giving the animals a voluntary choice to participate, ensures that the measurements reflect genuine perception rather than stress-induced artifacts, while keeping the subjects engaged and cooperative throughout the process.

The results revealed a clear peak of sensitivity. Both species were most responsive to sounds around 3.2 kilohertz, a frequency that falls squarely within the range of many natural communication signals and environmental cues. Above and below this optimum, sensitivity declined gradually, as it does in most mammals, with the animals becoming progressively less able to detect very low-frequency rumble and very high-frequency ultrasound. For the Australian sea lion, an endangered species with an estimated population of only about 12,000 animals, this baseline is more than an academic curiosity. Erbe emphasized that understanding what an animal can hear is the indispensable first step in any assessment of noise impacts, because a sound that falls outside an animal’s audible range, no matter how loud, cannot directly disturb its hearing or mask its acoustic signals.

The significance of that principle becomes obvious when the audiograms are overlaid with the spectra of human-made noise. The researchers compared the long-nosed fur seals’ hearing thresholds with the acoustic signatures of boats, ships, aircraft, trains and construction activity, and found that many of these sounds would be clearly audible to the animals when they were hauled out on land. This is a critical finding for coastal management. Pinnipeds spend substantial portions of their lives on beaches, rocky platforms and haul-out sites, often in close proximity to ports, harbors, airports and expanding urban development. If a passing vessel or an aircraft overhead produces sound energy within the animals’ most sensitive frequency band, it has the potential to mask communication between mothers and pups, interfere with the detection of predators or simply impose chronic acoustic stress on animals that are already conserving energy between foraging trips.

For the Australian sea lion, the stakes are particularly high. Endemic to Australian waters and already reduced to a small, fragmented population, the species relies on hearing for a suite of essential behaviors, including communication between individuals, locating prey, navigating through coastal waters and detecting threats. Unlike visual or chemical cues, sound travels efficiently both in air and underwater, making it one of the most versatile sensory channels available to a marine mammal that lives a double life across two media. Any degradation of that channel, whether through masking by vessel noise or disturbance at breeding colonies near busy waterways, could compound the other pressures facing the species, from fisheries interactions to prey scarcity and habitat limitation.

The studies, published separately in the journals Conservation and Animals, document the aerial audiograms of two long-nosed fur seals, Arctocephalus forsteri, and two Australian sea lions, Neophoca cinerea. The distinction between aerial and underwater hearing matters enormously here. Otariids, the family that includes sea lions and fur seals, are amphibious hunters whose ears must function in both environments, and their sensitivity in air and in water can differ substantially. The new measurements establish what these animals can hear in air, which is directly relevant to their time on land, but the underwater half of the picture remains to be filled in. As Erbe and Pitcher both noted, the next step is to understand how noise affects the animals under water, and at what point sound begins to interfere with their behavior or hearing.

That gap is not trivial. Underwater noise from shipping, seismic surveys, pile driving and offshore energy infrastructure has become one of the most pervasive and poorly regulated forms of marine pollution, and regulatory frameworks worldwide increasingly rely on species-specific hearing data to set exposure criteria. An animal’s audiogram determines whether a given noise source is merely background clutter or a genuine hazard. Without baseline hearing data, environmental impact assessments are forced to extrapolate from related species, often from different hemispheres or different ecological niches, introducing uncertainty that can undermine conservation decisions. By providing measured thresholds for two Australian species, the Curtin and Taronga team has given regulators and industry a defensible empirical anchor for the aerial component of those assessments.

The behavioral training regime that made the work possible also deserves attention, because it represents a humane and increasingly favored alternative to traditional methods of measuring animal sensory capabilities. Historically, hearing data for many marine mammals came from stranded or captive animals tested under anesthesia, or from invasive electrophysiological techniques that required restraint. Voluntary psychophysical testing, by contrast, allows the same individuals to be retested over time, produces thresholds that reflect the animal’s full attention and motivation, and doubles as a form of environmental enrichment for the animals involved. The fact that two sea lions and two fur seals could be trained to perform a discrimination task precise enough to yield publishable audiograms speaks to both the flexibility of these animals and the patience of the keepers and researchers who worked with them.

The broader context of the research is a coastline under increasing acoustic pressure. Australia’s shipping lanes are among the busiest in the Southern Hemisphere, coastal development continues to expand around major population centers, and offshore industries are projected to grow as energy demands evolve. Each of these activities adds sound to environments where pinnipeds haul out, breed and rest. For long-nosed fur seals, whose colonies increasingly overlap with urbanized coastlines in southeastern Australia, the finding that boats, aircraft, trains and construction noise fall within their audible range suggests that acoustic disturbance should be considered alongside more traditional concerns such as entanglement and prey depletion when managing haul-out sites and colony buffers.

What makes this research resonate beyond the specialist literature is the simplicity of its central insight: before we can ask whether our noise harms an animal, we must first learn what that animal can hear. For decades, the acoustic worlds of most marine mammal species have remained unmapped, leaving scientists and managers to guess at the boundaries of perception. The Australian sea lion and the long-nosed fur seal have now joined the short list of pinnipeds with measured aerial audiograms, and their contribution may ultimately extend well beyond their own populations. Every species-specific hearing curve that enters the literature strengthens the predictive tools used to protect the many other animals that still live in acoustic silence, at least as far as human science is concerned. As shipping lanes widen and coastal cities grow louder, the quiet paddle presses of four trained animals at Taronga Zoo may prove to be an unexpectedly consequential act of conservation.

Subject of Research: Aerial hearing sensitivity of Australian sea lions and long-nosed fur seals in relation to human-made noise

Article Title: Hearing tests reveal how sea lions and fur seals hear human-made noise

Article References: Hearing tests reveal how sea lions and fur seals hear human-made noise. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: Australian sea lion, long-nosed fur seal, audiogram, hearing sensitivity, noise pollution, marine mammals, pinnipeds, Taronga Zoo, Curtin University, behavioral audiometry, coastal development, conservation