The web of feeding associations among seabirds and marine mammals around South Georgia became far less connected over the past three decades, a study of shipboard surveys found. The share of possible species pairings judged statistically significant fell from about 40% in 1991 to 13% in 1993 and 7% in 2023.
The pattern is consistent with fewer opportunities for species to benefit from one another while finding food, a process the authors discuss as local enhancement. But the study did not directly measure individual local-enhancement behaviour, so it cannot show that the decline in pairings changed individual foraging or fitness.
A changing community, and fewer links
The analysis covered 24 bird species and seven marine mammal species recorded during shipboard transect surveys off eastern South Georgia in May to July of 1985, 1991, 1993 and 2023. Across the 38-year period, Antarctic prions were almost completely lost from the observations. Whale sightings also changed markedly, rising to 47 in 2023 from two in 1993, none in 1991 and one in 1985.
About 57% of the taxa formed a significant association in at least one survey year. Among the species that formed pairings, the typical number of links also fell: the median dropped from 10 in 1991 to five in 1993 and four in 2023. The ranges were three to 10, one to nine and one to seven, respectively.
Species abundance was positively correlated with the number of significant pairings in every survey year, with reported Spearman correlations ranging from 0.64 to 0.80. In 2023, the relationship remained after diving-petrels were removed, with a reported p-value of 0.002.
What counted as an association
The researchers used recurrent group analysis, a statistical test that asks whether two species appeared together more often than expected from their independent spatial distributions. They treated a pairing as significant when its reported p-value was below 0.05.
The pairwise comparison was possible for 1991, 1993 and 2023, using observations at about 18.5-kilometre spatial resolution. The 1985 survey supplied only daily species totals, so it could be used to examine community structure but not to calculate a comparable network measure.
The historical surveys used 300-metre strip transects. For comparability, observations from 2023 were collected using distance sampling and then truncated after the fact to a 300-metre comparison band.
The result points to a possibility, not a cause
The authors interpret the shrinking network as consistent with reduced opportunities for local enhancement and food acquisition in a changing predator community. The observations do not identify what produced the shift, and the authors note that abundance, prey availability and the unusual 2023 iceberg complicate attribution.
One unusual feature of the 2023 cruise was a 100-kilometre-long iceberg, A-76, grounded off eastern South Georgia. Most of the snow and Antarctic petrels observed that year were directly associated with the iceberg or pieces of it, adding another reason to be cautious about treating 2023 as a simple continuation of the earlier surveys.
The method captures positive co-occurrence between species, not direct use of local enhancement. It also excludes associations between individuals of the same species. The surveys were made in daylight, and ship observations may detect seabirds more readily than marine mammals or swimming species.
The study therefore describes a temporal change in observed species composition and species-level associations around eastern South Georgia. It does not establish that climate warming caused the decline, that fewer pairings caused poorer foraging or fitness, or that the pattern applies to other regions, seasons or night-time feeding.
What would answer the bigger question
The authors point to the need for additional winter surveys covering more of the season, along with dietary and prey-availability data. Tracking individual animals or recording their behaviour on video would be needed to test whether the observed species pairings translate into local-enhancement behaviour and any effect on foraging or fitness.
The data files and R code are provided in the supplementary material. The authors reported no competing interests. The work was supported by the Division of Antarctic Sciences and the U.S. National Science Foundation Office of Polar Programs, along with the College of Staten Island Biology Department and a City University of New York Graduate Center Fellowship.
Paper data and sources
Original title: Changes in species composition of South Georgia seabirds and mammals 1985–2023: decrease in opportunities for local enhancement?
Authors: Monier S, Manne L, Veit R
Journal/Repository: Philosophical transactions of the Royal Society of London. Series B, Biological sciences
Status: Peer-reviewed
First online: 2026-10-19
DOI: Not available
Original paper · Full text