When mahi-mahi and striped marlin hunted the same prey school, the pattern did not match the authors' prediction that mixed groups would have lower attack and capture rates for both species. Instead, the reported marlin capture interval fell from about 135.5 seconds in marlin-only sequences to 46.8 seconds in mixed sequences, while the probability of an isolated marlin capture rose from 56% to 87%. Mahi-mahi took fewer isolated prey and more prey from the school than expected from their share of the mixed group.
The evidence came from a record of free-ranging hunts rather than an assigned experiment. Researchers filmed predator aggregations from above with unmanned aerial vehicles offshore Baja California Sur, Mexico, in November 2021, 2022, 2023 and 2024. The main behavioral analysis covered 32 hunting sequences: 13 with only marlin, 10 with both species and 9 with only mahi-mahi. Continuous coding marked school attacks, successful captures, isolated captures and visible predator counts, then summarized attack and capture frequency as the time between events.
A faster, more divided hunt
In the mixed sequences, the interval between attacks was about 48% shorter than in marlin-only sequences. The interval between captures was about 60% shorter, and the interval between isolation events about 67% shorter. These are comparisons of observed event timing, so they show a faster sequence of recorded attacks, captures and isolations, not by themselves a cause-and-effect result.
Attack order showed a different kind of pattern. After a mahi-mahi attack, the next attack was by mahi-mahi in 86% of cases and by marlin in 14%. After a marlin attack, the next attacker was marlin in 52% of cases and mahi-mahi in 48%. After adjusting for the abundance of each species, the model supported the mahi-following pattern but not the marlin-after-marlin pattern.
Across predator compositions, mahi-mahi captured prey successfully in about 30% to 44% of school attacks, compared with about 21% for striped marlin. The within-species comparisons between alone and mixed conditions were not credible, so the result is a broad species contrast rather than a clean estimate of how mixing changed either predator's efficiency.
Different places in the feeding space
Location added another layer to the contrast. The study's same-cell measure, which tracks how often a species was closer to the prey school, was 0.65 for mahi-mahi and 0.35 for striped marlin. The spatial areas associated with marlin were about 1.73 times larger, and the researchers described marlin as occupying more peripheral positions.
That separation became clearer as an attack unfolded. Before contact, the distance difference was not credible. From about 3.5 seconds onward, striped marlin were farther from the prey school; by the end of the observation window, they were about 3.8 meters away, compared with 1.3 meters for mahi-mahi, roughly three times farther.
A pattern, not a cause
A separate abundance analysis used 41 hunting sequences from 2023. Visible mahi-mahi, marlin and California sea lions were counted in still frames taken every 30 seconds, and the sequences averaged 298 seconds with a reported standard error of 45 seconds. In that analysis, each additional mahi-mahi corresponded to about 5% fewer striped marlin, while each additional sea lion corresponded to about 25% fewer. The model did not provide reliable evidence for a specific mahi-mahi count that completely excluded marlin.
The abundance result should be read as an association within the filmed aggregations. Predator composition was not assigned, and the main behavioral sample was divided across only 32 sequences from four study years, while the abundance analysis came from 2023. That small, uneven set cannot settle whether the same relationships hold in other places, years or predator communities.
Taken together, the footage is consistent with resource partitioning: mahi-mahi stayed nearer the school and took more school prey, while marlin made more isolated captures and occupied larger, more peripheral spaces. It does not show that mahi-mahi caused marlin numbers to fall or caused marlin to leave, nor does it establish effects on daily energy intake, reproduction or population dynamics.
Paper data and sources
Original title: The presence of mahi-mahi at multi-predator feeding aggregations modifies striped marlin attack behaviour
Authors: Hansen M, Bartashevich P, Bideau P et al.
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