Killer whales (Orcinus orca, Linnaeus, 1758) have a wide distribution throughout the world’s oceans. They are most commonly found in higher latitudes in temperate and sub-polar waters with high marine productivity but also occur in offshore and less productive tropical waters (Leatherwood and Dalheim, 1978; Forney and Wade, 2006). They are considered top ocean predators, feeding on a variety of different prey, from invertebrates, fishes, and other marine mammals (Ford, 2002). Within their monotypic genus (Rice, 1998), killer whales from the both hemispheres form several ecotypes with distinguished external morphology, habitat use, behavior, social structure, acoustic repertoire, prey preferences, and genetics (Ford et al., 1998; Baird and Whitehead, 2000; Pitman and Ensor, 2003; Visser, 2007; LeDuc et al., 2008; Foote et al., 2009; Pitman, 2011; Durban et al., 2016; Foote et al., 2016). In the south-western Atlantic, killer whales were recorded along all Argentinean coast, mainly in Peninsula Valdés (Iñíguez, 2001). Small groups have been sighted around the Falkland (Malvinas) Islands (Yates et al., 2007) and in Uruguay, were observed in coastal waters and also related with depredation on catches of the pelagic longline fishery in offshore waters (Iriarte, 2006; Passadore et al., 2015).

Along most of the Brazilian coast, between latitudes 03°N to 32° S, killer whales have been reported as opportunistic sightings or stranding events of single individuals (Lodi and Hetzel, 1998; Siciliano et al., 1999; Dantas, 2007). Capture records by ex-soviet pelagic fleets during the whaling period (1969/70-1978/79) were also recorded in the region (Mikhalev et al., 1981). Their occurrence in coastal waters off south-eastern Brazil was associated with upwellings conditions and prey availability (Siciliano et al., 1999) and records have increased in the last two decades as more search efforts were applied (Pinedo et al., 2002; Santos and Netto, 2005; Meirelles et al., 2009; Santos and Silva, 2009; Santos et al., 2010; Batista et al., 2012; Ott et al., 2017; Lodi and Tardin, 2018; Santos et al., 2019; Renault-Braga et al., 2019). Few specimens had been studied in detail since fresh strandings are rare (Lemos et al., 2013; Laeta et al., 2019; Groch et al., 2020) and Morin et al. (2015) cite the stranding of an Antarctic killer whale type C in Brazil but without providing detailed information. The depredation by killer whales in the tuna longline fishery that operates from the north to south of Brazil to Uruguay indicates the regular presence of killer whales in offshore waters (Secchi and Vaske, 1998; Dalla Rosa and Secchi, 2007; Dantas, 2007; Passadore et al., 2012; Wedekin et al., 2014). Further evidence of their occurrence in offshore waters off Brazil comes from dedicated surveys and technical reports (Ramos et al., 2010; Andriolo et al., 2015; Di Tullio et al., 2016; PMC-BS, 2017).

Despite these observations off Brazil, the natural history of killer whales in these tropical and subtropical waters remains poorly understood, and data deficiency precludes the assessment of their conservation status (ICMBio, 2023). The first published record of killer whale movements in the region described a solitary male resighted over 11 years along the coast of Rio de Janeiro and Parana (22°S - 25°S) (Santos and Silva, 2009; Lodi and Farias-Junior, 2011). In 2016 and 2017, the PMC-BS (Santos Basin Cetacean Monitoring Project) that monitors the cetacean populations in coastal and oceanic areas of the Santos Basin offshore sedimentary basin to assess possible impacts of oil activities and gas on these animals, tagged four individuals in southeastern Brazil. Their data showed animals that remained approximately 225 km from the coast for 3 to 13 days, and a solitary male that moved from southern Brazil to the southern coast of Uruguay and Argentina during 33 days (PMC-BS, 2017). In addition, Santos et al. (2019) reported on a group of killer whales that travelled between Rio de Janeiro and São Paulo over a period of 27 days and Durban and Pitman (2012), through satellite telemetry, identified movements of five Type B Antarctic killer whales between the Antarctic Peninsula and the oceanic waters of southern Brazil and Uruguay. Such scattered reports on movements and behavior, although precious, hinders our understanding of the complex relationships of killer whale groups of this region. In this paper, new observations including group size, behavior and movements are presented contributing with more pieces of this puzzle in order to have a better understanding of the whole picture. Also, considering that social structure synthesizes a vital class of ecological relationships (Whitehead, 2008), can affect population dynamics, genetics (Wilson, 1975; Strier, 1997) and, is an important factor for conservation (Sutherland, 1998), using traditional photo-identification techniques, we present the first assessment of social structure of killer whales off the Brazilian coast.

There is a remarkable lack of consistent data on killer whales’ population characteristics, social structure and behavior in tropical and subtropical waters worldwide. Despite the opportunistic nature of our sighting data, our findings suggest that killer whales are present regularly along the coast of western south Atlantic, and provide novel insights on the occurrence, movements, foraging and social behavior.

These data suggest that killer whales are apparently more frequently sighted off southeastern Brazil during the austral spring and summer (see also Siciliano et al., 1999)—although caution is needed when interpreting seasonality because nautical activity, thus opportunistic sightings, increase in the region during these seasons. Remarkably, the re-sightings of the same 11 individuals in the same region over the years, including re-sightings of 3 individuals off the coast of Brazil and Uruguay, suggest that killer whales are widely distributed in this region and yet can show some degree of site fidelity to specific locations along this coast. Similar wide-ranging movements have been observed across the world. For example, between Mexico and Peru (Guerrero-Ruiz et al., 2005; Pacheco et al., 2019), Antarctic and New Zealand (Eisert et al., 2015), and between Alaska and California (Goley and Straley, 1994; Dahlheim et al., 2008). In the latter, some social units range more widely and transit hundreds of kilometers along the coast, while other units how higher fidelity to certain regions (Ford and Ellis, 1999).

These movements and occurrence of killer whales are typically associated to prey preferences and availability. In the southeastern coast of Brazil, there are upwelling systems (such as the rich waters of the Cabo Frio) that attract a variety of seasonally abundant sharks, rays, fish, and cetaceans (Siciliano et al., 1999). Previous studies on killer whales off Brazil have shown both predation and the presence in the stomach contents of a wide range of prey, such as rays, sharks, teleosts, salps, penguins, Burmeister’s porpoise (Phocoena spinipinnis), franciscana dolphins (Pontoporia blainvillei) and minke whale (Balaenoptera sp.) (See Table S1 on Supplementary Data ) (Castello, 1977; Bittencourt, 1983; Castello and Pinedo, 1986; Dalla Rosa, 1995; Lodi and Hetzel, 1998; Secchi and Vaske Jr., 1998; Ott and Danilewicz, 1998; Fernandes, 2001; Santos and Haimovici, 2001; Santos and Netto, 2005; Dalla Rosa and Secchi, 2007, Monteiro, 2008; Troina et al. 2020). Two videos released on the internet, one on September 11, 2019 in Bahia and the other on November 5, 2021, in Arraial do Cabo – Rio de Janeiro, show respectively, killer whales in a potential hunt and an attack on humpback whales (unpublished data). This region is also regularly used by Bryde’s Whales (Gonçalves et al., 2015; Lodi et al., 2015; Athayde et al., 2020) and as a migration route for humpback whales (Megaptera novaeangliae) and southern right whales (Eubalaena australis) (Groch et al., 2005; Siciliano et al., 2019; Renault-Braga et al., 2019)—all of which could be potential targets of killer whales. For instance, in 2008, a newborn female southern right whale was found dead with several tooth marks of adult killer whale in the body (Ott et al., 2017) and in 2012, killer whales were observed harassing a group of sperm whales (Physeter macrocephalus) to prey on a sperm whale calf (Andriolo et al., 2015; Sucunza et al., 2022). By contrast, on September 19, 2019, while conducting regular monitoring of southern right whales in the south of Brazil, researchers observed a group of at least four killer whales in direct interaction with a mother-calf pair of southern right whales. Despite the close proximity of the killer whales to the whales, no attack occurred (Renault-Braga et al., 2019). In our observations, killer whales were only seen feeding on elasmobranchs and fish: four times on rays (such as Gymnura altavela) and twice on a school of unidentified fish. No attacks on marine mammals have been recorded and just as observed with the mother-calf pair in 2019, despite the opportunity to prey on two Bryde's whales on November 30, 2018, north of the Ilhabela Archipelago, the group of eight killer whales preferred to attack a stingray. The killer whales found in these waters seem to be more generalist or opportunistic foragers (Lodi and Hetzel, 1998; Siciliano et al., 1999), in agreement with suggestions that killer whales in tropical waters tend to be less specialized on certain prey items than those found in temperate waters because diet breadth should increase as the availability of the most highly profitable prey decreases (Baird, 2002).

Prey type and availability typically shape group size and stability (Beck et al., 2012). Corroborating this idea, the mean size of the group of killer whales observed in this study was 5.61 individuals (SD = 2.91), similarly to previous observations off Brazil (4.3 individuals; Lodi and Hetzel, 1998; 3.9 individuals; Siciliano et al., 1999). This average group size is also similar to observations in other sub-tropical and tropical waters where killer whales show generalist feeding behaviors: Caribbean: 4.1 individuals (Bolaños-Jiménez et al., 2023), Galápagos: 4 individuals (Denkinger et al., 2020), Mexico Central Pacific coast: 4.6 individuals (Vargas-Bravo et al., 2020), Peru: 4.5 individuals (Garcia-Godos, 2004) and 4.3 individuals (Testino et al., 2019), Costa Rica: 3.4 individuals (Castro-Azofeifa, 2021), Hawaii: 4.2 individuals (Baird et al., 2006), Tropical West Africa: 5.6 individuals (Weir et al., 2010), Eastern Tropical Pacific: 5.4 individuals (Wade and Gerrodette, 1993), and Mozambique Channel: 6.1 individuals (Terrapon et al., 2021). By contrast, in temperate and polar zones, group size and is more variable and vary with the target prey type. For instance, the Antarctic Type A killer whales feeds mainly on Antarctic Minke whales and form groups of 13.6 individuals on average; the Type B (11.8 individuals) tend to prey upon pinnipeds, whales and penguins; and the Type C, (46.1 individuals) apparently feeds mainly on fish (Pitman and Ensor, 2003). In Norwegian waters, the mean size for herring-feeding killer whale groups is 15 individuals (Simila et al., 1996) while seal-feeding groups are 5 individuals (Jourdain et al., 2017); in addition, the group sizes of killer whales hunting seals off Scotland are smaller (5.8 individuals) than those on the Icelandic herring grounds (14.8 individuals) (Beck et al., 2012). These observations in Antarctic, Norway, Scotland and Icelandic are similar to those from the well-studied North Pacific population, with a proportional smaller more consistent group size of mammal-eating transient Killer whales (4.2 individuals; Baird and Dill, 1996) compared with fish- eating resident killer whales that often occur in groups of dozens to hundreds of individuals (Ford et al., 1998). According to Baird and Dill (1996) the small size of the mammal-eating groups could contribute to the management and maximization of energy intake. Associations with other groups, increasing the size of the group, would be useful to increase the success rates of prey difficult to capture or large prey allowing additional individuals to feed without increasing competition. Still in these zones, it is also possible to find more diverse groups where whales seem to be more generalists, as in Chilean Patagonia (4.2 individuals; Häussermann et al., 2013 and 5 individuals; Capella et al., 2014) and in New Zealand (mean of 12 individuals) where 27 different species of prey have been recorded, being ray the most common type of prey and, the diet varying according to the 3 sub-populations proposed for the region (Visser, 2007). Taken together, our data and this literature suggest that killer whales of southeastern Brazil are generalist foragers; however, one cannot yet discard the possibility that there may be specific groups with a specialized diet in the region.

Despite the paucity of encounters, our data provide novel insights on the social patterns of killer whales in the western south Atlantic waters. The association patterns among the most re-sighted 12 killer whale individuals suggest that they can form social groups of mixed age-sex, and engage in both long-term and more labile social associations. It is possible that these individuals are part of a single or more relatively stable social units in which members occasionally associate with each other with no clear preferences for sex and maturity stage. This pattern resembles the social structure of transient killer whales in the North Pacific (Ford and Ellis, 1999; Baird and Whitehead, 2000), Marion Islands (Reisinger et al., 2017), and Galápagos Islands (Denkinger et al., 2020), Argentinean Patagonia (Iñiguez et al., 2005) and, Mexico Pacific coast (Guerrero-Ruíz, 2013). However, the nature and scarcity of our association data is such that it remains premature to make such a conclusion. An increased and more homogeneous sampling effort in space and time to complement our current opportunistic and citizen science sampling will be crucial to further reveal the structure into social units, and to estimate the rates at which different units associate. Beyond odontocetes, such as killer and sperm whales, social patterns combining permanent and temporary associations can be found in several animal populations with different degrees of fission-fusion dynamics, such as orangutans, bats, elephants and hyenas (Kerth and König, 1999; Van Schaik, 1999; Wittemyer et al., 2005; Smith et al., 2008). Social plasticity in groups formation can reflect the underlying social and ecological conditions that modulate the costs and benefits that individual member experience. As in the abovementioned populations, associations among multiple social units of killer whales, temporarily increasing the size of the group, can be advantageous for mating opportunities, care of the young, or to increase foraging success on large prey that is difficult to capture; on the other hand, individuals in larger groups pay the cost of increased probability of competition for prey access (Hoelzel, 1993; Baird and Dill, 1996). In Galápagos, the average group size was estimated to be 4 animals. However, when attacking baleen whales, the groups appeared to consist of at least 5-10 animals, and even up to 25 animals when they attacked sperm whales (Denkinger et al., 2020). In northern Patagonia, the associations between killer whales are mostly long-term, but individuals from different groups have been observed to leave their maternal groups and form new groups (Iñíguez et al., 2005). This fluidity has also been identified along the Mexico Pacific coast, where individuals have been found to have indirect relationships with each other through key associations (Guerrero-Ruíz, 2013). In our study, we observed solitary adult males (BR01) that were not seen associated with any group and males that in certain sightings were alone or that were associated with different groups on different occasions, as is the case of males BR12, BR14 and, BR34 ( Figure 6 ). Solitary adult males have also been recorded preying in the region, including in association with the tuna longline fishery (Secchi and Vaske, 1998; Santos and Netto, 2005). According to Baird and Dill (1996), dispersal of individuals likely to occur due to the energetic benefits of foraging alone or in small groups and other groups may also allow them for a period of time to increased mating opportunities. In our study, we did not observe male-male pairs as described in Argentina (Hoelzel, 1991) and the Galápagos Islands (Denkinger et al., 2020).

This is the first integrated assessment of the movements and social behavior of killer whales in tropical and subtropical waters of Brazil. Our data suggest site fidelity by various groups that experience some degree of fission-fusion dynamics and engage in long and short-term movements in the waters of southern and southeastern Brazilian coast. Such movements seem to be erratic and driven by foraging opportunities, likely related to the productivity of western tropical South Atlantic and subtropical waters. Killer whales exhibit a generalist feeding behavior in this area; however, the possibility of there being groups with specialized diets in the region is not ruled out. More systematic surveys, covering both inshore and offshore waters over longer periods, are required to further clarify the status, social patterns, movements in western south Atlantic as well as an analysis of stable isotopes and fatty acids from biopsies to better characterize the diet of killer whales in the region. Given this, it appears premature to classify the killer whales in this particular region as belonging to any current or identified ecotype or a new yet to be defined.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the animal study because the analysis is based in photos of killer whales opportunistically captured.

AA: first authorship. SS: senior authorship. All authors contributed to the article and approved the submitted version.