Introduction
Birds of the family Ramphastidae (toucans) are known for their colourful feathers and long beaks. However, beyond their beauty, these birds play a crucial role in the maintenance of forest ecosystems. As a consequence of their predominantly frugivorous and generalist diet, along with their large gape width, large body size and ability to fly long distances, toucans are considered excellent seed dispersers: they can disperse seeds of various sizes over distances of hundreds of metres (Remsen et al., Reference Remsen, Hyde and Chapman1993; Galetti et al., Reference Galetti, Laps and Pizo2000, Reference Galetti, Pires, Brancalion and Fernandez2017; Holbrook, Reference Holbrook2011). This ability to consume and disperse large seeds is significant because there is a direct relationship between the body size of frugivores and the size of the seeds they can handle (Jordano, Reference Jordano2014). Plant species with smaller seeds tend to benefit from a greater diversity of potential dispersers as they can be consumed by small, medium and large frugivores. In contrast, plant species with larger seeds can only be dispersed by larger frugivores. Therefore, for large-seeded species, the presence of a large frugivore such as the toucan is crucial (Jordano, Reference Jordano2014; Rehling et al., Reference Rehling, Jaroszewicz, Braasch, Albrecht, Jordano and Schlautmann2021).
In the Anthropocene, with many ecosystems undergoing defaunation, the importance of large frugivores becomes more evident. Defaunation does not occur randomly; large vertebrates tend to disappear first because they are generally more sensitive to habitat fragmentation and are often targeted by hunters (Dirzo et al., Reference Dirzo, Young, Galetti, Ceballos, Isaac and Collen2014). The absence of these animals leads to the loss of interactions between large-seed dispersers and large-seeded plant species (Galetti & Dirzo, Reference Galetti and Dirzo2013; Galetti et al., Reference Galetti, Guevara, Côrtes, Fadini, Von Matter and Leite2013; Dirzo et al., Reference Dirzo, Young, Galetti, Ceballos, Isaac and Collen2014; Jordano, Reference Jordano2014; Gardner et al., Reference Gardner, Bicknell, Baldwin-Cantello, Struebig and Davies2019). Among the causes of defaunation, urbanization stands out because habitat loss caused by this process tends to be more permanent compared to loss caused by agriculture or logging, which can be reversed (McKinney, Reference McKinney2002).
The overall impact of urbanization on biodiversity depends on the extent of remaining vegetation (Marzluff & Ewing, Reference Marzluff and Ewing2008; Carbó-Ramírez & Zuria, Reference Carbó-Ramírez and Zuria2011; Melo & Piratelli, Reference Melo and Piratelli2023). Many studies have emphasized the role of urban green spaces in maintaining biodiversity because they serve as refuges, feeding grounds, breeding areas, ecological corridors and stepping stones between vegetation fragments (e.g. Ricketts & Imhoff, Reference Ricketts and Imhoff2003; Alberti, Reference Alberti2010; Zhou et al., Reference Zhou, Huang and Cadenasso2011; Aronson et al., Reference Aronson, La Sorte, Nilon, Katti, Goddard and Lepczyk2014; Gil & Brumm, Reference Gil and Brumm2014; Pena et al., Reference Pena, Martello, Ribeiro, Armitage, Young and Rodrigues2017; Callaghan et al., Reference Callaghan, Bino, Major, Martin, Lyons and Kingsford2019). Urban green areas can be a vital source of resources for frugivorous birds because of the frequent availability of fruits from exotic plant species (Rolando et al., Reference Rolando, Maffei, Pulcher and Giuso1997; Marzluff et al., Reference Marzluff, Bowman, Donnelly, Marzluff, Bowman and Donnelly2001; Kark et al., Reference Kark, Iwaniuk, Schalimtzek and Banker2007; Loss et al., Reference Loss, Ruiz and Brawn2009; Gasperin & Pizo, Reference Gasperin and Pizo2009; Philippsen et al., Reference Philippsen, Benedito and Zawadzki2010). However, toucans are secondary cavity nesters, meaning they rely on pre-existing tree cavities in the environment, rather than excavating their own (Skutch, Reference Skutch1971; Sick, Reference Sick1997). For these birds, urban ecosystems pose an additional challenge as cities often lack large, old trees with suitable cavities (Blewett & Marzluff, Reference Blewett and Marzluff2005; Cockle et al., Reference Cockle, Bodrati, Lammertink and Martin2015; Tomasevic & Marzluff, Reference Tomasevic and Marzluff2017).
Despite being one of the areas most affected by colonization and urbanization in Brazil (Herzog & Finotti, Reference Herzog and Finotti2013), the city of Rio de Janeiro still has a number and variety of green spaces, including protected areas, urban parks, squares and cultivated areas. The Tijuca National Park is one of the largest urban forests in the world (3,951 ha), embedded within the urban matrix of Rio de Janeiro. The Park’s fauna is impoverished and management interventions include reintroduction attempts to restore native biodiversity. As part of a conservation project, the Ariel toucan Ramphastos ariel was reintroduced during 1970–1971, after being considered locally extinct in the 1960s as a result of habitat degradation, fragmentation and hunting for their feathers for imperial cloaks (Sick & Pabst, Reference Sick and Pabst1968; Coimbra-Filho & Aldrighi, Reference Coimbra-Filho and Aldright1971; Coimbra-Filho, Reference Coimbra-Filho2000). After the release of 47 birds rescued from the illegal wildlife trade, no follow-up studies took place despite the known importance of post-release monitoring of reintroduced populations (IUCN, 2013). No information on the size of the current population was available even though the species can be readily observed feeding and reproducing both in the reintroduction area and surrounding green spaces.
In this study, we aimed to assess how the reintroduced population of R. ariel is utilizing an urban green area connected to its release site, and how this area is contributing to the success of the reintroduction. Although comprehensive monitoring is essential for a full evaluation, we focused on frugivory and reproduction, which are widely accepted indicators of population independence and sustainability (Sanz & Grajal, Reference Sanz and Grajal1998; Richards & Short, Reference Richards and Short2003).
Study area
The study was carried out in the city of Rio de Janeiro, one of the most populous cities in Brazil, in the Rio de Janeiro Botanical Garden (Fig. 1). The garden was established in 1808 and is adjacent to a busy road but is connected to the Tijuca National Park and lies within its buffer zone. It covers 137 ha and includes 54 ha of cultivated areas and 83 ha of Atlantic Forest remnants (a biome that encompasses the entire city). The cultivated area hosts c. 8,200 plant species, including species native to Brazil and exotics from other countries (Molinaro & Costa, Reference Molinaro and Costa2001; Peixoto & Guedes-Bruni, Reference Peixoto and Guedes-Bruni2010). A total of 186 bird species have been recorded in the area (Trindade & Rajão, Reference Trindade and Rajão2017) including R. ariel. Ariel toucans are readily observed in the Rio de Janeiro Botanical Garden and Tijuca National Park, as well as in other urban green areas and when passing through developed areas in and around the city.
Fig. 1 The location of Rio de Janeiro Botanical Garden, Rio de Janeiro city, Brazil, and its location in relation to Tijuca National Park (the site of an Ariel toucan Ramphastos ariel reintroduction in 1971) and its buffer zone.
Methods
Frugivory and dispersal potential
We used the feeding bouts method to assess frugivory (Galetti, Reference Galetti1993, Reference Galetti1996; Pizo & Galetti, Reference Pizo and Galetti2010). This method relies on visual tracking to record feeding events (bouts) regardless of the time spent interacting with the food resource or the number of items consumed (Plate 1a). We recorded one bout for R. ariel when it was observed feeding on a particular fruiting plant species, and a second bout when it fed on a different plant species, and so on. We searched for birds from the paths between the cultivated plant areas and recorded feeding whenever toucans were encountered.
Plate 1 The Ariel toucan Ramphastos ariel foraging and nesting in the Rio de Janeiro Botanical Garden, Rio de Janeiro, Brazil: (a) feeding on Virola surinamensis, (b) active toucan nest in tree cavity, (c) interior of tree cavity nest site showing eggs and seed debris, (d) newly hatched nestlings.
One or two observers followed the main pathways until toucans were detected visually or acoustically. As toucans are conspicuous and abundant in the area, encounters typically occurred soon after the search began. Once an individual or group was located, the observers followed it until it was lost from sight. Most plant species in the Botanical Garden are labeled with identification tags, facilitating species identification. When plants lacked identification tags, their locations were recorded using a GPS and later identified in consultation with the Botanical Garden staff. Toucans in this area appeared tolerant of human presence, allowing close approaches to fruiting plants during feeding events. Days without toucan encounters were rare, and the birds were most often observed in pairs or groups. On one occasion, a group of 23 toucans was recorded. Groups or pairs typically fed on the same individual plant; however, when one or more birds moved to a different plant, focal observations were shifted accordingly to increase the number of feeding bouts recorded. We carried out observations once per week during July 2017–March 2023 between 7.00 and 10.00, which is the toucan’s peak foraging time.
We counted the number of fruits consumed during feeding bouts, distinguishing between fruits that were eaten on the ground having fallen beneath the plant and fruits swallowed whilst feeding above ground and therefore considered to be potentially dispersed. We characterized the species of fruit eaten according to the plant’s origin (native to the Atlantic Forest or exotic, with the latter defined as those from other Brazilian biomes or from other countries), IUCN Red List category and CNC Flora category (Centro Nacional de Conservação da Flora, the national authority responsible for compiling the Red List of threatened plant species in Brazil) for nationally threatened species or for those not yet assessed for the IUCN Red List, and seed size. We categorized seeds as small, medium or large according to Brancalion et al. (Reference Brancalion, Bello, Chazdon, Galetti, Jordano and Lima2018), where small seeds have a diameter of < 6 mm, medium seeds have a diameter of 6–12 mm, and large seeds have a diameter > 12 mm. We took seed measurements from Bello et al. (Reference Bello, Galetti, Montan, Pizo, Mariguela and Culot2017) and Lorenzi (Reference Lorenzi1992, Reference Lorenzi1998, Reference Lorenzi2009).
Reproduction
We monitored the reproductive success of R. ariel over two breeding seasons (August–March) during 2019–2021. We identified potential nest cavities with the help of birdwatchers and gardeners who frequent the area daily. We accessed the cavities by climbing the trees or using ladders. We assumed that the presence of seed debris was an indication of reproductive activity. We located 10 nests (Plate 1b), which we monitored throughout the breeding period, defined as the interval between the date the first egg was laid and the date the last chick fledged and left the nest.
We monitored nests once per week, using an endoscopic camera (Panasonic HX-A100 HD Wearable Action Camera Digital Camcorder, Panasonic, Japan) for observations inside the cavity (Plate 1c,d). We installed camera traps (Bushnell Trophy Cam HD E3, Bushnell, USA) near the entrance of the nests to record feeding events and chick fledging. Cameras were set to run continuously and programmed with high sensitivity for both daytime and night-time recording, to ensure that the daytime sensor triggered in low light levels in the early morning and late afternoon. The recording duration was set to 30 s with a 1 s interval between recordings. We used 32 GB memory cards (the maximum supported by the camera) and rechargeable batteries. We recorded nest characteristics, including tree species, cavity position, tree height, diameter at breast height, tree diameter at the cavity entrance and height of the cavity from the ground, as well as vertical and horizontal measurements of the cavity entrance and cavity depth.
Results
Frugivory and dispersal potential
We recorded 850 feeding bouts of Ariel toucans during July 2017–March 2023 (monthly mean = 13; minimum = 3, June 2022; maximum = 35, April 2022). We observed the birds taking 4,358 fruits from 91 plant species of 28 families (Supplementary Table 1), of which 28 species (30%) were native to the Atlantic Forest and 65 (70%) were non-native, exotic species. They ate fruits of various sizes, with 57 speies (61%) containing medium-sized seeds, 20 (21%) small seeds and 14 (16%) large seeds. Most of the fruits were swallowed (3,961, 91%), indicating potential dispersal, and 397 (9%) were dropped. One feeding bout was recorded on flowers of Bombax ceiba and two feeding bouts on flowers of Pachira aquatica.
Table 1 The 12 plant species consumed most frequently by the Ariel toucan Ramphastos ariel in the Rio de Janeiro Botanical Garden, Brazil (Fig. 1), during July 2017–March 2023, with number of feeding bouts (and per cent of all feeding bouts), number (and per cent) of fruits removed (regardless of whether they were consumed or dropped), whether native to the Atlantic Forest or exotic (i.e. native to another Brazilian biome or other country), IUCN Red List category (IUCN, 2025) and national Red List category (CNC Flora, 2020).
1LC, Least Concern; NT, Near Threatened, VU, Vulnerable; EN, Endangered.
2Small, diameter < 6 mm; medium, 6–12 mm; large, ≥ 12 mm.
Twelve plant species accounted for 75% (n = 637) of feeding bouts and 81% (n = 3,543) of fruits removed (i.e. fruits either eaten or dropped; Table 1). Seven species had medium-sized seeds, three had large seeds and two had small seeds. Three species are native to the Atlantic forest, of which Euterpe edulis is categorized nationally as Vulnerable (CNC Flora, 2020) and Dendropanax arboreus and Aniba formula are categorized as Least Concern globally (Hargreaves, Reference Hargreaves2024). One of the exotic species (Eugenia crenata) is categorized as Vulnerable globally, one (Virola surinamensis) as Vulnerable nationally and one (Ardisia solanacea) as Endangered nationally (Table 1). The Arecaceae (palms) were the family most represented in the feeding bouts (53%, n = 454), in fruits removed (60% n = 2624) and in the number of species (32%, n=30).
Reproduction
Ariel toucan pairs usually breed annually, during September–February. We located four tree cavities being used by pairs of Ariel toucans in the Rio de Janeiro Botanical Garden. We recorded nests in three tree of these cavities, by 10 pairs of birds over two breeding seasons, indicating that cavities were reused during and between breeding periods by different pairs of birds (Ariel toucans can be identified individually by the markings on their bills). In two of these cavities, a new clutch of eggs was laid within the same breeding season shortly after fledglings from a different pair of birds left the nest. All cavities were located in large, living trees, and were characterized by a narrow entrance (c. 7 cm wide; Table 2).
Table 2 Details of three tree cavities used as nest sites by R. ariel in the Rio de Janeiro Botanical Garden, Brazil, during 2019–2021.
We monitored 10 nests, three during 2019–2020 and seven during 2020–2021. Clutch size was 3–4 eggs. Three nests already contained chicks when they were discovered. Of the 24 eggs found, 20 (83%) hatched and the chicks survived to the week following hatching. We monitored 29 chicks of which 21 (72%) survived the fledging period (causes of death were unknown) and left the nest at c. 46 days (Table 3). We observed a variety of food items brought to the nest by both parent birds, of which 75% were fruits, 7% were animal items (eggs, chicks, invertebrates, small vertebrates) and 18% were unidentifiable.
Table 3 Number of eggs laid by R. ariel, eggs hatched and chicks that fledged and left the nest in 10 monitored nests in the Rio de Janeiro Botanical Garden, Brazil, during 2019–2020 and 2020–2021.
Discussion
We have demonstrated that the Rio de Janeiro Botanical Garden provides the necessary resources (fruits and nesting sites) to support the reintroduced population of R. ariel. We observed toucans foraging on the fruits of 91 plant species and the flowers of two species over the 7-year study, indicating that the area provides a year-round food supply from the diverse range of native and exotic plants species. Toucans generally are reported to feed on a total of 115–120 plant species (Short & Horne, Reference Short, Horne, Del Hoyo, Elliott and Sargatal2002). The mix of species in the botanic garden supplies a greater quantities of fruit over a longer period than the plants found in more natural areas (Wang et al., Reference Wang, Zhan, Liao and Huang2023).
Medium-size seeds were consumed most frequently in terms of the number of feeding bouts and fruits removed, followed by large and small seeds. This underscores the role of toucans as potential dispersers of medium and large seeds, which tend to have fewer dispersers than small-seeded species. Our work highlights the importance of reintroductions for restoring ecological interactions in a defaunated ecosystem (Genes et al., Reference Genes, Cid, Fernandez and Pires2017; Mittelman et al., Reference Mittelman, Kreischer, Pires and Fernandez2020).
Although urban green areas can play an important role in supporting native biodiversity, these spaces frequently feature interactions between exotic and native species. In the Rio de Janeiro Botanical Garden, R. ariel mostly fed on plant species that are not native to the Atlantic Forest (Table 1, Supplementary Table 1). Although this is not unexpected in a botanical garden, it is of concern for the conservation of the native flora. Toucans are excellent seed dispersers and the botanical garden is connected to the Tijuca National Park, one of the few remnants of the Atlantic Forest. Botanical gardens are known to be potential sources of non-native invasive species (Mack, Reference Mack2005; Dawson, Reference Dawson, Mndolwa, Burslem and Hulme2008; Hulme, Reference Hulme2011), and the Rio de Janeiro Botanical Garden therefore poses a risk to the botanical integrity of the Tijuca National Park. However, the magnitude of the risk is difficult to assess because some exotic species, such as the palm Roystonea oleracea, were planted during restoration replanting in the Park in the 19th century (Drummond, Reference Drummond1996; Freitas et al., Reference Freitas, Neves and Chernicharo2006). Nonetheless, there are other species with invasive potential that are not present in the National Park but that interact with fauna in urban green areas and need to be monitored. These include the Amazon palm Euterpe oleracea, which could have a negative impact on the native and threatened palm E. edulis if it became established in the Atlantic Forest (Tiberio et al., Reference Tiberio, Sampaio-e-Silva, Matos and Antunes2016). Ramphastos ariel feeds on both species and could transport their seeds to new areas.
Cavity availability is a limiting factor for cavity-nesting birds, especially for secondary excavators such as R. ariel (Wiebe, Reference Wiebe2011; Cockle et al., Reference Cockle, Bodrati, Lammertink and Martin2015). Toucans select deep cavities with narrow entrances, possibly as a predator-avoidance strategy (Lill, Reference Lill1970; de Jesus et al., Reference de Jesus, Buzzato and de Araujo Monteiro-Filho2012; Perrella & Guida, Reference Perrella and Guida2019). Urbanization can be a major threat to breeding success because large, mature trees with decaying branches and soft heartwood prone to cavity formation are often removed on safety grounds in built-up areas. The density of cavity-nesting birds is lower in urban areas than in natural ecosystems and the reuse of tree cavities as nest sites may indicate a general scarcity in these environments (Blewett & Marzluff, Reference Blewett and Marzluff2005; Cockle et al., Reference Cockle, Bodrati, Lammertink and Martin2015; Tomasevic & Marzluff, Reference Tomasevic and Marzluff2017). Botanical gardens often serve as key refuges for cavity-nesting species in urban areas because they preserve older trees with cavities. This is demonstrated by the successful reproduction of R. ariel in the Rio de Janeiro Botanical Garden.
We conclude that the reintroduced population of R. ariel has colonized the Rio de Janeiro Botanical Garden because it offers a diverse range of food plants and suitable nesting sites in mature trees. Breeding success is high and its location in the buffer zone of the Tijuca National Park where the reintroduction took place has allowed the birds to occupy this new area. We were able to study the Arial toucan at close quarters within the botanic garden, where it is more approachable and observable than in high tree canopies in natural forests. We were able to collect data on foraging preferences and feeding behaviour more easily compared to studies of toucans in unmanaged environments (Galetti et al., Reference Galetti, Laps and Pizo2000, Ragusa-Netto, Reference Ragusa-Netto2006, Ragusa-Netto, Reference Ragusa-Netto2008, dos Santos & Ragusa-Netto, Reference dos Santos and Ragusa-Netto2013). We recommend further study and long-term monitoring of the reintroduced Ariel toucan population in the Rio de Janeiro Botanical Garden and in other urban green spaces across the city, to better understand the factors contributing to the success of the reintroduction in the adjacent Tijuca National Park.
Supplementary material
The supplementary material for this article is available at doi.org/10.1017/S0030605325101634.
Author contributions
Fieldwork: FBZ, JN, CLMF; data analysis, writing: all authors.
Acknowledgements
We are grateful to the Rio de Janeiro Botanical Garden for staff assistance and equipment; the Fauna Center of the Rio de Janeiro Botanical Garden and the Bird Watchers of Rio de Janeiro for information about nests; the Luísa Pinho Sartori Institute for funds to purchase camera traps; João Quental for donating equipment; and the Pontifical Catholic University of Rio de Janeiro for a scholarship to FBZ.
Conflicts of interest
None.
Ethical standards
This research abided by the Oryx guidelines on ethical standards. No animals were captured or collected during this study, which received support and authorization from the Fauna Center of the Rio de Janeiro Botanical Garden and was conducted under their supervision.
Open access
