Introduction
The Western Atlantic subpopulation of Red Knot (Calidris canutus rufa) breeds in the Canadian Arctic Archipelago and migrates to the southern hemisphere to overwinter (Baker et al. Reference Baker, González, Serrano, Júnior, Efe and Rice2005; Niles et al. Reference Niles, Sitters, Dey, Atkinson, Baker and Bennett2008; Lathrop et al. Reference Lathrop, Niles, Smith, Peck, Dey and Sacatelli2018). In Canada (Loi sur les especes menacees ou vulnerables, RLRQ cE-12.01) and USA the rufa Red Knot is listed as “Endangered” and “Threatened with extinction”, respectively (US Endangered Species Act in 2014). In Brazil, where the species overwinters, it is listed as “Critically Endangered” (Red Book of Endangered Brazilian Fauna of 2018), respectively, reflecting the ongoing declines of the species in these countries. The subspecies global population is an estimated 42,000 specimens, and 15,000–20,000 individuals are considered long-distance migrants, while the rest performs short migrations (Niles et al. Reference Niles, Burger, Porter, Dey, Minton and González2010; Andres et al. Reference Andres, Smith, Morrison, Gratto-Trevor, Brown and Friis2012). Short-distance migrants migrate to the Gulf of Mexico and the northern coast of South America (Maranhão, Brazil), while the long-distance group flies all the way to Tierra del Fuego, in the extreme south of Argentina and Chile (Baker et al. Reference Baker, González, Serrano, Júnior, Efe and Rice2005).
Due to the energy expenditure of their long flights, Red Knots depend on the existence of stop-over sites with an abundance of food and resting areas along their migratory routes (Harrington et al. Reference Harrington, Loch, Niles and Kalasz2010; Burger et al. Reference Burger, Niles, Porter, Dey, Koch and Gordon2012). During long-distance migrations, Red Knots stop along the mid coast of Rio Grande do Sul and at the Lagoa do Peixe National Park, Brazil to forage, rest, and perform post-reproductive feather moult (Harrington et al. Reference Harrington, Antas and Silva1986; Scherer and Petry Reference Scherer and Petry2012; Niles et al. Reference Niles, Burger, Porter, Dey, Minton and González2010). The passage of Red Knots occurs during the northbound migration in late March and early April and during the southbound migration in September and October (Harrington et al. Reference Harrington, Hagan and Leddy1988; Scherer and Petry Reference Scherer and Petry2012; Niles et al. Reference Niles, Sitters, Dey, Atkinson, Baker and Bennett2008). Some individuals may remain in Lagoa do Peixe for extended periods as seasonal residents. These are primarily young individuals or adult birds that do not migrate for reasons not yet understood (Lara-Resende and Leeuwenberg Reference Lara-Resende and Leeuwenberg1987; Martínez-Curci et al. Reference Martínez‐Curci, Isacch, D’amico, Rojas and Castresana2020). As Lagoa do Peixe and the coast of Rio Grande do Sul are one of the last stops before the extended flights to North America, these two areas have been considered critical to the viability of the rufa Red Knot (Oliveira et al. Reference Oliveira, Barbosa, Sousa, Lugarini, Lima and Nascimento2016).
Compared with the well-documented migration route of rufa Red Knots along the Atlantic coast of the USA (Burger et al. Reference Burger, Niles, Porter, Dey, Koch and Gordon2012; Niles et al. Reference Niles, Burger, Porter, Dey, Koch and Harrington2012; Newstead et al. Reference Newstead, Niles, Porter, Dey, Burger and Fitzsimmons2013), the migration route in South America is still poorly understood and mainly based on flag resightings. Niles et al. (Reference Niles, Burger, Porter, Dey, Minton and González2010) used geolocation tracking devices to access the southbound movements of Red Knots from Delaware Bay and detected the general use of the Central Brazil and Brazilian Atlantic coast routes. However, the length of flights and the use of stop-overs on the northward migration are even more poorly understood. The present study aimed to fill this knowledge gap by investigating rufa Red Knot migration routes starting in Lagoa do Peixe, southern Brazil, using PinPoint Argos-75 GPS Transmitters (Lotek Wireless, Ontario, Canada) (Argos Reference Argos2018). Migratory patterns, flight paths, stop-over duration, and distances travelled were analysed.
Methods
Red Knots were tagged at the Lagoa do Peixe National Park (31°00’46”S, 50°46’31”W; 31°29’00”S, 51°09’51”W), which includes a complex of coastal lagoons connected to the sea by an inlet, tidal marshes, dunes, and barrier beaches facing the Atlantic Ocean. The park covers 34,000 ha of the coastal plain of the State of Rio Grande do Sul in southern Brazil (Figure 1), and it is located between the municipal areas of Tavares and Mostardas (Oliveira et al. Reference Oliveira, Barbosa, Sousa, Lugarini, Lima and Nascimento2016). Since 1986, it has been classified as a Full Protection Conservation Unit (Presidência da Republica, Brasil 1986) and due to its importance for migratory shorebirds, it is internationally recognised as a Biosphere Reserve Site by the Ramsar Convention, to which Brazil is a signatory. The Migratory Wader’s Action Plan “PAN Waders” also recognises the importance of Lagoa do Peixe for hemispheric birds and highlights it as a stop-over for migratory shorebirds (ICMBio 2016).
Study Area: Lagoa do Peixe National Park, in Southern Brazil is shown in the marked area, and the capture location is shown with an arrow.
Red Knots were captured at the beach section of the park on 24 April 2019 using a cannon net (Figure 1). Five of the 51 captured individuals were selected for receiving PinPoint GPS Argos 75 tracking devices. Only birds with reproductive plumage and a body mass of >200 g were selected, which is the ideal body mass for an individual to be able to begin or continue its migration (Aldabe et al. Reference Aldabe, Rocca, Gonzalez, Caballero-Sadi and Baker2015). Tags were attached to the back of the birds, 1 cm above the uropygial gland (Scarpignato et al. Reference Scarpignato, Harrison, Newstead, Niles, Porter and van den Tillaart2016; Lathrop et al. Reference Lathrop, Niles, Smith, Peck, Dey and Sacatelli2018). An area of c.2 × 1.5 cm was cleared of feathers and devices were glued with high adhesion, long-lasting surgical glue directly on to the skin (Scarpignato et al. Reference Scarpignato, Harrison, Newstead, Niles, Porter and van den Tillaart2016) (Figure 2). A drop of blood was collected from the ulnar vein of the birds for molecular sexing. Tags weighed c.3.7 g, which constitutes less than the safety threshold of 3% of the bird’s body mass (Cooper et al. Reference Cooper, Hallworth and Marra2017). Tags were programmed to record fixes between 25 April 2019 and 4 October 2019, but with particular schedules according to the period (northward migration or breeding). Each bird that had not yet been marked with a flag received a coded blue flag (Brazilian country colour). Birds are hereafter identified by their flag codes (E8A, E7X, E7N, E9Y, and V53).
a) PinPoint GPS device attached to the back near the uropygium; b) Red Knot carrying the transmitter.
Total distances (distance from the place where the birds were marked to the last location recorded by the tag, in km), relative distances (distance between one position and the next position), travelling time (days), migration speed (km/day), average flight speed (km/hour), and time (days) spent on the stop-over were calculated for each individual.
Results
Birds were tracked from 10 to 90 days. Four individuals migrated, while one (E8A) remained at the Lagoa do Peixe National Park during the tracking period. All birds were females (see Supplementary material, Figure S3) and showed variation in migratory parameters and routes (Figure 3 and Table S1). Two females (E7X and E7N) migrated via the Central Brazil route, and the other two (E9Y and V53) migrated along the Brazilian Atlantic coast route (Figure 3). Only one (E7X) was tracked all the way to the Arctic breeding grounds.
Migratory route taken by Red Knot from the Lagoa do Peixe National Park - Brazil. The route of each bird is marked with different colors and the points are the recorded locations.
Bird E7X was tracked between 30 April and 10 June, when the tag sent its last fix. The bird made a non-stop, 8,289 km long flight from Lagoa do Peixe to Delaware Bay (north-east coast of the USA). It left Lagoa do Peixe on 30 April and took 14 days to reach Delaware Bay, flying over the Pantanal region, the Amazon Forest, and the Caribbean Sea (Figure 3). Stopover time at Delaware Bay was at least six days before departing to James Bay (Canada) and flying another 1,658 km. From James Bay, the bird flew an extra 1,206 km in nine days, reaching its breeding grounds at Southampton Island (Nunavut, Canada) around 10 June. Distances between fixes ranged from 1,206 to 4,916 km. Migration speed ranged from 105 km/day to 1,229 km/day, and the estimated flight speed reached a maximum of 51 km/hour. The total distance covered by this female was 11,170 km during a period of 45 days.
Bird E7N was tracked from 26 April to 20 May. Initially, it followed a similar route to E7X and also flew over the Pantanal (Figure 3). However, afterwards she made a turn to the north-east of Brazil and, after travelling 3,280 km in six days, she made two stops, one near the Xingu River and the other at Caxiuna Bay, State of Pará (Figure S1b). E7N remained in the region for four days before travelling another 614 km to the coast of Maranhão on 8 May, where she remained for at least 14 days until the tag stopped sending data. Relative distances ranged from 614 km to 1,646 km and the migration speed ranged from 307 km/day to 817 km/day. The estimated flight speed was 34 km/hour and the total distance covered by this female during a period of 12 days was 4,050 km.
Bird V53 was tracked between 26 April and 6 May. It migrated along the Brazilian coastal route, leaving Lagoa do Peixe on 26 April and, after six days, had a fix recorded at the coast of Bahia, north-east Brazil (Figure 3). From there it flew 1,619 km to the east in two days and arrived at Santa Rosa Bay, in the mouth of the Amazon River, Pará, on 6 May (Figure S1a). The bird stayed at this location for at least four days before the last fix was recorded. Relative distance ranged from 214 km to 2,028 km and the migration speed between 54 km/day and 1,014 km/day. Maximum estimated flight speed was 42 km/hour and total distance covered was 3,873 km.
The device from bird E9Y recorded only two fixes, one in south-western Brazil and another one off the Atlantic coast of Suriname. The bird covered a total distance of 4,524 km, during a period of 16 days.
Red Knots connected Lagoa do Peixe to Delaware Bay and the coast of Maranhão (Brazil), Caxiuana Bay, the interior of Pará and Santa Rosa Bay – mouth of the Amazon River – coast of Pará with the boundary of the State of Amapá (Brazil) (Figure 3 and Figure S1). Differences in covered distances and travel times for all females are shown in Figure S2.
Discussion
Our results reveal the occurrence of intrapopulation variation in migratory strategies of Red Knots. Despite slight variations, two major routes from Lagoa do Peixe were observed: the Central Brazil route (Pantanal, Amazon) and the Brazilian Atlantic coast route (Baker et al. Reference Baker, González, Serrano, Júnior, Efe and Rice2005; Niles et al. Reference Niles, Burger, Porter, Dey, Minton and González2010; Oliveira et al. Reference Oliveira, Barbosa, Sousa, Lugarini, Lima and Nascimento2016). Differences in migratory strategies within the same population of birds have been reported elsewhere (Alves et al. Reference Alves, Dias, Méndez, Katrínardóttir and Gunnarsson2016). An individual’s body mass and unfavourable climatic conditions can influence the direction that birds take along their migratory journey (Niles et al. Reference Niles, Burger, Porter, Dey, Koch and Harrington2012; Covino et al. Reference Covino, Holberton and Morris2015). In addition, quantity and quality of prey at foraging sites and risk of predation at stopovers can play an important role in migration strategies (Burger et al. Reference Burger, Niles, Porter, Dey, Koch and Gordon2012).
Bird E7X confirmed a direct connection between Lagoa do Peixe and Delaware Bay on a direct flight of approximately 8,300 km. Although the first fix in Delaware was registered only on 10 May, we estimated that the bird arrived earlier. Based on the average flight speed (51 km/hour) calculated between its first (30 April, Mato Grosso) and second fix (2 May, Sargasso Sea), we estimate that E7X left Lagoa do Peixe on 29 April and arrived in Delaware Bay on 6 May. The estimated average flight speed is within the average of 50–60 km/hour already observed for long-distance migratory flights in birds (Newstead et al. Reference Newstead, Niles, Porter, Dey, Burger and Fitzsimmons2013; Bishop et al. Reference Bishop, Buchanan, McCaffery and Johnson2016). Our estimate is also consistent with previous studies, which found that an 8,000 km route from southern Brazil to the coast of North Carolina was covered in 6 days (Niles et al. Reference Niles, Burger, Porter, Dey, Minton and González2010). The arrival date in Delaware Bay (6 May) matches data from previous studies, which registered Red Knots arriving in early May and departing after 26 May (Niles et al. Reference Niles, Burger, Porter, Dey, Koch and Harrington2012; Burger et al. Reference Burger, Niles, Jeitner and Gochfeld2018). Assuming this estimated arrival date, E7X appears to have stayed in Delaware Bay for a period of c.23 days, since the next record was made on 1 June in James Bay (Ontario, Canada). This estimate is compatible with previous studies (1–19 days) (González et al. Reference González, Baker and Echave2006; Newstead et al. Reference Newstead, Niles, Porter, Dey, Burger and Fitzsimmons2013). James Bay is a well-known stop-over for Red Knots, from where birds can fly directly to Hudson Bay (Niles et al. Reference Niles, Burger, Porter, Dey, Koch and Harrington2012; Andres et al. Reference Andres, Smith, Morrison, Gratto-Trevor, Brown and Friis2012). Red Knots typically take 2–3 days to complete the route from Delaware Bay to James Bay (Newstead et al. Reference Newstead, Niles, Porter, Dey, Burger and Fitzsimmons2013; Andres et al. Reference Andres, Smith, Morrison, Gratto-Trevor, Brown and Friis2012), which falls within our estimate. The bird’s last stop was in the south of Southampton Island (Nunavut, Canada), which is a known breeding site for the species (Lyons et al. Reference Lyons, Winn, Keyes and Kalasz2018; Lathrop et al. Reference Lathrop, Niles, Smith, Peck, Dey and Sacatelli2018).
Bird E7N connected Lagoa do Peixe to the Maranhão recesses, the overwintering area of short distance rufa migrants and a known important stop-over for long-distance migrants (Atkinson et al. Reference Atkinson, Baker, Bevan, Clark, Cole and Gonzalez2005; Baker et al. Reference Baker, González, Serrano, Júnior, Efe and Rice2005). E7N used unreported areas for this species; one wetland in the Extractive Reserve “Forever Green” between the Amazon and Xingu River and one in the Caxiuanã National Forest, in the southern portion of Caxiuanã Bay, both in the State of Pará. It appears that E7N was foraging in this inland area, since the two fixes recorded at the site indicated that it spent at least four days circulating in the area before heading to the Maranhão coast, where she stayed for another 14 days. However, such a long stay in Maranhão is unusual and represents a delayed migration and higher exposure to risks occurring at the area, which could compromise her reproductive success (Alves et al. Reference Alves, Dias, Méndez, Katrínardóttir and Gunnarsson2016; Bishop et al. Reference Bishop, Buchanan, McCaffery and Johnson2016).
Bird V53 apparently adopted a different migration strategy. We assume it continued in short flights along the Atlantic coast (Table S1), since it only covered 214 km within two days, a short distance compared with other birds. Santa Rosa Bay, at the mouth of the Amazon River, was used by this bird for at least four days. Use of this remote area as a stop-over has never been documented for the species. Selection of a stop-over site that is further north in Brazil highlights the occurrence of a wider stopover area for the population, since it is located between areas where the species usually stops: the State of Maranhão and the coast of French Guianas and Suriname (Baker et al. Reference Baker, González, Serrano, Júnior, Efe and Rice2005; Navedo and Gutiérrez Reference Navedo and Gutiérrez2019).
The single two locations from bird E9Y with such a spatial gap between them make it hard to know if it made any stops between the location where it was tagged and the last position off the Suriname coast. However, considering the migration speed (345 km/day), it was probably not a direct flight, and it is possible that the bird used the Maranhão recesses before crossing Central America.
The distances of displacement of the tracked birds (1,600–3,400 km) are similar to other studies with the same species: 2,000 km (Bishop et al. Reference Bishop, Buchanan, McCaffery and Johnson2016), 2,300 km (Carmona et al. Reference Carmona, Arce, Ayala-Perez, Hernández-Alvarez, Buchanan and Salzer2013), and 3,300 km (Niles et al. Reference Niles, Burger, Porter, Dey, Minton and González2010). Medium-sized migratory birds such as Red Knots can fly small 1,000–3,000 km (Brown et al. Reference Brown, Gratto-Trevor, Porter, Weiser, Mizrahi and Bentzen2017) or long distances, averaging from 5,000 to 8,000 km (Brown et al. Reference Brown, Gratto-Trevor, Porter, Weiser, Mizrahi and Bentzen2017; Hill and Renfrew Reference Hill and Renfrew2019). Our study corroborates Niles et al. (Reference Niles, Burger, Porter, Dey, Minton and González2010) study, which states that rufa Red Knots employ two migration strategies: direct long-distance flights from the south of Brazil to the USA and short-distance flights with stops either inland or along the Atlantic coast before jumping off northward. Our data confirm the relevance of several major stop-overs to the migration success, such as Lagoa do Peixe and the Maranhão coast, but also of minor stop-overs to support birds along their migratory routes. The use of Santa Rosa Bay as a minor stop-over is intriguing and more information about which habitats birds are using is needed. Migratory bird species play a key role in the connectivity of these environments (Gonzáles et al. 2006; Lyons et al. Reference Lyons, Winn, Keyes and Kalasz2018; Finch et al. Reference Finch, Butler, Franco and Cresswell2017), and we show that Red Knots connect national and international areas along its route.
Our study shows that the Lagoa do Peixe National Park also serves as a temporary residence for birds that for some unknown reason interrupt migration. There are several hypotheses to explain this pause, such as juvenile individuals, poor physical condition, and stop-over risks such as chemical contamination by pollutants and parasite load (Martínez-Curci et al. Reference Martínez‐Curci, Isacch, D’amico, Rojas and Castresana2020). However, E8A already had a breeding plumage and an ideal weight to migrate, so, it is possible that the bird suffered some contamination and weakened quickly after being tagged and so did not migrate. Either way, these data highlight the importance of Lagoa do Peixe for Red Knots even outside the migration season.
Acknowledgements
This project was partly financed by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ) (No. 425884/2016-3) and by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) through a Master’s scholarship to Antônio Coimbra de Brum and a PhD scholarship to Júlia V. G. Finger (Finance codes 001). The authors would like to thank colleagues from Unisinos Laboratório de Ornitologia e Animais Marinhos and Wildlife Restoration Partnerships of New Jersey for their fieldwork support. A special thanks to Dr Carmem Fedrizzi, Leonardo Dorneles, Douglas Ribeiro da Silva, Marlon Ferraz, Christopher Buidin, and Yann Rochepault for their contributions.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0959270923000308.
