DOI: https://doi.org/10.21203/rs.3.rs-1340663/v1
The Cattle Egret (Bubulcus ibis) is a native species of the tropical and subtropical regions of Africa, Asia, and the Iberian Peninsula. This species is mainly sedentary but includes migratory populations and inhabits diverse vegetation types including perturbed areas, which probably has facilitated its dispersion. In this study, we describe the geographic patterns during its invasion process in the Americas, characterize the environmental conditions that could favor this process and identify the areas of migration in North America. We obtained the occurrence records in chronological order from its arrival to the Americas to the present, and we made a climatic profile of the occurrence records. Then, we identified and characterized seasonally the environments associated with America's migration areas using temperature and precipitation variables. Starting in 1937 from Guiana, the species colonized the continent via three fronts: 1) through the Antilles until it reached Florida (United States); 2) towards South America until Patagonia; and 3) through Central America to Mexico, the central and western United States (US), and southern Canada. Since its arrival, the species continuously colonized America's new geographic and climatic conditions. However, in the 1970s, it seems to have reached its environmental limits in the northern extreme of its distribution (southern Canada and the US), where the minimum temperatures in winter (-25°C) apparently determine their physiological tolerance limits. Therefore, migrate from those regions to the south to search for more favorable climatic conditions. Currently, the Cattle Egret ranges from southern Canada to Patagonia (Argentina).
The Cattle Egret (Bubulcus ibis) is native to the tropical and subtropical regions of Africa, Asia, and the Iberian Peninsula. This species includes three subspecies: B. i. seychellarum that is distributed in the Seychelles archipelago; B. i. coromandus is native to southern Asia and has colonized large part of Oceania; and B. i. ibis which is native to Africa and Europe; and during the last century has naturally colonized the Americas (García-Fernández 2000, Ahmed 2011).
Bubulcus i. ibis (the Cattle Egret hereafter) is associated with lowlands and mainly inhabits Africa's savannas and tropical rain forests, where most of its native distribution occurs (Chapin 1932, Hancock and Elliot 1978). However, it has adapted to a great variety of tropical and temperate climates, including man-made areas; for example, cattle grazing sites, since it feeds on a great diversity of invertebrates present in all types of transformed environments (Abdelkrim et al. 2001, Dalio 2018). In addition, its diet includes a wide assortment of animals that consist of invertebrates, fishes, amphibians, as well as small mammals, birds and reptiles (Siegfried 1971, Fogarty and Hetrick 1973, Donald 1973, Amat and Soriguer 1981, Menon 1981, Abdelkrim et al. 2001, McKilligan 2005, Torres and Gutiérrez 2005, Sánchez-García 2011, Dalio 2018).
The Cattle Egret showed a migratory behavior in the populations found south of the Sahara Desert and in the extreme south of Africa, where they migrate towards central and northern Africa, particularly towards the Congo region between January and May (Chapin 1932, Browder 1973). It has also been suggested that specific populations from the Iberian Peninsula migrate south into Africa; however, this migratory behavior is not generalized since both migratory and resident populations are in the region (Riddell 1944). This species shows migration patterns in some colonized areas, such as the Americas. For example, populations from Canada and the central US migrate during the winter to the Antilles, Mexico and Central America (Browder 1973, Dunn and Alderfer 2011, Sibley 2014).
During the last century, the species colonized most of the Americas, arriving initially in northern South America. Massa et al. (2014) suggest that the air currents in the Atlantic Ocean favored the conditions for individuals from the coasts of central Africa to reach the coasts of southern America. The first collected specimen of the Cattle Egret in the Americas dates from 1937 in Buxton, Guyana (Blake 1939), representing the starting point to invasion.
Arendt (1988) described the presence of the Cattle Egret throughout different Caribbean islands until it reached Florida in 1952. Hengeveld (1989) described the invasion process and mentioned that in almost 40 years (1940 to 1976), the Cattle Egret colonized the extreme south of South America to southwestern Canada. However, that study included regions that the species have not colonized to that date. In this study we described the geographic and environmental patterns during its invasion process; we also identified the ecological conditions that could favor this process and characterized the areas that harbor migrant populations in North America.
To describe the Cattle Egret invasion process, we conducted an exhaustive literature review (e.g., Blake 1939, Phelps 1944, Drury et al. 1953, Haverschmidt 1953, Peterson 1954, Eisenmann 1955, Rice 1956, Bond 1957, Slud 1957, Stott 1957, Denham 1959, Davis 1960, Land 1963, Dickerman 1964, Hubbard 1966, Johnson and Goodall 1967, Post 1970, Zimmerman 1971, Holland and Williams 1978, Ffrench 1980, Vaz-Ferreira et al. 1981, Arendt 1988, Hengeveld 1989, Petracci and Delhey 2005). We obtained all available presence records from the Global Biodiversity Information Facility (GBIF 2020), Naturalist (2020), eBird (2020) and Ecoregistros (2020). We eliminated all duplicated records, without geographic coordinates, with no reference locality and those with no date.
We analyzed all the records chronologically from 1930 to 2020 in a geographic information system to identify the temporal colonization patterns. Then we compared our findings with the descriptions made by Arendt (1998) and Hengeveld (1989). We described the geographic invasion process across the Americas by decades through dispersion lines in a map (Fig. 1).
To analyze the process of environmental invasion, we performed a Principal Component Analysis (PCA) extracted from the values of the 19 temperature and precipitation variables from the Worldclim project (Fick and Hijmans 2017; Table 1). Then, we identified the eigenvalues of the records by decades. Finally, we performed ellipsoids in the environmental space described by the first three principal components by decade using the ellipsem package in R (Cobos et al. 2019).
Variables | PC1 | PC 2 | PC 3 |
---|---|---|---|
1 = Annual average temperature | 0.260 | 0.268 | 0.215 |
2 = Average daily range (mean monthly (max temp - min temp)) | -0.215 | 0.102 | -0.019 |
3 = Isothermality (P2/P7) (*100) | 0.275 | 0.023 | -0.268 |
4 = Temperature seasonality (standard deviation *100) | -0.301 | -0.076 | 0.227 |
5 = Maximum temperature of warmest month | -0.063 | 0.285 | 0.436 |
6 = Minimum temperature of coldest month | 0.317 | 0.162 | 0.008 |
7 = Annual temperature range (P5-P6) | -0.312 | -0.034 | 0.167 |
8 = Average temperature of wettest quarter | 0.133 | 0.194 | 0.338 |
9 = Average temperature of driest quarter | 0.243 | 0.193 | 0.070 |
10 = Average temperature of warmest quarter | 0.030 | 0.263 | 0.494 |
11 = Average temperature of coldest quarter | 0.308 | 0.201 | 0.014 |
12 = Annual precipitation | 0.276 | -0.240 | 0.079 |
13 = Precipitation in wettest month | 0.289 | -0.099 | -0.015 |
14 = Precipitation in driest month | 0.081 | -0.403 | 0.273 |
15 = Seasonality of precipitation (coefficient of variation) | 0.068 | 0.348 | -0.294 |
16 = Precipitation in wettest quarter | 0.288 | -0.108 | -0.004 |
17 = Precipitation in driest quarter | 0.109 | -0.411 | 0.252 |
18 = Precipitation in warmest quarter | 0.183 | -0.163 | 0.133 |
19 = Precipitation in coldest quarter | 0.216 | -0.248 | 0.038 |
% variance | 45.09 | 19.47 | 15.27 |
% cumulative variance | 64.56 | 79.84 |
We obtained 336,143 occurrences for the Americas from 1930s to 2020. Some authors suggest the Cattle Egret presence in South America since the end of the1800s and the beginning of the 1900s (e.g., Palmer 1962, Wetmore 1963, Crosby 1972, Arendt 1988). Even, there is a report of the species in in 1933 in Providence Island (Arendt 1988). Nevertheless, until 1937, the first specimen was collected in Buxton, Guyana (Blake 1939). From these records, three different invasion fronts appeared. The first front came by the expansion of the Cattle Egret following an insular pathway: the Antilles (Arendt 1988) with the first record in Providence Island (Arendt 1988) in the decade of 1930; subsequently in Aruba (Drury et al. 1953), Puerto Rico and Jamaica in 1940 (Bond 1957, Arendt 1988); the Antigua Island (Holland and Williams 1978); and Trinidad and Tobago (Ffrench 1980). This expansion continued to Florida (US) in the 1950s (Rice 1956). A second invasion front started in Guyana and continued in other South American countries, such as Venezuela (Phelps 1944); Colombia (Haverschmidt 1953, Wetmore 1963); and Peru (Stott 1957) in the 1950s; Chile (Post 1970) and Argentina (Petracci and Delhey 2005) in the 1970s; as far as Uruguay in 1980 (Vaz-Ferreira et al. 1981). The third invasion front followed Central America, towards Costa Rica (Slud 1957), Panama (Eisenmann 1955) and Guatemala (Land 1963) in the 1950s. This front continued towards Mexico, central and western US, to southern Canada in the following decades (Denham 1959, Dickerman 1964, Hubbard 1966, Zimmerman 1971).
We observed the presence records of the Cattle Egret in regions that were not considered previously by Hengeveld (1989). For example, in the decades of 1970 and 1980, we encountered the first records in central and western Mexico and the US; likewise, in Chile and central-eastern Brazil. Although in the most recent decades (1990 to 2020), the records have become more frequent in the latter regions suggesting the invasion completion, which currently includes from the southeastern of Canada to the extreme south of South America.
Regarding to the insular invasion, Arendt (1988) described the invasion process in the Caribbean region from 1930 to 1970. However, adding information to the insular front, we found, based on the bibliographic review, that in 1966 the species was recorded in the San Andres Island on the coast of Nicaragua (Paulson et al. 1969); and in 1968 in the Farallon Islands, US (Richardson et al. 2003). Then, it was recorded in 1976 in the Bay Islands, Honduras (Udvardy 1975); and in 1977 in the Chesapeake Bay Islands, US (Williams et al. 2007). In South America, the island invasion process included the South Georgia and the South Sandwich Islands, UK, in 1979 (Prince and Payne 1979); in 1980 the Vila dos Remédios Islands, Brazil (Nunes 2010); and the Falkland (Malvinas) Islands, UK-Argentine. In 1986, the species reached the Tierra del Fuego archipelago (Argentina) and Cabo de Hornos, Chile (Clark 1986, Chebez and Gómez 1988); and in 1987 in the Easter island, Chile (Marin and Caceres 2010).
There are records in diverse islands in the Americas that do not coincide with the expected geographic and temporal continuous invasion process observed for the Cattle Egret. However, this lack of correspondence may be due to the absence of local recordings rather than a lack of continuity during the species' dispersion. These records correspond to that in 1973 in Vancouver Island, Canada (Kragh 1982); in 1979 in the Channel Islands, US (Stewart and Kovach 1982); in 1988 in Revillagigedo Islands, Mexico (Howell and Webb 1995); in 1999 in Newfoundland and Labrador, Canada, and the Archipelago Bocas del Toro, Panama (Cooper 1999, Montevecchi et al. 2003); in 2000 in the Desventuradas Islands, Chile (Aguirre et al. 2008, Flores 2014); in 2003 in the Nueva Esparta islands, Venezuela (Hilty 2002); and in 2007 in the Swan Islands, Honduras and islands of the gulf of Panama (Aceituno and Medina 2007, Angehr and Kushlan 2007). Finally, in 2014 the species was recorded in the Chriqui Gulf Islands, Panama (Angehr et al. 2014).
On the other hand, the PCA showed that the first three components explained 80% of the variance. In the PC1 and PC3 the most important variables were those related to temperature, conversely to the most important described by the PC2 related to precipitation variables (Table 1). The invasion process in environmental space showed that from 1930s to 1960s the majority of the climate range currently used by the species was covered. Nevertheless, from the beginning of 1970 to date, we did not observe a significant increase in the occupation of new climatic conditions (Fig. 2).
The seasonal patterns described by the Cattle Egret includes its presence from southern Canada to Patagonia between March and November. However, during December, January and February, months that correspond to winter in the northern hemisphere, there is a lack of records in southern Canada and most of the US (Fig. 3). This temporal pattern suggests the existence of migration behavior in the populations from these regions. Such behavior can be associated with the climate limits described by the three principal components representing environmental space. We observed that the range of the values defined by the records from the migration regions are separated from those of the year-round records from other areas. The minimum temperature of the migration regions ranges between -25 to 0°C and the maximum between -10 to 20°C. In the case of the precipitation, we observed lower values in the migration regions in contrast to the rest of the continental records. Conversely, in the southern hemisphere, we did not observe any migratory behavior pattern across the different seasons of the year.
It is recognized that the invasion process by the Cattle Egret occurred naturally in the Americas during the last century, except for Hawaii (US) and the Eastern Islands (Chile), where the species was introduced (Marin and Caceres 2010, Davis and Krauss 1965). This process represents one of the quickest dispersions shown by an avian species, in contrast with other bird species that have been introduced into the Americas (e.g., Streptopelia decaocto, Stunus vulgaris, Passer domesticus). However, other species are expanding rapidly, similar to the Cattle Egret, such as the monk parakeet (Myiopsitta monachus; Gonçalves da Silva et al. 2010).
Our analyses allowed filling the lack of information on the previously described invasion process (e.g., Arendt 1988 and Hengeveld 1989). Currently, we can understand this process and identify the areas occupied by the Cattle Egret in a more accurate way throughout the Americas. The dispersion across the Americas could be favored by several distinctive characteristics of the Cattle Egret that probably contributed to its rapid invasion and the establishment of populations. An example is the species' ability to acclimatize to diverse environments that include temperate and tropical ecosystems, particularly those modified by man, which benefited its expansion. However, the development of cattle raising in America could have contributed due to the Cattle Egret consumes a significant quantity of invertebrates associated with the grazing zones (Franchimont 1986), even getting a benefit from the machinery that removes soil in the fields (Mukherjee 2000).
Another characteristic that could facilitate the Cattle Egret’ arriving to the Americas is its flight capacity, with the air currents from the Atlantic Ocean that could favour that the species crossed from Africa to America (Massa et al. 2014). Even it is possible that more individuals could continue recolonizing the Americas. This capacity could have contributed to the rapid dispersion throughout the Antilles to the coasts of Florida in a relatively short time (30 years), and it is possible that this flow of individuals among islands continues today.
Although the invasion process in the Americas was relatively fast, pulses of higher dispersion rates are observed, particularly between 1950s to 1980s, when the species covered most of its current range. The same rate of geographic invasion process can be observed in ecological terms; for example, from the 1930s to 1970s, the species continuously colonized new environments. However, it is noteworthy that from 1980s to the present day, the rate and magnitude of the environmental occupation towards new climates has been limited already, and it is not comparable with the speed of dispersion in the previous decades. This suggests that, just as in geographic space, the species has also reached its environmental limit, or in other words, it has reached a climatic equilibrium (Svenning and Skov 2004; Varela et al. 2009).
The geographical boundaries are particularly clearly defined in North America since it has been maintained to the 53° N latitude in the north of Canada and the US border in the last decades. However, such geographical limits also correspond to the environmental space, where the extreme minimum thermal limits have prevented its permanency along the year and probably promoted its migratory behavior in this region (Dunn and Alderfer 2011, Sibley 2014). It should be noted that across the western and eastern US coasts the Cattle Egret is present year-round, which also could suggest that the coastal ecosystems, as well as regions with lower extreme temperatures in the winter, might contribute to its annual permanence.
Due to the ease of acclimatization to different ecosystems and the dispersal capacity of the Cattle Egret, it has become a relatively common species throughout the Americas. However, its invasion also implies new interactions with other native species that could have significant ecological consequences (Donald 1969, Burger 1978). For example, the species is commonly observed nesting with other native egret species (e.g., Ardea alba, Egretta thula, E. tricolor, E. rufescens, E. caerulea, Nycticorax nycticorax, Cochlearius cochlearius) that could potentially compete for resources (Burger 1978, Rodgers 1987, Petry and Fonseca 2005, Ureña-Juárez 2015, Abdeldjalil 2019, Mera-Ortíz 2021). Besides, due to their abundance at the nesting sites, the presence of the Cattle Egret colonies can impact soil chemical changes; for example, hyperfertilization, which in turn can cause adverse effects on the surrounding vegetation (Denis-Ávila et al. 2019).
Given the current increase in invasions of exotic species into new regions throughout the world, mainly promoted by human intervention, it is essential to analyze the invasion processes and the maintenance of natural populations, particularly considering the threats that pose to the conservation of the natural systems. It is outstanding how tools like geographic information systems, the availability of biological and climatic information, and new software for ecological analyses is making this task easier.
We thank Alexander Peña-Peniche and Felipe Toro-Cardona for their support in data analysis and to the group of the Laboratorio de Bioclimatología, INECOL, A.C. for their valuable comments on the first versions of this manuscript.
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
The authors have no relevant financial or non-financial interests to disclose
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Trinidad-Domínguez C. D. and Mota-Vargas C. Trinidad-Domínguez CD and Mota-Vargas C. wrote the first draft of the manuscript and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript