Activity patterns of species are related to their physiology, their behaviour and the environment and can change in response to different factors, such as interactions between species. Bird species, typical of the understorey, show morphological and ecological similarities and must thus have some mechanism of ecological separation, such as temporal niche partitioning. The objective of this study was to provide information about activity patterns and activity overlap of bird species typical of the understorey. We expected temporal niche partitioning between ecologically-similar species. We placed camera traps in 29 sampling points in a high evergreen forest in the southeast of Mexico between 2011 and 2013. All species were mainly diurnal and, contrary to what we expected, there was temporal partitioning between tinamids, but not in galliforms and columbiforms. The degree of activity overlap might reflect a solitary or group lifestyle of the three sets of species, as well as shared behavioural preferences and similar adaptations. These results contribute to our knowledge of the basic biology and behavioural ecology of birds of the understorey.

Activity overlap, camera traps, Columbidae, Cracidae, Tinamidae

The activity patterns of species, that is, the time invested in rest and movement, are related to their circadian rhythm, physiology and behaviour, as well as to the environment where they develop (Ashby 1972; Kronfeld-Schor and Dayan 2003). Each species depends on these factors to adapt to the amount of light and darkness, to generate adaptive responses to environmental and their own changes and to inter- and intra-specific interactions (Bennie et al. 2014). Two species that use similar resources may avoid times when their competitors are active and minimise direct confrontation when resources are limited (Schoener 1974). The effects may be more intense when the species have taxonomic or morphological proximity (Schoener 1974; Kronfeld-Schor and Dayan 2003).

Various groups of bird species perform most of their activities at the ground level of tropical forests of the Neotropics and some of these species are morphologically, taxonomically and ecologically similar. For example, species from the orders Galliformes, Tinamiformes and Columbiformes forage at ground level, feed on fruits, seeds, leaves, shoots and arthropods, whose proportions vary with species and item availability and show a high diet overlap (Muñoz and Kattan 2007; Baur 2008; López et al. 2014). They nest and rest both on the ground and in the tree stratum and show a high overlap in the use of both strata (Baur 2008; López et al. 2014). Therefore, possible competitive interactions may be stronger between these similar species (Brooks et al. 2004; Roncal et al. 2019).

One of the regions with the most biodiverse avifauna in Mexico is Los Chimalapas, in the northeast of the State of Oaxaca. Bird diversity studies have been carried out in this region, but there are only a few ecological studies on ground-dwelling birds (Pérez-Irineo and Santos-Moreno 2017). This is a zone of difficult access with dense vegetation cover, which complicates the study of these species in such conditions. Furthermore, bird species typical of the understorey tend to be cryptic, shy and elusive (López et al. 2014; Negret et al. 2015). Camera trapping is a regular and widely-used method in bird studies and it is a useful alternative in topographically heterogeneous areas, such as Los Chimalapas.

The objective of this study was to provide information about the activity patterns and activity overlap between bird species present in the understorey of an evergreen forest. We expected temporal niche partitioning between morphologically, taxonomically and ecologically similar species as a mechanism to minimise the risk of competition. This information contributes to our understanding of how bird species respond to the presence of competitors in a dense tropical environment.

A total sampling effort of 8,529 days∙trap resulted in 5,269 photographs, 862 of which were classified as independent records of birds from 10 orders, 15 families, 22 genera and 27 species. Of the total number of records, 93.55% were obtained during the day, 6.3% during twilight (including sunrise and sunset) and 0.15% during the night (Fig. 1).

Figure 1. 

Distribution of the number of records of birds obtained by camera trapping in a 24 h cycle in the understorey of an evergreen forest in Los Chimalapas, Oaxaca, Mexico.

Of the total number of species, we recorded 10 species with 11 or more records. All these species were diurnal, with two activity peaks, one in the morning (06:00 – 10:00 h) and one in the afternoon (15:00 – 18:00 h; Fig. 1). Crypturellus boucardi (Slaty-breasted Tinamou), Tinamus major (Great Tinamou; Fig. 2a), Crax rubra (Great Curassow; Fig. 2b), Geotrygon montana (Ruddy Quail-dove; Fig. 2c) and Formicarius moniliger (Mayan Antthrush) were also active at dawn and dusk (Fig. 1).

Figure 2. 

Activity patterns of some Tinamiformes (a), Galliformes (b) and Columbiformes (c) with similar size and ecological characteristics, as well as between pair species of larger body size (d) in the understorey of an evergreen forest in Los Chimalapas, Oaxaca, Mexico.

Only two species had enough records in both seasons which allowed a comparison, C. rubra and T. major. The differences in activity between seasons were not significant (W dry-rainy 2012 = 3.64, P = 0.16 for C. rubra; W dry-rainy 2011 = 5.40, P = 0.06 and W dry-rainy 2012 = 0.81, P = 0.66 for T. major), except in 2011 for C. rubra (W dry-rainy 2011 = 9.81, P = 0.007). Considering this, we carried out the following analyses without considering seasonality.

Of the ten species with 11 or more records, we made comparisons for nine of them. We excluded F. moniliger because this species shows taxonomic and ecological differences with respect to the other species: it was the only representative of the order Passeriformes and is primarily insectivorous (Table 1).

For Galliformes and Columbiformes, there was no temporal niche partitioning between species: the pairs C. rubra and P. purpurascens, L. plumbeiceps and G. montana and C. rubra and T. major showed a high degree of overlap (Δ₁ = 0.79, 0.81 and 0.82, respectively; Table 2). For Tinamiformes, the pair C. boucardi and C. cinnamomeus showed the lowest degree of overlap (Δ₁ = 0.48) and the rest of the pairs showed a moderate degree of overlap (Table 2).

In general, the activity patterns of birds in Los Chimalapas are consistent with behavioural studies on cracids (Fernández-Duque et al. 2013), tinamids (Brooks et al. 2004; Kuhnen et al. 2013) and doves in other areas, where there is higher activity during the morning (06:00–08:00 h) and the afternoon (17:00–18:00 h; Dias et al. 2016; Roncal et al. 2019). This bimodal activity of birds might be due to a shared behaviour to avoid high temperatures at noon, which is similar to what has been observed in previous studies (Srbek-Araujo et al. 2012; Fernández-Duque et al. 2013; Dias et al. 2016) and because they continually descend to the ground in search for food during the first hours of the day and the afternoon, even those who nest in layers above the ground level (Skutch 1963a, b; López et al. 2014).

Our results indicate possible temporal niche partitioning between species of Tinamiformes, since the level of activity overlap was intermediate (e.g. T. major and C. cinnamomeus with Δ₁ = 0.63) or low (e.g. C. boucardi and C. cinnamomeus Δ₁ = 0.48). However, our results should be taken with caution due to the small amount of data for C. cinnamomeus (11 records) and the scale used in the analyses.

There are variations in temporal niche partitioning in some Tinamiformes. In some cases, the activity overlap is high (> 0.80), between the tinamous Tinamus guttatus (White-throated Tinamou) and Crypturellus cinereus (Cinereous Tinamou) (Roncal et al. 2019), as well as amongst T. major, Crypturellus variegates (Variagated Tinamou) and Tinamu tao (Gray Tinamou; Brooks et al. 2004); in others, it is intermediate (Δ₁ = 0.69), between the tinamous Crypturellus undulatus (Undulated Tinamou) and C. cinereous (Roncal et al. 2019); and in others, it is low (Brooks et al. 2004). On the other hand, Galliformes and Columbiformes showed no temporal niche partitioning, contrary to what we expected and had a high level of overlap (Δ₁ ≥ 75%). Previous studies have also shown high activity overlap between pairs of bird species that are ecologically similar, including the pair of galliforms Psophia leucoptera (Pale-winged Trumpeter) and Mitu tuberosum (Razor-billed Curassow; Roncal et al. 2019).

We consider the similarities between species as the most parsimonious factor for explaining our results. Galliformes and Tinamiformes, such as P. purpurascens, T. major, Meleagris ocellata (Ocellated Turkey) and Ortalis vetula (Plain Chachalaca), exhibit high diet overlap and shared behavioural preferences for resting and foraging times as a result of having similar adaptations (Baur 2008; López et al. 2014). These ecological similarities can generate high activity overlap between species.

Galliformes and Columbiformes are more sociable and form family groups (López et al. 2014; Pérez-Irineo and Santos-Moreno 2017), as observed in Los Chimalapas. Therefore, they may have fewer negative interactions with members of other similar species. In contrast, Tinamiformes are more solitary species, except in the reproductive season (Kuhnen et al. 2013; Brooks 2015); for example, C. soui has been observed to actively defend its territory (Brooks et al. 2004; Brooks 2015). Although many behavioural aspects of tinamiform species are unknown, these types of behaviour may occur in the species recorded in Los Chimalapas, favouring temporal niche partitioning (Brooks 2015).

Future studies will help to better understand if our results reflect niche separation between species of Tinamiformes or if there are factors other than competitive interactions. More studies on the activity patterns of bird species with ecological similarities will also help to determine whether what we observed in the present study is a generalised pattern.

The Instituto Politécnico Nacional de México provided funding for the implementation of this project (grant SIP-20180613). We thank the representatives of the communities of Los Chimalapas for allowing the use of their facilities, as well as all the volunteers who helped with the fieldwork. R. Pablo contributed with the identification of the species. IDEAWILD donated some of the camera traps used in this study. We thank two anonymous reviewers for their time and effort in improving the manuscript.