The response of mixed-species bird flocks to anthropogenic disturbance and elevational variation in southwest China

Abstract

Avian mixed-species flocks (MSFs) are an important example of species interactions threatened by the biodiversity crisis. They are found throughout the world in forested habitats but are generally reduced in size or frequency by human disturbance. In southern China, a unique MSF system is led by several species of closely- related fulvettas (Alcippe morrisonia, A. hueti, and A. davidi). Our objective was to understand how this system is distributed across elevational gradients, especially moving west into the Hengduan Mountains, and how it responds to human disturbance. We sampled leadership and composition of 375 MSFs over 2 yr in and surrounding 5 nature reserves in Guangxi Zhuang Autonomous Region and Yunnan Province, with transects placed in different land use types—protected forest, buffer areas, or agriculture—and ranging from 400 to 3,200 m. We also sampled birds outside of MSFs. We found MSFs led by fulvettas (A. fratercula and A. davidi) in forests across the region up to 2,900 m. Elevation was not a significant influence on MSF size or prevalence in models that also included land use. We found that MSFs were encountered at only one-third of the frequency in agriculture as in forest, and had strongly different composition and leadership. Although MSFs in buffer areas were more similar to those in forest, birds in buffer areas had lower flocking propensity, different flock leaders, and less complex social networks. In particular, buffer transects that were seeded pine had low numbers of fulvetta-led MSFs and forest specialists in MSFs. In the future, it is important to understand which vegetational characteristics allow MSFs, particularly fulvetta-led MSFs, to persist in buffer habitats.

Seasonal variation in community composition and distributional ranges of birds along a subtropical elevation gradient in China

“…We extracted body mass and dietary groups (insectivore, omnivore, plant‐seed eater and frugivore/nectarivore) from the Elton traits 1.0 database (Wilman et al., 2014). We extracted flocking status (flocking species vs. non‐flocking species) from Birds of the World (Billerman et al., 2020) and an empirical database of mixed‐species bird flocks in southwest China (Zhou et al., 2019). We classified two categories of forest dependency (forest birds: medium and high forest dependency; and non‐forest birds: low forest dependency) provided by BirdLife International Data Zone (BirdLife International, 2020).…”

Habitat alteration modifies the structure and function of mixed‐species flocks in an Andean landscape

“…The cluster coefficient marginally increased in areas with more open vegetation. This increase could be explained as an effect of the constant interaction among a reduced set of species compared to the more species‐rich flocks of forests (Zhou et al, 2019). Flocking species can be highly visible to predators when foraging in more open areas; thus, an increase in clustering among species in a flock could be beneficial for decreasing predation risks (Thiollay, 1999).…”

“…The interaction networks were constructed using the “group gambit” method (Franks et al, 2010), which defines all the individuals of a group of animals observed at any given time as associated (Whitehead & Dufault, 1999). This method assumes interactions among all the species that were part of a flock and is a common method employed in networks of mixed‐species flocks (Mokross et al, 2014; Montaño‐Centellas, 2020; Zhou et al, 2019). To build weighted networks, we used the frequency of co‐occurrence of species in flocks within a transect during each sampling period.…”

“…The cluster coefficient marginally increased in areas with more open vegetation. This increase could be explained as an effect of the constant interaction among a reduced set of species compared to the more species-rich flocks of forests (Zhou et al, 2019). Flocking…”

Seasonal dynamics of flock interaction networks across a human‐modified landscape in lowland Amazonian rain forest

“…For each subsampled network, we calculated five metrics at both the species and flock level, largely following Mokross et al (2014) and Zhou et al (2019): (1) species richness, (2) mean normalized degree, (3) mean weighted degree, (4) skewness, and (5) the global clustering coefficient (hereafter, clustering). Degree is the sum of the number of interspecific connections for a given species, which we normalized by dividing by the number of available species ( n − 1), before averaging to obtain a single overall estimate for the network.…”