Besides competition, positive interactions also play an important role in shaping the social structure of mixed-species bird flocks. This study aimed to illuminate the interspecific interactions of competition and facilitation in mixed-species bird flocks. We recorded the foraging behavior and microhabitat use of flocking species in montane coniferous forests of Taiwan under different social contexts. Foraging niche breadth and niche-overlap with other flocking species were compared between individuals inside and outside of mixed flocks. For the three microhabitat variables (foraging locations, vertical strata, and horizontal strata), relationships between niche-overlaps of heterospecific pairs of these flocking species and their corresponding interspecific associations were determined using a simple linear regression. While in mixed flocks, two understory species, Taiwan Fulvetta Fulvetta formosana and Yellowish-bellied Bush-Warbler Horornis acanthizoides, shifted their foraging from shrubs upwards into coniferous trees. Meanwhile, Flamecrests (Regulus goodfellowi) moved downwards vertically within the canopy, and Black-throated Tits (Aegithalos concinnus) spread out horizontally along branches. In addition, Flamecrests applied many more sally-hovers inside of mixed flocks than outside of flocks. All four species are insectivores which might find it more difficult to obtain sufficient food during the colder winters when food resources become scarcer. Therefore, they may be using the increased vigilance afforded by the flock to expand their foraging niches and thus to increase their foraging opportunities inside mixed flocks. Furthermore, niche-overlaps of heterospecific pairs of the 11 common flocking species were positively correlated with their corresponding interspecific associations on all three microhabitat variables. These results indicate that a greater foraging niche-overlap between two flocking species would result in higher coexistence of the two species in mixed flocks. Consequently, facilitative interactions occurred in these mixed-species flocks in addition to competitive interactions.

Fieldwork was carried out in the Tataka (N23.487546, E120.891723), Hehuanshan (N24.162243, E121.287129), and Dasyueshan (N24.255348, E121.008445) areas, Taiwan, during both the non-breeding (September-February) and the breeding (March-August) seasons in 2015 and 2016. Mixed-species flocks were regularly formed from September to February when cold weather prevails. We focused on flocking members of mixed flocks and recorded their foraging behavior and microhabitat use. Extensive trails within each study site were used for searching birds. A two-person team worked together and followed targeted bird species closely for as long as possible to collect data on each individual’s foraging behavior. Different social contexts (mixed-species flock, monospecific flock, solitary, or in pairs) were noted whenever a flocking species was encountered. All the species in a mixed flock were identified, and the number of individuals of each participant species was determined with the aid of binoculars. After the flock disappeared, both observers double-checked and compared the numbers of all species identified in the flock. If the recorded numbers for the individuals of a particular species were not the same between the two observers, we used the larger one, the reason being that one of the observers might have had a better view of the flock or was situated at the right spot to count the birds when they crossed a trail (Chen and Hsieh 2002, Kotagama and Goodale 2004). The proportion of all flocks taken together occupied by each flocking species was then calculated as the flocking frequency of each species; in other words, we calculated a mean proportion across all the data from the 81 flocks (Hutto 1994). We also collected foraging data of flocking species both inside and outside of the mixed flocks. The foraging behavior and categories for microhabitat use of a foraging bird as defined by Remsen and Robinson (1990) were used. The main foraging behaviors included gleaning, hanging, probing, sally-striking, and sally-hovering. Reaching and pecking occurred very rarely; thus, we lumped them with gleaning and probing. We made only one observation from each individual in a mixed flock to maintain the independence of our behavioral data. To achieve this, we always shifted to another species whenever observations of one individual had been completed. In some cases, we observed more than one individual of a particular species if we were able to assure that both individuals were not the same one. For example, we occasionally moved on to another individual of the same species which was located somewhere away from the previously targeted bird, or we switched to an individual of the opposite sex which we could distinguish by its distinctive sex-specific plumage. (For more details, please read the paper, which will be published in the Journal of Avian Biology)