Social Information About Predators In Two Mixed-species Bird Flock Systems

Mixed-species flocks of birds(MSFs)are a subset of avian assemblages worldwide and particularly prevalent in terrestrial forested habitats.Birds in MSFs usually travel together throughout the day;in contrast,birds gather together temporarily when mobbing predators.Due to the many species that engage in MSFs and mobs,these two kinds of communities provide good model systems to study how social information influences species interactions.Specifically,it is thought that some ‘nuclear species' are information providers,and other species listen to them and follow them.Yet these communities are complex,so the functional traits that make species nuclear are not fully understand.Further,the communities change seasonally,and it not clear how this influences information production and the role of species.Finally,the kinds of vocalizations that may be listened to by following species has not been explored,beyond alarm and mobbing calls.Given that,I asked three key questions in this study: 1)how social information about different kinds of predators is associated with the role of species(nuclear vs.follower),and changes seasonally(Chapter Two);2)how mobbing incidence,propensity and behavior are affected by external factors(temporal variables,as well as the conspecific and heterospecific audience)and intrinsic factors(the functional traits of the species involved;Chapter Three);3)to what extent following species can intercept information about predation encoded in the contact calls of nuclear species(Chapter Four).Beyond the theoretical interest of these questions for the understanding of community assembly,this information will be useful to plan strategies for community conservation that incorporate species interactions.To address the first key question,I investigated MSF composition across the annual cycle in a tropical seasonal rainforest near the Chinese-Vietnamese border.Simultaneously,I exposed birds to different predation risks when they were inside MSFs and outside of them,by simulating both a moving and a stationary predator.The identity of responding species,and their latency to response,duration of alarm call or mobbing call,minimum distance to model and aggressiveness were measured.In this chapter,I analyzed the differences between species in their vocal responses,and how vocal response was affected by seasonality.I found that MSFs were much less frequent in the breeding season,with fewer individuals of the nuclear species,David's Fulvetta(Alcippe davidi),participating,although the diversity of other species remained stable.Fulvettas were the dominant producer of alarm-related information both to the moving and stationary stimuli in MSFs,and were also among the most active mobbers to stimuli presented outside of MSFs.In the breeding season they tended to call less to the moving stimulus,and substantially fewer individuals responded to the in-flock stationary stimulus.This change may be due to breeding pairs dropping out of MSFs at this time.Other species increased their own information production in the breeding season,particularly when confronted with the stationary predator stimuli(inside and outside of MSFs),perhaps due to their increased investment in offspring.Yet even during the breeding season,David's Fulvetta remained the highest producer of information about predators in MSFs.To address the second key question,I used the mob-experiment data collected for the first question.I analyzed how external factors and intrinsic factors affected mobbing incidence,propensity and behavior.Results suggested that the incidence of mobbing was higher in the breeding season,partially resulting from an increase in the abundance of both migratory and resident species.Birds were more likely to be vocal in April than in later months.Similarly,birds were more likely to have high aggression in April and May than in June and July.The presence of conspecifics and heterospecifics affected how vocal mobbers were,and their level of aggression.Specifically,mobbers tended to be more vocal when there were more conspecifics and when they were the first species to arrive,and mobbers were more aggressive when David's Fulvetta was being highly aggressive.Mobbing propensity increased with species abundance but had no correlation with log-transformed body mass in a phylogenetically-corrected analysis.To address the third question,I conducted a playback experiment in a subtropical forest of north-central Florida,United States,broadcasting five different playback treatments that were associated with different predation risks.Three of these treatments were the vocalizations of the nuclear species(Tufted Titmouse,Baeolophus bicolor): alarm calls,and two types of contact calls associated with a high and a low level of risk,respectively.I classified how birds responded to the playback stimuli,and asked whether this depended on the treatment.The result indicated that most birds engaged in escape or freezing behavior in response to nuclear species' alarm calls,but produced ambiguous antipredation behaviors to the contact calls that carry predation threat,no stronger than their responses to insect calls.I conclude that 1)social information production in MSFs in this tropical seasonal rainforest can be somewhat dynamic across the years but is consistently asymmetric,with the MSF nuclear species important to the whole community;2)mobbing incidence is tightly associated with temporal factors and its characteristics are affected by both the presence of conspecifics and heterospecifics;3)contact calls appear to be a private channel of communication for conspecifics,one difficult to intercept by heterospecifics.