Environmental and social factors influence chorusing behaviour in a tropical frog: examining various temporal and spatial scales

Many animals use conspicuous display to attract mates, and there should be selection for displays to occur at times and places that maximise the proba bility of mating, while minimising energetic costs and predator attraction. To select the best times for display, individuals may use environmental cu es, the presence of other individuals, or both, but few studies have examin ed these sources of variation in display activity. In this study, we examin ed physical environmental and social factors triggering displays in a tropi cal, terrestrially breeding frog, Cophixalus ornatus. To measure the influe nce of physical environmental conditions on calling activity, we recorded t emperature, rainfall, moon illumination/visibility, humidity, barometric pr essure and intensity of calling activity throughout a breeding season at si x locations along a 560-m transect. The intensity of calling varied daily, seasonally, and at a small spatial scale. Variation in calling activity fro m day to day was large. There was also a strong seasonal trend in calling a ctivity: few males called at the start of the season, activity peaked short ly after the beginning of the season, and then declined linearly from the p eak to the end of the season. There was also consistent variation among sit es along the transect, which may have been due to variations in frog densit y at each site, or to consistent microscale variations in physical conditio ns, or both. After statistically removing consistent local variation among sites, a principal components analysis suggested that a maximum of 35.8% of the variation in calling activity among days was due to factors common to all sites, such as weather, moon illumination, or large-scale social facili tation (e.g. of choruses by other choruses). The remainder of the variation among sites (64.2%) was due to site-specific factors, such as small-scale social facilitation or unmeasured, apparently stochastic effects, such as m icroenvironmental physical factors that do not vary consistently over sites . Regressions of environmental variables on residual calling activity (afte r removing consistent effects of site and season), alone or in combination, accounted for very little of the variation in the number of calling males (maximum 10%). Thus, our data, showing strong seasonal effects and consiste nt variation among sites combined with large amounts of variation in the nu mber of calling males at small spatial scales, suggest that environmental c onditions, such as temperature, rainfall, moon illumination and barometric pressure, which act over large spatial scales, may determine the overall en vironmental envelope within which calling can occur but do not account for most of the variation in the number of calling males on a day-to-day or sit e-to-site basis. Similarly, variations in the number of calling males at sm all spatial scales suggest that social facilitation is a relatively unimpor tant trigger for displays on a large scale in these frogs. On the other han d, our data suggest that social facilitation may have important effects on variation in the number of calling males on a day-to-day and site-to-site b asis. We used playback experiments to assess whether the sound of calling c ould initiate displays. We played either a taped chorus or white noise in a reas where few (zero to two) males were calling. The number of calling male s increased both during and after the chorus stimulus, whereas there was no increase in calling in response to white noise. These data suggest that examining variation in calling activity at small sp atial scales can reveal the sources of variation for the number of calling males, and indicate that, in these frogs, males tend to use the calling of other individuals as a cue to determine when to display.