The effect of meteorology on the singing rate of 13 common bird species

Abstract

Birdsong is mainly used by birds to reproduce and defend their territory. Although the

intensity of singing activity follows circadian rhythms, various factors can modify it,

including weather conditions. The present study investigates the effect of 12 basic

meteorological variables on the singing activity of 13 common bird species in

Mediterranean areas during the reproductive period and evaluates the temporal

suitability of bird census to detect the maximum of species. 8 recording devices and the

data from a nearby weather station were used in the study. The registers were grouped

into 5-minute periods during daytime over 36 days. The determination of significant

variables at the community level was performed with a Random Forest analysis. The

variability in the intensity of birdsong for each species was studied using GLMM models.

Air temperature, relative humidity, solar irradiance, wind speed, and atmospheric

pressure, together with the time of the day, were selected as the variables that

significantly regulate avian singing activity. They explain 76% of the variability in the

singing activity of the community. The effects of each variable are species-specific, but

common patterns can be identified. Singing activity is low in extreme temperatures, but

the optimal temperature varies among species. The response to humidity and solar

irradiance are linked to the time of the day, during mornings most species sing the most

when humidity is ~80% and solar irradiance has little effect. Wind speed negatively

affects singing activity while atmospheric pressure affects it positively. Generally, the time

of the day with maximum singing activity is the early morning hours, following circadian

rhythms.

We concluded that the optimal time to conduct bird census is early in the morning, on

days when wind speed is minimum, atmospheric pressure is high, and temperature is

above 11ºC. The high interspecific variability makes it impossible to determine a more

specific condition for temperature, irradiance, and humidity. Therefore, we suggest

conducting replicas of the census, in order to include different atmospheric conditions,

this practice would maximize the detection of the majority of species

13

15