Age-dependent observer effects in Austrian long-term monitoring data of common breeding birds

Capsule Over time, the total number of bird species and individuals recorded on bird surveys decreased for older observers, whereas they increased for young observers.Aims We hypothesized that observer age and sex influence long-term bird survey results, with older observers recording progressively fewer birds, and with women performing better than men. We also assumed that the habitat in which the surveys are conducted plays a role in these effects.Methods We used data from 99 sites and a 20-year period of the Austrian Common Bird Monitoring Scheme. We calculated the total number of species and individuals recorded at each site and year. We fitted linear models for each individual site and for both variables, with count year as an independent variable. We then used the resulting slope values in linear mixed-effects models, with observer age, observer sex, habitat and start year as explanatory variables. Finally, we excluded birds with high-pitched songs from the dataset and re-ran the models.Results Observer age significantly influenced the linear regressions’ slopes of the total number of species and individuals. Young observers produced slight increases, while older observers produced slight decreases. Female observers tended to show increases up to a greater age than males, but this result was not significant, probably due to the small sample size for females. Habitat had no significant influence. When excluding birds with high-pitched songs, the effect sizes approximately halved for both dependent variables.Conclusion We found proof of observer effects in the analysed data. In relation to age, we concluded that there is an optimum curve, with middle-aged observers causing neither positive nor negative bias in long-term counts. We discuss how learning effects, hearing impairment, other ageing effects or sex-related differences could influence our results. Lastly, we address several options for dealing with observer effects.