Acoustic approach as an alternative to human-based survey in bird biodiversity monitoring in agricultural meadows

S4 Tables). From the 48 most common bird species that were analysed, we found a statistically indistinguishable difference in the detection rate by recorder and: (1) observer surveying birds within 50 m radius in the case of 12 species, (2) observer surveying birds within 100 m radius in the case of 22 species, and (3) observer surveying birds in unlimited radius in the case of 27 species (Table 2). The detection rate by a recorder in comparison with a human-observer sur- veying birds was: (1) significantly higher in the case of 36 species and was not lower for any species when 50 m radius was considered; (2) was significantly higher in the case of 23 species and lower in the case of three species when 100 m radius was considered; (3) significantly higher in the case of one species (Black-headed gull Chroicocephalus ridibundus) and lower in the case of 20 species when unlimited radius was considered. For more details see Table 2 and S1 Fig. Discussion Bird biodiversity estimation in unlimited detection radius The results of our study show that in open habitats an acoustic recorder detects fewer bird spe- cies compared to a highly skilled observer when unlimited radius is applied by the human- observer and more species when the observer surveys birds within a fixed radius (50 m or 100 m in our case). The difference between the methods may arise from visual detections of addi- tional species during the field survey by the observer [47] and due to the various areas sampled by recorder and observer [20,30]. Open habitats should strongly support visual detections by

observers [23]. Thus, birds that are at a great distance from a survey point, or species which are Fig 2. Number of detected bird species by various detection methods. Graphs show (a) all bird species, (b) songbird species, (c) farmland bird species and (d) meadow bird species by four detection methods: (1) recorder, (2) observer surveying birds within 50 m radius, (3) observer surveying birds within 100 m radius, and (4) observer surveying birds in unlimited radius. Boxplots show the median, interquartile range, and outliers. https://doi.org/10.1371/journal.pone.0266557.g002 PLOS ONE Acoustic bird biodiversity monitoring in agricultural meadows PLOS ONE | https://doi.org/10.1371/journal.pone.0266557 April 8, 2022 7 / 16