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

observers [32]. Therefore, with no distance limitation, experienced human-observers should detect more or the same number of species as recorders, because both techniques detect the same vocalising species, whereas observers have the additional opportunity to detect vocalising species from a further distance and all silent species that are within visibility range of the observer [32]. Moreover, difference in the number of species detected by recorder and observer should be greater in open than in closed habitats, since that chance for visual detec- tions only is higher in the former [23]. A lower number of species detected by human-observer than recorder within the unlimited distance suggests that the observer has overlooked some vocalising species during the field survey. The proportion of overlooked detections by observer in the field to the total number of species detections recorded by recorder can be used as a measure of the skills of observer. This way studies involving observers of varying skills could be compared. To effectively accomplish the monitoring of bird biodiversity, it is important to survey birds in the same locations, using the same method for many years [45]. Acoustic monitoring without distance limitation fulfils these assumptions. Data are collected in a highly standard- ised way: the same equipment and recording quality, constant acoustic properties of environ- ment at particular recording sites, constant within-species detection range (but diverse between-species). In this way the acoustic monitoring generates a presence-absence indicator,

which enables the tracking of environmental changes in a simple and reliable way, like in Aquatic warbler Acrocephalus paludicola, in which presence and abundance of singing males allows to rank the habitat quality and its changes [49]. Another issue is whether certain species are overlooked using acoustic recorders when conducting the biodiversity monitoring. In our study we found 20 species that were undetected by the recorder (S4 Table). Seven of them were birds of prey, which have broad territories and, excluding Montagu’s harrier, breed out- side the meadows. Another six species do not breed in meadows (herons, terns, cormorants, crossbills), five of them were only observed flying over the study area. Two other species, unde- tected by the recorder, (Ruff sandpiper Calidris pugnax and Green sandpiper Tringa ochropus) were typical spring migrants. Thus, the recorders only missed five species (in total 10 detec- tions: European serin Serinus serinus, Partridge Perdix perdix, Stock dove Columba oenas, Tur- tle dove Streptopelia turtur, River warbler Locustella fluviatilis) that breed in meadows and are important from the perspective of meadow bird biodiversity. On the other hand, field observ- ers did not detect five species whose vocalisations were recorded by the recorder. However, it was just seven detections belonging to five species (S3 and S4 Tables). None of them breeds in the meadows in Poland. Table 2. (Continued) Species (English name) Species (Latin name) N Obs. 50 m Obs. 100 m Obs. unlimited Great-spotted woodpecker Dendrocopos major 15 $ $ $

Eurasian blue tit Cyanistes caeruleus 15 $ $ $ Summary $ 12 22 27 # 0 3 20 " 36 23 1 #–detection rate is significantly lower in comparison to the detection by recorder; "–detection rate is significantly higher in comparison to the detection by recorder; $–no statistical significant difference in detection rate in comparison to detection by recorder. Bonferroni correction for multiple comparisons was applied (�a� ¼ a n, where n is the number of tests). N–number of surveys during which species was detected at recording site independently of the method of detection. For exact results of Wilcoxon two-related samples tests see S1 Fig. https://doi.org/10.1371/journal.pone.0266557.t002 PLOS ONE Acoustic bird biodiversity monitoring in agricultural meadows PLOS ONE | https://doi.org/10.1371/journal.pone.0266557 April 8, 2022 10 / 16