Bird species detection by an observer and an autonomous sound recorder in two different environments: Forest and farmland

Abstract

Birds are commonly used as bio-indicators of the quality of environments and the changes to them. Therefore, ecologists put a lot of effort into the monitoring of their population trends. One of the methods used for bird population monitoring is autonomous sound recording. Current studies provide inconsistent results when the number of detected species by autonomous sound recorders was compared with that delivered by an observer. In our study, observers counted birds using a point-count method at 64 random points in forest and farmland. At the same points, autonomous sound recorders recorded the soundscape four separate times (including counting by observer period) and the species present in the recordings were later identified by observers in the lab. We compared the number of species detected by simultaneous observations and recordings, as well as the number of species detected by recorders during four different surveys. Additionally, we calculated the Sorensen index to compare the species composition during different surveys at the same point. We found that observers detected more species than autonomous sound recorders. However, differences in the number of detected species were habitat dependent–observers detected more species than recorders in farmland, but not in the forest. When the time for recording was doubled, recorders were more effective than observers during a single survey. The average Sorensen index between the four repeated surveys performed by autonomous sound recorders ranged from 0.58 to 0.67, however we did not find significant differences in the number of species detected during different surveys conducted at the same point. Our study showed that 10-minutes sampling from the same point gives various species composition estimates but not species richness estimates between different surveys. Therefore, even when recorders detect less species than observers during the simultaneous surveys, increasing the survey duration of recorders may alter this difference. The use of autonomous sound recording for monitoring bird populations should be promoted, especially in forest habitats, as this technique is easier to standardise, eliminates many errors observed in the traditional point-count approach, enables conducting survey during adverse field conditions and delivers more reliable results for the majority of the species.

Farmland bird assemblages exhibit higher functional and phylogenetic diversity than forest assemblages in France

“…Lastly, a caveat of this study is that we did not account for imperfect detection of species in our analyses, a factor that may depend on the habitat type in which the survey is conducted. For instance, in a recent study, Kułaga and Budka (2019) compared the number of bird species detected through human observers (in the field) and recordings by autonomous sound recorders (manually analysed by observers in the lab) within two different habitats ‐forest and farmland‐ in Poland. They reported that observers detected more species than recorders in farmland, but not in the forest (Kułaga & Budka, 2019).…”

Autonomous sound recording outperforms human observation for sampling birds: a systematic map and user guide

“…Even non‐experts can attain high accuracy levels when using automated species classification methods (Goyette et al. ), and sound recordings are easier to process with little ornithological experience, thus increasing the number of available surveyors (Kułaga and Budka , Wheeldon et al. ).…”

“…In more heterogeneous montane habitats, McGrann and Furnas (2016) detected only 1% of birds just visually and in forest, Darras et al 2018b detected only 4% of birds just visually. Moreover, visual detections mostly concern birds flying over the sampling point, which have large ranges and are relatively unrelated to the sampled location (Kułaga and Budka 2019). In habitats where vegetation obstructs the observers' sight, the low proportion of visual detections is primarily due to visual ranges being much shorter than acoustic ranges.…”

“…Here, we depict updated results of the same meta-analysis, which now includes four new studies and one that was previously not considered (Campbell and Francis 2011, Hingston et al 2018, Kułaga and Budka 2019, Wheeldon et al 2019 in Fig. 3.…”

“…Funds for taxonomic experts can be optimized to assign them only to processing or reviewing recordings, or even postponed until funds become available. Even non-experts can attain high accuracy levels when using automated species classification methods (Goyette et al 2011), and sound recordings are easier to process with little ornithological experience, thus increasing the number of available surveyors (Kułaga andBudka 2019, Wheeldon et al 2019). Moreover, data can be sent to ornithologists or accessed online from anywhere (see, for example, BioSounds, Fig.…”