Comparing the sampling performance of sound recorders versus point counts in bird surveys: A meta‐analysis

Abstract

Autonomous sound recording is a promising sampling method for birds and other vocalizing terrestrial wildlife. However, while there are clear advantages of passive acoustic monitoring methods over classical point counts conducted by humans, it has been difficult to quantitatively assess how they compare in their sampling performance. Quantitative comparisons of species richness between acoustic recorders and human point counts in bird surveys have previously been hampered by the differing and often unknown detection ranges or sound detection spaces among sampling methods. We performed two meta‐analyses based on 28 studies where bird point counts were paired with sound recordings at the same sampling sites. We compared alpha and gamma richness estimated by both survey methods after equalizing their effective detection ranges. We further assessed the influence of technical sound recording specifications (microphone signal‐to‐noise ratio, height and number) on the bird sampling performance of sound recorders compared to unlimited radius point counts. We show that after standardizing detection ranges, alpha and gamma richness from both methods are statistically indistinguishable, while there might be an avoidance effect in point counts. Furthermore, we show that microphone signal‐to‐noise ratio (a measure of its quality), height and number positively affect performance through increasing the detection range, allowing sound recorders to match the performance of human point counts. Synthesis and applications. We demonstrate that when used properly, high‐end sound recording systems can sample terrestrial wildlife just as well as human observers conducting point counts. Correspondingly, we suggest a first standard methodology for sampling birds with autonomous sound recorders to obtain results comparable to point counts and enable practical sampling. We also give recommendations for carrying out effective surveys and making the most out of autonomous sound recorders.

Use of autonomous audio recordings for the rapid inventory of birds in the white-sand forests of the Peruvian Amazon

“…Although Darras et al. () found that height (~ 20–26 m) was a positive predictor for alpha richness in recorder species detection, the minor positioning differences in our study were less than local differences in altitude between sampling sites, and were therefore unlikely to have an effect on detection distances. Blake () found that tropical birds are most active from a half hour before sunrise until three hours after sunrise, and that species detection does not vary significantly during this period.…”

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

“…When less vocal birds are important, combining both methods can increase the chances of detection of relatively silent birds, even though this can also be achieved by processing longer duration recordings with automated detection methods (see 4.1 in Darras et al. ). Using both methods has been recommended for surveying rare bird species at risk (Holmes et al.…”

“…A recent meta‐analysis found no detectable difference between both methods in terms of species richness (Darras et al. ).…”

Spatio‐temporal scaling of biodiversity in acoustic tropical bird communities

“…In the current study, we used a probability of > 0.9 for counting a species detected, thus limiting ourselves to the 'sure species'. Adding to the favor of automated methods, a recent meta-analysis found a likely imprint of bird avoidance of the human observer in on-site point counts (Darras et al 2018). > 0.5, we would include more of the background species, reaching an average of 0.79 species per minute as compared to 0.24 species for the 0.9 threshold, i.e.…”

“…The first one relies on characterizing the soundscape for each time and frequency unit by summary indices (Buxton et al 2018). Of the two, the first one -soundscape analysis using indices -has been proposed as the most suitable tool to monitor the general state of habitats (Burivalova et al 2018, Buxton et al 2018, and indeed, acoustic indices have been shown to be closely related to the diversity and abundance of biological sounds across local recordings (Buxton et al 2018, Darras et al 2018). Here, the conversion of the sound data can be achieved by experts, semi-automated algorithms or machine learning techniques such as deep learning (Hill et al 2018, Ovaskainen et al 2018, Stowell et al 2018.…”