Mikusinski, Grzegorz
- School for Forest Management, Swedish University of Agricultural Sciences
- Polish Academy of Sciences (PAN)
Abstract
Ecoacoustic indices have been proposed as proxies for diversity measures such as species richness, however, their effectiveness remains a subject of ongoing debate. We examined how variation in recording sampling rate and computational parameters influences the strength of the relationship between bird species richness and two widely used ecoacoustic indices: the Bioacoustic Index (BI) and the Acoustic Complexity Index (ACI). We analyzed 5844 one-minute soundscape recordings from the Bia & lstrok;owiez(center dot)a Primeval Forest (Poland), down-sampling them from 192 kHz to 96, 48, and 24 kHz. The ACI and BI were calculated for each recording sampling rate using different configuration settings: seven Fast Fourier Transform (FFT) window lengths and two frequency range settings. We then related bird species richness to the ACI and BI across all combinations of FFT window length, sampling rate, and frequency range. We demonstrated that the relationship between species richness and ecoacoustic indices ranged from significantly positive to significantly negative, depending on the technical parameters applied, with a stronger effect observed for the ACI than for the BI. For both indices, adjusting the analysis frequency range to match the frequency range of bird vocalizations in our study area strengthened the relationship compared to the default settings, and the influence of technical parameters varied among habitats. In conclusion, the effectiveness of the ACI and the BI in representing bird species richness relies on technical parameters. When calculating ecoacoustic indices, particularly the ACI, we recommend adjusting the FFT window length to match the sampling rate of the recordings and the local ecoacoustic conditions. Furthermore, other calculation settings, such as the analysis frequency range, should be adjusted to the vocalisation characteristics of the studies taxa. Finally, we advise against using the ACI and BI without prior testing of their suitability to reflect local biodiversity measures.
Keywords
Acoustic complexity index; Bioacoustic index; FFT window length; Forest disturbance; Frequency range; Sound ecology; Sampling rate
Published in
Ecological Informatics
2025, volume: 92, article number: 103493
Publisher: ELSEVIER
SLU Authors
UKÄ Subject classification
Ecology
Publication identifier
- DOI: https://doi.org/10.1016/j.ecoinf.2025.103493
Permanent link to this page (URI)
https://res.slu.se/id/publ/144682