Fig 1.
Recording sites monitored during summer Jan-Feb 2024 at Hidden Vale Research Station (hereafter, Hidden Vale) and Mount Grandchester Conservation Estate (hereafter, Grandchester).
(a) location of the study area within Southeast Queensland, Australia (circled); (b) Southeast Queensland (SEQ) Regional Corridor, with the Little Liverpool Range (circled) linking the Main Range National Park and the Great Eastern Ranges; (c) boundaries of the two study areas and locations of Autonomous Recording Units (ARUs). Republished from Queensland LiDAR Data – Ipswich 2019 Project under a CC BY license, with permission from the Department of Natural Resources and Mines, Manufacturing and Regional and Rural Development, original copyright The State of Queensland 2025.
Table 1.
Summer season summary of nocturnal sonotype richness and corresponding acoustic indices for Grandchester and Hidden Vale reserves. Acoustic activity (ACT), acoustic evenness (AEI), spectral entropy (Hf), soundscape saturation (Sm). Bio = biophony (all vocal sonotypes); Ins = insects; Bir = birds; Fro = frogs; Mam = mammals; Filt = filtered audio files (0.3-20kHz); Raw = raw audio filles (0-20kHz).
Table 2.
Generalized Linear Models of annotated sonotype richness versus acoustic indices calculated from nocturnal soundscapes across 12 sites, using predictor variables as main effects (ACT, AEI, Hf, Sm) and interaction effect (equal-weighted sum of z-scored indices), with response variables total biophony richness and insect richness. Indices are acoustic activity (ACT), acoustic evenness (AEI), spectral entropy (Hf) and soundscape saturation (Sm). Datasets: filtered audio files (0.3-20kHz) and raw audio filles (0-20kHz). Biophony = the sum of annotated sonotypes from insects, frogs, birds and mammals. Insect = the sum of annotated sonotypes, predominantly from crickets, grasshoppers, katydids, and locusts.
Fig 2.
Non-metric multidimensional scaling of (a) Insect sonotype richness (Jaccard binary) and (b) indices interaction (acoustic activity (ACT), acoustic evenness (AEI), spectral entropy (Hf), soundscape saturation (Sm)) scaled composite index with Ružička distance) comparing Hidden Vale and Grandchester.
Points are labelled by survey site and coloured by reserve; ellipses were drawn using the within-reserve mean dissimilarity distance. Stress values are reported for each ordination, with small values indicating closer similarity.
Table 3.
Between and within similarity metrics of Hidden Vale and Grandchester from PERMANOVA and PERMDISP based on insect richness (sonotype count presence on five non-consecutive nights in summer season) and indices interaction (acoustic activity (ACT), acoustic evenness (AEI), spectral entropy (Hf), soundscape saturation (Sm)) calculated over the same audio clips. PERMANOVA R² and PERMANOVA p-value quantify centroid separation; PERMDISP p-value tests equality of multivariate dispersion, whereas non-significant indicates similar spread; Within Grandchester and Within Hidden Vale are mean pairwise distances among sites within each reserve, with lower values indicate more homogeneous.