The DOI matches its Crossref record; 79 of the 79 references it cites were found in a registry or an index; none of the 40 sources its claims cite is checked for support yet.
Soundscapes and deep learning enable tracking biodiversity recovery in tropical forests
Open accessAbstract
Tropical forest recovery is fundamental to addressing the intertwined climate and biodiversity loss crises. While regenerating trees sequester carbon relatively quickly, the pace of biodiversity recovery remains contentious. Here, we use bioacoustics and metabarcoding to measure forest recovery post-agriculture in a global biodiversity hotspot in Ecuador. We show that the community composition, and not species richness, of vocalizing vertebrates identified by experts reflects the restoration gradient. Two automated measures – an acoustic index model and a bird community composition derived from an independently developed Convolutional Neural Network - correlated well with restoration (adj-R² = 0.62 and 0.69, respectively). Importantly, both measures reflected composition of non-vocalizing nocturnal insects identified via metabarcoding. We show that such automated monitoring tools, based on new technologies, can effectively monitor the success of forest recovery, using robust and reproducible data.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
In a 43-plot chronosequence in Ecuadorian Chocó, soundscape-derived community composition tracked tropical forest recovery better than species richness, with acoustic indices and a CNN-derived bird community explaining expert-identified vocalizing vertebrate composition at adj-R² = 0.62 and 0.69, respectively. Both automated sound measures also reflected non-vocalizing nocturnal insect communities identified by metabarcoding.
Key findings
- Expert-identified vocalizing vertebrate community composition followed a linear recovery gradient from active agriculture to old-growth forest, while total species richness decreased along the gradient and old-growth-specific richness increased.
- A combination of five acoustic indices explained vertebrate community composition (adj-R² = 0.62), old-growth vertebrate richness (adj-R² = 0.45), and nocturnal insect community composition (adj-R² = 0.42), but had low explanatory power for total vertebrate richness (adj-R² = 0.20).
- The first CNN-derived bird community axis was the single best predictor of expert-derived vertebrate community composition (adj-R² = 0.69), despite the CNN identifying only ~25% of species detected by experts; restricting to 49 species shared between expert and CNN identifications raised explanatory value to R² = 0.85.
- From pastures to old-growth forests, Soundscape Saturation and Entropy of Frequency increased, while Acoustic Complexity and Events/Second decreased; both acoustic indices and CNN-derived composition also tracked nocturnal insect communities, which were predominantly non-vocalizing.
Methods
- The study used 43 plots in the Ecuadorian lowland Chocó: active pastures and cacao plantations, abandoned pastures/plantations regenerating for 1–34 years, and old-growth forest, sampled in October 2021 with one Bioacoustic Recorder per plot recording 2 min every 15 min for two weeks at 44.1 kHz.
- Experts manually identified vocalizing birds (183 species), mammals (3 species, excluding bats), and amphibians (41 species) from selected 2-min files; acoustic indices were calculated from recordings, and a CNN trained on an independent dataset of 401,685 1-min recordings from 55 sites identified 73 bird species in the study plots using a 0.8 score threshold.
- Nocturnal insects were sampled with one autonomous light trap per plot for one night and identified via CO1 metabarcoding, yielding 4,557 BINs from 24 orders; analyses used NMDS with Bray-Curtis distance, linear models, adjusted R², and cross-correlograms to check spatial independence of residuals.
Limitations
- The paper states that the CNN identified only ~25% of species detected by experts, that AI models for vocalizing animals need global improvement and regional validation, and that acoustic indices can saturate at high diversity and did not distinguish late recovery from old-growth communities.
- The paper notes that recorders were placed about 2 m above ground, possibly underestimating canopy species in old-growth plots, and that denser vegetation in old-growth forests may limit sound propagation, though it argues attenuation bias was likely minimal in this system.
- Evident from the text: the design is a space-for-time chronosequence in one tropical region, with acoustic sampling limited to two weeks and light trapping to one night per plot, so generality across seasons, regions, and taxa remains to be established.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 of 40 cited sources checked · 0 supported · 0 flagged
- Findingown data
The community composition, and not species richness, of vocalizing vertebrates identified by experts reflects the restoration gradient.
“We show that the community composition, and not species richness, of vocalizing vertebrates identified by experts reflects the restoration gradient.”
- Also stated in: § Vocalizing vertebrate communities represent the recovery gradient · boosted
“At our study sites in the Choco, Ecuador (Fig. 1), the community of vocalizing vertebrates, as identified by experts, showed a clear gradient along the first axis of a non-metric multidimensional scaling (NMDS) ordination (Fig. 2a).”
Quote in the paperSubstantial claim - Also stated in: § Vocalizing vertebrate communities represent the recovery gradient · boosted
- Findingown data
The richness of old-growth-specific species increased over time.
“In contrast, the richness of old-growth-specific species increased over time (Fig. S1).”
Quote in the paperSubstantial claim - Findingown data
Acoustic Complexity decreased along the recovery gradient from pastures to old-growths.
“As vocalizing vertebrate communities progressed towards old-growth forest conditions, the soundscapes showed increasing Soundscape Saturation and Entropy of Frequency, but decreasing Acoustic Complexity and Events per Second from pastures to old-growths (Fig. 4 and Table 1).”
Quote in the paperSubstantial claim - Findingown data
Events per Second decreased along the recovery gradient from pastures to old-growths.
“As vocalizing vertebrate communities progressed towards old-growth forest conditions, the soundscapes showed increasing Soundscape Saturation and Entropy of Frequency, but decreasing Acoustic Complexity and Events per Second from pastures to old-growths (Fig. 4 and Table 1).”
Quote in the paperSubstantial claim - Findingown data
The CNN model identified only ~25% of species detected by the experts in our data.
“Despite the model identifying only ~25% of species detected by the experts in our data, the model-derived first community axis was the single best predictor for expert-derived community composition.”
Quote in the paperSubstantial claim
- Findingown data
Excluding the 13 vocalizing insect species from the families Tettigoniidae, Gryllidae, Cicadidae present in light traps did not change the axis of species composition (Pearson correlation rho=0.986).
“In further testing, we excluded the 13 vocalizing insect species from the families Tettigoniidae, Gryllidae, Cicadidae, present in our light traps, which did not change the axis of species composition (Pearson correlation rho=0.986), indicating that light trap assemblages predominantly represent non-vocalizing insects.”
Quote in the paperSubstantial claim - Findingown data
Total species richness steadily decreased along the recovery gradient.
“Total species richness steadily decreased along the recovery gradient in line with one study in tropical dry forests47, but in contrast to another study on birds46.”
Quote in the paperNot a substantial claim - Findingown data
The main axis of community composition revealed a linear gradient from pastures to old-growth forests.
“As the main axis of community composition (Fig. 2a) revealed a linear gradient from pastures to old-growth forests, we used linear models in all further analyses as the most parsimonious approach.”
Quote in the paperSubstantial claim - Findingown data
A combination of acoustic indices had high explanatory power for the community composition of vocalizing vertebrates along the recovery gradient (adj. R2 = 0.62).
“Two automated measures – an acoustic index model and a bird community composition derived from an independently developed Convolutional Neural Network - correlated well with restoration (adj-R² = 0.62 and 0.69, respectively).”
- Also stated in: § Sound diversity as measure of biodiversity · plain
“Acoustic indices individually had low explanatory power, but we found that a combination of acoustic indices had a high explanatory power for the vertebrate community composition (adj. R² = 0.62), good explanatory power for richness of old-growth species and nocturnal insect composition, but low explanatory power for total vertebrate richness (Table 1).”
Quote in the paperSubstantial claim - Also stated in: § Sound diversity as measure of biodiversity · plain
- Findingown data
Acoustic indices do not distinguish between late recovery and old-growth communities.
“Despite the potential of acoustic indices as predictors of vocalizing vertebrate diversity, these do not distinguish between late recovery and old-growth communities, as seen by comparing observed versus predicted values in Fig. 2b.”
Quote in the paperSubstantial claim - Findingown data
Soundscape Saturation increased along the recovery gradient from pastures to old-growths.
“As vocalizing vertebrate communities progressed towards old-growth forest conditions, the soundscapes showed increasing Soundscape Saturation and Entropy of Frequency, but decreasing Acoustic Complexity and Events per Second from pastures to old-growths (Fig. 4 and Table 1).”
Quote in the paperSubstantial claim - Findingown data
Most old-growth forests were distinguishable in community composition from regenerating sites along the first ordination axis, with the exception of one pasture regenerating for 34 years.
“Most old-growth forests were rather distinguishable in community composition from regenerating sites along the first ordination axis, with the exception of one pasture regenerating for 34 years (Fig. 2a).”
Quote in the paperSubstantial claim - Findingown data
Community composition when restricting bird species to only those represented in both expert-led and CNN identification (n = 49) had an explanatory value of R2 = 0.85.
“This conjecture is supported by an even higher explanatory value (R² = 0.85) of community composition when restricting birds species to only those represented in both the expert-led and the CNN identification (n = 49, Fig. 3c).”
Quote in the paperNot a substantial claim - Findingown data
Entropy of Frequency increased along the recovery gradient from pastures to old-growths.
“As vocalizing vertebrate communities progressed towards old-growth forest conditions, the soundscapes showed increasing Soundscape Saturation and Entropy of Frequency, but decreasing Acoustic Complexity and Events per Second from pastures to old-growths (Fig. 4 and Table 1).”
Quote in the paperSubstantial claim - Findingown data
Both automated measures (acoustic index model and CNN bird community composition) reflected the composition of non-vocalizing nocturnal insects identified via metabarcoding.
“Importantly, both measures reflected composition of non-vocalizing nocturnal insects identified via metabarcoding.”
- Also stated in: § Sound diversity as measure of biodiversity · boosted
“Crucially, both automated sound-based measures (indices and CNN) had a good explanatory power even for species that are not part of the vocalizing tropical animal community, such as the largely silent nocturnal insects (Fig. 3b).”
Quote in the paperSubstantial claim - Also stated in: § Sound diversity as measure of biodiversity · boosted
- Findingown data
In our dataset, agricultural plots had more species than later recovery stages.
“In our dataset, we found more species and more vocalizations in agricultural plots (Fig. 4 and Fig. S2).”
Quote in the paperNot a substantial claim - Findingown data
The final data set of arthropods comprised 4557 BINs from 24 orders dominated by Lepidoptera (40%) followed by Diptera (32%), Hemiptera (10%), Hymenoptera (7%), Coleoptera (6%) and Trichoptera (2%).
“The final data set of arthropods comprised 4557 BINs from 24 orders dominated by Lepidoptera (40%) followed by Diptera (32 %), Hemiptera (10%), Hymenoptera (7%), Coleoptera (6%) and Trichoptera (2%).”
Quote in the paperNot a substantial claim - Findingown data
Nocturnal insects recover quickly with forest regeneration.
“Our data show not only that nocturnal insects recover quickly with forest regeneration but that CNN models track that recovery well.”
Quote in the paperSubstantial claim - Findingown data
CNN models track nocturnal insect recovery well.
“Our data show not only that nocturnal insects recover quickly with forest regeneration but that CNN models track that recovery well.”
Quote in the paperSubstantial claim - Findingown data
The richness of species recorded in old-growth plots increased with Entropy of Frequency.
“The richness of species recorded in old-growth plots increased with the Entropy of Frequency and decreased with Acoustic Complexity reflecting low species richness but a functional highly diverse community in old-growth forests (Table 1).”
Quote in the paperSubstantial claim - Findingown data
The first axis of the NMDS revealed a linear gradient of community recovery.
“As this axis revealed a linear gradient of community recovery (Fig. 2a), we generally hypothesized a similar relationship for other indicators, too, and applied linear models in subsequent analyses.”
Quote in the paperSubstantial claim - Findingown data
The richness of species recorded in old-growth plots decreased with Acoustic Complexity.
“The richness of species recorded in old-growth plots increased with the Entropy of Frequency and decreased with Acoustic Complexity reflecting low species richness but a functional highly diverse community in old-growth forests (Table 1).”
Quote in the paperNot a substantial claim - Findingown data
Predictions at 0.5 and 0.8 output score thresholds revealed very similar species communities.
“Both revealed very similar species communities, therefore, we used the more conservative threshold of 0.8 to declare a species as present in a specific file.”
Quote in the paperNot a substantial claim - Findingown data
The mean average-precision across the classes of the best model was 86.9%.
“The mean average-precision across the classes of the best model was 86.9%.”
Quote in the paperNot a substantial claim - Findingown data
The communities of vocalizing vertebrates and nocturnal insects were highly correlated with the main axis of species composition derived from CNN.
“The communities of vocalizing vertebrates, as well as the community of nocturnal insects, were highly correlated with the main axis of species composition derived from CNN (Fig. 3).”
Quote in the paperSubstantial claim - Findingown data
The total richness of vocalizing vertebrates increased with Events/Second.
“The total richness of vocalizing vertebrates decreased with Soundscape Saturation and increased with Events/Second, both in line with fewer species recorded in old-growth plots (Table 1).”
Quote in the paperNot a substantial claim - Findingown data
The total richness of vocalizing vertebrates decreased with Soundscape Saturation.
“The total richness of vocalizing vertebrates decreased with Soundscape Saturation and increased with Events/Second, both in line with fewer species recorded in old-growth plots (Table 1).”
Quote in the paperSubstantial claim - Findingown data
With this model, we identified 73 species in our 43 plots.
“With this model, we identified 73 species in our 43 plots (Tab. S1).”
Quote in the paperNot a substantial claim - Findingown data
There is no strong bias toward the most common classes (Pearson correlation ρ = 0.07, p = 0.48, Precision 0.8 ~ log(Presences)).
“There is no strong bias toward the most common classes (Pearson correlation ρ = 0.07, p = 0.48, Precision 0.8 ~ log(Presences)).”
Quote in the paperNot a substantial claim - Interpretationown data
Automated monitoring tools based on new technologies can effectively monitor the success of forest recovery, using robust and reproducible data.
“We show that such automated monitoring tools, based on new technologies, can effectively monitor the success of forest recovery, using robust and reproducible data.”
Quote in the paperSubstantial claim - Interpretationown data
Light trap assemblages predominantly represent non-vocalizing insects.
“In further testing, we excluded the 13 vocalizing insect species from the families Tettigoniidae, Gryllidae, Cicadidae, present in our light traps, which did not change the axis of species composition (Pearson correlation rho=0.986), indicating that light trap assemblages predominantly represent non-vocalizing insects.”
Quote in the paperSubstantial claim - Premisecitation
Bird vocalizations and soundscapes were altered with forest degradation from selective logging in Southeast Asia.
“Likewise, bird vocalizations and soundscapes were altered with forest degradation from selective logging in Southeast Asia22.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [22] Pillay, R., Fletcher, R. J., Sieving, K. E., Udell, B. J. & Bernard, H. Bioacoustic monitoring reveals shifts in breeding songbird populations and singing behaviour with selective logging in tropical forests. J. Appl. Ecol. 56, 2482–2492 (2019). source
- Premisecitation
Many taxonomic groups, including amphibians, birds, mammals, and insects include a considerable proportion of species that vocalize or otherwise use sound to communicate, making acoustic monitoring of these groups a particularly promising tool for biodiversity responses.
“Many taxonomic groups, including amphibians, birds, mammals, and insects include a considerable proportion of species that vocalize or otherwise use sound to communicate, making acoustic monitoring of these groups a particularly promising tool for biodiversity responses17,18.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [17] Teixeira, D., Maron, M. & van Rensburg, B. J. Bioacoustic monitoring of animal vocal behavior for conservation. Conserv. Sci. Pract. 1, e72 (2019). source
- cites [18] Vega-Hidalgo, A., Flatt, E., Whitworth, A. & Symes, L. Acoustic assessment of experimental reforestation in a Costa Rican rainforest. Ecol. Indic. 133, 108413 (2021). source
- Interpretationcitation
Measures based on composition are more powerful than richness alone.
“Moreover, we show that measures based on composition are more powerful than richness alone, as known for insect communities in tropical forests52.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [52] Basset, Y. et al. Changes in arthropod assemblages along a wide gradient of disturbance in Gabon. Conserv. Biol. 22, 1552–1563 (2008). source
- Premisecitation
Insect diversity, in general, has been shown to correlate with bird, frog and mammal diversity.
“Insect diversity, in general, has been shown to correlate with bird, frog and mammal diversity45.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [45] Kass, J. M. et al. The global distribution of known and undiscovered ant biodiversity. Sci. Adv. 8, eabp9908 (2022). source
- Interpretationown data
Using a set of acoustic indices instead of single ones is better for estimating biodiversity.
“Our findings add to the growing literature that shows that using a set of acoustic indices instead of single ones is better for estimating biodiversity50.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [50] Allen-Ankins, S. et al. Effectiveness of acoustic indices as indicators of vertebrate biodiversity. Ecol. Indic. 147, 109937 (2023). source
- Study design and conductnone given
Sound attenuation bias is minimal in our study system.
“Overall, we therefore think that the sound attenuation bias is minimal in our study system.”
Quote in the paperNot a substantial claim - Interpretationown data
Soundscape analysis is a powerful tool to monitor the recovery of faunal communities in hyper-diverse tropical forest.
“In summary, our results show that soundscape analysis is a powerful tool to monitor the recovery of faunal communities in hyper-diverse tropical forest.”
Quote in the paperSubstantial claim - Interpretationreasoning
Vocalizing vertebrate communities, sampled by audio recorders, can serve as indicators for the faunal recovery of tropical forests.
“Specifically, the vocalizing vertebrate communities, sampled by audio recorders, can therefore serve as indicators for the faunal recovery of tropical forests.”
Quote in the paperSubstantial claim - Interpretationown data
Soundscape diversity can be quantified in a cost-effective and robust way across the full gradient from active agriculture, to recovering and old-growth forests.
“Soundscape diversity can be quantified in a cost-effective and robust way across the full gradient from active agriculture, to recovering and old-growth forests.”
Quote in the paperSubstantial claim - Interpretationown data
Species composition, rather than species richness, serves as an indicator of forest recovery.
“Species composition, rather than species richness, thus serves as an indicator of forest recovery.”
Quote in the paperSubstantial claim - Study design and conductown data
Cross-correlograms showed spatial independence of the residuals in all six models.
“These showed spatial independence of the residuals in all six models (Fig. S4).”
Quote in the paperNot a substantial claim - Interpretationown data
Acoustic indices track the quick and consistent biodiversity recovery even in complex, hyper-diverse ecosystems such as tropical forests, independent of land use legacies.
“Importantly, we document that acoustic indices track the quick and consistent biodiversity recovery even in complex, hyper-diverse ecosystems such as tropical forests, independent of land use legacies.”
Quote in the paperSubstantial claim - Study design and conductown data
The additional sound files shifted by 15 min evaluated by another bird expert and the original dataset provided similar results.
“To assess inter-observer bias, additional sound files shifted by 15 min were evaluated by another bird expert (J.F.). Both datasets provided similar results (Fig. S3).”
Quote in the paperNot a substantial claim
- The quote only partly states this claim. 4
- It may state two separate claims as one. 21
- Probably not a substantial claim. 20
The quote only partly states this claim.
Findingown dataA combination of acoustic indices had good explanatory power for richness of old-growth species (adj. R2 = 0.45).
“Acoustic indices individually had low explanatory power, but we found that a combination of acoustic indices had a high explanatory power for the vertebrate community composition (adj. R² = 0.62), good explanatory power for richness of old-growth species and nocturnal insect composition, but low explanatory power for total vertebrate richness (Table 1).”
Quote in the paperSubstantial claimThe quote only partly states this claim.
Findingown dataA combination of acoustic indices had good explanatory power for nocturnal insect composition (adj. R2 = 0.42).
“Acoustic indices individually had low explanatory power, but we found that a combination of acoustic indices had a high explanatory power for the vertebrate community composition (adj. R² = 0.62), good explanatory power for richness of old-growth species and nocturnal insect composition, but low explanatory power for total vertebrate richness (Table 1).”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataAbandoned cacao plantations initially recovered vertebrate diversity faster than abandoned pastures, but converged during later succession.
“Abandoned cacao plantations initially recovered vertebrate diversity faster than abandoned pastures, but converged during later succession (Fig. 2a).”
Quote in the paperNot a substantial claimThe quote only partly states this claim.
Findingown dataA combination of acoustic indices had low explanatory power for total vertebrate richness (adj. R2 = 0.20).
“Acoustic indices individually had low explanatory power, but we found that a combination of acoustic indices had a high explanatory power for the vertebrate community composition (adj. R² = 0.62), good explanatory power for richness of old-growth species and nocturnal insect composition, but low explanatory power for total vertebrate richness (Table 1).”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataThere was a substantial overlap of species composition between the early and late recovery stages, with a most pronounced change in the early phase.
“First, we found a substantial overlap of species composition between the early and late recovery stages, with a most pronounced change in the early phase.”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataA bird community composition derived from an independently developed CNN correlated well with restoration (adj-R2 = 0.69) and was the single best predictor for expert-derived community composition, outperforming the combination of acoustic indices.
“Two automated measures – an acoustic index model and a bird community composition derived from an independently developed Convolutional Neural Network - correlated well with restoration (adj-R² = 0.62 and 0.69, respectively).”
- Also stated in: § Sound diversity as measure of biodiversity · boosted
“Despite the model identifying only ~25% of species detected by the experts in our data, the model-derived first community axis was the single best predictor for expert-derived community composition. With adjusted R² of 0.69, it outperformed the combination of acoustic indices, demonstrating that community composition estimated from CNN is a promising way to track faunal recovery of tropical forest”
Quote in the paperSubstantial claim- Also stated in: § Sound diversity as measure of biodiversity · boosted
It may state two separate claims as one.
Findingown dataAt an output score threshold of 0.5, the model achieved a mean F1-score across classes of 77.2%, a mean precision of 82.9% and a mean recall of 80.7%.
“At an output score threshold of 0.5, the model achieved a mean F1-score across classes of 77.2%, a mean precision of 82.9% and a mean recall of 80.7%.”
Quote in the paperNot a substantial claimThe quote only partly states this claim.
Findingown dataIn our dataset, agricultural plots had more vocalizations than later recovery stages.
“In our dataset, we found more species and more vocalizations in agricultural plots (Fig. 4 and Fig. S2).”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Findingown dataFor a threshold of 0.8, the mean F1-score, precision, and recall were 78.9%, 89.4%, and 77.5%, respectively.
“For a threshold of 0.8, the mean F1-score, precision, and recall were 78.9%, 89.4%, and 77.5%, respectively.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Findingown dataBlasts against the global database showed that a vast number of species identifications appear, which is completely implausible and complicates further ecological interpretations.
“Conversely, blasts against the global database showed that a vast number of species identifications appear, which is completely implausible and complicates further ecological interpretations.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Findingnone givenWith the community composition of insects shifting towards old-growth forests, Entropy of Frequency, Soundscape Saturation and Events per Second all increased.
“With the community composition of insects shifting towards old-growth forests, Entropy of Frequency, Soundscape Saturation and Events per Second all increased.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
PremisecitationCurrently, no single best acoustic index exists, as most indices have been tested only in a handful of habitats or land use contexts, and when used individually, they often yielded mixed results.
“Currently, no single best index exists, as most indices have been tested only in a handful of habitats or land use contexts, and when used individually, they often yielded mixed results27,28.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [27] Alcocer, I., Lima, H., Sugai, L. S. M. & Llusia, D. Acoustic indices as proxies for biodiversity: a meta-analysis. Biol. Rev. 97, 2209–2236 (2022). source
- cites [28] Sethi, S. S. et al. Limits to the accurate and generalizable use of soundscapes to monitor biodiversity. Nat. Ecol. Evol. 7, 1373–1378 (2023). source
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
PremisecitationSound diversity, expressed via the Soundscape Saturation Index, declined and became less synchronized with forest fragmentation and loss in Papua New Guinea and Borneo.
“Sound diversity, expressed via the Soundscape Saturation Index, declined and became less synchronized with forest fragmentation and loss in Papua New Guinea and Borneo19,20.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [19] Burivalova, Z. et al. Loss of temporal structure of tropical soundscapes with intensifying land use in Borneo. Sci. Total Environ. 852, 158268 (2022). source
- cites [20] Burivalova, Z. et al. Using soundscapes to detect variable degrees of human influence on tropical forests in Papua New Guinea. Conserv. Biol. 32, 205–215 (2018). source
It may state two separate claims as one.
PremisecitationA major bottleneck for machine learning approaches, including CNNs, is the need of large training datasets, particularly challenging in the hyper-diverse communities of tropical forests.
“A major bottleneck for machine learning approaches, including CNNs, is the need of large training datasets. This is particularly challenging in the hyper-diverse communities of tropical forests41.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [41] Sun, Y. R., Maeda, T. M., Solis-Lemus, C., Pimentel-Alarcon, D. & Burivalova, Z. Classification of animal sounds in a hyperdiverse rainforest using convolutional neural networks with data augmentation. Ecol. Indic. 145, 109621 (2022). source
It may state two separate claims as one.
Interpretationown dataBoth products from soundscape recordings and analyses-a set of acoustic indices and CNN-derived species composition-are suitable for tracking faunal recovery across a few decades.
“Our results demonstrate that both products from soundscapes recordings and analyses—a set of acoustic indices and CNN-derived species composition—are suitable for tracking faunal recovery across a few decades.”
Quote in the paperSubstantial claimProbably not a substantial claim. It may state two separate claims as one.
PremisecitationSome studies observed a strong correlation between the Acoustic Complexity Index and both the number of avian vocalizations and species richness, while others reported inconsistent results or no correlation.
“For example, some studies observed a strong correlation between the Acoustic Complexity Index, a measure for biotic activity29, and both the number of avian vocalizations and species richness24,29,30, while others reported inconsistent results31,32 or no correlation33,34.”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [24] Bradfer-Lawrence, T. et al. Guidelines for the use of acoustic indices in environmental research. Methods Ecol. Evol. 10, 1796–1807 (2019). source
- cites [29] Pieretti, N., Farina, A. & Morri, D. A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecol. Indic. 11, 868–873 (2011). source
- cites [30] Eldridge, A. et al. Sounding out ecoacoustic metrics: Avian species richness is predicted by acoustic indices in temperate but not tropical habitats. Ecol. Indic. 95, 939–952 (2018). source
Probably not a substantial claim.
PremisecitationDeep learning models, such as Convolutional Neural Networks (CNN), have been developed to identify birds, bats or amphibians.
“More recently, artificial intelligence deep learning models, such as Convolutional Neural Networks (CNN), have been developed to identify birds, bats or amphibians36–40.”
Quote in the paperSubstantial claim0 of 5 sources checked- cites [36] Ruff, Z. J., Lesmeister, D. B., Appel, C. L. & Sullivan, C. M. Workflow and convolutional neural network for automated identification of animal sounds. Ecol. Indic. 124, 107419 (2021). source
- cites [37] Ferreira, A. C. et al. Deep learning-based methods for individual recognition in small birds. Methods Ecol. Evol. 11, 1072–1085 (2020). source
- cites [38] Huang, C. J., Yang, Y. J., Yang, D. X. & Chen, Y. J. Applications of data mining techniques to automatic frog identification. Appl. Artif. Intell. 23, 553–569 (2009). source
- cites [39] Shivaprakash, K. N. et al. Potential for Artificial Intelligence (AI) and Machine Learning (ML) applications in biodiversity conservation, managing forests, and related services in India. Sustainability 14, 7154 (2022). source
- cites [40] Kahl, S., Wood, C. M., Eibl, M. & Klinck, H. BirdNET: a deep learning solution for avian diversity monitoring. Ecol. Inform. 61, 101236 (2021). source
It may state two separate claims as one.
PremisecitationAll conservation measures require cost-efficient and robust biodiversity monitoring, which is lagging behind carbon monitoring due in part to the lack of scalable, reproducible and cost-effective sampling methodologies.
“To be effective, all conservation measures require cost-efficient and robust biodiversity monitoring, which is lagging behind carbon monitoring due in part to the lack of scalable, reproducible and cost-effective sampling methodologies13,15.”
Quote in the paperSubstantial claim0 of 2 sources checkedIt may state two separate claims as one.
PremisecitationAdditional anthropogenic impacts such as logging or hunting can strongly affect the local fauna, but are not represented by recovery age.
“Second, additional anthropogenic impacts as logging or hunting20,49 can strongly affect the local fauna, but are not represented by recovery age.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [20] Burivalova, Z. et al. Using soundscapes to detect variable degrees of human influence on tropical forests in Papua New Guinea. Conserv. Biol. 32, 205–215 (2018). source
- cites [49] Burivalova, Z. et al. The sound of logging: Tropical forest soundscape before, during, and after selective timber extraction. Biol. Conserv. 254, 108812 (2021). source
It may state two separate claims as one.
PremisecitationThe recovery of tropical forest fauna varies widely, and depending on taxa, it is contentious and less predictable.
“However, the recovery of tropical forest fauna varies widely, and depending on taxa, it is contentious and less predictable9–13.”
Quote in the paperSubstantial claim0 of 5 sources checked- cites [9] Hua, F. Y. et al. The biodiversity and ecosystem service contributions and trade-offs of forest restoration approaches. Science 376, 839–844 (2022). source
- cites [10] Hoenle, P. O. et al. Rapid ant community reassembly in a Neotropical forest: Recovery dynamics and land-use legacy. Ecol. Appl. 32, e2559 (2022). source
- cites [11] Acevedo-Charry, O. & Aide, T. M. Recovery of amphibian, reptile, bird and mammal diversity during secondary forest succession in the tropics. Oikos 128, 1065–1078 (2019). source
- cites [12] Poorter, L. et al. Functional recovery of secondary tropical forests. Proc. Natl Acad. Sci. USA 118, e2003405118 (2021). source
- cites [13] Sethi, S. S. et al. Soundscapes predict species occurrence in tropical forests. Oikos 2022, e08525 (2022). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationCarbon storage and forest structure can be restored within several decades if appropriate species are planted and providing deforestation drivers have been addressed.
“Carbon storage and forest structure can be restored within several decades if appropriate species are planted and providing deforestation drivers have been addressed9.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [9] Hua, F. Y. et al. The biodiversity and ecosystem service contributions and trade-offs of forest restoration approaches. Science 376, 839–844 (2022). source
It may state two separate claims as one.
PremisecitationIn Puerto Rico, soundscapes became impoverished and subsequently recovered after a major hurricane, reflecting the reassembly of the vocalizing fauna.
“In Puerto Rico, soundscapes became impoverished and subsequently recovered after a major hurricane, reflecting the reassembly of the vocalizing fauna21.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [21] Gottesman, B. L. et al. What does resilience sound like? Coral reef and dry forest acoustic communities respond differently to Hurricane Maria. Ecol. Indic. 126, 107635 (2021). source
Probably not a substantial claim.
PremisecitationIt remains unclear if soundscapes reliably track the restoration of faunal biodiversity in tropical forests.
“These studies show that forest loss and degradation can be tracked by using soundscapes, but it remains unclear if soundscapes reliably track the restoration of faunal biodiversity in tropical forests23.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [23] Znidersic, E. & Watson, D. M. Acoustic restoration: Using soundscapes to benchmark and fast-track recovery of ecological communities. Ecol. Lett. 25, 1597–1603 (2022). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationThis kind of light attracts predominantly Lepidoptera and Dipteran.
“This kind of light attracts predominantly Lepidoptera and Dipteran.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [63] Brehm, G. et al. Moths are strongly attracted to ultraviolet and blue radiation. Insect Conserv. Divers. 14, 188–198 (2021). source
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
Study design and conductnone givenRandom resampling and data augmentation were applied to compensate for data imbalance and improve model generalization.
“Random resampling and data augmentation were applied to compensate for data imbalance and improve model generalization.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
Study design and conductnone givenOTUs were additionally removed from the results based on negative control samples, i.e. if the number of reads for the OTU in any sample was less than the maximum among negative controls, those reads were excluded from further analysis.
“OTUs were additionally removed from the results based on negative control samples, i.e. if the number of reads for the OTU in any sample was less than the maximum among negative controls, those reads were excluded from further analysis.”
Quote in the paperNot a substantial claimProbably not a substantial claim.
Study design and conductown dataThe model was finally evaluated against a dataset including the validation set and an additional, independent 200 expert-labelled one-minute recordings from similar habitats as in our study.
“The model was finally evaluated against a dataset including the validation set and an additional, independent 200 expert-labelled one-minute recordings from similar habitats as in our study (Supplementary Data 1).”
Quote in the paperSubstantial claimIt may state two separate claims as one.
PremisecitationSize filtering increases the likelihood of detecting small-bodied and rare species in samples, as individuals with a larger biomass provide disproportionate amounts of DNA and can therefore be over-represented when metabarcoding bulk samples.
“This was done to increase the likelihood of detecting small-bodied and rare species in samples, as individuals with a larger biomass provide disproportionate amounts of DNA and can therefore be over-represented when metabarcoding bulk samples65.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [65] Elbrecht, V. & Leese, F. Can DNA-based ecosystem assessments quantify species abundance? Testing primer bias and biomass—sequence relationships with an innovative metabarcoding protocol. PLoS ONE 10, e0130324 (2015). source
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
PremisecitationEven species belonging to the same genus may be differently impacted by the same DNA extraction/amplification/sequencing procedure, generating a difference in read numbers, which may result in loss of rare species.
“Here even species belonging to the same genus may be differently impacted by the same DNA extraction/amplification/sequencing procedure, generating a difference in read numbers, which may result in loss of rare species66.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [66] Martoni, F., Piper, A. M., Rodoni, B. C. & Blacket, M. J. Disentangling bias for non-destructive insect metabarcoding. Peerj 10, e12981 (2022). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationBINs avoid the situation that in certain lineages there are unequally more OTUs even within a species.
“BINs avoid the situation that in certain lineages there are unequally more OTUs even within a species.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [75] Ratnasingham, S. & Hebert, P. D. N. A DNA-based registry for all animal species: the Barcode Index Number (BIN) System. PLoS ONE 8, e66213 (2013). source
- cites [76] Ratnasingham, S. & Hebert, P. D. N. BOLD: The barcode of life data system (www.barcodinglife.org). Mol. Ecol. Notes 7, 355–364 (2007). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
Study design and conductreasoningSize filtering cannot guarantee all insects be detected.
“Size filtering is only one tool to improve the balance between small and rare species on the one hand and large and abundant species on the other hand in bulk samples, and even this cannot guarantee all insects be detected.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
Study design and conductreasoningTo reduce the risk of false positives, a cleaning step was employed that excluded read counts in the OTU table constituting <0.01% of the total number of reads in the sample.
“To reduce the risk of false positives, a cleaning step was employed that excluded read counts in the OTU table constituting <0.01% of the total number of reads in the sample.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
PremisecitationPassing samples through an 8-mm sieve separates larger and smaller insects, opening the avenue for using read numbers for abundance estimates.
“The samples were then passed through an 8-mm sieve, thereby separating larger and smaller insects, opening the avenue for using read numbers for abundance estimates64.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [64] Skelton, J., Cauvin, A. & Hunter, M. E. Environmental DNA metabarcoding read numbers and their variability predict species abundances, but weakly in non-dominant species. Environ. DNA early view https://doi.org/10.1002/edn3.355 (2022). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationPre-trained models for audio recognition (YAMNet, VGGish) are trained on sounds of limited time duration and resolution (rather than on images of sounds), which can impact their applicability to noisy and complex signals such as found in biodiverse soundscapes.
“While pre-trained models for audio recognition have recently been proposed (YAMNet, VGGish62), they are trained on sounds of limited time duration and resolution (rather than on images of sounds), which can impact their applicability to noisy and complex signals such as found in biodiverse soundscapes.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [62] Tsalera, E., Papadakis, A. & Samarakou, M. Comparison of pre-trained CNNs for audio classification using transfer learning. J. Sens. Actuator Netw. 10, 72 (2021). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
InterpretationreasoningBy restricting the library to BINs with georeferenced records from Central and South America, we limit the references in such a way that, for example, a newly introduced species from New Zealand would not be recognized without a reference from South America.
“By restricting the library to BINs with georeferenced records from Central and South America, we limit the references in such a way that, for example, a newly introduced species from New Zealand would not be recognized without a reference from South America.”
Quote in the paperNot a substantial claim- Rests on an assumption: The allocation to BIN units is a challenge, given today’s still very incomplete libraries.
- Rests on an assumption: In many regions, corresponding libraries are largely missing for large species groups, especially for groups harbouring “dark taxa” such as dipterans, hymenopterans and hemipterans, but also for large portions of other arthropod species, which are currently not referenced for South America.
Probably not a substantial claim.
Study design and conductown dataThe model was validated against a set of 200 one-minute audio files, each labelled with the set of present classes.
“The model was validated against a set of 200 one-minute audio files, each labelled with the set of present classes.”
Quote in the paperSubstantial claimProbably not a substantial claim.
Study design and conductown dataThe analyses used n = 43 biologically independent plots.
“Asterisks(***) indicate highly significant difference (p < 0.001) between both methods within a category of the recovery gradient based on a linear mixed-effects model (n = 43 biologically independent plots).”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationModels pre-trained on ImageNet are configured to extract a variety of generic image features that can be useful for spectrogram recognition.
“Models pre-trained on ImageNet are configured to extract a variety of generic image features that can be useful for spectrogram recognition40.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [40] Kahl, S., Wood, C. M., Eibl, M. & Klinck, H. BirdNET: a deep learning solution for avian diversity monitoring. Ecol. Inform. 61, 101236 (2021). source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
Study design and conductnone givenOur approach was standardized for all samples.
“However, our approach was standardized for all samples.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
InterpretationreasoningOn this basis, all sequences across all lineages receive a reasonably balanced assignment to taxonomic units (and/or interim species identifications such as the BIN) and information on ecological properties, e.g. pollination, is provided at different levels (species, genus, family).
“On this basis, all sequences across all lineages receive a reasonably balanced assignment to taxonomic units (and/or interim species identifications such as the BIN) and information on ecological properties, e.g. pollination78, is provided at different levels (species, genus, family).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [78] Ganuza, C. et al. Interactive effects of climate and land use on pollinator diversity differ among taxa and scales. Sci. Adv. 8, eabm9359 (2022). source
- Rests on an assumption: The allocation to BIN units is a challenge, given today’s still very incomplete libraries.
- Rests on an assumption: In many regions, corresponding libraries are largely missing for large species groups, especially for groups harbouring “dark taxa” such as dipterans, hymenopterans and hemipterans, but also for large portions of other arthropod species, which are currently not referenced for South America.
The paper relies on this without support.
The chronosequence of plots along a recovery gradient represents temporal forest recovery (space-for-time substitution).
“To pioneer the assessments of biodiversity recovery in tropical forests, we sampled soundscapes simultaneously with the same devices and protocol in 43 plots along a recovery gradient in a space-for-time substitute approach (Fig. 1).”
“Our plots followed a chronosequence of forest recovery, and included (i) active pastures and cacao plantations; (ii) formerly used pastures and plantations, which were secondary forests regrowing for 1 to 34 years at the time of the study, and (iii) old-growth forests, with no indication of recent use by humans.”
The authors state this as an assumption.
In the CNN training data, each song class is only present in its positive samples.
“However, it was assumed that each class is only present in its positive samples.”
The paper relies on this without support.
In comparison with randomly initialized models, models fine-tuned with transfer learning generally require less time and training data for strong performance.
“In comparison with randomly initialized models, models fine-tuned with transfer learning generally require less time and training data for strong performance.”
The paper relies on this without support.
Light traps with a wide range of species attracted are one of the most efficient methods for tropical nocturnal insects, even collecting some vocalizing insect species as cicadas.
“However, with the wide range of species attracted, it is one of the most efficient methods for tropical nocturnal insects, even collecting some vocalizing insect species as cicadas.”
The paper relies on this without support.
The allocation to BIN units is a challenge, given today’s still very incomplete libraries.
“The allocation to BIN units is a challenge, given today’s still very incomplete libraries.”
The paper relies on this without support.
In many regions, corresponding libraries are largely missing for large species groups, especially for groups harbouring “dark taxa” such as dipterans, hymenopterans and hemipterans, but also for large portions of other arthropod species, which are currently not referenced for South America.
“In many regions, corresponding libraries are largely missing for large species groups. This is true especially for groups harbouring “dark taxa” such as dipterans, hymenopterans and hemipterans, but also for large portions of other arthropod species, which are currently not referenced for South America.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
By listening in on soundscapes, researchers detect cryptic and endangered species non-invasively, infer population dynamics from vocal behavior, and track biodiversity change at community scales.
“By “listening in” on soundscapes, researchers detect cryptic and endangered species non-invasively, infer population dynamics from vocal behavior, and track biodiversity change at community scales”
From Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation (2026) · cites this paper as Müller et al. 2023
Acoustic complexity has been shown to decline following logging or fragmentation, serving as a sensitive indicator of habitat degradation.
“acoustic complexity has been shown to decline following logging or fragmentation, serving as a sensitive indicator of habitat degradation”
From Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation (2026) · cites this paper as Müller et al. 2023
In melampittas, song differences closely track genomic population structure, which is strong evidence that vocal divergence reflects evolutionary isolation.
“In melampittas, for example, song differences closely track genomic population structure—strong evidence that vocal divergence reflects evolutionary isolation”
From Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation (2026) · cites this paper as Müller et al. 2023
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Insect Events/Second t-value has the wrong sign
MajorFor nocturnal insect community composition, Events/Second has a negative coefficient and a positive standard error, so its t-value must be negative. The displayed values give approximately −3, not +2.33; the paper also interprets this predictor as increasing toward old-growth forests.
Reported 2.33 · recomputed -3
“| Events/Second | −0.8 ± 0.2 | −3.62 | 0.3 ± 0.2 | 2.04 | -0.3 ± 0.2 | −1.69 | −0.3 ± 0.1 | 2.33 |”
Figure 2 gives four acoustic predictors, but Table 1 lists five
MinorThe Figure 2 caption says its predictions use four acoustic indices and points to Table 1, which lists five acoustic-index predictors for the vertebrate community axis.
“b First axis of observed (expert identification) and predicted (four acoustic indices, Table 1) values of community composition.”
“| Events/Second | −0.8 ± 0.2 | −3.62 | 0.3 ± 0.2 | 2.04 | -0.3 ± 0.2 | −1.69 | −0.3 ± 0.1 | 2.33 |”
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755
Search and clustering orders of magnitude faster than BLAST
Edgar, R. C. (2010) · Bioinformatics · doi:10.1093/bioinformatics/btq461 · Cited by 21,056
Naive Bayesian Classifier for Rapid Assignment of rRNA Sequences into the New Bacterial Taxonomy
Wang, Q., Garrity, G. M., Tiedje, J. M., & Cole, J. R. (2007) · Applied and Environmental Microbiology · doi:10.1128/aem.00062-07 · Cited by 20,145
VSEARCH: a versatile open source tool for metagenomics
(2016) · doi:10.7717/peerj.2584 · Cited by 10,773
Gotelli, N., & Colwell, R. K. (2001) · Ecology Letters · doi:10.1046/j.1461-0248.2001.00230.x · Cited by 6,529
bold : The Barcode of Life Data System (http://www.barcodinglife.org)
RATNASINGHAM, S., & HEBERT, P. D. N. (2007) · Molecular Ecology Notes · doi:10.1111/j.1471-8286.2007.01678.x · Cited by 6,376
A DNA-Based Registry for All Animal Species: The Barcode Index Number (BIN) System
Ratnasingham, S., & Hebert, P. D. N. (2013) · PLoS ONE · doi:10.1371/journal.pone.0066213 · Cited by 2,301
MobileNetV2: Inverted Residuals and Linear BottlenecksPreprint
Sandler, M., Howard, A., Zhu, M., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1801.04381 · Cited by 2,255
Leray, M., Yang, J., Meyer, C., Mills, S. C., Agudelo, N., Ranwez, V., Boehm, J. T., & Machida, R. J. (2013) · Frontiers in Zoology · doi:10.1186/1742-9994-10-34 · Cited by 1,648
BirdNET: A deep learning solution for avian diversity monitoring
Kahl, S., Wood, C. M., Eibl, M., & Klinck, H. (2021) · Ecological Informatics · doi:10.1016/j.ecoinf.2021.101236 · Cited by 903
The biodiversity and ecosystem service contributions and trade-offs of forest restoration approaches
Hua, F., Bruijnzeel, L. A., Meli, P., Martin, P. A., Zhang, J., Nakagawa, S., Miao, X., Wang, W., McEvoy, C., Peña-Arancibia, J. L., et al. (2022) · Science · doi:10.1126/science.abl4649 · Cited by 831
Asynchronous carbon sink saturation in African and Amazonian tropical forests
Hubau, W., Lewis, S. L., Phillips, O. L., Affum-Baffoe, K., Beeckman, H., Cuní-Sanchez, A., Daniels, A. K., Ewango, C. E. N., Fauset, S., Mukinzi, J. M., et al. (2020) · Nature · doi:10.1038/s41586-020-2035-0 · Cited by 782
Elbrecht, V., & Leese, F. (2015) · PLOS ONE · doi:10.1371/journal.pone.0130324 · Cited by 765 · cited 1×
“This was done to increase the likelihood of detecting small-bodied and rare species in samples, as individuals with a larger biomass provide disproportionate amounts of DNA and can therefore be over-represented when metabarcoding bulk samples 65 .”
Pieretti, N., Farina, A., & Morri, D. (2011) · Ecological Indicators · doi:10.1016/j.ecolind.2010.11.005 · Cited by 658
Swarm v2: highly-scalable and high-resolution amplicon clustering
Mahé, F., Rognes, T., Quince, C., de Vargas, C., & Dunthorn, M. (2015) · PeerJ · doi:10.7717/peerj.1420 · Cited by 611
Hill, A. P., Prince, P., Piña‐Covarrubias, E., Doncaster, C. P., Snaddon, J. L., & Rogers, A. (2017) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12955 · Cited by 517
Real-time bioacoustics monitoring and automated species identification
Aide, T. M., Corrada-Bravo, C., Campos-Cerqueira, M., Milan, C., Vega, G., & Alvarez, R. (2013) · PeerJ · doi:10.7717/peerj.103 · Cited by 487
Self-citationHarnessing the potential of nature-based solutions for mitigating and adapting to climate change
Seddon, N. (2022) · Science · doi:10.1126/science.abn9668 · Cited by 471
ARTHROPOD ASSEMBLAGES ARE BEST PREDICTED BY PLANT SPECIES COMPOSITION
Schaffers, A. P., Raemakers, I., Sýkora, K. V., & ter Braak, C. J. F. (2008) · Ecology · doi:10.1890/07-0361.1 · Cited by 429
Guidelines for the use of acoustic indices in environmental research
Bradfer‐Lawrence, T., Gardner, N., Bunnefeld, L., Bunnefeld, N., Willis, S. G., & Dent, D. H. (2019) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13254 · Cited by 319
Connecting soundscape to landscape: Which acoustic index best describes landscape configuration?
Fuller, S., Axel, A. C., Tucker, D., & Gage, S. H. (2015) · Ecological Indicators · doi:10.1016/j.ecolind.2015.05.057 · Cited by 314
The contribution of insects to global forest deadwood decomposition
Seibold, S., Rammer, W., Hothorn, T., Seidl, R., Ulyshen, M. D., Lorz, J., Cadotte, M. W., Lindenmayer, D. B., Adhikari, Y. P., Aragón, R., et al. (2021) · Nature · doi:10.1038/s41586-021-03740-8 · Cited by 299
Self-citationFarina, A. (2013) · Springer eBooks · doi:10.1007/978-94-007-7374-5 · Cited by 282
Acoustic indices as proxies for biodiversity: a meta‐analysis
Alcocer, I., Lima, H., Sugai, L. S. M., & Llusia, D. (2022) · Biological Reviews · doi:10.1111/brv.12890 · Cited by 273 · cited 1×
“Currently, no single best index exists, as most indices have been tested only in a handful of habitats or land use contexts, and when used individually, they often yielded mixed results 27,28 .”
Tropical forests are home to over half of the world’s vertebrate species
Pillay, R., Venter, M., Aragon-Osejo, J. L., González‐del‐Pliego, P., Hansen, A. J., Watson, J. E. M., & Venter, O. (2021) · Frontiers in Ecology and the Environment · doi:10.1002/fee.2420 · Cited by 258
Eldridge, A., Guyot, P., Moscoso, P., Johnston, A., Eyre-Walker, Y., & Peck, M. (2018) · Ecological Indicators · doi:10.1016/j.ecolind.2018.06.012 · Cited by 209
Deep learning‐based methods for individual recognition in small birds
Ferreira, A. C., Silva, L. R., Renna, F., Brandl, H. B., Renoult, J. P., Farine, D. R., Covas, R., & Doutrelant, C. (2020) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13436 · Cited by 203
Relationship of insect biomass and richness with land use along a climate gradient
Uhler, J., Redlich, S., Zhang, J., Hothorn, T., Tobisch, C., Ewald, J., Thorn, S., Seibold, S., Mitesser, O., Morinière, J., et al. (2021) · Nature Communications · doi:10.1038/s41467-021-26181-3 · Cited by 201
Self-citationLeBien, J. G., Zhong, M., Campos‐Cerqueira, M., Velev, J. P., Dodhia, R. M., Ferres, J. L., & Aide, T. M. (2020) · Ecological Informatics · doi:10.1016/j.ecoinf.2020.101113 · Cited by 199
Self-citationEvidence complacency hampers conservation
Sutherland, W. J., & Wordley, C. F. R. (2017) · Nature Ecology & Evolution · doi:10.1038/s41559-017-0244-1 · Cited by 194
Shivaprakash, K. N., Swami, N., Mysorekar, S., Arora, R., Gangadharan, A., Vohra, K., Jadeyegowda, M., & Kiesecker, J. M. (2022) · Sustainability · doi:10.3390/su14127154 · Cited by 194
The global distribution of known and undiscovered ant biodiversity
Kass, J. M., Guénard, B., Dudley, K. L., Jenkins, C. N., Azuma, F., Fisher, B. L., Parr, C. L., Gibb, H., Longino, J. T., Ward, P. S., et al. (2022) · Science Advances · doi:10.1126/sciadv.abp9908 · Cited by 194
Self-citationAutomated high throughput animal CO1 metabarcode classification
Porter, T. M., & Hajibabaei, M. (2018) · Scientific Reports · doi:10.1038/s41598-018-22505-4 · Cited by 171
Aggregative response in bats: prey abundance versus habitat
Müller, J. C., Mehr, M., Bässler, C., Fenton, M. B., Hothorn, T., Pretzsch, H., Klemmt, H.-J., & Brandl, R. (2012) · Oecologia · doi:10.1007/s00442-011-2247-y · Cited by 161
Self-citationMorinière, J., de Araújo, B. C., Lam, A. W., Hausmann, A., Balke, M., Schmidt, S., Hendrich, L., Doczkal, D., Fartmann, B., Arvidsson, S., et al. (2016) · PLoS ONE · doi:10.1371/journal.pone.0155497 · Cited by 147
Functional recovery of secondary tropical forests
Poorter, L., Rozendaal, D. M. A., Bongers, F., Almeida, D. J. S., Álvarez, F. S., Andrade, J. L., Arreola Villa, L. F., Becknell, J. M., Bhaskar, R., Boukili, V., et al. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2003405118 · Cited by 147
Morinière, J., Balke, M., Doczkal, D., Geiger, M. F., Hardulak, L. A., Haszprunar, G., Hausmann, A., Hendrich, L., Regalado, L., Rulik, B., et al. (2019) · Molecular Ecology Resources · doi:10.1111/1755-0998.13022 · Cited by 143
Antich, A., Palacín, C., Wangensteen, O. S., & Turón, X. (2021) · BMC Bioinformatics · doi:10.1186/s12859-021-04115-6 · Cited by 140
Mammides, C., Goodale, E., Dayananda, S. K., Luo, K., & Chen, J. (2017) · Ecological Indicators · doi:10.1016/j.ecolind.2017.07.017 · Cited by 139
Bioacoustic monitoring of animal vocal behavior for conservation
Teixeira, D., Maron, M., & van Rensburg, B. J. (2019) · Conservation Science and Practice · doi:10.1111/csp2.72 · Cited by 138
Co-limitation towards lower latitudes shapes global forest diversity gradients
Liang, J., Gamarra, J. G. P., Picard, N., Zhou, M., Pijanowski, B. C., Jacobs, D. F., Reich, P. B., Crowther, T. W., Nabuurs, G. J., de‐Miguel, S., et al. (2022) · Nature Ecology & Evolution · doi:10.1038/s41559-022-01831-x · Cited by 133
Comparison of Pre-Trained CNNs for Audio Classification Using Transfer Learning
Tsalera, E., Papadakis, A., & Samarakou, M. (2021) · Journal of Sensor and Actuator Networks · doi:10.3390/jsan10040072 · Cited by 127
Workflow and convolutional neural network for automated identification of animal sounds
Ruff, Z. J., Lesmeister, D. B., Appel, C. L., & Sullivan, C. M. (2021) · Ecological Indicators · doi:10.1016/j.ecolind.2021.107419 · Cited by 123
Dröge, S., Martin, D. A., Andriafanomezantsoa, R., Buřivalová, Z., Fulgence, T. R., Osen, K., Rakotomalala, E., Schwab, D., Wurz, A., Richter, T., et al. (2020) · Ecological Indicators · doi:10.1016/j.ecolind.2020.106929 · Cited by 118
Self-citationSpecies Richness (of Insects) Drives the Use of Acoustic Space in the Tropics
Aide, T. M., Hernández-Serna, A., Campos‐Cerqueira, M., Acevedo-Charry, O. A., & Deichmann, J. L. (2017) · Remote Sensing · doi:10.3390/rs9111096 · Cited by 117
Self-citationAcevedo‐Charry, O., & Aide, T. M. (2019) · Oikos · doi:10.1111/oik.06252 · Cited by 113
Burivalova, Z., Towsey, M., Boucher, T., Truskinger, A., Apelis, C., Roe, P., & Game, E. T. (2018) · Conservation Biology · doi:10.1111/cobi.12968 · Cited by 110
Self-citationThe sound of a tropical forest
Burivalova, Z., Game, E. T., & Butler, R. A. (2019) · Science · doi:10.1126/science.aav1902 · Cited by 107
Self-citationInteractive effects of climate and land use on pollinator diversity differ among taxa and scales
Ganuza, C., Redlich, S., Uhler, J., Tobisch, C., Rojas-Botero, S., Peters, M. K., Zhang, J., Benjamin, C. S., Englmeier, J., Ewald, J., et al. (2022) · Science Advances · doi:10.1126/sciadv.abm9359 · Cited by 101
Self-citationSkelton, J., Cauvin, A., & Hunter, M. E. (2023) · Environmental DNA · doi:10.1002/edn3.355 · Cited by 99
Upscaling tropical restoration to deliver environmental benefits and socially equitable outcomes
Edwards, D. P., Cerullo, G. R., Chomba, S., Worthington, T. A., Balmford, A. P., Chazdon, R. L., & Harrison, R. D. (2021) · Current Biology · doi:10.1016/j.cub.2021.08.058 · Cited by 97
Limits to the accurate and generalizable use of soundscapes to monitor biodiversity
Sethi, S. S., Bick, A., Ewers, R. M., Klinck, H., Ramesh, V., Tuanmu, M.-N., & Coomes, D. A. (2023) · Nature Ecology & Evolution · doi:10.1038/s41559-023-02148-z · Cited by 95 · cited 2×
“This is in line with a recent meta-analysis using a global data set of audio-recordings paired with manual avifaunal point counts 28 .”
Toward a standardized quantitative and qualitative insect monitoring scheme
Hausmann, A., Segerer, A. H., Greifenstein, T., Knubben, J., Morinière, J., Bozicevic, V., Doczkal, D., Günter, A., Ulrich, W., & Habel, J. C. (2020) · Ecology and Evolution · doi:10.1002/ece3.6166 · Cited by 81
The effectiveness of acoustic indices for forest monitoring in Atlantic rainforest fragments
Jorge, F., Machado, C. G., Nogueira, S. S. D. C., & Nogueira-Filho, S. L. G. (2018) · Ecological Indicators · doi:10.1016/j.ecolind.2018.04.001 · Cited by 78
Utility of acoustic indices for ecological monitoring in complex sonic environments
Ross, S. R. P.-J., Friedman, N. R., Yoshimura, M., Yoshida, T., Donohue, I., & Economo, E. P. (2021) · Ecological Indicators · doi:10.1016/j.ecolind.2020.107114 · Cited by 78
Moths are strongly attracted to ultraviolet and blue radiation
Brehm, G., Niermann, J., Jaimes Nino, L. M., Enseling, D., Jüstel, T., Axmacher, J. C., Warrant, E., & Fiedler, K. (2021) · Insect Conservation and Diversity · doi:10.1111/icad.12476 · Cited by 76
How does FSC forest certification affect the acoustically active fauna in Madre de Dios, Peru?
Campos‐Cerqueira, M., Mena, J. L., Tejeda, V., Aguilar‐Amuchastegui, N., Gutiérrez, N., & Aide, T. M. (2019) · Remote Sensing in Ecology and Conservation · doi:10.1002/rse2.120 · Cited by 60
Self-citationChanges in Arthropod Assemblages along a Wide Gradient of Disturbance in Gabon
BASSET, Y., MISSA, O., ALONSO, A., MILLER, S. E., CURLETTI, G., DE MEYER, M., EARDLEY, C., LEWIS, O. T., MANSELL, M. W., NOVOTNY, V., et al. (2008) · Conservation Biology · doi:10.1111/j.1523-1739.2008.01017.x · Cited by 59
Effectiveness of acoustic indices as indicators of vertebrate biodiversity
Allen-Ankins, S., McKnight, D. T., Nordberg, E. J., Hoefer, S., Roe, P., Watson, D. M., McDonald, P. G., Fuller, R. A., & Schwarzkopf, L. (2023) · Ecological Indicators · doi:10.1016/j.ecolind.2023.109937 · Cited by 55
Sun, Y., Midori Maeda, T., Solís-Lemus, C., Pimentel-Alarcón, D., & Buřivalová, Z. (2022) · Ecological Indicators · doi:10.1016/j.ecolind.2022.109621 · Cited by 49
Self-citationBurivalova, Z., Purnomo, Orndorff, S., Truskinger, A., Roe, P., & Game, E. T. (2021) · Biological Conservation · doi:10.1016/j.biocon.2020.108812 · Cited by 46
Self-citationTobisch, C., Rojas‐Botero, S., Uhler, J., Müller, J. C., Kollmann, J., Moning, C., Brändle, M., Goßner, M. M., Redlich, S., Zhang, J., et al. (2023) · Oecologia · doi:10.1007/s00442-023-05345-6 · Cited by 45
Self-citationZnidersic, E., & Watson, D. M. (2022) · Ecology Letters · doi:10.1111/ele.14015 · Cited by 44
Farrell, C. A., Aronson, J. C., Daily, G. C., Hein, L., Obst, C., Woodworth, P., & Stout, J. C. (2021) · Restoration Ecology · doi:10.1111/rec.13613 · Cited by 43
Pillay, R., Fletcher, R. J., Sieving, K. E., Udell, B. J., & Bernard, H. (2019) · Journal of Applied Ecology · doi:10.1111/1365-2664.13492 · Cited by 40
Gottesman, B. L., Olson, J. C., Yang, S., Acevedo-Charry, O., Francomano, D., Martinez, F. A., Appeldoorn, R. S., Mason, D. M., Weil, E., & Pijanowski, B. C. (2021) · Ecological Indicators · doi:10.1016/j.ecolind.2021.107635 · Cited by 37
Disentangling bias for non-destructive insect metabarcoding
Martoni, F., Piper, A. M., Rodoni, B. C., & Blacket, M. J. (2022) · PeerJ · doi:10.7717/peerj.12981 · Cited by 36
Loss of temporal structure of tropical soundscapes with intensifying land use in Borneo
Burivalova, Z., Maeda, T. M., Purnomo, Rayadin, Y., Boucher, T., Choksi, P., Roe, P., Truskinger, A., & Game, E. T. (2022) · Science of The Total Environment · doi:10.1016/j.scitotenv.2022.158268 · Cited by 36
Self-citationRapid ant community reassembly in a Neotropical forest: Recovery dynamics and land‐use legacy
Hoenle, P. O., Donoso, D. A., Argoti, A., Staab, M., von Beeren, C., & Blüthgen, N. (2022) · Ecological Applications · doi:10.1002/eap.2559 · Cited by 35
Self-citationSoundscapes predict species occurrence in tropical forests
Sethi, S. S., Ewers, R. M., Jones, N. S., Sleutel, J., Shabrani, A., Zulkifli, N., & Picinali, L. (2022) · Oikos · doi:10.1111/oik.08525 · Cited by 34
Landscape-scale forest cover drives the predictability of forest regeneration across the Neotropics
Arroyo‐Rodríguez, V., Rito, K. F., Gutiérrez, M. F., Navia, I. C., Mora, F., Arreola-Villa, F., Balvanera, P., Bongers, F., Castellanos‐Castro, C., Catharino, E. L. M., et al. (2023) · Proceedings of the Royal Society B Biological Sciences · doi:10.1098/rspb.2022.2203 · Cited by 34
Acoustic assessment of experimental reforestation in a Costa Rican rainforest
Vega-Hidalgo, Á., Flatt, E., Whitworth, A., & Symes, L. (2021) · Ecological Indicators · doi:10.1016/j.ecolind.2021.108413 · Cited by 26
Owen, K. C., Melin, A. D., Campos, F. A., Fedigan, L. M., Gillespie, T. W., & Mennill, D. J. (2020) · Avian Conservation and Ecology · doi:10.5751/ace-01615-150125 · Cited by 24
Buxton, R. D. (1990) · Journal of Applied Entomology · doi:10.1111/j.1439-0418.1990.tb00055.x · Cited by 18
The glasgow leaders’ declaration on forests and land use: Significance toward “Net Zero”
Nasi, R. (2021) · Global Change Biology · doi:10.1111/gcb.16039 · Cited by 16
Sueur, J., Krause, B. L., & Farina, A. (2021) · Current Biology · doi:10.1016/j.cub.2021.08.063 · Cited by 15
Bird Assemblage Recovery in a Chronosequence of Tropical Dry Forests in Costa Rica
Hilje, B., Sánchez, N. V., Bayne, E. M., & Sánchez-Azofeifa, A. G. (2020) · Forests · doi:10.3390/f11060629 · Cited by 12
APPLICATIONS OF DATA MINING TECHNIQUES TO AUTOMATIC FROG IDENTIFICATION
Huang, C.-J., Yang, Y.-J., Yang, D.-X., & Chen, Y.-J. (2009) · Applied Artificial Intelligence · doi:10.1080/08839510903145223 · Cited by 5
Flagged first, then by citations
Cited by
Sorted by citations
The potential for AI to revolutionize conservation: a horizon scan
Reynolds, S. A., Beery, S., Burgess, N. D., Burgman, M. A., Butchart, S. H. M., Cooke, S. J., Coomes, D. A., Danielsen, F., Minin, E. D., Durán, A. P., et al. (2024) · Trends in Ecology & Evolution · doi:10.1016/j.tree.2024.11.013 · Cited by 136
Worldwide Soundscapes: A Synthesis of Passive Acoustic Monitoring Across Realms
Darras, K., Rountree, R. A., Wilgenburg, S. L. V., Cord, A. F., Pitz, F., Chen, Y., Dong, L., Rocquencourt, A., Desjonquères, C., Diaz, P. M., et al. (2025) · Global Ecology and Biogeography · doi:10.1111/geb.70021 · Cited by 91
Self-citationTowards a toolkit for global insect biodiversity monitoring
van Klink, R., Sheard, J. K., Høye, T. T., Roslin, T. V., Nascimento, L. A. D., & Bauer, S. (2024) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2023.0101 · Cited by 49
Drivers and benefits of natural regeneration in tropical forests
Chazdon, R. L., Blüthgen, N., Brancalion, P. H. S., Heinrich, V., & Bongers, F. (2025) · Nature Reviews Biodiversity · doi:10.1038/s44358-025-00043-y · Cited by 43
Self-citationEscobar, S., Newell, F. L., Endara, M., Guevara, J. E., Landim, A. R., Neuschulz, E. L., Hausmann, R., Müller, J. C., Pedersen, K., Schleuning, M., et al. (2025) · Ecosphere · doi:10.1002/ecs2.70157 · Cited by 42
Self-citationEcological restoration and artificial intelligence: whose values inform a project?
Woodrich, S. T., & Pape, T. (2024) · Restoration Ecology · doi:10.1111/rec.14128 · Cited by 28
Artificial Intelligence in Landscape Ecology: Recent Advances, Perspectives, and Opportunities
Frazier, A. E., & Song, L. (2024) · Current Landscape Ecology Reports · doi:10.1007/s40823-024-00103-7 · Cited by 27
Williams, B., Balvanera, S. M., Sethi, S. S., Lamont, T. A. C., Jompa, J., Prasetya, M. E., Richardson, L. E., Chapuis, L., Weschke, E., Hoey, A. S., et al. (2025) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013029 · Cited by 25
Kortmann, M., Chao, A., Schaefer, H. M., Blüthgen, N., Gelis, R. A., Tremlett, C. J., Busse, A., Püls, M., Seibold, S., Kriegel, P., et al. (2025) · Journal of Applied Ecology · doi:10.1111/1365-2664.14879 · Cited by 19
Self-citationJacuzzi, G., & Olden, J. D. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103294 · Cited by 16
Field Deployment of BiodivX Drones in the Amazon Rainforest for Biodiversity Monitoring
Geckeler, C., Kirchgeorg, S., Strunck, G., Thostrup, F. B., Sangermano, F., Desiderato, A., Lüthi, M. N., Jucker, M., Gonzalez, M. A., Franco‐Sierra, N. D., et al. (2025) · IEEE transactions on field robotics. · doi:10.1109/tfr.2025.3574945 · Cited by 15
Time and habitat structure shape insect acoustic activity in the Amazon
Nascimento, L. A. D., Pérez‐Granados, C., Alencar, J. B. R., & Beard, K. H. (2024) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2023.0112 · Cited by 15
Li, Y., Devenish, C., Tosa, M. I., Luo, M., Bell, D. M., Lesmeister, D. B., Greenfield, P. A., Pichler, M., Levi, T., & Yu, D. W. (2024) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2023.0123 · Cited by 15
Madhusudhana, S., Klinck, H., & Symes, L. B. (2024) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2023.0444 · Cited by 13
Deriving meaning from acoustic data in hyper-diverse ecosystems
Bradfer‐Lawrence, T., Buřivalová, Z., & Dent, D. H. (2025) · Trends in Ecology & Evolution · doi:10.1016/j.tree.2025.03.004 · Cited by 14
Self-citationHow AI can help to save endangered species
Thompson, T. (2023) · Nature · doi:10.1038/d41586-023-03328-4 · Cited by 13
Time series methods for the analysis of soundscapes and other cyclical ecological data
Yoh, N., Haley, C., & Buřivalová, Z. (2024) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14361 · Cited by 12
Self-citationHabitat niches of bird species along a recovery gradient in the Chocó tropical forest
Falconí‐López, A., Mitesser, O., Schaefer, H. M., Blüthgen, N., Busse, A., Feldhaar, H., Freile, J. F., Gelis, R. A., Grella, N., Heibl, C., et al. (2024) · Ecological Indicators · doi:10.1016/j.ecolind.2024.112260 · Cited by 12
Self-citationUrfan, M., Rajput, P., Mahajan, P., Sharma, S., Hakla, H. R., Kour, V., Khajuria, B., Chowdhary, R., Lehana, P. K., Karlupia, N., et al. (2024) · Research · doi:10.34133/research.0491 · Cited by 11
Ross, S. R. P. ‐., Friedman, N. R., Dudley, K. L., Yoshida, T., Yoshimura, M., Economo, E. P., Armitage, D. W., & Donohue, I. (2023) · Global Change Biology · doi:10.1111/gcb.17067 · Cited by 12
The potential of soundscapes as an ecosystem monitoring tool for urban biodiversity
Arzberger, S., Fairbairn, A. J., Hemauer, M., Mühlbauer, M., Weissmann, J. A., & Egerer, M. (2025) · Journal of Urban Ecology · doi:10.1093/jue/juaf002 · Cited by 10
Escobar, S., Newell, F. L., Endara, M., Guevara‐Andino, J. E., Landim, A. R., Neuschulz, E. L., Nußer, R., Müller, J. C., Pedersen, K., Schleuning, M., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.03.21.586145 · Cited by 10
Self-citationGindhart, R., Müller, J. C., Buřivalová, Z., Blüthgen, N., Busse, A., la Hoz, M. C. D., Donoso, D. A., Feldhaar, H., Grella, N., Kortmann, M., et al. (2024) · Insect Conservation and Diversity · doi:10.1111/icad.12793 · Cited by 8
Self-citationMichaud, F., Sueur, J., Sèbe, F., Cesne, M. L., & Haupert, S. (2025) · Ecological Indicators · doi:10.1016/j.ecolind.2025.114475 · Cited by 9
Automatic detection and annotation of eastern Caribbean sperm whale codas
Gubnitsky, G., Mevorach, Y., Gero, S., Gruber, D. F., & Diamant, R. (2025) · Scientific Reports · doi:10.1038/s41598-025-97009-z · Cited by 7
Vertebrate diversity and biomass along a recovery gradient in a lowland tropical forest
Grella, N., Pedersen, K., Blüthgen, N., Busse, A., Donoso, D. A., Falconí‐López, A., Fiderer, C., Heurich, M., la Hoz, M. C. D., Kriegel, P., et al. (2024) · Biotropica · doi:10.1111/btp.13417 · Cited by 9
Self-citationWang, S., Duan, Y., Cao, R., Feng, J., Ge, J., & Wang, T. (2025) · Biological Conservation · doi:10.1016/j.biocon.2025.111115 · Cited by 6
Rothacher, J., Mitesser, O., Müller, S., Scherer‐Lorenzen, M., Buřivalová, Z., & Müller, J. C. (2025) · Biological Conservation · doi:10.1016/j.biocon.2025.111116 · Cited by 6
Self-citationEmerging uses of artificial intelligence in deep time biodiversity research
Silvestro, D., & Pimiento, C. (2025) · Nature Reviews Biodiversity · doi:10.1038/s44358-025-00075-4 · Cited by 15
Kümmet, S., Müller, J. C., Buřivalová, Z., Schaefer, M. H., Gelis, R. A., Freile, J. F., Busse, A., Püls, M., Kriegel, P., Seibold, S., et al. (2025) · Conservation Letters · doi:10.1111/conl.13131 · Cited by 5
Self-citationInteligência artificial e mudanças climáticas
Artaxo, P. E., Rizzo, L. V., & Machado, L. A. T. (2024) · Revista USP · doi:10.11606/issn.2316-9036.i141p29-40 · Cited by 7
From valleys to peaks: characterizing soundscapes in the Northern European Limestone Alps
Ebner, M., Tappeiner, U., & Schirpke, U. (2025) · Landscape Ecology · doi:10.1007/s10980-025-02112-1 · Cited by 6
Distance-decay reveals contrasting effects of land-use types on arthropod community homogenisation
Decker, O., Uhler, J., Redlich, S., Chao, A., Kortmann, M., Steffan-Dewenter, I. D., Tobisch, C., Ewald, J., Englmeier, J., Fricke, U., et al. (2026) · Nature Communications · doi:10.1038/s41467-025-67612-9 · Cited by 5
Self-citationHussein, A. F., Al-Neami, A. Q., & Al-Qazzaz, N. K. (2025) · International Journal of Information Technology · doi:10.1007/s41870-025-02953-8 · Cited by 5
Gizachew, B. (2026) · Remote Sensing · doi:10.3390/rs18081193 · Cited by 6
Loo, Y. Y., Lee, M. Y., Shaheed, S., Maul, T., & Clink, D. J. (2025) · The Journal of the Acoustical Society of America · doi:10.1121/10.0034638 · Cited by 4
A Biodiversity Observation Network to support conservation action and mainstream knowledge in Canada
Gonzalez, A., O’Connor, M. I., Bates, A. E., Bobiwash, K., Burton, A. C., van Dam‐Bates, P., Eckert, I., Gravel, D., Idrobo, C. J., Pollock, L. J., et al. (2025) · FACETS · doi:10.1139/facets-2024-0206 · Cited by 8
Performance of four ecoacoustic indices as proxies of bird species richness in a Neotropical forest
Hutschenreiter, A., Aureli, F., Torres‐Araneda, A., & Andresen, E. (2025) · Biodiversity and Conservation · doi:10.1007/s10531-025-03208-5 · Cited by 4
Recovery of tree-related microhabitats in a tropical rainforest after agricultural abandonment
Hausmann, R., Escobar, S., Blüthgen, N., Schaefer, H. M., Newell, F. L., Endara, M., Guevara, J. E., Feldhaar, H., Kraus, D., Thom, D., et al. (2026) · Biological Conservation · doi:10.1016/j.biocon.2026.111739 · Cited by 3
Self-citationBerman, L. M., Perea, S., Sagar, H. S. S. C., Persche, M. E., Ingram, G., Maeda, T. M., Yoh, N., Müller, J. C., Schlüter, J., Zapata, N., et al. (2026) · Global Ecology and Biogeography · doi:10.1111/geb.70224 · Cited by 4
Self-citationFengler, B., Seyidbayli, C., Reinhardt, A., Anders, M., & Westphal, C. (2026) · Ecological Informatics · doi:10.1016/j.ecoinf.2026.103733 · Cited by 4
Securing the future of Hainan’s Tropical Rainforest: challenges and strategies for conservation
Wang, S., & Xie, C. (2025) · Frontiers in Forests and Global Change · doi:10.3389/ffgc.2025.1576900 · Cited by 3
IoT and M2M Applications in Satellite Networks
Iqbal, A., Shakeel, A., Rashid, A. A., & Kim, S. W. (2024) · Synthesis lectures on communications · doi:10.1007/978-3-031-63705-6_2 · Cited by 3
Huang, B., Yi, W., Lyu, Y., Yan, X., Tong, M., & Wang, X. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103280 · Cited by 3
A ready-to use workflow for automated, sound-based bird identification and localizationPreprint
Buchmann, C. M., & Schurr, F. M. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.07.02.601711 · Cited by 3
Diversity and host-networks of saproxylic beetles along a forest recovery gradient in the tropics
Falconí‐López, A., Grella, N., Donoso, D. A., Feldhaar, H., Huszarik, M., & Müller, J. (2026) · Forest Ecology and Management · doi:10.1016/j.foreco.2026.124102 · Cited by 2
Self-citationOkamoto, R., & Oguma, H. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103299 · Cited by 3
Habitat and land‐use intensity shape moth community structure across temperate forest and grassland
Achury, R., Staab, M., Seibold, S., Müller, J. C., Heidrich, L., Püls, M., Hacker, H., Fonseca, C. R., Fischer, M., Blüthgen, N., et al. (2025) · Journal of Animal Ecology · doi:10.1111/1365-2656.70132 · Cited by 2
Self-citationMüeller, S., Barrios, H., Campos, E., Casama, A., Forgues, K., Kawakami, S., Kunz, M., Marchena, B., Olea, D., Salazar, I., et al. (2025) · Global Ecology and Conservation · doi:10.1016/j.gecco.2025.e03920 · Cited by 2
An automated workflow that combines sound‐based bird identification and localization
Buchmann, C. M., & Schurr, F. M. (2025) · Ibis · doi:10.1111/ibi.70012 · Cited by 2
Duan, Y., Wang, S., Cao, R., Feng, J., Ge, J., & Wang, T. (2025) · Ecosphere · doi:10.1002/ecs2.70493 · Cited by 2
Brehm, G., Böttger, D., Diniz, U. M., Donoso, D. A., Kortmann, M., Müller, J., Rabl, D., Keller, A., & Laguerre, M. S. (2025) · Neotropical Entomology · doi:10.1007/s13744-025-01333-y · Cited by 2
Self-citationSuba, V. O., Achieng, O. E., Akter, A., Iheme, O. U., Sikuku, L., Basseya, E., Ochieng, W. A., Njoroge, B., Otieno, D. O., Tsingalia, M. H., et al. (2026) · Environmental Monitoring and Assessment · doi:10.1007/s10661-025-14973-9 · Cited by 2
In situ aerial bioacoustic monitoring from extremely noisy drone recordings
Wang, L., Clayton, M., & Rossberg, A. G. (2026) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.70324 · Cited by 2
Integrated development and future trends of restoration ecology
Ting, Z. H. O. U., BaoMing, C. H. E. N., HuiXuan, L. I. A. O., & ShaoLin, P. E. N. G. (2024) · Scientia Sinica Vitae · doi:10.1360/ssv-2024-0139 · Cited by 2
Perea, S. (2025) · Nature Reviews Biodiversity · doi:10.1038/s44358-025-00073-6 · Cited by 1
Huang, G., MuXin, L. I. A. O., WenBin, Q. I. A. N., & Wei, F. (2025) · Scientia Sinica Vitae · doi:10.1360/ssv-2025-0322 · Cited by 1
Development of ecocentric radiation protection: issues, challenges and approaches
Mothersill, C. E., Desai, R., Alonzo, F., Ariyoshi, K., Bonisoli‐Alquati, A., Bradshaw, C., Bréchignac, F., Byun, S. H., Chauhan, V., Cresswell, T., et al. (2025) · International Journal of Radiation Biology · doi:10.1080/09553002.2025.2595632 · Cited by 1
Bioacoustic Baselines for Intact Forests
Buřivalová, Z., Perea, S., Berman, L. M., Sagar, H. S. S. C., Maeda, T. M., Yoh, N., Ingram, G., Persche, M. E., Roncal, C. M., Lopera-Toro, A., et al. (2026) · Global Change Biology · doi:10.1111/gcb.70917 · Cited by 1
Self-citationUnderstory frog communities recover quickly during tropical forest regeneration
Neira-Salamea, K., Tartara, A., de la Cruz, L., Newell, F. L., Endara, M., Escobar, S., Guevara‐Andino, J. E., Donoso, D. A., Heethoff, M., & Rödel, M. (2026) · Ecological Applications · doi:10.1002/eap.70242 · Cited by 1
Self-citationDang, L. M., Danish, S., Fayaz, M., Arzu, G. E., Nguyen, A., Song, H.-K., & Moon, H. (2026) · Process Safety and Environmental Protection · doi:10.1016/j.psep.2026.109102 · Cited by 2
The role of conservation bioacoustics in achieving global biodiversity goals
Perea, S., Rhinehart, T. A., Bouffaut, L., Araya‐Salas, M., Davidson, I. K., Arcila-Pérez, L. F., Kitzes, J. A., & Buřivalová, Z. (2026) · Nature Reviews Biodiversity · doi:10.1038/s44358-026-00189-3 · Cited by 1
Self-citationAI-driven forest restoration: A governance framework for people and nature
Morales‐Rueda, F., Nyasulu, M. K., Fischer, H. W., Fetzer, I., Tabik, S., Alcaraz‐Segura, D., & Castro, J. (2026) · Environmental Science & Policy · doi:10.1016/j.envsci.2026.104463 · Cited by 1
Amazonian soundscapes: unravelling the secrets of insect acoustic niches in diverse habitatsPreprint
Nascimento, L. A. D., Pérez‐Granados, C., Alencar, J. R., & Beard, K. H. (2023) · doi:10.32942/x2102p · Cited by 1
Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation
Zsebők, S. (2026) · Biologia Futura · doi:10.1007/s42977-026-00322-5
“Because vocal signals influence behavior, population dynamics, and population dynamics from vocal behavior, and track bio-diversity change at community scales (Chhaya et al. 2021 ; Müller et al. 2023).”
Amazonian soundscapes: unravelling the secrets of insect acoustic niches in diverse habitatsPreprint
Nascimento, L. A. D., Pérez‐Granados, C., Alencar, J. R., & Beard, K. H. (2023) · doi:10.32942/x23s40
Acoustic Indices as Key Biodiversity Indicators in Agroforestry SystemsPreprint
Kahlau, C., Asbeck, T., Pistorius, T., & Müller, S. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4783753
Piccadilly full of people and other foul things
Leksa, V. (2024) · EMBO Reports · doi:10.1038/s44319-024-00143-4
Diversity and Host-Networks of Saproxylic Beetles Along a Forest Recovery Gradient in the TropicsPreprint
Falconí‐López, A., Grella, N., Donoso, D. A., Feldhaar, H., & Müller, J. C. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5151815
Self-citationBowden, J. J., Janke, L. A. A., Young, J. O., Moise, E. R. D., Schmidt, C., & Warren, J. (2025) · The Canadian Entomologist · doi:10.4039/tce.2024.49
Acoustic Trap Design for Biodiversity Detection
Seyidbayli, C., Fengler, B., Szafranski, D., & Reinhardt, A. (2025) · IoT · doi:10.3390/iot6040058
Williams, B., Mars Coral Restoration Project Monitoring Team, Naseem, A., Nava, G., Roberts, A., Nicholson, F., du Luart, A., Erasmus, D., Stoole, O., Whittick, A., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.09.24.678197
The Conservation of Forest Biodiversity of India Through the Integration of Law and AI/ML
Kothiyal, M., Singhal, A. K., Mehta, S., Tomar, A. K., Kaur, G., Dixit, A. K., & Saxena, S. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-6060-5_11
Pelanda, A. M., & Diógenes, A. N. (2025) · Revista Brasileira de Ciências Ambientais · doi:10.5327/z2176-94782514
Prado, C. V., Raya, J. P. B., & Sánchez, R. S. (2025) · Acoustics Australia · doi:10.1007/s40857-025-00370-2
Pallewatta, P., Arachchi, S. M., Karunanayaka, K. T., Kayuni, T., Aleksandr, T., & Halloluwa, T. (2025) · SLAAI International Conference on Artificial Intelligence (SLAAI-ICAI) · doi:10.1109/slaai-icai68534.2025.11318525
The application of AI technology in the field of sound
Wang, K., Shi, J., & Wang, C. (2026) · International Conference on Mechanical, Engineering, and Interaction Design (ICMEID) · doi:10.1117/12.3088780
Detection of Ecosystem Changes Using Bioacoustics Techniques
Fancy, C., Florence, S. M., Anand, M., & Raymond, V. J. (2026) · Springer eBooks · doi:10.1007/978-3-032-06088-4_10
Seibold, S., Rabl, D., Sorger, P., Bittermann, J., Blüthgen, N., Brehm, G., Busse, A., la Hoz, M. C. D., Kortmann, M., Kriegel, P., et al. (2026) · Insect Conservation and Diversity · doi:10.1111/icad.70059
Self-citationLuther, D., Campos‐Cerqueira, M., Medeiros, A. S. D. M., McCarthy, A. J., Roberts, E., Li, X., Bobrowiec, P. E. D., Wolfe, J. D., Leite, G. A., de Melo, T. N., et al. (2026) · Environmental and Sustainability Indicators · doi:10.1016/j.indic.2026.101212
Self-citationGrella, N., Donoso, D. A., Müller, J. C., Falconí‐López, A., Busse, A., Kriegel, P., Püls, M., Rabl, D., Seibold, S., & Feldhaar, H. (2026) · Oecologia · doi:10.1007/s00442-026-05875-9
Self-citationSelf-powered acoustic sensors with hierarchical porous structure for animal voiceprint recognition
Cai, Y., Xia, W., Wang, S., Chen, D., Huang, B., & Wang, Y. (2026) · Chemical Engineering Journal · doi:10.1016/j.cej.2026.176693
Saky, S. M. A. I., Yap, W., Nisar, H., Swee, T. T., Mokayed, H., & Hum, Y. C. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133774
Interacciones del cambio climático y pérdida de biodiversidad en ecosistemas vulnerables.
Zambrano, K. C. H., & Guerrero-Calero, J. M. (2025) · Revista de Investigación y Liderazgo · doi:10.63618/omd/revinvlid/v2/n3/39
Twelve quick tips for applying deep learning to animal sounds
Ghani, B., Baier, A. L., Kalkman, V. J., & Stowell, D. (2026) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1014604 · Cited by 1
Guan, S., Liao, Z., Han, X., & Niu, S. (2026) · iScience · doi:10.1016/j.isci.2026.116422 · Cited by 1
Püls, M., Kortmann, M., Blüthgen, N., Schaefer, M. H., Rabl, D., Thompson, M., Grella, N., Bittermann, J., Schreier, H., Sorger, P., et al. (2026) · Journal of Animal Ecology · doi:10.1111/1365-2656.70359
Self-citationDo Audio-Visual Large Language Models Really See and Hear?
Selvakumar, R. K., Jayakumar, K., Sakshi, S., Ghosh, S., Gao, R., & Manocha, D. (2026) · doi:10.1109/cvprf71909.2026.00566
Eisen, M. B. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.18.759993
Metadata
- Authors
- Müller, J., Mitesser, O., Schaefer, H. M., Seibold, S., Busse, A., Kriegel, P., Rabl, D., Gelis, R., Arteaga, A., Freile, J., Leite, G. A., de Melo, T. N., LeBien, J., Campos-Cerqueira, M., Blüthgen, N., Tremlett, C. J., Böttger, D., Feldhaar, H., Grella, N., Falconí-López, A., Donoso, D. A., Moriniere, J., & Buřivalová, Z.
- Affiliations
- University of Würzburg (DE)
- Fundación de Conservación Jocotoco (EC)
- Technical University of Munich (DE)
- Mpala Research Center and Wildlife Foundation (KE)
- Khamai Foundation (EC)
- Technische Universität Darmstadt (DE)
- Friedrich Schiller University Jena (DE)
- University of Bayreuth (DE)
- Universidad de Las Américas (EC)
- University of Wisconsin–Madison (US)
- Corresponding author
- Müller, Jörg
- ORCID
- Venue
- Nature Communications
- Published
- 2023-10-17
- Volume
- 14
- Issue
- 1
- Pages
- 6191-6191
- ISSN
- 2041-1723 (ISSN-L)
- PMC
- PMC10582010
- OpenAlex
- W4387701637
- Cited by
- 89
- References
- 78
- Citations by year
- 2026: 26 · 2025: 40 · 2024: 19 · 2023: 4
- FWCI
- 15.76
- Type
- journal article
- Language
- en
- Update notices
- correction (2023-11-02)
- Open access
- gold
- doi.org
- Locations
- Nature Communications (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Online Publication Service of Würzburg University (Würzburg University) (submittedVersion, cc-by)
- Common Library Network (Der Gemeinsame Bibliotheksverbund) (submittedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- TUbilio (Technical University of Darmstadt) (submittedVersion)
- Funders
- Fundación de Conservación Jocotoco
- Deutsche Forschungsgemeinschaft (FOR 5207)
- Prince Albert II of Monaco Foundation
- Topics
- Physical Sciences › Environmental Science › Nature and Landscape Conservation › Ecology and Vegetation Dynamics Studies
- Physical Sciences › Environmental Science › Ecology › Wildlife Ecology and Conservation
- Physical Sciences › Environmental Science › Ecology › Environmental DNA in Biodiversity Studies
- Keywords
- tropical forest recovery, biodiversity recovery, forest restoration monitoring, bioacoustics, soundscape, convolutional neural networks, bird community composition, metabarcoding, nocturnal insects, community composition, species richness, Ecuador
- MeSH terms
- Deep Learning, Animals, Conservation of Natural Resources, Trees, Tropical Climate, Ecosystem, Biodiversity, Forests
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
