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.
%0 Journal Article
%1 Müller2023
%A Müller, Jörg
%A Mitesser, Oliver
%A Schaefer, H. Martin
%A Seibold, Sebastian
%A Busse, Annika
%A Kriegel, Peter
%A Rabl, Dominik
%A Gelis, Rudy
%A Arteaga, Alejandro
%A Freile, Juan
%A Leite, Gabriel Augusto
%A de Melo, Tomaz Nascimento
%A LeBien, Jack
%A Campos-Cerqueira, Marconi
%A Blüthgen, Nico
%A Tremlett, Constance J.
%A Böttger, Dennis
%A Feldhaar, Heike
%A Grella, Nina
%A Falconí-López, Ana
%A Donoso, David A.
%A Moriniere, Jerome
%A Buřivalová, Zuzana
%D 2023
%J Nature Communications
%K fsfabrik joergmueller myown olivermitesser
%N 1
%P 6191
%R 10.1038/s41467-023-41693-w
%T Soundscapes and deep learning enable tracking biodiversity recovery in tropical forests
%U https://doi.org/10.1038/s41467-023-41693-w
%V 14
%X 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.
@article{Müller2023,
abstract = {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-R{\texttwosuperior} = 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.},
added-at = {2024-03-21T09:42:16.000+0100},
author = {Müller, Jörg and Mitesser, Oliver and Schaefer, H. Martin and Seibold, Sebastian and Busse, Annika and Kriegel, Peter and Rabl, Dominik and Gelis, Rudy and Arteaga, Alejandro and Freile, Juan and Leite, Gabriel Augusto and de Melo, Tomaz Nascimento and LeBien, Jack and Campos-Cerqueira, Marconi and Bl{\"u}thgen, Nico and Tremlett, Constance J. and Böttger, Dennis and Feldhaar, Heike and Grella, Nina and Falconí-López, Ana and Donoso, David A. and Moriniere, Jerome and Buřivalová, Zuzana},
biburl = {https://www.bibsonomy.org/bibtex/2c4f2c415b60167c84619c539ac62dda2/fsfabrik},
day = 17,
doi = {10.1038/s41467-023-41693-w},
interhash = {2bf18216e6f00f10a6b886fdbffc6623},
intrahash = {c4f2c415b60167c84619c539ac62dda2},
issn = {2041-1723},
journal = {Nature Communications},
keywords = {fsfabrik joergmueller myown olivermitesser},
month = oct,
number = 1,
pages = 6191,
timestamp = {2024-03-21T09:49:26.000+0100},
title = {Soundscapes and deep learning enable tracking biodiversity recovery in tropical forests},
url = {https://doi.org/10.1038/s41467-023-41693-w},
volume = 14,
year = 2023
}