Biodiversity Monitoring in Brazil’s Juruá Region

Brazil

Overview

DEEP partnered with Rainforest Connection (RFCx) and Instituto Juruá to understand biodiversity patterns in and around floodplain lakes of the Juruá River, in the Brazilian Amazon.

Rainforest Connection (RFCx) partnered with Instituto Juruá to understand biodiversity patterns in and around floodplain lakes of the Juruá River, in the Brazilian Amazon, using passive acoustic monitoring (PAM).

More specifically, the project aims to test if lake protection by the community through fisheries management has a positive effect on other non-aquatic fauna such as bird, primate and anuran species.

Partners

  • Instituto Juruá

  • DEEP

Objectives

  • Evaluated how different protections and community-based conservation strategies influenced biodiversity levels near these floodplain lakes.

  • Test if lake protection by the community through fisheries management has a positive effect on other non-aquatic fauna such as bird, primate and anuran species.

  • To understand if local community protection of floodplain lakes has a positive impact on
    overall biodiversity and soundscape.

  • Study focused on comparing the acoustic characteristics of protected and unprotected lakes, as well as within the forest surrounding these lakes.

Implementation

Deployed specialized acoustic recorders at 144 sites across 18 floodplain lakes (9 under community protection & nine without protection) from September-November 2022.

Passive acoustic monitoring (PAM) was conducted in 18 floodplain lakes in the region of the Juruá River, Western Brazilian Amazon, with nine designated as community protected and nine as unprotected.

AudioMoth acoustic loggers from Open Acoustic Devices were used to collect 60-second audio recordings every eleven minutes (around 130 recordings per day) at each site for a minimum period of approximately one week from September to November 2022 . A total of 131,706 1-minute recordings were collected by AudioMoth recorders and uploaded to the Arbimon platform by on-ground Instituto Juruá partners.

Two biologists from the RFCx science team with extensive experience in audio surveys (GL and TM) used the Arbimon online platform to listen to recordings and manually inspect the spectrograms of one day from each sampling site (i.e., first or last survey day) to create a preliminary list of species from the study area.

We created PM models for 51 species (Appendix B) in Arbimon to automatically identify the species in the audio recordings using the parameterization.

Overall, 4,685 detections generated by the PM analyses were manually validated.

Limitations

Data collection was carried out during the dry season, and in floodplains the season may influence the detection of some species as well as the soundscape patterns, which can change drastically during the rainy season (Melo et al. 2022). With additional data collection in the flood season we would be able to compare species occurrences and soundscape outcomes in both seasons and assess the variation of acoustic variability between the seasons and population dynamics for many vertebrate species.

Impact

Outcomes:

  • We detected 174 species in total (164 birds, 5 mammals, 5 amphibians), including 4 species threatened with extinction:

  1. Wattled curassow (Crax globulosa)

  2. Ringed woodpecker (Celeus torquatus)

  3. Zimmer's woodcreeper (Dendroplex kienerii)

  4. Razor-billed curassow (Mitu tuberosum)

  • We deployed specialized acoustic recorders at 144 sites across 18 floodplain lakes (9 under community protection & nine without protection). from September-November 2022. We detected 174 species in total (164 birds, 5 mammals, 5 amphibians), including 4 species threatened with extinction: wattled curassow (Crax globulosa), ringed woodpecker (Celeus torquatus), Zimmer’s woodcreeper (Dendroplex kienerii), and razor-billed curassow (Mitu tuberosum). We developed species detection and identification models for 51 of the most ecologically important of these species.

  • We used proven acoustic-based proxies (soundscape metrics) for species richness and composition to show that protected lake sites have higher species diversity than unprotected lake sites. This result highlights the positive impacts that community-led conservation and management can have on biodiversity, and underscores the need to continue supporting grassroots conservation initiatives.

  • This project has enabled real-time ecoacoustic monitoring capabilities in the region, providing important insights on the effects that protected status has on biodiversity around floodplain lakes in the Brazilian Amazon. Project results have been made publicly available through an intuitive and digestible Insights page to facilitate collaboration and knowledge-sharing.

  • These analyses suggest that protected floodplain lakes have a unique, and more variable soundscape signature, indicating that soniferous animal species differ between protected and unprotected lakes.

  • The results suggest that lake protection conducted by local communities through fisheries management is spilling over beyond the lake boundaries and positively impacting soniferous fauna, creating a higher richness of soniferous species around protected lakes than unprotected ones.

Resources

Resources

Stay Connected

Sign up for our newsletter
Receive real-impact updates

Get early access to tours, volunteer missions and invitations to exclusive webinars and events about bioacoustics, climate tech, and sustainability.

We respect your privacy. We also hate spam.

Stay Connected

Sign up for our newsletter
Receive real-impact updates

Get early access to tours, volunteer missions and invitations to exclusive webinars and events about bioacoustics, climate tech, and sustainability.

We respect your privacy. We also hate spam.

Stay Connected

Sign up for our newsletter
Receive real-impact updates

Get early access to tours, volunteer missions and invitations to exclusive webinars and events about bioacoustics, climate tech, and sustainability.

We respect your privacy. We also hate spam.