BioChocó

An integrated biodiversity monitoring network across a forest-to-farm landscape in the Chocó of western Ecuador.

Download
Data current September 2, 2026Camera traps · passive acoustics · habitat · endangered species

Objectives

The Chocó is one of the most biodiverse rainforests on Earth, and more than 95% of its original forest is already gone. What remains is a mosaic of forest, cacao farms, and pasture. How does biodiversity respond to these shifts in land use? How can we design conservation interventions to maximize benefits for local communities and biodiversity? FCAT is working to answer these questions.

BioChocó is one part of a broader, community-led conservation and research program led by FCAT since 2003, employing two dozen local residents, pairing field research with education and capacity building, and protecting >750 ha and ~50 endangered species with a multi-use field station that receives hundreds of visitors per year. Our approach has produced more than 50 peer-reviewed scientific papers with local coauthors and awards including the Whitley Prize for Nature. BioChocó is run by four local field biologists — known as FCATero/as — and works with 50 local farmers who host monitoring stations on their land.

Our main objectives are to:

01

Track conservation outcomes

Follow the real-time conservation outcomes of FCAT's corridor-building and restoration interventions, so we can see how biodiversity recovers as it happens.

02

Understand land-use change

Measure how biodiversity and ecosystems respond to land-use change across a gradient of land use types, from primary forest through cacao agroforestry to open pasture.

03

Assess priority species

Provide scientifically rigorous habitat-use assessments for the Chocó's endangered and keystone species across birds, mammals, frogs, and insects.

04

Elevate the Chocó locally & globally

Raise local & global awareness of and engagement with this understudied rainforest through personalized farmer-reports and an open, real-time data platform.

How each biodiversity monitoring station works

Every biodiversity monitoring station within the Biochocó network is built around a stack of automated sensors paired with field observations, so a single visit captures multiple aspects of biodiversity, local microclimate, and forest structure together.

Motion-activated camera

Trail camera · photo mode

Runs at least 30 days per deployment. We pass photos through a custom species classifier we fine-tuned for the Chocó and each image is reviewed and verified.

Passive acoustic recorder

Song Meter Micro 2 · 48 kHz

Records soundscapes to capture calls and songs from birds, mammals, frogs, and insects, recording one minute in every ten.

Microclimate logger

iButton Thermochron

Logs temperature at 30 minute intervals through the deployment, giving a paired record of the microclimate each camera and recorder experienced.

Habitat structure

Field survey · drone overflight

The field team measures canopy cover, tree size and density, and forest cover around each station, then links structure to what the sensors detect.

What types of habitats are we monitoring?

Monitoring stations are spread across the full gradient of land use types in the region, from primary and regenerating forest to open pasture, along with three cacao farming systems.

Primary forest
Primary forest
Relatively undisturbed old growth
14 sites sampled
Secondary forest
Secondary forest
5–30 years of regeneration
15 sites sampled
Cacao (Nacional)
Cacao (Nacional)
Mature shade-grown cacao, native variety
15 sites sampled
Cacao (GIZ)
Cacao (GIZ)
Young shade-grown agroforestry, under 5 years
15 sites sampled
Cacao (CCN)
Cacao (CCN)
Full-sun cacao, no shade trees
15 sites sampled
Reforestation
Reforestation
Active restoration plantings as part of the Choconexion project
12 sites sampled
Pasture
Pasture
Open, grazed land
14 sites sampled

Where the network stands today

66
monitoring deployments
64 camera · 65 audio · 65 climate
2,168
camera-trap days
Jan 2026 – Jul 2026
26,830
camera-trap photos
collected in the field
14,706
animal detections
animals found in photos
38
species on camera
19 mammals · 18 birds
435,933
audio recordings
2.34 TB of sound
346,172
bird detections by sound
BirdNET at ≥0.8 confidence
118,973
microclimate readings
84 temperature loggers

Where we are working

On the map below, each point is a monitoring site, colored by habitat type, spanning a network of ~50 local farms. Pan and zoom to explore; the dashed line marks the FCAT reserve boundary.

Species detections

Camera identifications are verified or corrected by FCAT biologists. Bird detections come from automated BirdNET analysis, filtered to a confidence of 0.8 or higher.

On camera

38 species identified so far. The most-detected wild species:

Central American agouti Dasyprocta punctata1,160
Western Andean Long-nosed Armadillo Dasypus fenestratus738
Collared peccary Dicotyles tajacu509
Northern Black-eared Opossum Didelphis marsupialis311
Lowland paca Cuniculus paca177
Common pauraque Nyctidromus albicollis128
White-nosed Coati Nasua narica112
Tome's Spiny-rat Proechimys semispinosus82
Crab-eating Raccoon Procyon cancrivorus75

By sound

Most-detected birds across 435,933 recordings:

Yellow-throated Toucan Ramphastos ambiguus102,604
Common Pauraque Nyctidromus albicollis23,254
Chocó Toucan Ramphastos brevis19,815
Little Tinamou Crypturellus soui17,072
Rufous-headed Chachalaca Ortalis erythroptera13,349
Pacific Antwren Myrmotherula pacifica11,789
Slaty Spinetail Synallaxis brachyura10,709
Ruddy Pigeon Patagioenas subvinacea9,668
Blue-black Grassquit Volatinia jacarina9,201

From the field

Ocelot walking along a trail at night
Leopardus pardalis
Ocelot walking along a trail at night
A collared peccary often seen traveling in groups
Dicotyles tajacu
A collared peccary often seen traveling in groups
Lowland paca foraging at night
Cuniculus paca
Lowland paca foraging at night
Central American agouti eating a fallen fruit, the forest's most-photographed mammal
Dasyprocta punctata
Central American agouti eating a fallen fruit, the forest's most-photographed mammal
A white-nosed coati nosing through the leaf litter by day
Nasua narica
A white-nosed coati nosing through the leaf litter by day
Tayra, a large diurnal weasel, on the move
Eira barbara
Tayra, a large diurnal weasel, on the move
Crested guan foraging on the forest floor
Penelope purpurascens
Crested guan foraging on the forest floor
Great tinamou, more often heard than seen
Tinamus major
Great tinamou, more often heard than seen
  • Ramphastos ambiguus — A Yellow-throated Toucan calling from the canopy
    Spectrogram — Ramphastos ambiguus
  • Ramphastos brevis — Chocó Toucan, a Chocó endemic
    Spectrogram — Ramphastos brevis
  • Cephalopterus penduliger — Long-wattled Umbrellabird, a threatened Chocó endemic and flagship species for FCAT
    Spectrogram — Cephalopterus penduliger
  • Ortalis erythroptera — Rufous-headed Chachalaca, an endangered endemic, at dawn
    Spectrogram — Ortalis erythroptera
  • Pulsatrix perspicillata — Spectacled Owl calling after dark
    Spectrogram — Pulsatrix perspicillata
  • Tinamus major — Great Tinamou's tremulous whistle at dusk
    Spectrogram — Tinamus major

An integrated end-to-end platform

FCAT has built an end-to-end platform that integrates raw sensor files into verified biodiversity data to share with local residents and the broader public. Camera traps, passive audio recorders, and microclimate loggers all feed a single pipeline, where our custom camera trap species classifier and BirdNET propose identifications, biologists verify the results, and species habitat use models refit as new data is processed. The data flows both back to the landowners hosting the sensors through their own personalized site pages and also to the global research and conservation community via standardized data outputs (e.g., Camtrap DP).

Custom species classifierA fine-tuned BioCLIP model for the species of the Chocó is built into a custom camera-trap annotation pipeline to review and correct identifications.
FCAT Portal · Clasificador de especies
Custom species classifier
Bioacoustics with BirdNETContinuous day-and-night recordings run through BirdNET, with an interactive spectrogram for verifying species identifications.
FCAT Portal · Audio
Bioacoustics with BirdNET
Live habitat use modelingSingle-species occupancy models that refit automatically as new camera and audio detections arrive, rendered as predicted-occurrence maps.
FCAT Portal · Ocupación
Live habitat use modeling
Personalized landowner pagesEach participating landowner gets their own page that includes species recorded on their land and information about species' ecological role management.
FCAT Portal · Fichas por sitio
Personalized landowner pages
Also in the platform
  • Open-data export in the Camtrap DP standard, published to GBIF and the Environmental Data Initiative.
  • Per-station microclimate records from iButton loggers — temperature series, deployment window, and coverage.
  • A field schedule that tracks every sensor installation and retrieval across the network.

Where collaborators come in

We designed the BioChocó network as a shared foundation for researchers. FCAT collaborates with computer scientists and ecologists at Tulane University, researchers at Universidad San Francisco de Quito, Virginia Tech University, University of Hawaii, the Cornell Lab of Ornithology, among many others, so collaborators plug into an active, well-connected effort.

Opportunities for collaboration

  • Occupancy and community modelingRepeated visits across the land-use gradient support single- and multi-species occupancy and biodiversity studies on birds, mammals, frogs, and insects, drawing on verified species records from camera traps and passive audio recorders.
  • Restoration and corridor outcomes over timeFCAT is building a biodiversity corridor and running two large-scale restoration and agroforestry projects now, with monitoring stations already in the ground. Collaborators get a before-and-after record with the land-use gradient as built-in reference sites, which supports work on how quickly animal communities recover and whether a new corridor promotes the movement it was designed for.
  • Machine learning across images and audioA labeled, growing image dataset for classifier training and detection benchmarks, built on our fine-tuned BioCLIP and MegaDetector, plus terabytes of day-and-night audio recordings.
  • Agroforestry and sustainable farm designMonitoring stations sit on roughly 50 working farms, spanning three cacao management systems alongside forest and pasture. That setup can show which agricultural practices promote biodiversity on the farm while still being profitable: shade tree composition, canopy structure, patch size, distance to remaining forest.
  • Microclimate and thermal ecologyEvery station logs temperature every 30 minutes for the length of its deployment, alongside its own camera, audio, and forest-structure record. That pairing supports work on how much heat cacao shade and forest canopy buffer compared to open pasture, and where thermal refugia sit for frogs and insects as the Chocó warms.
  • People and land-use decisionsRoughly 50 landowners host stations and receive a personalized page showing what lives on their land. What practices can help promote and incentivize biodiversity-friendly land-use decisions?
  • Additional taxa or sensors!We've built the network of 50 farms with landowner agreements in place, and a resident field team that visits every station on a schedule. Collaborators can add a method to sites that already run, whether that is eDNA sampling, bat detectors, arthropod traps, or vegetation plots, and build on the paired camera, audio, microclimate, and structure data at each one.

Get in touch!

If your research could use camera-trap, acoustic, or microclimate data from a Chocó forest landscape, or you want to co-design a study around the network, please reach out.

Luke Browne
Monitoring lead
lukebrowne@fcat-ecuador.org
Luis Carrasco
FCAT Reserve Director
luiscarrasco@fcat-ecuador.org
Jordan Karubian
FCAT co-founder
jordankarubian@fcat-ecuador.org
Melissa Loayza
Program Director
melissaloayza@fcat-ecuador.org
Karla Zambrano
Field Coordinator
Luis Zambrano
Field Coordinator
Gregory Paladines
Local biologist (FCATero)
Gloria Loor
Local biologist (FCATera)
Julio Loor
Local biologist (FCATero)
Darwin Zambrano
Local biologist (FCATero)

Funders and acknowledgements

Funders

Support for the BioChocó monitoring network comes from:

  • Wedgetail Foundation
  • National Science Foundation
  • Private donors

Institutional support

  • Ministry of the Environment, Water and Ecological Transition (MAATE)
    Research permitting and institutional support
Fundación para la Conservación de los Andes Tropicales
BioChocó biodiversity monitoring network · Chocó, Ecuador
Data current as of September 2, 2026