Téo Barracho

Téo Barracho

Auto-generated profile, based on publicly available information

Téo Barracho

Département de Biologie Polaire, Centre Scientifique de Monaco, Monaco, Monaco

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 August 2023 in Methods in Ecology and Evolution
Verified
Journal Article
Open access
GBGaël Bardon
Gaël Bardon
RCRobin Cristofari
Robin Cristofari
AWAlexander Winterl
Alexander Winterl
TBTéo Barracho
Téo Barracho
MBMarine Benoiste
Marine Benoiste
CBCéline Le Bohec
Céline Le Bohec
Automatic monitoring of wildlife is becoming a critical tool in the field of ecology. In particular, Radio‐Frequency IDentification (RFID) is now a widespread technology to assess the phenology, breeding and survival of many species. While RFID produces massive datasets, no established fast and accurate methods are yet available for this type of data processing. Deep learning approaches have been used to overcome similar problems in other scientific fields and hence might hold the potential to overcome these analytical challenges and unlock the full potential of RFID studies. We present a deep learning workflow, coined “RFIDeep”, to derive ecological features, such as breeding status and outcome, from RFID mark‐recapture data. To demonstrate the performance of RFIDeep with complex datasets, we used a long‐term automatic monitoring of a long‐lived seabird that breeds in densely packed colonies, hence with many daily entries and exits. To determine individual breeding status and phenology and for each breeding season, we first developed a one‐dimensional convolution neural network (1D‐CNN) architecture. Second, to account for variance in breeding phenology and technical limitations of field data acquisition, we built a new data augmentation step mimicking a shift in breeding dates and missing RFID detections, a common issue with RFIDs. Third, to identify the segments of the breeding activity used during classification, we also included a visualisation tool, which allows users to understand what is usually considered a “black box” step of deep learning. With these three steps, we achieved a high accuracy for all breeding parameters: breeding status accuracy = 96.3%; phenological accuracy = 86.9%; and breeding success accuracy = 97.3%. RFIDeep has unfolded the potential of artificial intelligence for tracking changes in animal populations, multiplying the benefit of automated mark‐recapture monitoring of undisturbed wildlife populations. RFIDeep is an open source code to facilitate the use, adaptation, or enhancement of RFID data in a wide variety of species. In addition to a tremendous time saving for analysing these large datasets, our study shows the capacities of CNN models to autonomously detect ecologically meaningful patterns in data through visualisation techniques, which are seldom used in ecology.

13 February 2025 in Polar Biology
Verified
Journal Article
Closed access
PDF file
ESEwen Le Scornec
Ewen Le Scornec
JAJimmy Allain
Jimmy Allain
GBGaël Bardon
Gaël Bardon
TBTéo Barracho
Téo Barracho
NCNicolas Chatelain
Nicolas Chatelain
CBChristophe Barbraud
Christophe Barbraud

13 March 2026 in Science Advances
Verified
Journal Article
Open access
GBGaël Bardon
Gaël Bardon
TBTéo Barracho
Téo Barracho
JDJoël M. Durant
Joël M. Durant
NLNicolas Lecomte
Nicolas Lecomte
YMYvon Le Maho
Yvon Le Maho
CBCéline Le Bohec
Céline Le Bohec
Climate-driven phenological mismatches threaten avian reproduction by disrupting food availability during critical breeding stages. In marine ecosystems, time lags between environmental changes and their effects on food webs are challenging to model, yet they can have a profound impact on reproduction at higher trophic levels. We disentangle how oceanic and climatic variability influence the breeding phenology and success of a major predator species in the Southern Ocean, drawing on 24 years of data from 17,000 marked king penguins. We document a 19-day advancement in breeding phenology, alongside increased breeding success (44% in 2000 and 62% in 2023). A sliding-window analysis reveals that sea temperature and primary production in key foraging zones predict both phenology and breeding success, with lags up to several weeks and 2 years, respectively. The amplitude of their recent response, as well as their dependence on multiyear environmental conditions, underscores a potential vulnerability in a context of decreasing environment predictability. However, so far, king penguins appear to be keeping pace with environmental changes and demonstrate a remarkable resilience and important flexibility in the face of variable oceanographic conditions.

26 August 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
TBTeo Barracho
Teo Barracho
GBGael Bardon
Gael Bardon
RCRemi Choquet
Remi Choquet
AHAymeric Houstin
Aymeric Houstin
AWAlexander Winterl
Alexander Winterl
CBCeline Le Bohec
Celine Le Bohec
As summer sea ice around Antarctica reaches historical lows, quantifying the demographic response of polar species to such environmental changes becomes critical. To achieve this, synthesizing results from across species' ranges and elucidating the environmental factors driving population dynamics are key. Adelie penguins are considered reliable indicators of changes in Antarctica but the pathways through which sea ice and other environmental factors shape their population dynamics are still unclear, especially for the juvenile stage. Using a 17-year dataset of electronically tagged Adelie penguins from Adelie Land (Antarctica), we found that juvenile survival probability was most impacted by sea ice concentration near their natal colony right after fledging, with lower ice concentrations detrimental to survival. Importantly, we found that juvenile survival declined by 32% from 2007 to 2020, mirroring trends at other distant colonies. The emergence of similar patterns at opposite ends of the continent may be an early signal for shifts in population trends expected from climate change.

25 August 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
AWAlexander Winterl
Alexander Winterl
SRSebastian Richter
Sebastian Richter
AHAymeric Houstin
Aymeric Houstin
TBTéo Barracho
Téo Barracho
MBMatthieu Boureau
Matthieu Boureau
DZDaniel P. Zitterbart
Daniel P. Zitterbart
Emperor penguins ( Aptenodytes forsteri ) are under increasing environmental pressure. Monitoring colony size and trends of this Antarctic seabird relies primarily on satellite imagery recorded near the end of the breeding season, when illumination levels are sufficient to capture images, but colony occupancy is highly variable. To correct population estimates for this variability, we develop a phenological model that accurately predicts the number of breeding pairs and fledging chicks, as well as key phenological events such as arrival, hatching and foraging times, from as few as six data points from a single season. The ability to extrapolate occupancy from sparse data makes the model particularly useful for monitoring remotely sensed animal colonies where ground-based population estimates are very rare or unavailable. Teaser: The Emperor penguin becomes the Southern Ocean’s canary in a coal mine through remote sensing its annual breeding success.

27 March 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
GBGaël Bardon
Gaël Bardon
RCRobin Cristofari
Robin Cristofari
AWAlexander Winterl
Alexander Winterl
TBTéo Barracho
Téo Barracho
MBMarine Benoiste
Marine Benoiste
CBCéline Le Bohec
Céline Le Bohec
Automatic monitoring of wildlife is becoming a critical tool in the field of ecology. In particular, Radio-Frequency IDentification (RFID) is now a widespread technology to assess the phenology, breeding, and survival of many species. While RFID produces massive datasets, no established fast and accurate methods are yet available for this type of data processing. Deep learning approaches have been used to overcome similar problems in other scientific fields and hence might hold the potential to overcome these analytical challenges and unlock the full potential of RFID studies. We present a deep learning workflow, coined “RFIDeep”, to derive ecological features, such as breeding status and outcome, from RFID mark-recapture data. To demonstrate the performance of RFIDeep with complex datasets, we used a long-term automatic monitoring of a long-lived seabird that breeds in densely packed colonies, hence with many daily entries and exits. To determine individual breeding status and phenology and for each breeding season, we first developed a one-dimensional convolution neural network (1D-CNN) architecture. Second, to account for variance in breeding phenology and technical limitations of field data acquisition, we built a new data augmentation step mimicking a shift in breeding dates and missing RFID detections, a common issue with RFIDs. Third, to identify the segments of the breeding activity used during classification, we also included a visualisation tool, which allows users to understand what is usually considered a “black box” step of deep learning. With these three steps, we achieved a high accuracy for all breeding parameters: breeding status accuracy = 96.3%; phenological accuracy = 86.9%; breeding success accuracy = 97.3%. RFIDeep has unfolded the potential of artificial intelligence for tracking changes in animal populations, multiplying the benefit of automated mark-recapture monitoring of undisturbed wildlife populations. RFIDeep is an open source code to facilitate the use, adaptation, or enhancement of RFID data in a wide variety of species. In addition to a tremendous time saving for analyzing these large datasets, our study shows the capacities of CNN models to autonomously detect ecologically meaningful patterns in data through visualisation techniques, which are seldom used in ecology.