Biology subjects

Houstin, A.

Publications and source records attributed to Houstin, A..

3 recordsLinked to original sources

bioRxiv2023

RFIDeep: Unfolding the Potential of Deep Learning forRadio-Frequency Identification

O_LIAutomatic 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. C_LIO_LIWe 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. C_LIO_LITo 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%. C_LIO_LIRFIDeep 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. C_LI

ecology↗

bioRxiv2021

Biologging of emperor penguins - attachment techniques and associated deployment performance

An increasing number of marine animals are equipped with biologgers, to study their physiology, behaviour and ecology, often for conservation purposes. To minimise the impacts of biologgers on the animals welfare, the Refinement principle from the Three Rs framework (Replacement, Reduction, Refinement) urges to continuously test and evaluate new and updated biologging protocols. Here, we propose alternative and promising techniques for emperor penguin (Aptenodytes forsteri) capture and on-site logger deployment that aim to mitigate the potential negative impacts of logger deployment on these birds. We equipped adult emperor penguins for short-term (GPS, Time-Depth Recorder (TDR)) and long-term (i.e. planned for one year) deployments (ARGOS platforms, TDR), as well as juvenile emperor penguins for long-term deployments (ARGOS platforms) in the Weddell Sea area where they had not yet been studied. We describe and qualitatively evaluate our protocols for the attachment of biologgers on-site at the colony, the capture of the animals and the recovery of the devices after deployment. We report unprecedented recaptures of long-term equipped adult emperor penguins (50% of equipped individuals recaptured after 290 days). Our data demonstrate that the traditional technique of long-term attachment by gluing the biologgers directly to the back feathers is detrimental to the birds. It causes excessive feather breakage and the loss of the devices at an early stage. We therefore propose an alternative method of attachment for back-mounted devices. This technique led to successful year-round deployments on 37.5% of the equipped juveniles. Finally, we also disclose the first deployments of leg-bracelet mounted TDRs on emperor penguins. Our findings highlight the importance of monitoring potential impacts of biologger deployments on the animals and the need to remain critical towards established and new protocols.

ecology↗

bioRxiv2021

Juvenile emperor penguin range calls for extended conservation measures in the Southern Ocean

To protect the unique Southern Ocean biodiversity, conservation measures like marine protected areas (MPAs) are implemented based on the known habitat distribution of ecologically important species. However, distribution models focus on adults, neglecting that immatures animals can inhabit vastly different areas. Here, we show that current conservation efforts in the Southern Ocean are insufficient for ensuring the protection of the highly mobile Emperor penguin. We find that juveniles spend [~]90% of their time outside the boundaries of proposed and existing MPAs, and that their distribution extends far beyond (> 1500 km) the species extent of occurrence as defined by the International Union for Conservation of Nature. We argue that strategic conservation plans for Emperor penguin and long-lived ecologically important species must consider the dynamic habitat range of all age classes.

ecology↗