Search bioRxiv⌕ Search

Biology subjects

Kohles, J. E.

Publications and source records attributed to Kohles, J. E..

2 recordsLinked to original sources
bioRxiv2026

BatSpot: a retrainable neural network for automatic detection and classification of bat echolocation and detection of buzzes and social calls

O_LIBats are a diverse taxonomic group that display a wide range of interesting behaviours. Many bats are keystone species for their ecosystem, are IUCN Red-listed as vulnerable to critically endangered, and subject to human-wildlife conflicts arising from anthropogenic expansion. Yet bats remain understudied both with respect to behaviour, population ecology and conservation status. One of the major challenges when studying bats is obtaining data. Their nocturnal lifestyle and use of ultrasonic echolocation makes them difficult to track and record using traditional methods. Recent advances in passive acoustic monitoring have allowed researchers to record large amounts of data, but the detection and classification of vocalisations remain a challenge. Most available tools are either for profit or are limited to a narrow geographic range, and mostly focus on echolocation search phase calls. C_LIO_LIHere we present BatSpot, a convolutional neural network trained to detect search phase calls, buzzes and social calls. It also offers the option to classify the search phase calls to species(-complex) level. We provide a GUI that allows researchers to retrain or transfer-train the models for their specific needs and validate the performance. C_LIO_LIWe test the performance of all models and show that they perform better than both commercial and open-source solutions (search phase file level F1: 0.97 vs 0.96, buzz detector F1: 0.95 vs 0.11). We furthermore show that retraining the search phase call detector for a new country with examples from just 59 recordings massively improves the performance (F1: 0.48 to 0.79). C_LIO_LIBatSpot will enable bat researchers globally to automate detection and classification with minimal effort and includes novel options for social call and buzz detection, typically not featured in other automated tools for bat monitoring. C_LI

ecology↗
bioRxiv2025

Fine-scale animal proximity detection and localization via multi-sensor biologgers

O_LIAccurately quantifying spatial interactions is central to understanding social behavior, information flow, predator-prey dynamics, and disease transmission. Proximity loggers that record received signal strength indicator (RSSI) offer a promising approach for estimating pairwise distances, particularly in environments where GPS is unavailable or imprecise. However, RSSI is often dismissed as too noisy for fine-scale inference, with performance that depends on environmental conditions, tag orientation, and between-device variability. Incorporating additional tag-measured data may improve RSSI performance and enable its use as a continuous measure of distance in variable environments. C_LIO_LIHere, we assess the utility of continuous RSSI as a fine-scale distance estimator and localization tool using a novel multi-sensor WiFi biologger (WildFi). We conducted four experiments: (1) testing how tag orientation affects RSSI-distance relationships; (2) evaluating whether environmental covariates measured by onboard sensors improve proximity estimates; (3) assessing the accuracy of trilateration-based tag localization using fixed gateway arrays; and (4) comparing RSSI- and GPS-inferred proximity in free-ranging Egyptian fruit bats (Rousettus aegyptiacus). C_LIO_LIWhile RSSI alone could predict distance with reasonable accuracy, incorporating additional tag-sensed information (e.g., temperature, humidity, barometric pressure) and accounting for tag-level heterogeneity significantly improved predictive accuracy. Based on RSSI predictions, we could estimate tag location with a median error of 2.6 meters, accurate enough to indirectly estimate proximity networks without tag-to-tag communication. In deployments on free-flying bats, we found that RSSI and GPS were only weakly concordant, with GPS unreliable for detecting fine-scale interactions (<50 m). In contrast, RSSI could capture both fine-scale and some long-range interactions up to [~]250m. C_LIO_LIThese findings highlight RSSIs potential as a robust metric for proximity logging, particularly when combined with multi-sensor data and pre-deployment validations. Integrating multi-sensor data streams further enhances RSSI interpretability. Future biologger designs should prioritize synergy among data streams for integrated insights into proximity and animal behavior. C_LI Data and code for peer review statementData and code to reproduce the results of the paper are provided in a zip folder for peer review. We also provided our compiled code.

ecology↗