bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2024.01.23.576628

MAARU: Multichannel Acoustic Autonomous Recording Unit for spatial ecosystem monitoring.

Heath, B. E.·Le Penru, N. P.·Skinner, J.·Orme, C. D. L.·Sethi, S. S.·Ewers, R. M.·Picinali, L.
View original source ↗Export metadata ↓

Abstract

1.1 - Acoustic localisation, which relies on simultaneous multi-microphone recording, adds spatial information to recorded audio and has been used in ecosystem monitoring to count individuals to improve abundance estimates, locate illegal activities such as logging/poaching, and monitor behaviour such as habitat use or species interactions. Studies have shown many advantages of acoustic localisation, but uptake remains limited as the equipment is often expensive, inaccessible, or only suitable for short-term deployments. 2 -Here, we present a low-cost, open-source, 6-channel recorder built entirely from commercially available components which can be integrated into a solar-powered, networked system. The MAARU (Multichannel Acoustic Autonomous Recording Unit) works in long-term autonomous, passive, and ad-hoc deployments. We introduce MAARUs hardware and software and present the results of lab and field tests investigating the devices durability, localisation accuracy, and other applications. 3 -MAARU provides multichannel data with similar costs and power demands to equivalent omnidirectional recorders. MAARU devices have been deployed in the UK and Brazil, where we have shown MAARUs can accurately localise pure tones up to 6kHz and 65dB bird calls as far as 8m away ({+/-}10{degrees} range, 100% and >60% of signals respectively), louder calls may have even further detection radii. We also show how beamforming can be used on MAARU devices to improve species ID confidence scores by 20%+ and recall by 10%+ when using the BirdNET automated bird identification algorithms. 4 -MAARU is an accessible, low-cost option for those looking to explore spatial soundscape ecology accurately and easily. Ultimately, the added directional element of the multichannel recording provided by MAARU allows for a new type of exploration into sonic environments. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=118 SRC="FIGDIR/small/576628v1_ufig1.gif" ALT="Figure 1"> View larger version (63K): org.highwire.dtl.DTLVardef@1078827org.highwire.dtl.DTLVardef@17bd5b1org.highwire.dtl.DTLVardef@10a6696org.highwire.dtl.DTLVardef@1d518f4_HPS_FORMAT_FIGEXP M_FIG C_FIG

Explore related subjects

ecologyPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Heath, B. E., Le Penru, N. P., Skinner, J., Orme, C. D. L., Sethi, S. S., Ewers, R. M., Picinali, L.. 2024-01-26. MAARU: Multichannel Acoustic Autonomous Recording Unit for spatial ecosystem monitoring.. https://doi.org/10.1101/2024.01.23.576628

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

PlanktonLake-CEREEP- A Freshwater Plankton Image Dataset with Semi-Automated Label Cleaning

Plankton plays a fundamental role in aquatic ecosystems, influencing biogeochemical cycles and serving as a key food source for many organisms. Recent high-throughput imaging technologies enable the rapid acquisition of large volumes of microscopic images, creating new opportunities for monitoring planktonic ecosystems. However, the manual processing and annotation of the vast amounts of data generated by these devices remain time-consuming tasks. In this context, machine learning-based classification models offer a promising solution. In this data paper, we introduce a new labeled freshwater plankton dataset comprising approximately 88,000 images distributed across 43 taxa. We also present the labeling assistance method we used to facilitate dataset annotation. Finally, we present a baseline based on a convolutional neural network (CNN), which achieves a classification accuracy of 93% on our dataset.

ecology↗
bioRxiv2026

A training protocol for human classification of Asian elephant images from trail cameras

Trail cameras have become ubiquitous tools for ecological data collection over recent decades. Despite progress in the development of automated algorithms and artificial intelligence for image classification, our ability to process large volumes of data remain limited by the need for trained human observers to make refined judgements. We provide guidance on placement of trail cameras for observing Asian elephants (Elephas maximus) and outline a protocol for training and testing naive human observers in performing image classifications (age/sex class and group composition) that cannot yet be automated. This process can be used to develop a high-throughput workflow capable of extracting useful data from large volumes of images. Our training material consisted of 14,007 images collected from 6 trail cameras around Udawalawe National Park in Sri Lanka from 2017-2019. In the first stage, expert observers (n=3) trained a group of inexperienced participants (n=4), who engaged in an iterative process to develop a protocol document. The document was then tested on a second set of subjects (n=6) each of whom classified 350 test images in four separate sequential batches using quantitative measures of precision and accuracy. The test set was sampled from 54,435 images from an additional 25 cameras. When compared to expert observers, they achieved a fair level of precision (Fleiss' kappa = 0.247) and 82.6% accuracy. Our approach can usefully be extended to other species and contexts.

ecology↗
bioRxiv2026

Forest belowground productivity and carbon allocation predominantly driven by soil properties rather than climate

Forests are threatened by a multitude of stressors, including anthropogenic disturbances and climate change. Assessing how forests will respond to these stressors requires a comprehensive understanding of net primary productivity (Npp), environmental constraints on growth, and adaptive capacity. A parameter of significant uncertainty is belowground Npp (bNpp), which can account for up to 80% of total Npp but is poorly estimated and rarely measured directly. We used a cross-biome dataset of direct, field-based measurements of aboveground and belowground primary productivity and 21 climatic and soil variables to identify potential constraints on bNpp and belowground carbon allocation in boreal and cold temperate forests. Soil variables, rather than climate variables, were the main drivers of bNpp and belowground allocation across biomes. The importance of soil variables suggests that soil nutrient dynamics, especially soil nutrient pool and flux variables, must be explicitly modeled to more accurately predict feedbacks between climate, productivity, and within-tree carbon allocation. Within biomes, environmental drivers of belowground allocation varied between low versus high allocation forests, indicating that environmental drivers are site-specific and the development of within-biome, site-scale classifications for forest ecosystems could be useful. Changes in soil variables, such as increasing soil nitrogen pools, caused abrupt and large decreases in bNpp for boreal, but not cold temperate forests. Threshold-like shifts indicate that boreal forests might have lower adaptive capacity and higher sensitivity to disturbances than cold temperate forests. With 70% of boreal forests characterized by low bNpp, disturbances such as anthropogenic nitrogen deposition could cause large-scale decreases in bNpp that could push these forests beyond their adaptive capacity.

ecology↗