Stephen Marsland

Stephen Marsland

Auto-generated profile, based on publicly available information

Stephen Marsland

Victoria University of Wellington

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
23 May 2018 in Journal of Avian Biology
Verified
Journal Article
Closed access
PDF file
NPNirosha Priyadarshani
Nirosha Priyadarshani
SMStephen Marsland
Stephen Marsland
ICIsabel Castro
Isabel Castro

9 January 2020 in Methods in Ecology and Evolution
Verified
Journal Article
Open access
PDF file
YFYukio Fukuzawa
Yukio Fukuzawa
WWWesley H. Webb
Wesley H. Webb
MPMatthew D. M. Pawley
Matthew D. M. Pawley
MRMichelle Mary Roper
Michelle Mary Roper
SMStephen Marsland
Stephen Marsland
AGAndrew Gilman
Andrew Gilman
Classifying acoustic units is often a key step in studying repertoires and sequence structure in animal communication. Manual classification by eye and ear remains the primary method, but new tools and techniques are urgently needed to expedite the process for large, diverse datasets. Here we introduce Koe, an application for classifying and analysing animal vocalizations. Koe offers bulk‐labelling of units via interactive ordination plots and unit tables, as well as visualization and playback, segmentation, measurement, data filtering/exporting and new tools for analysing repertoire and sequence structure – in an integrated environment. We demonstrate Koe with a real‐world case study of New Zealand bellbird Anthornis melanura songs from an archipelago metapopulation. Having classified 21,500 units in Koe, we compare repertoires and sequence structure between sites and sexes. Koe is web‐based (koe.io.ac.nz.) and easy to use, making it ideal for collaboration, education and citizen science. By enabling large‐scale, high‐resolution classification and analysis of animal vocalizations, Koe expands the possibilities for bioacoustics research.

25 January 2020 in Methods in Ecology and Evolution
Verified
Journal Article
Closed access
PDF file
NPNirosha Priyadarshani
Nirosha Priyadarshani
SMStephen Marsland
Stephen Marsland
JJJulius Juodakis
Julius Juodakis
ICIsabel Castro
Isabel Castro
VLVirginia Listanti
Virginia Listanti

12 July 2022 in Methods in Ecology and Evolution
Verified
Journal Article
Open access
JJJulius Juodakis
Julius Juodakis
SMStephen Marsland
Stephen Marsland
Sound recordings are used in various ecological studies, including wildlife monitoring by acoustic surveys. Such surveys often require automatic detection of target sound events in the large amount of data produced. However, current processing methods, especially those relying on sound intensity for detection, are severely impacted by wind, which causes transient intensity peaks. The rapid dynamics of this noise invalidate standard noise estimators, and no satisfactory method for dealing with wind exists in bioacoustics, where simple training and generalization between conditions are important. We estimate the transient noise level by fitting short‐term spectrum models to a wavelet packet representation. This estimator is then combined with log‐spectral subtraction to stabilize the background level. The resulting adjusted wavelet series can be analysed by standard detectors. We use real data from long‐term acoustic monitoring to tune this workflow, demonstrate its denoising capabilities and test the improved detection in two population surveys of birds. The proposed short‐term estimator was more effective than standard (constant) noise estimates in both denoising and detection tasks. In the surveys, the noise‐robust workflow greatly reduced the number of false alarms. As a result, the survey efficiency (precision of the estimated call density) improved for both species. In contrast to existing methods, the proposed estimator can adjust for transient broadband noises without requiring additional hardware or extensive tuning to each species. It improved the detection workflow based on very little training data, making it particularly attractive for detection of rare species or general soundscape analysis.

17 June 2019 in Methods in Ecology and Evolution
Verified
Journal Article
Open access
PDF file
SMStephen Marsland
Stephen Marsland
NPNirosha Priyadarshani
Nirosha Priyadarshani
JJJulius Juodakis
Julius Juodakis
ICIsabel Castro
Isabel Castro
The routine collection of long‐time acoustic recordings of animals in the field presents new challenges in data analysis. While many terabytes of data are collected annually, effective use of this noisy, highly variable data require skilled humans to manually identify calls. While computer programs to automatically analyse these recordings are becoming available, it is important that they are user‐friendly and easy‐to‐use, so that everybody – citizen scientists, wildlife managers, researchers – can take advantage of them, and that they keep the human in the loop so analyses carried out this year are comparable both to manual call counts from the past, and more accurate automated analyses performed in the future. We present the AviaNZ program, which is designed to achieve these goals: the software includes methods for simple, rapid manual annotation of recordings, denoising and segmentation methods, and a training procedure by which the user can prepare their own filters to automatically recognize individual species. The software can run in batch mode, automatically processing folders of field recordings, and then present the outputs to enable the quick and easy review of the results. Finally, the outputs are presented in a variety of spreadsheets to enable different statistical analyses to be performed. We describe the various workflows of manually and semi‐automatically processing sound files, annotating them to train automatic filters, using those filters in batch mode, and how the software facilitates rapid evaluation of the automated analysis. A demonstration of the software, comparing manual and automatic detection of calls of the little spotted kiwi Apteryx owenii is given. It shows that while the automatic detection does produce false positives, human correction of these is far faster than manual review of the whole sound file. AviaNZ is a freely available open‐source standalone program. Our experience shows that it can be used by anybody quickly and easily. However, for experienced users it is easily customizable and extendable. By enabling everybody involved with acoustic bird recording to quickly and easily analyse their own data, while future‐proofing it by keeping the human in the loop, we are enabling acoustic field recordings to meet their potential.

10 February 2019 in Ecology and Evolution
Verified
Journal Article
Open access
PDF file
ICIsabel Castro
Isabel Castro
ARAlberto De Rosa
Alberto De Rosa
NPNirosha Priyadarshani
Nirosha Priyadarshani
LBLeanne Bradbury
Leanne Bradbury
SMStephen Marsland
Stephen Marsland
Autonomous recording units are now routinely used to monitor birdsong, starting to supplement and potentially replace human listening methods. However, to date there has been very little systematic comparison of human and machine detection ability. We present an experiment based on broadcast calls of nocturnal New Zealand birds in an area of natural forest. The soundscape was monitored by both novice and experienced humans performing a call count, and autonomous recording units. We match records of when calls were broadcast with detections by both humans and machines, and construct a hierarchical generalized linear model of the binary variable of correct detection or not, with a set of covariates about the call (distance, sound direction, relative altitude, and line of sight) and about the listener (age, experience, and gender). The results show that machines and humans have similar listening ability. Humans are more homogeneous in their recording of sounds, and this was not affected by their individual experience or characteristics. Humans were affected by trial and location, in particular one of the stations located in a small but deep valley. Despite recorders being affected significantly more than people by distance, altitude, and line of sight, their overall detection probability was higher. The specific location of recorders seems to be the most important factor determining what they record, and we suggest that for best results more than one recorder (or at least, microphone) is needed at each station to ensure all bird sounds of interest are captured.