Tag: dataset

Unraveling nature’s chorus: AI detects bird sounds in Taiwan’s montane forests

Researchers developed an AI tool which identifies 169 species native to Taiwan from the sound of their calls.

Spectacular subtropical montane forest scenery in Yushan National Park. Credit: Ms. Wen-Ling Tsai

Montane forests, known as biodiversity hotspots, are among the ecosystems facing threats from climate change. To comprehend potential impacts of climate change on birds in these forests, researchers set up automatic recorders in Yushan National Park, Taiwan, and developed an AI tool for species identification using bird sounds. Their goal is to analyze status and trends in animal activity through acoustic data.

Prof. Hsueh-Wen Chang and Ph.D. Candidate Shih-Hung Wu from National Sun Yat-Sen University, Taiwan, Dr. Ruey-Shing Lin, Assistant Researcher Jerome Chie-Jen Ko from the Endemic Species Research Institute, and Ms. Wen-Ling Tsai from Yushan National Park Headquarters have published a paper in the open access journal Biodiversity Data Journal, detailing their use of AI to detect 6 million bird songs.

Compared to traditional observation-based methods, passive acoustic monitoring using automatic recorders to capture wildlife sounds provides cost-effective, long-term, and systematic alternative for long-term biodiversity monitoring.

The authors deployed six recorders in Yushan National Park, Taiwan, a subtropical montane forest habitat with elevations ranging from 1,200 to 2,800 meters. From 2020 to 2021, they recorded nearly 30,000 hours of audio files with abundant biological information.

An automatic recorder was installed on a tree to capture the surrounding soundscape. Credit: Ph.D. Candidate Shih-Hung Wu

However, analyzing this vast dataset is challenging and requires more than human effort alone.

To tackle this challenge, the authors utilized deep learning technology to develop an AI tool called SILIC that can identify species by sound.

SILIC can quickly pinpoint the precise timing of each animal call within the audio files. After several optimizations, the tool is now capable of recognizing 169 species of wildlife native to Taiwan, including 137 bird species, as well as frogs, mammals, and reptiles.

In this study, authors used SILIC to extract 6,243,820 vocalizations from seven montane forest bird species with a high precision of 95%, creating the first open-access AI-analyzed species occurrence dataset available on the Global Biodiversity Information Facility. This is the first open-access dataset with species occurrence data extracted from sounds in soundscape recordings by artificial intelligence.

The Gray-chinned Minivet (left) displays a secondary non-breeding season peak (right) which is possibly related to flocking behavior. Credit: Shih-Hung Wu, Ph.D. Candidate

The dataset unveils detailed acoustic activity patterns of wildlife across both short and long temporal scales. For instance, in diel patterns, the authors identify a morning vocalization peak for all species. On an annual basis, most species exhibit a single breeding season peak; however, some, like the Gray-chinned Minivet, display a secondary non-breeding season peak, possibly related to flocking behavior.

As the monitoring projects continue, the acoustic data may help to understand changes and trends in animal behavior and population across years in a cost-effective and automated manner.

The sound of Gray-chinned Minivet. Credit: Ph.D. Candidate Shih-Hung Wu

The authors anticipate that this extensive wildlife vocalization dataset will not be valuable only for the National Park’s headquarters in decision-making.

“We expect our dataset will be able to help fill the data gaps of fine-scale avian temporal activity patterns in montane forests and contribute to studies concerning the impacts of climate change on montane forest ecosystems,”

they say.

Original source:

Wu S-H, Ko JC-J, Lin R-S, Tsai W-L, Chang H-W (2023) An acoustic detection dataset of birds (Aves) in montane forests using a deep learning approach. Biodiversity Data Journal 11: e97811. https://doi.org/10.3897/BDJ.11.e97811

You can also follow Biodiversity Data Journal on Twitter and Facebook.

Tag team: a tale of two Antarctic blue whales

For the first time, the satellite tracks of two Antarctic blue whales, tagged a decade ago, have been published in the open-access Biodiversity Data Journal.

Ten years ago, Dr Virginia Andrews-Goff was riding the bowsprit of a six-metre boat, as a 30-metre, 120-tonne Antarctic blue whale surfaced alongside.

That day in the Southern Ocean, she became the first and, so far, the only person, to deploy satellite tags on two of these critically endangered and rarely sighted giants.

Scientists approach a 30 metre blue whale in their six metre boat. ©Kylie Owens/Australian Antarctic Division

At the time, her success added weight to a case in the United Nations International Court of Justice, demonstrating that scientific research on whales could be conducted without killing them.

Dr Andrews-Goff and her colleagues at the Australian Antarctic Division have now published the two satellite tracks generated by that 2013 work, in the open-access Biodiversity Data Journal.

This is a unique data set that was incredibly challenging to get.

Dr Virginia Andrews-Goff

The tracks give an insight into the animals’ movement and behaviour on their feeding grounds, and illustrate the significant logistical challenges needed to successfully locate, tag, and track Antarctic blue whales.

“This is a unique data set that was incredibly challenging to get, and, unfortunately, for 10 years no-one has been able to generate more data,” Dr Andrews-Goff said.

“We know very little about the movement and distribution of Antarctic blue whales, where they migrate, where they forage and breed, and we don’t understand the threats they might face as they recover from whaling.”

Two satellite tagged Antarctic blue whales have provided the first insights into the movement and behaviour of these critically endangered ocean giants on their feeding grounds. ©Australian Antarctic Division

Part of the issue is that the animals are incredibly difficult to find. Commercial whaling in the 1960s and ‘70s killed about 290,000 Antarctic blue whales, accounting for 90% of the population. By the late 1990s, the world’s population of Antarctic blue whales was estimated at 2280 animals.

Back in 2013, the research team used novel acoustic tracking techniques to detect blue whale calls and hone in on their location from up to 1000 kilometres away. Once the whales were in sight (in two separate locations), an expert crew manoeuvred close to their fast-moving targets.

The satellite tags showed that the whales travelled 1390 kilometres in 13 days and 5550 kilometres in 74 days, with an average distance of more than 100 kilometres per day.

“The two whales did entirely different things, but what became obvious is that these animals can travel really quickly,” Dr Andrews-Goff said.

“If you consider how far and fast these animals moved, protecting the broader population against potential threats will be tricky because they could potentially circumnavigate Antarctica within a single feeding season.”

his map shows the movement of two satellite tagged Antarctic blue whales. The track on the bottom right are the movements of one whale over 13 days. The other three tracks capture segments of movement by the second whale over 74 days. The tag for this second whale did not transmit data consistently, resulting in data gaps throughout the tracking period.
The blue portions of track show where the whales were moving quickly and directly, suggesting they were in transit, while the orange locations show where they slowed down and appeared to be searching or foraging.    ©Australian Antarctic Division

Since the tracks were obtained, new analytical methods have added some behavioural context to the data.

Two movement rates were observed – a faster ‘in transit’ speed averaging 4.2 km/hr and a slower speed of 2.5 km/hr, thought to correspond with searching or foraging.

“It looks like the whales might hang around in one area to feed and then move quickly to another area and hang around there for another feed,” Dr Andrews-Goff said.

“There may be certain areas that are better feeding grounds than others. From a management perspective, it would be good to understand what is it that makes these areas important?”

Even at a sample size of two, Dr Andrews-Goff said the satellite tracks will assist the International Whaling Commission’s management of Antarctic blue whales, by providing initial insights into blue whale foraging ecology, habitat preferences, distribution, movement rates, and feeding. These will inform an in-depth assessment of Antarctic blue whales due to begin in 2024.

Original source:

Andrews-Goff V, Bell EM, Miller BS, Wotherspoon SJ, Double MC (2022). Satellite tag derived data from two Antarctic blue whales (Balaenoptera musculus intermedia) tagged in the east Antarctic sector of the Southern Ocean. Biodviersity Data Journal 10: e94228 https://doi.org/10.3897/BDJ.10.e94228

***

Conditions of Use – strictly non-commercial, once only, no archive + no sales
The Australian Antarctic Division welcomes your interest in Australia’s Antarctic Program. The Commonwealth of Australia, represented by the Australian Antarctic Division of the Department of Climate Change, Energy, the Environment and Water (Commonwealth), hereby consents to you publishing each item of material listed below ONCE ONLY on a non-commercial, royalty-free and non-exclusive basis. Where material is published in a printed or electronic format (including on the internet), you are requested to acknowledge the photographer or videographer as listed below. If used on social media please tag the Australian Antarctic Division @AusAntarctic #AusAntarctic

This Consent does not entitle you to use the material specified below in any future article, feature or broadcast without further specific prior written permission of the Commonwealth or to adapt, modify, exploit or sublicense the material specified below in any way. By using this material, you agree to the terms of this Consent.

Call for data papers describing datasets from Russia to be published in Biodiversity Data Journal

GBIF partners with FinBIF and Pensoft to support publication of new datasets about biodiversity from across Russia

Original post via GBIF

In collaboration with the Finnish Biodiversity Information Facility (FinBIF) and Pensoft Publishers, GBIF has announced a new call for authors to submit and publish data papers on Russia in a special collection of Biodiversity Data Journal (BDJ). The call extends and expands upon a successful effort in 2020 to mobilize data from European Russia.

Between now and 15 September 2021, the article processing fee (normally €550) will be waived for the first 36 papers, provided that the publications are accepted and meet the following criteria that the data paper describes a dataset:

The manuscript must be prepared in English and is submitted in accordance with BDJ’s instructions to authors by 15 September 2021. Late submissions will not be eligible for APC waivers.

Sponsorship is limited to the first 36 accepted submissions meeting these criteria on a first-come, first-served basis. The call for submissions can therefore close prior to the stated deadline of 15 September 2021. Authors may contribute to more than one manuscript, but artificial division of the logically uniform data and data stories, or “salami publishing”, is not allowed.

BDJ will publish a special issue including the selected papers by the end of 2021. The journal is indexed by Web of Science (Impact Factor 1.331), Scopus (CiteScore: 2.1) and listed in РИНЦ / eLibrary.ru.

For non-native speakers, please ensure that your English is checked either by native speakers or by professional English-language editors prior to submission. You may credit these individuals as a “Contributor” through the AWT interface. Contributors are not listed as co-authors but can help you improve your manuscripts.

In addition to the BDJ instruction to authors, it is required that datasets referenced from the data paper a) cite the dataset’s DOI, b) appear in the paper’s list of references, and c) has “Russia 2021” in Project Data: Title and “N-Eurasia-Russia2021“ in Project Data: Identifier in the dataset’s metadata.

Authors should explore the GBIF.org section on data papers and Strategies and guidelines for scholarly publishing of biodiversity data. Manuscripts and datasets will go through a standard peer-review process. When submitting a manuscript to BDJ, authors are requested to select the Biota of Russia collection.

To see an example, view this dataset on GBIF.org and the corresponding data paper published by BDJ.

Questions may be directed either to Dmitry Schigel, GBIF scientific officer, or Yasen Mutafchiev, managing editor of Biodiversity Data Journal.

The 2021 extension of the collection of data papers will be edited by Vladimir Blagoderov, Pedro Cardoso, Ivan Chadin, Nina Filippova, Alexander Sennikov, Alexey Seregin, and Dmitry Schigel.

This project is a continuation of the successful call for data papers from European Russia in 2020. The funded papers are available in the Biota of Russia special collection and the datasets are shown on the project page.

***

Definition of terms

Datasets with more than 5,000 records that are new to GBIF.org

Datasets should contain at a minimum 5,000 new records that are new to GBIF.org. While the focus is on additional records for the region, records already published in GBIF may meet the criteria of ‘new’ if they are substantially improved, particularly through the addition of georeferenced locations.” Artificial reduction of records from otherwise uniform datasets to the necessary minimum (“salami publishing”) is discouraged and may result in rejection of the manuscript. New submissions describing updates of datasets, already presented in earlier published data papers will not be sponsored.

Justification for publishing datasets with fewer records (e.g. sampling-event datasets, sequence-based data, checklists with endemics etc.) will be considered on a case-by-case basis.

Datasets with high-quality data and metadata

Authors should start by publishing a dataset comprised of data and metadata that meets GBIF’s stated data quality requirement. This effort will involve work on an installation of the GBIF Integrated Publishing Toolkit.

Only when the dataset is prepared should authors then turn to working on the manuscript text. The extended metadata you enter in the IPT while describing your dataset can be converted into manuscript with a single-click of a button in the ARPHA Writing Tool (see also Creation and Publication of Data Papers from Ecological Metadata Language (EML) Metadata. Authors can then complete, edit and submit manuscripts to BDJ for review.

Datasets with geographic coverage in Russia

In correspondence with the funding priorities of this programme, at least 80% of the records in a dataset should have coordinates that fall within the priority area of Russia. However, authors of the paper may be affiliated with institutions anywhere in the world.

***

Check out the Biota of Russia dynamic data paper collection so far.

Follow Biodiversity Data Journal on Twitter and Facebook to keep yourself posted about the new research published.