Bat Acoustic Data Submission Instructions

TERN is publishing bat acoustic files and call information for further reuse. These datasets will be made discoverable from TERN infrastructure for the wider use of the bat acoustic community. The flowchart below explains the bat acoustic data publication process.

All bat acoustic audio files and pulse calls will be uploaded to TERN storage through https://sftpgo.tern.org.au/ service. Please follow the instructions below:

All individual data submission happens using the TERN data submission tool so that all required metadata are captured. If any data providers want to publish many files, please get in touch with the TERN help desk via esupport@tern.org.au for alternative submission. If submitting an individual collected dataset, go to the TERN SHaRED Data Submission Tool to start the submission processes. The SHaRED online tool will enable users to author metadata about the data they want to submit and upload all acoustic data files before agreeing to the terms of use. For more information on how to submit datasets via SHaRED, please see the TERN SHaRED User Guide. Before submitting all sound files, please create all audio files as instructed below:

What files to submit

Sound Files

Both .zc and .wav files are acceptable.
All file names should include the recording device's name as a suffix to the sound file name.
Include files that have at least four bat-call pulses and are of reasonable quality.
Files may have a species name attributed to them, but ‘unnamed’ files are also valuable and acceptable.
If a Species name is provided:
1. Species names should be included in a metadata field titled “Species”.
2. The file MUST also include a metadata field titled “Identified by” that lists the name(s) of the individuals(s) who identified the species. This is a key component in ensuring the accuracy of species identifications.
3. Files with a Species name can include other unnamed species
4. Files can have multiple species names attributed to them

Pulse Metric Files

data providers are requested to include a summary of the pulse metrics within each file, allowing database users to search files on a ‘per pulse’ basis. This will allow users to search the database per pulse basis without regenerating this information with every search.
please use the following approach to generate pulse data if using Anabat Insight.

Name the generated .csv with the same filename as the folder containing the sound files it describes, add ‘_PULSE.csv’ as a filename suffix and save the .csv file into the same folder as the sound files (it will be read automatically when sound files are submitted to the database). For example, the following image shows three pulse metric .csv data files describing the contents of the folders Swift1, Swift2 and Walkabout respectively

Using metadata fields to provide file-related information

important information should be included in the metadata fields associated with each file. This will ensure that the information does not become separated from sound file
- Metadata information can be accessed/edited using programs such as AnaLook, Insight, and Kaleidoscope.
- All metadata fields and their contents will be read automatically when sound files are loaded into the database
The minimum information required for each sound file includes:
- A metadata field called Location that provides latitude and longitude in decimal minutes (see below)
- A metadata field called “Record context” (see below)
- A metadata field called “Collected by” that includes the name(s) of the individuals that collected the calls (see below)
Except for Location and Timestamp metadata fields, DO NOT INCLUDE a comma (’,’) in a data field. This confuses the GUANO metadata sound format protocols.

Example metadata entries and formatting requirements.

Metadata name	Sample data format	Explanation
Minimum information required
Location	-19.678967, 146.976445	Please use decimal minutes (4 to 6 decimal points only)
Collected By	Robson SKA Robson SKA; Edwards W	Surname initials Separate fields with a `;`
Site	Boodjamulla (Lawn Hill) NP Carrington Cave Ross River	List the name of any national Parks first, followed by more specific locations. Some NP’s have the indigenous name first, others the reverse. National Park and State Forest can be abbreviated to NP and SF respectively
Required if Genus or Species provided
Species	Rhinonicteris aurantia Rhinonicteris aurantia; Macroderma gigas	Genus and Family can also be used if this was the only taxonomic level resolved. Multiple species names can be attributed to a single file.
Identified by	Robson SKA Robson SKA; Edwards W	please provide the name of the person(s) who identified the species if, the species name is provided. Surname initials Separate individuals with a ‘;’ Multiple individuals can be attributed to a single file.
ID method	Captured Visual Call	Captured= identified in the hand Visual = observed flying or at roost Call = default for most observations
ADDITIONAL VALUABLE INFORMATION IF AVAILABLE
Record context	Free flying Release In bag Cave entrance Mine entrance Roost entrance Enclosure Overwater	'Record context is a critical element that should help with species identification. The majority of calls collected outside a known Miniopterus maternity cave for example will be Miniopterus, but not all. Most calls will be ‘Free flying’ Additional call types may be relevant
Sex	Female Male
Life stage	Adult Juvenile Pregnant Lactating Postlactating
Call Features	Buzz Social Song Subharmonics Harmonics Steep

Once the data is submitted, Simon Robson will review the data submission and approve it for publication. Data providers will be notified when sound files are accessible from the TERN portal for the wider community.