About AusMT

1 · What AusMT is

Australian MT results are scattered across university groups, government agencies, industry collaborations, supplementary files and ageing legacy archives. AusMT exists to fix that: one national catalogue where every transfer function can be found, cited and downloaded with its provenance attached.

AusMT is a national MT research infrastructure: a discovery and provenance layer over work that other people own, indexing processed transfer functions and their provenance. It doesn't host raw time series; where an archive has verified an openly accessible file, it routes you straight to the archive that does, such as the NCI-AuScope MT collection.

A magnetotelluric transfer function is a processed estimate of the Earth's electromagnetic response at a site, derived from measured electric and magnetic field variations. It's the result most MT work reuses, and it's what AusMT indexes. For each one the catalogue records which file it came from, who produced it, which licence applies and how to cite it.

2 · Who enables AusMT

AusMT is enabled by AuScope and the Australian Government via the National Collaborative Research Infrastructure Strategy (NCRIS).

AuScope is Australia's national provider of research infrastructure for the geoscience community. AusMT provides national digital research infrastructure for discovering, accessing and reusing Australian magnetotelluric data and its provenance.

Learn more about AuScope.

Operation and governance

AusMT is maintained by AuScope with contributions from Australia's magnetotelluric community. Data custodians retain authority over their survey data, licences and attribution. AusMT's governance, correction and preservation arrangements are documented in Governance & Operation.

3 · What you can do here

Contribute a survey →

4 · Licence and access

Custodians keep ownership of their data. AusMT claims no ownership over anything it indexes, and what it may redistribute depends on the licence and the access level each custodian sets. Three things are licensed separately and it's worth keeping them apart.

• Survey data is licensed by its custodian, survey by survey. The licence is on every survey record and travels with the bytes as a LICENSE.txt inside each download.
• Catalogue metadata is CC0-1.0, so you can harvest and republish the catalogue itself freely.
• Portal and engine software is licensed under Apache-2.0.

Every survey carries one of three access levels:

• open: the record and the data files are both served.
• metadata_only: the record is served; the files are not distributed.
• embargoed: the record is served; the files are withheld until the custodian lifts the embargo.

Anything AusMT does not recognise is treated as closed. An embargo never hides discovery: an embargoed survey keeps its full catalogue record, its stations and its footprint on the map. Station positions follow the custodian's instruction: a withheld position is served without coordinates, and a generalised one is rounded to a 0.1 degree cell.

Privacy. AusMT's server logs truncate IP addresses at the edge, IPv4 to /24 and IPv6 to /48, so a full address is never stored; they are kept briefly, about seven days, and used only for aggregate reporting such as downloads by dataset and visit counts. There are no cookies, no cross-site tracking and no personal data.

Who runs AusMT, how to request a correction, and what happens to the catalogue if the project ends are set out in Governance & Operation.

5 · Citing and credit

Cite the survey; acknowledge AusMT. Each survey record carries the custodian's own citation, and the station drawer's Cite tab renders it in APA with the DOI as a live link where one exists. Where a custodian recorded no citation, the drawer says so plainly and points you at the survey package instead of inventing an author list.

Credit is recorded as two separate lists. creators[] is the citation author list and its order is the author order, so reproduce it as given. contributors[] records who did what, each row tagged with a DataCite contributor role, and its order carries no citation meaning. People carry an ORCID and organisations a ROR where one was declared.

A bulk download brings the paperwork with it. The citation pack contains CITATIONS.txt, citations.bib and citations.ris, and each survey's EDIs sit beside a LICENSE.txt carrying that survey's licence and its required attribution. The two credit lists are documented field by field in the MTCAT schema reference, what gets recorded about how a product was made is in the Provenance Model, and Versioning covers what a survey version means and why you should cite the one you used.

Cite the survey wherever its data appears in your work (the data availability or methods section is the usual home) and make sure the survey's citation lands in your reference list, not only in the text. Please also acknowledge AusMT:

Magnetotelluric transfer functions were accessed through AusMT, Australia's Magnetotelluric Data Portal (https://ausmt.auscope.org.au), enabled by AuScope and the Australian Government via the National Collaborative Research Infrastructure Strategy (NCRIS).

If your talk or paper shows AusMT as well as citing it, the Brand and assets page carries the logos, the standalone mark and the presentation artwork, with the terms for using them.

6 · Contributing a survey

Bring your EDI, EMTF XML or MTH5 transfer functions to the Contribute a survey page. It extracts the stations, plots them, and walks you through the metadata AusMT needs.

After you submit, the gateway runs the AusMT validator over the package and a curator reviews it before anything is published. The full validate, preview, package, review and publish path is written up in the Submission Workflow.

7 · Fetching data via API

Everything the portal reads is a read-only static JSON file on this site, and you can read the same files. There's no API key and no sign-up. Every response carries Access-Control-Allow-Origin: *, so a browser application can fetch it cross-origin.

Start from mtcat.json, the MTCAT document that describes every survey in one file. Each survey record carries its slug under survey_id, and the slug is what the download bundles are named after. This lists the catalogue, then pulls one survey's EDIs down as a single zip.

import json, urllib.request
BASE = "https://ausmt.auscope.org.au"
cat = json.load(urllib.request.urlopen(f"{BASE}/data/mtcat.json"))
for s in cat["surveys"]:
    print(s["survey_id"], s["access"], s["license"], s["title"])

slug = "vulcan-2022"                        # any survey whose access is "open"
body = urllib.request.urlopen(f"{BASE}/data/bundles/{slug}-edi.zip").read()
open(f"{slug}-edi.zip", "wb").write(body)

Every published survey is bundled three ways, as /data/bundles/<slug>-edi.zip, /data/bundles/<slug>-xml.zip for EMTF XML and /data/bundles/<slug>-tf.h5 for MTH5. A LICENSE.txt rides inside each zip. If you need one station rather than a whole survey, want a bounding-box query, or want to verify what you downloaded against its published SHA-256, the worked patterns are in the API reference on the documentation site.

An embargoed survey has no rows in the download manifest at all. Its bytes are withheld by construction, so there's nothing to request and no access error to handle, while its catalogue record stays public.

Machine contract. The catalogue document mtcat.json is written against the MTCAT schema served next to it at data/mtcat.schema.json, so a second implementation can validate the document without resolving anything off-site. The document names its own contract in its portal block, where schema is mtcat, version is 2.0 and metadata_license is CC0-1.0. That licence covers the catalogue metadata only. A survey's DATA licence is the separate license field on its own survey record, and it varies by survey. The field guide to reading a served record is on the MTCAT schema reference.

8 · Documentation

For further information, see the AusMT documentation.

Quarterly citable snapshots of the corpus are listed on the releases page; each one is a frozen tree with its own identifier.