ANSIR web

Public web pages and data pipeline for ANSIR Research Facilities for Earth Sounding, a national research facility program of AuScope, funded under the National Collaborative Research Infrastructure Strategy (NCRIS).

ANSIR provides seismic, magnetotelluric and related instrumentation to researchers at Australian and international universities, publicly funded research organisations, and their industry partners.

What is here

Four public pages, each served by GitHub Pages and embedded in the AuScope website through an iframe:

Page	Path	What it shows
Research projects	/	Every project on the ANSIR register: map, filters, search, deep links
Instrument availability and schedule	/schedule/	What is available now, what is committed to which experiment, and when it is due back
Portfolio statistics	/stats/	Projects, researchers, institutions, instrument deployments and published outputs
Equipment loan application	/apply/	The application form for instrument time

Live on the AuScope site at /ansir-projects, /ansir-schedule and /ansir-application-form.

How it works

Everything public is a static file in this repository. Data is refreshed by scheduled GitHub Actions, not by a server.

Google Sheets (maintained by ANSIR staff)        Public APIs
  |  project register (published CSV view)         |  DataCite, ORCID, CrossRef
  |  seismic fleet status (published tabs)         |
  v                                                v
      GitHub Actions, nightly and on demand
                      |
                      v
        data/*.json committed to this repo
                      |
                      v
              GitHub Pages serves the pages
                      |
                      v
        embedded in the AuScope Squarespace site

The pages hold no data of their own: each one fetches the JSON files below and renders whatever it finds. When staff update a sheet, the next sync run publishes the change with no code edit and no manual step.

Data files

File	Built by	Source
data/data.json	scripts/fetch-projects.js	The project register: a published, filtered view of the master sheet holding only rows and columns approved for publication
data/schedule.json	scripts/fetch-schedule.js	The ANSIR Seismic Fleet Status sheet: availability, deployments and rapid-response records
data/instrument-register.json	scripts/fetch-instruments.js	The AuScope instrument register: one DataCite DOI per physical instrument, under prefix 10.82388
data/contributor-verification.json	scripts/validate-contributors.js	The public ORCID register: which identifiers are confirmed to belong to which researcher
ro-crate-metadata.json	scripts/generate-ro-crate.js	Generated from data/data.json after DOI enrichment

Each fetch script validates what it receives and exits with a named error rather than publishing something wrong. A sheet that has been restructured, unpublished, or emptied stops the sync and leaves the last good data in place; the failure is reported in the workflow run summary. Every page displays the date its data was gathered, so stale data is visible rather than silent.

Research metadata

ro-crate-metadata.json is RO-Crate 1.1 compliant, using Schema.org JSON-LD to describe the project collection, each project, its contributors (linked by ORCID), organisations, funding, instruments, publications, FDSN networks, and spatial and temporal coverage. This supports discovery by ARDC Research Data Australia, institutional repositories and other RO-Crate-aware harvesters.

Publication and dataset metadata is resolved from DOIs through the CrossRef and DataCite APIs.

Repository structure

.
├── index.html                  Research projects page
├── schedule/index.html         Instrument availability and schedule
├── stats/index.html            Portfolio statistics
├── apply/index.html            Equipment loan application form
├── data/                       Published JSON, rebuilt by the workflows
├── scripts/                    The fetch, enrichment and validation scripts
├── lib/                        Self-hosted Leaflet and Leaflet.draw
├── assets/                     Images
├── gas/                        Google Apps Script source for the intake endpoint
└── .github/workflows/          Scheduled sync and DOI enrichment

The one piece that is not static

gas/Code.gs is the application intake endpoint. Accepting a form submission from a member of the public, storing an uploaded PDF and sending email all require a server, so this single Apps Script deployment remains. It exposes exactly four web-callable functions and writes submissions to a separate review tab, never to the published project data. gas/README.md is the deployment guide.

Working on it locally

No build step, no package manager, no framework. Serve the directory and open a page:

python3 -m http.server 8000

Then http://localhost:8000/, /schedule/, /stats/ or /apply/.

To rebuild a data file locally (Node 20 or newer, no dependencies):

node scripts/fetch-schedule.js       # fleet schedule, from the published sheet
node scripts/fetch-instruments.js    # instrument register, from DataCite
node scripts/validate-contributors.js --warn   # ORCID verification

scripts/fetch-projects.js additionally needs ANSIR_PROJECTS_CSV_URL in the environment, matching the repository variable of the same name.

Conventions

• Vanilla HTML, CSS and JavaScript. No frameworks, no build tooling, no CDNs: every dependency is vendored under lib/.
• Pages are embedded in iframes and report their height to the parent page, so each one is written to work standalone and embedded.
• No credentials or internal identifiers in this repository. The intake endpoint reads its sheet ID from an Apps Script property; the workflows read the published sheet URL from a repository variable.

Documentation

Document	Covers
docs/PIPELINE.md	How project data is published: the sheet, its public view, the nightly sync, and how to run and check it
docs/SCHEDULE_DATA.md	The fleet spreadsheet layout the schedule parser depends on, and how to extend it
docs/SCHEDULE_PAGE.md	How the schedule page derives availability, and how it is embedded
docs/INTAKE.md	What happens to an equipment loan application from submission to review
docs/FORM.md	The application form: its steps, autofill, drafts and transport
docs/PIDINST.md	The AuScope instrument register as a source of fleet totals
gas/README.md	Deploying the intake endpoint

Licence

CC BY 4.0. Data sourced from ANSIR research projects; publication metadata resolved via CrossRef and DataCite.