PRD-028 · Earth-based observatories — `ground-observatory` fleet category

Status · Draft (prep for v0.8) Date · 2026-06-09 Owner · Marko Closes into · GH epic v0.8 content track (#313 sibling); per-slice PRs Builds on · PRD-012 (Spaceflight Fleet) · existing observatory fleet category (space-based: Hubble, JWST, Chandra, etc.)

Why this is a PRD. Marko's idea (2026-06-09): orrery currently surfaces only space-based observatories under /fleet (Hubble, JWST, Chandra, Compton, Euclid, Gaia, Hitomi, Kepler, Spektr-RG, Spitzer, TESS, XMM-Newton — 12 entries). But Earth-based radio telescopes are critical companions to those space platforms: FAST, Arecibo (collapsed but historically dominant), Green Bank, VLA, ALMA, Parkes, Effelsberg, Jodrell Bank, MeerKAT, plus the Deep Space Network tracking complexes that talk to every interplanetary probe orrery already renders. This PRD adds a new ground-observatory fleet category alongside the existing observatory (which stays space-only), with a curated set of 12–14 entries spanning radio + deep-space-tracking infrastructure on every populated continent. Carves a separate category (matching the surface-base precedent from PRD-025) rather than mixing space + ground under one umbrella, so filter UX stays clean.

What

A new ground-observatory category in the fleet schema, with 12–14 curated entries. Existing observatory category remains space-only and unchanged. New category lands as a sibling chip on /fleet's filter row, with a brief category description explaining the space-vs-ground split.

Each entry surfaces: telescope/array name, location (mountain or site + country), agency/consortium, operational status (ACTIVE / RETIRED / FAILED / DEVELOPMENT), key science (what it observes + what makes it notable), first light date, links to operator + Wikipedia.

Curated set (~18, globally balanced)

Coverage target: every populated continent + ≥ 1 entry per major radio-astronomy operator. Original draft skewed USA-heavy (5 of 13); revised cut rebalances toward representative non-US facilities so the catalogue honestly reflects where Earth-based radio astronomy actually happens.

Single-dish radio telescopes (8):

id	Name	Location	Status	Notes
`arecibo`	Arecibo Observatory	Puerto Rico, USA	RETIRED (collapsed Dec 2020)	305 m fixed dish; first millisecond pulsar, Hulse-Taylor binary, 1974 Arecibo message; ancillary 12 m antenna + lidar continue post-collapse
`fast`	FAST (Five-hundred-meter Aperture Spherical Telescope)	Guizhou, China	ACTIVE	World's largest filled-aperture radio telescope (500 m); first light 2016; hundreds of new pulsars + FRB localisations
`gbt`	Green Bank Telescope	West Virginia, USA	ACTIVE	World's largest fully-steerable radio telescope (100 m); inside the National Radio Quiet Zone
`effelsberg-100m`	Effelsberg 100m Radio Telescope	Bad Münstereifel, Germany	ACTIVE	Largest fully-steerable in Europe; MPIfR / Max Planck
`parkes-64m`	Parkes 64m Radio Telescope ("The Dish")	New South Wales, Australia	ACTIVE	Apollo 11 lunar landing comms; ongoing pulsar + FRB science; CSIRO
`lovell-telescope`	Lovell Telescope at Jodrell Bank	Cheshire, UK	ACTIVE	76 m; UNESCO World Heritage Site; operational since 1957
`ratan-600`	RATAN-600	Zelenchukskaya, Russia	ACTIVE	World's largest single-piece reflector — 576 m circular ring antenna; SAO RAS; operational since 1974
`nobeyama-45m`	Nobeyama 45m Radio Telescope	Nagano, Japan	ACTIVE	NAOJ; flagship mm-wave single dish in Asia; star-forming-region + molecular-cloud science

Arrays + interferometers (6):

id	Name	Location	Status	Notes
`vla`	Very Large Array (Karl G. Jansky VLA)	New Mexico, USA	ACTIVE	27 antennas, 25 m each, Y-config; NRAO; Contact filming location
`alma`	ALMA (Atacama Large Millimeter / submillimeter Array)	Chajnantor, Chile	ACTIVE	66 antennas at 5,058 m altitude; ESO + NRAO + NAOJ multinational; EHT contributor (2019 M87 image)
`meerkat`	MeerKAT	Karoo, South Africa	ACTIVE	64 dishes; SARAO; SKA pathfinder; Galactic Centre imaging breakthroughs
`gmrt`	GMRT (Giant Metrewave Radio Telescope)	Pune, India	ACTIVE	NCRA-TIFR; 30 dishes at 45 m each; world-leading at 150–1500 MHz low-frequency
`lofar`	LOFAR (LOw Frequency ARray)	Netherlands + 7 European partner countries	ACTIVE	ASTRON-led; phased-array (no moving dishes); pan-European baselines
`lmt`	Large Millimeter Telescope (LMT / GTM)	Sierra Negra volcano, Mexico	ACTIVE	INAOE + UMass; 50 m mm-wave; EHT contributor

Future / under construction (1):

id	Name	Location	Status	Notes
`ska`	Square Kilometre Array (SKA-Low + SKA-Mid)	Australia + South Africa	DEVELOPMENT	World's largest radio observatory when complete; SKAO multinational; first light ~2028–2030

Deep Space Network (3 sites — tracking infrastructure for every interplanetary probe orrery renders):

id	Name	Location	Status	Notes
`dsn-goldstone`	Goldstone Deep Space Communications Complex	Mojave Desert, California, USA	ACTIVE	JPL; primary US site; doubles as Goldstone Solar System Radar for asteroid imaging
`dsn-madrid`	Madrid Deep Space Communications Complex (Robledo de Chavela)	Spain	ACTIVE	JPL + INTA; Europe / Africa / W-Asia coverage
`dsn-canberra`	Canberra Deep Space Communications Complex	Tidbinbilla, Australia	ACTIVE	JPL + CSIRO; Asia / Australasia coverage; only DSN site with direct Voyager 2 line-of-sight (southern declination)

Geographic + operator coverage check (post-rebalance):

Continents: NA (Arecibo, GBT, VLA, LMT, Goldstone), SA (ALMA), EU (Effelsberg, Lovell, LOFAR, Madrid), Asia (FAST, GMRT, Nobeyama), Africa (MeerKAT), Australia (Parkes, SKA-Low, Canberra), Russia / Eurasia (RATAN-600). All 6 inhabited continents represented.
Operators: NRAO (US), MPIfR (Germany), CSIRO (Australia), NAOC (China), SARAO (South Africa), NCRA-TIFR (India), ASTRON (Netherlands), INAOE + UMass (Mexico), NAOJ (Japan), SAO RAS (Russia), Jodrell Bank Centre (UK), SKAO (multinational), JPL DSN (NASA + INTA + CSIRO). 13 distinct operators.
USA share: 5 of 18 = 28% (down from 38% in original cut).

Optional candidates if the cut needs trimming or expanding: VLBA (USA, 10 sites — overlap with VLA), CHIME (Canada, FRB-detection star), MWA / Murchison Widefield Array (Australia, SKA-Low precursor — overlap with SKA entry), Sardinia Radio Telescope (Italy 64 m), Allen Telescope Array (SETI, USA), KVN (Korea VLBI Network, 4 antennas), Nançay (France, historic). Marko's editorial call. Aim for 18 in v0.8 ship list; defer the rest to a follow-up wave.

Schema sketch — new `ground-observatory` category

Mirrors observatory shape; new category enum value + a few ground-specific optional fields.

json

{
  "id": "fast",
  "name": "FAST (Five-hundred-meter Aperture Spherical Telescope)",
  "category": "ground-observatory",
  "agency": "NAOC / CAS",
  "country": "China",
  "manufacturer": "NAOC",
  "first_flight": "2016-09-25",          // first light date — same field, repurposed
  "status": "ACTIVE",
  "era": "2011-now",
  "epoch": "commercial-era",
  "best_known_for": "World's largest filled-aperture radio telescope (500 m dish); pulsar discoveries + FRB localisations",
  "instrument_type": "single-dish radio telescope",   // new optional field
  "wavelength_band": "radio (70 MHz – 3 GHz)",         // new optional field
  "site_location": {                                    // new optional field
    "lat": 25.65,
    "lon": 106.86,
    "elevation_m": 1110,
    "site_name": "Dawodang depression, Pingtang County, Guizhou"
  },
  "credit": "...",
  "links": [ ... ],
  "linked_missions": []     // ground observatories rarely link to discrete missions; may carry e.g. SETI@home, Breakthrough Listen, EHT campaign refs
}

Schema-touch points:

static/data/schemas/fleet-entry.schema.json — add ground-observatory to category enum + register the three new optional fields (instrument_type, wavelength_band, site_location)
static/data/fleet/index.json — append ground-observatory entries
static/data/fleet/ground-observatory/<id>.json — new directory
src/lib/fleet/ filter chips + category labels — surface ground-observatory in chip row
src/lib/paraglide/messages.json — add fleet_category_ground_observatory label string (en-US; other locales fall back)
/fleet route — add ground-observatory to the category query param + filter render
ADR-002 (if it enumerates categories) — note the addition

Slice order

Slice 0 (category foundation, shared with PRD-025 surface-base) —
  · if PRD-025 Slice 0 is in flight: merge ground-observatory into the same schema + filter UI work so both new categories land together
  · if PRD-025 Slice 0 already shipped: standalone schema + filter UI slice for ground-observatory

Slice 1 (radio telescopes + arrays PR) —
  · add the 6 single-dish + 4 array entries (10 entries)
  · full en-US strings, 3+ links per entry, credit string sourced
  · `validate-data` green; `preflight` green

Slice 2 (DSN trio + SKA PR) —
  · add the 3 Deep Space Network sites + Square Kilometre Array (4 entries)
  · DSN entries cross-link from every interplanetary probe / orbiter entry in fleet that uses DSN comms (Cassini, Voyager 1/2, New Horizons, Juno, Europa Clipper, Tianwen-1, JUICE, BepiColombo, etc.) — adds "tracked via DSN" surface
  · SKA entry honestly flagged as DEVELOPMENT

Done definition

Slice 0

ground-observatory registered in fleet-entry schema enum
New optional fields (instrument_type, wavelength_band, site_location) registered
Filter chip live on /fleet route
i18n key fleet_category_ground_observatory added (en-US)
npm run validate-data + preflight green

Slice 1 + 2

12–14 entries land under static/data/fleet/ground-observatory/
Every entry has full en-US copy: name, agency, country, status, best_known_for, credit (1–2 sentences sourced), ≥ 3 links, instrument_type, wavelength_band, site_location
DSN entries surface "tracked via DSN" cross-link on the relevant interplanetary probe fleet entries
validate-data + preflight green

v0.8 ship bar

All 12–14 entries merged across Slice 0 + 1 + 2
≥ 1 entry per continent (NA, SA, EU, Asia, Africa, Australia) — concentrated coverage for radio astronomy globally
≥ 1 entry per major operator: NRAO, MPIfR, CSIRO, NAOC, SARAO, JPL DSN, SKA Organisation
PRD-028 status → Accepted (v0.8)

Out of scope (v0.8)

Optical telescopes (Mauna Kea Keck / Subaru / Gemini-N, ESO VLT Paranal, GTC La Palma, Magellan, Vera Rubin / LSST). User's brief was "radio telescopes where we observe space"; optical is a substantial separate scope deferred to v0.9. Note: Vera Rubin first-light campaign is ongoing in 2026 — strong candidate for v0.9 priority.
Solar observatories on the ground (DKIST, GREGOR, SDO is space and lives in existing category)
Cosmic-ray + neutrino observatories (IceCube, KM3NeT, Auger) — different physics scope; would need its own category if added
Atmospheric Cherenkov telescopes (HESS, MAGIC, VERITAS gamma-ray ground array)
Historical defunct sites beyond Arecibo (e.g., Big Ear / WOW! signal, Lebedev) — only Arecibo represents "historically dominant + recently lost" in v0.8
Cover / hero imagery for each entry — image-pipeline catch-up, not blocking content adds
Per-site visitor information (Jodrell Bank + Parkes both have visitor centres but that's PRD-027 visit-list cross-link territory, not this PRD)
3D rendering of ground-observatory sites on /earth — separate visualisation scope, future PRD

Risks

DSN classification. The Deep Space Network sites are tracking infrastructure, not science observatories in the traditional radio-astronomy sense. The entry's best_known_for + credit must be honest about that — "primary communications gateway for every interplanetary probe" rather than implying these are doing original astronomy. They are sometimes used for radar imaging (Goldstone Solar System Radar) so the science overlap is non-zero.
Arecibo's status. The main 305 m dish collapsed December 2020. The site itself still operates a 12 m antenna + lidar for atmospheric science as of 2026. Entry needs both facts; can't be marked simply RETIRED. Recommend status: RETIRED with a credit-line caveat that ancillary instruments continue.
SKA classification. SKA-Low (Australia) and SKA-Mid (South Africa) are technically two arrays but one observatory. Single entry covers both phases honestly; flag the construction timeline (first phase first-light ~2028–2030 per current schedule, slip-prone) in credit.
location_type ambiguity for non-ground non-space stations. Stratospheric balloon observatories (SOFIA retired 2022) and high-altitude airborne platforms don't fit either category cleanly. Not in v0.8 scope; if they come back via PRD-029+ they get their own category.
Cross-link bloat. Every interplanetary probe in fleet uses DSN. Auto-cross-linking every probe entry to all three DSN sites could clutter the panel. Recommend: one "tracked via Deep Space Network" link that resolves to a category-level page or one canonical DSN entry, not three per-site links per probe.

References

Project memory: "Orrery celebrates global space programs, not just NASA"
Existing space-based observatories: static/data/fleet/observatory/* (12 entries, all space)
PRD-012 (Spaceflight Fleet)
ADR-002 (data shape) — categories enum (if maintained centrally)
v0.8 content epic: chipi/orrery#313 (PRD-025) — this PRD is a sibling slice in the same v0.8 content track
Sister PRD: PRD-027 (museums "what to visit") — shares the v0.8 prep wave
Sister category precedent: surface-base from PRD-025

PRD-028 · Earth-based observatories — ground-observatory fleet category ​

What ​

Curated set (~18, globally balanced) ​

Schema sketch — new ground-observatory category ​

Slice order ​

Done definition ​

Slice 0 ​

Slice 1 + 2 ​

v0.8 ship bar ​

Out of scope (v0.8) ​

Risks ​

References ​