HADS stars in NightScribe — field guide & references
Bilingual twin: HADS.es.md (Spanish). Implementation plan: PLANS/hads-stars.md. Decision record: adr/ADR-034-hads-stars.md.
This file is the research dossier for the HADS feature: the science, the data files bundled in the repo, the design rationale (transits vs. HADS) and the reference trail — so that any human or AI agent can keep working on it without re-doing the investigation.
1. What a HADS star is
HADS = High-Amplitude Delta Scuti star, a subclass of δ Scuti pulsating variables. They sit at the intersection of the classical instability strip and the main sequence (early F to late A spectral types).
Key physical facts that drive the feature design:
| Property | Value | Source |
|---|---|---|
| Pulsation periods | ~1–3 h (class up to ~6 h) | Kotysz 2020; McNamara 2000 |
| Peak-to-peak amplitude | ≥ 0.3 mag in V for the HADS subclass (class δ Scuti starts at > 0.1) | AAVSO VSO; OGLE |
| Light-curve shape | Asymmetric "sawtooth": fast rise, slow decline | OGLE atlas |
| Modes | Mostly fundamental (F) radial; some double-mode F + first overtone (period ratio 0.76–0.78, Petersen diagram); very few triple-mode | Kotysz 2020; Wils et al. 2008 |
| Anything else | Some are multiperiodic / show non-radial modes | Kotysz 2020; catalog flags |
| History | Once called "dwarf Cepheids" (shapes resemble Cepheids) | OGLE |
Because periods are 1–5 h and amplitudes ≥ 0.3 mag, a full light curve is captured in a single short night — you literally watch a star change brightness in real time. AAVSO recommends a HADS star as the perfect first target for digital photometry ("Your First Observing Target"), with images every ≤ 15 min to follow the curve.
Amplitude caveat: for visual observers AAVSO advises ΔV ≥ 0.5 mag; smaller amplitudes need photometry.
2. The reference figure: Patrick Wils
Patrick Wils is a Belgian amateur astronomer and a world reference in the study of variable stars. He coordinates the long-term photometric monitoring of HADS stars linked to the Belgian Vereniging Voor Sterrenkunde (VVS), and he is part of the technical team of the AAVSO VSX (International Variable Star Index) — co-compiler, together with Sebastián Otero, Patrick Schmeer and Klaus Bernhard, of the index's million-plus objects. His catalogues of double-mode and triple-mode δ Scuti stars (Wils et al. 2008) are cited across the literature. The HADS target list bundled with NightScribe comes from his monitoring programme.
References (verbatim)
HADS — talk, BAV meeting, Hamburg 2016: https://www.bav-astro.eu/images/BAV_Tagungen/Hamburg_2016/04_HADS.pdf
AAVSO — 20 million observations milestone: https://www.aavso.org/20-million-observations
VSX — About variable types: https://vsx.aavso.org/index.php?view=about.vartypes
Wils / variable stars (austriaca.at): https://www.austriaca.at/0xc1aa5572%200x00166b7b.pdf
BAA — Short Period Pulsator Program: https://britastro.org/forums/topic/short-period-pulsator-program
Supplementary sources consulted (for the narrative and scoring)
Wils, P. et al. 2008 — double/triple-mode HADS classification (cited in arXiv:2510.24444: "Wils et al. (2008) list only four known HADS that pulsate in three radial modes simultaneously").
VSX founding paper bibcode:
2006SASS...25...47W.AAVSO — "Delta Scuti and the Delta Scuti variables" (Variable Star of the Season) https://www.aavso.org/vsots_delsct
AAVSO — "Your First Observing Target" (HADS as first photometry target) https://wp-test.aavso.org/your-first-observing-target
OGLE Atlas of variable star light curves — δ Scuti https://ogle.astrouw.edu.pl/atlas/delta_Sct.html
Kotysz, K. 2020, "Variability of HADS stars in TESS", PTA Proceedings, vol. 10, 180–182, pta.edu.pl/proc/v10p180 — URL only (PDF not committed, ~9.7 MB). Key content extracted in §4.
3. Data files in the repo
Source of truth & refresh (2026-09-11, decision H-b/H-j)
The live catalog is Patrick Wils' Google Sheets workbook: <https://docs.google.com/spreadsheets/d/1oGA2HaEHE8L6eX19ZoHqQQTu0LYV56HX3Srg7oCtOHo/> — public, one tab per year (2010…now), updated daily. NightScribe downloads it at runtime as .xlsx (the only export carrying the font colors), caches it 12 h via core/db.py and parses it stdlib-only in core/sources/hads_sheet.py (two-level cache: raw XLSX + parsed JSON, so the ~1-2 s parse happens once a day). The bundled CSV below is the offline fallback and alias source; the merge lives in core/hads.py (catalog()).
Color legend (star name / coordinates font color in the workbook):
| Color | Meaning | Priority |
|---|---|---|
| Red name | period changes found | Priority! |
| Orange name | period changes possible | Priority! |
| Blue coordinates | not yet observed in the programme | opportunity (+6) |
| Purple name | multiple pulsation modes (observe on consecutive nights) | none |
To refresh the bundled snapshot when it drifts: run the functional test and read its informational drift report — .venv/bin/python -m pytest tests/functional -k hads_live -s — then hand-edit the few drifted rows (periods/magnitudes; the colors never live in the CSV, they come from the live sheet at runtime).
nightscribe/assets/HADS-stars.csv (runtime catalog, v1)
168 stars. ASCII, CRLF line endings, some Name fields are quoted (they contain commas inside (=…=…) aliases). Columns:
| Column | Example | Notes |
|---|---|---|
Name | GP And (=GSC 01739-01964) | primary + parseable aliases; may carry trailing flags: , multiperiodic?, -- Non-radial, -- change in amplitude? |
RA | 00 55 18.1 | HH MM SS.s → degrees |
DEC | +23 09 49 | ±DD MM SS → degrees |
Max | 10.4 | magnitude at maximum |
Min | 11 | magnitude at minimum |
Period_h | 1.89 | period in hours |
Derived: amp = Max − Min, mag_median = (Max+Min)/2.
Copy source: the observatory workspace /home/boreal/Develop/astronomy/ns-hads (snapshot; the live sheet above is fresher and wins at runtime — see §Source of truth & refresh). Attribution added to nightscribe/assets/ATTRIBUTION.txt.
nightscribe/assets/hads-coverage/HADS-Project-YYYY.csv (2011–2026, stretch data)
The same star list plus 12 monthly columns whose cells hold comma-separated observer codes (e.g. JH, dsu08, pv28, SD28, CKH). This is the real monthly coverage of Wils' monitoring programme: who measured which star in which month. Header row ends with stray /,,,/ columns (ignore; likely an artifact of the authoring tool). Planned use (stretch): an "nobody is covering this star this month" urgency signal. Beware it is a snapshot that goes stale; refresh = replace the asset.
4. Science extracted from the Kotysz poster (v10p180)
URL: https://pta.edu.pl/proc/v10p180 — "Variability of HADS stars in TESS", Krzysztof Kotysz (Astronomical Institute, University of Wroclaw). PTA Proceedings, Oct 2020, vol. 10, pp. 180–182.
A sample of ~29 HADS + SX Phe stars from TESS sectors 1–4 was analyzed (proposal of 213 such stars for 2-min cadence).
Pulsation types found: (i) single-mode with harmonics only; (ii) double-mode with period ratio 0.76–0.78 (fundamental + first overtone); (iii) single or double-mode with additional (likely non-radial) modes; some low-frequency terms interpretable as g modes.
Example non-radial HADS: ASAS 032246-7237.8 = TIC 431589510, main frequency f1 = 8.1919 d⁻¹ (period ≈ 0.12207 d ≈ 2.93 h), amplitude 81.7 ppt, with at least three non-radial modes below 0.6 ppt.
Light-curve shapes studied via Fourier decomposition (R21, φ31); two stars of the sample turned out to be RR Lyrae.
The final sample will include ~300 HADS/SX Phe stars; the Petersen diagram separates them from Cepheids and RR Lyrae.
5. Design: transits vs. HADS
| Exoplanet transits (existing) | HADS (new) | |
|---|---|---|
| Event | Predictable t0 + n·P | No known phase: only the cycle period |
| Catalog data | t0, period, duration_h | period_h, Max/Min range |
| Observability gate | star above horizon at mid-transit | star up + contiguous window with ≥ 1 full cycle (2 recommended) |
| Capture advice | baseline + transit + baseline | continuous 2×P capture (confirm repeat + fold) |
| Key metric | transit coverage (ingress/mid/egress) | complete cycles = hours_up / P |
| Cadence | resolve ingress (≥ 3 points) | ≥ 12 points/cycle, ≤ 15 min real (AAVSO) |
| Outcome | light curve of the transit | watch the star pulsate live |
Core module: core/hads.py (built on core/transits.py patterns — coords.samples_tonight, horizon gate + margin per ADR-020). No t0, no phase, no epoch in v1. If a future version enriches via the VSX API, VSX stores an Epoch (HJD of max/min) which would enable transit-style maximum prediction — a documented stretch.
See PLANS/hads-stars.md for the full integration map (file-by-file, scoring, GUI, projects, tests, phases).