# Citation Audit — JPL_ASTRO_EARTH_10

**Candidate:** JPL_ASTRO_EARTH_10
**Source audited:** `dissertation.md` (read-only; the dissertation was not modified)
**Method:** Every reference identifier was resolved live. DOIs were checked against
`https://api.crossref.org/works/<doi>` (HTTP 200 = resolves, 404/410 = dead,
403/429/timeout/network = unverifiable). NTRS citations were checked via the
hardened `https://ntrs.nasa.gov/api/citations/<id>` API rather than the SPA route
(the NTRS single-page app returns 200 for every path). Other URLs were checked by
HEAD then GET. Every dead DOI was passed to the ARGOS near-miss resolver
(`argos/near_miss_resolver.py`) for classification and a suggested correction.

## Summary

| Metric | Count |
|--------|-------|
| Total references | 141 |
| Resolved (live, 200) | 140 |
| Unverifiable (403 / timeout / publisher bot-block) | 0 |
| Flagged dead (confirmed 404/410 or near-miss defect) | 1 |

### Flagged breakdown by ARGOS classification

| Classification | Count |
|----------------|-------|
| mistyped_recoverable | 1 |
| misattributed | 0 |
| fabricated_neighbor | 0 |
| not_found | 0 |
| unverifiable | 0 |

Identifier mix: 115 DOIs, 26 non-DOI URLs (arXiv abstract pages, 21 NASA NTRS
citations, 2 RePEc EconPapers chapters, 1 Utah State Digital Commons SmallSat
paper). All 26 URLs reach a live target; all 21 NTRS IDs resolve through the
citations API.

This is a clean reference list. Exactly one DOI is a genuine defect, and it is a
recoverable typo pointing at a real, resolving AIAA source. No reference shows
any sign of fabrication.

## Repair Worklist — flagged references

| Ref | Author-Year (cited) | Cited DOI | HTTP | Classification | Suggested fix |
|-----|---------------------|-----------|------|----------------|----------------|
| 17 | M. W. Maier, P. Collopy, E. Hollingsworth, "Value-Driven Design," *Journal of Spacecraft and Rockets / AIAA*, 2012 | `10.2514/1.54033` | 404 | mistyped_recoverable | Replace with **`10.2514/1.C000311`** — P. D. Collopy & P. M. Hollingsworth, "Value-Driven Design," *Journal of Aircraft*, 2011 (Crossref-verified, resolves 200). Also correct the author list and venue: drop "M. W. Maier," and change the journal to *Journal of Aircraft* (the cited venue and the "Maier" author are misattributions on a recoverable real source). |

### Resolver evidence for ref 17

The ARGOS resolver returned for cited DOI `10.2514/1.54033`:

- `"exact": { "resolves": false, "http": 404, "unverifiable": false }`
- `"classification": "mistyped_recoverable"`
- `"note": "a topic-matched paper resolves and closely matches the cited title (similarity 0.60); likely a typo of a real source, recoverable by correcting the DOI"`
- `"archivist_action": "soft finding: fail backward to Stage 1 to correct the DOI, the source is real"`

The resolver's own top topic-neighbor (`10.1002/9780470686652.eae553`,
"Value-Driven Design and Operational Value," Encyclopedia of Aerospace
Engineering) is a different work and should not be used as the fix. A direct
Crossref bibliographic search on the cited author/title string returns the
canonical archival paper:

- DOI `10.2514/1.C000311` resolves (200), title "Value-Driven Design,"
  *Journal of Aircraft*, 2011, authors **Paul D. Collopy** and
  **Peter M. Hollingsworth**.

This is the intended source. The cited DOI `10.2514/1.54033` is a malformed AIAA
DOI for the same author pair; the citation additionally carries a spurious lead
author ("M. W. Maier") and the wrong AIAA journal. The recommended repair keeps
the reference (it is real) and corrects the DOI, authors, and venue.

## Notes on guardrail compliance

- No reference was flagged on a 403 or a timeout. The audit produced zero
  unverifiable rows; every identifier returned a definitive 200 except ref 17's
  confirmed 404.
- The single dead DOI was confirmed by Crossref 404 AND independently corroborated
  by a near-miss to a real, resolving source, so it is a true defect and not a
  publisher bot-block artifact.
- The NTRS rule was applied: all 21 NTRS references were checked via the citations
  API, not the SPA, so none were spuriously marked "resolves" by the SPA's blanket
  200s.