DOW-UAP-D3, Mission Report, Arabian Gulf, 2020
Evidence media
- Official PDF: Open the verified Release 01 PDF. The release-file copy is a seven-page redacted MISREP; it does not contain object photographs, cockpit/sensor frames, a plotted track, or a map.

Page 1 derived render from the official PDF: MISREP 8799515 narrative/admin/classification/operation fields with heavy redactions. This is source-document context, not object imagery.

Page 6 derived render from the official PDF: UAP form fields for observer/platform/location/signature details. Most fields are blank or redacted; the visible UAP Advanced Capabilities And/Or Materials entry is NO, and the event serial field is a dash.

Page 7 derived render from the official PDF: GENTEXT/UAP language describing four reported UAP around 1736Z, cloud obstruction, and the sequence at 17:36:22, 17:36:30, and 17:36:49. The page is text/redactions only, not a sensor frame or photograph.
Investigation reading
This released file is a seven-page, redacted MISREP-style mission report titled DOW-UAP-D3, Mission Report, Arabian Gulf, 2020. The official release description summarizes the reported observation as a military operator seeing a “line of dots followed by a trailing dot.” The document itself gives the clearest surviving detail in the final GENTEXT/UAP section: at approximately 1736Z, four UAP were reported under a redacted reference point, with one observed at 17:36:22, two side by side at 17:36:30, and one at 17:36:49.
The available source is a form report, not an imagery exhibit. Review covered the verified release PDF, its complete seven-page OCR, the selectable text layer, and rendered page images. The visual pass shows typed/redacted form pages only: no embedded object photograph, no map, no radar scope, no plotted track, no cockpit/sensor frame, and no diagram. Page 5 is visually a redacted image page headed “Global Campaign Plan”; the selectable text layer is effectively blank there, while the OCR and rendered page confirm the page is not actually blank.
This draft does not resolve the event. It preserves what the released file says, what it does not show, and what should be checked before any stronger interpretation.
What the file appears to contain
| Page | Public reading from the released PDF |
|---|---|
| 1 | MISREP 8799515 cover/narrative page with heavy redaction, classification metadata, operation fields, AFCENT listed as MAJCOM, and report type MISREP. |
| 2 | Message/tasking and personnel routing fields: MSNID, POC, QC, and approver sections. Names, unit details, and contact fields are redacted. |
| 3 | Aircraft/equipment section (ACEQUIP) with prompts for aircraft callsign, radar, RWR, MWS, IRCM, ECM, chaff/flares, weapons, target pod, and avionics. Most entries are blank or redacted. These are form fields, not evidence that each listed system generated a UAP return. |
| 4 | Timeline and ISR section. The visible text says weather was not a factor in a takeoff/details field. The ISR block lists time-on/time-off station, aircraft callsign, mission type, primary sensor, sensors available, tasking, activity description, and an FMV/image-file field, but the key values are redacted or blank. |
| 5 | Redacted image-like page headed “Global Campaign Plan.” Rendered review confirms it is not blank, but the substance is covered by redactions. |
| 6 | UAP form fields: initial contact, event type, maneuverability, observer response, aircraft state, first/last seen locations, altitude/velocity/trajectory, physical state, number sighted, signatures, RF fields, and event serial. Many fields are blank or redacted. The visible populated negative entry is UAP Advanced Capabilities And/Or Materials: NO; the event serial field shows a dash. |
| 7 | GENTEXT/UAP continuation. It reports 4X UAP OBSERVED flying in a redacted field of view; cloud coverage obstructed a redacted observer/sensor from following and getting a clear visual; the event description records the 1736Z sequence: 1 UAP at 17:36:22, 2 UAP side by side at 17:36:30, and 1 UAP at 17:36:49. |
The key released language is narrow. The document supports a report of four observed UAP/dots in a short time sequence, with cloud obstruction and limited visual clarity. It does not provide a clear object image, raw sensor data, independent track, velocity estimate, altitude, trajectory, recovery, effects on people/equipment, or a populated RF/signature field in the visible text.
Source custody and provenance
- Official/source URL: https://www.war.gov/medialink/ufo/release_1/dow-uap-d3-mission-report-arabian-gulf-2020.pdf
- Open Sky release-file route: war-gov-dow-uap-d3-mission-report-arabian-gulf-2020-fd8bfb94
- Agency: Department of War
- Release: PURSUE Release 01
- Official CSV row: 46
- Source type: PDF, 7 pages, 101,111 bytes
- SHA-256:
58219b8000454aa27424ead9530bf80420365a587b95d4c770a5dce916783b45 - OCR coverage: Frontier OCR complete for 7 / 7 pages; rendered-page review was used to check the page-5 redaction page and the UAP/GENTEXT pages.
The title and official file name place the report in the Arabian Gulf, 2020, but the manifest row does not provide an unredacted incident date or incident location. The PDF’s visible form text also leaves the key operational location, platform details, and many times redacted.
Graph context
Open Sky’s graph has this item as an official primary-source Document asset linked to the Release 01 record for CSV row 46. The graph currently preserves 40 extracted claims, 30 entity mentions, 4 sensor-event records, and 0 table rows for this file.
The most useful graph claims match the source-level reading: document identity (DOW-UAP-D3), title/location/year metadata (Arabian Gulf, 2020), the official manifest wording about a “line of dots followed by a trailing dot,” and the page-7 UAP/GENTEXT language about four UAP around 1736Z. The graph also points to repeated redaction and missing-data claims, which are important here because most operational specifics are withheld.
The four graph sensor-event records should be treated as navigation cues, not conclusions. The source contains aircraft/equipment and ISR/FMVs prompts, plus redacted sensor fields, but the released PDF does not show raw radar returns, IR frames, FMV stills, a video file name, or a sensor plot. The page-3 equipment prompts and page-4 ISR form fields can inflate the appearance of sensor coverage if read without the rendered PDF.
No candidate crosslinks are currently recorded for this asset. Related graph records include the Release 01 row record for this PDF and other MISREP-style Release 01 documents; those are structural neighbors, not evidence that the incidents are connected.
Leads to check
- Locate any paired ISR/FMVs or still imagery referenced by “SEE ISR 1” and by the page-4 FMV/image-file form field. The released PDF itself does not include that media.
- Clarify whether the official manifest phrase “line of dots followed by a trailing dot” is a human-readable paraphrase of the page-7 timing sequence or comes from another unreleased field.
- Check whether the redacted “beneath [redacted]” reference point on page 7 can be resolved through any separately released mission context without exposing protected information.
- Separate form prompts from populated evidence fields in future graph cleanup: radar, RWR, IR, and FMV words appear in template/equipment areas, but the visible UAP entry does not present raw instrument data.
- Compare this PDF with any related Release 01 MISREPs that include paired video/still records, especially to see whether the same reporting template produces similar graph sensor-event inflation.
Lead check notes
- Blocked — paired ISR/FMVs or still imagery: The D3 release metadata and Open Sky release-file copy expose only the PDF. The PDF references
SEE ISR 1and includes an FMV/image-file form field, but the current linked Release 01 record has no paired PR/video row, DVIDS ID, still image, or released ISR file for this item. Resolving this lead requires the underlying ISR/FMVs or still package, if it exists and can be released. - Partial — manifest phrase versus page-7 source wording: The phrase “line of dots followed by a trailing dot” is present in the official release metadata, while the PDF text/render gives the source-level sequence: one UAP at 17:36:22, two side by side at 17:36:30, and one at 17:36:49. The phrase is not visible in the PDF text itself, so it remains unclear whether it is a manifest paraphrase or language from another source field.
- Blocked — redacted reference point: The
BENEATH [REDACTED]wording appears in the D3 GENTEXT/UAP page; the current linked Release 01 text does not expose the redacted reference point. Resolution would require separately released mission context that can be cited without exposing protected operational details. - Checked — form prompts versus populated evidence fields: The rendered/source pages support keeping radar, RWR, IR, FMV, and equipment words as form/template context unless a populated return, filename, frame, or sensor product is separately sourced. The visible UAP section does not present a raw instrument return or object image.
- Partial — related MISREP template comparison:
SEE ISR 1appears in other Release 01 MISREP-style documents, including records that do have separate PR/video pairings. For D3, the current record links only the PDF, so this comparison supports metadata/graph cleanup rather than an evidentiary connection between events.
Limits
This is a redacted official report, not an adjudication package. Most platform, sensor, location, speed, altitude, trajectory, and observer details are blank or redacted. The visible text says cloud coverage prevented following the objects and getting a clear visual. There are no object photos, no maps, no radar plots, no sensor frames, and no independent corroborating media in this PDF.
The page should therefore be used as a source-custody and reading aid: it preserves a reported short-duration observation of four UAP/dots and the document’s constraints, while leaving explanation, corroboration, and prosaic checks for later human review.
Deep investigation — graph + web reconnaissance
Source reread
- The verified Open Sky release-file copy still matches the published asset metadata: SHA-256
58219b8000454aa27424ead9530bf80420365a587b95d4c770a5dce916783b45, 101,111 bytes, seven PDF pages, title DOW-UAP-D3, Mission Report, Arabian Gulf, 2020. - Page 7 is the controlling source passage. OCR and rendered-page review both show:
4X UAP OBSERVED FLYING IN [REDACTED] FOV; cloud coverage obstructed the observer/sensor from following and getting a clear visual; and the GENTEXT sequence records one UAP at17:36:22, two side by side at17:36:30, and one at17:36:49. - Page 6 is a redacted UAP form page, not sensor media. The visible populated entries include
UAP Advanced Capabilities And/Or Materials ... NOandUAP Event Serial Number: -; the observation/interrogation, signatures, RF, altitude/velocity/trajectory, and many platform/location fields are blank or redacted. - The derived renders checked for pages 6 and 7 show only typed form text and redactions. They do not show a UAP photograph, video still, radar/IR/EO frame, map, plotted track, or other object imagery.
Graph connections checked
- A read-only exact SHA/URL graph check finds a single official asset
Documentfor this PDF and a separate Release 01 manifest-rowDocumentwith the same title. Same-hash and same-URL checks do not show duplicate official asset nodes. - The asset node currently carries 40 machine-extracted
Claimnodes, sevenTextChunknodes, fourSensorEventnodes, and one officialSourcelink to the WAR.GOV/PURSUE Release 01 landing-page spine. The claim/sensor records are markedmachine_extracted_needs_human_review/not_a_findingand should remain navigation aids until checked against source text. - The four
SensorEventrecords are deterministic extracts from form/template language such as aircraft, IR, and FMV fields. The PDF itself does not provide raw sensor returns, an FMV filename, a still frame, or a plotted sensor track, so these records should not be read as confirmed sensor corroboration. - Relationship audit found the correct current manifest relationship for CSV row 46. It also found a stale
RELATED_TOrelationship to the D32/Syria row 47 record withcurrent_in_official_csv: false; that is a manifest-reconciliation cleanup lead, not evidence connecting the D3 and D32 events. - Entity extraction collapses mainly to Department of War, AARO, Arabian Gulf, redaction markers, generic platform/sensor terms, dates/times, and report IDs. No named witness, exact coordinate, platform type, or unredacted operational location is exposed by this page.
External provenance and web reconnaissance
- Direct official HTTP probes to the WAR.GOV/PURSUE PDF, CSV manifest, and landing page returned Akamai
403responses during this check. The verified Open Sky release-file copy, graph manifest record, and stored OCR therefore remain the usable source path until canonical access is reachable again. - DVIDS exact-title reconnaissance did not produce an accessible result: the DVIDS search endpoint returned a CloudFront WAF challenge (
202) rather than results, and the Release 01 manifest row has no PR number or DVIDS video ID for D3. This is not a negative finding; it only means no paired DVIDS media is currently exposed in the checked release metadata. - AARO public-page/search probes also returned
403, so no independent AARO page was confirmed for this item in this pass. - The graph points to a secondary UFO-USA GitHub markdown conversion for row 46. Its page-7 transcription reproduces the same GENTEXT wording and is useful as a comparison lead, but it is not canonical over the official WAR.GOV/PURSUE PDF.
Prosaic checks and limits
- The source itself flags cloud obstruction, which is a major limit on visual clarity. A separate takeoff/weather line is template-like and redacted, so it should not override the page-7 cloud statement.
- Astronomy, weather, satellite/launch, aircraft, flare, and line-of-sight checks are blocked by missing basics: unredacted event date, exact location/reference point, azimuth/elevation, platform position, observer/sensor identity, field of view, and whether
SEE ISR 1refers to releasable media. The manifest phrase “line of dots followed by a trailing dot” is compatible with several prosaic check lanes, but no correlation should be claimed without those fields. - The strongest next evidence target is the referenced
ISR 1/ FMV or image-file material, if it exists in a releasable package. Without it, this page remains a redacted source-custody record for a short reported observation, not an adjudication package.
Follow-up leads
- Locate any releasable
ISR 1, FMV, still-image, or mission attachment referenced by the form. - Clean or annotate the stale graph relationship to D32 row 47 so future graph traversals do not imply an event-level connection.
- If the date/location/platform details are ever unredacted, run a bounded prosaic correlation set: cloud/weather at the observation point, satellite trains/reentries/launches around 1736Z, military/civil aircraft tracks where available, flare/munition activity, and sensor field-of-view geometry.
Audit note
This section adds source reread, read-only graph context, and external provenance checks. It does not create a finding, hypothesis, or resolution decision.
Sources
- Department of War, PURSUE Release 01, DOW-UAP-D3, Mission Report, Arabian Gulf, 2020, official PDF: https://www.war.gov/medialink/ufo/release_1/dow-uap-d3-mission-report-arabian-gulf-2020.pdf
- Open Sky release-file route for the verified PDF: /api/explore/war-gov/release-file/war-gov-dow-uap-d3-mission-report-arabian-gulf-2020-fd8bfb94
- Release 01 manifest record, CSV row 46, Department of War, redacted PDF asset.