DOW-UAP-PR38, Unresolved UAP Report, Middle East, 2013
Evidence media
- Open Sky release-file copy of the official PR38 MP4 — verified as a 30,707,537-byte MP4 with SHA-256
bd3f5269c3f123b90db76c9496f8ddbb145184c66b0c11cd3cfd9e54f5b683b6.
Derived frame context/contact sheets from the official MP4 are included below for orientation. They are review images made from the released video frames, not standalone object identifications.
Sparse full-frame samples across the released video show the monochrome sensor-display layout, fixed reticle/symbology, black masking/redaction blocks, and the changing position and appearance of an unresolved high-contrast feature. The sheet documents display context and timing samples only; it does not establish range, size, speed, altitude, or identity.
Central sensor-crop samples cover the publisher-described timing windows. They preserve the starburst-like high-contrast appearance, faint smear/trail-like image feature, lower/right field position around 00:30–00:31, and later post-cut samples, without deciding whether those appearances are physical scene content, sensor/display behavior, compression/processing, or another ordinary effect.
Investigation reading
This draft reads PR38 as a short official DVIDS video record released through WAR.GOV/PURSUE Release 01. The verified release copy is a 30,707,537-byte MP4 with SHA-256 bd3f5269c3f123b90db76c9496f8ddbb145184c66b0c11cd3cfd9e54f5b683b6. Its media stream is H.264 video at 1920×1080, 30 frames per second, about 1 minute 46 seconds long, with an AAC audio stream that is effectively silent in the reviewed copy.
The DVIDS source page lists the title as DOW-UAP-PR38, Unresolved UAP Report, Middle East, 2013, category B-Roll, Video ID 1006088, VIRIN 130102-D-D0360-8893, date taken 01.01.2013, date posted 05.08.2026 07:49, and location (UNDISCLOSED LOCATION). The release metadata gives the incident location as Middle East and does not provide a specific incident date beyond the 2013 DVIDS date field.
The official description says U.S. Central Command submitted a report to AARO consisting of one minute and 46 seconds of infrared-sensor video from a U.S. military platform in 2013. It also states that the reporter did not provide an oral or written description of the observation. The source's own caution is important: the video description is informational and should not be treated as an analytical judgment, investigative conclusion, or factual determination about the event.
What the file appears to contain
The video is a monochrome, infrared-like sensor-display recording. Most frames include fixed display symbology, a central reticle/crosshair, edge markings, and multiple black rectangular masks or redaction blocks. Those masks and overlays are part of the released video presentation, not scene content.
The visible feature is an unresolved high-contrast area that often appears as a compact dark/bright mark with radial or starburst-like smear. At early sampled times it is subtle; around the official 00:10 note it becomes more distinct as the field of view changes. From roughly 00:11 through 00:29, the feature is visible in the lower half of the frame and shifts position relative to the reticle and masking. Several frames show a faint elongated smear or trail-like image feature, but the release does not establish whether that is physical trailing, sensor integration, motion blur, contrast processing, compression, or background interaction.
Around 00:30 to 00:31, frame review is broadly consistent with the official note that the feature leaves the visible field near the lower/right side. After an apparent cut near 00:35, the feature is not obvious in the first post-cut sample, then reappears as a similar high-contrast/starburst-like mark in the right half of the frame and remains intermittently visible through much of the remaining clip. By about 01:44 it is faint and poorly resolved near the upper portion of the frame, and by the end of the clip it is no longer clearly separable from background texture and display artifacts.
No narration, transcript, cockpit audio, written witness statement, range/range-rate readout, platform track, target track file, exact location, weather record, or aircraft/airspace context is visible in the released MP4.
Source custody and provenance
Primary source: DVIDS video 1006088, canonical URL https://www.dvidshub.net/video/1006088/dow-uap-pr38-unresolved-uap-report-middle-east-2013. The Open Sky release-file route is /api/explore/war-gov/release-file/war-gov-dow-uap-pr38-unresolved-uap-report-middle-east-2013-1006088.
Release 01 associates this video with WAR.GOV/PURSUE CSV row 95 in the video inventory. The graph also has a WAR.GOV release-record document for the current release row view, with a canonical WAR.GOV release anchor and DVIDS URL. Those two records should be kept distinct: one is the video evidence record, the other is the release/manifest record that points to it.
The date/location custody is limited. DVIDS gives date taken 01.01.2013 and location (UNDISCLOSED LOCATION), while the release metadata summarizes the place as Middle East and the incident date as N/A. That is sufficient for source provenance, but not enough for independent weather, astronomy, air-traffic, satellite, launch/reentry, or local operating-area checks.
Graph context
The graph has an exact VideoEvidence record for DVIDS 1006088 and a related WAR.GOV release-record Document. The current semantic integration for this item contains 11 extracted claims, 9 entity mentions, 1 sensor-event record, and no table rows. These graph facts are useful navigation and provenance context, not findings.
The single sensor-event cue is based on the release text's reference to a U.S. military platform/aircraft and infrared-sensor video. It should not be read as an independent radar, multi-sensor, or track-file confirmation. The public video itself is the available evidence item; it does not expose raw instrument data or platform telemetry.
Candidate crosslinks in the graph include broad anchor matches such as Middle East, Unresolved UAP Report, and unrelated or weak contextual matches. They remain review leads only. Nothing in this page resolves PR38 by reference to those crosslinks.
Leads to check
- Locate any paired mission report, operations log, or AARO case record that can tie PR38 to an exact collection date/time, platform, sensor, location, range, and operator context.
- Compare the released MP4 against any higher-quality original sensor file, if available, to test whether the starburst/radial appearance and faint trail-like smear persist outside compression and export processing.
- Run a frame-by-frame stabilization/registration pass before interpreting apparent motion; the feature's path in the display is affected by reticle position, zoom/framing, masking, and possible sensor movement.
- Check whether the
01.01.2013DVIDS date is an actual event date or a placeholder/custody date for older media. - If location/time are later obtained, perform ordinary prosaic checks: aircraft, balloons, drones, munitions, flares, birds, atmospheric effects, sensor artifacts, weather, traffic, satellite/reentry/launch windows, and operating-area activity.
Lead check notes
- Checked — PR38 release-file copy and video metadata: The Open Sky release-file copy matches the Release 01 video inventory record for DVIDS
1006088: 30,707,537 bytes, SHA-256bd3f5269c3f123b90db76c9496f8ddbb145184c66b0c11cd3cfd9e54f5b683b6, H.264 video at 1920×1080/30 fps, and about01:46duration. The DVIDS page supplies VIRIN130102-D-D0360-8893, date taken01.01.2013, and(UNDISCLOSED LOCATION), but not an exact event time, platform, sensor settings, range, or coordinates. - Checked — frame-context media: Derived contact sheets from the official MP4 now cover sparse full-clip samples and the publisher-described timing windows. They support a cautious visual reading of a monochrome sensor display, masks/redactions, fixed reticle/symbology, a starburst-like high-contrast feature, faint smear/trail-like image behavior, a lower/right field position around
00:30–00:31, and later post-cut samples. They do not identify the feature or measure its distance, speed, size, altitude, or independent motion. - Partial — paired-report and AARO context: Current release metadata and graph context expose the PR38 video record and a WAR.GOV release-record document, but no public paired mission report, operator statement, AARO case file, platform log, or raw sensor export for PR38. Broad related records and candidate crosslinks remain navigation leads only until exact source pages are compared.
- Blocked — original-file and stabilization checks: A higher-quality original sensor file, unredacted frame sequence, telemetry, stabilization/registration pass, platform track, sensor calibration, and range/range-rate data are not present in the public release copy. Without those sources, the starburst/radial appearance and smear/trail cannot be separated from sensor response, display processing, compression, focus, motion blur, or background interaction.
- Partial — date custody: DVIDS lists
01.01.2013, while the Release 01 incident date field remainsN/Aand the location is onlyMiddle East/(UNDISCLOSED LOCATION). Treat the date as source metadata until a WAR.GOV/DVIDS/AARO manifest history or underlying report confirms whether it is an actual collection date or a custody/placeholder date. - Needs external source — prosaic/environment checks: Aircraft, balloon, drone, munition/flare, bird, atmospheric, sensor-artifact, weather, traffic, satellite/reentry/launch-window, and operating-area checks remain open because the public source lacks exact time, coordinates, platform geometry, viewing direction, altitude/range, and raw collection data.
Limits
This draft does not identify the feature, estimate size, altitude, speed, distance, or intent, or decide whether the feature is anomalous. The released video is compressed, masked, and lacks raw sensor metadata. The reporter's underlying oral or written description is absent from the released source. The visual feature's changing starburst-like appearance may be partly or wholly shaped by sensor response, display processing, compression, focus, saturation, or motion effects.
The page should therefore be treated as a provenance and source-reading draft: it documents what the released video and graph currently support, what they do not support, and what a human reviewer should check next.
Deep investigation — graph + web reconnaissance
Source reread
The live DVIDS page for Video ID 1006088 was reachable during this review and still resolves to the canonical PR38 title, DOW-UAP-PR38, Unresolved UAP Report, Middle East, 2013. It lists Date Taken 01.01.2013, Date Posted 05.08.2026 07:49, VIRIN 130102-D-D0360-8893, filename DOD_111689051, length 00:01:46, and location (UNDISCLOSED LOCATION). A range check against the official MP4 delivery URL returned bytes 0-15/30707537 with MP4 ftyp magic, matching the Open Sky release-file copy's 30,707,537-byte size and SHA-256 bd3f5269c3f123b90db76c9496f8ddbb145184c66b0c11cd3cfd9e54f5b683b6.
The source description is narrow: U.S. Central Command submitted an AARO report consisting of one minute and 46 seconds of infrared-sensor video from a U.S. military platform in 2013, and the reporter did not provide an oral or written description. DVIDS describes the visible feature as an area of contrast resembling an eight-pointed star with arms of alternating length; notes a sensor field-of-view narrowing at 00:10; describes apparent movement with a visible trail from 00:11 to 00:29; notes exit near the bottom right around 00:30; and describes post-cut visibility from 00:35 to 01:44. The page's own disclaimer is important provenance language: the description is informational only and should not be read as an analytical judgment, investigative conclusion, or factual determination.
Frame/media review
The verified MP4 stream is H.264 video at 1920×1080/30 fps, about 106.43 seconds long, with an AAC audio stream. Review of the derived contact sheets from the official MP4 supports only a cautious visual reading: a monochrome sensor-display frame with fixed reticle/symbology, black masking/redaction blocks, mottled background texture, and a compact unresolved high-contrast feature that sometimes appears starburst-like or bloomed. Several publisher-window samples show pale linear or curved smear/trail-like image features, including a clearer curved/loop-like smear near the 00:31 sample, but the release does not establish whether those marks are physical trailing, sensor persistence, optical/display behavior, compression, focus, motion blur, or processing artifacts.
The source-visible feature changes position relative to the reticle and masks, including lower/right-side appearances near the publisher's exit window and later smaller/fainter post-cut appearances. Those frame observations are consistent with the DVIDS timing notes, but they do not measure distance, size, speed, altitude, heading, independent motion, or identity. The masks and export/compression context prevent treating the starburst shape as a reliable physical outline.
Graph connections and provenance hygiene
The Neo4j record with stable identity is the exact VideoEvidence asset official:video:war-pursue-uap-release:dvids-1006088, tied to the DVIDS URL, the CloudFront MP4 URL, Release 01 video row 95, incident date N/A, incident location Middle East, and the same MP4 hash and byte count above. Its semantic integration currently contains 11 machine-extracted Claim nodes, 9 EntityMention nodes, and 1 SensorEvent node, all still marked as machine-extracted review material rather than findings.
The single sensor-event cue is not a radar, track-file, telemetry, or multi-sensor confirmation. It is an unreviewed machine extraction from manifest wording about a U.S. military platform/aircraft and infrared-sensor video. The graph also preserves one text chunk containing the official manifest description, including the AARO/CENTCOM source language and the DVIDS timing notes.
Graph provenance review surfaced manifest-row drift that should be cleaned separately: one current release-record document carries the PR38 title and DVIDS ID while its final-file URL points to a D35 Greece PDF, and the row-95 release-record relationship to this video carries the PR38 final URL while retaining stale PR41/DVIDS 1006094 title fields. Those are custody/indexing issues, not corroborating reports. For PR38, the safe provenance anchor is the exact DVIDS 1006088 video record plus the verified MP4 hash, not mixed row-record titles.
No direct CANDIDATE_CROSSLINK relationship was found from the exact PR38 video asset in this check. Broad inventory-level anchors such as Middle East, Unresolved UAP Report, or unrelated regional/theme matches should remain navigation leads only unless a source page, timestamp, platform, or exact media identity is later verified.
External context and prosaic checks
The official public record still lacks the key variables needed for ordinary correlation work: exact event date/time, coordinates, platform, sensor model/settings, look direction, range/range-rate, target track, operator notes, weather, airspace, and surrounding mission context. DVIDS Date Taken 01.01.2013 is source metadata, but Release 01 keeps the incident date as N/A; treat the DVIDS date as unconfirmed event timing until a paired report or manifest history proves it. Date-only graph probes for LaunchEvent, AstronomyEvent, and WeatherEvent records on 2013-01-01 returned no modeled rows, but that is a graph-coverage/data-availability note, not a prosaic exclusion.
The first prosaic lanes are imaging and collection-context checks: sensor bloom/saturation, focus or optics, display/export processing, compression, reticle/mask interaction, platform/sensor motion, stabilization/registration, atmospheric contrast, and ordinary airborne or surface activity such as aircraft, drones, balloons, birds, flares/munitions, or operating-area traffic. Satellite/reentry/launch checks remain blocked until the event time, location, and viewing geometry are released.
Limits and follow-up leads
This page still does not identify the feature or decide whether it is anomalous. The released evidence is a compressed, masked public MP4 with no raw sensor export, telemetry, platform track, range data, unredacted frame sequence, witness statement, or paired mission report. Follow-up should prioritize locating the underlying CENTCOM/AARO report package, any PR38 mission or sensor logs, original sensor-file provenance, and a row-manifest cleanup pass that separates the stable DVIDS video asset from stale release-record fields.
Sources
- DVIDS official video page: DOW-UAP-PR38, Unresolved UAP Report, Middle East, 2013
- Open released MP4: war-gov-dow-uap-pr38-unresolved-uap-report-middle-east-2013-1006088
- WAR.GOV/PURSUE Release 01 source dataset:
war_pursue_uap_release_2026_05_08 - Open Sky semantic dataset:
war_pursue_release01_semantic_2026_05_12 - MP4 SHA-256:
bd3f5269c3f123b90db76c9496f8ddbb145184c66b0c11cd3cfd9e54f5b683b6