The Diamond Seen Only in Short-Wave Infrared

What the mission report says

The report is a seven-page mission report headed “Misrep undefined-9629373”, from the 33rd Special Operations Squadron (“Originator (Unit or Squadron): 33 SOS”). USCENTCOM declassified it on 24 October 2025. The aircraft “CONDUCTED SLR TAKE OFF FROM LGLR”, Larissa in Greece, at 01:09Z on 25 January 2024, arrived on station at 06:35Z and landed back at Larissa at 21:49Z. Its type, tail number and call sign are blacked out, and so are the unit and the operation it supported.

The sighting takes one line of the narrative, “AT 0509Z, […] OBSERVED AN UNIDENTIFIED AERIAL PHENOMENON (SEE UAP 1).”, and fills the UAP section on pages 6 and 7. Its account in full:

“AT 0509Z, […] OBSERVED 1X UAP WHILE IN TRANSIT. UPON INITIAL OBSERVATION, THE UAP WAS FLYING AT A SPEED OF APPROX 434 KNOTS. THE UAP APPEARED DIAMOND IN SHAPE WITH A NON-MANUEVERING PROBE AT THE BOTTOM. THE UAP ONLY APPEARED ON THE SWIR CAMERA. THE EVENT LASTED APPROX 2 MINUTES ENDING AT 0511Z, WITHOUT ANY OTHER INCIDENT OCCURING.”

The description field reads “ROUND DIAMOND SHAPE WITH STRAIGHT, NON MANUEVERABLE “TAIL”“. SWIR is short-wave infrared. The fields around it give the aircraft at “FL250”, on “Friendly Aircraft Trajectory: 162”, at “176 KTS”, and the object as “Solid”, “UAP Signatures: SWIR WHT”, at an estimated “FL200”, moving “W”. Under anomalous characteristics the report has “MAINTAINED STEADY FLIGHT PATH, INCREASED AND DECREASED ALTITUDE PROFILE, DID NOT CHANGE TRAJECTORY”. It answers “NO” on advanced capabilities, “NONE” on third-party observers, “NO” on whether anything interrogated the object, and gives the observers’ assessment as “Benign”. The event carries the serial number “250509ZJAN2024-CENTCOM 001”, and the contact given in the section is a “SENIOR AIRMAN/E-4”.

What the recording shows

The recording is DOW-UAP-PR28, in the release 01 video bundle as DOD_111688954.mp4 and on the publisher’s video page. It runs 1 min 6 s. Every overlay that would give time, position or sensor settings is blacked out. By our viewing of the file:

• 0:00–0:10. The screen is split in two. The left half shows pale, hazy cloud under a bright sky; the right half shows the same clouds with more contrast under a dark sky.
• 0:04.5–0:07. A small bright point sits beside the crosshair of the right half. The left half shows nothing there.
• 0:08.3–0:09. The object, larger, at the upper left of the right half, with a faint trail below it. Again nothing in the left half.
• 0:10.5–0:10.7. The display goes full screen. The full-screen picture carries on the right half: the same clouds, rendered the same way, with the same object.
• 0:12–0:56. The operator holds the object near the crosshair against sky and cloud. It is a bright, compact body, round to diamond-shaped, with a fainter straight element below it that stays upright in the frame throughout.
• 0:56.3–0:57.2. The picture breaks up and then goes blank while the camera changes.
• 0:57.5–0:58.8. A colour picture of blue sky. The object cannot be picked out.
• 0:58.9–1:06. A grainy black-and-white picture. The object does not come back.

The publisher’s description agrees on the broad course: a split screen, a full-screen “SWIR feed” from 00:10, a switch “to visible spectrum, losing the subject against the background” at 00:56, and a switch to “SWIR (Black-Hot)” that “does not reacquire the area of contrast”.

Which half is short-wave infrared

On the split screen, the publisher’s description and the recording do not fit together. The description says “the right side displaying electro-optical footage and the left side displaying SWIR footage”, and that at “00:04: An area of contrast becomes distinguishable against the background in the center of the right frame.” The report says the object “ONLY APPEARED ON THE SWIR CAMERA”, and the description repeats that it “was only detectable via short-wave infrared (SWIR) sensor”.

By our viewing, the object is in the right half only, from about 0:04.5, and at 0:10.5–0:10.7 the full-screen picture that the description calls “the SWIR feed” continues the right half. Either the right half is the short-wave infrared camera and the description has its sides the wrong way round, or the object was visible to the electro-optical camera, and then the report and the description are both wrong to say it appeared only in short-wave infrared. Nothing on the recording labels the halves: the overlays are blacked out.

434 knots without a range

The only speed in the report is in its account: “UPON INITIAL OBSERVATION, THE UAP WAS FLYING AT A SPEED OF APPROX 434 KNOTS.” The field meant for it is empty: “Kinetic Velocity:” is followed by nothing, although the line above gives “Kinetic Velocity Accuracy: Estimated”. Nothing measured a range to the object: “Observation Interrogation of UAP (yes/no; if yes list sensor(s) and returns):: NO”. A camera without a range gives angles, not speeds. The recording, with its overlays blacked out, carries no range, angle or field of view either, so the figure cannot be checked from what was released.

The report also disagrees with itself. Its “First Coordinate” and “Last Coordinate” lie about 81 km apart from east to west by our reading of the grid references, for an event that “LASTED APPROX 2 MINUTES”. If those were the object’s positions, it would have moved at about 1,300 knots by our calculation; at 434 knots it would cover about 27 km in two minutes. The report does not say what the two coordinates are. Both are marked “Estimated”, each with a radius of “20” and no unit. They agree with the direction, “Kinetic Trajectory: W”, and not with the speed.

Where it was

The report gives three grid references in the military grid (MGRS), each partly blacked out: the aircraft’s “Friendly Aircraft Location: 35SQT67[…]8[…]”, the “First Coordinate: 35SQT44[…]7[…]” and the “Last Coordinate: 35SPT63[…]5[…]”. We measured where the surviving digits stand in the text. After the two digits that follow the square letters, the black box takes the place of exactly four characters. Those two digits are therefore the start of the easting alone, and the single digit after the box is the second digit of the northing; the first, which gives the tens of kilometres north to south, is under the box. The report confirms the pattern itself: on page 1 the narrative gives the point where the aircraft started its tasking with spaces, “36R XV 19[…] 63[…]”, and on page 5 the same point appears without them, “36RXV19[…]3[…]”, with the 6 under the box.

So the report gives longitude but not latitude. By our conversion the aircraft was at about 29.88°E, the object was first seen at about 29.63°E and last seen at about 28.76°E, all somewhere between about 33.4° and 34.3°N: open sea some 250–350 km north of Alexandria. The map pin stands at 29.63°E and 33.86°N, the middle of that range. The crew saw the object “WHILE IN TRANSIT” towards their tasking. The publisher’s title, “Greece, January 2024”, follows the aircraft’s base.

The same silhouette in June, over the Gulf of Oman

Five months later a crew of the 3rd Special Operations Squadron reported an object of the same outline. DOW-UAP-D27 describes a “GLOWING HOT SPHERICAL UNIDENTIFIED OBJECT WITH A VERTICAL UNWAVERING CYLINDRICAL POLE/BAR ATTACHED ON THE BOTTOM OF THE OBJECT POSS REFLECTION FROM THE OBJECT IN THE WATER, MOVING AT 140KNOTS” (p. 7), “FLYING STRAIGHT JUST OVER THE WATER” (p. 6), seen “DURING RTB AT 0457Z”, on the way back to base, over the Gulf of Oman by our reading of its grid references. Its recording, DOW-UAP-PR29, runs 21 seconds.

The two reports share an object given as “Solid”, assessed as “Benign”, with “NO” advanced capabilities; a straight path, “MAINTAINED STEADY FLIGHT PATH … DID NOT CHANGE TRAJECTORY” against “FLYING STRAIGHT”; and a straight upright element under the body, a “PROBE AT THE BOTTOM” against a “POLE/BAR ATTACHED ON THE BOTTOM”. They differ in everything else. D25’s object was bright in short-wave infrared only (“SWIR WHT”), estimated at “FL200” and 434 knots, and seen in transit. D27’s was “GLOWING HOT”, “JUST OVER THE WATER”, at 140 knots, and its crew offered a reflection in the water.

The publisher describes both recordings in the same words: “an inverted teardrop with a vertically linear trailing mass suspended below”. By our viewing the two silhouettes look alike: a bright body with a fainter straight element below it, upright in the frame. In PR29 the body is plainly diamond-shaped at 0:06–0:08 and later shows a small dark spot at its lower right. The backgrounds differ. In PR28 the camera looks out at sky above cloud tops; in PR29 it looks down at the sea, whose waves fill the frame.

D27 does not agree with itself on the date. It gives initial contact as “070457:00ZJUN24”, 7 June at 04:57Z (p. 5), and the same time as the date of acquisition (p. 7), but its event serial number reads “060457ZJUN2024-CENTCOM” (p. 6), a day earlier. The publisher’s title for it, “United Arab Emirates, October 2023”, matches neither.

What the publisher got wrong

The publisher’s page for the recording (DVIDS video 1006073) names the accompanying mission report as “DoW-UAP-D7”. D7 describes something else: an object that “LOOKS LIKE A BALLOON”, held as a “WEAPONS QUALITY 1 TRACK” and identified “IN THE TFLIR”. The diamond, the 434 knots and the short-wave infrared all come from D25. Our registry pairs PR28 with D25.

The description gives the recording as “one minute and five seconds”; the file runs 1 min 6 s, and the same page gives its length as 00:01:06. The page lists the location as “GR”, Greece, where the aircraft took off. The sighting was over open sea some 900 km south-east of Larissa.

Why this scores where it does

Witnesses score 2: one crew, trained to watch through its sensors, and no one else (“Third-party Observers and/or Reporters: NONE”). Radar scores 0: nothing interrogated the object. Optics scores 3: a minute of recording shows the object and its outline, but the overlays that would give time, position and pointing are blacked out. Physical trace scores 0.

Multi-channel scores 1: the report and the recording come from the same sensor and the same crew, so they are one channel, not two. Documentation scores 3: a full mission report with times, the aircraft’s height, heading and speed and the object’s estimated height and direction, but every grid reference is cut to its easting. Anomaly scores 1: the only unusual feature is a speed estimated without a range, which the report’s own coordinates contradict; the report itself answers “NO” on advanced capabilities, and being visible to one camera and not another is not unusual in itself. Data today scores 2: the report and the recording are public, but the one claim worth testing needs data blacked out in both.

That puts the case at 1.4 overall, class D, with the status insufficient data. The class is labelled “testimony only” even though there is a recording, because the recording carries no measurement.

What would settle it

The grid references without their redactions. The northing digits would place the aircraft and the object, and show whether “First” and “Last” are positions of the object at all.

The overlays blacked out on the recording. Time, the aircraft’s position, where the sensor pointed and its field of view would let the object’s movement across the sky be measured in angles.

A range. The report says nothing measured one. A radar or laser return, or a second aircraft, would turn 434 knots into a measurement or remove it. The report lists “Data Link: LINK 16”; the picture on that link for 05:09–05:11Z might show whether anything else held the object.

AARO’s file on event “250509ZJAN2024-CENTCOM 001”. Whatever it concluded, if it concluded anything.

The wind at FL200 over that sea that morning. A drift to the west at the speed of the wind would fit a balloon; 434 knots would not.