The Al Taqaddum thermal object

The recording and its provenance

The official archive identifies this as infrared footage from a force-protection aerostat near Al Taqaddum, Iraq, in October 2017. Its published file runs for approximately seventeen minutes, much longer than the short extracts that circulated under the nickname “jellyfish.” The camera view contains the object, ground structures and tracking graphics. Those contextual features matter: a crop that removes the background also removes much of the information needed to test motion. The archive supplies the recording, but does not release every calibration or operational detail needed to reproduce the analysis independently.

What the image actually shows

At 2:51 in the recording, an irregular bright silhouette is visible against buildings and open ground. Its outline has hanging-looking extensions; the appearance alone does not establish appendages, propulsion or a biological form. Infrared brightness describes the sensor response, not an ordinary daylight colour. Changes in contrast and background can alter the apparent edges of a small target. Viewers can examine whether the same features persist through the longer sequence, while avoiding conclusions about material or temperature from a compressed public copy without radiometric calibration.

The official reconstruction

AARO’s later resolution attributes the object to a cluster of fully or partly inflated balloons and reports high confidence that the recording demonstrates no anomalous performance. The agency says it used the full motion video, metadata, line-of-sight reconstruction and weather information. This is a substantive explanation with a stated analytical basis, but the published conclusion is not a substitute for all of its underlying inputs. A reproducible comparison would expose the assumed object height, camera position, wind profile and uncertainty bounds so another analyst could test the same trajectory.

Independent discussion and competing readings

Before the official resolution, contributors to the Metabunk discussion examined balloon shape, infrared behaviour and the camera system. These are independent proposed explanations, not a single formally established result. Participants also questioned what could be inferred from a target that remains approximately fixed in a tracked image. The discussion is useful for identifying testable alternatives: camera tracking can conceal apparent ground-relative movement, and an unusual outline can have several ordinary causes. Neither a resemblance to balloons nor a resemblance to a creature, by itself, measures the object’s distance or speed.

Chronology and a date discrepancy

The DVIDS record carries an October 1 metadata date, whereas the later AARO resolution identifies October 23, 2017. This entry therefore uses the month rather than silently choosing one exact day. The official video was posted in April 2025, and the resolution followed in September 2025. Publication dates should not be mistaken for observation dates. Claims about later water entry or underwater travel are not demonstrated by the ground-view sequence presented here; any such claim requires a separate original recording and a traceable connection to this event.

What remains available to test

This case offers unusually useful duration, visible terrain and an explicit proposed explanation. The public can compare changes in outline and image scale, but an angular measurement still needs range before it becomes a physical size. A ground feature and an airborne target need not share the same distance. The important remaining question is whether an independently reproducible reconstruction supports the reported balloon trajectory across the complete sequence. The released evidence does not support turning an unidentified silhouette into an extraordinary vehicle merely because a short extract initially looks unfamiliar.

Sources & further reading

• Al Taqaddum Object — original release — AARO / DVIDS
• Al Taqaddam case resolution — AARO
• Independent infrared and balloon discussion — Metabunk contributors