Case No. 339·Omán
2021·Gulf of Oman, near MGRS coordinates 40REN6 (international/territorial waters)
Omán.Gulf of Oman · the 'cold orbs' that dodged an AC-130's cannon
On the night of September 8, 2021, a U.S. Special Operations Forces AC-130 gunship observed roughly 25 instances of unidentified aerial phenomena during a live-fire exercise in the Gulf of Oman. The electro-optical/infrared sensor captured 'cold orbs' —with no thermal signature of their own— that fled without changing altitude just as the 105mm cannon opened fire, flew in coordinated formations 'like dolphins swimming in a pod,' and reached up to 1,300 mph with no visible means of propulsion. The backup video was corrupted by an onboard recorder failure. The Air Force intelligence report (IIR), declassified in 2026 via PURSUE, explicitly catalogs it as 'not finally evaluated.'
Year
2021
Tier
S
Probability
65%
Category
Incident
Strong evidence: instrumental sensor record + multiple witnesses (typically military). Three independent axes: the «tier» measures the strength of the evidence; the «probability» estimates how genuinely unexplained the case is —a natural phenomenon can remain unexplained, so it does not equal «non-prosaic»—; and the partition of explanations (below) says what it most plausibly was. So a well-documented case can have a possible hoax as its most plausible cause, and a Tier B is not, for that reason, a hoax.
Primary documents · viewer
DOW-UAP-D101 — AFSOC Intelligence Information Report (IIR), Gulf of Oman (published Apr 27, 2022)
US Air Force Special Operations Command / AFSOC A2OX (PURSUE Release 05) · Dominio público · US Government work · open original
Part 01
The night in question
On the night of September 8, 2021, a U.S. Special Operations Forces (USSOF) crew aboard an AC-130 gunship was conducting a live-fire exercise in the Gulf of Oman, in the vicinity of MGRS coordinates 40REN6. The aircraft dropped a water-activated flare to use as a mock target and entered an orbit around it while the Combat Systems Officer (CSO) conducted collateral scans with the aircraft's electro-optical/infrared (EO/IR) sensor. When the CSO slewed the sensor in IR mode back to the flare, they observed two roughly four-foot-diameter 'cold' orbs —with no detectable thermal signature of their own— hovering between 0 and 20 feet above the water's surface, stationary over the flare for about 15 seconds. After that time passed, the CSO fired the aircraft's 105mm cannon. In the brief delay between trigger pull and the weapon's recoil, the CSO observed the pair of orbs rapidly fly away from the flare and out of the sensor's field of view without changing altitude —an immediate reaction to the detonation.
Over the rest of the mission, the crew logged approximately 25 additional instances of these objects, visible in infrared in both white-hot and black-hot polarity. Two flew in a loose but coordinated formation, rising and falling in altitude 'like a wave' while maneuvering around, over, and under each other — the report directly quotes a source comment: 'they knew they were flying together,' explicitly likening it to dolphins' social behavior. Other objects were seen in a formation of three, and separately maneuvering erratically and aggressively. The crew attempted to save the entirety of the EO/IR footage, but the aircraft's Digital Video Recorder (DVR) 'inexplicably failed' and the data became corrupted. The CSO calculated the objects' speeds using the aircraft's onboard tactical tool (TACTOOL): maximum observed maneuvering speed reached 1,300 miles per hour, and minimum observed maneuvering speed was 250 mph. No means of propulsion was observed on any of the objects.
The report was formalized as an Air Force Intelligence Information Report (IIR), originally classified SECRET//NOFORN, published April 27, 2022 —more than seven months after the incident— and distributed to an extensive list of commands (CENTCOM, SOCOM, DIA, NGA, ONI, FBI, CIA, among dozens more). It was coordinated with the U.S. Defense Attaché Office in Muscat and the SOCCENT intelligence cell 'without comment.' The document was formally declassified for mandatory review (MDR) on March 26, 2026 and released via PURSUE that August. The six video recordings attached to the original report, however, remain under the same SECRET//NOFORN classification and are not part of this release.
Part 02
Why this case moved the needle
The Gulf of Oman case earns Tier S for the same reason as Nimitz or Aguadilla: it isn't loose testimony, it's an instrumental sensor record —an AC-130's EO/IR feed, captured in two thermal polarities— corroborated by multiple military crew members and quantified with a formal tactical tool. The Department of War itself labels it 'intelligence information, not finally evaluated': it neither confirms it as an anomalous phenomenon nor closes it as explained, which is precisely the most honest and least-cited category in the public debate.
Three concrete elements sustain the case's analytical weight. First, the thermal signature: objects described as 'cold' yet visible in both infrared polarities don't cleanly match a conventional drone (which would emit engine heat) or marine wildlife (which wouldn't hover at low altitude or react to cannon fire). Second, reactivity: the pair of orbs left position at the exact moment of firing, within the brief window between trigger pull and the weapon's recoil —a timing coincidence that, if real, suggests active awareness of the environment rather than pre-programmed autonomous flight. Third, the DVR failure: the 'inexplicable' loss of the backup footage is, absent further evidence, indistinguishable from an ordinary electronics glitch, but it's exactly the kind of detail that, accumulated across other similar cases in the corpus, feeds suspicion of interference over coincidence.
The posterior calibration reflects genuine uncertainty, not a settled lean toward the anomalous: the Gulf of Oman is one of the world's most congested and contested waterways, with constant Iranian naval activity and frequent allied exercises, so a drone swarm —domestic, allied, or Iranian— remains a serious hypothesis (hence the combined weight on `humana_clasificada` and `adversaria`). But the report's own language —'not finally evaluated,' no observed means of propulsion, immediate reaction to weapons fire— leaves a considerable mass on `indet` and a non-trivial fraction in the non-human categories, honestly proportional to what the document itself is willing to say.
Part 03
What's left on paper
Documented evidence
- AC-130 EO/IR sensor logs ~25 instances of 'cold orbs' during a live-fire exercise (Sept 8, 2021, Gulf of Oman)
- Pair of ~4-foot-diameter orbs, stationary over a flare, flee without changing altitude at the instant the 105mm cannon fires
- Coordinated formation of two objects 'like dolphins swimming in a pod'; source comment: 'they knew they were flying together'
- Maneuvering speeds calculated via TACTOOL: maximum 1,300 mph, minimum 250 mph; no means of propulsion observed
- 'Inexplicable' failure of the onboard Digital Video Recorder (DVR); backup data became corrupted
- AFSOC Intelligence Information Report (IIR), classified SECRET//NOFORN, explicitly cataloged as 'not finally evaluated'; declassified for mandatory review on March 26, 2026
Sources
- war.gov/UFO — DOW-UAP-D101, IIR de AFSOC sobre el Golfo de Omán (PURSUE Release 05) — Document embedded in this case's viewer.
Location
Gulf of Oman, near MGRS coordinates 40REN6 (international/territorial waters) · 24.85°, 58.75°
Distribution of explanations
This case is classified among the model's hypotheses: the bar splits 100% by how much each explanation weighs (the uncertainty is spread across the hypotheses the case supports). Summed across the corpus they produce the comparable partition. It is a different question from the Probability above: that one estimates how likely the case is a genuinely unexplained phenomenon; this one splits which the explanation would be.
Modal hypothesis: Misidentification 29% · sums to 100%
Structured analytical judgment, not a calibrated frequency. Forced classification: the mass the evidence cannot assign is spread across the hypotheses the case does support.