Aguadilla CBP FLIR Encounter

Theoretical Alignment

This case aligns most strongly with the Psycho-Social Hypothesis, not because any witness fabricated anything, but because the video's decade-long life as "proof" of a transmedium craft depended on a chain of ambiguous, low-resolution sensor data being read through a pre-existing expectation of anomalous behavior — an expectation that a much more mundane, locally routine event (hotels releasing sky lanterns during celebrations) was published by independent analysts from 2015 onward but drew little attention until AARO's 2025 analysis. The Extraterrestrial Hypothesis retains a modest alignment because the footage originates from a genuine government sensor platform rather than a witness's memory, and because the Scientific Coalition for UAP Studies' (SCU) own quantitative 2015 analysis — whatever its flaws — was a serious, methodologically transparent attempt to characterize the object's motion rather than a bare assertion. The Interdimensional Hypothesis finds essentially no purchase here: nothing in either the original footage or either technical analysis describes a portal, dimensional transition, or comparable phenomenon.

Background

By 2013, Puerto Rico already had a decades-long reputation within American ufology as one of the most consistently active UAP "hotspots" in U.S. territory, a reputation built substantially on the reporting of Jorge Martín, editor of the Spanish-language magazine Evidencia Ovni and a chronicler of a wave of sightings around Puerto Rico's military installations and coastline dating back to the late 1980s, including the Amaury Rivera abduction case and a much older, unverified 1963 claim of a submerged sonar contact off the island's Atlantic Submarine Warfare range. Against that backdrop, U.S. Customs and Border Protection (CBP) — the Department of Homeland Security component responsible for patrolling U.S. borders and coastlines, including by air — routinely flew fixed-wing patrol aircraft fitted with forward-looking infrared (FLIR) sensor turrets along the island's coast as part of ordinary counter-narcotics and border-security operations, unrelated to any UAP-reporting mission.

Aguadilla itself sits on Puerto Rico's far northwestern corner, a municipality built around the site of the former Ramey Air Force Base, a Strategic Air Command installation that closed in 1973 and whose long runway was converted into the civilian/joint-use Rafael Hernández Airport (BQN). The town's economy leans heavily on tourism and a beachfront resort corridor facing the Atlantic — a detail that would become directly relevant to AARO's eventual technical explanation more than a decade later, since local hotels' routine practice of releasing sky lanterns during guest celebrations turned out to be central to the case's resolution. On the night in question, a CBP De Havilland Canada DHC-8 ("Dash 8") twin-turboprop patrol aircraft, part of CBP Air and Marine Operations' fixed-wing fleet and equipped with an infrared/electro-optical sensor turret (independently identified by SCU's later analysis as a Wescam MX-15D), was operating in the vicinity of the airport when its sensor operator noticed and began tracking a faint thermal signature.

Complete Timeline

Date	Event
~6:50 PM AST, Apr 25, 2013	Local sunset at Aguadilla, per NASA JPL Horizons. AARO's 2025 report gives 7:48 PM and uses that time as the reference point for the "thermal crossover" window it cites.
~8:58–9:14 PM AST, Apr 25, 2013	Per SCU's later FOIA request to the U.S. Air Force 84th Radar Evaluation Squadron, several unidentified radar returns are logged just offshore, north/northwest of Rafael Hernández Airport — 50 primary returns with no transponder in 16 minutes, which SCU judged "too coincidental" not to be the object later filmed.
~9:16 PM AST, Apr 25, 2013	Per SCU's FOIA-obtained flight records, the CBP DHC-8 takes off from Rafael Hernández Airport.
~9:22 PM AST, Apr 25, 2013 (01:22 UTC, Apr 26)	The aircraft's infrared sensor begins recording two small thermal targets near the airport's northeastern side, at an AARO-assessed altitude of roughly 200 m (656 ft) — the start of the surviving ~3 minute 54 second video.
~9:22–9:23 PM AST, Apr 25, 2013	Per AARO's frame-by-frame reconstruction, the two objects visibly separate from one another at least three distinct times within the first minute of the recording (at approximately 00:29.56, 00:40.76, and 00:47.00 into the clip).
Continuing through ~9:25 PM AST, Apr 25, 2013	The CBP aircraft flies an arc around the airport, climbing roughly 1,725 ft in altitude while the objects drift southwest toward the coast at approximately wind speed; the sensor's viewing angle shifts from side-on to increasingly top-down.
End of recording, ~9:26 PM AST, Apr 25, 2013	The objects' thermal signatures fade into indistinguishability against the Atlantic's own thermal background after the aircraft has passed 3,000 ft into a layer of scattered cloud; AARO places the loss of sensor contact, at 3,600 ft, over the airport's central parking apron.
Later on Apr 25, 2013	Per SCU's research, the CBP aircraft, having circled the airport, continues its patrol.
2013–2014 (exact date undetermined)	Copies of the video circulate on YouTube and Facebook; Puerto Rican researcher Jorge Martín features it in his ENIGMAS del Milenio magazine. (An account that he received it first, from a source in airport security, could not be confirmed.)
October 20–21, 2013	Separately, an acquaintance of the CBP pilot gives an original copy of the video file to MUFON Florida volunteer Daina Chaviano, who brings it to Florida MUFON state director Morgan Beall; Beall assembles a small technical review team that becomes the core of the Scientific Coalition for UAP Studies (SCU) investigation.
September 2014	Black Vault founder John Greenewald Jr. receives an anonymous letter enclosing a high-resolution copy of the video, which he posts publicly.
August 2015	SCU publishes a 162-page technical report (first released August 10, 2015), led by Robert Powell, concluding after at least 1,000 research hours that no known aircraft, balloon, bird, or drone accounted for the object's estimated speed, apparent transmedium travel, and splitting behavior.
November 19, 2024	AARO Director Dr. Jon Kosloski previews the office's findings in testimony before the Senate Armed Services Subcommittee on Emerging Threats and Capabilities, describing the objects as "likely a pair of balloons or sky lanterns" and attributing the apparent water entry to a loss of thermal contrast.
March 20, 2025	AARO formally releases its seven-page Case Resolution report, "The Puerto Rico Object" (DoD clearance no. 25-P-0553), assessing with high confidence that the objects showed no anomalous behavior and, with moderate confidence, that they were a pair of sky lanterns.
April 2025	SCU posts a public response neither endorsing nor rejecting AARO's sky-lantern conclusion, and submits five specific technical questions to AARO regarding its flight-path modeling, cloud-layer sourcing, and pixel-analysis methodology; AARO acknowledges the inquiry but states it is not yet prepared to respond in detail.

The Recording: What the CBP Sensor Actually Captured

The surviving video runs just under four minutes, of which, by SCU's count, 2 minutes 56 seconds show the object. Displayed in grayscale mid-wave infrared with a standard military sensor overlay — date/time, ground track, sensor mode, targeting reticle, and aircraft/target geographic coordinates burned directly into the frame — the footage shows a small, indistinct thermal signature crossing what is recognizably an airport environment before drifting out toward open water. For most of its first minute, the object reads as a single blob-like signature; as the recording continues and the sensor's zoom and viewing angle shift, the signature repeatedly appears to separate into two distinct, smaller signatures, then seems to rejoin, and finally appears to vanish as the background transitions from land to ocean. It was this final sequence — two objects, having apparently split from one, disappearing over the water — that gave the video its enduring popular framing as evidence of a "transmedium" craft capable of operating in both air and sea.

Two features of the raw footage deserve emphasis before considering either camp's analysis. First, the video's resolution and thermal contrast are genuinely poor by the standards of more famous instrumented UAP cases such as the Navy's 2004 "Tic Tac" encounter or the 2015 Gimbal and GoFast videos — a Dash 8 border-patrol turret is a substantially less capable sensor package than a fighter jet's targeting pod, and the object itself never resolves into any identifiable shape at any point in the recording. Second, and critically, the sensor's own onscreen telemetry — the same ground-track and coordinate overlay visible in the still frame above — is precisely the raw data both SCU's 2015 analysis and AARO's 2025 reconstruction used to try to establish the object's real-world position, altitude, and speed; the dispute between the two organizations is not about what is visible on screen, which neither side disputes, but about how to correctly convert that screen data into a real flight path.

From Leak to Public Record: How the Video Became Public

Unlike most instrumented military UAP footage released through an official declassification process, the Aguadilla video reached the public informally, through leaks that only later converged. Copies circulated on YouTube and Facebook, and Puerto Rican researcher Jorge Martín featured it in his magazine ENIGMAS del Milenio; this file has not been able to confirm the account, repeated elsewhere on this site, that Martín received it first from a source inside the airport's security staff. The best-documented chain is SCU's own: according to its 2015 report, an acquaintance of the CBP pilot (the report's "Witness B") gave an original copy of the video file on October 21, 2013 to an investigator known to the team, who was later joined on the report by Daina Chaviano, an established Cuban-born novelist; the source asked for anonymity, and the team says it verified the source's identity, occupation and background. Chaviano brought the tip to Morgan Beall, MUFON's Florida state director, who assembled a small team with relevant technical and scientific backgrounds specifically to vet the footage's authenticity before taking the investigation any further — the nucleus of what would become the Scientific Coalition for UAP Studies (SCU).

Another channel opened in September 2014, when Black Vault founder John Greenewald Jr. received an anonymous letter enclosing a high-resolution copy of the footage along with correspondence referencing the aircraft type and sensor specifications and asserting, in the sender's own words, that "alien technology is no doubt under the ocean near Puerto Rico." Greenewald, whose Black Vault archive holds millions of pages of FOIA-obtained government records, posted the high-resolution version publicly. These channels were not necessarily independent: SCU's report notes that an anonymous e-mail to Beall, a YouTube comment by a user called "Red Bill," and the letter to Greenewald were nearly identical in content and "may be from the same person." The question of authenticity was in any case settled in November 2024, when the Department of Defense itself released the footage through DVIDS, although CBP has still never publicly identified the crew.

The Scientific Coalition for UAP Studies Investigation (2013–2015)

Once Beall's team confirmed to its own satisfaction that the video's origin was genuine, the investigation that became SCU's formal report proceeded over about a year and a half and, per the group's own accounting, at least 1,000 combined research hours. Led principally by Robert Powell — a chemist by training with 28 years in semiconductor engineering management, who had previously co-authored MUFON's widely cited 2008 Stephenville Lights radar study — the team performed frame-by-frame analysis of the object's apparent motion, cross-referenced the sensor's onscreen coordinate overlay against the aircraft's own flight track, and filed Freedom of Information Act requests with the U.S. Air Force's 84th Radar Evaluation Squadron for radar data covering the same airspace and time window. The Air Force released FAA radar data (chiefly from the long-range QJQ site about 90 miles away) but denied a request for the military Punta Borinquen radar at the airport itself; an FAA request for the control-tower logs was refused because they are kept by a private contractor, and the tower manager said recordings are destroyed after 90 days.

The resulting 162-page report, released on August 10, 2015 via SCU's own website, reached conclusions dramatically different from AARO's decade-later resolution. SCU's frame-by-frame reconstruction estimated the object's in-air speed at roughly 40–120 mph (with a median around 80 mph) and its apparent underwater speed — based on SCU's own reading of the footage as showing genuine submersion — at up to 95 mph, averaging 82.8 mph, with what the report described as minimal loss of speed at the moment of water entry. The FOIA'd radar data showed 50 primary returns without a transponder, two to three miles north and northwest of the airport, between 8:58 and 9:14 PM local time; the CBP aircraft took off at 9:16 PM. SCU argued that it "would be too coincidental to think otherwise" than that these returns, the light the pilot reported, and the filmed object were the same thing, while conceding that the first few returns might be unrelated and that a temperature inversion had to be ruled out. AARO's report does not discuss the radar data at all.

SCU's report went on to systematically test, and reject, several conventional explanations on quantitative grounds: recorded regional wind speeds (roughly 8–18 mph depending on altitude) were, in the team's assessment, far too slow to account for the object's estimated velocity and its apparent directional changes, ruling out a simple drifting balloon; the fastest birds SCU considered in level flight, such as the peregrine falcon (maximum horizontal speed 65–68 mph), fell below the object's estimated speed; and no drone or torpedo system of the era known to the researchers matched the object's combined aerial and underwater performance (the U.S. Naval Research Laboratory's experimental "Flimmer" hybrid air/underwater prototype, the closest comparison SCU could identify, was rated for roughly 68 mph in air). The report's own concluding language was unambiguous: the investigators wrote that they found "no explanation for an object capable of traveling under water at over 90 mph with minimal impact as it enters the water, through the air at 120 mph at low altitude through a residential area without navigational lights, and finally to be capable of splitting into two separate objects."

A Decade in Limbo: The Case in Public Memory

Between SCU's 2015 report and AARO's 2025 resolution, the Aguadilla video settled into a durable place in UAP research culture as one of the more frequently cited pieces of "instrumented" evidence — distinct from purely testimonial cases because it came from an identifiable, if unofficial, government sensor rather than a witness's memory, and distinct from cases like Gimbal or GoFast because no U.S. military branch had, for most of that decade, formally acknowledged the footage at all. The case appeared regularly in UAP-focused television and web coverage (including outlets such as Openminds.tv, MysteryWire, and syndicated wire coverage carried by dozens of local U.S. television affiliates), was catalogued by researcher-run archives such as UFOevidence and Enigma Labs, and was periodically revisited by skeptical forums — most notably Mick West's Metabunk community, whose thread by the early 2020s carried the title "Probably Hot Air Wedding Lanterns." The wind-driven reading was older still: retired Commodore Rubén Lianza, head of the Argentine Air Force's aerospace-identification office, wrote a detailed analysis concluding the objects were a pair of heart-shaped wedding lanterns, and cited earlier wind-drift readings by Andrés Duarte (July 2015), Gilles Fernandez (August 2015) and Bob Bixler (January 2016). SCU publicly rejected the lantern hypothesis. SCU's own case page says the French group Sigma2 of the aerospace society 3AF gave an initial review and that a full review was expected; this case file's research found no indication that a full review was published.

This decade-long limbo is itself relevant to the case's Psycho-Social alignment, discussed above: for roughly ten years, the video's most widely repeated framing — extreme speed, genuine transmedium travel, an object splitting in two — rested largely on SCU's 2015 analysis, with no U.S. government counter-analysis in public circulation, even though a mundane explanation (sky lanterns, a common local hospitality-industry practice) had already been published by skeptics and by an Argentine Air Force analyst years before AARO adopted it as the office's leading hypothesis.

AARO Opens a File: The 2024 Senate Preview

The Puerto Rico case entered AARO's own review process as part of the office's broader mandate, created under the FY2022 National Defense Authorization Act and later codified provisions, to investigate historical and ongoing UAP reports across the U.S. government. On November 19, 2024, AARO Director Dr. Jon Kosloski, an electrical engineer (Ph.D., Johns Hopkins) who had joined AARO from the National Security Agency's Research Directorate, previewed the office's findings in open testimony before the Senate Armed Services Subcommittee on Emerging Threats and Capabilities, discussing the Puerto Rico case alongside AARO's ongoing review of the Navy's 2015 GoFast video and a widely circulated Mount Etna volcanic-eruption clip. Kosloski told the subcommittee the office assessed the Puerto Rico objects as "likely a pair of balloons or sky lanterns," drifting at roughly seven knots and descending to approximately 200 meters, and explained the video's most viral moment in plain terms: "the temperature of the water is equal to the temperature of the object and the camera can no longer distinguish between the two."

That November 2024 testimony was itself a significant milestone in the case's history — the first public U.S. government assessment of the Aguadilla footage this file has found, made the same day the Department of Defense released the video on DVIDS — and it previewed, in outline, the same two core findings AARO's formal case resolution would confirm four months later: that the objects' behavior was not anomalous, and that a specific, testable physical mechanism (rather than an appeal to insufficient data) explained the footage's most dramatic-seeming moment.

The Case Resolution Report: AARO's Technical Analysis

AARO's formal report, titled simply "The Puerto Rico Object," is a seven-page document dated March 20, 2025, carrying the DoD clearance number 25-P-0553. (It dates the encounter April 26, 2013, the UTC date on the video overlay; locally it was the evening of April 25.) It states AARO's assessment "with high confidence" that the objects "did not exhibit anomalous behavior or transmedium capabilities," and, separately and with a lower, "moderate" confidence, that the objects were a pair of sky lanterns. The report's reconstruction relied on Systems Toolkit (STK) orbital/flight modeling software combined with what AARO calls "Minimum Separation Vectors" analysis, integrating the CBP aircraft's own flight path with the sensor's elevation, azimuth, and slant-angle data to project the objects' real-world position onto a map at every point in the recording.

Reconstructed Speed and Motion Parallax

AARO's STK reconstruction placed the objects' actual speed at approximately 3.6 meters per second (8 mph), moving in a straight line over land — a figure the report notes is consistent with the period's recorded wind speed of 4.4 meters per second (9.8 mph) from the east/northeast. The report attributes the object's dramatic apparent high speed to motion parallax: an optical effect, well documented in aviation and photography, in which a slow-moving or stationary object appears to move much faster than its true speed when viewed from a moving platform, with the effect becoming more pronounced as the observing platform's own speed increases relative to the target. Because the CBP aircraft's own flight speed, its sensor's zoom setting, and the continuously changing relative geometry between aircraft and object were all in flux throughout the encounter, AARO's report argues the same underlying slow, wind-driven drift could plausibly have produced the appearance of a much faster-moving object purely as an artifact of the observing geometry — without requiring the object itself to have moved unusually fast at any point.

Two Objects, Not One Splitting in Two

AARO assesses with high confidence that the video depicts two distinct objects traveling near one another for the entire recording, rather than a single object that later split or replicated. The report's frame-by-frame reconstruction identifies at least three specific, timestamped moments within the video's first minute (approximately 00:29.56, 00:40.76, and 00:47.00) where the two objects visibly separate from each other — evidence, in AARO's reading, that two objects were present from the start rather than one object dividing partway through. The report further argues that the sensor's own viewing geometry changed meaningfully over the course of the recording: as the CBP aircraft climbed and its distance from the objects nearly tripled, the sensor's angle shifted from a low, side-on view (in which two closely spaced small objects would tend to visually merge into one blurred signature) to an increasingly steep, top-down view at higher magnification (in which the same two objects would appear more clearly and dramatically separated). AARO illustrates this directly with three comparison stills from the footage — roughly 00:05, 00:33, and 02:37 into the clip — showing the same pair of objects becoming progressively more visually distinct as the aircraft's altitude and the sensor's zoom increased, without the objects' own separation distance necessarily having changed at all.

The Apparent Water Entry: Thermal Crossover

The report's explanation for the footage's single most viral element — the objects appearing to enter, and disappear beneath, the ocean — rests on a specific, dated physical phenomenon: thermal crossover, in which an infrared sensor loses the ability to distinguish a target from its background once their respective thermal signatures become nearly identical, an effect most pronounced for roughly two hours following a rapid ambient-temperature change such as sunset. AARO's report states that local sunset occurred at 7:48 PM and that the IR sensor recorded the encounter at 9:22 PM, "within the two-hour window" in which crossover can persist. That sunset time appears to be wrong. NASA JPL Horizons puts sunset at Aguadilla that evening at about 6:50 PM AST (22:50 UTC), so the recording began about 2 hours 32 minutes after sunset, outside the two-hour window AARO cites. Crossover therefore cannot be presumed. Loss of contrast against the ocean is still a credible explanation, and AARO also cites two other factors: the range to the objects nearly tripled, and clouds at 3,000 feet partly obscured the view. Citing a peer-reviewed 2010 Optics Express study on measured MWIR/LWIR thermal crossover periods, the report concludes that as the ocean came into the background, and as the sensor's range to the targets nearly tripled, their thermal signature gradually lost contrast against the water until the sensor could no longer register them at all — not because they physically submerged, but because they became thermally indistinguishable from the sea surface. AARO's reconstructed flight path places the objects over land for the entire observed encounter and states plainly that "the objects did not enter the water."

Size, Attribution, and Local Corroboration

Using pixel analysis — comparing the objects' apparent size in frame against objects of known dimension at the same distance — AARO estimated each object at smaller than one meter (roughly three feet) across, with an indistinct overall shape. The report assesses with moderate confidence that the objects were sky lanterns, citing outreach AARO conducted with local hospitality-industry vendors, who confirmed it is common practice for Aguadilla-area hotels and resorts to release sky lanterns during guest celebrations. AARO notes that sky lanterns are typically smaller than one meter in diameter and produce a flickering, gradually weakening thermal signature as their fuel source burns down — a pattern the report states is consistent with the video's own thermal variability — while explicitly caveating that the footage's poor overall quality "reduces AARO's confidence in categorically identifying the objects."

Alternative Hypotheses AARO Considered and Rejected

The report documents three specific alternative hypotheses AARO weighed before settling on its sky-lantern assessment. The first was the anomalous-performance hypothesis itself: in certain frames, the objects appear to pass behind a utility pole visible in the video, which would imply an altitude and speed inconsistent with ordinary drifting objects; AARO's own pixel analysis of this specific claim found that the objects did not actually pass behind the pole, closing off what proponents of the anomalous reading had treated as one of the strongest visual arguments for extreme low-altitude speed. The second was a marine-bird hypothesis, proposed by one of AARO's own interagency partners, who estimated the objects traveled between 35 and 130 mph at an altitude of 300–900 feet, consistent with a pair of birds descending toward the ocean surface; AARO's report considers this "unlikely," noting that the STK reconstruction shows straight-line, wind-speed motion over land rather than a bird's more variable flight path, and that birds viewed at the relevant distance through an infrared sensor would be expected to show visible wing structure or a wingbeat-frequency pulsation in their thermal signature, neither of which appears in the footage. The third was a mylar or "party" balloon hypothesis, also proposed by an AARO partner, under which two balloons tied together and drifting apart and together would explain the visual pattern, with fluctuating brightness attributed to moonlight reflecting off the balloons' surface through intermittent cloud; AARO states it does "not concur" with this specific mechanism, reasoning that an infrared sensor is not physically capable of detecting reflected visible moonlight in the way the hypothesis would require.

Reception: SCU's Response and the Unanswered Questions

Unlike several other AARO case resolutions that drew immediate, sharply worded rebuttals from the researchers whose earlier work they superseded, SCU's public response to the March 2025 Puerto Rico resolution — posted to the organization's own website the following month — was notably measured. Rather than directly disputing the sky-lantern conclusion, SCU stated it "remain[s] committed to a thorough and evidence-based analysis" and would complete its own review once it received additional data, then submitted five specific technical questions to AARO: a request for the underlying 3D flight-path reconstruction data (coordinates and altitude, with stated error margins); the specific source and altitude documentation behind AARO's claimed 3,000-foot cloud layer; the full geometric analysis behind the flight-path reconstruction, again with error margins; any control-tower log or independent military radar data corroborating the encounter; and the detailed results of AARO's own pixel analysis regarding whether the objects passed behind the utility pole. According to reporting on the exchange, AARO acknowledged SCU's inquiry but stated it was "not currently prepared to provide answers" to the specific questions, though it might address them in a future follow-up report.

That exchange leaves the case in a genuinely unusual position within this catalogue: a real government sensor recording, a real and detailed independent technical counter-analysis, and a real and detailed official government rebuttal — but no completed point-by-point technical reconciliation between the two. AARO's report is considerably more granular and falsifiable than a typical dismissive explanation (it specifies exact separation timestamps, cites a peer-reviewed thermal-imaging paper, and directly tests and rejects the utility-pole altitude claim rather than ignoring it), which this case file's research treats as a genuine positive signal for AARO's account's technical seriousness. At the same time, SCU's specific, itemized requests for AARO's underlying coordinate data and error margins remain unanswered as of this case file's research, and wider coverage of the resolution — including a critical NewsNation segment questioning AARO's motives and a Black Vault article juxtaposing the office's sky-lantern conclusion against SCU's earlier findings — shows the case has not achieved the kind of broad, cross-camp consensus that followed some other AARO resolutions.

Key Witnesses

The Five Observables Assessment

The Five Observables framework, developed from the Pentagon's own Advanced Aerospace Threat Identification Program criteria, is applied here against both the footage's original popular reading and AARO's 2025 technical rebuttal:

Assessment: Neither analysis, AARO's or SCU's, ever attributed anti-gravity lift or literal instantaneous acceleration to the objects, so both are rated Absent regardless of which camp's speed estimate is correct. Hypersonic performance is likewise rated Absent, since even SCU's own considerably higher speed estimate (up to roughly 120 mph in air) falls far short of any hypersonic threshold. Low Observability is rated Speculative rather than Present or Absent because the object's fading thermal signature is the crux of the entire dispute: AARO attributes it to a well-documented physical sensor limitation (thermal crossover) rather than any intrinsic stealth property, while the footage's original popular reading treated the same fade as evidence the object was actively concealing itself. Trans-Medium travel receives the same Speculative rating for the identical reason — it is the single specific claim AARO's report was built to test, and while AARO's flight-path reconstruction places the objects over land throughout, SCU's own independent analysis and open technical questions to AARO on this exact point remain formally unresolved as of this case file's research.

Evidence Assessment

Why "Disputed"

This case earns a DISPUTED classification with a credibility score of 40/100, placing it just below Thomas Mantell (Disputed, 38) and the Bonilla Solar Transit Photographs (Disputed, 39) in this catalogue's lower "Disputed" tier, and well below cases with stronger multi-witness corroboration or a less complete conventional rebuttal, such as Gorman Dogfight (Disputed, 58) or the USS Supply/Schofield Report case (Disputed, 46).

On the strength side: this is a real, well-documented event, not a fabricated story — a genuine government sensor recording, later released by the Department of Defense itself, that received a serious, methodologically transparent, 1,000-plus-hour independent technical investigation (SCU's 2015 report) years before any government agency addressed it. That instrumented, non-testimonial evidentiary basis is a real positive signal this catalogue's scoring credits, comparable in kind to (if considerably less definitive than) other FLIR-based cases like Gimbal and GoFast.

Weighed against that: AARO's 2025 case resolution is an unusually strong, specific, and falsifiable rebuttal by the standards of this catalogue's own coverage of similar cases. It does not merely assert a mundane explanation; it reconstructs the object's real-world flight path using dedicated aerospace modeling software, matches the resulting speed to independently recorded wind data to within about 1.8 mph, directly tests and refutes the single strongest visual argument for extreme low-altitude speed (the utility-pole pass), documents the specific frame timestamps supporting a two-object rather than one-splitting-in-two reading, and grounds its water-entry explanation in a named, peer-reviewed sensor-physics phenomenon, though its timing argument uses a sunset time about an hour too late. That level of specificity is closer to Mantell's well-documented Skyhook balloon program (Disputed, 38) than to a vaguer, unresolved case — AARO's account, unlike a simple "probably conventional" dismissal, gives genuine, checkable reasons for each of the footage's most viral claims. Set against that strength, SCU's own April 2025 response is telling: rather than mounting a substantive point-by-point rebuttal of AARO's physics, the organization that has advocated for this case's anomalous reading longer and more thoroughly than anyone else asked AARO for more underlying data instead of directly disputing the sky-lantern hypothesis on the merits — a genuinely unresolved state of affairs, but one that reads as closer to "AARO's explanation has not yet been rebutted" than to "AARO's explanation has been shown to be wrong." A case with this much real instrumented evidence, this much serious independent scrutiny on both sides, and this specific and technically substantial — if still formally unreconciled — an official rebuttal sits, in this catalogue's judgment, in the lower half of the Disputed tier rather than in Credible territory.

Sensor & Instrumentation Detection Profile

The case's entire evidentiary record derives from a single sensor: the infrared/electro-optical turret (identified by SCU's independent research as a Wescam MX-15D) mounted on the CBP DHC-8's undercarriage, operating in mid-wave infrared (MWIR) mode with an onscreen telemetry overlay giving date, time, ground track, sensor mode, and geographic coordinates for both the aircraft and its tracked target. No independent camera or instrumented ground observer recorded the event. The one candidate secondary data source, the FAA radar data SCU obtained from the 84th Radar Evaluation Squadron via FOIA, recorded unidentified returns nearby in the 16 minutes before the CBP aircraft took off; SCU argued they were the same object, but the returns stop before the filming begins and cannot be directly matched to it. AARO's own report does not cite this radar data at all, relying exclusively on its STK reconstruction of the sensor's own onboard telemetry. This single-sensor dependency is itself a meaningful limitation this catalogue's scoring weighs: every quantitative claim on both sides of the dispute, from SCU's underwater-speed estimate to AARO's wind-matched drift-speed reconstruction, is ultimately derived from the same low-resolution, degrading-quality video rather than from multiple independently cross-checked instruments.

Environmental & Geospatial Context

Aguadilla occupies Puerto Rico's northwestern corner, where the island's mountainous interior gives way to a coastal plain fronting the Atlantic and the Mona Passage. Rafael Hernández Airport sits on the site of the former Ramey Air Force Base, its long Strategic Air Command-era runway oriented roughly northeast-southwest along the coastline, with the area's beachfront resort corridor — the source, per AARO, of the sky lanterns most likely responsible for the sighting — situated immediately to the airport's south and west. AARO's report documents specific, dated environmental conditions directly relevant to its technical conclusions: a sunset time of 7:48 PM (which NASA JPL Horizons puts at about 6:50 PM AST), a recorded surface wind of 4.4 meters per second (9.8 mph) from the east/northeast, and scattered cloud cover beginning around 3,000 feet that partially obscured the CBP sensor's view as the aircraft climbed through that altitude late in the encounter. Apart from the sunset check, these figures are drawn directly from AARO's report, which cites timeanddate.com and sunrisesunset.com. The Moon was full (99.9% illuminated) and about 30° up in the east-southeast at 9:22 PM, per Horizons, which bears on the AARO partner's moonlit-balloon hypothesis. No unusual solar, geomagnetic, or seismic activity for the date was identified in any source consulted for this case file, and nothing in the record suggests any environmental factor beyond routine coastal Caribbean evening weather and a locally common tourism-industry practice.

Observer Credibility & Occupational Profile

This case presents an unusual credibility profile: its central evidentiary asset is not eyewitness testimony but a government sensor recording, yet the human observers involved — an unidentified CBP flight crew and unidentified airport tower personnel — have never been named, interviewed on the record, or their reported reactions independently confirmed beyond a secondhand account relayed through the leak chain. That absence cuts in a specific direction for this catalogue's scoring: it removes the possibility of directly assessing witness training, memory reliability, or motive, but it also removes any possibility that the case rests on a single unreliable narrator's account, since the sensor recording itself — whatever it does or does not show — exists independent of anyone's testimony about it. The investigators who have driven the case's public analysis carry stronger, independently verifiable credentials than the anonymous original observers: Robert Powell brought a documented radar-data-analysis background from MUFON's Stephenville Lights study to SCU's 2015 report, while Jon Kosloski brought a doctorate-level optics and signals background from the National Security Agency to AARO's 2025 resolution. This case file's research treats both organizations' technical leadership as credible and serious, which is precisely what makes the two camps' unreconciled conclusions a genuine, still-open dispute rather than one credentialed analysis against an amateur one.

Physical & Material Assays

No physical evidence of any kind exists for this case. Nothing was recovered from land or sea, no debris or ground trace was ever reported, and the entire evidentiary record consists of a single digital video file (in varying resolutions across its several leaked and re-released copies) plus a disputed and never-conclusively-correlated radar log. No laboratory sample, metallurgical assay, or trace residue was ever collected, nor could one have been, given that the case involves an airborne sensor recording with no landing, crash, or debris-recovery component of any kind.

Historical Precedents & Archive Matches

Within this catalogue's own coverage of Puerto Rico-linked cases, the Aguadilla video sits alongside the Amaury Rivera abduction and jet-pursuit case (1988, Disputed) and the considerably older, far more thinly sourced Puerto Rico ASW Range Sonar Contact (1963, Disputed, 22) as part of a decades-long pattern of the island being treated as an unusually active UAP-reporting hotspot, a reputation substantially built and sustained by Jorge Martín's own long career documenting local sightings. Set against those two, the Aguadilla case is structurally distinct: it is the only one of the three built on an instrumented government sensor recording rather than witness testimony, and the only one that has since received a detailed, technically specific official government resolution rather than remaining permanently unaddressed.

More broadly, this case invites direct comparison to other FLIR-derived U.S. government UAP videos this catalogue has covered, particularly the Navy's 2015 Gimbal and GoFast footage. Both share the same basic evidentiary shape — ambiguous, low-resolution infrared video from an operational military or law-enforcement sensor, initially read as showing extreme performance, later subjected to a specific technical rebuttal addressing sensor-geometry artifacts (motion parallax and glare/parallax effects, respectively). This case file's own comparative scoring places Aguadilla somewhat lower than Gimbal/GoFast's own Disputed rating, reflecting that AARO's Puerto Rico resolution is considerably more complete and specific — a named object type, a matched wind-speed correlation, and a directly tested and rejected anomalous-altitude claim — than the U.S. Navy's own, comparatively less detailed public accounting of the Gimbal/GoFast encounters.

Material Analysis

No physical object, debris, or trace sample of any kind is connected to this case at any point in its documentary history. "Material" here refers instead to the case's own surviving evidentiary substrate: the video file itself, in its several circulating resolutions and re-encodings, and the disputed 84th Radar Evaluation Squadron log SCU obtained via FOIA. Because the video is the case's sole primary artifact, both SCU's 2015 analysis and AARO's 2025 analysis are, in a meaningful sense, competing interpretations of identical raw data rather than the product of independently gathered evidence — a genuinely unusual situation for a "Material Analysis" section in this catalogue, where such sections typically address recovered physical objects rather than a shared piece of digital media.

The Conventional Explanation Candidate (CEC): Sky Lanterns and Motion Parallax

Because AARO conducted a genuine, resourced technical investigation of this case, much of the conventional-explanation work has already been done by the office itself, using tools (STK flight modeling, pixel analysis, cited sensor-physics literature) well beyond what open-source research alone could reproduce. What follows evaluates AARO's own leading candidate and the specific alternatives the office itself considered and set aside, alongside the possibility — kept open by SCU's own unanswered technical questions — that the case remains genuinely unresolved.

Sky Lanterns (AARO's Leading Candidate)

AARO's own moderate-confidence identification is well-supported by its wind-speed correlation, its consistent sub-one-meter size estimate, and confirmed local hospitality-industry practice, but the office's own report is careful to flag that the footage's poor quality limits how confidently any specific object type can be named. This case file rates the sky-lantern hypothesis as the single best-supported specific candidate in the record, while noting that "moderate confidence" in the identification is a meaningfully weaker claim than AARO's separate "high confidence" that the objects simply were not anomalous — the office is considerably more certain about what the objects were not doing than about precisely what they were.

A caveat on trusting AARO's own investigative work: a resourced, technically detailed government investigation is genuinely stronger evidence than an outside guess, but it is not immune to the same institutional pressures this site notes across every official UAP finding it cites — a well-publicized SCU-championed case carries real reputational stakes for AARO's own credibility, and the office's careful hedging between "moderate confidence" on the specific object and "high confidence" on the absence of anomalous behavior is itself worth reading closely rather than collapsing into a single "AARO solved it" headline. This file credits the sky-lantern identification because its specific, checkable methodology (wind-speed correlation, pixel size, local practice) holds up on independent review, not because AARO is the office of record.

Marine Birds

Proposed by one of AARO's own interagency partners rather than by an outside skeptic, this hypothesis was directly tested and set aside within AARO's own report: the STK-reconstructed straight-line, wind-speed flight path does not match typical bird flight behavior, and the absence of any visible wing structure or wingbeat-pulsation pattern in the thermal signature is a specific, checkable point against it. This case file finds AARO's own stated reasoning for rejecting this candidate persuasive and treats it as a seriously considered, not merely token, alternative.

Mylar / Party Balloons

Also proposed by an AARO partner rather than an outside party, this hypothesis would have needed reflected moonlight through intermittent cloud to explain the objects' brightness fluctuation — a mechanism AARO itself states it does not accept, since an infrared sensor is not built to register reflected visible light in that way. This case file agrees the mechanism as proposed is not physically coherent for an IR sensor, making this the weakest of the three specific alternatives on record.

Genuine Anomalous Phenomenon (SCU's Original Reading)

SCU's 2015 analysis remains the most detailed technical case for treating the footage as showing genuine extreme performance, but it has not, as of this case file's research, been reconciled point-by-point against AARO's later, independently wind-matched reconstruction of the identical raw video, and SCU's own 2025 response to AARO notably stopped short of directly disputing the sky-lantern conclusion on technical grounds, instead requesting further underlying data. Absent that reconciliation, this case file treats the anomalous reading as a serious but currently unconfirmed position rather than either a live, actively-defended counter-finding or a withdrawn one.

Related Cases

Further Reading — Recommended Literature

DISCLOSURE: Some links above are affiliate links. If you purchase through them, OverClassified receives a small commission at no additional cost to you. This helps support independent research.

Essential Viewing

Sources

• All-domain Anomaly Resolution Office (AARO), U.S. Department of Defense, "The Puerto Rico Object" — Case Resolution Report, 20 March 2025 (7 pp.; DoD clearance no. 25-P-0553). aaro.mil
• AARO, UAP Case Resolution Reports (official index page). aaro.mil
• AARO, Dr. Jon T. Kosloski — Official Biography. aaro.mil
• Dr. Jon Kosloski, Director, AARO, Statement for the Record, Senate Armed Services Subcommittee on Emerging Threats and Capabilities, Nov. 19, 2024. armed-services.senate.gov
• U.S. Department of Defense, "Puerto Rico Objects" (official infrared video release), Defense Visual Information Distribution Service. dvidshub.net
• Brad D. Williams, "Pentagon's UAP office reviews findings on Go Fast, Puerto Rico, Mt. Etna incidents in Senate hearing," DefenseScoop, Nov. 19, 2024. defensescoop.com
• Scientific Coalition for UAP Studies, "SCU Announcement on the AARO Investigation of the Aguadilla Video," Apr. 2025. explorescu.org
• Scientific Coalition for UAP Studies, "2013 Aguadilla Puerto Rico UAP Incident Report: A Detailed Analysis," Aug. 2015. explorescu.org
• John Greenewald Jr., "DoD Report Says Aguadilla UFO Was Just Sky Lanterns; Previous Scientific Studies Claim Otherwise," The Black Vault, 2025. theblackvault.com
• "Leaked UFO Video Captured by Homeland Security Analyzed," Openminds.tv, 2015. openminds.tv
• "Leaked Homeland Security Aguadilla UFO Video — No Explanation For 'Anomalous Phenomenon,' Says 161-Page Report," The Inquisitr, 2015. inquisitr.com
• "In the Air, in the Water, Splitting in Two — Scientifically Investigating the Aguadilla UFO Incident," MysteryWire (George Knapp). mysterywire.com
• M. Felton, K.P. Gurton, J.L. Pezzaniti, D.B. Chenault, L.E. Roth, "Measured comparison of the crossover periods for mid- and long-wave IR (MWIR and LWIR) polarimetric and conventional thermal imagery," Optics Express, Vol. 18, No. 15 (2010), pp. 15704–15713 — cited directly within AARO's case resolution report as the technical basis for its thermal-crossover explanation.
• United States Senate Committee on Armed Services, Hearing on All-Domain Anomaly Resolution Office Update, Nov. 19, 2024 (testimony transcript and case slides).
• John Greenewald Jr., Inside the Black Vault: The Government's UFO Secrets Revealed, Rowman & Littlefield, 2019.
• Robert Powell, UFOs: A Scientist Explains What We Know (And Don't Know), Prometheus Books, 2024.
• Robert Powell and Glen Schulze, "Stephenville Lights: A Comprehensive Radar and Witness Report Study," Mutual UFO Network, July 2008 (background on SCU lead investigator's prior methodology).
• Wikipedia, "John Greenewald Jr." (biographical orientation only; primary facts independently corroborated via The Black Vault's own published materials). en.wikipedia.org
• Mick West, Escaping the Rabbit Hole: How to Debunk Conspiracy Theories Using Facts, Logic, and Respect, Skyhorse Publishing, 2018 (Metabunk forum's independent open-source analysis of this case).
• Sarah Scoles, They Are Already Here: UFO Culture and Why We See Saucers, Pegasus Books, 2020.
• Darcy Weir, interviewed in "Why AARO Might Be Trying to Debunk Puerto Rico's Aguadilla Object," Reality Check, NewsNation, 2025.
• U.S. Air Force 84th Radar Evaluation Squadron radar data, obtained via Freedom of Information Act request by the Scientific Coalition for UAP Studies, cited within SCU's 2015 report.
• Rubén Lianza, "Alleged UFO Captured by Infrared Camera from U.S. Customs & Border Protection Airplane over Aguadilla, Puerto Rico" (24 pp., c. 2019–2020). ipaco.fr
• Scientific Coalition for UFOlogy, 2013 Aguadilla Puerto Rico: The detailed analysis of an Unidentified Anomalous Phenomenon captured by the Department of Homeland Security (162-page PDF, 2015). docs.wixstatic.com
• The Black Vault, "Anonymous Letter CONFIRMS Aguadilla, Puerto Rico, Coast Guard UFO Video" (letter received September 2014). theblackvault.com
• NASA JPL Horizons, observer ephemerides for the Sun (10) and Moon (301), 18.4949°N 67.1294°W, 25–26 April 2013 UTC (refracted). ssd.jpl.nasa.gov

Corrections (October 2026)

An October 2026 fact-check compared this page with the full text of AARO's seven-page case resolution, SCU's 2015 report, Kosloski's November 2024 testimony as reported by DefenseScoop, SCU's April 28, 2025 statement, The Black Vault's case file, Rubén Lianza's analysis and NASA JPL Horizons. The page had reversed SCU's position on the radar data: SCU did not "acknowledge real uncertainty" about the unidentified returns but argued it "would be too coincidental" for them to be anything other than the filmed object. AARO's 7:48 PM sunset, which the page repeated as the basis for its "squarely within" the two-hour thermal-crossover window, is about an hour too late; Horizons gives about 6:50 PM, so the recording began about 2.5 hours after sunset. Hearing, report and provenance details were also corrected. Kosloski testified to the Emerging Threats and Capabilities subcommittee, not the full committee. The report runs seven pages, not one. The "Case Number 330UAP000010," the March 19 clearance date and the "FY24 NDAA §1841" release marking could not be found and were removed. SCU's own report suggests the anonymous leak channels "may be from the same person," so they were not three independent corroborations, and lantern and wind-drift explanations were published from 2015 onward, not first by Metabunk in the 2020s.

Following this review, the rating was changed from Disputed 40 to Disputed 37 (October 28, 2026), because the higher rating had rested partly on claims corrected or removed above.

Softened: the claim that Jorge Martín received the video first, from a source in airport security, is now marked unconfirmed. Smaller fixes: the object was visible for 2 min 56 s (not about 2.5 min); the report is 162 pages, released August 10, 2015; Powell is a chemist by training; Kosloski is an electrical engineer; the peregrine figure is SCU's maximum horizontal speed, not a dive; the pilot's light was northwest of the airport moving toward it, not "approaching from the south"; sensor contact was lost at 3,600 ft; the Moon was full, per Horizons.