Aguadilla’s infrared recording became a dispute about how to reconstruct a small target from a moving camera. Investigators disagreed about its distance, speed, apparent division and relationship to the sea. AARO later assessed ordinary wind-driven objects, while the Scientific Coalition for UAP Studies questioned that reconstruction. The disagreement is documented; the extraordinary performance claimed for the target is not an agreed observation.
Map location. The pin is an approximate Rafael Hernández Airport area reference, rounded from GeoNames’ airport record. It does not identify a measured target position, a landing point or an invitation to enter restricted airport grounds. The recording and proposed paths span the surrounding area.
The recording and the first published analysis
The SCU report first released in August 2015 dates the encounter to about 21:20 local time on 25 April 2013. Its account concerns a U.S. Customs and Border Protection DHC-8 aircraft near Rafael Hernández Airport, Puerto Rico. A secondary source supplied the video and described what the crew had experienced; the report distinguishes that intermediary from the primary witnesses. Its authors say they checked the source’s background, but the individual remains anonymous in the public document.
The authors interpreted the thermal sequence as a small object moving rapidly, apparently entering water and later dividing. They used radar data to corroborate the recording aircraft’s movements and discussed earlier unknown radar returns nearby. Their suggestion that those returns and the infrared target belonged to one event is an inference. Matching the aircraft’s telemetry does not independently establish the target’s identity, and earlier radar returns are not automatically a continuous track of the object in the video.
Why the sea in the picture became important
In Rubén Lianza’s 2017 analysis, the proposed explanation is two wind-driven, heart-shaped sky lanterns. He argues that camera movement and zoom can leave water filling the frame while a nearer object remains over land. His proposed range depends on an assumed lantern size and selected image measurements. That is a testable model, rather than a directly measured distance to a recovered object.
Lianza also investigated a possible source upwind. His paper reproduces a resort manager’s reply that lanterns had been released there in the past, but that the exact date could not be confirmed. This supports the existence of a local practice, not a demonstrated launch on the encounter night. The distinction matters: identifying somewhere capable of releasing lanterns does not establish that those particular objects originated there. Nor does a plausible shape comparison establish every detail of the proposed flight.
SCU’s objections to the lantern interpretation
On its report and alternative-hypotheses page, SCU rejects Lianza’s interpretation. It argues that a wind-driven object would require at least 16 mph to fit its lines of sight, questions whether lanterns could remain airborne in those conditions, and disputes how that proposal explains apparent occlusions and disappearance. It also argues that two lanterns should have separated visibly earlier in the sequence. These are the investigators’ stated objections, not an independent finding that all lantern models are impossible.
The page’s publication and update dates should not be confused with the encounter or the original report release. A later web presentation can collect earlier material without becoming a new sighting. Its value here is to preserve the authors’ position and provide their source files, allowing the competing calculations to be compared rather than reducing the dispute to rival descriptions of a strange-looking clip.
A separate balloon response makes the geometry explicit
The linked six-page Balloon Alternative Hypothesis: SCU Response constructs successive camera lines of sight six seconds apart. The authors ask whether an object drifting west with the reported wind could occupy positions on all those lines. Their illustrated example requires 567 feet of travel between two positions in six seconds, equivalent to roughly 64 mph. They consider that inconsistent with a simple drifting-balloon explanation.
That result belongs to their selected reconstruction. It should not be presented as a speedometer reading from the target. The response also invokes opinions from unnamed technicians and engineers, but these are reported consultations rather than separately published measurements available for examination. The useful part of the argument is the set of positions, intervals and constraints that another analyst could attempt to reproduce or challenge. Disagreement about which positions are justified is therefore central to the case.
AARO’s 2025 resolution and its confidence levels
AARO’s 20 March 2025 resolution assesses with high confidence that the objects did not demonstrate anomalous performance or enter the water. Its reconstruction places two objects over land, moving at about 8 mph, and attributes apparent speed to motion parallax. AARO gives the more specific identification as sky lanterns only moderate confidence. Those confidence levels are different claims and should remain distinct.
The report discusses changing view angle, range, clouds and thermal contrast, and identifies apparent separations within the first minute. It dates the case 26 April 2013, whereas SCU uses 25 April locally; that difference is retained here rather than silently harmonised. The official PDF host was inaccessible during this research, so the seven-page document was read through an archival copy of the report. That is a copy of AARO’s analysis, not a new independent investigation.
The questions SCU published afterwards
In its 28 April 2025 statement, SCU said it had asked AARO for the reconstructed three-dimensional path, frame-linked coordinates and error margins, the basis for the cloud statement, relevant tower or radar records, and the methodology behind a utility-pole pixel analysis. SCU reported that AARO had acknowledged the inquiry but was not then ready to answer, and said its own review remained incomplete.
The statement documents the position at that date. It is not evidence that no later exchange could occur, or that asking questions invalidates the official reconstruction. Equally, an agency’s resolution does not make the reproducibility questions disappear. The appropriate comparison concerns the underlying measurements, assumptions and uncertainty, rather than the institutional title of either participant.
What the record can support
There are several different evidential steps: obtaining a recording, locating its camera, interpreting the target’s image, estimating its range and deciding what caused it. Success at one step does not automatically settle the others. A camera can record something real while depth, motion and disappearance are misinterpreted. A proposed ordinary explanation can also remain less certain in its specific identification than in its rejection of extraordinary performance.
This case file compares the published analyses; it does not claim to have reproduced their frame measurements, radar processing or flight reconstruction. AARO’s resolution is the official assessment described here, while SCU’s objection and request for further data are attributed positions. The central unresolved disagreement between those published positions concerns geometry and sensor interpretation. It is not, by itself, positive evidence for an exotic craft, and the alleged transition through water should not be narrated as an established capability.
For a different dispute over infrared imagery, compare the Chilean Navy’s 2014 video. Explore UFO / UAP / USO and the worldwide Cases directory for other source-led investigations.