Scientists Find Critical Data Gaps in 112 Pentagon PURSUE Program UAP Videos, Urge Release of Sensor Range Information

Researchers Jacob Haqq-Misra (Blue Marble Space) and Ravi Kopparapu (NASA Goddard Space Flight Center) published a preprint paper on August 20, 2026 analyzing 112 UAP videos released by the U.S. Department of War under the Presidential Unsealing and Reporting System for UAP Encounters (PURSUE). The PURSUE releases began on May 8, 2026, and are publicly available on the Department of War website. The paper, "Limits on Velocity Recovery from the PURSUE Sensor Videos," was posted to the arXiv preprint server and had not yet undergone peer review at the time of publication.

Both authors serve on the UAP Science Advisory Council to the U.S. government. In statements to The Debrief, Kopparapu clarified that his views are his own and do not represent the position of NASA. The videos analyzed were acquired by targeting pods and turret-mounted camera systems aboard military aircraft or similar government platforms, and consist primarily of infrared and electro-optical imagery showing unidentified objects crossing or being actively tracked within the camera's field of view.

The central finding of the study is that apparent object motion — measured in pixels per frame — cannot be converted to a real-world velocity without several additional data points that are either missing or redacted in the current PURSUE releases. Kopparapu identified range (the distance between the object and the observing platform, ideally tracked over time) as "perhaps the single most important missing quantity." He noted that without range data, an object moving rapidly across an image could be either a small nearby object moving slowly or a large distant object moving at extreme speed. Additional missing data identified by the researchers includes camera field-of-view specifications, observing platform motion, and viewing geometry.

The paper includes a case study of PURSUE video DOW-UAP-PR113, an infrared video reportedly recorded over the western United States in 1996. The footage shows a compact dark object — or possibly two objects in close proximity — crossing the field of view at approximately 142 pixels per frame, equivalent to roughly 2.2 frame widths per second, remaining visible for approximately one-tenth of a second. The researchers noted that embedded imagery markings allowed a limited reconstruction of angular scale and horizontal field of view, but that the available data was still insufficient to draw conclusions about the object's actual motion or physical nature.

The authors use the well-known Navy GOFAST video as a comparative case study in transparency. They note that in the GOFAST case, sufficient telemetry was available to determine that the object's apparent speed largely resulted from parallax caused by the observing aircraft, and that the object's motion was consistent with wind-borne drift — most likely a balloon. They contrast this with the current PURSUE releases, where equivalent telemetry is absent. An April 2020 Pentagon statement on the GOFAST release — "After a thorough review, the department has determined that the authorized release of these unclassified videos does not reveal any sensitive capabilities or systems" — is cited by the researchers as a precedent suggesting that more complete data could, in at least some cases, be safely disclosed.

Haqq-Misra told The Debrief that the UAP Science Advisory Council is already working to recommend what additional data future PURSUE releases might include, and described an ideal future UAP observatory as "a multi-modal and multi-site observatory, leveraging video observations from infrared to visible and including other modalities such as radio, radar, and audio data collection," citing the Galileo Project at Harvard as an existing privately funded effort following this design. The researchers concluded that "the PURSUE sensor videos in their present form cannot fully resolve unidentified cases or conclusively indicate anomalous velocities," while characterizing their analysis as a meaningful early step toward identifying strategies for obtaining better data.