1957 RB-47 radar-visual UFO incident

Case dossier

Mission and crew. The RB-47H belonged to the 55th Strategic Reconnaissance Wing and was flying a composite training mission from Forbes Air Force Base. The six officers were aircraft commander Lewis D. Chase, copilot James H. McCoid, navigator Thomas H. Hanley and electronic-warfare officers John J. Provenzano, Frank B. McClure and Walter A. Tuchscherer. Their roles matter because the visual and electronic reports came from different stations aboard the same aircraft, not from a single observer retelling one impression.

First electronic detection. Near the Meridian-Gulfport area, McClure's number-two monitor received a radar-like signal around 2,995–3,000 MHz, with a roughly two-microsecond pulse, about 600 pulses per second and an apparent four-rpm sweep. The equipment was a passive direction-finding receiver: it listened for an emitter and displayed relative bearing, but it did not measure a target's range or produce a conventional radar echo. That distinction limits what the signal alone can prove.

First visual report. At 1010Z near Winnsboro, Louisiana, Chase and McCoid reported an intense white light with a blue tint crossing from about the aircraft's eleven-o'clock position to roughly two-thirty before vanishing. After Chase alerted the electronic-warfare crew, the receivers found a strong signal in approximately the same relative sector. The written record supports a reported temporal and directional correlation; it does not establish that the light and radio-frequency emission necessarily came from one object.

Texas phase and ground radar. Farther west, the crew reported a red light ahead while the passive receiver logged changing bearings. Chase contacted the 745th Aircraft Control and Warning Squadron at Duncanville, call sign Utah, and received permission to leave the flight plan and investigate. The contemporary wing summary states that Utah confirmed an object on its scopes northwest of Fort Worth and that visual, electronic and ground-radar contacts were lost and later reacquired during turns. Those correlations are preserved in typed reports, not in publicly available scope film or raw tracking data.

End of the encounter and surviving paperwork. Fuel requirements ended the investigation near Mineral Wells, although later visual and electronic contacts were still reported on the northbound leg. The surviving file includes a three-page message filed by the Duncanville squadron, a four-page intelligence summary prepared by the 55th Wing, observer questionnaires and later analysis. These records establish that the Air Force processed the incident; they do not preserve every operational exchange or instrument output.

The Condon date error. When the University of Colorado team revisited the case about a decade later, it initially worked from an incorrect fall 1957 date and could not locate the Air Force file. Case 5 therefore relied heavily on delayed crew recollections and concluded that the available descriptions were inadequate for identification. The error matters because weather and radar-propagation work performed for 19 September could not test an event that occurred on 17 July.

McDonald's reconstruction and its limits. Atmospheric physicist James E. McDonald later found the file, interviewed crew members and published a detailed reconstruction in the American Institute of Aeronautics and Astronautics journal in 1971. Aircraft commander Chase confirmed that account to the AIAA subcommittee. The paper is valuable because it quotes contemporary records and corrects the date, but its conclusions remain a later scientific interpretation rather than an Air Force finding.

Why the case remains unresolved. Later testimony conflicts over whether film, wire recordings or other instrument records ever existed: Chase remembered intelligence personnel collecting material, while McCoid and McClure said the aircraft carried no such recording media. Skeptical proposals include reception of ordinary ground radar, switching or relay effects, aircraft lights and mistaken correlation of separate events. None can be tested against a complete raw dataset. The defensible status is therefore unresolved, credibility B: unusually well documented as a report, but unable to identify the reported source or establish exotic technology.