Overview
Researchers Beatriz Villarroel, Stephen Bruehl and colleagues have reported a possible relationship between historical nuclear tests and fleeting, star-like markings found on astronomical photographic plates taken before the launch of Sputnik. The findings, presented in two arXiv studies published in September 2026 and supported by a recent Scientific Reports paper, are being interpreted by the researchers as evidence that at least some of the markings may represent real objects rather than photographic defects. The interpretation remains contested, and the studies do not establish that the objects were artificial or extraterrestrial.
The work centers on a catalog of 107,875 candidate transients identified in the first Palomar photographic survey. Because the relevant observations predate Sputnik, the researchers argue that conventional satellites and much of the human-made orbital debris now commonly invoked to explain similar flashes cannot account for the entire sample. The candidates appear as point-like sources on one plate but are absent from later observations, giving them their transient character.
Studies Examine Geometry and Distribution
In a paper titled Modeling Palomar Transients: Constraints from Reflection Geometry and Orbital Altitude, posted on September 4, the team modeled what the sources might look like if they were caused by sunlight reflecting from objects near Earth. The analysis considers the geometry required for a reflective object to produce a brief optical signal and attempts to estimate possible orbital altitudes and physical characteristics.
A second study, Earth-Projected Clustering of Historical Optical Transients, posted on September 8, examines the candidates’ geographic distribution. The researchers say the data include apparent clusters and alignments, as well as a reduction in detections in areas corresponding to the Earth’s shadow. Villarroel has described a possible explanation involving a relatively flat object that rotates and becomes visible only when its surface is favorably oriented toward the Sun and the observatory.
The estimated dimensions in the analysis range from centimeters to approximately three meters, although the researchers caution that these numbers are not necessarily the dimensions of an underlying object. A larger, less reflective body with several reflective surfaces could produce a comparable signal. In an interview with Ross Coulthart on Reality Check, Villarroel identified the apparent absence of features within the Earth’s shadow as one of her strongest arguments against a purely photographic explanation. “There is no reason why features would know where the Earth’s shadow is,” she said, describing the relevance of the changing Sun–Earth geometry.
Nuclear-Test Correlation Remains Disputed
The team’s earlier and more recent analyses also report a temporal association between some higher-probability transients and nuclear tests. In the Scientific Reports study, machine-learning methods were used to remove likely false detections and refine the candidate sample. The researchers say that, after this filtering, the association with nuclear tests became stronger.
They have presented the pattern as potentially consistent with surveillance or monitoring of nuclear activity. However, a correlation in historical data does not by itself demonstrate a causal relationship, identify the objects, or show that they were intelligently controlled. Selection effects, observing schedules, photographic artifacts, atmospheric conditions and the statistical treatment of the catalog remain important issues for independent review. The findings therefore represent a hypothesis for further testing rather than a confirmed detection of unknown technology.
Debate Over Methods and Future Research
Villarroel has also faced criticism over the interpretation of the archival material and has advocated greater use of existing observatory records to investigate unusual aerial phenomena. Her position is that valuable evidence may already exist in photographic archives and other astronomical datasets, even if it was not collected specifically for UAP research. Critics, meanwhile, have questioned whether the transient markings can be reliably separated from defects and whether the reported patterns could emerge through ordinary statistical or observational biases.
The Beyond Institute is planning a Stockholm panel in October 2026 addressing UAP research and related topics. The event is expected to contribute to a wider discussion about how historical astronomical data should be analyzed, what standards of evidence are appropriate, and how extraordinary interpretations can be tested against conventional explanations. For now, the Palomar candidates remain an unresolved population: potentially meaningful astronomical observations, but evidence that requires replication, transparent data analysis and independent scrutiny before stronger conclusions can be drawn.
