The nuclear pattern: what the data actually says
Three peer-reviewed studies have now tested whether UAP reports cluster around US nuclear facilities. The answer is yes, and the magnitude is larger than most commentary admits.
The recurring claim that UAP activity clusters around US nuclear-weapons facilities, ICBM launch fields, and atomic-research labs has, until recently, lived in oral history — Robert Hastings' 'UFOs and Nukes' (2008), the Salas Malmstrom testimony, the Bentwaters story. Since 2022, three peer-reviewed and SCU-reviewed studies have tested the claim statistically. This briefing summarises what they found and what the methodology can and cannot support.
What's being tested
The question, stated carefully, is: do UAP reports from the 1945–1975 window distribute non-randomly with respect to the geographic and temporal footprint of the US Atomic Warfare Complex — i.e. the weapons-design labs (Los Alamos, Sandia, Livermore), the materials-production sites (Hanford, Oak Ridge, Pantex, Rocky Flats, Savannah River), the testing ranges (NTS, the Pacific Proving Grounds), and the operational ICBM and SLBM bases (Malmstrom, F.E. Warren, Minot, Whiteman, Ellsworth, the Atlantic-coast submarine bases)?
The null hypothesis the studies test against is straightforward: UAP reports might cluster around nuclear sites simply because those sites are in geographies where reports were more likely to be made (proximity to military airspace with active radar, proximity to defended airspace where unauthorised aircraft are noticed quickly, proximity to populated areas with reporting infrastructure). The methodological work is in controlling for that.
The Hancock 2022 study
Larry Hancock and colleagues, working under the Scientific Coalition for UAP Studies, published 'UAP Pattern Recognition Study 1945–1975: US Military Atomic Warfare Complex' as an SCU white paper in 2022. The study used a curated dataset of approximately 1,500 UAP reports drawn from Project Blue Book, the FBI Vault, the NICAP archive, and contemporary newspaper indices. It mapped report density against the Atomic Warfare Complex sites enumerated above.
The headline finding: UAP reports cluster around AWC sites at densities significantly higher than baseline civilian-population-weighted expectations. The clustering is most concentrated in the 1945–1955 window (the early-Cold-War weapons-design and -testing era) and 1965–1975 (the Minuteman ICBM operational deployment era). The 1955–1965 window shows a relative dip that the authors attribute primarily to reporting-channel changes rather than to a real decline in events.
The Grosvenor 2025 follow-up
Robert Grosvenor and colleagues extended the Hancock methodology in 'UAP Indications Analysis 1945–1975: United States Atomic Warfare Complex,' published in Limina (the open-access peer-reviewed UAP journal) in 2025. The follow-up refined the per-facility intensity scoring, controlled more rigorously for military-airspace and base-aviation effects, and added a temporal-correlation analysis against the major weapons-development and -deployment milestones.
The refined finding: the clustering is robust to the additional controls. The strongest single-site signal sits at Oak Ridge — the Tennessee uranium-enrichment and weapons-research complex — over the 1948–1952 window. The Hanford signal is a close second, concentrated in the 1949–1952 plutonium-production years. Malmstrom Air Force Base in Montana produces the strongest ICBM-era signal over 1966–1968.
The Limina paper is careful about the interpretive limit: a statistically significant correlation does not specify a mechanism. The data establishes that something happened more often near AWC sites than baseline. It does not establish what.
The Palomar transient evidence
The third 2020s entry is methodologically different. Beatriz Villarroel, Mattias Brühl, and colleagues published 'Transients in the Palomar Observatory Sky Survey May Be Associated with Nuclear Testing and Reports of Unidentified Anomalous Phenomena' in Scientific Reports in 2025. The paper is the most prestigious peer-review venue any UAP-pattern claim has reached.
The methodology: the Villarroel group analysed photographic-plate transients from the original 1950s Palomar Observatory Sky Survey — point sources that appear in one plate and not in subsequent imaging of the same field. The hypothesis was that transients might cluster temporally with atmospheric nuclear tests (which create reflective ionospheric disturbances) and with documented UAP report dates.
The finding: the transient count rises during periods of atmospheric nuclear testing at the p<0.05 level, and shows secondary correlations with documented UAP-report dates from the contemporaneous record. The authors are explicit that this is correlation, not causation, and offer multiple potential explanations including reflective-debris artefacts. But the temporal coupling is real and replicable.
What the synthesis supports
Three independent methodologies — case-density mapping, refined indications analysis, photographic-transient counting — point in the same direction. UAP report density and adjacent observational anomalies do cluster around the US nuclear apparatus over the 1945–1975 window. The clustering is statistically significant at the per-study levels, and the cross-methodology coherence makes the overall claim difficult to dismiss as a reporting-bias artefact.
What the synthesis does not support is any specific physical or intentional explanation. The pattern is compatible with several distinct causes: routine intelligence reconnaissance by adversary states (the most parsimonious), atmospheric-physics interactions with nuclear-testing artefacts (the Villarroel direction), instrumented surveillance by an as-yet-unidentified non-human agent (the maximalist reading), and observer-side selection effects that the authors have tried but cannot fully exclude.
What it does demand is a continuing research programme. AARO's Historical Record Report Volume I (2024) treats the nuclear-site pattern as part of the historical baseline but does not characterise the underlying phenomena. The peer-reviewed literature now exceeds AARO's own analytical engagement with the question. That gap is, itself, part of the modern UAP story.