Weather Correlation Study (2020-01-01 to 2026-03-10) — Orb, Triangle

The analysis utilizes an aggregated dataset of 11,962 UAP sighting reports, sourced predominantly (99.97%) from the National UFO Reporting Center (NUFORC). The date range spans from 2020-01-01 to 2026-03-10. The study focuses exclusively on reports where the described shape was 'Orb' (N=10,322) or 'Triangle' (N=1,640), constituting the two most frequently reported structured shapes in the broader global dataset. The primary analytical method is a descriptive statistical examination of the dataset's intrinsic properties, including temporal trends, geographic clustering, and witness metrics. The study assesses the dataset's suitability for weather correlation by evaluating the density and reliability of the spatiotemporal data points. Specific dimensions analyzed include the monthly sighting frequency, the geographic distribution across U.S. states and specific latitude/longitude clusters, and the witness count and corroboration statistics.

The shape distribution is heavily skewed, with 'Orb' reports constituting 86.3% (10,322/11,962) of the dataset and 'Triangle' reports comprising 13.7% (1,640/11,962). Geographically, reports are overwhelmingly concentrated in the United States, with the top three states—California (1,228), Florida (764), and Texas (610)—accounting for 21.7% of all reports. The location point data reveals intense micro-clustering around major metropolitan areas (e.g., Phoenix, AZ [lat: 33.5, lng: -112, count: 160]; Los Angeles, CA [lat: 34, lng: -118, count: 152]; Portland, OR [lat: 45.5, lng: -122.5, count: 108]). Temporally, the monthly data from April 2024 onward shows a mean of approximately 150 reports per month. However, December 2024 is a profound outlier with 452 reports, representing a 201% increase over the period's mean. Witness statistics show an average of 2.2 witnesses per report across 26,692 total witnesses. The corroboration rate is effectively 0% (1 corroborated event out of 11,962 total sightings).

The extreme geographic clustering strongly correlates with population centers, indicating that reporting frequency is a function of human population density and potential cultural factors rather than a uniform atmospheric phenomenon. The dramatic December 2024 anomaly suggests a significant exogenous stimulus to reporting, such as media coverage, a specific viral social media event, or a coordinated reporting campaign, which overwhelms any baseline signal. The near-zero corroboration rate (0.008%) and the low average witness count are critical data quality indicators. They establish that this dataset is composed almost entirely of uncorroborated, anecdotal accounts, which severely limits the confidence with which any single report can be treated as a reliable observation of an external physical event. The predominance of the simple 'Orb' shape category, which is susceptible to misidentification of common objects (e.g., planets, stars, aircraft landing lights), further challenges the dataset's utility for isolating anomalous phenomena. These findings do not preclude the existence of anomalous events within the dataset but demonstrate that the signal is deeply embedded in noise dominated by reporting artifacts.

The dataset, as constituted, is not suitable for a meaningful weather correlation study. The high degree of geographic bias, the presence of massive temporal outliers driven by non-meteorological factors, and the critically low level of observational reliability (as measured by corroboration) mean that any attempted correlation with atmospheric conditions would be statistically invalid and scientifically inconclusive. The confidence level in deriving physical insights from this dataset is assessed as Low. The primary limitation is the uncontrolled, anecdotal nature of the source data, which lacks the validation protocols necessary for physical science. Recommendations for further research are twofold: First, future studies must prioritize the collection and analysis of sensor-based or multi-witness-corroborated cases where time and location are precisely known. Second, research should employ a case-control methodology, comparing the weather conditions at reported sighting locations/times against randomly selected control locations/times within the same geographic region to account for local climate baselines and population distribution.