Modelling Palomar Transients: Constraints from Reflection Geometry and Orbital Altitude

Abstract: Recent searches of digitised photographic plates from the Palomar Observatory have uncovered tens of thousands of short-lived transients, each visible in only a single exposure. Their morphologies indicate sub-second flashes, consistent with specular reflections from highly reflective objects in near-Earth orbits. In this paper, we present a modelling study using geometric shadow modelling, Monte Carlo simulations, and photometric constraints to infer their physical properties. We examine the observed groupings and alignments, which appear to be consistent with sudden changes in attitude. Assuming a spherical-shell model, the angular profile of the measured transient deficit around the antisolar point is consistent with a population at characteristic altitudes of $\sim 20,000$--25,000 km. A complementary estimate, based on the altitude dependence of the global Earth-shadow deficit, yields a broader characteristic range of $\sim$ 20,000--35,000 km above Earth's surface, extending into the geosynchronous orbital (GSO) region. Under simplified geometric and photometric assumptions and assuming $\sim20{,}000$ to $35{,}786$ km above the Earth's surface, the inferred sizes of the specularly reflecting facets range from centimetre scales up to $\sim3$ m, with characteristic flash durations of $\sim320$ ms and slow rotation rates. We explore both natural and non-natural toy models consistent with the data, and provide constraints to guide future observational searches.

• The paper models Palomar Transients as sub-second optical flashes from specular reflection in near-Earth orbit, constraining orbital altitudes to 20,000–35,000 km.
• Using Earth-shadow geometry and Monte Carlo simulations, the study infers reflecting areas ranging from centimeters to 3 meters, with characteristic flash durations near 320 ms.
• These results imply slow rotational dynamics and a possible orbital alignment with the celestial equator, providing constraints on geomagnetic and nuclear test correlations.

Scope and central premise

“Modelling Palomar Transients: Constraints from Reflection Geometry and Orbital Altitude” (2609.05105) develops a phenomenological model for short-lived point sources identified in digitized POSS-I photographic plates. The paper treats the transients as genuine, sub-second optical flashes produced by specular reflection from objects in near-Earth orbit, rather than as plate defects or scanning artifacts. It then uses Earth-shadow geometry, Monte Carlo simulations, photometric scaling, and simplified rotational models to constrain orbital altitude, reflecting area, flash duration, and angular velocity.

The analysis is explicitly conditional. The authors do not claim that the data uniquely establish an artificial or non-human origin. Instead, they adopt an orbital-reflection hypothesis and ask what physical parameters are compatible with the observed transient distributions. The principal inferred altitude range is approximately $20{,}000$–$35{,}000$ km above Earth’s surface, extending into the geosynchronous-orbit region. The corresponding effective reflecting facets range from centimetre scales to approximately $3$ m, while the simulated population is dominated by slowly rotating objects producing flashes with characteristic durations near $320$ ms.

The paper builds on a POSS-I catalogue containing 107,875 transients, derived from an initially larger detection set through filtering procedures described in earlier work. A machine-learning classifier is also used to define a higher-confidence subset of 11,418 candidates with transient probability $p>0.7$. This distinction is important: many of the strongest statistical effects appear after removing ambiguous detections, but the filtering itself introduces selection effects that complicate the interpretation of absolute rates and physical population properties.

Observational basis and interpretation

The transient candidates occur in single photographic exposures, whose typical durations are approximately 40–60 minutes. Their point-spread functions are described as star-like but somewhat narrower than ordinary stellar profiles. The paper interprets this morphology as consistent with flashes much shorter than the plate exposure, because temporal integration of a short flash can produce a compact image rather than a trail. This interpretation is not unique: emulsion defects, scratches, and other plate artifacts can also generate compact detections, and previous work has specifically argued for an instrumental origin. The paper responds by emphasizing the recovery of similar events in independent plate archives and reports of optical coma consistent with light passing through the telescope rather than defects located in the emulsion.

The authors also point to grouped events, alignments, and close pairs. Earlier studies reported nine transients within a $10\times10$ arcmin$^2$ region and a separate triple-transient event. In the present paper, these structures are treated as potentially related to common motion or attitude changes. The interpretation is strongest for configurations in which flashes occur along an apparent track and where paired flashes are displaced transverse to that track. The authors argue that such morphologies could arise from multiple reflecting facets or from changes in the orientation of an extended object.

The broader evidential context includes reported temporal correlations with nuclear tests and unidentified-anomalous-phenomena reports, as well as an anticorrelation with geomagnetic-storm activity. These results are not rederived in detail here; they are imported as constraints from related analyses, including the machine-learning study “Machine Learning Supports Existence of Previously Unrecognized Transient Astronomical Phenomena in Historical Observatory Images” (Bruehl et al., 20 Apr 2026). Consequently, the physical modelling in this paper cannot independently validate those correlations. Its principal new contribution is the attempt to translate the Earth-shadow deficit and transient morphology into orbital and geometric parameters.

Earth-shadow deficit and altitude inference

The central observable is a deficit of transients in the direction of Earth’s antisolar point. If the sources are illuminated by the Sun and reflect sunlight toward the telescope, objects in the umbra should produce substantially fewer optical glints. The angular size of the shadow at the object’s location decreases with increasing altitude, so the angular profile of the deficit provides a geometric altitude diagnostic.

The paper compares the transient sample with a control catalogue generated at the same plate coordinates and observation times. Earth-shadow membership is calculated using the earthshadow software, with the analysis repeated across assumed altitudes from 20,000 to 80,000 km. The observed deficit is largest near the antisolar point and declines with angular separation, qualitatively matching the expected shadow structure.

For the complete catalogue, the reported global deficits are approximately $0.44$ at 20,000 km, $0.57$ at 25,000 km, $0.56$ at 35,000 km, and $35{,}000$0 at geosynchronous altitude. For the quality-filtered sample, the deficit rises to $35{,}000$1, $35{,}000$2, $35{,}000$3, and $35{,}000$4 at the same respective altitudes. At 35,000 km, the quality-filtered value is reported as $35{,}000$5, corresponding to $35{,}000$6 under the quoted Poisson uncertainty; at 25,000 km, the formal significance reaches $35{,}000$7 because more objects fall within the corresponding shadow region.

These large significances should not be interpreted as direct measurements of altitude. The paper correctly notes that statistical significance depends on the number of objects included in the assumed shadow region, whereas the physical deficit amplitude is more relevant to characteristic altitude. The authors therefore favour approximately 25,000–35,000 km rather than simply selecting the altitude with the largest formal significance. They also acknowledge that the complete sample contains contamination that is approximately independent of Earth–Sun geometry and therefore dilutes the deficit.

The angular-profile analysis provides a second estimate. Monte Carlo populations are distributed in spherical shells, and the predicted transition from strong deficit to recovery is compared with the observed profile. Fits restricted to the initial drop region favour approximately 20,000–25,000 km, especially for the complete catalogue. The quality-filtered sample shifts modestly toward 25,000 km when the fitting interval is extended. Combining both procedures, the paper adopts $35{,}000$8–$35{,}000$9 km as a characteristic range rather than a formal confidence interval.

Figure 1: Observed transient deficit as a function of angular distance from the antisolar point, showing the strongest suppression near the expected Earth-shadow direction.

The spherical-shell model is a substantial simplifying assumption. It excludes detailed orbital inclinations, longitude distributions, illumination phase functions, object cross sections, attitude states, and correlations between object location and detectability. The paper explicitly notes that an equatorial ring or another anisotropic distribution could alter the mapping between angular deficit and altitude. Thus, the altitude constraints are model-dependent geometric estimates, not orbit determinations for individual sources.

Declination structure and orbital distribution

The high-probability sample exhibits a non-uniform declination distribution. Relative to the control sample, transient counts exceed expectations by approximately 30–75% at declinations near $3$0–$3$1, decline at intermediate declinations, and become deficient at high declination. A secondary excess near $3$2 is reported but remains unexplained.

The low-declination excess is interpreted as evidence for concentration near the celestial equator, consistent with an equatorial or near-equatorial orbital population. This is qualitatively compatible with geosynchronous and other high-altitude orbital objects, for which optical glints are preferentially observed near the equatorial plane. It is also presented as a difficulty for asteroid, cometary, or zodiacal-debris explanations, since those populations are generally associated with the ecliptic rather than the celestial equator.

Figure 2: Relative excess of high-probability transients versus declination, including the low-declination enhancement and the unresolved secondary feature near $3$3.

The inference is limited by the restricted right-ascension and declination cuts, plate-edge treatment, survey geometry, and the classifier’s possible dependence on morphology or location. In particular, the first declination bin is sensitive to the imposed $3$4 cut, and the high-declination tail contains fewer detections. The distribution is therefore evidence for non-uniformity, but its conversion into an orbital inclination distribution requires a forward model of sky coverage and transient detectability.

Photometric constraints on reflecting area

The photometric calculation converts observed plate magnitudes into effective reflecting areas. The authors account for the long POSS-I exposure by assuming that a flash lasting 0.1–1 s is diluted over an exposure of approximately 3,000 s. They examine both Lambertian and specular reflection, but correctly note that steady diffuse reflection would generally produce a persistent trail rather than a compact flash. The physically relevant interpretation in their model is therefore a short-lived specular reflection from a planar facet.

For a representative bright candidate at $3$5, a perfect reflector at the adopted orbital altitudes requires a facet area of approximately $3$6–$3$7 m$3$8, depending on flash duration. At reflectivity $3$9, the required area increases to approximately $320$0–$320$1 m$320$2, corresponding to equivalent linear dimensions of roughly $320$3–$320$4 m.

Across the full magnitude, altitude, reflectivity, and duration ranges, the paper obtains equivalent facet sizes from centimetres to approximately $320$5 m. More representative transients near $320$6 and with a characteristic duration of approximately $320$7 s require projected areas of roughly $320$8–$320$9 m$p>0.7$0 for $p>0.7$1, or $p>0.7$2–$p>0.7$3 m$p>0.7$4 for $p>0.7$5. These values correspond to equivalent facet dimensions of approximately 4.7–8.3 cm and 15–26 cm, respectively.

Figure 3: Required reflecting area as a function of plate magnitude for specular and diffuse models at 20,000 km and geosynchronous altitude.

The estimates are highly sensitive to reflectivity and flash duration. The perfect-reflector case is an optimistic lower bound, while the assumed $p>0.7$6 case is an illustrative low-reflectivity scenario rather than a measured material property. The model also treats the reflector as planar and omits a full bidirectional reflectance distribution function, phase-angle dependence, surface roughness, and optical-system effects. The authors identify BRDF modelling and calibration against known satellite glints as necessary steps before the areas can be regarded as robust physical sizes.

Rotational dynamics and flash durations

The rotational analysis uses Monte Carlo simulations of cubes, icosahedra, and a saucer-like multifaceted geometry. The simulations vary spin, precession, nutation, object geometry, and solar phase angle, selecting configurations that produce approximately one detectable flash per plate rather than repeated flashes throughout a 50-minute exposure.

For cube geometries, viable spin rates extend from approximately $p>0.7$7 to $p>0.7$8 revolutions per minute, with flash durations from roughly 60 to 955 ms in the explored cases. Icosahedral models generally require slower rotation and produce longer flashes, with durations from approximately 90 to 1,775 ms. The combined cube–icosahedron simulations yield a mean spin rate of approximately $p>0.7$9 RPM and a mean flash duration of approximately $10\times10$0 ms. The saucer-like model has a characteristic rotation rate near $10\times10$1 RPM, or about one revolution per hour, and produces approximately one to two detectable flashes during passage through a POSS-I field.

Figure 4: Simulated flash-duration distribution, with most modelled events concentrated between approximately 60 and 600 ms.

The authors emphasize that longer flashes may be incompatible with the observed compact PSFs because orbital motion would elongate the image. At 20,000 km, flashes longer than roughly 0.2–0.3 s may produce detectable elongation; at geosynchronous altitude, the corresponding threshold is approximately 0.5 s. This provides an independent morphology-based constraint on duration, although it depends on angular velocity, field position, plate resolution, and the PSF model.

Figure 5: Relationship between spin rate and flash duration for cube and icosahedral models, illustrating the trade-off between rotation and the number of detectable flashes.

The rotational results should not be read as population estimates. The simulations use selected geometries and parameter combinations designed to reproduce a low per-plate flash count. They do not employ empirically calibrated priors for debris spin states, surface BRDFs, or attitude dynamics. The inferred slow rotation is therefore a compatibility condition within the toy models, not a statistically identified distribution.

Velocity and alignment constraints

The paper estimates transverse velocity from the angular extent of aligned transients divided by an assumed time interval. For compact alignments spanning only a few arcminutes, the resulting minimum velocities are generally sub-orbital. For alignments spanning 3–6 degrees across a full $10\times10$2 plate during a 50-minute exposure, the inferred minimum velocity can correspond to circular orbital motion at an altitude near 66,580 km. If the same displacement occurs over 25 minutes, the implied velocity is compatible with a circular orbit near 41,149 km.

These calculations are deliberately permissive. They provide lower bounds because the exposure interval is not known, the object may not have produced flashes continuously, and the geometry is simplified by choosing an overhead trajectory. Upper velocities remain unconstrained. Moreover, a set of aligned point sources need not be a single moving object; multiple objects in an orbital structure, plate artifacts, or detection-selection effects remain alternative explanations.

The discussion of transverse doublets is more speculative. The paper argues that enhanced flash activity near the beginning and end of apparent tracks could reflect changes in attitude, while perpendicular separations may indicate multiple reflecting surfaces or an extended structure. A separation of approximately 5 arcsec is converted into a projected scale of order 1 km at geosynchronous distance. That scale is too large for multiple facets on a compact rigid body, but this conclusion depends directly on the assumed altitude and on interpreting the paired detections as simultaneous components of one physical system.

Comparison of explanatory models

The paper considers natural bodies, magnetospheric plasmoids, and non-human artificial objects. Natural explanations include reflective debris, asteroidal fragments, and cometary material. The authors argue that diffuse natural surfaces would tend to produce sustained or elongated signals rather than sub-second point-like flashes. They also emphasize the apparent avoidance of the ecliptic and concentration near the celestial equator as a demographic mismatch for ordinary Solar-System populations.

However, the natural-body analysis is not a complete population synthesis. The paper acknowledges that small diffuse bodies could evade streak detection, particularly when their light is distributed over many pixels. It estimates that low-reflectivity diffuse objects with projected areas below approximately 120–170 m$10\times10$3, corresponding to diameters of roughly 12–15 m, may not produce clearly detectable trails under the adopted criterion. Under more sensitive matched-filter assumptions, objects as small as approximately 0.8–1.2 m could become detectable. This range demonstrates that the exclusion of natural objects depends on the definition of detectability and on the treatment of faint trails.

The dusty-plasma interpretation is judged inconsistent with the inferred centimetre-to-metre reflecting areas, although the paper notes that geomagnetic modulation could motivate such a model. This is a quantitative tension rather than a definitive exclusion, because the optical-emission mechanism and collective scattering properties of a plasma ensemble are not modelled in detail.

The strongest claim concerns the non-human-intelligence hypothesis. The paper states that this is the only considered framework that can, in principle, accommodate the combination of compact PSFs, groupings, Earth-shadow avoidance, correlations with nuclear tests and UAP reports, and geomagnetic-storm anticorrelation. This is a comparative claim among the paper’s selected toy models, not a positive identification of NHI. The argument depends heavily on external statistical results and on accepting a common origin for phenomena that may instead be heterogeneous or affected by unrecognized selection effects. The orbital-reflection model itself is more directly supported by the Earth-shadow geometry than by the proposed interpretation of the temporal correlations.

Limitations and open questions

The paper’s principal limitation is the dependence on a prefiltered transient catalogue. Bright sources with $10\times10$4 were excluded, while streaks and elongated objects were removed because the analysis focuses on point sources. The inferred population is therefore biased toward fainter, compact events and may omit larger reflective objects and fast-moving objects in lower orbits. The reported size distribution cannot be extrapolated to the full orbital population.

The spherical-shell geometry is another major assumption. It is useful for converting shadow angular scales into characteristic altitudes, but it does not represent a realistic orbital phase-space distribution. A ring-like equatorial population, a mixture of inclinations, or multiple altitude components could produce similar profiles with different physical interpretations. The quoted $10\times10$5–$10\times10$6 km range is consequently a model-dependent characteristic scale, not an orbital census.

The photometric and rotational models are likewise underconstrained. Perfect specular reflection, simplified planar facets, selected flash durations, and ad hoc geometric shapes are used to derive effective areas and spin rates. The simulations do not yet reproduce the full joint distribution of magnitude, PSF width, alignment, sky position, and event rate. A calibrated BRDF, realistic debris-orientation priors, and forward modelling of plate detection efficiency are required to determine whether one physical population can simultaneously account for all observed properties.

Several specific questions remain open: whether the shadow profile persists in independent plate archives; whether modern surveys recover the predicted transverse doublets; whether parallax measurements place individual events at the inferred distances; and whether the correlations with nuclear testing and geomagnetic activity survive preregistered analyses with fully specified selection functions. These tests bear directly on whether the Earth-shadow signal is an orbital-reflection phenomenon or a compound result of catalogue construction and observational systematics.

Conclusion

The paper presents a coherent but conditional geometric interpretation of Palomar photographic-plate transients. Under the assumption that the detections are short specular flashes from objects near Earth, the Earth-shadow deficit favours characteristic altitudes of approximately $10\times10$7–$10\times10$8 km, with possible extension into geosynchronous orbit. Photometric scaling implies effective reflecting facets from centimetres to approximately $10\times10$9 m, while rotational simulations favour slow tumbling and flash durations centred near $^2$0 ms.

The strongest contribution is the conversion of a population-level shadow statistic into quantitative constraints on altitude and reflection geometry. The strongest limitation is that each inference depends on simplified spatial, photometric, and dynamical models applied to a selection-biased point-source catalogue. The results therefore establish a set of testable parameter ranges rather than a unique physical identification.