At least some of the disputed pre-Sputnik flashes were sunlight reflected from objects physically near Earth (at geosynchronous-like altitudes) rather than light from distant astronomical sources.

Depends entirely on TRN-C020 plus contested shadow geometry; no geometric signature has yet survived an independently constructed catalog.

The argument for a near-Earth reflective origin rests primarily on geometric signatures like the Earth-shadow deficit, which currently replicate only conditionally on unvalidated working catalogs. The strongest opposing consideration is that independent analyses of more stringently vetted datasets lack this deficit, leaving the geometric case unsupported.

The main line of support for a near-Earth reflective population is the reported Earth-shadow deficit and nuclear-test timing correlation in the 107,875-object working catalog (TRN-E010, TRN-E015), which are consistent with objects shining by reflected sunlight and avoiding Earth’s umbra. However, these effects vanish or become insignificant when the spatially non-uniform background is modeled properly and when more heavily vetted subsets are used (TRN-E011), and the nuclear correlation disappears in an independently extracted catalog with per-night exposure modeling (TRN-E009). The multi-source events (TRN-C001, TRN-C003) do not by themselves demonstrate near-Earth geometry, and TRN-C020 (real flashes) is not established. Thus any support for a near-Earth origin is very weak and heavily conditional on a contested catalog. The strongest opposing consideration is the Watters/Hayes combination showing that ensemble effects are highly sensitive to catalog definition and likely reflect survey logistics and artifacts rather than orbital geometry.

A near-Earth reflective population is a geometric hypothesis, and its only published geometric signature — the Earth-shadow deficit — replicates only inside the VASCO working catalog and fails in Watters et al.'s reanalysis with a non-uniform background and harder vetting. Multiple simultaneous point sources would argue against most distant-source models if the flashes were real, but that premise is TRN-C020, which is open. The strongest opposing consideration is that the near-Earth reading currently rests on the least-vetted rows and a null that the spatial data visibly violate.

The only positive geometric evidence is the Earth-shadow deficit, which reproduces on the author-supplied catalog but vanishes under a non-uniform background null and in heavily vetted subsets (Watters), with the proponents' rebuttal resting on statistical power rather than a counter-demonstration. The strongest opposing consideration is the simultaneous-multiple argument: nine point flashes within 10 arcminutes, or a bright triple within 10 arcseconds, are genuinely awkward for distant-source astrophysics — if the flashes were made by incident light. That premise (TRN-C020) is exactly what is unresolved, and the multiple-transient events would dissolve if shown to be copy-stage defects. Neither established nor refuted; the preregistered multi-catalog shadow test (TRN-R005) is the live instrument.

The near-Earth reading rests almost entirely on the Earth-shadow deficit, and that deficit is computed against a uniform-random background that Watters et al. show the data visibly violate, with no significant deficit surviving in more heavily vetted subsets. The proponents' rebuttal is a power argument, not a counter-demonstration, and the claim is in any case conditional on TRN-C020, which is open. The strongest opposing consideration in the claim's favor is genuinely non-trivial: multiple simultaneous point flashes within ten arcminutes, or three within ten arcseconds vanishing inside an hour, are geometrically awkward for most distant-source astrophysics if the flashes were real at all — so a demonstration of incident light would immediately make this rung far more plausible than it now is.

Still depends entirely on TRN-C020 plus contested shadow geometry that has not survived independent catalog reconstruction (TRN-E009, TRN-E011). TRN-E020 adds a candidate natural mechanism for near-Earth objects but is weak, unreplicated, and premised on the transients being real.

Still wholly dependent on TRN-C020 plus contested shadow geometry that has not survived an independently constructed catalog (TRN-E009, TRN-E011). TRN-E020 adds a candidate natural mechanism for near-Earth reflectors, which broadens the hypothesis space at this rung without adding weight — it presupposes the transients' reality and shares the contested catalog lineage.

Depends entirely on TRN-C020 plus the contested shadow geometry; no geometric near-Earth signature has yet survived an independently constructed catalog (TRN-E009, TRN-E011).

Depends entirely on TRN-C020 plus the contested shadow geometry; TRN-E017 removes one artifact route for the geometry but no geometric signature has yet survived an independently constructed catalog (TRN-E009, TRN-E011).

The near-Earth reflective origin hypothesis depends entirely on ensemble geometric signatures, primarily the Earth-shadow deficit, which has yielded deeply conflicting results based on dataset definitions. Furthermore, recently proposed natural mechanisms, such as storm-driven dusty plasmas, offer alternative explanations for near-Earth optical phenomena without requiring solid reflective objects. The strongest opposing consideration is that the foundational Earth-shadow deficit fails to appear when the analysis is run on more heavily vetted candidate sets that properly account for the spatial non-uniformity and clustering inherent in the background data.

The primary geometric evidence for a near-Earth origin is the Earth-shadow deficit. However, this signature is strongly contested; independent reanalysis indicates the uniform-random background assumption is false and finds no significant deficit when spatial heterogeneity is properly modeled. Until the shadow deficit is demonstrated robustly across independently extracted catalogs, the near-Earth interpretation remains unresolved.

Support for a near-Earth reflective origin comes mainly from the reported Earth-shadow deficit (TRN-C005) and nuclear-test correlation (TRN-C006), which are replicated statistically on the VASCO working catalog and persist after some internal vetting (TRN-E010, TRN-E015, TRN-E017). However, Watters et al. show that the spatial distribution of features violates the uniform-background assumption and that vetted subsets yield no significant deficit or robust nuclear effect (TRN-E011), and Hayes’s independently extracted catalog finds no significant nuclear association once nightly coverage is modeled (TRN-E009). Moreover, whether the marks are real flashes at all (TRN-C020) remains open; if many are defects, their geometry is uninformative. The strongest opposing consideration is thus that all claimed geometric signals depend on a contested, unvalidated catalog and vanish or weaken under stronger controls.

The near-Earth interpretation relies heavily on the Earth-shadow deficit (TRN-C005) and, secondarily, on nuclear-test timing (TRN-C006). Those effects are reproduced in independent statistics applied to the same VASCO working catalog (TRN-E010, TRN-E015) but are strongly undermined when spatial inhomogeneities and observing schedules are accounted for (TRN-E011) and when the catalog is independently rebuilt (TRN-E009). While simultaneous multi-source events could, if confirmed as real flashes, disfavor standard distant astrophysical sources, that rests on TRN-C020, not yet established. The strongest opposing consideration is that all geometric and temporal correlations presently supporting a near-Earth population are contingent on a single, unvalidated catalog lineage and do not survive more robust null modeling and independent extraction.

The near-Earth reading rests almost entirely on the Earth-shadow deficit (TRN-C005), which replicates only inside the VASCO working catalog and disappears once spatial non-uniformity is modeled or the catalog is rebuilt. Simultaneous multi-point flashes would disfavor distant astrophysical sources if the flashes were photons, but that premise is TRN-C020, which is open. The strongest opposing consideration is Watters et al. (TRN-E011): the geometric signature is computed against a uniform background the data violate, and it is absent from the subsets vetted hardest against plate and scan artifacts.

The near-Earth reading rests almost entirely on the Earth-shadow deficit, which is mixed: present in the working catalog and among ML-scored subsets, absent once spatial heterogeneity is modeled and in more heavily vetted samples. Simultaneous multi-point events would be awkward for distant astrophysical sources if the flashes were real, but that premise is TRN-C020 and remains open. Strongest opposing consideration: Watters et al. find no significant shadow deficit after replacing the false uniform-background null, so the geometric signature currently lives in the least-vetted data.

The supporting geometry — the Earth-shadow deficit — reproduces on the working catalog with independent code (E010) and reportedly strengthens among ML-vetted candidates (E015), but Watters et al. show the uniform-background null is false for these data and find no significant deficit in the hardest-vetted subsets (E011); the proponents' rebuttal (E012) argues statistical power, not a counter-demonstration. The qualitative argument that nine simultaneous flashes within ten arcminutes sit awkwardly with distant astrophysical sources is real but fully conditional on TRN-C020, which is unresolved. Cann's preregistered post-storm profile and natural near-Earth mechanism (E020) keep the rung conceptually alive without adding probative weight. The strongest opposing consideration: the deficit's significance is lineage-bound and has never been shown in an independently constructed catalog. Unresolved, moderate confidence.

The only direct geometric support is the Earth-shadow deficit, which is mixed at best (see TRN-C005): reproducible on the working catalog against a naive expectation, absent under a defensible null and in vetted subsets. The strongest opposing consideration to a skeptical dismissal is the simultaneity argument: nine points within ten arcminutes in half an hour, and a triple within ten arcseconds, strain most distant-source astrophysics — if the flashes were real. But that premise is TRN-C020, which is unresolved, and Watters et al. document geometric clusters and edge concentration among the features themselves, so clustered marks are also what plate and scan artifact populations produce. This rung is hostage to TRN-C020 and to a shadow analysis nobody has run on an independently built catalog; unresolved is the warranted grade.

The near-Earth reading rests almost entirely on the shadow deficit, which reproduces only conditional on the VASCO working catalog (TRN-E010, TRN-E015) and fails once the background's demonstrated spatial heterogeneity is modeled and vetted subsets are used (TRN-E011). The rebuttal (TRN-E012) is a power argument, not a counter-demonstration. Multiple simultaneous point sources across tens of arcminutes would indeed argue against distant astrophysical origin if the flashes were real, but that premise is TRN-C020 and is unresolved. The strongest consideration on the other side is that TRN-E020's dusty-plasma proposal, if it survived, would satisfy this claim by a wholly natural route — so this rung is not intrinsically exotic and could in principle resolve positively without the headline following.

The near-Earth reading depends on the shadow geometry, which reproduces only within the unvalidated 107,875-row working catalog (TRN-E010, TRN-E015) and fails once the non-uniform background is modeled or the sample is vetted harder (TRN-E011); the proponents' reply argues power loss rather than demonstrating the effect survives (TRN-E012). The genuinely strong point in this direction is that multiple point sources appearing and vanishing together within arcminutes on one exposure is awkward for most distant-source astrophysics — but that argument presupposes TRN-C020, which is open, and clustered plate defects are not excluded. Doherty's alignment preprint (TRN-E018) adds a pattern claim with post-hoc criteria and modest p-values, not a geometric confirmation.

Depends wholly on TRN-C020 plus shadow geometry that does not survive a corrected null or an independently built catalog (TRN-E009, TRN-E011). TRN-E020's preregistered storm-recovery profile is a methodological credit but weak, unreplicated, and same-lineage.

Still depends entirely on TRN-C020 plus contested shadow geometry that has not survived an independently constructed catalog (TRN-E009, TRN-E011). TRN-E020's storm-recovery result is preregistered but weak, same-lineage, and unreplicated.

Depends wholly on TRN-C020 plus shadow geometry that does not survive a corrected null or an independently built catalog (TRN-E009, TRN-E011). TRN-E020's preregistered storm-recovery profile is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that does not survive a corrected null or an independently built catalog (TRN-E009, TRN-E011). TRN-E020's preregistered storm-recovery profile is a methodological credit but weak, unreplicated, and same-lineage.

The geometric evidence for near-Earth origin (the Earth-shadow deficit) replicates on the proponents' working catalog but fails when spatial clustering and background nonuniformity are controlled for. Furthermore, even if near-Earth origin were true, natural mechanisms like dusty plasmas have been proposed. The strongest opposing consideration is that the core geometric signal dissolves under strict independent reanalysis of spatial distributions.

For a near-Earth reflective population, we would expect robust geometric signatures (e.g., Earth-shadow avoidance) across independently constructed catalogs and, ideally, per-event distance constraints. The shadow deficit appears in the proponent working catalog and persists after some internal robustness checks (TRN‑E010, TRN‑E015, TRN‑E017), but disappears when background heterogeneity is modeled and stricter vetting is applied (TRN‑E011), and no comparable signature has been demonstrated on Hayes’s independent catalog. The strongest opposing consideration is that all ensemble patterns so far are contingent on disputed candidate lists that likely mix defects and stars; without a validated, flash-dominated catalog and any direct distance measurement, a near‑Earth origin remains unproven.

A near-Earth reflective origin is the one reading that would make an Earth-shadow deficit and simultaneous multi-point flashes geometrically natural, but the deficit (TRN-C005) replicates only inside the VASCO working catalog and is disputed once spatial non-uniformity is modeled, and the multi-point events only constrain distant astrophysics if the marks are real light (TRN-C020), which is the open crux. Cann’s post-storm dusty-plasma paper, if it held, would actually support this rung while undercutting artificiality, yet it inherits the same unvalidated catalog and is a single-author preprint. The strongest opposing consideration is Watters et al.: after discarding the false uniform-background assumption, they find no significant shadow deficit in more heavily vetted subsets, so the geometric case currently rests on the least-vetted data.

The near-Earth rung rests almost entirely on the Earth-shadow deficit (TRN-C005), which reproduces on the author-supplied catalog but not in vetted subsets with realistic backgrounds (Watters) and has never been reproduced in an independently constructed catalog. Multiple simultaneous point flashes across arcminutes would, if real, strain most distant-source explanations — but the conditional is the whole case, since TRN-C020 is unresolved. The strongest opposing consideration is that the multi-event geometry (nine at once; a tight triple) is genuinely awkward for independent plate defects, so this rung is not merely parasitic on statistics — but copy- and emulsion-defect clustering (Watters's edge/corner concentrations) is awkward in the same places. Unresolved, with the preregistered multi-catalog shadow test (TRN-R005) as the designated arbiter.

The near-Earth reading rests almost entirely on the Earth-shadow deficit, which reproduces only within the contested 107,875-row working catalog and fails once the background null is derived from the data's actual spatial heterogeneity (TRN-E011); the proponents' reply (TRN-E012) argues power loss rather than demonstrating the deficit in a vetted set. The secondary argument — that multiple simultaneous point flashes across arcminutes disfavour distant astrophysical sources — has real force but is conditional on the flashes being flashes, which is exactly TRN-C020. The strongest opposing consideration in the other direction is that if the shadow deficit did survive an independently built catalog it would be a geometric signature with no plausible plate-chemistry analogue; that test has not been run, so the claim sits open rather than refuted.

Depends wholly on TRN-C020 plus shadow geometry that does not survive a corrected null or an independently built catalog (TRN-E009, TRN-E011). TRN-E020's preregistered storm-recovery profile is a methodological credit but weak, unreplicated, and same-lineage.

The primary evidence for this claim is the Earth-shadow deficit in the VASCO catalog (TRN-E010, TRN-E015). However, this geometric argument collapses when independent extractions or more rigorous spatial background modeling are applied (TRN-E011). The strongest opposing consideration is that even if the transients are real, a natural mechanism like storm-enhanced dusty plasmas (TRN-E020) could explain near-Earth reflections without invoking artificial glinting objects.

The main geometric support—the Earth-shadow deficit in the 107,875-row working catalog (TRN-C005, via TRN-E010, TRN-E015)—is countered by Watters et al.’s demonstration that the background is highly non-uniform and that more vetted subsets show no significant deficit (TRN-E011). Additionally, Hayes’s independently extracted catalog shows no robust nuclear-window effect once nightly coverage is modeled (TRN-E009), weakening arguments for an orbital population tied to human activity. On the other hand, multi-source events (TRN-C001, TRN-C003) and Cann’s proposal of natural near-Earth dust aggregates (TRN-E020) keep a near-Earth reflective population plausible in principle, so the near-Earth interpretation cannot be rejected on present evidence.

A sunlight-reflecting near-Earth population makes a specific geometric prediction — avoidance of Earth's shadow at geosynchronous altitude — that is mixed in the working catalog and absent in Watters's vetted, non-uniform-background treatment. Simultaneous multi-point flashes would be awkward for most distant-source models if the marks were photons, but that premise is TRN-C020 and is not established. Strongest opposing consideration: the near-Earth reading currently rests on the least-vetted sample and on a uniform-sky null the plates visibly violate, so the geometry has not been shown to survive independent catalog construction.

The only geometric evidence is the Earth-shadow deficit, which replicates within the working catalog (TRN-E010) and reportedly strengthens among ML-scored candidates (TRN-E015), but vanishes under proper non-uniform nulls in vetted subsets (TRN-E011) — the signature currently lives in the least-vetted data, which is the strongest opposing consideration. In its favor, multiple simultaneous flashes across tens of arcminutes are hard for distant-source explanations, but only conditional on TRN-C020, the unresolved crux. Cann's dusty-plasma proposal (TRN-E020) keeps the near-Earth-natural space live, which would support this rung while undercutting artificiality. Genuinely open, pending a preregistered cross-catalog shadow test.

The near-Earth reading rests almost entirely on the Earth-shadow deficit, which replicates only conditional on the working catalog (TRN-E010, TRN-E015) and disappears once the non-uniform background is modeled and vetted subsets are used (TRN-E011). It also inherits the unresolved status of TRN-C020: geometry computed over marks of unknown origin is geometry over an unknown population. The genuinely strong argument in its favour is structural rather than statistical — nine point sources within ~10 arcmin on one exposure, and a 16-mag triple gone within 50 minutes, are hard to place at astrophysical distances if they were light at all. That argument's force is entirely conditional on a premise the file does not supply, and TRN-E020 shows the rung is also compatible with natural near-Earth reflectors rather than the case's headline reading.

Depends wholly on TRN-C020 plus shadow geometry that fails the corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails the corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistration is a methodological credit but the result is weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and inside the same contested catalog lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

The near-Earth attribution relies heavily on the Earth-shadow deficit, which replicates on the contested working catalog but fails in subsets subjected to stricter vetting for spatial background artifacts. The strongest opposing consideration is that the geometric signatures of orbit collapse once uniform-background assumptions are replaced with the data's actual plate-edge cluster patterns.

The primary geometric support is the reported Earth-shadow deficit (TRN-C005), which does replicate on the VASCO working catalog and appears stronger after ML artifact scoring, and multiple simultaneous point events would, if real flashes, be hard to reconcile with many distant-source scenarios. However, Watters et al. show that the background is highly non-uniform and edge-dominated, and find no significant deficit in vetted subsets; Hayes’s independent catalog with explicit exposure modeling also fails to recover an independent nuclear correlation, suggesting catalog definition strongly shapes population signatures. Since TRN-C020 (incident light) is itself unresolved and the ensemble evidence is mixed and lineage-bound, a near-Earth reflective population cannot presently be established. The strongest opposing consideration is this clear dependence of the signal on one unvalidated working catalog and on modeling choices.

The only geometric signature claimed — an Earth-shadow deficit at geosynchronous altitude — replicates inside the VASCO working catalog and fails once spatial non-uniformity is modelled or the candidate set is more heavily vetted. Simultaneous multi-source flashes across tens of arcminutes would argue against most distant-source explanations, but only if TRN-C020 is already true, which it is not shown to be. The strongest opposing consideration is Watters et al.: the deficit is computed against a uniform background the data visibly violate, and it is absent in the subsets vetted hardest against plate and scan artifacts.

The geometric logic is genuine: multiple simultaneous flashes across tens of arcminutes, if real, resist most distant-source explanations, and the Earth-shadow test is exactly the right kind of discriminant. But the shadow deficit replicates only on author-supplied rows (TRN-E010), fails when the background's demonstrated non-uniformity is modeled and vetting is tightened (TRN-E011), and the premise — that the flashes are real, TRN-C020 — is itself open. The strongest opposing consideration is TRN-E015's finding that the deficit persists and strengthens among ML-scored likelier-real candidates, a real internal-robustness result; it cannot overcome training labels made by expert judgment rather than physical ground truth, inside the same contested lineage.

The near-Earth reading rests almost entirely on the Earth-shadow deficit, and TRN-E011 shows the deficit's expectation was computed against a uniform background the data visibly violate, with no significant deficit surviving in more heavily vetted subsets; TRN-E012 answers with a power argument rather than a counter-analysis. Against that, TRN-E010 and TRN-E015 do reproduce the deficit within the working catalog, and it reportedly strengthens among ML-scored likely-real candidates. The strongest opposing consideration for a negative verdict is that the geometry is the one prediction plate defects cannot make, so a clean replication would be near-decisive — but no independently constructed catalog has been tested (TRN-R005 is unrun), and the claim is additionally gated by TRN-C020.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

Depends wholly on TRN-C020 plus shadow geometry that fails a corrected null and independent extraction (TRN-E009, TRN-E011). TRN-E020's preregistered result is a methodological credit but weak, unreplicated, and same-lineage.

The supporting geometry (Earth-shadow deficit) replicates only conditional on the VASCO working catalog (TRN-C005) and is disputed in vetted subsets; multiple simultaneous point flashes across tens of arcminutes do argue against most distant-source explanations if the flashes were real — but that premise (TRN-C020) is exactly what is unresolved.

The hypothesis of a near-Earth reflective population relies primarily on the reported Earth-shadow deficit and declination excesses in the VASCO working catalog. However, independent critical evaluation has shown that the shadow deficit is an artifact of non-uniform spatial defect distributions and plate-edge concentrations, disappearing under an appropriate null distribution. The strongest opposing consideration is that the deficit reproduces across independent statistical code on the supplied catalog and reportedly strengthens when applying machine-learning quality filters that exclude obvious artifacts.

Support is indirect and conditional: the shadow deficit reproduces under independent code on the authors' rows (TRN-E010), persists under ML artifact scoring (TRN-E015), and the low-declination excess (TRN-E058/E059) is the signature a GSO-like belt would give, while TRN-E074's one-third-solar-reflection inference is explicitly an assumption. Against it, Watters derives no significant deficit once the false uniform-random background null is corrected (TRN-E011), and the whole inference is conditional on TRN-C020, which is unresolved — if the marks are defects there is nothing to attribute. Strongest opposing consideration: TRN-E011's corrected-null null result, which the proponents' rebuttal answers with a power argument rather than the background-model charge.

The reproducible Earth-shadow statistic and TRN-E074's corresponding reflection inference provide some evidence for a sunlit near-Earth population, while the Hamburg coma result supports the broader premise that at least some analogous marks may be optical images. The altitude modelling is conditional on those disputed catalog rows and is not a direct distance measurement. The strongest opposing consideration is TRN-E011, which obtains no significant shadow deficit under a non-uniform spatial null, compounded by the unresolved flash-versus- emulsion question at the preceding rung.

The near-Earth glint reading rests on the Earth-shadow deficit and related geometry, which reproduce on the author catalog and strengthen under ML filtering yet vanish or lose significance under Watters’ non-uniform-background null and are undefined on Hayes’ independent extraction. The claim also presupposes TRN-C020, which remains open. Strongest opposing consideration: schedule, spatial-heterogeneity, and catalog-composition artifacts can manufacture an apparent shadow deficit without any near-Earth population.

The claim's principal geometric support is the Earth-shadow deficit, which replicates on the authors' own rows (TRN-E010) and strengthens under their ML filter (TRN-E015, TRN-E052) but loses significance once the uniform-background null is replaced (TRN-E011), and whose magnitude and preferred altitude shift with sample definition and post-hoc spatial cuts (TRN-E025, TRN-E051). It also inherits TRN-C020: if the marks are emulsion defects, no reflection geometry is needed. The strongest opposing consideration for a purely sceptical reading is that a shadow deficit is a genuinely non-obvious prediction of the near-Earth hypothesis and no defect mechanism has been shown to produce an antisolar-angle dependence — but nor has the background been shown uniform enough for the deficit to be measured.

Rests on TRN-C005 (mixed) and on TRN-C020 (unresolved). TRN-E074 records that the modelling programme assumes real Earth-orbit sources and infers a one-third solar-reflection fraction from the shadow deficit — an elaboration, not a test. TRN-E020 offers a natural route to the same geometry. TRN-E021 takes reality as premise. No record tests near versus far directly; the parallax test is infeasible for the canonical events (TRN-E024).

TRN-E074 (weak, supports) infers a one-third solar-reflection fraction from the shadow deficit while explicitly assuming real orbital sources; TRN-E020 (weak, qualifies) and TRN-E021 (context) take the reality of the flashes as premise. The claim stands or falls with TRN-C005, which is mixed, and with TRN-C020, which is unresolved.

The attribution of plate transients to near-Earth solar reflections depends primarily on the Earth-shadow deficit, which dissolves when analyzed with proper non-uniform background models and independent candidate extraction. Furthermore, it has not yet been demonstrated that these marks were produced by incident optical light rather than localized emulsion flaws. Strongest opposing consideration: The persistence of a statistically significant shadow deficit across multiple assumed altitudes and after machine-learning quality filtering within the proponent working catalog.

The geometric case (Earth-shadow deficit) replicates as a statistic on the original rows (TRN-E010) and survives ML artifact scoring (TRN-E015), with one environmental confounder measured and closed (TRN-E017); but the comprehensive skeptical reanalysis finds no deficit once the proper null is used and shows the background is spatially non-uniform (TRN-E011), and the proponents' own modelling preprint concedes the deficit magnitude is definition- and cut-dependent (TRN-E051). Everything is additionally conditional on TRN-C020, which microscopy explicitly cannot settle. Strongest opposing consideration: TRN-E011's corrected null on closely related datasets.

TRN-E010 reproduces the Earth-shadow deficit on the supplied catalog, and TRN-E015 reports that it strengthens among candidates scored less artifact-like; these are genuine but lineage-dependent support for sunlight-sensitive objects. TRN-E011 obtains no significant deficit after replacing the uniform background with a spatially heterogeneous null, and no independent extraction has reproduced the effect. The strongest opposing consideration is therefore that the principal geometric signal may be a property of candidate selection and plate position rather than of near-Earth reflectors.

The principal geometric support is the Earth-shadow deficit on the working catalog (E010, E015, E025), yet the same deficit is absent once spatial non-uniformity is modeled (E011) and the nuclear correlation that is sometimes paired with it fails under independent extraction (E009). All altitude and hotspot elaborations assume the deficit is astrophysical. Strongest opposing consideration is that every positive ensemble result remains inside the contested 107875-row lineage while the decisive independent-catalog tests are null or unperformed.

The main near-Earth-specific evidence is the shadow deficit (C005, E025, E074); the geomagnetic pattern (C103, E020) is thinner still. The deficit loses significance under a null that allows for an uneven background, tested on related data (E011). It has never been computed on an independently extracted catalog. The shell fits are poor and prefer 20,000 km to geosynchronous altitude (E055), and C020 remains unresolved beneath it. Strongest opposing consideration: the deficit has the required sign, survives centre-of-plate and ML cuts (E010, E015), and no record explains why defects would avoid the umbra.

The only attached support, TRN-E074, is the modelling preprint's estimate that a third of the sample is solar reflection. That estimate is inferred from the shadow deficit the paper assumes real. TRN-E020 qualifies by offering a natural near-Earth route. TRN-E021 is context that presumes the flashes real. The claim therefore rests on TRN-C005, which is mixed, and on TRN-C020, which is unresolved.