The Eight-Planet Photograph

At pp. 162–163 the book stops arguing and asks a question: “If we fully simulated the Solar System on a computer for Feb 22, 2025, would we be able to reproduce the eight-planet photo using the same coordinates? … How closely would the simulation match reality?” It is a fair question, it names a date and a criterion, and it accepts in advance that a close match would count for something. So we ran it. But the answer that matters here is not the one the question asks for.

What the claim is

The chapter is Unknown Luminaries and its thesis is at p. 161: “In reality, we have no idea what these luminaries in the sky really are or how they work.” The question at p. 162 is offered as a test of it. Earth is the eighth planet; the photograph carries the other seven.

The catch, which we think is real

Before answering, it is worth noticing where the question sits. Three pages earlier, at pp. 155–157, the book establishes that planetary imaging is a processing pipeline — lucky imaging, frame stacking, wavelet deconvolution — under the heading “How to manipulate a photo to achieve the desired planet 101.” Two pages later, at p. 164, the reader is told that “a brief introduction to geocentrism is highly recommended before continued reading.”

The question is set between those two, and answering it carelessly walks into both. Say yes, it matches perfectly, and the reply is available on either side: that you have matched a composite image, or that a geocentric model produces the same sky and so the match decides nothing. Neither reply is unfair. We would rather concede them at the top than have them sprung.

What the photograph actually is

The QR at p. 163 resolves to a Live Science report on an image by the astrophotographer Josh Dury, taken from the Mendip Hills in Somerset just after sundown on 22 February 2025. Dury describes his method openly: the picture is a panorama of several panes, each pane captured in multiple exposures, shot with HDR settings, and he used astronomy software to locate the fainter planets.

None of that is a criticism of him. It is the only way the picture could exist, and he said so unprompted. But it does mean the photograph cannot serve as an independent check on an ephemeris, because an ephemeris is part of how it was made. Comparing the simulation to this image tests the stitching, not the sky. That is the book’s point, it is a good one, and it is the reason this page does not lead with the match.

Why it had to be built that way

Computed for the Mendip Hills at 17:50 UT, twelve minutes after sunset:

The Sun set at 17:38 UT and Mercury followed it at 18:37. The entire window in which all seven stood above that horizon was fifty-nine minutes, and it ran through civil and nautical twilight, with the two faintest members never visible to an unaided eye at all. A single exposure could not hold Venus and Neptune together, and no exposure could find Neptune without being told where to point.

So: does it match?

Yes. Positions computed from JPL’s DE440 ephemeris, apparent geocentric — corrected for light-time and aberration, so where the planets are seen rather than where they are — reproduce the arrangement. The chain outward from the sunset point:

All seven fall within 5.6° of a single great circle. Note the ordering: outward from the Sun it runs Mercury, Saturn, Neptune, Venus, Uranus, Jupiter, Mars — nothing like their order by distance. That scrambling is not a defect in the model; it is what the model predicts, because what you see is the arrangement in longitude from a vantage point inside the system, not a cross-section of it.

His question also assumes the camera’s coordinates are a necessary input, and for half of it they are: where the line sits relative to the horizon, and its tilt, depend entirely on latitude, longitude and clock. The separations between the planets do not. Recomputed from London, Sydney and Quito, no pair separation moves by more than 25 arcseconds. Parallax is the only term that cares where you stand, and at these distances it is negligible.

The check anybody can run

None of this needs an ephemeris library. Open a planetarium program, set it to the Mendips on the evening of 22 February 2025, and look. Doing that from Priddy at 18:39:23 UT, against our own computation for the same instant and place:

The caveat matters more than the agreement: a planetarium’s planetary positions descend from the same ephemeris lineage as ours, so this tests our arithmetic and not the ephemeris. It is not independent corroboration and we do not offer it as such.

Two things fall out of it that bare agreement would not have shown. The first is that 18:39 is two minutes after Mercury set — computed altitude −1.0°, and the program draws it sitting on the horizon where a refracted setting object sits. An evening’s casual look a quarter of an hour later would have found six planets, not seven. The window is real, and it is short.

The second is that at default settings the program does not label Uranus or Neptune at all, though both stand on the line at +53.9° and +10.5°. The two faintest members of the “seven planets” are objects a planetarium declines to draw for a person standing there — which is the same fact that made the composite necessary in the first place.

What the match is worth, stated plainly

Less than it looks. A photograph records directions, and directions are the part of an arrangement that more than one arrangement can share. Matching this frame does not establish where anything is — only that where everything would appear was known in advance. Anyone offering the picture as proof of a heliocentric solar system is overclaiming, and the book is entitled to say so.

What it does settle is the chapter’s title. A model with far more observational constraints than free parameters, tabulating positions years ahead to a fraction of a planet’s apparent width, is not a description of something unknown. The luminaries may be mysterious in composition; they are not mysterious in behaviour, and “we have no idea how they work” is not a statement one can make about something predictable.

The version of his test that does decide

Since the arrangement is fixed everywhere on Earth, every difference between two observers is orientation — which is precisely where a flat plane and a globe part company. Two frames, same instant, 22 February 2025 at 18:45 UT:

Largest disagreement between the two sites in any pair separation: 21 arcseconds. Identical arrangement; 149° apart in azimuth.

What would change our mind

• The two-hemisphere frames, disagreeing. A matched pair, same instant, showing Jupiter in the same compass direction from both sites. That would be a real result and we would publish it.
• The composite’s panes falling outside the window. If the constituent exposures are timestamped outside 17:38–18:37 UT, the “all seven at once” framing does not hold — a criticism of the photograph the book could have made and did not.
• A demonstration that the ephemeris was tuned to the answer. Recomputing this date on DE200 or DE405, fitted decades before 2025, should leave the arrangement unchanged. If it does not, the framing above weakens and this page must say so.
• A luminary breaking the elongation rule. Mercury and Venus never exceed about 28° and 47° from the Sun; Mars and Saturn reach 175°. One documented sighting of Venus at opposition and the structure behind all of this is gone.
• To do: complete this list.

Where this page could be wrong

• The published description of the planet order may not match ours. The Live Science account lists the sequence differently from the ephemeris in the Saturn–Neptune–Venus section, where Venus and Neptune sit at almost the same azimuth but very different altitudes and “order along the line” is genuinely ambiguous. This needs resolving against the image rather than against the caption.
• DE440 is a fit, not a revelation. It is adjusted to radar ranging, spacecraft tracking and centuries of optical observation. The claim is that it has far more constraints than parameters and predicts events it was never shown — not that it was derived from first principles. The older-ephemeris check would demonstrate that, and we have not run it.
• The two-hemisphere test has not been performed. It is proposed here, not reported.
• To do: the pane timestamps, and a measurable copy of the frame.

Sources & further reading

• Levi Miller, Globe Deconstruction? (prerelease) — pp. 162–163 (the question and the QR), p. 161 (the chapter thesis), pp. 155–157 (the imaging-process argument this page treats as load-bearing), p. 164 (the geocentrism pointer).
• Live Science, “Planet parade photo captures 7 planets in a line over Earth” — the target of the p. 163 QR, carrying Josh Dury’s account of how the composite was assembled. livescience.com
• Stellarium Web (stellarium-web.org) — used for the independent-of-our-code sky check from Priddy, Somerset. Screenshots of it are not reproduced here: the program’s own guidance permits non-profit use with credit, but the photographic landscape textures are third-party assets carrying separate licences, so the comparison is given as a table instead.
• JPL Development Ephemeris DE440 (Park et al., 2021), via the Skyfield library. Apparent geocentric positions for 2025-02-22; observer at 51.28°N, 2.72°W, 300 m.
• Cross-references on this site: Jupiter’s Shadows (the other specific argument in this chapter), Celestial Globes Exposed, The Sky Turns at One Rate, Is the Moon a Ball?.