Fig. A1 — “The Blue Marble,” December 7, 1972. Credited to the Apollo 17 crew. Hasselblad 70mm camera, approximately 29,400 km from Earth. (NASA / Public Domain)

Taken on December 7, 1972, during the translunar coast phase of the Apollo 17 mission, at an altitude of approximately 29,400 km (18,300 miles). The crew used a Hasselblad 70mm film camera with an 80mm lens. Because the spacecraft was oriented with the Sun behind them, the entire illuminated face of Earth was visible in the frame — the only time this geometry has been achieved by a crewed mission. NASA describes it as “one of the most reproduced photographic images in history” and the “first photograph in which Earth is in full view.”

• Unverifiable Photographer Identity The photograph’s official credit reads “Apollo 17 crew” rather than a named individual. Both astronauts Jack Schmitt and Ron Evans claimed to have taken the shot; NASA’s own records list the photographer as unverifiable. Skeptics argue that in any genuine photographic event, the photographer would be identifiable — and that the confusion suggests a later assembly rather than a spontaneous snap.
• Implausibly Perfect Composition The photograph centers Africa and the Indian Ocean almost exactly in the middle of the frame, with Antarctica clearly visible at the bottom and cloud formations arranged in a visually symmetrical, aesthetically pleasing distribution. Skeptics argue this level of compositional precision — Africa centered, poles visible, cloud patterns balanced — is statistically improbable for a handheld shot taken without a viewfinder from a rotating spacecraft. The framing resembles that of a globe model viewed from a set distance.
• Ocean Color and Tone Anomalies The original negative reveals oceans with a muted, grayish-blue quality inconsistent with the vivid deep blue of modern Earth imagery. Some skeptics argue this is evidence of post-processing on the negative itself or that the image was color-corrected in ways not disclosed. Others contend the color difference simply reflects the Ektachrome film stock of the era and the atmospheric scattering of light at that geometry.
• Antarctica’s Unprecedented Clarity The South Polar ice cap is visible in extraordinary detail. Meteorological records from December 7, 1972, indicate significant cloud cover over Antarctica at the time. Skeptics argue that the ice cap should not be so clearly visible — and that its crisp appearance is consistent with a composite or an idealized model, rather than a photograph of actual atmospheric conditions.
• The “Only One” Argument Skeptics note that for a moment described as a historic photographic achievement, only a very small number of frames were taken. The ubiquitous version of the image is one of only a handful of exposures. If a crew of professional astronauts had a clear view of the fully lit Earth — an event never previously achieved — the argument is made that they would have taken far more exposures.

Fig. B1 — “Blue Marble 2002,” released January 2002. Composite assembled by Robert Simmon from MODIS / Terra satellite data. (NASA / Public Domain)

The 2002 Blue Marble is a composite assembled by Robert Simmon, a data visualizer at NASA’s Earth Observatory. The cloud-free land surface data was collected over four months by the MODIS (Moderate Resolution Imaging Spectroradiometer) instrument aboard NASA’s Terra satellite. Cloud data was composited from two days of visible-light imagery and one day of thermal infrared imagery over the poles. The land surface, ocean color, sea ice, and cloud cover were combined into a single image and then mapped onto a sphere. The black background surrounding the Earth is a solid color fill; it is not a photograph of space. The atmosphere’s blue halo was created using Photoshop’s Gaussian blur tool.

The image was adopted by Apple as the default iPhone lock screen in 2007 and was seen by hundreds of millions of people with no context identifying it as a visualization rather than a photograph. Robert Simmon later confirmed in the Quartz interview that this was never intended to be concealed, but acknowledged the distinction was not made explicit to the public.

• Duplicated Cloud Formations — Confirmed by Creator This is the most extensively documented anomaly. Online analysts — using tools such as image overlay, contrast enhancement, and pattern-matching — identified multiple cloud formations that appear verbatim in different parts of the image. These duplicates are consistent with the use of Photoshop’s clone-stamp or copy-paste tools to cover data gaps left by the satellite’s orbital gaps. Robert Simmon confirmed in the Quartz interview that clouds were “collaged together using Photoshop’s clone tool to cover gaps in the satellite’s coverage.” This confirmation is cited as the single most powerful piece of evidence in the entire “Fake Blue Marbles” theory — not because it proves fabrication, but because it establishes that the image contains deliberate digital manipulation that was not disclosed to the public for years.
• “Photoshopped Because It Has to Be” — The Admission Simmon’s phrase is treated by skeptics as an inadvertent disclosure. The argument is not that Simmon confessed to making up a fictional planet, but that the phrase “it has to be” implies that there was no other way to produce such an image — i.e., that a true photograph of the full Earth from space was not available in 2002. Skeptics use this to argue that the public was for decades presented with a fabricated image in place of a real photograph, and that this substitution was institutional policy.
• Atmosphere as “Photoshop Blur” The thin blue atmospheric halo visible around the Earth in the 2002 image is not derived from any camera data. Simmon confirmed it was produced using Photoshop’s blur effect. Skeptics argue that if the image were a genuine photograph, the atmosphere would appear naturally. The deliberate addition of a simulated atmosphere is cited as evidence that the surrounding visual elements were constructed to look like a photograph rather than derived from observation.
• Background Space Is a Solid Fill The deep black void surrounding the Earth in the 2002 image is not a photograph of space. Simmon confirmed it was a solid color painted onto the canvas in Photoshop. Skeptics note that a true photograph taken from orbital altitude would capture the actual visual environment — stars, faint stars, cosmic background — not a perfectly uniform black.
• Terra’s Orbital Constraints Make a Single-Frame Earth Impossible The Terra satellite orbits at approximately 705 km altitude. At this altitude, the satellite’s field of view covers approximately 2,330 km per swath — roughly one sixth of the Earth’s circumference. It is physically impossible for the MODIS instrument to photograph an entire hemisphere in a single exposure. The entire image is necessarily a mosaic of many passes assembled over time. Skeptics argue this technical constraint was not communicated to the public, who were left to assume the image was a photograph.
• North America’s Size Discrepancy Across Versions Analysts have noted that the apparent size of North America relative to Africa differs between the 2002 image and other Blue Marble versions. On a true globe photographed from a fixed distance, the proportions of continents should remain consistent. Skeptics argue the discrepancy reveals the use of different base sphere models or projections rather than consistent photographic documentation of the same planet.

Fig. C1 — “Blue Marble: Next Generation,” 2005. MODIS / Terra & Aqua. 500-meter resolution composite. (NASA Earth Observatory / Public Domain)

Released in 2005 as an update to the 2002 image, “Blue Marble: Next Generation” offers 500-meter-per-pixel resolution — a significant improvement — and provides twelve distinct monthly versions to show seasonal changes in vegetation, snow cover, and sea ice. The land surface data came from MODIS instruments aboard both the Terra and Aqua satellites. Cloud cover is, again, a composite layer assembled from multiple days of satellite passes. The image was produced by Reto Stöckli at NASA’s Goddard Space Flight Center.

• Persistence of Duplicated Cloud Patterns Community analysts identified the same pattern of duplicated cloud formations in the 2005 image as were found in the 2002 version — a finding interpreted as evidence that the same Photoshop clone-stamp methodology was applied. If the image were a fresh capture of real atmospheric conditions in 2005, the cloud formations should bear no resemblance to those in the 2002 image. The apparent re-use of cloud tile elements across versions is treated as a smoking gun for institutional recycling of fabricated imagery rather than fresh photographic capture.
• “Copy-Pasted Segments” — The Eric Dubay Claim Flat-Earth theorist Eric Dubay — one of the most prominent figures in this intersection of conspiracist communities — published annotated versions of the 2005 image claiming to identify rectangular copy-paste segments visible as subtle tonal seams across the cloud layer. This claim was widely circulated. Independent image analysts have both partially confirmed and disputed the claim: some seams are artefacts of satellite data mosaicking (which NASA acknowledges), while others appear consistent with manual digital editing. The distinction between the two is technically complex and has not been definitively resolved.
• Resolution Implausibility Some skeptics argue that a 500-meter-per-pixel resolution composite assembled from a satellite at 705 km altitude — producing imagery covering the entire globe — represents a data density implausible for legitimate satellite imaging in 2005. This argument is less widely held and is disputed by satellite imaging professionals who note the MODIS instrument’s documented specifications.
• Twelve-Version Monthly Series Anomaly The release of twelve monthly versions of the composite raised questions among skeptics: if each version represents real satellite data from a given month, why do certain geographical features — particularly vegetation boundaries in contested regions — appear to shift in ways inconsistent with known seasonal patterns? Analysts mapped the vegetation index changes across the twelve versions and identified several regions where the seasonal shift appeared to exceed what remote-sensing data for that period supported.

Fig. D1 — Western Hemisphere version. Released January 2012. (NASA / Public Domain)

Fig. D2 — 2002 Blue Marble, shown for side-by-side comparison. Note differences in North America’s apparent scale. (NASA / Public Domain)

Released in January 2012 and composed of data from the VIIRS (Visible Infrared Imaging Radiometer Suite) instrument aboard the Suomi NPP (National Polar-Orbiting Partnership) satellite. The Suomi NPP satellite is a joint project of NASA, NOAA, and the United States Department of Defense. The composite is 8,000 × 8,000 pixels — 64 megapixels — and was assembled from data collected over January 4, 2012. The image was published in two versions showing the western and eastern hemispheres, and became one of the most viewed images on Flickr at the time of release.

• North America’s Disproportionate Scale This is the most widely cited anomaly in the 2012 image. When the western hemisphere version is compared against the 2002 Blue Marble and against standard geodetic measurements, North America appears significantly larger relative to other landmasses than it should be. Skeptics have produced annotated comparisons showing the apparent width of North America in the 2012 image exceeding its proportion in the 2002 image by a margin inconsistent with any change in observation altitude or lens geometry. The argument is that different CGI sphere models were used across different Blue Marble versions — models with different scaling parameters applied to different regions.
• Implausibly Uniform Cloud Coverage The 2012 image presents a cloud layer that many meteorological analysts describe as implausibly uniform in texture. Real cloud systems have fractal complexity and visible differentiation between cloud types, altitudes, and densities. The cloud layer in the 2012 image appears, to some analysts, to have a processed or rendered quality — smoother, more uniform in pixel-level noise structure — than would be expected from raw VIIRS sensor data.
• Department of Defense Involvement The Suomi NPP satellite is a tri-agency project, with the Department of Defense as a partner. Some skeptics argue that the involvement of the US military in the production of the satellite imagery introduces a chain of custody concern — that the image could be modified at the DoD processing stage without NASA’s knowledge or disclosure. This argument is more speculative and represents a minority position even within the skeptical community.
• Single-Day Data Claim Under Scrutiny NASA states the 2012 image was assembled from data collected on January 4, 2012. The Suomi NPP satellite completes approximately 14.5 orbits per day. Skeptics calculate that 14.5 orbits at the satellite’s swath width would not be sufficient to provide seamless full-hemisphere coverage in a single day, and that some data gaps would be expected. The seamless appearance of the composite is cited as evidence of gap-filling beyond what is officially acknowledged.

Fig. E1 — DSCOVR / EPIC image, July 6, 2015, taken from Lagrange Point 1, approximately 1.6 million km from Earth. (NASA / Public Domain)

The DSCOVR (Deep Space Climate Observatory) satellite was launched on February 11, 2015, and positioned at the Sun–Earth Lagrange Point 1 (L1) — approximately 1.5 million km (1 million miles) from Earth, on the line between Earth and the Sun. From this vantage point, the EPIC (Earth Polychromatic Imaging Camera) instrument can photograph the fully sunlit face of Earth in a single exposure. NASA and NOAA described the July 6, 2015, image as “the first picture of the entire sunlit side of Earth since the Apollo 17 photograph taken December 7, 1972.” Images are released publicly with an approximately 12-to-36-hour delay from capture.

DSCOVR originated as “Triana,” a project proposed by then-Vice President Al Gore in 1998, intended to provide a continuous live feed of Earth from the L1 point. The satellite was assembled but shelved by the George W. Bush administration in 2001; it was revived by the Obama administration and launched in 2015.

• The “First in 43 Years” Admission — Retroactive Condemnation of All Intervening Images This is, for many theorists, the single most significant piece of evidence across the entire Blue Marble file. NASA’s own press release described the DSCOVR image as the first full-disc single-exposure photograph of Earth since 1972. Skeptics argue that this statement constitutes an institutional admission that the 2002, 2005, and 2012 images — which were presented to the public as images of Earth — were not photographs. The community’s position is that the public was for 43 years shown fabricated visualisations in place of photographs, and that this substitution was not adequately disclosed.
• Color Discrepancy Between DSCOVR and Apollo 17 Side-by-side comparison of the DSCOVR image and the Apollo 17 photograph reveals a significant color difference. The 2015 image appears notably bluer and more saturated than the 1972 original, which exhibits a greener, more subdued tone over the oceans. Skeptics argue that if both images are genuine photographs of the same planet taken from similar distances with similar lighting geometries, they should produce comparable color rendition — and that the chromatic divergence suggests one or both images have been color-processed beyond what would be expected for straightforward photography.
• The EPIC Camera’s Multi-Band Compositing Process The EPIC instrument does not produce true-color photographs in a single exposure. It captures ten separate narrow-band wavelength images in sequence, which are then combined into a single colour composite. The red, green, and blue channels are captured approximately 30 seconds apart. During those 30 seconds, Earth rotates approximately 0.2 degrees at the equator — a shift of roughly 22 km. Skeptics argue that the combination of out-of-time exposures, each at a different spectral band, means that even the DSCOVR image is a processed composite — not a single-moment photograph — and that this was not clearly communicated in NASA’s “first real photograph” framing.
• Political Origins — The “Al Gore Satellite” The satellite’s origin as a personal initiative of Al Gore — a politically divisive figure in the United States — and its decade-long suppression by a subsequent administration, followed by its revival under a different administration, is cited by some skeptics as evidence that the imagery produced by DSCOVR serves a political function. The argument is that a satellite whose deployment was determined by political factors cannot be presumed to produce politically neutral imagery. This argument represents a more speculative position and is noted without independent evidentiary support.
• The Delay in Release and Real-Time Feed DSCOVR images are not released in real time. There is a 12-to-36-hour processing delay between capture and public release. Skeptics note that this window is sufficient time for processing decisions — including potential alteration — to be made before the image enters the public record. The original “Triana” proposal called for a live feed; the implementation with a processing delay is characterised by some analysts as a structural vulnerability to manipulation, regardless of whether that vulnerability has been exploited.

Fig. F1 — “Hello, World.” Photographed by Commander Reid Wiseman aboard the Orion spacecraft, April 2–3, 2026, approximately 9,889 km from Earth’s surface. Nikon D5, 14–24mm f/2.8 lens at 22mm, 1/4 sec, f/4, ISO 51,200. (NASA / Public Domain)

On April 1, 2026, NASA launched the Artemis II mission aboard the Orion spacecraft — the first crewed lunar mission since Apollo 17 in December 1972. The crew of four astronauts — Commander Reid Wiseman, Pilot Victor Glover, Mission Specialist Christina Koch, and Canadian Space Agency Mission Specialist Jeremy Hansen — completed a translunar injection burn approximately one day after launch, placing the spacecraft on a free-return trajectory toward the Moon. Shortly after that burn, on April 2–3, 2026, Wiseman photographed Earth through one of Orion’s four-sided windows using a Nikon D5 DSLR camera. The image, titled “Hello, World,” was downlinked to Earth and published by NASA on April 3, 2026.

NASA described the image as among the first photographs of a full Earth disk taken by a human since the Apollo 17 “Blue Marble” in 1972 — a designation that immediately positioned it within the Blue Marble lineage and, by extension, within the existing analytical framework of this dossier. The photograph is notable for several technical reasons: it was taken at night, with Earth illuminated by reflected moonlight rather than direct sunlight, requiring an extreme ISO of 51,200 — roughly nine stops more sensitive than the Ektachrome film used in 1972. Two auroras are visible at the top right and bottom left of the frame, and zodiacal light glows in the lower right as Earth eclipses the Sun. Africa dominates the center of the frame, with the Sahara clearly visible and the Iberian Peninsula at lower left. City lights — including Madrid — are discernible in the darkness.

The mission proceeded to a lunar flyby on April 6, 2026, during which the crew captured additional images of Earth from deep space — including an “Earthset” photograph showing a crescent Earth setting behind the Moon’s limb, deliberately evoking the 1968 Apollo 8 “Earthrise.” The Artemis II crew set a new distance record for humans from Earth during the flyby before splashing down on April 10, 2026.

Exhibition F introduces an anomaly category not present in any prior Blue Marble entry: the mass production of AI-generated fake imagery falsely attributed to the Artemis II mission. Within days of launch, social media platforms were flooded with images purporting to show Artemis II photographs — depicting Earth through a five-sided spacecraft window (the real Orion windows have four sides), the lunar Orientale Basin with Earth in the background, astronauts in front of what appeared to be a green screen, and a recreation of the 1968 Apollo 8 “Earthrise” image with identical cloud patterns relabelled as Artemis II imagery. Fact-checking organisations including Full Fact and Lead Stories confirmed these viral images were AI-generated, identifying SynthID watermarks — an embedded digital signature used by Google’s Gemini AI — in multiple circulated photographs.

This creates a documented analytical complication specific to the AI era: the most viral “evidence” of Artemis II being fake was itself fabricated. Commentators and digital forensics experts noted the irony — AI-generated fakes were being deployed to argue that genuine NASA imagery was fake. UC Berkeley image-verification expert Hany Farid described the situation as emblematic of the new epistemic environment: a world in which the ease of fabricating photographic evidence makes it simultaneously easier to fake events and easier to dismiss real events as fake. One survey cited in coverage found that 82 percent of respondents reported their confidence in media had decreased as a result of AI-generated content. The community of analysts working on the “Fake Blue Marbles” theory now faces a methodological challenge it did not previously have: distinguishing between anomalies in official NASA imagery and anomalies in AI-generated content falsely attributed to NASA.

• The “Too Similar to Blue Marble” Objection — Africa-Centered Composition The most immediately noted and widely discussed anomaly is the striking compositional similarity between the 2026 “Hello, World” image and the 1972 Apollo 17 “Blue Marble.” Both images feature Africa centered in the frame. Both images show the South Pole oriented toward the top. Both images were taken at approximately similar distances — 29,400 km in 1972, approximately 9,889 km in 2026 — from similar angular perspectives. Online communities rapidly noted that out of all possible orientations of Earth visible from a spacecraft traveling toward the Moon, the odds of obtaining an Africa-centered, south-pole-up composition matching the most iconic Earth photograph in history are vanishingly small. The question asked by the community: did NASA deliberately orient the spacecraft, select the timing of the photograph, and choose this specific image from many exposures because it visually echoes the 1972 original — or was the geometric coincidence genuine? NASA itself drew explicit attention to the similarity by posting a side-by-side comparison of the two images on its official social media channels, captioning it: “We’ve come so far in the last 54 years, but one thing hasn’t changed: Our home looks gorgeous from space!” Skeptics argue that this institutional promotion of the similarity, far from being innocent pride, suggests the resemblance was engineered for symbolic effect rather than discovered by chance.
• Color and Luminosity Discrepancy — “Earth Looks Dirtier” A widespread public reaction to the “Hello, World” image was that Earth appears notably darker, more muted, and less vividly blue-green than in the 1972 and 2002 Blue Marble images. The reaction spread across social media platforms, with users comparing the two images side-by-side and characterizing the 2026 version as showing a “dirty,” “washed-out,” or “duller” planet. Some commentators initially attributed this to atmospheric pollution or climate change — which generated significant secondary coverage. Within the skeptical community, the color discrepancy is framed differently: if multiple photographs of the same planet, taken at broadly similar distances under similar geometry, produce dramatically different color profiles, the question arises as to whether consistent photographic documentation of a real object is being shown, or whether different production processes — different CGI models, different image-processing pipelines, different atmospheric simulations — are generating images with insufficiently consistent output.
• Extreme ISO and the “Processed to Look Like Day” Argument The “Hello, World” image was taken at ISO 51,200 — an extreme sensitivity setting that, in terrestrial photography, would produce severe digital noise, color aberration, and loss of fine detail. The published image shows none of these artefacts. It presents as a clean, sharp, well-resolved image of Earth. The community notes two possibilities: either the Nikon D5’s noise-reduction capabilities at high ISO are so exceptional that they produce a clean image indistinguishable from a daytime photograph — which some photographers dispute — or the image has undergone significant post-processing, including noise reduction, sharpening, and colour correction, that transforms a genuinely dark, noisy nighttime exposure into something that reads as a pristine daytime image. The Guardian and other outlets noted that the image “has been enhanced to look like daylight.” NASA has not published the raw, unprocessed file — only the processed version — making independent assessment of the degree of post-processing impossible from public data alone.
• The Window Frame Absence — No Spacecraft Context The official “Hello, World” image presents Earth against a pure black background with no visible window frame, spacecraft structure, or other contextual element. Skeptics note that the Orion spacecraft’s windows are relatively small — described as four-sided rectangular portholes — and that photographing a perfectly circular Earth at 22mm focal length through such a window from the stated distance would require a precise framing that leaves the window frame entirely out of shot. Some community analysts argue that the complete absence of any spacecraft context is unusual for a genuinely handheld shot through a porthole, and that the composition suggests the Earth has been extracted and placed against a black background in post-processing. Comparison AI-generated fakes that circulated online ironically depicted Earth through a five-sided window — a detail inconsistent with the real Orion craft — while the official image shows no window at all.
• The “First Human Photo Since 1972” Designation — And the DSCOVR Contradiction NASA described the “Hello, World” image as among the first photographs of a full Earth disk taken by a human since 1972 — directly echoing the same framing used for the 2015 DSCOVR image, which was also described as the “first” full-disc Earth photograph since Apollo 17. The community notes an apparent contradiction: in 2015, NASA stated DSCOVR was the first full-disc Earth photograph since 1972. In 2026, NASA applied a near-identical “first since 1972” designation to the Artemis II image. Skeptics argue that the repeated deployment of a “first since Apollo 17” superlative — applied now to two different images a decade apart — suggests either that the claim is being used as a marketing device irrespective of its technical accuracy, or that the qualifications attached to each “first” are being shifted post-hoc to accommodate the new image. The fine print in each case involves distinctions between “full hemisphere,” “full disc,” “taken by a human,” and “single-exposure” — distinctions that the community argues are sufficiently elastic to be applied retroactively to whichever image NASA wishes to elevate.
• The Earthset Image — Too Perfect an Echo of Earthrise On April 6, 2026, during the lunar flyby, the Artemis II crew captured the “Earthset” image — showing a crescent Earth setting behind the Moon’s limb, deliberately designed to echo the 1968 Apollo 8 “Earthrise.” NASA had published simulations of what the Earthset would look like before the flyby, and the crew was explicitly tasked with recreating the iconic framing. Skeptics note that NASA’s release of pre-mission computer simulations of the expected image — followed by an official photograph that closely matches those simulations — raises the question of which came first: the photograph confirming the prediction, or a photograph produced to match the prediction. The community further notes that the Earthset image shows a crescent Earth of a specific angular size and phase angle that matches the pre-published simulation with a precision that some analysts describe as implausibly exact for a handheld shot taken through a spacecraft window during a seven-hour flyby.
• The Green Screen Claim — And Its Paradox Within days of the Artemis II launch, viral clips on TikTok and X purported to show the four astronauts floating in what appeared to be a harness system in front of a green screen. These clips spread rapidly and were cited as proof that the mission was being filmed in a studio. Fact-checkers including the BBC confirmed the clips were AI-generated — identified by SynthID watermarks from Google’s Gemini image generation system. The paradox documented here is distinct from the debunking: the green screen clips were so widely circulated and so visually convincing that they functionally contaminated the information environment around the Artemis II mission. Even after debunking, the association between “Artemis II” and “green screen” persisted in search results, social media comments, and online discussions. The community of analysts working on the “Fake Blue Marbles” theory now faces the documented reality that AI fabrications seeded in the environment around a genuine mission make it structurally harder to assess genuine anomalies — because the noise floor of fake “evidence” has been dramatically raised.