The Danjon Limit: Why Some Crescents Are Physically Impossible to See

In the world of crescent moon observation, there is a line in the sky beyond which no amount of magnification, camera sensitivity, or observer skill can make the hilal appear. This line is not drawn by clouds or atmospheric haze; it is carved by the physics of sunlight grazing the surface of the Moon and by the way the human eye struggles to separate a faint sliver from a bright twilight sky.

This boundary is called the Danjon limit, and understanding it is essential for anyone serious about lunar observation, Islamic calendar determination, or the science of celestial visibility. It is the hard floor beneath every prediction made by the Yallop and Odeh criteria, and it explains why some announced sightings can be ruled out before an observer ever looks up.

The Discovery: 1932 and 1936

The history of the limit unfolds across two dates that are often conflated. In 1932, French astronomer André Danjon published observations in the journal L'Astronomie noting a striking pattern among the youngest crescents on record: no observer had ever reliably reported seeing a crescent when the angular separation between the Moon and the Sun, the elongation (arc of light, or ARCL), fell below roughly 7 degrees. He reported the effect in 1932 and then quantified the limit in 1936, again in L'Astronomie, by plotting the measured length of the visible crescent arc against elongation.

This was not a statistical fluke or a coincidence of bad weather. Danjon argued that something physical, not merely meteorological, switched the crescent off as the Moon approached the Sun in the sky. What that "something" actually is remains, as we shall see, an open question. Danjon's work sits within a much longer history of crescent observation, from Babylonian records to the modern era, which is what first made the pattern visible to him.

What Danjon Actually Measured

Danjon's key result was not simply "you cannot see a crescent below 7 degrees". It was a quantitative trend. He measured the angular length of the visible crescent arc, the span from one cusp (horn) to the other, and found that it does not stay constant as the Moon thins. Instead, the cusp-to-cusp arc shrinks as elongation decreases. A crescent near full illumination spans almost the entire 180 degrees of the lit limb; a young crescent at, say, 20 degrees elongation shows a much shorter arc; and as elongation falls toward about 7 degrees, the extrapolated arc length collapses toward zero. In other words, the horns of the crescent retreat inward from the poles toward the central, sunward point of the limb until, at the limit, there is effectively no continuous illuminated arc left to perceive.

This "shortening of the cusps" is the geometric heart of the Danjon limit. Whatever the underlying cause, the observable consequence is the same: the crescent does not merely dim uniformly as it approaches the Sun, it loses length from both ends.

The Physics: Three Competing Explanations

Here is the single most important point that popular accounts get wrong. The cause of the cusp shortening is still debated. Danjon offered one hypothesis, but it is not settled fact, and at least three distinct mechanisms have been proposed. A careful reader should hold all three in mind rather than treating any one as proven.

Hypothesis 1: Danjon's lunar topography and cusp foreshortening

Danjon's own explanation was geometric. Think of the Moon not as a smooth sphere but as a world of mountains, craters, and sharp ridges with no atmosphere to soften the boundary between light and shadow. When the Moon is close to the Sun in the sky (low elongation), sunlight strikes the limb at an extremely shallow, grazing angle. Near the cusps, where the terminator meets the limb, this grazing light is partly blocked:

1. Mountains cast long shadows. Peaks along the Moon's limb block sunlight from reaching the terrain behind them.
2. Craters become shadow pools. Depressions that would otherwise catch some light fall entirely into shadow.
3. The cusps foreshorten. The illuminated arc breaks up and retreats from the poles, so the effective length of the visible crescent shrinks toward the centre.

In this picture the limit is a property of lunar geology: the roughness of the surface eats away the tips of the crescent. It is an elegant idea, and it is the explanation most often repeated, but it was a hypothesis, not a measured mechanism.

Hypothesis 2: Schaefer's photometric brightness fall-off

Astronomer Bradley Schaefer challenged the topographic account on photometric grounds. He argued that the dominant effect is not shadowing by mountains but the intrinsic brightness fall-off of the lunar surface toward the cusps. The surface brightness of the crescent depends strongly on the local angle of illumination and on the Moon's scattering behaviour at large phase angles. Toward the horns, the geometry drives the surface brightness down so steeply that the cusps fade below the eye's contrast threshold long before any mountain shadow matters. On this view the cusp shortening is a smooth photometric effect of a roughly spherical Moon, not a consequence of jagged topography, and it would occur even on a perfectly smooth body.

Hypothesis 3: Atmospheric seeing and the contrast threshold

A third strand, associated with D. McNally and with contrast-threshold modelling more broadly, locates the limit partly in Earth's atmosphere and the observer's eye rather than on the Moon at all. Near the horizon and near the Sun, atmospheric turbulence (astronomical "seeing") blurs the already-thin crescent, while the bright twilight sky raises the background against which it must be detected. The decisive quantity becomes the contrast threshold: the minimum brightness difference the eye can register between the crescent and its background. When the crescent's surface brightness, smeared by seeing and set against a luminous twilight sky, drops below that threshold, it vanishes, regardless of what the lunar surface is doing.

So which is correct?

Honestly, the question is unresolved. The three mechanisms are not mutually exclusive; the real limit is almost certainly a combination of lunar photometry, surface roughness, and atmospheric contrast, with the balance shifting depending on whether the observer uses the naked eye, binoculars, or a CCD. What everyone agrees on is the observable outcome Danjon first charted: the crescent's effective length collapses as elongation approaches roughly 7 degrees. The mechanism is contested; the boundary is robust.

The Role of Earthshine

You might wonder: what about earthshine? When the Moon is a thin crescent, the rest of the lunar disc is faintly illuminated by sunlight reflected off Earth, a phenomenon Leonardo da Vinci was the first to explain correctly. Does this help?

Unfortunately, no. Near conjunction, the Moon is so close to the Sun in the sky that the overwhelming brightness of the solar glare (and the blue sky of twilight) completely drowns out the faint earthshine. Even for the crescent itself, the signal-to-noise ratio against the bright twilight background plummets at low elongations.

The Brightness Contrast Problem

Whichever of the three mechanisms above dominates, they share a common consequence: contrast collapses. The crescent is brightest when it is far from the Sun (high elongation) and observed against a dark sky. As elongation decreases, three things happen at once:

• The crescent gets thinner and dimmer (less reflecting area, smaller topocentric width W)
• The sky background gets brighter (closer to the Sun means more twilight glow)
• The crescent's surface brightness drops (shallower illumination angle and stronger fall-off toward the cusps)

All three factors work against detection simultaneously. It is not just that the crescent is dimmer; it is dimmer against a brighter background, so the contrast ratio collapses steeply. This is the same contrast machinery that underlies the arc of vision and atmospheric constraints explored elsewhere on the blog.

The Exact Value: 7 Degrees, or Something Else?

Danjon originally placed the limit near 7 degrees, based on the observational data available to him in the 1930s. Since then the precise figure has been re-derived several times, and the published values cluster in a narrow but real spread rather than agreeing on a single number. It is worth naming who derived what, because "there is significant debate" is too vague to be useful.

The empirical spread

The headline points are these. Fatoohi, Stephenson and Al-Dargazelli (1998), re-examining the historical record in The Observatory, derived a naked-eye limit of about 7.5 degrees, slightly higher than Danjon's original figure. Schaefer's photometric models pushed the theoretical floor down toward roughly 5 degrees, reflecting his view that the eye can in principle detect a smooth, faint cusp closer to the Sun than topography would allow. And modern observers using telescopes and CCDs have repeatedly recorded crescents down to an optical-aid limit of about 6.4 degrees of elongation, which is also the empirical threshold built into the Odeh visibility criterion. The honest summary is that the naked-eye limit sits somewhere in the 5 to 7.5 degree range, with most field evidence clustering around 7 to 7.5 degrees.

Challenges from photography

Photography has pushed the recordable boundary far below the visual limit, but it is important to be precise about what these records mean.

• In 2013, Thierry Legault photographed a crescent at an elongation of about 4.4 degrees, captured at essentially the instant of conjunction (age near zero hours), using a highly specialised setup with a tracking mount in broad daylight. This was an extreme case involving purpose-built equipment, careful solar avoidance, and extensive processing, far removed from anything resembling naked-eye observation, and it tells us about the sensitivity of cameras rather than about the Danjon limit for the eye.
• Several observers have captured crescents at 6.0 to 6.5 degrees using consumer-grade telephoto lenses and CCD cameras.

These observations confirm that the Danjon limit is not a single hard cutoff but a graduated transition zone whose position depends on the detection method:

What Does This Mean for the Islamic Calendar?

For the purpose of determining Islamic months, the relevant limit is the naked-eye or binocular limit, not the camera limit. The Prophetic instruction refers to seeing the crescent (رؤية الهلال), and even scholars who accept optical-aid sighting generally draw the line at binoculars and small telescopes. (For how the new crescent itself is defined, see what is the hilal.)

This means the effective Danjon limit for Islamic calendar purposes remains at approximately 7 degrees, with a narrow grey zone between about 6.4 and 7.5 degrees where optical-aid sightings are possible but naked-eye sightings are essentially impossible.

ARCV and the Danjon Limit: Two Different Constraints

It's important not to confuse the Danjon Limit with the ARCV (Arc of Vision) constraint. They describe different aspects of the visibility problem:

• Danjon Limit: A constraint on elongation, the total angular separation between Moon and Sun. Below ~7°, the crescent fragments regardless of how high the Moon is above the horizon.
• ARCV: A constraint on altitude difference, the Moon's height above the Sun at sunset. Below ~4 to 5° ARCV, the Moon is too close to the horizon's twilight glow to be distinguished, even if the crescent itself is physically intact.

A crescent can fail the Danjon limit while passing the ARCV test (the Moon is high enough but too close to the Sun in total angular distance), or fail the ARCV test while passing the Danjon limit (the Moon is far enough from the Sun but too close to the horizon). The Yallop and Odeh criteria combine both constraints into a single score, but understanding them independently is crucial for interpreting marginal cases. See the methodology overview for how these constraints are implemented in the moonsighting.live prediction engine.

The Danjon Limit as a Falsification Tool

One of the most practically valuable applications of the Danjon Limit is as a falsification criterion for sighting claims. If a sighting committee announces that the crescent was observed on a night when the elongation was, say, 5°, the astronomical community can say with high confidence that a naked-eye sighting was physically impossible.

This is not a matter of opinion or probability; it is a statement about the physics of light hitting a rocky surface. No amount of clear skies, experienced observers, or high-altitude locations can overcome the fragmentation of the crescent below the Danjon Limit.

This falsification capacity makes the Danjon limit a crucial tool for calibrating sighting report databases. When researchers build models like Yallop or Odeh, they use the Danjon limit to filter out implausible claims from their training data, ensuring the resulting visibility criteria are trained on physically possible observations. For a deeper look at how calculation compares with reported sightings, see moon sighting versus calculation.

How the Danjon Limit Relates to Moon Age

"Moon age", the elapsed time since conjunction, is an intuitive but imprecise way to think about crescent visibility. A very young moon (say, 10 hours since conjunction) will typically have an elongation below the Danjon Limit. But the exact relationship between age and elongation depends on the Moon's orbital velocity at the time, which varies due to the eccentricity of the lunar orbit.

At perigee (when the Moon is closest to Earth), the Moon moves faster through its orbit, gaining elongation more quickly after conjunction. A 15-hour-old moon at perigee might have the same elongation as an 18-hour-old moon at apogee.

This is why "moon age" alone is a poor predictor of visibility. Two crescents can have the same age but very different elongations, and therefore very different visibility prospects. If you are just starting out with crescent observation, the beginner's guide to spotting the crescent covers the practical implications of this in plain language.

Approximate age-to-elongation relationships:

These are rough averages and vary by ±2 to 3 hours depending on orbital position.

Record-Breaking Crescents

The pursuit of the youngest visible crescent has become a competitive niche within amateur astronomy, and a handful of extreme observations are frequently cited but often with inaccurate details. It is worth stating the verified facts precisely.

• Youngest naked-eye crescent: Generally cited as approximately 15.5 hours after conjunction, achieved under ideal atmospheric conditions at low latitude with the ecliptic nearly vertical.
• Youngest verified binocular crescent: Mohsen Mirsaeed (2002) recorded a crescent at approximately 11 hours 40 minutes after conjunction, a figure verified by independent observers and cited in the ICOP database. Earlier claims of "13 to 14 hours" typically conflate binocular and naked-eye categories or lack independent corroboration.
• Youngest CCD crescent: Thierry Legault (2013) photographed a crescent at an elongation of about 4.4 degrees, captured at essentially the instant of conjunction (age near zero hours), using a specialised tracking mount in broad daylight with extensive post-processing. This was an engineering achievement with purpose-built equipment, far removed from anything resembling visual observation, and it tells us about camera sensitivity rather than the Danjon limit for the human eye.

These records illustrate the graduated nature of the Danjon limit. The physical fragmentation of the crescent is real, but advanced optics can partially resolve features that the naked eye cannot. The ICOP archive, accessible via the historical sightings archive on this site, is the most comprehensive database of verified crescent records available.

Conclusion

The Danjon limit is one of the most elegant constraints in observational astronomy: a boundary where the photometry of sunlit rock, the contrast sensitivity of the human eye, and the turbulence of the atmosphere converge in a way that no telescope can fully overcome. What Danjon charted geometrically in 1932 and 1936 as the "shortening of the cusps" has since been interpreted through three distinct lenses (lunar topography, photometric fall-off, and atmospheric contrast), and the debate about which mechanism dominates is genuinely unresolved. The observable boundary is robust; its physical cause is contested.

For Islamic calendar determination, the Danjon limit provides a firm practical floor for naked-eye sighting: if the elongation is below about 7 degrees, the evidence firmly suggests that the crescent cannot be seen without optical aid, and any claim to the contrary demands extraordinary supporting evidence. Optical-aid sightings can push toward 6.4 degrees; CCD photography can go lower still, but those instruments operate in a different regime from the historical and religious tradition of visual crescent sighting.

Understanding the Danjon limit transforms the crescent from a simple calendar question into a window on lunar photometry, atmospheric optics, and the fundamental limits of human vision. You can explore how these limits play out across the globe on the global visibility map, or compute your own location's prediction directly at moonsighting.live. Pro-tier users gain access to cloud-cover overlays and the extended ICOP archive, tools that bring the full depth of this science to bear on a single night's sky.

Clear skies and happy sighting.

References and Further Reading

• Danjon, A. (1932). L'Astronomie, vol. 46. (First report of the elongation effect.)
• Danjon, A. (1936). L'Astronomie, vol. 50. (Quantification of the cusp-arc limit near 7 degrees.)
• Fatoohi, L.J., Stephenson, F.R. and Al-Dargazelli, S.S. (1998). "The Danjon limit of first visibility of the lunar crescent." The Observatory, 118. (Derived approximately 7.5 degrees from historical naked-eye records.)
• Odeh, M.Sh. (2004). "New Criterion for Lunar Crescent Visibility." Experimental Astronomy. (737 observation records; optical-aid limit approximately 6.4 degrees; ICOP founded 1998.)
• Yallop, B.D. (1997). "A Method for Predicting the First Sighting of the New Crescent Moon." HM Nautical Almanac Office, NAO Technical Note No. 69.
• Schaefer, B.E. (1991). "Length of the Lunar Crescent." Quarterly Journal of the Royal Astronomical Society, 32. (Photometric model for cusp brightness fall-off.)
• McNally, D. (1983). "The Length of the Lunar Crescent." Quarterly Journal of the Royal Astronomical Society, 24. (Atmospheric contrast-threshold model.)
• The ICOP database of crescent-sighting records is maintained by the International Astronomical Center at https://www.astronomycenter.net.