Table of Contents (click to expand)
- How Far Away Is The Horizon?
- How Far Can You See From A Mountain?
- Can A Human See 20 Miles Away?
- Why Should The 443 km Photograph Be Impossible?
- How Does Air Bend Light Around The Curve?
- The 7/6 Trick: How Much Farther Does Refraction Let You See?
- Why Did The Record Photo Need A Cold Dawn?
- Is Seeing Past The Horizon A Mirage?
- What Is The Longest Line Of Sight Ever Photographed On Earth?
- So, How Can You Photograph A Mountain 443 km Away?
On 16 July 2016, Marc Bret photographed Pic Gaspard in the French Alps from Pic de Finestrelles in the Pyrenees, 443 km (275 miles) away, then the longest line of sight ever captured on Earth. On a perfectly round, airless planet the two summits could see no farther than about 412 km, because each one’s own horizon lies about 190 km and 223 km away, so the far peak should sit more than a kilometer below the sightline. Air is denser near the ground, so light curves gently downward and follows the planet’s curve a little, which stretches the reach to roughly 442 to 445 km in ordinary conditions and made the shot possible on a cold, still dawn.
Picture a few minutes before sunrise on 16 July 2016. Marc Bret is standing on Pic de Finestrelles, a 2,820 m (9,252 ft) summit in the Pyrenees. To the east the sky is turning orange, and along the bottom of the glow a thin, jagged line of peaks appears: the Alps.
Bret raises a Panasonic Lumix FZ72, a bridge camera with a built-in zoom, and takes the picture. The farthest peak in the frame is Pic Gaspard, 443 km (275 mi) away. That is about the distance from New York City to Washington, D.C. For eight years, nobody photographed anything farther.
The problem is that Earth is a ball, and balls curve. Stand on any hill and the ground drops away in every direction, which is why ships vanish hull-first over the horizon. Do the math for two mountains 443 km apart and the Alps should be hidden below the curve, and yet the photo exists.
The answer involves the air itself, and a margin so thin it needed a cold dawn to close it.

How Far Away Is The Horizon?
Stand on a beach and look out to sea, and the water seems to end in a sharp line. That line is your horizon, and what hides everything beyond it is the planet itself.
Earth curves away beneath your feet, and your eye sits a small height above the surface. So your line of sight skims out, touches the surface at one point, and carries on into the sky. The touching point is the horizon, and everything past it is below the curve.
The rule is short: the higher your eye, the farther the horizon. Working out how much farther takes a little arithmetic.
Andrew Young at San Diego State University works it out with Pythagoras. Call Earth’s radius R and your eye height h. Then:
d = √(2Rh)
Earth’s radius is about 6,378 km at the equator, per NASA. Plug that in and, with h in meters, d ≈ 3.57 × √h kilometers. An adult on the beach has eyes about 1.7 m above the sand. √1.7 is 1.3, so the horizon is about 4.7 km (2.9 mi) away. We walk through that case in how far away the horizon is from a beach.
Notice the square root: to see twice as far, you need four times the height. A tall friend helps less than you would hope.
How Far Can You See From A Mountain?
Now climb: a 1,000 m (3,280 ft) peak puts your horizon about 3.57 × √1000 ≈ 113 km (70 mi) away. Mountains are good at this, which is most of why people climb them at 5 a.m.
Pic de Finestrelles is 2,820 m high. In steps:
- 2 × 6,378 km × 2.82 km = 35,972 km2
- √35,972 ≈ 190 km
So from Bret’s summit, the horizon was about 190 km (118 mi) away, less than half of 443 km.
What people tend to miss is that the target has a horizon too. Pic Gaspard is 3,883 m (12,740 ft) tall, so its summit pokes above the curve as well. Its own horizon lies √(2 × 6,378 × 3.883) ≈ 223 km (139 mi) away.
If the two horizons meet, one straight line can graze the surface at the meeting point and touch both summits. So the longest straight sightline between two peaks is the sum of their horizon distances:
190 km + 223 km ≈ 412 km (256 mi)
On a round, airless Earth, no straight line between those two mountains can be longer than that.

Can A Human See 20 Miles Away?
It depends on how high your eye is and how tall the thing is.
Twenty miles is 32 km. To push your own horizon that far, you need √h = 32 ÷ 3.57 ≈ 9, so h ≈ 80 m (about 260 ft). That puts your eye at the top of a 25-story building.
But the horizon is only where the ground disappears, and a tall object past it still shows its top. A ship’s bridge 30 m above the water adds √(2 × 6,378 × 0.030) ≈ 20 km of reach. So beachgoer and ship can see each other from about 25 km (15 mi), hull already gone. That vanishing hull is a classic clue in how we know the Earth is round.
So a human can see 20 miles, and far more with a mountain at either end. What nobody can do is see past two horizons at once, and that is what the Alps photograph appears to do.
Why Should The 443 km Photograph Be Impossible?
Put the numbers side by side. The straight-line limit between Finestrelles and Pic Gaspard is 412 km. Yet the photograph shows a peak 443 km away, which is 31 km (19 mi) too far.
Bret’s horizon point sat 190 km out, which left 253 km to go. To meet a straight ray from that far past a horizon point, a summit would need to be (253 ÷ 3.57)2 ≈ 5,000 m tall. Pic Gaspard is only 3,883 m, so its summit should sit more than a kilometer below the sightline. It is not that tall, and it has shown no interest in growing.
Bret knew this. In his own write-up, the taller Barre des Écrins (4,102 m) shows as a silhouette with the Sun “in close position below the horizon.” Then comes the key line. “Refractive favorable circumstances,” he writes, let him see peaks even farther off, including Pic Gaspard.
That one adjective, refractive, holds the whole answer. To unpack it, we have to stop treating air as nothing.

How Does Air Bend Light Around The Curve?
Air has real substance, even though it is good at looking like nothing.
Light slows a tiny bit in air, and more in denser air. That slowing is the index of refraction. When light passes from thinner air into denser air at an angle, it bends toward the denser side. HyperPhysics at Georgia State University puts it this way: light in air “will tend to bend toward the area of greater pressure.”
The weight of the whole atmosphere squeezes the air near the ground. So air is densest at the surface and thins with height. A ray skimming the landscape has thinner air above it and denser air below, so it keeps nudging downward.
The surface curves down and away, and the ray curves down too, just less. So the light follows the planet a little way around the bend, and hidden objects get lifted into view. The same effect lets you see the Sun for minutes after it has set.
Surveyors measure the bending with one number, the refraction coefficient, written k. It is the curvature of the ray divided by the curvature of the Earth. If k were 1, light would follow the surface all the way around and there would be no horizon at all. Young’s SDSU pages are blunt: “Values of k around 0.13 have been used in correcting surveyors’ data for a century or more.” The ray curves about one-seventh as hard as the planet does.

The 7/6 Trick: How Much Farther Does Refraction Let You See?
Here is where surveyors, a cautious people, do something clever. Instead of tracing curved rays, they keep the rays straight and pretend the Earth is a bit bigger. A ray that curves one-seventh as much as the planet behaves, on paper, like a straight ray over a planet with one-seventh less curvature. That planet has a radius 7/6 as large.
In symbols, the effective radius is:
R′ = R ÷ (1 − k)
With k = 1/7, that is 6,378 × 7/6 ≈ 7,441 km. Young uses this 7/6 factor to get a new rule of thumb: “about 3.86 km times the square root of the height in meters.”
Now redo the record on the generous planet. Every horizon distance grows by √(7/6) ≈ 1.08:
- Finestrelles: 190 km × 1.08 ≈ 205 km
- Pic Gaspard: 223 km × 1.08 ≈ 240 km
- Total: 205 + 240 ≈ 445 km
With the surveyors’ more conservative k = 0.13, the factor is √(1 ÷ 0.87) ≈ 1.072 and the total is about 442 km.
The airless limit was 412 km, and everyday refraction lifts the reach to 442 to 445 km. The photograph shows a peak 443 km away. So the air bent the rules by only about two kilometers, and on a bad day it would not have bent them at all.
Why Did The Record Photo Need A Cold Dawn?
That two-kilometer margin explains how the shot had to be taken.
The refraction coefficient changes with how fast the air cools with height. Young’s pages give the range: slightly negative on sunny days, and well above 1 at night or over cold water. On a hot afternoon the ground heats the air above it, so the dense layer sits on top, rays bend upward and the horizon shrinks. On a cold night the ground chills the air, the layering is strong, and rays bend down hard.
A 2010 study in the Journal of Geophysical Research by Christian Hirt and colleagues tracked k minute by minute, a couple of meters above a lawn. On sunny days it swung “with amplitudes of 1–1.5 at time scales of 10–30 min.” Down near the ground, the air changes its mind every half hour.
So a sightline that needs k of at least 0.13 wants cold, calm air and no Sun on the ground. That describes the minutes before sunrise better than any other time of day. It is when Wikipedia’s summary places the shot: “photographed at dawn on 16 July 2016.” Bret’s own account describes a January 2015 try that gave only a faint result. The mountain was always there, but the air had to be in the right mood.
Dawn has a bonus, too. With the Sun just below the horizon behind the Alps, the peaks stand as black cut-outs against a glowing sky. A silhouette a few pixels tall needs that contrast. That is why long-distance photographers set alarms for hours nobody else uses.

Is Seeing Past The Horizon A Mirage?
Not in this case, though it lives on the same family tree.
Push the bending far enough and you get looming. HyperPhysics calls it “the phenomenon which can allow you to see a distant ship when it is geometrically below the horizon.” Warm air over cold water bends light down hard. A ship beyond the curve then seems to hang in the sky, sometimes upside down, sometimes stacked and stretched. That is a superior mirage, which distorts the image instead of just lifting it.
HyperPhysics repeats an Arctic story about the polar explorer Fridtjof Nansen. He “once nearly shot one of his sled dogs, thinking it was a polar bear because it formed an enlarged mirage.” The dog, one assumes, had opinions.
The Alps photograph is the tamer relative. Pic Gaspard was not stretched, doubled or flipped. It sat where it belonged, lifted a whisker by the same bend that surveyors correct for every day. A mirage would have given it a second, floating head.

What Is The Longest Line Of Sight Ever Photographed On Earth?
Bret’s 443 km stood for more than eight years. Then, on 15 December 2024, a Slovak photographer named Richard Jezik beat it. Guinness World Records lists the current record as 493.07 km (306.37 mi), photographed from Karagöl in Giresun province, Turkey.
Read the Guinness account and every ingredient from this article shows up. Jezik “monitored weather conditions closely to find the right window and right location.” The village below was snowed in, so he hiked ten hours and set up by moonlight. At midnight it was around −12 °C (10 °F). He stayed all night, and he had “planned to capture the image at sunrise to improve contrast.”
Cold air, a high summit, a winter night and a sunrise silhouette are all there. Nobody has found a way around the physics, only ways to wait for it. At −12 °C, that is a level of dedication most of us reserve for concert tickets.

So, How Can You Photograph A Mountain 443 km Away?
Every summit has a horizon, and the distance to it grows with the square root of the height. Finestrelles reaches about 190 km, Pic Gaspard about 223 km. Add them and a straight line can span 412 km and no farther. The photograph shows a peak 443 km away, so on an airless Earth it could not exist.
But Earth has air, densest at the ground and thinning with height. So a ray skimming the landscape bends downward and follows the curve a little. Surveyors bundle that into a refraction coefficient of about 0.13, or a planet 7/6 its real size. Rerun the sums and the reach grows to 442 to 445 km. The record squeaks in with two kilometers to spare, and only when the air is cold, still and layered.
The flat-Earth crowd in the comments under Bret’s post has it backwards. On a flat Earth you would not need refraction, a cold dawn or a 2,820 m summit to see the Alps from the Pyrenees. You would see them from the beach, every clear day, hull and all. The picture is a world record because the curve is real and the atmosphere lets you cheat it by about half a percent.
The planet is round, and on a cold, still morning its air bends light just far enough to lift a distant summit into view.
References (click to expand)
- Pic Gaspard — Wikipedia
- Pic de Finestrelles — Wikipedia
- Distance to the Horizon — Andrew T. Young, San Diego State University
- Dip of the Horizon — Andrew T. Young, San Diego State University
- Calculating ray-bending — Andrew T. Young, San Diego State University
- Earth — Imagine the Universe, NASA Goddard
- Atmospheric Refraction, Mirages and Looming — HyperPhysics, Georgia State University
- Hirt, C., Guillaume, S., Wisbar, A., Bürki, B., & Sternberg, H. (2010). Monitoring of the refraction coefficient in the lower atmosphere using a controlled setup of simultaneous reciprocal vertical angle measurements. Journal of Geophysical Research: Atmospheres, 115, D21102.
- Longest line of sight on earth photographed — Guinness World Records







