Artificial Wasteland · a record-correction you can operate

The Colour the Sky Didn't Lend It

Ask why the ocean is blue and you'll be told it reflects the sky. It doesn't — or barely. Water is blue on its own, in the dark, in a white-walled pool, because it absorbs red light. And the strangest part: it does this not with its electrons, the way every dye and pigment does, but with the O–H bond's own vibration — the only everyday colour in nature that comes from a molecule shaking rather than its electrons jumping. Everything below is computed live in your browser from the measured absorption spectrum of pure water. There is no sky anywhere in the math.

Fill a glass and the water looks colourless — correctly, because half a metre of it barely touches the light. The blue is faint, and faint things need depth to gather. Look down a few metres of clean water and it is unmistakable. So the first instrument is a column you can make as deep as you like, lit from above by a perfectly white lamp — equal energy at every wavelength, so any colour that survives the trip is the water's doing and nothing else's.

Instrument I — a column of pure water, lit by white light

Drag it down. Near the surface the light is white; by a few metres it has cooled to a pale blue-green; by ten metres it is a clear cyan-blue; by thirty it is deep ocean blue and most of the light is gone. The colour the water keeps is the blue–violet end, around 475 nm — exactly where, in the next instrument, the water hardly absorbs at all.

Why blue? Because red goes first.

White light is every colour at once. Send it through water and the water removes wavelengths at wildly different rates. Here is the measured absorption of pure water across the visible — Pope & Fry's integrating-cavity data, the modern gold standard. It plunges to almost nothing in the blue (its minimum is at 418 nm) and climbs steeply, by more than a hundredfold, toward the red. So red is stripped out in a few metres while blue runs on for hundreds. The strip beneath the curve is the light that survives at your chosen depth: watch its red end die first.

This is also why a diver's world turns blue and a red wetsuit looks grey at depth: by the time you're down ten metres there is essentially no red light left to reflect. In pure water, deep red (700 nm) is cut to 1% within about 7 metres; blue (450 nm) survives to nearly 500. That seventy-fold gap is the colour.

The deep strangeness: it's the bond, vibrating

Almost everything coloured around you — leaves, blood, paint, the sky — gets its colour from electrons absorbing light as they jump between energy levels. Water is the rare exception. Its colour comes from the O–H bonds stretching, like two masses on a spring. That vibration's fundamental note is deep in the infrared, far from any visible light. But vibrations have overtones, just like a plucked string, and water's fourth overtone lands at 698 nm — right at the red edge of vision (Braun & Smirnov, 1993). That faint overtone, and its neighbours, are what eat the red.

Here's the clean proof that it really is the vibration. A spring's pitch depends on the masses on it. Swap each hydrogen for its heavier twin, deuterium — that's heavy water, D₂O — and the bond, now loaded with a heavier atom, vibrates slower. Toggle it:

So what about the sky?

The reflection isn't nothing. A still water surface does mirror the sky, most strongly when you look across it at a low, grazing angle — that's the glare on a lake at noon. But that is a surface effect, and it is a contributor, not the cause. The blue you see looking into deep clean water is light that went down, had its red absorbed away by the water itself, and was scattered back up to you still blue. It is there under an overcast grey sky; it is there in a white-tiled pool; it is there in the blue glow deep inside a glacier, where the same O–H overtone absorbs the red out of the light rattling around in the ice. Take the sky away entirely and the water is still blue. Take the absorption away and no sky could colour it.