Sunday, April 6, 2014

Hidden realities 2 - Sky structures

Structures inside the monochrome
Take a picture of a perfectly blue sky. Do some image processing. Enhance contrast, fiddle with brightness and gamma. Finally convert to binary. You will discover hidden structures inside that monochrome blue.
This is not an earth-shattering discovery. But I'm surprised  how much can be seen with very simple tools. 
What do we see? We have amplified very subtle differences in picture intensity. Too subtle to see with the naked eye. But are these differences really there? Are they in the sky? Or in the optics of the camera? Are they differences in light absorption or internal reflections? Or just differences in noise level of the sensor?
Is it only my camera? Or is it a universal phenomenon? To test this I downloaded a few monochrome pictures from Flickr. Similar structures are in all of them. They reveal mysterious flocks, horizons, intrusions, coronas, protuberances and ellipsoids. Shapes I never saw before and whose existence I didn't suspect.
And even scientists can be surprised by the subtle shapes they find in the sky. They speak poetically about their discoveries inside the known, their expeditions into the (seemingly) obvious:
To the uninitiated, the clear daytime sky seems such a commonplace that its radiance and brightness distribution surely must be well known. Researchers in fields ranging from solar energy engineering to atmospheric optics have repeatedly measured and modeled the angular distribution of clear-sky radiances, and they have published scores of papers on the subject. What can be left to know?
In fact, a great deal is left to know. In simple models of scattering by the clear atmosphere, radiance increases monotonically from the zenith to the astronomical (i.e., dead-level) horizon. However, a persistent feature of our cloudless atmosphere is a local maximum of radiance several degrees above the horizon, not at it. We have detected this near horizon radiance maximum in clear daytime skies ranging from mid latitudes to the Antarctic, and from mid continent to the open sea. However, no one, to my knowledge, has written about it. Why?
And more discoveries hide inside the everyday. Logical assumptions proven untrue:
One of the oldest assumptions about cloudless skies is that their chromaticity and luminance distributions are symmetric about the solar meridian, or principal plane. Skylight symmetry also agrees with the evidence of our eyes—the clear daytime sky’s color and brightness indeed look balanced on either side of the principal plane. But what experimental evidence exists for this symmetry?
Although the apparent dome of the clear sky may summon thoughts of celestial perfection and symmetry, our research shows something quite different. No matter how clear the sky may appear to us, on many days its color and luminance are asymmetric about the solar meridian.

Horizon brightness revisited: measurements and a model of clear-sky radiances, Raymond L. Lee, Jr.
Color and luminance asymmetries in the clear sky, Javier Hernandez-Andres, Raymond L. Lee, Jr., and Javier Romero

No comments:

Post a Comment