Nick Carver, a photographer, had a really interesting video recently where he explained the progression of light at sunset. Prior to watching the video, I understood the primary phenomenon of Rayleigh scattering, so I didn't really expect to learn anything. However, there's a lot more going on as the sun sets than I realized. The properties of sunlight (color, hardness, direction) shift in interesting and surprising ways even after the sun falls below the horizon.
The segment begins at 7:10, but if you already have a basic familiarity with the science, you can probably skip to 12:00.
What a great video, thanks! Especially the airlight phenomenon is fascinating with the softness related to its size.
I've always loved that magic almost pink shadowless very dreamlike illumination you'll sometimes get right when the arch is large, and the subtle change in feeling when it gets sharper and dimmer.
Yeah this videonwas great. Some years ago I visited Kitt Peak observatory and the astronomer pointed out something I had never appreciated before. If you have a big enough open sky to observe and turn against the sunset you can observe the Earth's shadow in atmospher. A cool site that few l ever turn to appreciate.
Hey, thanks a lot for sharing Nick's video! I hadn't seen it before, but the details about colours, light and atmosphere were super interesting, especially if you are into photography and painting.
What surprised me was that with enough exposure time, the sky was blue, not black (I’d also note that this being 2008, the color change cannot be attributed to algorithmic processing of the image in camera).
If that was surprising to you, have you seen full moon long exposure images? It looks just like the day time. I have long exposures during a full moon, where at first glance it just looks like a random daytime shot, but when you look more closely, you can clearly make out the stars/milky way/etc. The water also has that telltale smoothing. I actually enjoy this style of photography, but it is so constrained by the short availability window. Essentially, it's a golden hour type thing. Once the moon gets too high, it's just overpowering. So just before/after moon rise is best.
This shouldn't be that big of a surprise. The atmosphere is still there, it does the same thing to incoming light that it always does, your eye normally can't see it because the total amount of light drops the color filtering below the detection threshold of a human eyeball.
It’s the sort of surprise that seems obvious once you see it, but isn’t so easy to predict. Without this experience, most people would guess that in a long exposure picture of the night, you would see a black sky with perhaps brighter luminaries and more easily seen objects on the ground, not a blue sky.
I shot a timelapse of the Tokyo skyline at sunset that is similar to this but with much more detail [0], and then motion tracked it so that time is traveling across the frame from left to right[1]. In a similar process to the linked NASA image, the video is made of strips, but each strip is 4 pixels wide. It turned out very different compared to how I imagined it would, but that is due to the stages explained in the linked article. Super interesting!
Cheers! I became a bit obsessed with timelapses and this technique during the pandemic. It was shot in Ichikawa, Chiba, close to the station.
And yes, definitely, it's quite stark. I guess a combination of being shot in summer and Tokyo's location relatively close to the equator. I'm originally from southern Australia and the dusk there is far longer.
Olbers's paradox, also known as the dark night paradox, is an argument in astrophysics and physical cosmology that says that the darkness of the night sky conflicts with the assumption of an infinite and eternal static universe. In the hypothetical case that the universe is static, homogeneous at a large scale, and populated by an infinite number of stars, any line of sight from Earth must end at the surface of a star and hence the night sky should be completely illuminated and very bright. This contradicts the observed darkness and non-uniformity of the night sky.
The line of sight argument is nice and succinct, and of course in some sense correct, but I've come to view the paradox in a somewhat different way: if all you have in a static and eternal universe are everlasting sources of energy, then of course you're going to run into trouble when considering equilibrium concerning energy, because it won't exist.
So while it is true that the true resolution to the paradox is that our universe is finite in age and expanding, that doesn't mean a static and eternal universe is in principle untenable. One could for example imagine as of yet unknown sinks of energy, or perhaps starlight gets recycled back into new stars as the old stars disappear. Without speculating on the mechanism, basically any universe where conservation of energy holds will have (on a large enough scale) a constant energy density, and hence, a dark sky at night.
The paradox also assumes several things about light, specifically that light propagates forever. Only in recent years have we managed to prove that assumption true. But if light did degrade over astronomical distances, a static and infinite universe could still have a dark sky at night. If one postulates that light degrades into lower and lower frequencies over time/distance, maybe we are indeed living in a non-expanding universe? Given the thermodynamic issues of an accelerating expansion (dark energy) photons that degrade over distance seems at least a less-strange option.
> light propagates forever .... Only in recent years have we managed to prove that assumption true
I thought the dark of night was explained by redshift (given distance, it shifts out of the visual spectrum). I guess the infra-red still continues on so that is in line with light propagating forever
Because very distant objects describe a very different universe, meaning the light has traveled in time without degrading. For instance, the cmb doesn't match the current universe, proving both that light travels forever and that the universe is not static.
Distant objects also do not show degradation in brightness beyond square area expectations. They are redshifted, but not dimmer than would be normal.
I don’t really see how Olbers’s paradox proves that the Universe is finite, merely that the observable universe is finite. The observable universe is finite because (a) it’s expanding and (b) the speed of light is finite. Both of these can be true in an infinite universe (an infinite universe can still expand everywhere, which can be counterintuitive but it’s true.)
I don't see how it would be a contradiction in terms; we can easily speak of the possibility of an infinite amount of time in the future, why not also in the past?
(It may be factually wrong, but that's not a contradiction in terms).
After writing a path tracer following 'Raytracing in a weekend' I checked out the paper of T. Nishita about implementing a sky model.
It also describes how light is scattered.
For those interested, a copy can be found at multiple websites: Display method of the sky color taking into account multiple scattering
At the lodges of the Grand Canyon they have pamphlets with the sunrise times for the whole year. If one arrives at night and leave the same night. One was not there really. But that prompts the question too: How does the earth turn dark at night?
this is why I love hacker news. the actual article is shit. NASA source, but shit. terms you fucking nothing about the "how". then you come to the HN comments and it's a bunch of smart shit, each comment 100iq higher than the clickbait NASA bullshit they are nested beneath.
The segment begins at 7:10, but if you already have a basic familiarity with the science, you can probably skip to 12:00.
1. https://youtu.be/_q9i_HEOkxo?si=7YYRq6bK-Bcjiw_i&t=430