4 ms·

Do rainbows have ultraviolet bands and infrared bands? (2015)

gilrain 5y ago
Indeed, Isaac Newton noticed that a thermometer placed just beyond the red band of a prism-produced rainbow showed a higher temperature than shadow or indeed the red band itself. From this he deduced and described infrared light.
- jeffwass 5y ago
  That was William Herschel, not Newton.
  - gilrain 5y ago
    Ah, I misremembered; thanks! Newton proved that prisms split light into colors it was already comprised of, rather than coloring the light by themselves. Herschel then discovered there were more of those “colors” than could be seen.
  - dotancohen 5y ago
    Good of you for mentioning Herschel. The contrast between him and Newton is stark. Newton was considered a genius, but Herschel was considered to have average intelligence at best. However Herschel's curiosity is as legendary as Newton's indignation is legendary. Herschel made many discoveries, infrared and ultraviolet light among them, just by being curious and testing and trying new things. Such as measuring the temperature of the colors of the rainbow, and even the areas off to the sides.
    - mmmpop 5y ago
      > Herschel was considered to have average intelligence at best. I just looked the guy up and he seems like he was well above average intelligence... what's your source? I feel like people here often are disengaged with "average" people. Think of the "dumbest" person you know in your day-to-day; I bet they're actually fairly average.
      - dotancohen 5y ago
        > I just looked the guy up and he seems like he was > well above average intelligence... what's your source? University professor. Reading Herschel's wikipedia page, I see that the guy was exceptionally curious. I don't see mention of his intelligence going one way or the other.
        perl4ever 5y ago
        "In addition to the oboe, he played the violin and harpsichord and later the organ. He composed numerous musical works, including 24 symphonies and many concertos, as well as some church music." (emphasis mine) Obviously he wasn't Newton, but who is? It's kind of amusing that Uranus could have ended up named "George" if it wasn't for the French.
        dotancohen 5y ago
        Such interest in music was not uncommon for someone of Herschel's social standing in his time. Note that we're not listening to Herschel's Ninth. This is like saying today "In addition to Python, he coded in Javascript and PHP, and later in C#. He wrote numerous scripts, including 24 websites and many text-processing utilities, as well as some CRM plugins for his local church."
        perl4ever 5y ago
        No, it isn't like that. People are engineers, scientists, or programmers, and also have musical interests today. Usually they're way above average in their STEM career.
        cyberbanjo 5y ago
        I agree, I think it's more like being good at golf
        perl4ever 5y ago
croddin 5y ago
It is interesting that the shortest wavelength the human eye can see is very close to half of the longest wavelength the human eye can see. (~380nm - 740nm) In musical terms it is close to one ‘octave’ of light that is visible to us. If we could see more frequencies maybe we would see multiple rainbows stacked on top of each other, but the other rainbows would somehow be different (like notes sound the same but can be up or down an octave). I know that sound works a little differently than light with overtones etc but still something to think about.
- ethbr0 5y ago
  That's a good follow-up. If rainbows have non-visible wavelength components (as it appears they do, from experimentation), where would those spacially appear on the rainbow? The assumption would be "on either side", but I suppose there's also the possibility that you start to get repetition at the same angle from viewer, so you'd have multiple bands overlapping?
  - Ekaros 5y ago
    Or somewhere else. Looking at prism it seems possible that they would be totally reflected.
- VLM 5y ago
  Two cool topics to wikipedia/google for are achromatic lens and achromatic tech in general and chromatic aberration. Turns out to be VERY difficult to manufacture optics that focus correctly even across that one octave. I just think its technologically interesting how camera lenses balance dispersion using multiple types of glass with different properties. There's a resolution issue where eyes and optics run near enough optical law limits such that twice the wavelength is half the resolutions so in theory an ideal camera or eye could run twice the resolution at blue color as red. So deep enough IR you get into snake territory where deep IR "vision" would be like one analog pixel in resolution for a reasonable sized eye. There's a book "Exploring Ultraviolet Photography" about hacking DSLR cameras into UV cams, not to rain on their parade but it seems there's not much "cool" to see out there in nature that radiates in UV except flowers to attract bees. Anyway that's probably why there's not much evolutionary pressure to see UV; "flowers glow a little brighter and more obvious" MIGHT help gather fruiting plants a little better, but apparently not enough to make evolutionary pressure. Its still a cool book from a technological perspective. AFAIK the "problem" with very near IR photography is some dyes and cloths become semi-transparent in very near IR so it turns into creepshot women territory so the community ecosystem is a bit off. From a mother nature evolutionary standpoint I don't recall ever reading anything interesting about near IR forest pictures or whatever. So there's nothing to see until humans invented clothes and dyes and near IR cams to translucently see thru them.
VLM 5y ago
This stack exchange question shows everything wrong with SE and why I no longer use it. 99% of the answers are sophistry. Along the lines of "I'm authoritative" "You need to respect my job title" "I mentioned a pop culture icon so updoot me" "I don't believe in imaginary property so here's a stolen encyclopedia article that doesn't even apply to your question" "Here's someone else's picture, but at least I gave them citation credit" "savaged for not providing his answer in the form of an answer and various other site meme infractions that don't matter to people outside SE" "The usual making fun of the asker for asking a question too poorly to properly answer" PlasmaHH provides the only answer, and its only a partial answer, that they personally used mirrors to turn the UV band on some UV-sensitive ink on money and it lit up, so yeah rainbows clearly output enough UV to activate currency inks. Cool experiment and it answers the question! As for the UV, if it were not for PlasmaHH's IRL experimental results I'd think air and rain would filter UV and IR pretty effectively in general such that you'd get no UV and some narrow "water-proof" IR transmission bands (think of how military FLIR sees thru clouds in specific narrow IR bands but in general IR is blocked by water pretty well, clearly the bottom of the ocean is not cooked by sunlight all the heat gets dumped in the first few feet of water) I know from optical fiber work that non-linear stuff is stereotypically non-linear so its possible the rainbow doesn't emit all wavelengths of EM radiation; an ideal prism or droplet would, but its not ideal, so ... Experimentally I've never heard of a microwave relay path having its SNR knocked down by GHz range "rainbow interference" or wifi does not shut down for miles because there's rainbow interference. Certainly aircraft navigation does not shut down because of rainbow interference. At some frequency range, the total light energy emitted over the life of the rainbow drops below one photon's worth across that range, so you don't get to detect it, so you could usefully argue its not there. It could be fun, but its stack exchange, so its not fun.
- alanbernstein 5y ago
  Did you comment on the wrong article or something? There are only four answers, and the first one links to a page with direct photographic evidence. I get that there are some issues with SE culture, but I don't really see a problem with the top-level question and answers on this page. It's also, you know, not a site for requesting original research from strangers.
TehCorwiz 5y ago
If I understand correctly, the "spread" of a rainbow is due to the higher frequency light refracting at a different angle than the lower frequency light. So wouldn't this be determined first by the frequencies of light (is there IR/UV light to refract?) entering the refractive medium (ie. atmospheric water or maybe a prism) and second by the absorptive / refractive properties of the medium (some materials are opaque in some light bands but not others) with regard to different frequencies of light?
- _Microft 5y ago
  Yes and yes. The range of colors is limited by our vision, though. A hint at that is that the colors of the rainbow, so pretty in the sky, cover the full range (red to violet) of our vision. If rainbows only contained a subset of the colors we could see, e.g. from yellow to green, we might better assume that this is all that there are. > So wouldn't this be determined first by the frequencies of light (is there IR/UV light to refract?) entering the refractive medium I have photographed IR bands of a rainbow while the visible ones had already faded for example. That was right around sunset, so I suspect it was an extreme form of "red sunset" playing into this.

Indeed, Isaac Newton noticed that a thermometer placed just beyond the red band of a prism-produced rainbow showed a higher temperature than shadow or indeed the red band itself. From this he deduced and described infrared light.

jeffwass 5y ago
That was William Herschel, not Newton.
- gilrain 5y ago
  Ah, I misremembered; thanks! Newton proved that prisms split light into colors it was already comprised of, rather than coloring the light by themselves. Herschel then discovered there were more of those “colors” than could be seen.
- dotancohen 5y ago
  Good of you for mentioning Herschel. The contrast between him and Newton is stark. Newton was considered a genius, but Herschel was considered to have average intelligence at best. However Herschel's curiosity is as legendary as Newton's indignation is legendary. Herschel made many discoveries, infrared and ultraviolet light among them, just by being curious and testing and trying new things. Such as measuring the temperature of the colors of the rainbow, and even the areas off to the sides.
  - mmmpop 5y ago
    > Herschel was considered to have average intelligence at best. I just looked the guy up and he seems like he was well above average intelligence... what's your source? I feel like people here often are disengaged with "average" people. Think of the "dumbest" person you know in your day-to-day; I bet they're actually fairly average.
    - dotancohen 5y ago
      > I just looked the guy up and he seems like he was > well above average intelligence... what's your source? University professor. Reading Herschel's wikipedia page, I see that the guy was exceptionally curious. I don't see mention of his intelligence going one way or the other.
      - perl4ever 5y ago
        "In addition to the oboe, he played the violin and harpsichord and later the organ. He composed numerous musical works, including 24 symphonies and many concertos, as well as some church music." (emphasis mine) Obviously he wasn't Newton, but who is? It's kind of amusing that Uranus could have ended up named "George" if it wasn't for the French.
        dotancohen 5y ago
        Such interest in music was not uncommon for someone of Herschel's social standing in his time. Note that we're not listening to Herschel's Ninth. This is like saying today "In addition to Python, he coded in Javascript and PHP, and later in C#. He wrote numerous scripts, including 24 websites and many text-processing utilities, as well as some CRM plugins for his local church."
        perl4ever 5y ago
        No, it isn't like that. People are engineers, scientists, or programmers, and also have musical interests today. Usually they're way above average in their STEM career.
        cyberbanjo 5y ago
        I agree, I think it's more like being good at golf
        perl4ever 5y ago

croddin 5y ago

It is interesting that the shortest wavelength the human eye can see is very close to half of the longest wavelength the human eye can see. (~380nm - 740nm) In musical terms it is close to one ‘octave’ of light that is visible to us. If we could see more frequencies maybe we would see multiple rainbows stacked on top of each other, but the other rainbows would somehow be different (like notes sound the same but can be up or down an octave). I know that sound works a little differently than light with overtones etc but still something to think about.

ethbr0 5y ago
That's a good follow-up. If rainbows have non-visible wavelength components (as it appears they do, from experimentation), where would those spacially appear on the rainbow? The assumption would be "on either side", but I suppose there's also the possibility that you start to get repetition at the same angle from viewer, so you'd have multiple bands overlapping?
- Ekaros 5y ago
  Or somewhere else. Looking at prism it seems possible that they would be totally reflected.
VLM 5y ago
Two cool topics to wikipedia/google for are achromatic lens and achromatic tech in general and chromatic aberration. Turns out to be VERY difficult to manufacture optics that focus correctly even across that one octave. I just think its technologically interesting how camera lenses balance dispersion using multiple types of glass with different properties. There's a resolution issue where eyes and optics run near enough optical law limits such that twice the wavelength is half the resolutions so in theory an ideal camera or eye could run twice the resolution at blue color as red. So deep enough IR you get into snake territory where deep IR "vision" would be like one analog pixel in resolution for a reasonable sized eye. There's a book "Exploring Ultraviolet Photography" about hacking DSLR cameras into UV cams, not to rain on their parade but it seems there's not much "cool" to see out there in nature that radiates in UV except flowers to attract bees. Anyway that's probably why there's not much evolutionary pressure to see UV; "flowers glow a little brighter and more obvious" MIGHT help gather fruiting plants a little better, but apparently not enough to make evolutionary pressure. Its still a cool book from a technological perspective. AFAIK the "problem" with very near IR photography is some dyes and cloths become semi-transparent in very near IR so it turns into creepshot women territory so the community ecosystem is a bit off. From a mother nature evolutionary standpoint I don't recall ever reading anything interesting about near IR forest pictures or whatever. So there's nothing to see until humans invented clothes and dyes and near IR cams to translucently see thru them.

VLM 5y ago

This stack exchange question shows everything wrong with SE and why I no longer use it. 99% of the answers are sophistry. Along the lines of "I'm authoritative" "You need to respect my job title" "I mentioned a pop culture icon so updoot me" "I don't believe in imaginary property so here's a stolen encyclopedia article that doesn't even apply to your question" "Here's someone else's picture, but at least I gave them citation credit" "savaged for not providing his answer in the form of an answer and various other site meme infractions that don't matter to people outside SE" "The usual making fun of the asker for asking a question too poorly to properly answer" PlasmaHH provides the only answer, and its only a partial answer, that they personally used mirrors to turn the UV band on some UV-sensitive ink on money and it lit up, so yeah rainbows clearly output enough UV to activate currency inks. Cool experiment and it answers the question! As for the UV, if it were not for PlasmaHH's IRL experimental results I'd think air and rain would filter UV and IR pretty effectively in general such that you'd get no UV and some narrow "water-proof" IR transmission bands (think of how military FLIR sees thru clouds in specific narrow IR bands but in general IR is blocked by water pretty well, clearly the bottom of the ocean is not cooked by sunlight all the heat gets dumped in the first few feet of water) I know from optical fiber work that non-linear stuff is stereotypically non-linear so its possible the rainbow doesn't emit all wavelengths of EM radiation; an ideal prism or droplet would, but its not ideal, so ... Experimentally I've never heard of a microwave relay path having its SNR knocked down by GHz range "rainbow interference" or wifi does not shut down for miles because there's rainbow interference. Certainly aircraft navigation does not shut down because of rainbow interference. At some frequency range, the total light energy emitted over the life of the rainbow drops below one photon's worth across that range, so you don't get to detect it, so you could usefully argue its not there. It could be fun, but its stack exchange, so its not fun.

alanbernstein 5y ago
Did you comment on the wrong article or something? There are only four answers, and the first one links to a page with direct photographic evidence. I get that there are some issues with SE culture, but I don't really see a problem with the top-level question and answers on this page. It's also, you know, not a site for requesting original research from strangers.

TehCorwiz 5y ago

If I understand correctly, the "spread" of a rainbow is due to the higher frequency light refracting at a different angle than the lower frequency light. So wouldn't this be determined first by the frequencies of light (is there IR/UV light to refract?) entering the refractive medium (ie. atmospheric water or maybe a prism) and second by the absorptive / refractive properties of the medium (some materials are opaque in some light bands but not others) with regard to different frequencies of light?

_Microft 5y ago
Yes and yes. The range of colors is limited by our vision, though. A hint at that is that the colors of the rainbow, so pretty in the sky, cover the full range (red to violet) of our vision. If rainbows only contained a subset of the colors we could see, e.g. from yellow to green, we might better assume that this is all that there are. > So wouldn't this be determined first by the frequencies of light (is there IR/UV light to refract?) entering the refractive medium I have photographed IR bands of a rainbow while the visible ones had already faded for example. That was right around sunset, so I suspect it was an extreme form of "red sunset" playing into this.