Random Question

[Henrik Persson]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

This is indeed a very interesting off-topic discussion, thank you for
starting it, Dr. Brown.

I have been thinking about it for a bit in the past few days and also want
to weigh in.



Perception of colour is a very interesting topic, combining physics,
biology and cognition. We have to keep in mind the spectrum of the sunlight
(in space) and the wavelength sensitivity of our cones (maxima around 419,
531 and 539 nm [https://www.ncbi.nlm.nih.gov/books/NBK11059/], assuming we
are not tetrachromats). On top of that, our brains will try to adjust what
we’re seeing to interpret the colours correctly. A banana will appear more
or less yellow regardless of the actual wavelength of the light it
reflects, be it in the supermarket or in a spring forest. Recall the
obviously black and blue dress from 4 years ago [
https://en.wikipedia.org/wiki/The_dress]



(Or in microscopy terms – we have to be aware of the dye’s emission
spectrum, our emission filter’s band pass and camera sensitivity as well as
any camera auto white balance we accidentally left on.)



The purple colour can potentially be caused by either purple wavelengths or
by blue+red wavelengths hitting the retina. If it is the former, then for
whatever reason, there has been very little scattering (purple has shorter
wavelengths than blue light and would be even more prone to scattering –
there might be conditions where purple light has a chance to scatter but
not blue light, in say a very thin atmosphere or one with different
composition).  Reduced scattering can be caused by light travelling a
shorter distance through the atmosphere and/or cleaner air. (Here we can
start discussing the colour of the atmosphere on other planets with
completely different atmospheric composition and density.)



The explanation of red+blue light for the purple sunsets might be more
accurate. This could simply be caused by red light from the sunset being
scattered down from the blue areas above. E.g. scattering of the underside
of clouds or non-visible particles/vapour/ice crystals in the
atmosphere.  Indeed,
when scrolling through pictures of purple sunsets, they all seem to contain
clouds. (The non-cloud areas are also purple but may contain traces of
vapour.)



Searching a bit more on the topic seems to support this. There are several
articles talking about the purple skies seen over Florida in the wake of
hurricane Michael last October (lots of pictures and live recordings).
Articles attribute the purple to scattering from an increased amount of
water in the atmosphere but do a poor job of explaining the phenomenon
further (i.e. why purple and not blue).



This turned into a very long email, I hope you made it this far and found
something interesting.



Cheers

/Henrik





*Quick recap on blue sky research (literally)*

As I’m sure you know, the light from the sun is white and contains all
wavelengths albeit at different intensities. As the light enters the
atmosphere, it scatters in a wavelength dependent manner, causing blue
(short wavelengths) to bounce around multiple times, spreading out across
the sky and reaching our eyes from all over. If you look at the sun
(Don’t!), it appears yellow, basically white minus blue.

When the sun is low on the horizon, its rays has to travel through more
atmosphere and slightly longer wavelengths are also scattered – only orange
to red maintains a roughly straight path from the sun. The sun appears
orange-red.

If you add particles to the air, scattering of longer wavelengths (than
blue) will increase and the sky will be more orange, especially at sunset
with the long travel distance. One of the many gifts industrialization has
given us – bright orange sunsets. Prior to the amount of soot, ash and
pollution we have today, we had much more timid sunsets. See for instance
this paper [https://www.atmos-chem-phys.net/14/2987/2014/] where Zerefos et
al. studied old paintings to figure out what the sunset looked like prior
to the industrial revolution. They also compare sunsets before and after
volcanic eruptions. For instance, in 1883, the volcano Krakatoa erupted in
modern day Indonesia and caused sunsets across the world to turn a deep red
- people even called the fire brigade, thinking nearby towns were on fire [
https://en.wikipedia.org/wiki/1883_eruption_of_Krakatoa#Global_optical_effects].





On Thu, 22 Aug 2019 at 14:20, Claire Brown <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> I have really enjoyed reading the replies to this question! Fascinating
> topic that I guess many of us are interested in.
>
> I have only been to Australia once. Last September and I saw a very violet
> sunset. It stood out to me because I had never seen a sunset of that colour
> before. Then a friend of mine posted a sunset photo from Australia just a
> few weeks ago and it had the same violet colour.
>
> Really interesting if this is happening all over the world. I have mostly
> lived in the Northern part of the Northern hemisphere and as I mentioned I
> have never seen a violet sunset before.
>
> I love the idea of testing this out with filters but I'd have to ask
> someone to do that when they witness a violet sunset since I don't live in
> Australia.
> I was trying to understand how uv/violet light would be visible at sunset
> so I favour the idea of it being a mix of colours and like the idea that it
> involves a certain particulate or mix of particulate in the sky.
>
> Thanks for the engaging responses!
>