Planetary Atmospheres Light Scattering

Description: General discussion about Celestia that doesn't fit into other forums.

scalbers
scalbers
Posts: 138
With us: 12 years 4 months

#41by scalbers » 28.03.2017, 16:45

CO2 is pretty dominant and has a large Rayleigh scattering cross-section (see below). The thing I'm wondering is with all that Rayleigh scattering, how any of the light makes it to the surface instead of being bounced. Here is a good summary.

http://link.springer.com/referenceworkentry/10.1007/978-3-642-27833-4_1349-2#page-1

We can assume for now that 30% of the light makes it through the clouds (if no gas is present). If there were no clouds, the optical thickness of the CO2 (and N2) would be about 20. This should reflect 90% of the light and transmit 10%. I suppose the net effect of clouds + gas would be to transmit 3%. It's unclear to me where the absorption (about 27%) of the light happens, since 70% is the albedo of Venus. Perhaps Venus reflects more light when the sun is overhead rather than lower where somewhat more can be absorbed by the clouds.

https://books.google.com/books?id=lL57o9YB0mAC&pg ... us rayleigh scattering&f=false

Yes, Huygens recorded lots of relevant data from upward and downward spectral radiometers (link earlier posted). I think asymmetry parameters were derived with single scattering albedo. This is where I'm endeavoring to derive phase functions for single scattering and what I call an effective phase function for multiple scattering. We can see Titan also has an N2 gas optical depth of 1.45.
http://stevealbers.net

FarGetaNik M
Topic author
Avatar
FarGetaNik M
Topic author
Posts: 283
With us: 5 years
Location: Germany

#42by FarGetaNik » 31.03.2017, 18:33

The article gives a great description on the light considtions on Venus, I almost felt like I was on Venus' surface!

Well Bond albedo is about 90%, so overall 90% is reflected (hm I might have to adjust my cloud texture once more...). For some reason this is higher than the geometric albedo of 67%... does this mean Venus scatters most light forward? For the 10% transmittet light, considering Rayleigh scattering and absorption, very little should reach the surface.

Ok it should be possible to model Titan's atmosphere accurately, once we get Celestia to render atmosphere better.

scalbers
scalbers
Posts: 138
With us: 12 years 4 months

#43by scalbers » 31.03.2017, 19:03

Sounds good - the first link two posts up is from a reissue of a book from Carl Sagan.

I think Venus' planetary albedo is around 75% so there would be three definitions. I'm unsure what the difference between planetary and Bond albedo is, they appear to be the same as per the values given in the link (77% for Bond). The geometric albedo (at zero phase angle) seems to be less due to more forward scattering as you suggest.

https://nssdc.gsfc.nasa.gov/planetary/factsheet/venusfact.html
http://stevealbers.net

scalbers
scalbers
Posts: 138
With us: 12 years 4 months

Titan views

#44by scalbers » 02.04.2017, 13:19

I'm still considering how to render Titan's sky from the surface. This University or Arizona animation shows a more uniform sky even near the sun without an aureole. This makes sense the more I think about it. The sun and sky color look pretty good. It's a quite well done rendering of the descent and surface views. The descent is a fish-eye lens type view with a field of view of over 180 degrees, so the edge of the the field actually looks up somewhat. We can see the blue sky looking up from higher altitudes, apparently the effect of scattering in a thinner haze. So the Rayleigh scattering does happen though its effects would be mostly swamped by the Mie scattering. Small particles are exhibiting a large negative Angstrom exponent.

https://www.youtube.com/watch?v=9L471ct7YDo (surface is around 4:20 time in the video).

At ground level the overall sky could darken a bit near the horizon, depending on the regional surface albedo. Here is a good reference in case I've yet to post this:

http://www.ciclops.org/media/sp/2010/6514_15623_0.pdf
http://stevealbers.net

FarGetaNik M
Topic author
Avatar
FarGetaNik M
Topic author
Posts: 283
With us: 5 years
Location: Germany

#45by FarGetaNik » 03.04.2017, 09:22

Hm the different definitions of albedo are still confusing, often it isn't stated if the value refers to bond, geometric or other albedos. But well, 77% sounds reasonable...

This is a great video! The upper atmophere surely is dominated by Rayleigh scattering. I didn't incluse any Mie scattering in my Titan definition, as I can't get a good approximation with Celestia's current models. The clouds hide this effect anyways. But from the ground it's a different story... I wonder if the sun is really visible from the surface of Titan? Also which color of light reaches the surface. This paper surely will help to model Titan's atmoshpere better.

scalbers
scalbers
Posts: 138
With us: 12 years 4 months

#46by scalbers » 03.04.2017, 17:41

Indeed quite the video and I think it does a good job with the details of seeing the sun from the surface and with the color of the light. Judging from the paper though the blue sky at high altitudes could (depending on phase angle perhaps) be just as much from fine aerosols as from Nitrogen. This relates to stating what the turbidity is, namely the ratio of extinction coefficients of the aerosols and gas at a given altitude. If the angstrom exponent of the aerosols approaches -4 it effectively becomes the same thing as Rayleigh scattering. The aerosols at some levels have some unusual properties, such as a large angstrom exponent and relatively more forward scattering at the same time.

The methane clouds are only spotty in coverage most of the time. Thus most of the orange and blue colors we see can be considered to be from aerosols.
http://stevealbers.net

scalbers
scalbers
Posts: 138
With us: 12 years 4 months

Titan: Blue and Orange

#47by scalbers » 05.04.2017, 16:27

If we assume most of the scattering effects are from aerosols, why does Titan look blue at the high levels and orange when looking lower down? An important factor could be that at high altitudes the aerosols are thin. Single scattering and the effects of the Angstrom exponent are dominant with more scattering being done at shorter wavelengths. When looking at light coming from deeper in the atmosphere multiple scattering is going on, and the differences in single scattering albedo (compounded by the multiple scattering) switch the preference over to the red and green and less blue. The Angstrom exponent effect becomes less of a factor in a thick atmosphere since the "potential" reflected light increases only asymptotically with the optical depth for the non-absorbing case.

I will try to come up with an algorithm that captures these relationships and we can check how the Celestia code handles this.
http://stevealbers.net


Return to “Celestia Users”

Who is online (over the past 5 minutes)

Users browsing this forum: 2 guests
Bots: Baidu [Spider]