Hi Dawoon,

Yes, of course, your observation is correct. I was fiddling quite a bit trying to reduce the intensity of the blue color in head-on lighting, but given the existing approximations of exact Mie theory in the code this turned out pretty hard.

This remaining deficiency was also what I meant in my above statements:

• "Reproducing the blue halo of backlit Pluto turned out to be a challenge that was less easy than I thought at first."
• "It's not final but shows that I am on the right track"

I tend to agree, but unfortunately some more quantitative imaging data are needed to decide. It's an easy task to weaken the overall intensity

Cheers,
Fridger

_________________

Dawoon,

here are some second thoughts concerning my previous agreement with your claims above:


This guess of yours seems to be contradicted by "Pluto's Blue Sky" release of October 8, 2015. This release showed the blue color of the backlit haze for the first time officially:

Pluto's Blue Sky

A crucial quote from there reads:

This image was generated by software that combines information from blue, red and near-infrared images to replicate the color a human eye would perceive as closely as possible.

In principle, you may still argue that the intensity of this phenomenon displayed in Natural color is too low to be seen visually, albeit its color being as the eye would see it. This interpretation --if correct-- would be a VERY "academic" one ... According to my understanding, Natural color implies also displayed intensities to match the human eye's sensitivity range.

There is actually further support for the naked-eye visibility of the backlit haze from this very recently released greyscale photo of the crescent Pluto:
http://pluto.jhuapl.edu/Multimedia/Scie ... age_id=333

The deep haze layers are seen to extend all the way around Pluto with the sun-lit side (on the right) appearing with normal i.e. NOT significantly enhanced intensity.

Released Oct 29 2015:
[Please click on image for a better view]


And slightly older zoom versions of the sunlit part of the crescent:
Released Sept 17 2015:



What do you think?

Fridger

_________________

While none of the above published images give intensity scales, the apparent brightness of the Pluto sky is known to be equivalent to about twilight or dusk on Earth (http://solarsystem.nasa.gov/plutotime/).

Here's a twilight sky photo on Earth for reference (credit: Thierry Lombry):

Twilight is really quite dark (just look at the ground in the photo, it is nearly black), bright enough for shadows to be visible perhaps but not for white specular highlights. But if we check the intensity values in the above Pluto images, especially the sunlit parts of the surface, you will notice wide regions with luminosity values of 240, 250, etc out of a maximum of 255.

I think that the images have been scaled to luminosity values [0, 255], and the true brightness to the human eye is darker. The colors as you say might be natural, but without some idea of the intensity scales the publicity images are not extremely accurate references for scientific purposes.

Hi all,

thanks to Dawoon's remarks above

I did some re-thinking about the Mie parameters, trying to realize the desired goal of a nice blue backlit haze ring and no significant haze visibility in head-on lighting of Pluto.

Actually, the solution turned out VERY simple: Just multiply the 3 entries of the Rayleigh parameter triplet by 1/3! That's it...

Atmosphere {
Height 130.0 # official value [km]

Mie 0.00012
MieAsymmetry -0.50
Rayleigh [1.66667e-6 1.8e-5 8.984733e-5]
Absorption [0.000367 0.000245 0.0]
MieScaleHeight 49.0
}

As a test, here are one king-size screenshot for backlit and front-lit Pluto, respectively:

backlit:
[For a bigger view, click on images by all means!]


front-lit
[For a bigger view, click on images by all means!]


It's now quite perfect, isn't it?

Here is an MP4 video of the changing haze illumination upon rotation :
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++
From backlit to crescent Pluto
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Due to the numerous similarities of the Pluto and Titan atmospheres, let me check whether the rescaling trick of the Rayleigh triplet also works for my Titan texture, where also the color of the haze ring in head-on lighting was too strong. The spectacular rose backlit ring should not be affected:

Here is what I get for Titan with the same 1/3 rescaling of the Rayleigh triplet:

Atmosphere {
Height 500.0

Mie 0.0001
MieAsymmetry -0.55
# Rayleigh [ 0.0 0.0 0.00017 ]
Rayleigh [ 0.0 0.0 5.666667e-5 ]
Absorption [ 0.000075 0.00030 0.00025 ]
MieScaleHeight 220.0
CloudHeight 220.0
CloudSpeed 65.0
CloudMap "titan-clouds.*"
}

backlit:
[For a bigger view, click on images by all means!]


head-on lighting
[For a bigger view, click on images by all means!]


Fridger

_________________

Looks better! I'm not sure why you chose 1/3 as the scaling factor though?

The main constraint is of course that each entry of the Rayleigh triplet is rescaled with the same factor < 1. Otherwise the color properties of the backlit haze ring would be sensitively affected! The rescaling value is an empirical one, such as to leave a tiny bit of bluish sky near the horizon if the ground is viewed in the crescent configuration. If you prefer, you may use a somewhat smaller value like 1/4 or even 1/10. Unfortunately we are missing reliable imaging info in order to completely fix all parameters.

_________________