Celestia Forums

Hi all,

since I had some spare time this weekend (at last!), I attempted to
finally implement a /scientifically correct/ Pluto-Charon texture
system into Celestia and proceeded to the "final challenge":

------------------------------------------------------------------
Tests of precisely known /mutual Pluto-Charon (eclipse) events/!
------------------------------------------------------------------

During the period from 1985 through 1990, Pluto and its satellite
Charon underwent a series of transits, eclipses, and occultations,
which are collectively called "mutual events."

The latter have been extensively studied, since they serve
to derive photometrically the best Albedo maps of Pluto and
Charon, that I have implemented below into scientifically accurate
Celestia textures!

a) The Pluto-Charon Textures/Albedo maps.
-----------------------------------------
There are direct surface images of Pluto from the HST but the resolution
is much inferior to the albedo maps one gets by means of /eclipse
photometry/ in the Pluto-Charon system.

Note that Pluto and Charon are completely tidally locked in their
orbit. Standing on either Pluto or Charon, you would always see the
same face if you are on the side toward the other body.

I extracted and checked lots of material directly from the
/original/ astronomical publications (that I can read online through
my laboratory server) . This allowed to also eliminate various
mistakes in WEB images, like incorrect vertical flips etc.

Let me display my final Pluto/Charon textures that involve /all/ reliable
information we have about them. Virtually NO phantasy. They are
compared to reference images (from Marc W. Buie, Lowell Observatory,
http://www.lowell.edu/users/buie/pluto/plutomap1.html
). These images, however, are incorrectly flipped vertically compared
to the original papers...I have corrected for this flip in the
following comparison (the relative Pluto-Charon texture patterns are
unaffected though):



Of course, the correct Pluto colors from
http://antwrp.gsfc.nasa.gov/apod/ap010319.html

are precisely mapped by GIMP onto my Pluto texture. These amazing results
are in fact from a most interesting paper in Astronomical Journal,
http://www.journals.uchicago.edu/AJ/jou ... t.html#fg1
but I do not know whether you may read it on line...

b) Mutual Pluto-Charon events
-----------------------------

This is just incredible stuff! From that paper I display the precise
event listings as well as the respective schematic drawings that will
allow everyone to perform his/her own checks and orbit adjustments ( Hi
Grant!!)...

Schematic drawings of the events:

Precise timings of the events in UTC:


Now let's see how Celestia is doing. Here is one example that works
/very/well! The time is precisely the one predicted with light time
delay accounted for! Note the shadow of Charon on Pluto and compare
with the corresponding drawing above of the event on May 22, 1987,
7:26 UTC (16 years back!)



We can now use all the above events and start tuning the orbit
parameters, look for buggies in the code etc;-)...

Enjoy!

Bye Fridger

Incredible! Can't wait to download them. When'll they be avalible?

In order that everyone interested can play with the mutual events etc. I produced 1k JPG versions of my Pluto-Charon textures along with a solarsys.ssc setup:

"Pluto" "Sol"
{
Texture "pluto1k.jpg"
SpecularTexture "pluto1k-spec.jpg"
SpecularColor [ 0.135 0.12 0.08 ]
SpecularPower 9.5
HazeColor [0.62 0.85 1.0 ]
Radius 1137

CustomOrbit "pluto"
EllipticalOrbit
{
Period 248.54
SemiMajorAxis 39.48168677
Eccentricity 0.24880766
Inclination 17.14175
AscendingNode 110.30347
LongOfPericenter 224.06776
MeanLongitude 238.92881
}

RotationPeriod 153.293904
Obliquity 115.60
EquatorAscendingNode 228.34
RotationOffset 320.75

Albedo 0.55
}

"Charon" "Sol/Pluto"
{
Texture "charon1k.jpg"
Radius 586

InfoURL "http://www.nineplanets.org/pluto.html#Charon"

SpecularTexture "charon1k-spec.jpg"
SpecularColor [ 0.135 0.12 0.08 ]
SpecularPower 9.5
HazeColor [0.62 0.85 1.0 ]

EllipticalOrbit
{
Period 6.387246
SemiMajorAxis 19636
Eccentricity 0.0002
Inclination 0.0048
AscendingNode 93.5582
ArgOfPericenter 133.7606
MeanAnomaly 93.4395
}
Obliquity 1.0
EquatorAscendingNode 287.6
RotationOffset 213.3
Albedo 0.35
}
pluto1k.jpg:

pluto1k-spec.jpg:

charon1k.jpg:

charon1k-spec.jpg:


Bye Fridger

Fridger:

There is a minor mismatch between the .jpeg extensions of the filenames you linked to, and the .jpg extensions used in the solarsys.scc file. On my Windows machine, I tried just changing the extension name in solarsys.scc, and got nothing but black balls where the planets should be. I had to rename the files themselves with .jpg extensions for it to work.

Tim

timcrews wrote:Fridger:

There is a minor mismatch between the .jpeg extensions of the filenames you linked to, and the .jpg extensions used in the solarsys.scc file. On my Windows machine, I tried just changing the extension name in solarsys.scc, and got nothing but black balls where the planets should be. I had to rename the files themselves with .jpg extensions for it to work.

Tim

Tim:

I consistently use .jpg as ending (not .jpeg) and this should work. Both endings are used, however, in the UNIX world, .JPG is used exclusively...

Bye Fridger

I have now tested the first 7 mutual Pluto-Charon events in the above table. The results are displayed in the image below. I have always entered the nominal (mid) event timing from the table above and accounted for the LT-delay.

NOT bad, NOT bad;-)...

Bye Fridger

t00fri wrote:NOT bad, NOT bad;-)...

Why thank you, Fridger. I'll take that as a compliment to my orbital and rotational definitions.
I might be able to tighten it slightly, though, with these precise timings - especially for the grazing shadow transit of 1990, which seems to be adrift by 30 or 40 minutes in Celestia. I'd previously suppressed most of the trailing decimals in the angular parameters of Charon's orbit, because I despise spurious accuracy in these things - but now we have a real-world referent to which they might make a difference, it seems worth putting them back in and seeing what happens ...

Now ... ahem. Might I suggest that your texture maps are the wrong way up for Celestia? Buie's maps use an east longitude convention, suggesting they are drawn with reference to an IAU north pole (ecliptic north) - whereas Celestia, as you know, uses a rotational north pole (ecliptic south, in the case of Pluto and Charon). But I notice that you haven't inverted Buie's data.

But these, I reckon, are the textures that should be included with the next Celestia release, since they include the maximum amount of real-world data, and will still look fine at relatively low res.

Grant

granthutchison wrote:

t00fri wrote:NOT bad, NOT bad;-)...

Why thank you, Fridger. I'll take that as a compliment to my orbital and rotational definitions. :wink:
I might be able to tighten it slightly, though, with these precise timings - especially for the grazing shadow transit of 1990, which seems to be adrift by 30 or 40 minutes in Celestia. I'd previously suppressed most of the trailing decimals in the angular parameters of Charon's orbit, because I despise spurious accuracy in these things - but now we have a real-world referent to which they might make a difference, it seems worth putting them back in and seeing what happens ...

Now ... ahem. Might I suggest that your texture maps are the wrong way up for Celestia? Buie's maps use an east longitude convention, suggesting they are drawn with reference to an IAU north pole (ecliptic north) - whereas Celestia, as you know, uses a rotational north pole (ecliptic south, in the case of Pluto and Charon). But I notice that you haven't inverted Buie's data.

But these, I reckon, are the textures that should be included with the next Celestia release, since they include the maximum amount of real-world data, and will still look fine at relatively low res.

Grant

Well Marc Buie was apparently confused about his own data (which makes me always suspicious...). If one compares the images he displays on his Website (stating that north is /up/), http://www.lowell.edu/users/buie/pluto/plutomap1.html

with the paper by him and collaborators in AJ,
http://www.journals.uchicago.edu/AJ/jou ... t.html#fg1

then obviously there is a vertical flip missing in the images on his WEBsite.

Now concerning longitude, I could nowhere find a really clearcut statement. In Buie's notation, the face of Pluto that is facing Charon is centered at 0 degrees. In my Celestia texture (cf above), the hemisphere facing Charon is centered at 180 degrees. I thought this is just right?

What I seem to notice though is that everyone is doing things differently than we do it in Celestia;-)...

Bye Fridger

Fridger, a long reply has just been lost by the forum when I tried to post it.
So, a quick reconstruction:
1) The third image on Buie's website has Pluto's central meridian at 281 degrees with the Charon facing side to the right, and north at the top - this implies an east longitude system. There also exists a gridded version of Buie's Charon map (by Tayfun Oner), which employs east longitude - I had a link in my lost posting, and can provide one again if you need it. This made me feel that Buie was following the standard IAU system, also employed by JPL - which meant that his maps needed to be rotated so that their top edge was at the bottom of a Celestia texture.

2) I agree that your central meridians are correct - only the north-south orientation is a cause for concern. I notice you've adjusted for Buie's eccentric prime meridian for Charon, which faces away from the planet - Buie's choice is inconsistent with the IAU convention, and the prime meridian used by JPL Horizons, as well as being out of step with the mapping of every other synchronously rotating moon in the solar system. Sigh.

3) Buie's website maps are very strange, I agree - there's no physical way his north-south flip can be brought into agreement with the original images from the paper you reference.

4) I now belatedly notice that the paper you reference employs a Celestia-style rotational system (with an anticlockwise north) but also a mixture of west and east longitudes - both out of step with the IAU standard.

5) This blithe disregard for mapping conventions undermines any deduction of mine based purely on choice of longitude system - but I do wonder if the reason for the north-south flip on Buie's website wasn't some attempt to transform from one mapping convention to the other, with a momentary lapse of concentration over how that should be done.

6) So since there seems to be no solid way of deducing the correct pole orientation from the data provided, I withdraw all objection to your version. In fact, I now believe it's the more probable of the two options, if we assume that the original paper maintained some internal consistency, Buie's website contains an error, and both ignore planetary mapping conventions.

7) I agree everyone's doing it differently from us - but on the strength of this performance, that seems to be because they're all doing it differently from each other as well.

Grant

MAN! I can tell you that an astronaught got in the way of the pluto picture.
Look closely...

AAHHHH! You're right!! That's pretty neat

I've been eyeballing this Pluto map compared to the one based on the hubble direct observation (posted in the other current pluto thread) and I just don't get it. How are these 2 maps, supposedly both of Pluto, supposed to line up? they look like maps of different objects entirely, except for the general colour.

btw; At the moment I'm just using the basic Hubble observation texture as a texture - if we're going for scientific accuracy I'd only like to see what we are sure is there... even if it's just a low resolution blur.
then again, that's just me - and I probably only do this to feel better about having such a crappy graphics adapter

ogg wrote:I've been eyeballing this Pluto map compared to the one based on the hubble direct observation (posted in the other current pluto thread) and I just don't get it. How are these 2 maps, supposedly both of Pluto, supposed to line up? they look like maps of different objects entirely, except for the general colour.

btw; At the moment I'm just using the basic Hubble observation texture as a texture - if we're going for scientific accuracy I'd only like to see what we are sure is there... even if it's just a low resolution blur.
then again, that's just me - and I probably only do this to feel better about having such a crappy graphics adapter ;-)

Ogg,

the HST results are indeed well compatible, as was also discussed in the scientific paper I was quoting above. Originally, I had even planned to make a texture with both results overlaid. But the HST texture is so much worse than the one above...

Unfortunately, since the conventions/displays involved are not well explained/partly incorrect (see also the discussion with Grant above!), I had not yet added the special Celestia offset of 0.5*width to the texture I gave in the Pluto threat. This was in the initial stages of my investigations and first required looking up the original papers. This I only managed to do over the weekend. Also, the HST texture only extends to -75 degrees in latitude and thus needs 15 degrees of 'phantasy' to be added.

Tonight, I shall replace the HST texture by one that can be directly used (or overlaid) in Celestia.

Bye Fridger

The Hubble map is another one to make you lose faith in mapping conventions. The version available on Views of the Solar System is marked up with an east longitude grid, but unequivocally uses rotational rather than ecliptic north. The map was compiled in the mid-nineties and the missing data are in the map south - the invisible pole at that time was rotational south, IAU north. So at least we're sure which way up that one goes ...
So the fact that the authors of the paper you reference discuss correspondences between the two maps is circumstantial evidence that the orientations should be kept the same.

Grant

ogg wrote:I've been eyeballing this Pluto map compared to the one based on the hubble direct observation (posted in the other current pluto thread) and I just don't get it. How are these 2 maps, supposedly both of Pluto, supposed to line up? they look like maps of different objects entirely, except for the general colour.

btw; At the moment I'm just using the basic Hubble observation texture as a texture - if we're going for scientific accuracy I'd only like to see what we are sure is there... even if it's just a low resolution blur.
then again, that's just me - and I probably only do this to feel better about having such a crappy graphics adapter ;-)

Below you find the promised blend of HST and 'photometric eclipse' textures. I have also not filled the 15 degrees of 'southern phantasy' in the HST texture.

Bye Fridger

PS: I have also replaced the 1k HST texture in the "Pluto thread" by one in accordance with Celestia conventions.

If I stare a this map long enough, I could swear I see canals!

selden wrote:If I stare a this map long enough, I could swear I see canals!

:)

To me they look more like German Autoroutes;-)

Quite advanced people out there on Pluto...

Bye Fridger

Fridger:

I am enjoying the thread on the accuracy of Pluto and Charon's mutual orbits, but I'm confused about one thing. Your ssc file matches the Pluto-Charon ssc that Grant has posted earlier, with one exception. Charon's semimajoraxis in Grant's file is 17456. Yours is 19636. That is significant. Which should we use?

Frank

fsgregs wrote:Fridger:

I am enjoying the thread on the accuracy of Pluto and Charon's mutual orbits, but I'm confused about one thing. Your ssc file matches the Pluto-Charon ssc that Grant has posted earlier, with one exception. Charon's semimajoraxis in Grant's file is 17456. Yours is 19636. That is significant. Which should we use?

Frank

Good you noticed this. I do not know to which of Grant's modifications you are referring to. However, I am always using the latest CVS data, where Grant is usually committing his /final/ adjustments.

I have just checked again: SemiMajorAxis 19636
is what is in there...

But you see, given those high-precision mutual event timings and topologies we can now really tune the orbit parameters to an even better level of accuracy.

Bye Fridger

Frank:
I can't seem to find the thread right now, but what you're quoting looks like my original back-of-the-envelope estimate of Charon's distance from the Pluto-Charon barycentre, rather than the distance from Pluto itself. See http://shatters.net/forum/viewtopic.php?t=2101 for my final take on a barycentric Pluto-Charon orbit - slightly different, but of the order you quote. (Now, of course, you don't need to use my empty object - there's an updated version for Celestia 1.3, using the "invisible" object, at http://www.lns.cornell.edu/~seb/celestia/hutchison/pluto-charon.ssc.)
Fridger is quoting the Pluto-centred orbit for Charon as it is currently defined in solarsys.ssc.

Either the barycentric definition or the planetocentric definition will work for the mutual events since the difference between the two is relatively trivial - less than the diameter of Pluto at ~30AU, or <1 part in a million.
But obviously you shouldn't mix values from the two definitions!

Grant