Scientifically accurate ?

[t00fri]

There are different JPL ephemerides. I think that the one of most interest to Celestia users is DE406, not DE405. The DE406 ephemeris covers the time span from years -3001 to +3000, while DE405 spans 1600 to 2200. Each 300 year block of DE406 is about 10 MB, so 200 MB for the whole 6000 years covered. DE405 runs to 4.7 MB for each 50 years, so it's roughly three times as large as DE406 for the equivalent timespan. DE405 also includes nutations and librations; Celestia doesn't make use of these yet.

Regarding the difference in accuracy between DE405 and DE406, the JPL documentations says this:

This is the same ephemeris as DE405, though the accuracy of the
interpolating polynomials has been lessened (interpolation on the
64-day mesh points remains exact, however). For DE406/LE406, the
interpolating accuracy is no worse than 25 meters for any planet and
no worse than 1 meter for the moon.

Good enough for most Celestia users, I think . . .

--Chris

Thanks, Chris,

I used DE 405 as an example since --independently of Celestia-- it is supposed to be the more accurate set, notably including nutations and librations. I think it's about time to think about n's & l's also for Celestia

Bye Fridger

[chris]

Thanks, Chris,

I used DE 405 as an example since --independently of Celestia-- it is supposed to be the more accurate set, notably including nutations and librations. I think it's about time to think about n's & l's also for Celestia

Yes, it's necessary to start accounting for nutation and libration, as well as precession. The JPL ephemerides aren't required for nutation and libration of course, but it would be nice to offer them as an option, I think via the currently unused CustomRotation property. For precession, I favor this theory by Vondrak:

http://www.astronomy2006.com/files/vondrak.pdf

It's the only one I've seen that doesn't diverge after two or fewer precession cycles.

But this is really a topic for another thread.

--Chris

[t00fri]

http://www.astronomy2006.com/files/vondrak.pdf
--Chris

Interesting!

F.

[cartrite]

There are different JPL ephemerides. I think that the one of most interest to Celestia users is DE406, not DE405. The DE406 ephemeris covers the time span from years -3001 to +3000, while DE405 spans 1600 to 2200. Each 300 year block of DE406 is about 10 MB, so 200 MB for the whole 6000 years covered. DE405 runs to 4.7 MB for each 50 years, so it's roughly three times as large as DE406 for the equivalent timespan. DE405 also includes nutations and librations; Celestia doesn't make use of these yet.

Regarding the difference in accuracy between DE405 and DE406, the JPL documentations says this:

This is the same ephemeris as DE405, though the accuracy of the
interpolating polynomials has been lessened (interpolation on the
64-day mesh points remains exact, however). For DE406/LE406, the
interpolating accuracy is no worse than 25 meters for any planet and
no worse than 1 meter for the moon.

Good enough for most Celestia users, I think . . .

--Chris

I ran the program brief from the spice toolkit and found my copy of de405.bsp only covers from 1959 to 2060. Anyhow such a file, files would be best offered as accessories to Celestia. In other words, use the spice tool kit to create files covering different centuries. That way the 405 file could be segmented into smaller files and be used for it's times and the 406 used only if and when the user wants to. An abbreviated 405 could then be used only covering a couple of decades or so and would be small enough to include in Celestia's files.
cartrite

[cartrite]

This ephemeris is a time subset of the JPL planetary
ephemeris DE405. All of the bodies whose positions are modelled
by DE405 are included.

As of June 23, 1997 this is the official ephemeris planned
for use by the navigation team for Mars Pathfinder.

We note that "formally" the positions of Mercury, Venus and Mars
(planet centers) are not present in DE405. However, we have included
these objects in this SPK file and set the offsets from to their
respective system barycenters to zero. Mercury and Venus have no
satellites. Consequently the system barycenters and centers of the bodies
coincide. In the case of Mars, the offset of Mars from the Mars
system barycenter cannot be detected with current technology.
Moreover, simple calculations show that this offset can never exceed
more than 10 cm. For this reason we include Mars with a zero offset
from the system barycenter.

This is the latest and most accurate ephemeris distributed by
JPL for the planets, sun and moon.

No changes or updates are expected between June 1997 and the end of
the Pathfinder nominal mission in September of 1997.

Usage

This ephemeris is suitable for use by the Mars Pathfinder project.

However, satellites of planets, asteroids and comets are not
provided in this ephemeris. The ephemeris for the Martian
satellites (Phobos and Deimos) to be used by Mars Pathfinder is
MAR033.2.

Some of the text in my copy of 405. Chris, does the copy of 406 have satellities, asteroids? To cover the whole solarsystem would require extra files? Anyhow, I'm starting to remember that these files are already separated and available from links you provided when your were developing SPICE ephemerides.
cartrite

[einstein_1961]

can anyone please inform me in very simple terms, how close is Celestia's accuracy to the calculations used by NASA in their missions?? I hope I could make myself clear.

Thanks for all your efforts.
Raffi

[selden]

The ephemerides included with Celestia should be good to within a small number of arc-seconds as viewed from the Earth. They can differ by a couple of thousand km at the distance of Saturn.

If you want to get exactly the same results as NASA, then you need to use a recent pre-release of Celestia v1.5.0 and install the same SPICE ephemerides that NASA uses.

For more information, please use the Forum's search page to look for threads with the word SPICE in their titles.

[Ajaja]

Excellent! DE405/406 and SPICE in v1.5.0. Very accurate! But still for scientific accuracy we need precise rotation model for Earth/Moon and planets. Default "UniformRotation" in solarsys.ssc is too inexact. Maybe, is it possible to take correct Earth/Moon/planets orientation angles from VSOP87/DE40x/SPICE (with precession, nutation and libration)?