Hi Andrew,
Late reply here...
Your post caused me to realize that the U parameter is not all that
well documented, either by me or by MPC. MPC does have a page on it,
https://minorplanetcenter.net/iau/info/UValue.html
which has left me mostly in the dark as to what U is "supposed" to
mean. I think it was sort of bodged together to give MPC some sort of
rough guidance as to when an object is considered to be numberable.
I've written up some documentation for U and just posted it at
https://www.projectpluto.com/uparam.htm
The problem with U for planetary orbits is that it's really intended
for (elliptical) heliocentric orbits, and it's not at all clear to me
how it's supposed to generalize for geocentric orbits. (There are a
variety of other problems with U, as you'll see from my mini-rant
at the end of the above page.)
The criteria used for numbering of outer satellites are somewhat
unclear. I suspect that MPC or JPL run ephemerides and look at the
ephemeris uncertainty; if the object is unlikely to be lost (the
uncertainty remains less than a few arcseconds for the next few
decades), it's considered numberable. As you've probably seen, I
have pseudo-MPECs for most natural satellites at
https://www.projectpluto.com/natsats/summary.htm
They tend to fall into two categories. If an object has been seen
at two or more oppositions, it's numbered (and its ephemeris is indeed
determined to good precision for the next few decades). Some of these
have U parameters as high as 6, but they do have well-determined orbits.
The second category is objects observed at one opposition, and I
don't think any are numbered and I don't think any have ephemerides
good enough for me to say, "These guys really ought to be numbered."
I would expect anyone studying the stability and long-term evolution
of these objects would generate Monte Carlo clones, and would apply
that to _all_ of the objects. Even those with poorly determined orbits
frequently are well enough determined so that you can examine their
stability and evolution, even though you may not be able to say exactly
where they were ten years ago. You'd end up seeing the clones disperse
into a ring around the planet, and could then see how that ring changes
over time.
Some of my pseudo-MPECs are elderly. There hasn't been an update
to the MPC's natural satellites file since May. However, their site
says that they are "preparing MPCs", which may mean they'll cough up
observations made since then and I'll be able to do some updates.
-- Bill
On 2018-09-20 07:13, Andrew.Walker@... [find_orb] wrote:
>
>
> Bill, I've noticed from your web pages and on here that findorb gives a U value for orbits of planetary
>
> satellites (natural and artificial).
>
>
> Can you comment on how these compare with the U values for regular minor planets.
>
>
> Also, especially for the outer satellites of Jupiter and Saturn, is there a U value which roughly corresponds with a numbered status (i.e. the orbit is well determined?). I'm interested as hopefully in a few years many of these objects will have their orbits reach this level of accuracy. Then people can study how stable thay are, whether Saturn has a major influence etc etc!
>
>
> Thanks!
>
>
> Andrew
>
>
>
>
>