Hello all,
I've just posted an update with some fairly significant improvements.
As usual, you can download it from
http://www.projectpluto.com/find_orb.htm
...with the latest changes described at
http://www.projectpluto.com/findhist.htm
Summary of changes :
-- You can get ephemeris uncertainty data;
-- You can suppress ephemeris output below a certain magnitude;
-- You can get ephemerides and information over any desired time
range (though you may have to be quite patient if you go really far!);
-- You can get ephemerides in a more "computer-readable" form;
-- A couple of smaller changes.
The most important change, I think, is the ability to show ephemeris
uncertainties. Basically, if the orbital elements show sigmas, you
can toggle display of ephemeris uncertainties. These will be shown in
arcseconds, along with the position angle of the line of variations.
This works, however, only if the orbital arc is long enough so that
you can compute meaningful sigmas (which basically means, "you can hit
the 'full step' button and things don't diverge"). I expect to overcome
this limitation eventually, using statistical ranging to generate both
ephemeris uncertainties and to generate charts showing the uncertainty
area, much as MPC currently does on NEOCP. There are at least six
different ways of computing ephemeris uncertainties, as described
by Steve Chesley at
http://groups.yahoo.com/neo/groups/mpml/conversations/topics/9488
The uncertainties currently shown for both elements and ephemerides
are covariance-based, the first of the six methods Steve described.
Eventually, I expect to add statistical ranging for really short arcs;
observational Monte Carlo (this has actually been in Find_Orb for years,
but you can't get ephemeris uncertainties from it yet); and the fifth
method Steve describes, "multiple solutions", suitable for doing
impact predictions for very low-probability events. I've figured out
how to do all these; it's just a matter of sitting down and pounding
the keyboard a bit.)
You can also configure ephemerides to suppress output if the object will
be too faint, and there's a switch for 'computer-readable' ephemerides.
When this is turned on, RA/dec are in decimal degrees instead of in
Babylonian/sexagesimal/base-60 format, and distances are always expressed
in AUs, instead of in kilometers when less than a million km and in AU
when further than that. The resulting ephemerides are much easier for
programs to parse through.
Also, you can reset the available time range if you want to examine
ancient/prehistoric/far-future events, and there are a couple of other
(mostly cosmetic) changes. For example, when you click on an observation
or generate a pseudo-MPEC, the observatory is marked with something like
"US/Arizona" or "Australia/QLD" to give you an idea as to where the
observatory is. And in pseudo-MPECs, you can have NEOCP data rendered
actually visible, instead of replaced with random text and then blacked
out; so far, this is enabled for CSS, which has kindly given
permission for its NEOCP data to be re-distributed in this manner.
As always, please let me know if you see anything wrong!
-- Bill