Asteroid occultation predictions
Sites providing asteroid occultation predictions: An excellent source for asteroid occultation predictions is Steve Preston's site. You can also get info on occultations visible in Europe at this link and this link to the EAON (European Asteroid Occultation Network). Also, Mike Kretlow's home page contains some general occultation information. The IOTA page is an extremely useful resource as well.
Another EAON page, in Portuguese (though much of the raw data can be understood with no knowledge of the language.)
The Web site of the R.A.S.N.Z. (Royal Astronomical Society of New Zealand) Occultation Section has some good data on events in Australia and New Zealand. The above link is to the main page; click here for the occultation update section.
I've ceased posting asteroid occultation predictions of my own; it's a tough thing to keep up with, and several people are doing a better job of it than I probably would anyway.
Doing your own asteroid occultation predictions with Guide 8.0 or 9.0: You can use Guide to produce good predictions for these events. Steve Preston did some interesting tests comparing Guide's predictions to those of his software of choice, OCCULT. There are some occasional disagreements of several kilometers, which have driven us both a little crazy in our efforts to puzzle them out; we have no reason to think either program is in error by that amount. For practical purposes, it doesn't matter, because the currently-available data does not allow for an accuracy of better than a few dozen kilometers. (Which is why we can't say, "Compare the results to the real sky and see which program is right.")
Asteroid occultations are probably the most accuracy-demanding thing you can do with a planetarium program. The asteroid is usually a small fraction of an arcsecond in diameter, which means that if your position for the star or the asteroid is off by that amount, your prediction will be off by an entire path width.
The first, simplest way to generate a prediction in Guide is as follows:
Doesn't sound too daunting, does it? And in fact, doing the above really is easy. Unfortunately, the resulting paths will usually be "close, but not quite right." There are three sources of error, all so minor that you could ignore them in almost any other situation. Two appear with every occultation. One appears on some occultations of fainter stars. All three can be fixed:
In almost all "practical" situations, the best thing to do is to make use of the UCAC-2 catalog. This is now the catalog of choice for really exact star positions. You can get it on three CD-ROMs (click on the above link for details) and then display it in Guide. Once displayed in Guide, you can right-click on the UCAC-2 star, and Guide will know that the occultation is supposed to be relative to that point, not some older catalog's coordinates.
Usually, you'll have three superimposed objects: the star as shown by UCAC-2, the star as shown by an older catalog built into Guide, and the actual asteroid. You may have to zoom in quite a bit until the three separate a bit; then you can right-click on the asteroid, then on the UCAC-2 star.
It's vaguely possible that you'll have a target star that is not in the UCAC-2. This catalog goes down to about 17th magnitude, and doesn't cover much of the sky above declination +45. If the star can't be found in UCAC-2, you will have to somehow enter its exact coordinates by hand.
To get that to happen, download this user-added dataset (about 1 KByte) and unZIP it in your Guide directory. You will see that the file lma_star.dat gives the RA/dec of several "historic" occultation target stars. You can add yours to the list with any text editor.
Once you've added the accurate coordinates, you can fire up Guide and go through the above procedure, presumably now getting an accurate asteroid with "current" elements.
After all that, click on "Extras... Show Eclipse", and you'll finally have a truly accurate, world-class asteroid occultation prediction.
Tips on making the occultation charts in Guide "look good": Once you've zoomed in/out and panned over to get the sort of field of view you really want, you may decide that it would be nice to have more cities shown, or fewer cities shown, or no cities at all shown. All that can be accomplished by hitting the '+' and '-' keys. The former increases the density of displayed cities, adding in smaller, "less important" cities. The latter decreases the density, leaving only the more "major" cities. Hit '-' often enough, and cities go away completely.
Next, it can help to have time-tick markings on the chart indicating what area may be eclipsed at (say) 10:35, 10:40, 10:45, and so on. To do this, click on "Settings... Mark Interval" and (for this example) enter "5". Guide will add black circles on the eclipse path at five-minute intervals.
Having gone that far, you might decide that it would be nice to have labels indicating every fifteen minutes: 10:30, 10:45, 11:00, etc. To do that, click again on "Settings... Mark Interval" and enter (for this example) "i3". Guide will add time labels to every third circle; since these are at five-minute intervals, we'll have time labels for each fifteen minutes.
And finally: you may know that the second Guide CD-ROM has a database of two million places worldwide (outside the U.S.). You can click here for information on how to display them. Be warned that the dataset lacks any concept of "level of importance", though, so towns of population 3 are shown just the same as, say, London or Tokyo. (U.S. users can just make use of the "U.S. Cities" database.)