Last-minute astrometry with Guide
What is last-minute astrometry?
How is last-minute astrometry done in Guide?
How do I get this data to observers?
A while ago, the ability to show the paths of eclipses and occultations was added to the Guide software. In addition, Guide 7.0 makes use of the ACT star catalog. Put together, these allow you to get accurate asteroid occultation charts. Furthermore, you can make last-minute astrometric corrections with Guide.
What is last-minute astrometry? Guide makes fairly good predictions of the paths made by asteroids occulting stars. But the precision required here is about as stringent as one gets in optical astronomy. The asteroids usually have diameters of about .05 arcsecond. Current theories are really just not up to that level of precision, and in the past, it was quite common for asteroid occultation predictions to be not much more than educated guesses. Even now, Guide's paths (and those computed by IOTA and similar groups) may be a few hundred kilometers off.
At one time, there were two main sources of error: inaccuracy in the star position, and inaccuracy in the asteroid position. Stars are no longer a real problem. The original Tycho catalog (a.k.a. "Tycho-1") vastly improved star position accuracy; ACT made another vast leap. The new Tycho-2 catalog, which will be used in Guide 8, doesn't improve star position accuracy a lot, but does extend coverage from the brightest million stars down to the brightest 2.5 million stars.
Asteroid accuracy is still a pain, though. The orbits were computed using positions relative to older catalogs, and while this gets you down to maybe better than an arcsecond for numbered asteroids, it's still not good enough for this demanding application. To get an orbit that is good enough, you have to get a new set of astrometry for the object, with Tycho or ACT (or, if you can do it, UCAC) based positions. And even then, getting the necessary astrometric quality is a real bear.
How is last-minute astrometry done in Guide? This falls into two pieces. You have to get the high-quality astrometry for the asteroid, as described in the previous paragraph; and you then have to process that data into an orbit for use in Guide.
I'll describe the easy part first: getting an occultation prediction, given a high-quality orbit. The example I'll use is the 20 November 2000 occultation by (752) Sulamitis of the star Mu Geminorum. This event is the subject of an article in the November 2000 issue of Sky & Telescope, and is probably the most significant asteroid occultation of a star this year.
To start, here's the "high-quality orbit". Fortunately, some new observations have been made since the Sky & Telescope article. Combining these with older FASTT data and running them through Find_Orb, I was able to get the following orbit.
Orbital elements: (752) Perihelion 2001 Feb 25.619028 TT Epoch 2000 Nov 20.0 TT = JDT 2451868.5 M 335.11485 (2000.0) P Q n 0.25492107 Peri. 23.73798 -0.32352712 -0.94055386 a 2.4633846 Node 85.26951 0.85068149 -0.33696391 e 0.0744458 Incl. 5.95454 0.41433223 -0.04258833 P 3.87 H 10.4 G 0.15 q 2.2799959 From 49 observations 1998 Feb. 12-2000 Oct. 9; RMS error 0.206 arcseconds
You'll notice that the arc runs close to the actual event on 20 November; the epoch has been set to that date; and the RMS error is quite low. All three are necessary to getting a good prediction.
Anyway, you can save this Web page to your hard drive, then run Guide, click on "Extras... Add MPC comets/asteroids", and select the file you saved. Guide will read through the text and extract new orbital elements for (752) Sulamitis when it sees the above element set. (Guide is bright enough to recognize orbital elements in several formats; the above is one such format.)
Next, go into "Data Shown", and in the "Asteroid labels" section, click on the "Label name, prov" box. Also, check the "Full precision" checkbox, then "OK".
As you may know, Guide can handle occultations in a very general way. You can click on, say, the Moon and Antares, then on "Extras... Show Eclipse/Occultation", and Guide will find the next occultation of Antares by the Moon and will show the path on the earth for that event.
Similarly, for an asteroid occultation, you can click on the asteroid and then on the star that will be occulted. Here's a small trick to simplify this part. Click on "Extras... Miscellaneous Tables", and look for either "Asteroid occn worldwide, 2000" or "Asteroid occ'ns in North Amer 2000". Select it, then scroll down to the 20 November event. Click on the part highlighted in red. Guide will set the date and time to that of the occultation, and recenter on the objects.
You may have to zoom in, and/or go into "Data Shown" to change asteroid display, to get (752) to show up. A tricky point: the "new" asteroid you've added (from the orbital elements shown above) will appear, but so will the "old" asteroid (from Guide's built-in dataset). Guide isn't bright enough to figure out that they are one and the same; if it could, it would drop the old object. The "new" object will be labelled (752). The "old" object will be labelled by name.
So: click on the version of the asteroid labelled (752), and click "OK". Then click on the target star, and click "OK". Guide now knows which object is supposed to occult which, and if you go into "Extras", you'll see that the "Show Eclipse/Occultation" option is no longer grayed out. Select it, and a chart showing the eclipse path will appear. You can zoom in (in this case, on the United States, and get a chart that looks like this.
In the above case, you were supplied with a high-quality orbit with the right epoch of osculation. I ignored the issue of where that orbit came from. That's a somewhat more difficult subject, and I'll post something about that later. In the meantime, I should mention that, on the page of asteroid occultation predictions, I've posted a link to the orbital elements I created in the course of making those predictions. So you can at least get high-quality orbits for those events.
Getting high-precision, Tycho-based astrometry: Under construction...
To my knowledge, only three people/groups currently are getting astrometry of the necessary quality: Ron Stone's FASTT group, at the Naval Observatory in Flagstaff, Arizona; Bill Owen's project at Table Mountain in California; and Gordon Garradd, sometimes from his home in Australia and sometimes from Columbia University's Biosphere-2.
Of these, FASTT currently produces the vast majority of the data.
How do I get this data to observers? UNDER CONSTRUCTION... I'll gladly post charts from your data (it's probably best to send me your astrometric data; then I can create a .GIF chart based on all received data and put it on my Web site. Click here for examples of these charts. ) However, it may be best to send me a URL for your own site. The advantage of this is that, if I post a link to your site, the updates can be nearly instantaneous; there's no risk of a delay while data makes its way to me, often while I'm asleep, then to be processed and posted by me.
It would be better still if IOTA and similar groups (the EAON, European Asteroid Occultation Network) were able to get this data... and I will find out where such data should be sent.