Updated 3 May 2003

Stuff relating to using Microsoft Streets & Trips 2002 and Macro Express to automate plotting of transit tracks is at greenville.html

10 April 2003

Chicago area moon transit tonight - also see matches.txt

Some older content is preserved below.

I have a high level of confidence in various critical aspects of the program that has produced these results, such as its ability to precisely compute the position of the sun & moon- I make use of Mark Huss' AstroLib, which is a Java port of fellow-Techer Bill Gray's Project Pluto C code, that in turn is based largely upon the algorithms in Jean Meeus' Astronomical Algorithms, 2nd Edition.  However, I won't know whether or not the program is free of serious bugs until observations confirm or contradict the predictions that have been made.

As one potential source of serious error- at least at low angles of elevation- the correction I use for atmospheric refraction assumes an observer at sea level.  I should- and expect that I soon will- adjust the predictions to account for the effect of elevation on the amount of refraction that needs to be factored in.  At some point, I may write a Java applet that will allow you enter the actual elevation and air temperature of the location where you intend to observe the transit, which the applet will then use to compute a final adjustment to the transit location.

These predictions have been made by my WorldView program (though I haven't yet released this version of the program online).  One thought is to add "bright object occultation" prediction to the program, that would predict where you'd have to be to see the space station pass in front of Jupiter, Saturn, Sirius, etc.  It's highly unlikely that the program could ever be made accurate enough to let you set up a telescope, and see the space station pass in front of Jupiter*, but such observations would make it possible to measure the prediction error, and thereby fine-tune the predictions.  (John Locker has captured some images that suggest what such an occultation might look like, see: http://www.satobs.org/seesat/Mar-2003/0281.html.)

* you might be able to pick up a few surplus Iraqi anti-aircraft radars pretty cheap, however, that would help you do such a thing.

Before you get in your car and drive 50 miles to try to observe a transit, at least compare the predictions with those made by the NASA websites & applets, such as in the example below.  This is especially useful in verifying the prediction time if, say, the time zone offset in your database record doesn't reflect Daylight Savings Time, which has gone into effect in most of the U.S. now- if you're looking through your binoculars an hour ahead of time, you're not going to see much!

liftoff.msfc.nasa.gov/temp/StationLoc.html can be used to show the location of the space station in semi-real time, though of course it requires a connection to the Internet.

liftoff.msfc.nasa.gov/RealTime/JPass/25/JPass.asp looks like:

This appears to be in good agreement with WorldView's predictions, but the J-Pass graphic does leave a little to be desired, in terms of the precision with which it can be read.

Finally, using the SkyTrack feature of spaceflight.nasa.gov/realdata/sightings/SSapplications/Post/JavaSSOP/JavaSSOP.html, it appears that WorldView's predictions are in excellent agreement with those of NASA- it also shows that if you were to play your cards right tonight in Indiana, you could see the International Space Station ever so briefly obscure the disk of Jupiter!

matches.txt gives the predicted transit tracks for the next 10 days; the format is:

 0 - database record#
 1 - transit type (sun or moon)
 2 - travel distance (miles; individualized emails will be in the units specified during signup)
 3 - date
 4 - time (in UTC, if you haven't entered your timezone at TransitAlertSignup.html)
 5 - latitude of the International Space Station at the given time
 6 - longitude of the ISS
 7 - elevation angle of the ISS
 8 - azimuth angle of the ISS
 9 - ISS range (miles)
10 - elevation corrected latitude for observing the transit
11 -     "         "     longitude
12 - uncorrected (i.e., sea level) latitude for observing the transit
13 -      "      longitude
14 - is space station sunlit?
15 - sun/moon separation angle
0-  1  2----  3--------- 4-------  5--------   6--------  7-----  8-------  9--  10-------  11--------   12-------   13------- 14 15--
52) s    0.4 10 Apr 2003 16:05:07  49.8441 N,  11.5204 W  25.267  111.0840  505  52.4459 N,   2.5437 W  (52.4520 N,   2.5177 W) 
 4) m   22.0 11 Apr 2003 16:48:06  51.3179 N,   3.1721 E  42.532 -112.7280  345  52.7755 N,   1.9188 W  (52.7760 N,   1.9210 W) y 110.6

First, find your name in the list of subscribers below, then search for the corresponding record number in matches.txt .

In the first example above (subscriber record 52), a transit across the sun is predicted for Thursday at (4) 16:05:07 UTC (using the time zone offset specified by the subscriber), (2) 0.4 miles distant from the subscriber, at latitude (10) 52.4459 N, longitude (11) 2.5437 W.

At that time, the space station will be above latitude (5) 49.8441 N, longitude (6) 11.5204 W, and will be (7) 25.267 above the horizon; (8) 111.0840 west from north (oops- I've got the sign reversed from the convention!), at a range of (9) 505 miles.  Because of the subscriber's elevation, the optimum viewing location is west (i.e., in the direction of the space station itself) of where it would be [(13) 2.5177 W] if his elevation were sea level.

In the second example, a transit across the moon is predicted.  At that time, the sun/moon separation angle is (15) 110.6, and the space station is sunlit (14).  Unfortunately, the transit occurs well before sunset.

These scenarios are depicted in the following WorldView screen captures:

(What follows is a bit outdated, but still useful as general information.)

Below, I used Microsoft Streets & Trips 2001 to mark this transit location:

This location is near Asheville, NC, for which Topozone gives an elevation of 2134 feet; hence, I'd want to adjust my observation point accordingly, using a simple linear interpolation.

Below, I used spaceflight.nasa.gov/realdata/sightings/SSapplications/Post/JavaSSOP/JavaSSOP.html to verify these predictions (note that the closest approach of the space station in general will not coincide with the closest approach of the transit point, which essentially is just the shadow cast by the space station):

I use spaceflight.nasa.gov/realdata/sightings/SSapplications/Post/JavaSSOP/orbit/ISS/SVPOST.html as my source of 2-line tracking elements.  Evidently, they update that page on some sort of schedule (I'd guess each Friday).  In the future, my transit predictions will therefore be tied to NASA's schedule for providing fresh elements; when NASA updates their tracking elements, I'll use them to make transit predictions for the next 9 days.  I expect that by Friday, I'll have implemented individualized email alerts.

Daylight Savings Time is a problematic issue, and at present I don't intend to try to deal with it.  If DST is a factor for you, you can always update your timezone offset to take it into account.  In most of the U.S., DST goes into effect Sunday, April 6.

 1 Thomas Fly
 2 Bill Arnold
 3 Brian Cook
 4 Michael Gill
 5 Robin R. Wier
 6 Calvin Hansen
 7 Jay Galvin
 9 Robert Smathers
10 Dick Emmons
11 sciteach@mchsi.com
12 Willie Koorts
13 Willie Koorts
14 Vossinakis Andreas
15 Patrick Faucon
16 Josef Huber Munich Observatory
17 Podgor
18 Mike Boschat
19 Joe Hurley
20 jeep49
21 Jimmy Herrera
22 Will Marchant
23 Bob Forrester
24 Gene Stratton
25 Anthony Ayiomamitis
26 Paul Maley
27 Moshe Nativ
28 Ralph McConahy
29 rblair
30 Ulhas Deshpande
31 Pat McNally
32 Sherry Underhill
34 Paolo Morini
35 TomS
36 Fred Burger
37 skyguy
38 Yoshiro Yamada
39 Suhas
40 Mark Hanning-Lee
41 Wim Holwerda
42 Sandro Boschetti
43 Russ Pinizzotto
44 Jim Opalek
45 Arturo R Montesinos
46 Oliver Kloes
47 Marco Fiaschi
49 Mark Elowitz
50 Robert Fuchs
51 rcurry@astroedu.org
52 Mike Tyrrell
53 Tim Farris
54 Matija Perne
56 Eddie Bellhouse
57 Peter Kok
58 Pietro Basello
59 Dennis Jones
60 Gian Pietro Basello
61 Kevin McMillan
62 Ron Lee
63 Prof. Dick Emmons
64 Forrest Hamilton
65 mark belleisle
66 franco foresta martin

Shown below is a WorldView screen capture showing the ISS (gold dot) and its moon transit (red dot) predicted for the morning of Sunday (no pun intended), March 23; the moon transit path is from St. Louis, Missouri, to Jacksonville, Florida.  Unfortunately, the time is a bit too much before dawn, and the space station will just miss being sunlit in Jacksonville.

Also shown (the displayed time is Eastern U.S.) is a sun transit predicted just before sunset, near Capetown, South Africa, a couple orbits later (another moon transit occurs in the western U.S., in between the two orbits).

WorldView's 3D animation capability adds a valuable dimension to a plain text set of results such as matches.txt; Robert Smathers, at least, was successful in installing it on his Pentium IV / Windows XP system, after some initial hiccups due to leaving the hyphen out of j2ee-consultants.

Sunday moon transit:

Moon transit near St. Louis, Missouri

And (almost) a Monday sun transit!

  Sun transit for Capetown, South Africa

page accesses

Transit track plotted with the help of Microsoft Streets & Trips 2001 (red arrow courtesy of Adobe PhotoShop Elements; screen capture by SnagIt):