Above: a rare color photograph of a transit by the International Space Station, 2004-06-08
A week after the 2004-06-08 fact, the above is the only photo / video known to me of the International Space Station transiting the Sun simultaneously with Venus. This remarkable feat was accomplished by Tomáš Maruška: http://atm.zaciatok.sk^ (DivX 5 format video)
European Space Agency story^
Tomáš made this observation from his uncle's patio, near the Slovakian village of Stupava, at 48.2579° N, 17.0272° E, elevation 164 meters above MSL (208 meters above the WGS84 ellipsoid) at about 10:09:17 UTC, after receiving my email alert. I regenerated that alert to reflect his actual observation site:
name: ISS Venus
latitude: 48.2579 N
longitude: 17.0272 E
elevation: 208 m
time zone: using UTC
A - travel distance (kilometers) and direction
B - date
C - time
D - elevation angle of the ISS
E - azimuth angle of the ISS ( + is East from North; - is W from N)
F - range (kilometers)
G - latitude for observing the transit
H - longitude
I - how far (kilometers) can I be from the centerline?
J - planetary encounters: 1=Mercury; 2=Venus; 4=Mars; 5=Jupiter; 6=Saturn
K - sun elevation angle
L - sun/moon or sun/planet separation angle
A------- B----- C----- D--- E----- F--- G------- H-------- I---- J K---- L----
8.7 SW 8 Jun 100916 63.3 157.8 410 48.2109 16.9334 2.1
1.9 S 8 Jun 100917 63.3 158.0 410 48.2413 17.0222 2.1
1.6 SE 17.16 48.2461 17.0361 2.1
6.4 E 8 Jun 100918 63.3 158.3 410 48.2717 17.1111 2.1
7.4 SW 8 Jun 100916 63.2 158.2 412 48.2256 16.9392 - 2 63.5 0.2
0.3 SE 8 Jun 100917 63.2 158.4 412 48.2560 17.0281 - 2 63.5 0.2
7.4 NE 8 Jun 100918 63.1 158.6 412 48.2863 17.1170 - 2 63.5 0.2
Tomáš subsequently used CalSKY^ and its link to MapQuest to plot the solar transit path. Based upon his MapQuest map (below), CalSKY apparently was predicting a solar transit path that was about 100 meters NNW of what turned out to be the actual path.
The first set of data gives the path of the ISS across the center of the Sun, while the second set represents the predicted path across Venus itself. Tomáš actually would not have received the second set of data, since I normally report only sunlit planetary encounters, when the Sun is at least 8° below the horizon (the ISS was plenty sunlit, but from directly behind!), and was apparently unaware that I'd posted the Venus track online:
Simultaneous transit of Venus and the International Space Station
To do some preliminary analysis, I ran my WorldView program to determine the altitude and velocity of the ISS in orbit; I also used "Image Preview" to magnify Tomáš' photo, until it was nearly the size of my monitor. I then used a ruler to measure the onscreen size of Venus, and the distance between the video frames. This analysis resulted in the following observations:
ISS velocity is 25233 fps -> 7693 meters per second
altitude is 230.0 mi -> 370.0 km
range is 256 mi -> 412 km
Venus 0.55 inches? -> about .01 inch per 1" of arc
ISS travels 12 inches in 10 frames -> 1.2 inches per frame -> 2.0' of arc per frame
-> 239.7 meters per frame (at 412 km range)
This equates to a frame rate of 32.1 fps (in reality, obviously 30 fps, mono)
Information defining the precise orbit of the ISS is encoded in the 6 following "Two-Line Elements." The first was generated on June 7 by NASA's Mission Control Center in Houston, Texas; the others were derived from radar data, and posted by NASA's Orbital Information Group. My predictions were based upon the first TLE; the interactive website CalSKY would have based its predictions on the OIG TLEs, as they became available.
The last OIG TLE "ISSd" actually represents about 10:10:25.6 UTC, which was after this observation was made. "ISSc" represents a time of about 2 hours- little more than 1 orbit- before the observation.
ISSmcc 1 25544U 98067A 04159.51929753 .00020000 00000-0 20000-3 0 9009 2 25544 51.6330 15.0298 0005537 207.1571 152.9300 15.68796178 36803 ISSoig 1 25544U 98067A 04159.93711806 .00014804 00000-0 13139-3 0 9452 2 25544 51.6330 12.8933 0005507 205.7066 355.6711 15.68799899316862 ISSa 1 25544U 98067A 04160.11829168 .00014969 00000-0 13275-3 0 9469 2 25544 51.6331 11.9695 0005534 204.9163 300.3650 15.68806244316892 ISSb 1 25544U 98067A 04160.18136574 .00014863 00000-0 13184-3 0 9471 2 25544 51.6330 11.6482 0005508 205.6040 296.1391 15.68807781316908 ISSc 1 25544U 98067A 04160.34703704 .00014830 00000-0 13154-3 0 9486 2 25544 51.6332 10.8038 0005500 206.3688 151.6668 15.68812729316928 ISSd 1 25544U 98067A 04160.42390752 .00014992 00000-0 13290-3 0 9491 2 25544 51.6329 10.4117 0005395 206.7073 225.7658 15.68815833316945
These TLEs were run thru Rob Matson's SkyMap^ program, using the location 48.2579° N, 17.0272° E, along with an elevation of 208 meters (164 meters above Mean Sea Level, plus 44 meters representing the local difference between MSL and the WGS84 ellipsoid), from CalSKY's 50 GB Shuttle Radar Topography Mission 3" (SRTM-3) Digital Elevation Model database:
(It appears from these SkyMap graphics that Tomáš' camera wasn't quite aligned with the horizon.)
Venus had an angular diameter of 58" of arc (almost 1' of arc = 1/60°). At a range of 412 km, and perpendicular to the line of sight, a "weather balloon" ...
412000 meters * 58" / (180°/Pi * 3600 seconds-of-arc per degree) = 116 meters (380 feet)
in diameter would have the same apparent size as Venus. However, the elevation angle is 63.2° so moving 116 meters perpendicular to the line of sight is equivalent to moving ...
116 meters / sin(63.2°) = 130 meters (425 feet)
on the ground (toward or away from Venus). Venus, in effect, gives us a "ruler in the sky" (with diamonds) that we use for making other inferences. One such inference is that the ISS is traveling about 260 meters (2 Venus's) between frames, which is something that was previously worked out (i.e., 239.7 meters per frame).
Another inference is that- since the ISS appears to be about 1/2 the size of Venus (although it's not quite oriented perpendicular to the line of sight, and so it would appear a little smaller than it actually is)- it must be about 58 meters in size. In fact, the main solar panels of the ISS are 73 meters (240 feet) across.
Another important conclusion is that Tomáš was only about 1/2 a Venus- or 65 meters (210 feet)- north of the actual center of the path of the ISS Venus un-shadow!
Shown below is an error analysis of Tomáš' photo / video, in the form of a 454 kB GIF. Its actual dimensions are 2780 x 2284, so you can save the image locally, then look at it with more resolution (or even magnified), using Image Preview for example.
Because the error is so small, it's rather difficult even to determine. However, in the graphic, the lines that Tomáš plotted represent the transit path of the ISS across the center of the Sun, as well as the transit path "boundary," outside of which the ISS would miss the Sun (and not show up in his video!).
For analysis, I rotated Tomáš' photo 180°, because any given point on the Sun's (or Venus') surface is "projected through" the center of the ISS (e.g., the "top" of the Sun is projected thru the center of the ISS to the south, on the earth's surface, relative to the "bottom" of the Sun).
The x marks the center of the Sun (and hence a point on the actual center line of the solar transit path). It appears to be about "2/3 of a Venus" SSE of Tomáš' center line; or in other words, the center line from CalSKY was about 97 meters NNW of the actual solar transit path (and so, if Tomáš' had located along that path, he would have observed the ISS pass about 43" of arc below the center of the Sun - note, once again, that because the error is so small, this figure is not likely to be precise).
In fact, it appears likely that CalSKY was using the MCC TLE, and predicted about the same path as SkyMap predicts using the OIGd TLE (which is very close to where Tomáš was actually located). If this is the case, then- using the same TLE- the ground position predicted by CalSKY is about 220 meters NNW from that of SkyMap; equivalently, its predicted sky track is about 1.64" of arc NNW from that of SkyMap's, for the same observer site.
Compared to my ground track, Tomáš was essentially 218 meters NNW of the 10:09:17 UTC point (my "travel distance" computation is slightly off, apparently), meaning that my MCC track was about 153 meters SSE of the actual track; or identical to SkyMap's results within my ability to measure.
This doesn't prove that CalSKY's computation is wrong, but it does suggest that Rob Matson and I are doing our computations virtually identically.
Here's how WorldView depicts the transit from high above the Earth (the gold dot represents the ISS; the white dot represents the shadow of the ISS, as cast from the center of the solar disk), in a nice fat 787 kB PNG graphic:
15 June 2004