In a series of four tweets starting with this one (found in 2 satellites will narrowly avoid colliding at 32,800 mph over Pittsburgh on Wednesday)

1/ We are monitoring a close approach event involving IRAS (13777), the decommissioned space telescope launched in 1983, and GGSE-4 (2828), an experimental US payload launched in 1967.

LeoLabs mentions a predicted conjunction of two defunct satellites with a worrisome separation

2/ On Jan 29 at 23:39:35 UTC, these two objects will pass close by one another at a relative velocity of 14.7 km/s (900km directly above Pittsburgh, PA). Our latest metrics on the event show a predicted miss distance of between 15-30 meters.

Since neither spacecraft can be controlled we just have to sit back and watch what does or does not happen 900 km above Pittsburgh, PA.

Question: How did these two spacecraft end up in nearly counter-propagating orbits? Since their altitude is so high they must have both started not far above 900 km. Wouldn't it have been a good idea not to have them going opposite directions? (Or am I misunderstanding the situation?)

enter image description here

  • 1
    $\begingroup$ companion question: What does “MSR” represent in the context of this predicted satellite conjunction? $\endgroup$
    – uhoh
    Commented Jan 29, 2020 at 3:29
  • 4
    $\begingroup$ I didn't know there was a Pittsburgh in Panama (=PA) ;-) $\endgroup$
    – gerrit
    Commented Jan 29, 2020 at 8:57
  • 4
    $\begingroup$ They were launched in 83 and 67. I'm yet to hear about predictions being accurate over a month, let alone after 37 years. There was just no way to predict that over time these two particular satellites would go into conjunction. And by the way, those kind of conjunctions happen often in LEO, they just don't usually make news. But notice that over the next revolution, these two satellites won't pass close to each other, hence why this usually isn't a problem. Every LEO operator without propulsion is afraid of being hit by another satellite, this just didn't happen (by accident) yet. $\endgroup$
    – Mefitico
    Commented Jan 29, 2020 at 19:54
  • 1
    $\begingroup$ Looks like they tunneled through each other. Simulation time step too long ... :-) $\endgroup$
    – user34755
    Commented Jan 30, 2020 at 13:52
  • 1
    $\begingroup$ @ebv there are advantages to living in a simulation ;-) $\endgroup$
    – uhoh
    Commented Jan 30, 2020 at 13:53

1 Answer 1


How? Simple, because they launched into those orbits.

Why? Well, first, let me explain what their orbits actually are.

IRAS (13777) and GGSE-4 (2828) are both in high-inclination orbits, 70° and 99°, respectively. The latter is slightly retrograde, as is common for sun-synchronous orbits. However, to fully understand in what plane they are orbiting, we need two elements, not just the inclination. (Why two? Because you need two measures to indicate a unique direction in 3 dimensions.)

This second element is their right ascension of the ascending node, or, more simply, at what longitude (not relative to the rotating surface but rather to to the First Point of Ares, a fixed point in the sky) they cross the Earth's equator traveling north. These are 214.5008° and 027.7275°, respectively.

You can see those are nearly 180° apart. Combined with the inclinations, which are 10° and 20° off of a true polar orbit but in different directions, and we've got two satellites in nearly the same orbit, except moving in opposite directions.

So back to why...

IRAS was launched into a sun-synchronous orbit because it is an infrared survey satellite and it is useful to always have the sun in the (nearly) same position when shooting infrared photos. That way, they're all taken at the same time of day & you don't have to consider the changes in lighting (as much) for remote sensing applications. GGSE was launched into a high subpolar orbit because it was a '60 spy sat and it needed to take photos of high-latitude Soviet things.

However, it seems IRAS was launched just at the right (wrong?) time of day that GGSE's orbit was directly overhead, but going the opposite direction that IRAS launched. Because their inclinations are similarly off of 90°, but in different directions, that gave their two orbits a very low relative inclination of around 10° (actually, technically, it's around 180, because they're going nearly opposite directions, but that's not my point here, my point is they're in the same plane).

So if IRAS had launched 12 hours later or earlier from when it did, they would have been going the same direction but with a 30° relative inclination difference instead of a 11° one.

  • 21
    $\begingroup$ tl;dr: NASA either didn't do their homework, or made a calculation error, re IRAS; or they didn't know about GCSE-4 (possibly its existence was still classified in 1983?) $\endgroup$
    – Ian Kemp
    Commented Jan 29, 2020 at 11:52
  • 13
    $\begingroup$ @IanKemp I don't think that's necessarily a fair conclusion. At the time (and to a certain extent, today) most operators assumed that the number of objects in space was so low that it would be nearly impossible for any two of them to collide. Even now, nearly 40 years after their launches, they haven't collided (though how many close approaches there have been, I couldn't say) $\endgroup$
    – costrom
    Commented Jan 29, 2020 at 15:41
  • 9
    $\begingroup$ Are you suggesting that a sun-sync and a non-sun-sync vehicle have maintained the same orbital relationship since 1983? They will be precessing at different rates. $\endgroup$
    – BowlOfRed
    Commented Jan 29, 2020 at 20:03
  • 3
    $\begingroup$ Shouldn't "around 180" be "around 190"? $\endgroup$ Commented Jan 29, 2020 at 20:19
  • 1
    $\begingroup$ "because they launched into those orbits" Does a satellite really stay in the same orbit for 50 years? $\endgroup$
    – JiK
    Commented Jan 30, 2020 at 13:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.