In current (2021) graph from heavens above (https://www.heavens-above.com/IssHeight.aspx) there are sudden falls - more than 1km in many cases and then after other sudden changes, the graphs return on path of previous almost smooth decay:

iss orbit height for 2020 and 2021

For example in the middle of April 2020, middle of July and beginning of October 2020.

Those are not present in older ISS height graphs, for example: iss orbit height for 2017 and 2018

I understand what sudden jumps up are (engine boosts), and what "slow" falling is (orbit decay).

What I don't understand are those sudden falls of 1-2km in less than one day apparently.

What are those?

  • 2
    $\begingroup$ After the jump up in March 2020 and the following jump down in April the ISS is at the same slope as before. Just imagine a straight line drawn from February to May. $\endgroup$
    – Uwe
    Commented Feb 10, 2021 at 23:11
  • 1
    $\begingroup$ @DarrelHoffman Until 2011 the orbit was intentionally low to allow a larger payload for the Shuttle. The current higher orbit reduces drag and fuel consumption. $\endgroup$
    – asdfex
    Commented Feb 11, 2021 at 15:28
  • 1
    $\begingroup$ Does this answer your question? What caused the ISS's sudden loss of altitude in January 2015? $\endgroup$ Commented Feb 11, 2021 at 17:39
  • 1
    $\begingroup$ @MaxQLagrange That does explain why it went down around 2014, but still not why it went back up again in 2018. $\endgroup$ Commented Feb 11, 2021 at 20:04
  • 1
    $\begingroup$ not a duplicate! $\endgroup$
    – uhoh
    Commented Feb 12, 2021 at 1:57

2 Answers 2


They've done a few "de-boost" burns to set up orbital phasing for an upcoming visiting vehicle rendezvous from time to time. That's likely what you're seeing.

Edit: looking at the first plot -- I'm not necessarily sure this is the case now. Notably, the "de-boosts" in that first plot bring the altitude right back to the decay trajectory it was on previously. And the shape of the mean altitude spike in the September-October timeframe looks suspiciously similar to the March-April one. This has me wondering if it is in fact a data or a calculation problem and not reflective of what ISS was actually doing.

All that said, de-boosts have been utilized to set up phasing in the past.

  • 2
    $\begingroup$ There's a "Soyuz phasing reboost" mentioned in this April 2020 report, but no details blogs.nasa.gov/stationreport/2020/04/page/18 $\endgroup$ Commented Feb 10, 2021 at 18:31
  • 2
    $\begingroup$ A reference would improve this answer. $\endgroup$ Commented Feb 10, 2021 at 19:14
  • $\begingroup$ Haven't found anything public in recent memory -- only internal ops notes. Edited my post with more thoughts now. $\endgroup$
    – Tristan
    Commented Feb 10, 2021 at 19:25
  • 3
    $\begingroup$ bring the altitude right back to the decay trajectory it was on previously Well, yes, that's what phasing is all about. On the same trajectory as before, but now displaced in time. Having spent time in a higher orbit, briefly, the station will be later arriving at the same point over the earth some time in the future. $\endgroup$
    – J...
    Commented Feb 11, 2021 at 13:57
  • 1
    $\begingroup$ @uhoh For practical purposes, there's almost always a progress vehicle docked to SM aft which does the reboost/deboost maneuver IIRC. Pointing the engine in the right direction is just a matter of maneuvering station as appropriate. (They prefer to use the progress thrusters whenever possible instead of station's thrusters, as thrusters are limited life items) $\endgroup$
    – Tristan
    Commented Feb 11, 2021 at 15:36

I had a thought about inaccuracies of orbit estimation like @uhoh...

But I then asked - what changed since 2018 (the lower picture of OP)?

Two major changes:

  1. a crewed Dragon

  2. Soyuz and Progress spacecraft have switched to a 3-hour flight scheme.

After comparing launch dates, it's clear that the second is the reason.

It requires much more precise orbits phasing than previous schemes of Soyuz/Progress.

If we look at the upper graph - we see orbit fall in April, and Soyuz MS-16 was launched April 9, 2020.

Then falling in July - Progress MS-15 was launched July 23, 2020.

Falling in October - Soyuz MS-17 was launched October 14, 2020.

In comparison, Crew Dragon doesn't require such precise orbit phasing because it currently (2020) uses longer flight scheme.

  • 1
    $\begingroup$ This might be a coincidence. Why would you do that three times in a way that puts the station back to the extrapolated decaying height? For a pure phasing change that wouldn't be necessary - and it happens three times during the year. $\endgroup$
    – asdfex
    Commented Feb 11, 2021 at 13:29
  • 7
    $\begingroup$ @asdfex: A boost followed a whole number of orbits later with a matching deboost is exactly what you'd need for a pure phase change. It will put you back in your original (or originally predicted) orbit, but delayed by some amount that depends on the size of the boost and how long you spend in the boosted orbit. As for why there are two boosts and one deboost, I don't know, but maybe they wanted to start with an initial boost and observe its effects for a while before following up with a second one to hit the exact desired phase shift. $\endgroup$ Commented Feb 11, 2021 at 15:15
  • 1
    $\begingroup$ @IlmariKaronen Right. Actually, it's a bit more complicated. In April they raised the perigee and lowered the apogee to get into an almost circular orbit. In September there was a raising of perigee and apogee separated and a combined lowering. heavens-above.com/OrbitHeight.aspx?satid=25544 $\endgroup$
    – asdfex
    Commented Feb 11, 2021 at 15:24
  • 2
    $\begingroup$ Another change is that the ISS is 20 km higher than it used to be, so the decay rate is lower. $\endgroup$
    – Tristan
    Commented Feb 11, 2021 at 15:38

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.