This article, this article, and this earlier question indicate that, given enough satellites in LEO, it's possible to substitute between GEO and LEO. The first article and the question have some indication of the substitution ratios (from the article, 40-80 LEO sats to 3 GEO sats; from the question, 100 LEO sats to 1 GEO sat) for what seem to be telecom applications.

My question: For which applications are LEO and GEO substitutable?

I'm assuming telecom and imaging are among them; is that correct, and what else might be?

Any thoughts or suggestions would be much appreciated. Thank you!


The choice of LEO or GEO is often a trade off between different systems and frequently comes down to cost. There's not a lot that you can do in GEO that you can't manage in LEO if cost isn't a problem (or visa versa).

A few examples of where this is the case;

  1. Earth observation missions - typically your first impulse might be that LEO is a clear winner here, it's closer to the Earth, so your resolution is higher, you can get the data back to Earth quicker and using less power - win win right? What if you want to observe a specific country frequently, your own country for monitoring deforestation, soil moisture levels or air quality. You can do that with a single Geo satellite in position over your country, or many LEO sats. If you want to collect data every few minutes, but don't care too much about the resolution of the data, then GEO would almost certainly be a cheaper option than LEO.

  2. Communications satellites - The further away from the Earth a satellite is, the greater proportion of the surface of the Earth that satellite has line of sight to (up to a point of course). So Geo seems like an easy win, sure it'll cost more power to transmit coms over the greater distance, but in space you've got solar panels and power's easy! But what about making sure you're using that line of sight? If your satellite is facing the wrong way then you can put in as much power as you like, your directional antenna won't make a blip on Earth. Pointing accuracy requires either momentum wheels or attitude thrusters, and usually both (thrusters to de-load the wheels after periods of use). Propellant isn't anywhere near as cheap or easy in space, and the pointing accuracy resolution for GEO is much higher than LEO.

  3. Access to space - so you're a 2nd world country looking at dragging yourself into the intellectual elite by launching your very own space station. You can examine crystal growth, ant colonies and spring oscillations in 0g - who wouldn't want to! Do you go LEO or GEO. Certainly in LEO you're more greatly affected by the mass concentrations on Earth, but GEO is so much farther away. It's not cheap to get the bulk material needed for a space station to GEO, let alone the tons of food, water, air and springs that astronauts consume on a daily basis. LEO is closer and therefore cheaper.

I can only come up with one type of mission that requires you to operate in either LEO or GEO (there may be more);

  1. Removal of GEO Debris/Servicing of GEO satellites - replace GEO with LEO, it can work either way. The key thing here is that you are targeting a specific area of space, it's one of your mission requirements. You want to clean up GEO, you can't send a satellite to remove debris from LEO, it would have a negligible effect.
  • 2
    $\begingroup$ Observing from LEO also means that you have to contend with much higher angular velocities during the pass. Not only is this a problem for spacecraft-ground communications, it could quite conceivably in some situations limit the achievable data resolution. As for "power's easy" in space, well, power might be okay but getting rid of waste heat certainly isn't so easy! $\endgroup$ – a CVn Sep 1 '16 at 21:41
  • $\begingroup$ Thanks! This is exactly what I was looking for: potential situations where substitution is impossible (not just costly). So, in general then, it seems like the answer is "unless your application is LEO/GEO-specific, you can substitute". Is that right? $\endgroup$ – penGuinKeeper Sep 2 '16 at 17:17
  • $\begingroup$ *situations where substitution is possible and those where it is impossible (not just costly) $\endgroup$ – penGuinKeeper Sep 2 '16 at 17:24
  • 1
    $\begingroup$ In general I would yes, it's almost always possible to substitute LEO/GEO operations it just comes at a cost. The only other scenario (other than having orbit specific requirement as in servicing/debris removal) might be hardware being too large to launch. If you need mm resolution on images then you might not be able to launch an optics system to get that from GEO because the mirrors (or w/e) are too big to fit in a launch vehicle. You could get around that with deployable device (+cost) or a bigger launch vehicle (+lots of cost). $\endgroup$ – ThePlanMan Sep 2 '16 at 19:39

Moving a system from LEO to GEO has impacts far beyond the number of satellites required to provide constant line of sight to the ground:

  • GEO systems are further away, which increases the latency by 60x and the power required by 3600x for telecom applications, compared to a 600km LEO orbit. It also degrades the resolution by a factor 60x for imaging systems, which makes high-resolution imaging satellites in GEO infeasible with current technology. For active systems such as RADAR and LIDAR, the power required in GEO is 13 million times higher.

  • GEO systems are fixed in the sky: the ground terminals do not need a tracking antenna to follow them, which decreases a lot the cost and makes it possible to reach a wide market (this is important for satellite TV for instance).

So depending on your exact application (satellite TV, communications on the move with lightweight terminals like Iridium, hourly low-resolution high-coverage imaging of the Earth, or high-resolution imaging), the tradeoffs will tell you that a LEO or GEO system is optimal. In that sense, they are not subtitutable. I can't think of any application where a GEO and a LEO system deliver the same service at the same price.

  • $\begingroup$ Unless you are taking the position that the answer to 'which applications are substitutable' is 'none', this isn't an answer. It needs to make clear that is your position, and cover any remaining type of satellite as to why only one of the orbits makes sense. I can't think of any such case, but it should be clear everything is covered. $\endgroup$ – kim holder Sep 2 '16 at 15:59
  • $\begingroup$ Thanks. I'm not thinking about the costs/benefits of either just yet, but this is helpful for me to know going forward. $\endgroup$ – penGuinKeeper Sep 2 '16 at 17:19

TV transmission using LEO satellites would be difficult and expensive for a consumer. A tracking antenna would necessary and for uninterrupted TV even two tracking antennas with a delicate handover from one satellite to the next one. The tracking antennas would need a clear view to the sky for the complete path of the satellite. Compared with a GEO TV satellite a lot of TV consumers could not mount a tracking antenna at the same place useful for a fixed antenna.

  • 1
    $\begingroup$ This seems like an answer to the opposite question: When would LEO and GEO not be substitutable? $\endgroup$ – a CVn Sep 1 '16 at 21:37
  • $\begingroup$ Unless you are taking the position that the answer to 'which applications are substitutable' is 'none', this isn't an answer. So you would have to go farther with this to eliminate other categories of satellites as well. $\endgroup$ – kim holder Sep 2 '16 at 15:52

Geostationary orbits only come in one flavor: zero inclination.

That means that there is no access to the near-polar areas from GEO (geostationary). No imaging, no reliable data links, nada, ничего, zilch. Ignoring refraction, the geostationary ring is below the horizon when you are beyond roughly 81 degrees North or South latitude.

However, the first, and very useful geosynchronous orbit - the Molniya orbit - can have excellent coverage of one pole by sacrificing good (good = slow moving, long time) coverage of the opposite hemisphere. Three satellites in that orbit will provide continuous coverage, but may require some attention to the pointing direction for antennas which have gain.

enter image description here

above: A Molniya orbit illustration:"The Molniya orbit. Usually the period from perigee +2 hours to perigee +10 hours is used to transmit to the northern hemisphere." from here.

enter image description here

above: A sphere inside a cone, from http://mathcentral.uregina.ca.

  • $\begingroup$ I don't see why a Molniya orbit would need to be geosynchronous - there just needs to be satellites in position to cover the desired area at any time. The Wikipedia article on the topic doesn't mention geosynchronous. The diagram shows an orbit that is half of geosynchronous, and that is described as 'usual', not universal. $\endgroup$ – kim holder Sep 2 '16 at 14:57
  • $\begingroup$ While a Molniya is a good example of an orbit that combines LEO with a high-altitude portion, the question doesn't ask about that. An answer really needs to directly compare LEO and GEO. $\endgroup$ – kim holder Sep 2 '16 at 15:01
  • $\begingroup$ @kimholder I'm pretty sure that the term geosynchronous can be used as a broad term to cover other-than-1:1 ratios. An approximately 16 hour orbit could be synchronized to a 3:2 ratio for example, and still fall within the category of geosynchronous. The Wikipedia article for geosynchronous orbit does mention the "subsynchronous" Molniya orbit in the introduction. I've got to read the Molniya article a few more times to understand what's going on with $\Delta \Omega$ vs $\Delta \omega$ and the difference between a synodic day and sidereal. $\endgroup$ – uhoh Sep 2 '16 at 15:34
  • $\begingroup$ Note the use of a different word - semi-synchronous. :P Note also the part where the article mentions a specific radius for that orbit. $\endgroup$ – kim holder Sep 2 '16 at 15:40
  • 1
    $\begingroup$ Thank you, GSO instead of GEO also works for my purposes. The general idea I was getting at was substituting between the higher orbits in and around GEO and the orbits in LEO. $\endgroup$ – penGuinKeeper Sep 2 '16 at 17:21

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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