I understand the use of Metric measurement, the real question is: with how vast space truly is, why do we not use other prefixes like we do for computer memory?

Unit In meters In AU
1 kilometer $1\ 000$ meters (km) $6.68459\cdot 10^{-9}\ AU$
1 megameter $1\ 000\ 000$ meters (Mm) $6.68459\cdot 10^{-6}\ AU$
1 gigameter $1\ 000\ 000\ 000$ meters (Gm) $6.68459\cdot 10^{-3}\ AU$
1 terameter $1\ 000\ 000\ 000\ 000$ meters (Tm) $6.68459\ AU$


I get that we also have $AU$, which is more commonly referred to in terms of long-distance space travel (because it's the most relative thing we have). But I made a joke about megameters (Mm) earlier in a comment and started honestly wondering if there is a specific reason? I found this generic Quora answer (completely unrelated to Space Travel), is it actually that simple-- human relatability?

EDIT: Either we can close this or, for an "accepted" answer, I'm honestly looking for examples (opinion or otherwise) of these measurements being used, practically. For instance was there ever a mission done in anything other than AU/km/m? Was there a rejected proposal for software that involved measurements like this? I understand all of the arguments presented, but wonder if there's instances of this in existing software.

  • 3
    $\begingroup$ Very related: astronomy.stackexchange.com/q/20466/6 $\endgroup$
    – called2voyage
    Jul 5, 2018 at 17:58
  • 4
    $\begingroup$ Metric prefixes like M G T for the length unit meter allow to express distances in space easily, even for distances like many astronomical units or many lightyears. I can think of no other reason than tradition. Wikipedia has a very nice and very long list with samples for small and large distances. The diameter of the observable universe is 886 Ym. $\endgroup$
    – Uwe
    Jul 5, 2018 at 18:00
  • 2
    $\begingroup$ We use metric prefixes for electrical units like Volt, Ampere, Henry, Farad, Cycles. There is no logic reason not to use them for distances too. megameters and gigameters are part of the same logical system as kilometer and millimeter. $\endgroup$
    – Uwe
    Jul 5, 2018 at 18:14
  • 2
    $\begingroup$ I don't understand how the "correct" answer to this question could be identified, nor how other answers ruled out. How is this not a pristine example of primarily opinion-based? $\endgroup$
    – uhoh
    Jul 6, 2018 at 1:46
  • 2
    $\begingroup$ The design tools in the late 70's and 80's used E engineering notation for every calculation: "1.00E9" would print, referring to the base unit type. No prefixes visible anywhere. We were always going back and forth between the numbers on the machines and in documents, etc, so the practice was to always use the machine's "engineering notation" - powers of 10^3 in floating point numbers - instead of using milli/kilo/etc prefixes that might permit careless notation to make a conversion mistake. Not sure that was a written standard or just a practice, though. $\endgroup$ Jul 10, 2018 at 16:07

2 Answers 2


Astronomers were among the first groups that needed to make measurements that were impractical to express in the units then available. In 1900, distances in the solar system were measured accurately for the first time, and the Astronomical Unit was born. The SI system of units was proposed in 1875, but the SI prefixes ended at M for 106.

It took until 1960 for the list of prefixes to be expanded.

So from 1900 to 1960, to express '150 million km' you needed to write 150,000 Mm. With the SI system less established than it is now, it's understandable that astronomers settled on their own units.
The same goes for the unit light-year, which came into use in (1838).

The next branch of physics and engineering to need really big or small numbers, was electronics in the early 1900s. The SI system and prefix were universally adopted there (from pF to G$\Omega$ and THz). So the "risk of screwing up a unit conversion" argument doesn't sound too convincing to me.

What we see these days seems to me to be a result of inertia (with astronomers being unwilling to switch to SI because AU/ly/parsec is what they're used to). Some astronomers are advocating to switch to SI.

As it is, astronomers use a series of units based on physical objects (AU, solar mass, parsec) while the SI units have mostly been redefined to be independent of physical objects. The AU and solar mass both change over time, so these days they are defined in terms of SI units instead, making them a derivative unit.

  • 1
    $\begingroup$ "So to express '150 million km' you need to write 150,000 Mm." That is not necessary anymore, there are now (since more than two decades) SI system prefixes like Gm and Tm and more. So '150 million km' is 150 billion meter and therefore 160 Gm $\endgroup$
    – Uwe
    Sep 6, 2018 at 15:48
  • 1
    $\begingroup$ The point I was making: until 1960, you needed to write 150,000 Mm. $\endgroup$
    – Hobbes
    Sep 6, 2018 at 16:37
  • $\begingroup$ Ack! I never upvoted your answer for this, I apologize. It was well written and showed a lot of cool stuff about the history of units. It's logical, when you don't have a name for something, you use something that's relative and makes sense! $\endgroup$ Sep 6, 2018 at 19:59
  • $\begingroup$ Another difficulty is the SI, right now, may still need a few more prefixes added to make it best for astronomical "prime time". While for the dimensions of length and time things are pretty good, mass and energy not so much. The largest SI multiple unit for mass is a yottagram (Yg), or $10^{21}$ kg, and already the Earth's mass is 5972 Yg. To get to a replacement for solar mass ($M_\odot$), that's now ~$2 \times 10^9$ Yg. That's 3 more prefixes needed above Yotta to make for a convenient number. $\endgroup$ Sep 7, 2018 at 6:49

I think I may be able to provide an answer to this: as @Russell Borogove said, units are an engineers bread and butter and an engineers bane, and the more their are of them the more likely there will be a mistake (take a certain infamous NASA/ESA mission that got broke because of the lbs/kg divide!). Even highly trained pro's make these seemingly "silly" mistakes.

Secondly, and this is just my opinion, but I think it has a little to do with sort of "average scale" (a phrase I just made up). The most common size of things in the solar system we cared about at the beginning of space exploration was measurable in km (for instance - radius of Earth is approx 6000km). Introducing Mm would've served only to make writing (which is how the pioneers did it) such measurements annoying (rEarth being 6e-3Mm, and believe me, constantly writing x10^blah on paper is annoying!)

Added to that, thousands of km is imaginable in the human mind as the size of certain countries/oceans, so slightly easier to grasp.

Asteroids are hundreds of km, planets are 1000's of km, orbits around planets are 100's of km - the next thing to think is interplanetary distances. Which are in the region of Tkm's, in your units. Here's where relatability comes in again! We as a kind have no concept of Tkm's or billions/millions of kilometres, and when we don't have a concept of something, any problems involving it just get harder in our head. However, if we can relate it to something we know (like Earth - Sun distance), then we can get our heads sort of around it.

So you see, because of the jumps in measurements from 1000's km to AU's, and because of relatability, the inbetween units do not have any use (Mm,Gm,etc) - so that's probably why we don't use them. My first answer, hope you found it useful!

  • 5
    $\begingroup$ rEarth being 6e-3Mm - that's not how SI prefixes work. You use either a prefix or exponent notation, not both. rEarth is 6 Mm, by the way. You don't use 2 prefixes either, so no Tkm (you use Em instead). $\endgroup$
    – Hobbes
    Aug 7, 2018 at 5:59
  • $\begingroup$ Ah, my answer was littered with mistakes - which kinda further illustrates my point. $\endgroup$ Aug 7, 2018 at 11:06
  • $\begingroup$ Also, I was trying to use the unit prefixes the questioner provided in the example to illustrate my answer, the point being they don't really belong. $\endgroup$ Aug 7, 2018 at 11:07

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.