Disturbing the Universe

David L Clements, science and science fiction

Interstellar War: Part 1


There is a long history of interstellar warfare in SF, using both real and less real physics. Various discussions of space combat can be found at Strange HorizonsGizmodo and elsewhere, to name but three. Those three discussions cover a lot of ground, most of which I agree with, but with one notable exception. What I’d like to discuss here is the somewhat more specific issue of interstellar warfare – of how one might conduct a war between two solar systems separated by several light years. Hopefully, being an astrophysicist, I can bring some quantitative analysis to the subject.


Before we get started, various assumptions need to be clearly stated. The most important of these is that I will be trying to keep within the accepted laws of physics. Thus no faster than light travel will be permitted. If that were possible it would be a total game changer, and the details of how it worked would define the way interstellar warfare would be conducted. So, no warp drives, jump drives, stutterwarps, Guild navigators or hyperspace.

Secondly, I will take as an assumption that the resources to conduct this war are available in the opposed systems. This is not a simple assumption. Accelerating a ship to 10% of the speed of light (0.1c), which is what I’m going to assume our interstellar battleships are travelling at, takes a lot of energy. For a 1000 Tonne vehicle, probably the minimum size craft one might consider, that will take 4.5×10^20 J. That amount of energy is equivalent to the whole of the UK’s power output (about 110 GW) if dedicated to this purpose, and no other, for 130 years. And that’s just to get it up to speed – I haven’t included any allowance for inefficiencies and deceleration at the destination system. However, the total power from the Sun received by the surface of the Earth is about 2×10^17 W. If this could all be collected and used, you could generate the power required to accelerate our interstellar battleship to 0.1c in about 2 hours. Of course, you probably wouldn’t want to pave the whole planet with solar cells for this. Instead you’d build a solar collector the size of the planet. In fact, since you would want a large battlefleet, you would probably build several.

Thirdly, I will assume, for the moment, that magic nanotechnolgy pixie dust isn’t available. This, like FTL travel, would be a massive game changer. Our 1000 Tonne ship could be replaced with untold billions of nanomachines, any one of which, on arrival at the victim system, could construct anything it wanted from local resources.

However, the use of local resources in general, albeit using more conventional construction techniques, is something that I will allow, and in fact is probably something that will be essential.

Interstellar Warfare will be Asymmetric

In terms of sheer numbers and resources, the system being attacked will have huge advantages. There will be one or more inhabited planets, along with assorted habitats on moons, asteroids, and floating in space. In the event of a pitched, face to face battle, the overwhelming numbers in favour of the locals would mean almost inevitable defeat for the invaders, in the long run if not in any specific engagement.

However, the vastly superior resource base for the locals is also a disadvantage, since, as we shall see later, this makes for a very target rich environment.

Hide and Seek with Bazookas

I borrow this phrase from one of my favourite space battle games – Star Cruiser from the late lamented GDW’s 2300 universe. While SC uses FTL and other exotic physics, the basic idea that the main defence of a starship is stealth is a conclusion I agree with.

This isn’t universally accepted. Memphet’ran’s discussion of space combat concludes that the amount of energy released by starship propulsion would be easy to detect even at very great distances, and that the waste heat from a ship, even in cruise mode, would also be easy to spot.

Interestingly, recent projects in astrophysics have a lot to say about these problems.

Boost Phase Detection

Memphet’ran claims that an accelerating starship could be spotted as it boosts to cruise velocity in its own system, while still light years from its target. In a sense, this detection problem is equivalent to trying to detect the light of a terrestrial planet orbiting another star, and a lot of astrophysicist effort is current aimed at exactly that.

Let’s look at our 1000 Tonne battleship again. It needs to use at least 4.5×10^20 J to get up to speed. In fact it will be more than that, at least twice to account for exhaust gasses, unless something more sophisticated, like a laser-driven solar sail, is used (more on these in a moment). If we assume our invaders are coming from a sun-like star, then the power output of our ship’s engines has to be seen against a background stellar output of ~3.8×10^26 W. If our starship was to burn all its fuel, and go from a standing start to 0.1c in just 1 second (which would produce some rather extreme G forces), the ship’s energy output would amount to only about a 1 in a million perturbation on the stellar output. Spreading the boost phase over 20 minutes, which would still turn your crew into a monomolecular layer of raspberry jam, would turn that into a 1 in a billion perturbation. A more realistic boost phase, say using 1g acceleration, would take 34 days to reach 0.1c, meaning that the light of the engine would be essentially undetectable against the light of the star.

But wait, I hear you say, by the time that 34 day burn has finished, the ship will be something like 300 AU away from its starting point. Won’t that be far enough to be able to resolve its emission from that of its parent star? That is more technically feasible, except for the fact that the invaders know where their enemy are, and can arrange to have the sun at their backs for the whole of the boost phase.

This analysis also ignores the possibility of hiding the light of the boost phase with some kind of screen, which would probably be necessary anyway since otherwise you might irradiate your ship and crew. The laser-drive light sail, mentioned briefly above, is just a large scale example of such a screen, but with the ‘engine’ left behind. Any light overspill from a laser booster would be quite conspicuous – indeed searches for laser lines superimposed on stellar spectra have been proposed as a method for detecting extraterrestrial intelligence. In our case, though, we want stealth, and this can easily be achieved by oversizing the light sail, so that any weak, sidelobe spillover from the boosting laser would still be obscured by the sail.

We must therefore conclude that boost phase detection of a launched invasion fleet is not at all possible. The victims of the attack will be unprepared for what is going to happen.

In the next instalment we will look at the arrival phase and conduct of the actual war.


Author: davecl

Astronomy, science, science fiction

2 thoughts on “Interstellar War: Part 1

  1. Detection depends a lot on the bleed of power in the propulsion system, as an ideal one would direct it all away from the target. So its uncaptured laser light or orion particles that escaped the shield. If interstellar trajectories were dead straight you might hope for some shadowing, but it makes bleed detection at the edge of the target system easier. So if you had internal power or clever mirrors you might want to aim off axis a fair way to avoid the bleed going near the target.

    Also if you assume both sides had similar tech, then each would have seeded nearby space with detectors which are much cheaper to launch than battleships. Jolly hard to keep media lockdown on the start of an interstellar war too.

    So if the combatants know each other I suspect launch detection is very likely.

    • I’m envisioning that all of the boost to cruise velocity (assumed to be 0.1c here) occurs in, or within a few hundred AUs of, the home system of the invaders. There would thus be no bleed to detect at the end of the target system (unless I’m misunderstanding you).

      The deceleration phase will be close to the target system, but I’ll be discussing that in the next instalment.

      You do make a good point about the possibility of aiming off axis.

      I’ve been assuming that there isn’t much interstellar colonisation by either party at this point. The possibility of HUMINT (ETINT?) about the launch of the warships would thus not apply. However, even if the target system knew the invaders were coming, I think there wouldn’t be much they could easily do. The volume of space between the systems, and the volume of space in Oort cloud around the target system, is so vast that you couldn’t easily patrol it. I’ll be discussing some of the issues around detecting invading ships on arrival and afterwards in part 2. The generic description of space warfare as hide and seek with bazookas will continue to apply 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s