Monday, June 15, 2009

Space Warfare II - Stealth Reconsidered

Second in a sporadic series.

Space is vast and dark, so for diurnal creatures like us it seems like a good place to hide in. But the darkness of space is exactly why it is mostly not a good place to hide: Objects such as spacecraft tend to be bright, if not in visible light then in the infrared. Drive engines are really bright, as is any large power source not perfectly efficient. Thus the popular rule of thumb in haunts like SFConsim-l, that 'everyone sees everything.'

In the rocketpunk era, before automated detectors, this was not so much of a problem. Human lookouts get bored and nod off or goof off; even if conscientious they may slip into a sort of highway hypnosis and miss a faint 'star' drifting across the starfield. Automated scan is annoyingly non-deficient in these respects.

Physics and technology are similarly unfriendly to decoys. For an accelerating decoy to be mistaken for a ship it must have the mass of a ship, or else passive scan will reveal a suspiciously low power backwash. But if the decoy is as big and heavy as a ship, with a drive engine as powerful as a ship's drive, it will be more or less expensive as a ship, and you may as well go the rest of the way and arm the thing.

A great many electrons have been spilled in vain trying to wiggle out of these conundrums. Of course with a sufficiently large array of oscillating hands you can defeat any inconvenient physics; tactical FTL for example, or (less problematic from a causality standpoint) discontinuity fields that space subs can dive into.

For this series of discussions, however, I am ignoring outright magitech and concentrating on plausible midfuture technologies using known physical principles. Yes, 'plausible midfuture' is a slippery phrase (and a topic for a future post), but it gives us something to hang our hat on. The realm of coilguns and fusion drives is at least defined enough for people to know what they disagree about. In this realm the arguments outlined above are pretty robust.

That said, politics is more complicated and difficult than physics, and so is warfare. Targets will go out of their way, perhaps literally, to be unhelpful. 'Stealth' in the currently popular sense of making yourself invisible to tracking tech looks like a nonstarter, but there are many other ways to be stealthy, even in space.

In the comments to Part I of this series, commenter Mark brought up orbital decoys. The relationship between mass and thrust that gives away low-mass accelerating objects does not apply to non-accelerating objects, in orbit for example. Long range scan then reveals only how bright things are, not how heavy.

This generally is an advantage to the defender. We can assume that in a space setting, important planets (or moons, whatever) have a lot of orbital clutter, from major stations to space junk, as well as defensive forces and decoys. It will be hard for an approaching attacker to tell which is what. Even visual imagery may not be helpful, if warcraft are not obviously and outwardly different from civil types (or can be made so). The Beziers solution, blow it all away, may not be practical for a variety of reasons.

For deep space attackers hiding in the clutter is problematic, but there are other forms of stealth and deception. In Heinlein's Between Planets the Venus rebels seize Earth's military-civil orbital station, including its stockpile of nukes, by arriving aboard a scheduled space liner. Better security measures should have foiled this, but you can say that of many successful operations, and it is not hard to come up with scenarios for military raiders approaching on civil orbits.

For major forces this is harder to bring off. Vast space armadas probably won't be mistaken for Hanseatic merchantmen, but there is the opposite ploy of using relatively cheaper civil craft (perhaps obsolescent ones) to bulk up a task force – or simulate one, drawing the other side into a wild goose chase. How this works depends in part on secondary factors that are dealer's (or author's) choice. Drive engines may have a distinct signature that identifies models or even individual ships, but this is by no means a given. Just don't have 'drive signature characteristic of a Vikrant class cruiser' and 'estimated drive power of 2.5 to 7.0 gigawatts' in the same story. (Or at least have a quick explanation for the differing specificity of results.)

All of these deceptions are on the operational scale, but there are also smaller-scale tactical or 'micro-tactical' deceptions. Opaque 'smoke' can throw off precision targeting. The hidden ship presumably can't fire lasers through it, but it can launch missiles through it (at least if some other ship is providing targeting data). A small target-seeker warhead probably does not carry a scan suite equivalent to that carried by a large warcraft, and cheap flare- or chaff-like decoys only need to confuse its sensors for a few second, and create a few meters' worth of error, to turn a hit into a miss.

In space, everyone sees everything. But in war, no one can be quite sure what they are seeing.


Related links: Part I of this series is here. And some of these points grew out of discussion of space piracy here, here, and here.


Update: This post also generated discussion at SFConsim-l, much of it covering different ground from the comment thread. Yahoo Groups has lame threading, but the discussion spins off from my pimping post here.

100 comments:

Kedamono said...

I ran into this problem when I was working on a section of FTL:2448 that dealt with infiltration of a hostile alien sphere of influence back in the late 1980's. The "Stealth" ships were more like subs, in that they were spheres, using their supercooled fuel tankage as a heat sink and camouflage. No, it doesn't work, but at the time, as I was working out the numbers, I was amazed at how bright a ship was. I chose spherical ships, as the sphere has the smallest apparent size when it comes to radar. Then I dithered between a mirror finish and a matte black finish. I went with the black finish.

Of course these days I realized that it doesn't make a bit of difference. The "stealth" ships would still be visible.

So when I worked on my FTL:2448 novel Memories Long Forgotten and wrote a section that dealt with these aliens, I made use of the fact that the alien "empire" was actually more akin to warring baronies, just as likely to shoot at each other, and one of the barons wants to trade with the Humans. So a freighter does what a multi-trillion dollar stealth ship couldn't do.

Oh, and the Stealth Fighter and Bomber are only stealthy to ground base or horizontal radar. If you could bounce a low power radar beam off the ionosphere, you'd see them as plain as day. A sphere is the best shape, because except for dead on hits, everything else bounces off at an angle. And you can improve on that by using radar absorbent material for the hull's skin.

Rick said...

It is always sort of shocking when you do the numbers on how BRIGHT drive engines are. (Night launches from Vandenberg are way cool! Well, so are daytime launches, for that matter.)

It's so natural to think that black ships would be hard to see!

Your workaround for FTL:2248 is rather elegant. As we've learned off Somalia, one thing sophisticated sensors can't register is crew motivation. Back in the days of John W. Campbell, you'd have had psi specialists to handle that part.

As for stealth aircraft, I guess we don't see spherical ones because spheres are not very well suited to fast atmospheric flight. :-)

Anonymous said...

There's also strategic stealth. You spend a lot of time, effort, and money making a great deal of noise about your plan to conquer Ganymede, but at the last moment land your forces on Rhea. Or you spend that time, effort, and money on making sure that your plans don't get out, and that the people of Ganymede think your shiny new peace treaty will hold up and they support your plans to invade Rhea.

Politics and orbital clutter seem to be the best options for stealth, barring superscience. Other than that I suspect stealth will boil down to concealing what's going on inside the craft, hiding all interior EM emissions so sensitive spy arrays don't tap into internal communications.

Low-emissions/low-reflectivity coatings might also work on missiles. You're not trying to hide the missile itself, just hide what class of missile it is and what range of warheads it could be carrying. Throw in flares and hot reflective chaff and the defending ship won't be able to prioritize its point-defence efforts.

Long-term strategic misdirection and short-term sensory overload, but no invisible ships. Sounds a bit like modern naval conflict.

Ian_M

Rick said...

Damn, I completely left out the strategic level of deception!

I suspect that sub types would disagree about modern naval warfare, but otherwise your comment makes the point exactly.

Kedamono said...

Ian, sure, who needs stealth ships if you can use politics to do your battles? If you can figure out how to influence a "friendly" political group on a world, you might end up with a local government that is friendly to the Terran Federation/UN/United Earth.

Of course you have to be careful not to let the locals know you're backing one group or another. Nothing can kill a political group than knowing it's being backed by the Evil Empire™.

Of course you'll still have the military mavens who will say "You always need boots on the ground to conquer someone."

And the astute politician will point out, "But they don't have to be our boots."

Marktheother said...

A legitimately civilian cargo carrier is extremely dangerous on its own, all it has to do is jettison it's cargo before it begins decelerating, leaving it's impromptu kinetic weapons to cut a swath before it. (and who says it's cargo isn't made up of space combat drones)

An attack could be proceeded by an otherwise completely legitimate and scheduled cargo fleet, it probably wouldn't work twice, and any ship coming from the attackers would be suspect any where it went, but it would probably be devastating the first time.

-Mark

Nyrath said...

Re: orbital stealth
Correct me if I am wrong, but I think that technically one can determine the mass of a satellite by using Kepler's third law.
In practice, though, the mass difference between the primary planet and the satellite is so great that the measurement could not be made with sufficient accuracy to determine the satellite's mass.

Rick said...

Mark - Yes. In practice, I imagine that civil approach orbits will be laid out so that cargo pods full of rubble will not hit important stuff - but if the cargo pods contain target seekers ....

The ploy you mention shouldn't work twice. But (though the analogy is hardly exact) think of Pearl Harbor. We were caught napping even though a) US naval aviators had repeatedly 'attacked' Pearl in fleet exercises during the 30s, AND b) the Brits did essentially the same thing at Taranto only 6 months earlier. Sucker punches work, because human beings are suckers.


Nyrath - You are correct both on theory and proviso. You might be able to determine that a gigantic balloon was not a real Death Star, but for dinky little million ton spacecraft, not so much. :-)

Rick said...

Kedamono - (Oops, got out of sequence.) Venturing into the Real World [TM], note that your point applies very much to the current situation in Iran. The US has to step very lightly to avoid too blatantly favoring the side we obviously favor, to keep from giving the other side a free shot at linking their opponents to the Great Satan.

Re 'boots on the ground,' this gets into a really major issue that I'll be taking up in a future post. (Natch!)

Short form: Historically, maritime powers have tended toward indirect control, because a) ships can't wear boots, and b) marines, who can wear boots, are by general military standards exceptionally difficult to deploy. But space powers seem inherently 'maritime,' since armies can't march from planet to planet.

This should have fairly fundamental implications for interplanetary (or interstellar) power politics. A significant minority of military/political SF embraces this, but by and large the mainstream does not.

Sundog said...

It seems to me, that if you're willing to spend the money, decoys become valid.

What, after all, do you see in a long range scan? You get approximate heat emissions, from which you may be able to guess the amount of thrust, and speed of acceleration, which, with the thrust, should give you a range of mass. Hopefully, you can then match up that approximation to the solid numbers your naval intelligence division can give you regarding the opponent's fleet to know what you're actually facing.

You might also get wash from his active sensors, and be able to determine information from them - type and capability of his Radar, for instance.

However, what's to prevent your opponent building drone ships, with the mass and acceleration of his real ships, but just metal shells with weights and a remote control reception system? They'd be expensive, but MUCH cheaper than actual ships (no life support, no weapon systems, no sensors...) and best of all, they'd still be useful even if the enemy knows about them.

Consider the scenario: you have twelve warships, all of which will be needed to subdue one of the enemy's outer bases. You also have twelve decoy ships. You send your invasion force at base A, and the decoys at base B - on the other side of the system. He must choose whether to send his reinforcements to A or B, and he has to commit to the burn long before you reach either. If he sends them to A, you break off. If he sends them to B, the Decoys break off - and you take A. If he splits the two, he is inviting defeat in detail, and potentially losing far more than just an outer base. It doesn't matter if he knows one of the forces is a decoy - he doesn't know which one, and thus must treat both as being the dangerous group.

Anonymous said...

Like I've said before: if you can't use stealth, then do the opposite; blind them and confuse them. Obscuring what type of weapon by using special hull coatings; using chaff, flairs, and decoys; using EW to overload, 'spoof'(creating ghost sensor imagies), and outright blind your enemy's sensors to overwhelm their defenses. The first one who can't see the incoming missile/penitrator rod/war-drone may not get killed, but may lose it's abilty to effectively fight back. If you're blinded, you have to withdraw or surrender.

Picking out the wheat from the chaff (i.e. DefSats from orbital junk or civilian satellites) is a daunting and time consuming task; even for a high speed computer controlled targeting system, you might still get zapped before you can shot them all down; swapping a super interplanetary battleship for a (relitively) cheap single-shot satellite doesn't seem like a fair trade.

Using masking to keep people from being able to tell the difference between civilian and military ships is interesiting; makes it hard to tell if the ships comming toward your colony are a civilian convoy or a military flotilla.

"Of course you'll still have the military mavens who will say "You always need boots on the ground to conquer someone."
And the astute politician will point out, "But they don't have to be our boots.""
Of course, at this level of technology, you should have a choise of organic or steel boots. UAV's and UGV's would go a long way to suppliment human soldiers. Space warfare isn't impossible, but nor is it easy; the outcome (like any armed conflict) can't be decided until the shooting is over.
Ferrell

Anonymous said...

"Of course, at this level of technology, you should have a choise of organic or steel boots. UAV's and UGV's would go a long way to suppliment human soldiers."

For an invasion or short-term occupation those are great tools. Recent experiences in Afghanistan and Iraq suggest that in long-term occupations those tools just piss people off. Even the US-style vehicle patrols, with real soldiers in the vehicle, seem to be less effective at making the sort of connections that boots actually on the ground build. And UAVs can actually be used against you in the propaganda war; "Those faceless cowards refuse to come out from behind their machines."

The machines have their uses, but human faces seem to be better suited for long-term occupation and patrols.

In space-warfare your wheat from the chaff problem gets even worse when you realize that the chaff is hot. Very hot. And moving fast enough to damage any sensitive equipment that gets in the way. It's a minor issue, but you're still going to have to either deal with the mess to avoid sensor-damage. In that sense it's almost like pepper spray.

Ian_M

Rick said...

Sundog - The (not-quite) counterargument re thrusting decoys is that once you've got spent the money to put a drive engine on it, it might not cost that much more to weaponize it. Depending on how much drive engines cost relative to other stuff such as sensor packages and armament.

In kinetic-dominant space warfare the whole situation is ambiguous. A self propelled decoy is automatically a kinetic missile: Just steer it into a target. Add a small bursting charge to get a cloud of kinetic fragments (since a 100 ton decoy/missile is huge overkill against most targets; even at 10 km/s it packs a kiloton of kinetic punch).

You can easily have a situation where a vehicle can be deployed as a recoverable drone, a kinetic missile, or a decoy, depending on loadout and mission.


Ferrell - 'Dazzle' is an effective and underrated antisensor technique, probably underrated because of the contemporary fascination with technical stealth.

And here's something to think about. Suppose you have weapon lasers that can physically burn through and wreck stuff at some range, say 1000 km. That same laser must be able to fry delicate sensor elements at a much greater range. So laser armed forces might systematically blind each other long before they get into physical damage range.

This also applies to target seeker missiles. Zap it with a laser and it is Enlightened, and seeks no more. :-)

As a further consideration what happens when lasers zap straight at each other? The main telescope mirrors concentrate the beam right onto the laser. I don't know what happens, but I suspect that laser battles might become in effect staring contests.


Ian - For occupation I lean toward your view, but robotics do become multipliers. If a patrol takes fire from a building, it's nice to be able to send robotics into the building instead of troops. Admitting that there are limits when the building is also full of terrified civilians.

Partly this is a situation where doctrine has not caught up with technology. Plus, we 'Murricans are really good at some things, but amazingly naive about others. Someone says 'Those people are Taliban,' we're not so good at realizing maybe we were lied to, and they're just local rivals of the informant.

On the other hand, and although the Iraq war was an incredibly bad idea, some midlevel US commanders have proven to have a totally unexpected talent as Ottoman pashas, sitting down to coffee with local shaikhs, and succeeding in turning a total strategic botch into a mere muddle. Of course it helped that 'al Qaeda in Iraq' was such a vicious bunch that they were damaging the brand.

Kedamono said...

In a previous comment, you said Rick "As we've learned off Somalia, one thing sophisticated sensors can't register is crew motivation."

Which reminds me of the RPG Traveller and it's "Anti-Hijack" software, which was described to make it hard for stowaways and hijackers from gaining access to the ship, as well as identifying potential hijackers based on their actions.

Rick said...

Kedamono - Very odd - why would they come up with a rather implausible tech that has the sole functionality of killing plots?

Kedamono said...

Rick, they came up with this software, which has to be loaded into the small computer's memory to be effective, and it just gives you a chance to detect such activity. Most of the ships in the game have computers with such small memory spaces that they can't run the Anti-Hijack software and maneuver effectively! A 1-bis computer is tiny indeed!

Anonymous said...

'Plot' in an RPG is very different than plot in a story. I've never played Traveller, but as an RPG geek I'd say the Anti-Hijack software simultaneously gave ship-owning player-characters a feeling of safety while also giving those same player-characters one more obstacle to overcome when stealing someone else's ship.

Back on topic - Decoy ships not outfitted as warships may be a result of logistics. If the shipyards can churn out hulls and drives but the arms factories have a shortage of parts/volatiles for weapons, or the electronics factories can't build tracking/targeting hardware fast enough, then empty hulls outfitted with drives and filled with waste-mass may be all they can build.

Politics can come into it as well. The German High Command wasted material and money keeping the factories pumping out consumer goods up until late 1942. That kept the German middle class happy, but also kept the Allies feeling pretty good.

Ian_M

Rick said...

I've never played Traveller (or any RPG except for a bit of D&D eons ago). I guess I'm just reflecting my bias that motivation is a human-factors thing that only a pretty highlevel AI could analyse. But as explained it makes sense.

True that if I've got drones and no sensor/weapon packages for them I can at least use them as decoys!

Regarding Nazi Germany, I recall reading that when New Dealers got there in 1945-46 their jaws dropped at the lameness of German economic mobilization.

Anonymous said...

Did you change your site from
http://rocketpunk-manifesto.blogspot.com/
to
http://www.rocketpunk-manifesto.com/

My bookmark is being redirected and I want to make sure it's legit.

Ian_M

Rick said...

Yes, I did - I guess I should have mentioned it!

Anonymous said...

Fixed.

The jaw-drop response is common among people who study the decisions of the German High Command. Continuing to churn out tanks and aircraft when they didn't have fuel for existing vehicles (Including the navy), forcibly relocating ethnic Germans living outside of Germany to Poland, looting the homes of ethnic Germans outside of Germany and sending the loot to middle-class Germans in Germany, the list of self-defeating decisions goes on and on...

Actually the Second World War brings up another form of strategic stealth: Have greater resources and industrial capacity than your enemy thinks is possible. This is how Japan decided to pick a fight with a nation with more people, iron, fuel, food, water, money... I suppose you could call this Invisibility to Idiots.

Ian_M

Rick said...

Don't forget despising 'Jewish physics.'

Yamamoto, as I recall, warned the Japanese leadership of what they were getting into, and of course was ignored.

The best stealth of all is indeed choosing self-blinding enemies.

Sundog said...

Unfortunately, opponent-wise you're just as likely to be facing a Napoleon as a Hitler - or worse, a Rommel working for a Napoleon!

It seems to me, Stealth as such (low-observability) is just as dead as ever. What we're really saying here is that Deception and Tactical and Strategic Surprise are still alive, well and bollixing up the enemy and yourself as ever.

Rick said...

Alas, true; you can't choose your enemies. Or if you can, it probably means you're making 'wars of choice' - and down that road lies an enemy who will surprise you. Disagreeably.

And yes, I fudged a bit on the title. Stealth is dead; long live surprise and deception!

Daniel said...

Repost:

With all due respect, your article seems to be operating under a number of assumptions that are far from assured.

1) The ability to detect the advisory. My objection here is not at the physics of it, but at the assumption that the ability will be on hand. Ship/station mounted telescopes would have to deal with background vibration from onboard equipment, compensate for movement through space, and most importantly limited resolution. One can attempt to compensate for the first two (though the end result would be similar to the operation of the human visual system, which is frankly not that good) but the last is restricted by the size of the craft. The baseline for your interferometer won't be longer than the largest dimension of the craft, which effectively caps your abilities within any technology. Beyond a certain range, you are going to know there is a bright spot there and nothing more, because you cannot generate a better image. This is not very helpful.

While most scenarios get around this by positing the existence of a large network of automated tracking telescopes, this also has severe problems. While such a system can effectively reveal everything in the star system, the presence of such a network is unlikely. Firstly if such a system exists in the scenario, there is no guarantee that it actually impacts it. It may belong to a neutral party. In fact one could well argue that this is the most probable scenario as the neutral ones may be the only ones left after the shooting starts. Your remote telescopes would be emitting signals by transmitting information to you and radiating heat while operating. They would also be in relatively stable, constant positions. This would mean that their location would be easy to determine. So given the advantage such a system offers and that you know where the enemy's is, one of the first actions any commander would take would be to target and destroy it.

Additionally, one must factor in the time it takes to pass the information along. The images would need to stream to the information cells, be compiled and analyzed (which given computer speeds may be minimal depending on the efficiency of the intelligence agency) and then sent out to the remote ships. If you are getting signals from a telescope near the moon that may not be a problem, but if you are talking about Jupiter you will be running over a hour behind.

Daniel said...

2) The assumption that detection matters. Raw intelligence is different from actionable intelligence. Simply being able to see what they are doing doesn't actually help that much unless you can act on it. We've seen similar scenarios multiple times in the past few decades, where a government may discover that a meeting is taking place at a certain time and place, but cannot get a plane or missile there in time to take out the target. Just as in real life, unless the ships are able to fire upon each other as soon as one is seen, knowing that the enemy is there doesn't really change things very much. It just alters the scale of force that must be brought to bare. For another example, try playing an online version of Risk where you can engage the "fog of war" and one where it is disabled and compare. The game changes very little.


3) That a warfleet will not be confused with "Hanseatic merchantmen". This is an implicit scenario assumption - you have previously written about how technology and "power density" come into play, but here you disregard your own notes. At the varying extremes of the scale, there isn't really much of a difference in an armed merchantman and a cruiser. And within the realm of space warfare this seems to be the more likely case. Any ship capable of moving a considerable mass at a high velocity (as a cargo ship would be designed to do) must also be capable of generating and directing a considerable amount of energy and discharging it at a high rate. Going off remote detection, distinguishing the 1000 ton warship from the 1000 ton freighter is going to be very difficult and ultimately academic, as both would have the capabilities to be a sincere threat

Daniel said...

>
> These factors could potentially handled by letting telescopes fly free for observing, and having several fairly well separated ones. What is achieved in actual practice may vary.

Off the top of my sleep deprived head, this strikes me as a potential source of some kind of space fighter. Have telescopes be on returnable drones to save you on mass and man-hours, enemy tries to shoot them down to deprive you of the range advantage you would enjoy when you engage, at that point game theory and economics take over how the design changes. You'd have fighters that would keep trying to shoot each other down for dogfights, but would be trying to get images of the other ship rather than attack it explaining the firepower difference. I'll have to give it more thought after I get some sleep.


>
> The argument I've gotten from telescope geeks is that the required network is not all that big by defense-budget standards

I don't doubt it, and given the sheer advantages in intelligence gathering, even if it was a major expense it would be worth it. The problem is that the enemy knows what a major advantage it is, and they are essentially fixed, remote, undefended targets. Blinding you would be one of his first priorities.



>
> Quite true! But I'd argue that this is a different problem from 'stealth.'

Ah, but stealth coverings are multifunctional - while they prevent the enemy from seeing you coming (which is null here) they also make it very difficult to get a radar lock to shoot you down (preventing them from being able to take action on the knowledge that you are there). But that is splitting hairs.


>
> What I really had in mind here was Really Big Fleets - hence the allusion and link to Drake and the Armada. You are right that individual warships need not be obviously distinguishable from merchies at a distance, or for that matter task forces from convoys.

Ah, I was thinking more in the midfuture still, when ships are basically bricks atop torches, and what you stuff inside the brick is the only difference.


>
> > First post, hope I have struck the proper tone for this forum.
>
> Absolutely - my only grump is that I hope you'll also post this to my blog comments! :>
>

Done, and I will cross post this as well.

Rick said...

Daniel - Thanks for reposting here!

(Note: This is crossposted from SFConsim-l, message #80067 - go there for additional discussion!)

1) Free-flying telescopes these 'should' be well beyond tactical strike range of the enemy, and have already performed their early warning function before they could be hit. How things work out in practice could be more complicated.

2) The last point is also applicable to how much of this stuff there is. Telescope mavens can say a lot more than I can about the cost of a surveillance system. But I'll acknowledge that the parties to a scenario may have a lot fewer resources, relatively speaking, than the US military. What sort of scan suite do mercantile craft carry?

Flip side: In an interplanetary setting (or an interstellar one with well-constrained jump points), strategic surveillance can concentrate on logistic networks. Mars might want to construct a secret base somewhere in the asteroid belt, but ships have to leave Mars to get there. Or even if the Martians route their logistic operation via neutral Europa, traffic leaving Europa to 'somewhere in the asteroid belt' can be detected as it leaves Europa on an orbit not headed for a known facility.

On the third hand, this gets to the resources, technical and human, available for strategic surveillance. Search can be automated, but analysis mostly cannot be. Or else the AI required is so high level that it is the moral (or amoral!) equivalent of an intel officer.

Regarding radar, I don't really see it playing a big role in space operations. But that goes deep into the technical weeds.

3) Ships may indeed be 'black boxes,' with civil and combat craft indistinguishable from a distance. (Or even fairly close up; how do you tell a cargo pod from a missile pod?) What will stand out as suspicious is ships on unusual orbits.

Stealth situations in space can't be totally ruled out. What is the Clarke story of a guy in a spacesuit, on one of the moons of Mars, who eludes a cruiser searching for him?

But broadly, I think that space is much more favorable to 'hiding in plain sight' than stealth in the usual sense. The more traffic in a region of space, the harder it is to monitor.

And as I noted in my 'Tough Guide to the Known Galaxy,' a well placed bribe to the operator can defeat any scan technology. :-)

Anonymous said...

"(Or even fairly close up; how do you tell a cargo pod from a missile pod?)"

The missile pod has some poor Able Spacer crawling over it, scrubbing the pod with a toothbrush as part of punishment duty for that incident with the captain's son...

Any hypothetical warship will almost certainly operate as part of a fleeet. Even if that fleet is its own launches, drones, or other subcraft, that greatly extends the size of the observational array. As the battle goes on the array will be degraded, partly due to material losses but also to ECM. Lose of observational capacity may be considered a valid reason to withdraw from the conflict.

Another point, this one from the Atomic Rockets' discussion of tactics: Baring superscience space conflict may be limited to a 1 lightsecond effective range (A bit less than 300 000 kilometres). Even a low-tech hypothetical warcraft can move nearly ten meters with a 1-second burn. That's enough to disperse energy weapons over a large surface area (Or in some cases avoid it altogether), force incoming missiles to burn fuel re-targeting sensitive points, or otherwise maneuver to reduce threats from incoming fire. A 1-second delay between 'fire' and 'hit' is a long time. Even flesh-units can dodge attacks in less time than that.

So for the kind of near-future/plausible midfuture tech Rick is talking about here - What kind of resolution can we get at ~300 000 KM or less? How large does the virtual array of warcraft, drones, and scouts need to be to achieve targeting solutions at those ranges?

Ian_M

Rick said...

Especially with beam weapons, which I suspect particularly target sensors (and each other), sensor degradation could be a very common reason for wanting to bail out of an engagement.

As for sensor - and beam weapon - resolution, someone around here probably knows the formula. I had it around somewhere, but am not sure if I had it right. The formulae for telescope resolution and laser spot size are similar, but not identical (IIRC).

But from memory, a laser near the visible band firing through a 5 meter telescope will give you a spot size of a few cm at 1000 km. Like a killer SLR camera, you most likely aim the laser through the same optics, so target resolution is also a few cm. At uber-ranges like 100,000 km you'll need enormous power, humongous optics - or a very short wavelength laser, such as an X-ray laser.

With fast moving kinetics, required target resolution can be quite coarse. If impact speed is 100 km/s, impact energy is about 1000 Ricks, i.e. 1 ton of TNT per kg of impactor mass. For this speed range my weapon of choice is a kinetic fragmentation missile that can sweep out an area up to hundred of meters across You don't need precision targeting for hypervelocity grapeshot!

Anonymous said...

"Like a killer SLR camera, you most likely aim the laser through the same optics, so target resolution is also a few cm."

Laser optics as secondary targeting systems would probably work, but I'm leery of using them for primary systems. Redundancy is a big advantage in battle. If you have a targeting scanner and a laser that can also be used for targeting, you can lose the laser and still use the scanner for another weapon or lose the scanner and still target with the laser optics. But if you combine them as a single package, then losing that package is a disaster.

I suspect that one of the major visual differences between a merchant ship and a war ship will be sensors. The merchant ship can and probably will get by on one or two sensor packages, to control costs. The warship will probably have three or four primary packages and multiple smaller packages on each 'side' of its hull.

And merchants will probably not have a lot of hatches on their ships. Every hatch is a slow leak, and every slow leak is $$$ draining away in volatiles that need to be replaced. Warships will have pods, hatches, and bays to keep weapons and scanners out of sight during peacetime. Why give the enemy easy intelligence by hanging everything out in the open? There might even be workspace behind the hatches so drones or crew can repair equipment during battle.

Ian_M

Rick said...

As usual, how these things works out in practice depends on granular technical details. But the main argument for aiming the laser using its own telescope is 'because you can.' It provides hi-res imagery, and is guaranteed to be pointing the way the laser will zap. It's the equivalent of putting gunsights on the barrel.

You'd also surely have separate optics for target acquisition - my comments were about the fine-precision aiming of the laser.

I mostly disagree with warships carrying multiple 'main' sensor packages, simply on the presumption that you'll put on the most powerful main sensors that the ship can carry / that you can budget.

This relates to beam armament as well. The factors that made battleships carry several main battery guns probably won't apply, so I'd expect laser warcraft to be more like tanks in that regard, armed with the single most powerful zapper they can carry, plus PD armament.

Generally I'd expect warcraft and cargo carriers to differ much as you describe. If deceptive outward appearance is a major concern, designers will do what they can, without too much compromising, to disguise at least some types of military craft.

Slightly meta to this, but in a future post I'll be taking up how much / whether space weapon platforms will fit the template of 'warships.' (I tend to think probably not that much.)

Anonymous said...

I still think that until you have a situation where dedicated combat spacecraft are common, the main difference between a merchant ship and a warship will be the weapons load and the addition of sensor/PD drones/manned daughter ships. I doubt that a cargo container would look much different (at a 1000Km), than a missile-pod or a pop-up beam-cannon. You could even put some company's logo on it.

"This relates to beam armament as well. The factors that made battleships carry several main battery guns probably won't apply, so I'd expect laser warcraft to be more like tanks in that regard, armed with the single most powerful zapper they can carry, plus PD armament."

I'd think that unless your spacecraft is fairly agile, you'd want to have more than one beam turret. Otherwise you'd run the risk of losing your only offensive weapon; so unless you want your warship to fight the same way as a zepplin-sized jetfighter (your Lancer concept); point-nose-and-shoot... Maybe you're just thinking too small when you dismiss space fighters... :)

The only way to actually determine what a ship or ships are doing in an off-the-beaten-track location (building a 'secret' military base or a new mining/research station) is to go out there and see for yourself.

Ferrell

Rick said...

Other things equal, I tend to agree about missile-armed ships. Other things might not be equal if, say, either the sensors or the missile buses are big, and can't be disguised as cargo pods.

On beam weapons, the advantage of one big main zapper is not just more punch but longer range - the bigger your main telescope mirror, the farther you can zap with a concentrated beam.

Anonymous said...

"On beam weapons, the advantage of one big main zapper is not just more punch but longer range - the bigger your main telescope mirror, the farther you can zap with a concentrated beam."

So, your combat spacecraft design is to build a ginormous laser, put a cluster of engines & propellent tanks on the end, stick-on a couple of radiator fins, encircle the mid-ships with a habitat torus, place multiple RCS clusters just inbord of the engines and main emitter, and then stick PD and sensor mounts all over it? You do know what that sounds like, right? :)
Ferrell

Rick said...

You do know what that sounds like, right? :)

A Victorian torpedo ram, like HMS Thunderchild?

I suspect your intended answer is 'space fighter.' But the defining characteristic of a classic space fighter isn't really a keel-mounted armament, but a swaggering top-gun pilot. :-P

Rick said...

Ferrell - And now I wonder if that I took more than the daily recommended dose of stupid pills, because re-reading your comments ...

Death Star? :-)

Anonymous said...

:)

Ferrell

Rick said...

Which also does bring in space fighters, though, since we know they're the only thing Death Stars are vulnerable to!

Anonymous said...

"Which also does bring in space fighters, though, since we know they're the only thing Death Stars are vulnerable to!"

So, the brave robots piloting the multi-digit-G-drawing-nuk-armmed space fighters belly up to the bar and order high-end motor oil to swap tales of blasting Death Stars and flirting with the AI barmaid...
Ferrell

Rick said...

LOL

But lurking here is an interesting question about sentient AIs. What will be their pleasures and vices?

Not the least pleasure that I get from the comment threads here is that they supply me with material for future blog posts!

Jim Baerg said...

Speaking of Artificial Intelligence see some interesting commentary from someone working in the field

http://computing.dcu.ie/~humphrys/philosophy.html

especially go a bit more than halfway down the page to:

AI is possible ..
.. but AI won't happen

Rick said...

Jim - Interesting linked analysis. I'll have to give it a much more thorough read!

I think high level AI is possible, and for the moment I'd say it could eventually happen, though his argument that we have already filled the ecological niche is way interesting.

But I don't think we have the faintest clue of how to build one now, because we haven't the faintest clue of how the human mind works. (Expert system type AI strikes me as a projection of human intelligence, mental leverage if you will.)

For this reason I don't see true AI (actual or merely possible) as a foreseeable 'midfuture' thing. We might discover the relevant knowledge tomorrow, but so far it has been outside the scope of the industrial revolution.

Z said...

My apologies if I was just not reading carefully enough, but I smell one more micro-style variant- stealthed/chilled projectiles. Sure, you can't hide life-support radiators or drive plumes or open cycle cool a ship for long enough to be interesting, but I could conceive of radar-absorbent, liquid-helium jacket missiles being lobbed by a cold launch mechanism like a coilgun (preferably in groups fired askance to the target so their trajectory could not be inferred from from motions of the firing ships) that only ignite drives when they are close enough to have high enougn angular velocity and low enough response time to elude a point-defense system. Maybe make them spherical, maybe make them rodlike and use some of their pumped liquid helium to orient their smallest profile towards the target even as they are not approaching it strictly CBDR. If both ships are keeping each other bathed in enough light to scramble a LIDAR, lighter, unguided but more numerous chilled delayed-ignition flares to overload point defenses relying on quick and dirty temperature detection and not trajectory/jet power analysis, might be able to get a couple shots through.

There's a lot of ifs and buts in there. But fundamentally, the thermodynamic arguments against stealth cover things that at any given moment in time need to be exothermic for an indefinite amount of time- there's no reason why something producing no power, having no minimum internal temperature and not accelerating could not be temporarily open-cycle cooled to 3K. I say no reason excluding the fact they might be godawful to build, but cyrogenic shrouds are a real near-term countermeasure concern for IR BMD interceptors, and I hadn't seen them bandied much, except in the thermodynamically untenable case of jacketing an entire ship. It's not subs in space, but it does at least restore us to a battleship-uncertainity of where the bullets are.

Rick said...

Well, this is rather cool ... so to speak. :-)

I can easily think of limitations and provisos, but no obvious deal breakers, because you're not trying to keep a large object cold for a long time. (Caution: More knowledgeable people than me may come up with deal breakers.)

I don't think you could shoot it from a coilgun, though, because I doubt there's any such thing as a cold launch. Any coilgun with real oomph puts such powerful magnetic forces on the shell, or ferric sabot, that it would surely get very hot.

But a lancer type or big missile bus might be able to deploy them as submunis, especially if it is also deploying penaids to make it hard for the target's sensors to detect small, cold shells.

Z said...

A good point regarding the coilgun, but really, so long as the projectile is cold before its trajectory is established, it doesn't matter-a powder gun would work fine too. I just liked the idea of a gun vs. a booster stage/missile bus/whatever, because so long as the ship fire more than one at a time on different trajectories or at different speeds, you can't work backwards from the motion of the firing platform as seen by your telescope to determine the whereabout of the projectiles. Depending on again, minor variations up or down in the size of ships, the amount and magnitude of acceleration in battle, and the power of sensors, that may be an irrelevant detail- the sensor platform might not be able to detect any apparent motion from a thousand-ton ship firing a kilogram missilcicle, but the potential is there. I also admit I liked the dramatic potential of The Heroes knowing the other ship has fired, but having next to no idea when or where they will be hit, and just waiting for their drives to light up so they can get shot down...or not. But sure, start em' on drones, lancers, whatever. So long as too much is going on to intuit their motion from the changes in the motion of the launch platform, you're golden.

Just as another maybe-maybe not addendum, this might be a decent place to employ an actual explosive warhead. If the key is to minimize the amount of time the drive is burning back towards the target from wherever in the vicinty it has been thrown, I can conceive of something akin to a cryo-jacketed Gator mine with an explosively forged projectile being lobbed into the vicinty, keeping its nose aimed at the target until it was close enough and effectively expending all its propellant as once. It's another place to play with the technical details- if they can go hot far enough out without getting pegged that they can accelerate, they'll be KE killers- if they need to get closer and lack the time to accelerate, they'll have warheads. Granted, given the burn rate of some solid propellants, there may not be a "too close."

Once again, big caveats all around. The Atomic Rocket indictment of spaceship stealth seems pretty thorough, and I don't want to be noobishly parroting discredited notions, but it just stuck out to me as an obvious element to toss into the soup.

Rick said...

The hitch with any sort of gun is that the shell's trajectory is established when it leaves the barrel. I think these will work better if they are released 'quietly' by a ship/bus that then veers away.

I wouldn't quite call a fast final kick a 'warhead,' since its purpose is course adjustment. (Or am I misunderstanding the idea here.)

Z said...

Misunderstanding a little-but that's what I get from long posts.

My notion was that the liquid-helium jacketed shells are probably missiles themselves, fired from some kind of gun with their booster unactivated, so most of the travel time between the ships is covered in a stealthy mode. Once close enough to have a chance to out-slew point defense and the like, they go hot and ramp up their speed fast enough to be kinetic killers- or they just use a warhead if they don't have time/distance to accelerate at that close, "inside" range. The Gator mine was just one more variant of the maybe KE/maybe warhead off-axis game- you lob CBU-97 Skeet submunitions at a few hundreds meters per second in the general direction of the enemy with a nice soft launch to avoid detection, they keep themselves pointed at the target via jets of their cryogenic bath, and at some arbitrarily close range they go BANG and shoot off an explosively forged pentrator at some kilometers per second.

The other fun part of it is that by laying them down by gun (though it would work laying them by lancer too) is that they don't all have to approach CBDR before the engines kick in. Presumably, any trace signal on a constant bearing will merit extra attention, as will the bearing directly between two ships if the relative motion were small, say in orbit. You can lob the missiles, engines off, in the general vicinity as a sort of cloud, perhaps none on an actual collision course- and when they go hot, they could converge from any and all sides, and now your point defense can't be directionalized and gets that much harder.

The other reason I preferrred gun to lancer or booster stage deployment was that if a lancer drop ones, and the missile is any appreciable fraction of the mass of the lancer, you can work backwards from the change in the motion of the lancer under thrust and figure out where you should be pointing your LIDAR. If you have a ship with at least two guns, firing off pairs of shells at different angle and/or speeds, there are an infinite number of solutions to the vectors that summed to produce the new relative motion of the firing ship, and so you can't even theoretically work backwards even if you had the sensor resolution to watch the launching ship's "recoil" motion.

Granted, that may be an unecessary detail fraught with issues. The core is the cryojacketed radar-absorbent shells. Unguided pistol slugs or lancer-deployed frozen kinetic-kill AMRAAMs that go hot in the last kilometer all work just fine too, and what size, terminal propulsion (if any) guidance (if any) and deployment scheme are again all products of how good the drives, sensors, defensive armament, and the like are. The more I think about it, the more I think you have the better idea, actually, but I was just playing with some of the possibilities it seemed at first glance to open up.

Rick said...

The shells are 'target seekers,' equivalent to the top stage of any kinetic missile, with a deflect motor for final intercept.

At typical space encounter speeds, though, they don't need to 'ramp up their speed' - forces pretty much have to orbit match NOT to have encounter speeds of a few km/s, at which any explosive or fuel aboard the impactor is purely icing on the cake.

My concern with guns is purely the heat resulting from putting LOTS of kinetic energy into the shell very quickly. But low power coilguns could be mounted internally as 'torpedo tubes.' A kick of a few tens of meters per second, even a couple of hundred, probably won't heat the shell that much. Mount them in opposite-facing pairs and you can kick out a pattern without any net effect on ship motion.

But they will have to be nearly on intercept course, or the target will have a chance for defensive fire after they telegraph their presence by their deflect burns.

Z said...

Sounds like we're on the same page- just that any simultaneously fired pair works too- the ship will move, it's just impossible to figure where the torpedoes are as a result. Invisible frozen torpedoes- might be fun :-)

Rick said...

Z - any simultaneously fired pair works too- the ship will move, it's just impossible to figure where the torpedoes are as a result.

I hadn't thought of that point - true! I'll have to mention this to Winch, of the Atomic Rockets site. He follows this blog, but might not notice comments added to older posts.

Geoffrey S H said...

Having read the earlier posts about hiding in traffic, I remembered Atomic rocket's thoughts about how drives can be instantly identified down to the class of ship/spacecraft in view. Added to the fact that the entire solar system can see you thanks to infrared (light speed lag not withstanding), thus any power with a supercomputer handy can see you. Now add to the fact that they would have the class of every spacecraft on a database somewhere, thanks to HUMINT, and factor in the processing power of the computer. Thus hiding in traffic means they wiz up a list of the classes of ship in the cloud of traffic you’re hiding in and simply get the computer to go over the list until they find you. Of course, it helps they know what you're flying, but it seems to me that hiding there simply means they have afew extra seconds before they find you instead of doing it instantly in open space. Not to mention the fact that they know your speed and trajectory and will have calculated precisely where you are in the traffic long before you have hidden in it.

P.S: A rather stupid question: these “spherical stealth ships”- were they intended to run on maximum power with the heat sinks (explaining thus why it would not work?) or would the sheer power outage of life support and computers powered on standby have done it anyway, thus rendering the heat sink impractical? I’d like to read up on this more, but Atomic Rockets and others sites dedicated to showing why stealth does not work don’t touch on heat sinks at all.

Rick said...

Welcome to the comment threads!

One proviso about identifying drive signatures - that's an 'optional' feature, so to speak, plausible but not inevitable. Scan might only be able to report the general characteristics of the drive, such as estimated power output. It all depends on fine details of drive technology (and what you want in a setting).

The biggest problem for heat sinks is sunlight, assuming you're anywhere near 1 AU. If your hull reflects it, that's not very stealthy. If you paint your ship black to absorb it, that's a constant 1400 watts/m2 your hull is absorbing, and it will overwhelm any heat sink pretty quickly.

But a couple of people have pointed out that heat sinks might work well for small kinetics that only need to stay cold for a short time. Chill the kinetic and give it a gentle kick out the hatch, and it will be very hard to detect till it warms up.

Geoffrey S H said...

That potentially creates some interesting possibilities for settings. Barring a “cloaking device” (though the UK/US militaries seem to be making progress there I might sadly add- I’ve never liked the “trek” cloak concept to be honest, too much cheating) hoping you’re not too bright to be seen or that you’ve got some phenomenal heat sinks seems to be the only options (barring that nifty spring fired weapon idea). Anyway, the setting:

1. The inner planets are the peaceful areas (the "inner core") as sunlight is so intense that stealthed systems are impossible to maintain (if internal storage systems become even vaguely possible through superior heat sinks). Combat, if ever, is done thorough drones and planetary silo based missiles, due to the total lack of reasons for strategic manoeuvring. The "outer rim" as it were is perhaps more friendly to raiders and manoeuvring forces, and thus is slightly more conducive to scouts with radar and telescopes sent out to pick up the occasional "heat-dump" and drive flare. The only possibility for manoeuvring in the inner core is within the shadow of a planet- probably impossibly from your target anyway.
2. If a black ship absorbs huge amounts of sunlight, and more is radiated off- then some could be creamed off as power. While solar panels are probably more efficient, I kinda like the idea of black ships creaming off some heat, and using it to superheat water, produce steam and power electrical turbines which goes to life support, ion engines, etc and use some of the remaining heat to simply warm up the interior- essentially a "steam-powered" spacecraft. A setting could consist of the inner planets using these for transportation and the outer worlds using more conventional vehicles.

Of course, if heat sinks powerful enough to absorb sunlight just outside 1AU were developed, then that would immeasurably help me develop one setting I was thinking about (the space between Earth and Mars would feature prominently), and I’m pondering whether Jupiter’s space might be enough (around 300 W/m2 there).
I don’t even want stealth per se to be honest. I’d call it more privacy. A ship can get from point A to B without being watched all the time, apart from the occasional main burn, heat dump or thrusters burn. The idea that everything everywhere can be seen from anywhere has taken me a long time to get my mind around. I rather hope we won’t get to the point where any mode of transportation (or any person!) could be seen anywhere on the planet either.
One last “stealth” thought- liquid hydrogen thrusters to change course (slightly) without anyone seeing you? Or some other medium cooled pre-voyage to 3 Kelvin (and somehow squeezed out of a nozzle manually- extreme pressure?).

Ok, one last one: I read a thriller sometime ago that talked of a (probably fictional) Soviet program to put missiles on concrete rafts near to strategic assets that were NATO aligned. Hearing about the Japanese concept of a solar station to literally “beam power” to a terrestrial receiver, I had the idea of weapons platforms somehow placed near important potential threats, with no power source. That would be right next to your terrestrial headquarters- everyone knows its there but that doesn’t change the fact that the platforms themselves are the temperature of space and radiating no heat. When you want to activate them, beam the power to them, warm them up and give them instructions. Before your foe can react. Now give them the spring fired “torpedo”. Not only does the foe have to find the platforms, he now has to find the ordinance coming towards him. Nasty.

Please feel free to shoots these concepts down gentlemen, I have my parachute ready!

Rick said...

In the outer system, the problem becomes not so much about sunlight as the need to keep the crew compartment at room temperature, and that heat will leak out. This will rule out keeping a (manned) ship chilled for any extended period.

Alas, I can't really see any way to get 'privacy' in the inner system, short of losing yourself in a crowd of other traffic on the same orbit.

Unknown said...

I've read a fair amount of the "no one can hide in space" argument, and I don't buy it.

Drive flares are visible from a long way away, and everyone is radiating ~300K just to keep the place survivable to humanoids. Occultation is unavoidable.

However, past this, it seems that the practical considerations which actually determine the stealth situation are completely ignored. What kind of sensors are you using? How many, and how are they placed, and if they're optical or IR, what kind of optics do they use? Do you have a few fixed sensors with enough field of view to cover everything around your vehicle? Do those provide enough resolution at the relevant distances to get you solid data?

How do you determine ranging? There are several ways, but all of them are going to require a significant image size (and image size is exactly what you sacrifice with a wide FOV). Sure, you can see a drive flare from halfway across the solar system, but unless you know what the dimensions of that flare are supposed to be, you can't determine orientation to any precision. If you can't determine orientation, there is very little about their orbit you can determine.

Even if you can determine their orientation, is your platform stable enough and your imaging system high resolution enough to determine which direction the craft is moving? The same orientation of the flare could give either an acceleration, or a retrograde deceleration.

Now, I concede the possibility that modern imaging and automated analysis techniques are good enough to make space impossible to hide in. I just haven't seen the evidence that this is the case. Seeing a drive flare in the distance, and even figuring out exactly how much power they're making, doesn't do you much good if they're a spec. And if you have to have dozens of telephoto zoom lenses plastered all over your ship to get the data needed for your scenarios, that'll be a financial barrier to entry. Maybe you can't hide from the CIA black ops ship, but you sure can hide from a cargo carrier you want to sneak up on.

Rick said...

Welcome to the comment boards!

The key point here may be your last one: hiding from whom? First a couple of minor responses.

Range information can come from parallax - you don't need a huge baseline, though big scan platforms may deploy some free flying scopes.

Actually tracking a ship and determining its acceleration is a matter of orbital mechanics. If it is moving in a nonballistic way you can calculate its acceleration (and which way it is accelerating). Then measure the drive flare to estimate drive power, and drive power and acceleration give you approximate mass.

A million important technical details aside,sSo far as I can tell, it is very hard to hide from major military grade scan capability, the spacegoing equivalent of an Aegis cruiser. Hiding from a civil ship that just has basic commercial gear - and may not make proper use of that - may be quite another matter.

But I think the biggest factor will be 'terrain,' not only the physical terrain of planets, but the clutter of spacecraft in orbital space. Warcraft may not be obviously different in appearance, from a distance, or in behavior until they take warlike action.

One form of stealth is strolling along like you are minding your own peaceable business, instead of skulking around.

Unknown said...

Thanks for the welcome :)

I'm trying to look at this issue not from the perspective of someone trying to hide, but from the perspective of someone trying to keep watch, or an engineer trying to design a space vehicle with the best scanning abilities that are currently plausible. I think there's a big difference between "You there, trying to hide in orbit, will almost certainly be spotted by someone" and "I need to reliably spot potential aggressors at the maximum possible range and with the most information I can possibly extract using modern sensors. How do I do it, and more importantly, how often will I fail?"

So I'm looking at this from the perspective of a sensor suite engineer, trying to figure out the real values involved. I think you're partly right, in that intentionally hiding is going to be hard. But I think the perspective that seems to dominate in discussions of stealth/sensors in space is a bit misleading, the "you can't hide" perspective. I prefer the "can I always see bad guys" perspective, because in the long run, sometimes failing to spot the bad guy can be very, very lethal. I don't buy that "you can't hide in space" implies "I can always see the bad guy".

Accuracy and precision in tracking are also very important. It's not enough to say that you can't hide movement in space. If you're trying to engage in boarding maneuvers, or combat maneuvers, you must know the others orbit to an extremely precise degree, and I am cautious of simply assuming that the stipulated capability of noticing a drive flare implies the ability to calculate the velocity and orbit of a ship to the necessary degree of precision. To some degree of precision surely, but "some" degree of precision doesn't make a targeting solution.

It sounds like the discussion here tends to prefer more abstract or social forms of obscurity, which are easier to discuss without reference to mathematics and engineering. These are important points, but I think the difficulty of doing the math has caused an interesting discussion to fall by the wayside in favor of simply assuming that, if you're in space, you're being tracked by anyone on your side of the planet to an arbitrary degree of precision. In practice, the degree of precision that different sensor suites give for different kinds of information will heavily inform combat and police operations, unless it turns out that sensor systems are so useful in space that literally anyone can track literally anything to an arbitrary degree of precision. I highly doubt that is the case.

Unknown said...

I've been playing with the math a bit, and I want to point out a couple of specific areas of technical issues that are being overlooked. This isn't to say that these technical areas rule out the "no hiding in space" issue, but that before it can even be determined if you can hide in space, these technical questions need to be answered.

First: I'm assuming right now that the only thing you get is passive IR emissions. Drive flares can be timed to coincide with periods when you're invisible to the observer, so I don't think any paranoid patrol boat captain is going to rely solely on drive flare visibility for his intel gathering.

So, quick googling shows that passive IR is going to be in the longwave range, 8-15µm. Given plausible detection technology at that wavelength range, what optical focal length is required for a 50m craft that is 1840km away (the maximum separation attainable while everyone is in what is considered LEO) to be resolved to greater than the size of a detector pixel? What field of view does this imply? Two degrees of FOV is going to be a bitch, when you have to cover 360 degrees of view, in two dimensions. If you can get away with a lens that's got 180 degrees of FOV, on the other hand, not bad, only needing 4 cameras at a minimum.

How about Signal/Noise? What's the optical SNR of a 50m IR radiator at 1840km against a background of stars? Near or over the solar disc? Near or over the lunar disc? Near or over the terrestrial disc? If it's not possible to distinguish a 50m radiator from noise when it's between you and the Earth, then LEO just turned into stealth mode, although with the ironic tradeoff that in very low LEO, you need more maintenance burns to maintain an orbit.

Related to the above: angular resolution isn't the only thing you need to pick out a radiator a couple of megameters away from you on IR band, you also need enough photon gathering power for it to show up on the exposure you can get. The exposure you can get is going to depend on your relative velocities, the aperture of your lenses, and the frame rate you're capturing with. It's quite possible that a ship is quite a bit brighter than the background, but that you can't see it anyway, because your exposure is too dim.

So these are some technical issues regarding passive detection of IR radiators in orbit. If you know of some place that answers these technical questions, I'd be much obliged. If not, I'm working through the problem myself, and eventually I hope to have a better idea of the plausible technical detection capabilities of spacecraft. Can you see, though, why I think these high level discussions of stealth/detection in space don't really answer the question one way or the other? It is possible in principle to detect anything out there and calculate everything about its kinematics, but "in principle" isn't the same as knowing a solution can be engineered to do that.

Unknown said...

Alright, so I've spent more time with the math. Back-of-the-envelope calculations for a 30x30mm longwave photodetector system at 16MP density (~4k x 4k pixels) requires a 3.6 degree FOV to detect our hypothetical 50m ship, at the upper edge of LEO from the lower edge of LEO, as a single pixel. Anything wider FOV, or wider separation (like, for example, when they don't share a terrestrial radius), and the vehicle we're attempting to detect will subtend less sensor area than a single photosite. This has lots of problems for the notion of being able to "spot" another spacecraft which is in an arbitrary orbit. Your SNR looking into the light of any local body is going to be 0dB.

According the Wikipedia article on thermal imaging the resolution of modern longwave detectors is limited to about 320x240 pixels. That's an order of magnitude less resolution than I'm stipulating, and that plausible futuristic system still requires dozens of cameras to implement.

An alternative to complete coverage I've come upon at the Atomic Rocket page about stealth in space is that of slow scan detection. At one point it is determined that a full sky scan can be completed in 4 hours. To which I must ask: a full sky scan of what? in LEO your orbital period does not exceed 2 hours, 10 minutes, and at the lower end is near 90 minutes. How does any ship in LEO have any guarantee of detecting another ship in LEO when it takes longer to scan the sky than the orbital period of either? How do you expect to get any kind of useful information about a passive emitter when it takes you 4 hours to get a single still image of the vehicle?

Note that the slow-scan analysis was also targeting a vehicle engaged in a burn.

I'm still not entirely convinced that the "stealth in space" thesis is wrong, I need to work out the numbers in more detail, and look at more of the Atomic Rocket arguments, but its seriously starting to look untenable from a technical standpoint.

Byron said...

The only problem with the theory that resolution would have to be very high is that we're not dealing with trying to find a black object on a white background. We're trying to find a light in a dark room.
For example, Epsilon Eridani has an apparent visual angle of 9.711e-11 degrees (if I did the math right), which is a lot less than the minimum human perception, which is about 1.2 arc minutes according to wikipedia. You can't treat the ship as a non-radiating object.

Anonymous said...

Chase,

Keep up the good work and posting your comments. I understand and agree with the base argument, but as with your thoughts, there's a lot of "noise" to consider.

At a far enough distance a "burn" signature may not be distinguishable from a star for certain burn temperatures and without multiple detections to work out parallax or trajectory.

Furthermore if a burn was conducted out of view (or when the observing ship was looking elsewhere) the approaching ship in ballistic trajectory will be significantly more difficult to spot. After all how much energy does a craft emitting just life-support and waste heat from the computer and power systems generate? Surely many orders of magnitude than that generated by the craft's engines (for any self-respecting SF story ;) ).

Another aspect to all of this is just how much of the torch ship's "plume" is visible to a ship not looking straight up it's tail pipe? A very efficient drive such as photonic or plasma based with a high level of collimation would not through much signal in other directions.

Some "nozzle" designs would direct the exhaust mechanically (i.e. SSME) while others (such as many fusion variants might do so electromagnetically (e.g. fusion and fission fragment type engines).

I would think that electromagnetically directed plumes might have open structures that allowed more visibility to the released gamma rays, etc. but would radiate MUCH less IR (making a case for a multi-frequency sensor suite).

Also all of the arguments I've seen mostly consider *thermalized* exhausts and don't discuss those not in thermal equilibrium (such as fission or fusion fragment engines). This also argues in favor of a multi-frequency sensor suite (keep an eye out for those gamma rays!).

Anonymous said...

Just a follow-up:

>>If the Oscar's crew was shivering at the freezing point, the maximum detection range of the frigid submarine would be 13.4 * sqrt(1510) * 2732 = 38,800,000 kilometers, about one hundred times the distance between the Earth and the Moon, or about 129 light-seconds.<<

I'm STILL not buying this. I don't doubt that the math and theory are correct, but if what this says is true, then our current space program should have no problem detecting ALL Earth crossing asteroids larger than 100m in less than 4 hours (the time claimed in many sections).

A 100m Earth crossing asteroid would be larger than many space ships, should have an average surface temperature about that of room temperature, and should be visible against the back drop of the 3K CBR.

What actually happens is something like a 1/2 life. We detected some right away, but over time we continue to detect new Earth crossing asteroids.

And this is with a LOT more equipment and computational power than will be available to a single spacecraft. It seems to me that every assumption has been made in favor of the seeking ship and to the disadvantage of the hiding ship.

I agree that if one side has multiple sensor platforms distributed throughout the solar system all looking in the right spot, they will always detect a ship trying to hide. But I don't agree that one side will always have multiple sensors deployed, or that they'll be looking in the right direct and at the right time.

Also consider that a ship in Earth Orbit will be difficult to detect with the Earth as its backdrop (especially in IR). Of course this orientation changes as it orbits and it will present the best chance of detection when it's furthest to the side of the planet. Consider that astronomers are still not certain there are no bodies in our solar system between the orbit of Mercury and the Sun even though a large region which could support a stable orbit exists. This is a region 60,000,000 miles in diameter!

At best when writing such a novel, the protagonists might be able to declare, "our Monte Carlo simulators indicate that we've detected ~95% of all space craft down to 100m in the region." or some such. While good in theory, the missing 10% in combat could be quite lethal.

Anonymous said...

Also another comment on detecting drives...

As I mentioned earlier a collimated drive would be difficult to detect outside of a small arc directly behind the ship. The ultimate in collimation would be that of a mass driver.

Oh sure, if you were directly observing a vessel and saw acceleration it wouldn't be tough to figure out what you were seeing, but consider a ship "hiding" but running on minimal power. It could maneuver some but I would think that there's not much delta V in mass driver propulsion system.

Still that 10% uncertainty of not seeing a ship "hiding" and that the ship could still maneuver could be a BAD combination.

All of these advantages decline if the hiding ship moves out from the "clutter" though (away from planets, moons, asteroids, the disk of the sun, etc.).

And no amount of LO technologies is going to keep an Orion project class ship from being detected while under acceleration :)

I wonder, how detectable will a mag sail be? The shear surface area of a solar sail will make it a non-starter for hiding.

One final comment is what about the cooling requirements for IR detectors as well as a host of other engineering problems? Long wavelength IR imposes substantial refrigeration problems on the detection equipment (cryogenic - in fact some require LHe!) to keep the IR "noise" in the equipment from swamping the signal in the sensors. Also what about Cosmic ray interaction with the detectors. These also generate substantial noise in most detectors (though I suppose sufficient CPU filtering ought to remove the worst effects).

I'm left with the impression that the discussion on sensor theory here has left out the reality of what NASA faces in doing a asteroid and comet surveys. Some of these objects are HUGE, with high albedo, and reflecting significant amounts of light. If things are so easy to see in space, completing these surveys should be a piece of cake, and yet they are not.

Rick said...

One factor here is just plain money. If we were willing to spend as much on detecting Earth-crossing asteroids as the US spends on AEGIS cruisers, we would detect a lot more of them.

This does potentially go both ways - in classic operatic scenarios, a patrol corvette in some remote corner of the galaxy is not going to carry a battlecruiser grade scan suite.

In Solar System scenarios you have to assume that all major players will have access to first class equipment. That is what makes them major players.

Having said all that, there is a meta issue here. One phenomenon that astronomers have never dealt with is the fog of war, and how weapons and other systems perform on the battlefield is hardly ever like their performance on the bench, let alone on the back of an envelope.

With one extremely conspicuous exception.

Anonymous said...

Ah, I see part of my disconnect with the conventional wisdom.

Case 1) Future Sol system interplanetary warfare.
All participants are familiar with each other and the solar system. They know what's supposed to be there and what's not. Each side has numerous sensor platforms (scientific, military, trade, etc.). Anytime ANYTHING new or unexpected is found, it gets quite a lot of sensor time devoted to it.

Case 2) Sensing ship is solitary sensing platform AND unfamiliar with the "terrain".
This can come about if an interstellar ship arrives in the Sol system OR if one of ours goes there OR if two survey ships meet in a newly discovered system.

In the first case, odds of detection and discover are quite high.

In the second, it may take quite a long time for one side to figure out where to look. Upon finding "something" it may take another long time to determine that "something" is a hostile ship. It'll take NASA 10 years and MANY sensing platforms to discover 90% of NEOs down to 1 km!

Some additional reading:
http://en.wikipedia.org/wiki/Near-Earth_object#Close_approaches

http://en.wikipedia.org/wiki/Spaceguard#History

On one page I saw an estimate that a 100 m ship at room temperature (say 300K) could be detected at 38,000,000 km. Yet the NEO close approaches show much worse detection ranges!

Consider the last 2 moons of Jupiter found and the last 5 NEO approaches and I get an average *detection* range for a 100 M sphere at 300K as 625,000 km (with a Std Dev of about 1,250,000 km). Even worse I have no means of incorporating the objects we DIDN'T even see.

"The difference between theory and practice is larger in practice than it is in theory."

I think this means my previous suggestion is a good one. At best a detector can detect some % of objects. That % goes up asymptotically as the power emissions go up and as an inverse asymptote to the range. Those slopes can be quite steep but you're never going to be 100% mathematically certain (even though you can be pretty certain if your detection crosses some real life threshold - 99.99%, 3 signma, etc?).

Anonymous said...

Rick said:

>>With one extremely conspicuous exception.

Ok, I'm dying to know. What's the one extremely conspicuous exception?

Rick said...

Detection will be more difficult in an unfamiliar star system, where vast numbers of natural bodies need to be tracked and cataloged. A 'coasting' ship could be difficult to detect among this clutter.

A ship lighting up a high power drive is another matter, because these are VERY bright, and have a distinct spectrum.

But ordinary civil ships in familiar space probably keep no scan watch to speak of, depending on some traffic control service. Ships intended for travel to unfamiliar space will usually have decent scan suites, and will take scan more seriously.

Military ships range from comparable to civil 'adventure' ships, on up to battlecruiser grade scan suites intended to carry out a fast, thorough survey of a system.

Rick said...

Oops. The brilliant exception is nukes, which fully lived up to expectations.

Anonymous said...

Rick,

I think I see part of the disconnect for this discussion. The Atomic Rockets page provides equations and numbers for the *maximum* possible detection ranges.

It seems that many here assume that this number will be the *actual* detection range. They also assume they'll get Hubble type precision imaging from a sensor with a field of view 10deg x 10deg AND that their sensor suite will enable viewing of all frequencies from IR through gamma range.

Effectively giving them the abilities of Hubble, Compton, Spitzer, and Chandra telescopes as far as resolution and precision AND also over a large section of the sky.

But that's not how these things work. You get resolution OR field of view, not both. If you want to cover the spectrum of all four of the mentioned telescopes, then you need to carry their weight/size in sensor equipment.

That's quite a hefty payload for a 100 m warship (though not for a 1 km sized one).

Spreading your sensors out to drones/support ships helps a bit but ultimately you'll need "light" collection and that requires a lot of collection area. Small 'scopes on secondary craft will not be as effective as a large 'scope on the primary.

IOW, I think that the assumption that everyone gets to see everything (especially in the 1 on 1 scenario) is doubtful. It assumes that there are no required engineering trade-offs between sensor power and size or sensor resolution and field of view. IMO both of these are "bad" assumptions.

In the other case (long existing players in a known system) the requirement to look everywhere at once significantly declines and the need for full-sky surveys every 4 hours dramatically relaxes.

On the laser thread they mentioned 4 zones (can't see, can see with passive sensors, illuminate with active sensors, and weapons range). As an engineer I would think very carefully about those 4 zones and how to best prepare any craft for potential trouble while making trades between the different constraints. If your tech level didn't include drives capable of producing gamma rays, I'd leave off the gamma ray detectors.

This would almost certainly mean that a ship "running silent" could slip within 1,000,000 km to perhaps 500,000 km.

Also remember the drive detection ranges listed were also the *maximum* possible detection ranges. If you had only had Hubble type (high angular resolution, small field of view) optical/near IR detection and were not looking in that direction during their burn, then you wouldn't see a thing!

Of course this plays right back into your discussion of the difference between stealth and tricky. Being tricky and running silent could confer the equivalent of stealth. Meaning a warship *could* spot you if they knew where to look. It's sort of like a magician's gag, don't be too obvious (high-g burn) and a little misdirection ("hey, look over there") could provide the tactical equivalent of "stealth".

At some point though, even a ship "running silent" is going to pass close enough to tip off the large field of view sensor suite and then you get to discover just how good your misdirection works ("we're only a cargo ship carrying volatiles to Ceres.")

Jim

Turbo10k said...

I'm bookmarking all these classic threads.

Locki said...

I've been reading through the various posts on space warfare degenerating into the proverbial "eyeball burning" contest. Remember destroying an opponents sensors is a mission kill.

Stealth may be possible because useing a passive sensor is to risk being blinded.


It strikes me long range IR sensors will be extremelely vulnerable to "blinding lasers" and will be disabled long before effective weapon ranges are reached.

In fact any passive sensor mounted on a ship/guided missle/drone could be easily disabled by blinding lasers.


Likewise, your really really expensive multi-meter laser can be disabled by a much cheaper and smaller blinding laser.

If you were postulating a an exciting setting for space warfare plentiful cheap and nasty blinding lasers may mean the whole battle degenerates into a relatively close-in-battle relying on active sensors (read radar) with unguided kinetics and high thrust spacecraft.

Basically some sort of fighter that is relatively stealthy to radar!


Disabling an opponents sensors is a mission kill afterall and there is no need to physically tear the opposing ship apart if it is blind.

Rick said...

Welcome to the comment threads!

'Everyone blinds everyone' does raise some interesting tactical conundrums! I will probably deal more with some of these issues in upcoming posts.

Anonymous said...

I've been reading about the stealth in space issue since I discovered it a few days ago and there a few things that seem incorrect to me.

First, the equation for determining distance that a ship can be detected applies to surface temperature, but people frequently use it to apply to cabin temperature.
With enough insulation you keep the outer hull as cold as you like without cooling the cabin. What's important is the amount of heat being generated in the ship. Heat in = heat out.

Second, the detection range of an engine burn. The calculation for this seems to assume that all waste heat is radiated, but most of it is in the exhaust. As the exhaust is ejected into a vacuum, its pressure will drop rapidly, which will also cause the temperature to drop. Since radiated heat is T to the power of 4, only a small fraction of the heat in the exhaust will be radiated.

Or am I missing something here?

Gareth.

Anonymous said...

I think Daniel's objections have not really been adequately adressed in the discussion. The distance at which something can be detected is rather meaningless - unless you have reason to expect that anything moving in the area is a potential threat. Given the delay in measurement on interplanetary distances, you might not even note a divergence from known "traffic lanes", if such exist at all, until it's already too late. A telescope network is a nice idea, but in reality introduces further delays, because the information has to go from the source to the telescope first, and then from the telescope to the defender, which is a larger distance than straight from the source to the defender.

I think the discussion has been too focused on the basic detectability. But stealth is not about the "invisible ships" someone mentione. When an F-117 flies by on a bright day, you can easily see it. When you stand next to its landing strip, you can see it at night. Stealth is not a Klingon cloaking device, which is what the arguments presented really rule out. Stealth is making it hard to identify you as what you are and making it harder to target you. In the end, that is perfectly possible in space. Will there always be a way to detect a ship? Sure. Will there be a way to identify a ship as a threat and target it and take it out IN TIME? Much more difficult. Especially when you're an unsuspecting target either as a sitting duck (by space standards) or travelling on a consistent vector.

As RPGs were mentioned, I loved how the space combat system to a previous edition of SpaceMaster really reduced space battles to fighting with maneuvering thrusters on a scale of a few dozen km. Kick your main engine online and you're out of the battle.

Brett said...

I wonder how easy it would actually separate an IR decoy "lamp" drone from an actual warship. I've heard the usual argument about emission lines, but that usually involves doing spectrography on it, something that takes hours in real life to get the proper exposure.

A smart enemy would do engine burns only for a couple of minutes at a time, and you'd never be the wiser. You'd have very little ability to tell whether those interesting pixels of IR that your telescopes picked up are attacking spaceships or decoy emitter drones.

Rocketeer said...

Forgive me if this has already been talked about, but a cold gas CO2 thruster would not have much of a heat sig, and there does not appear to be a theoretical (though there certainly is a practical) problem with radiating all your heat into a .05 degree beam while not using your main (bright) engines. Even then you could radiate only once a day for thirty seconds at a time. As I understand it, the *only* heat you have to radiate is the heat generated by the life support equipment and by the (shut down) reactor.

That much heat is sufficiently small that you could put that in an on-board heat sink, and radiate it off quickly, and into a very small volume of space. And then change direction very slightly using your CO2 jets to avoid being hit by the nearby battleship on the off chance that you are detected.

The other advantage of that is that any small unmanned sensor platform unlucky enough to detect you just got destroyed by the same beam of heat that it detected. And if a larger ship detected you, you just damaged it enough to be at an advantage in the fight that just broke out :)

Rick said...

Apologies to Gareth for not noticing his comment for so long!

And welcome to all the new commenters in this thread!

Cabin temp is often used as a baseline for skin temperature because over long travel times it will be hard to keep the skin much colder. Especially if the ship is near 1 AU (or equivalent distance from other stars), which is where human habitable planets tend to congregate.

Exhaust will cool rapidly, but it does start out hot! Generally I think it's a fair bet that at least a few percent of waste energy will be in radiative form. And even 1 percent only reduces detection range / increases aperture for a given range by a factor of 10.

Cold gas is very hard to detect, but it has very limited specific impulse (~30 seconds at room temp, less at lower temps). So the deflection available is limited.

Having said on this, hydroxide touches on the core point of this post, that 'stealth' is not identical to invisibility. Knowing the trajectory, size, and mass of a spacecraft does not tell you its intentions.

'Everyone sees everything' should be understood as a general warning that invisibility-style stealth is not easily achievable in space. Which is not intuitive to most of us, starting out, because after all we see space as dark, and we learn at an early age that it is hard to see things in the dark!

Anonymous said...

Rick, the point that is missed in posts such as the one on projectrho.com that you linked is that "everyone sees everything" is meaningless for conflicts. Yes, I can see a whole lot of things if I put up a telescope and take a 30 minutes exposure. In battle, that's just not going to cut it. It is meaningless that I see that there's something moving, unless I can reliably pinpoint the characteristics. To do that, it's not enough that a handful of photons hit my sensors. I need the confidence interval for the deduced characteristics to be reasonably small for them to be meaningful.

Automated scans do a lot. They're not omnipotent, however. And at the end of things, no one can beat statistics. Statistics however says that I'll always have a confidence interval, and the less events I have, the broader it will be. As such, any reasonable effort to reduce emissions WILL help in clouding the character of what's there in terms of a per-time basis. It will prolong the period of observation until you can actually make sense of what's moving there and separate noise from signal.

"Seeing" and "characterizing" something are two very different things. When the realization that you're looking at where an enemy was 8 minutes ago hits you at the same time as his thermonuclear warhead, you might just as well not have bothered....

Rick said...

No dispute, really, on these points!

But discussions about stealth in space generally start from the intuitive sense that space is night writ big, and thus that sneaking up on someone is inherently easy. When the contrary is pointed out, people put huge efforts into Rube Goldberg-ish ways to make ships invisible.

That is the context of 'everyone sees everything' as an aphorism.

Exploiting uncertainty is a very different matter, and far more fruitful.

Anonymous said...

Remember that you don't really need spectrography to determine the bearing of a suspect burn using doppler shifts when you can point a nice, big optical telescope at the area you saw the burn and note the direction the ship is pointing.

For this reason I envision that future space navies might employ something similar to WWI "dazzler" camouflage. It was dramatic and erratic counter-painting designed to break the contours of the ship to make it more difficult to identify against the background and, more importantly for our purposes, to make it more difficult to determine the ship's class, bearing, and speed.

I could see a future ship having sharp black-and-white counter-painting on the hull, oddly distributed insulation around the reactor, oddly aligned or oddly shaped thermal radiators (think NASA AD-1 only with radiators instead of a wing), and panels around the engine exhaust to occlude enemy sensors to give the burn a deceptive shape from the side.

Of course, these things would have the disadvantage of throwing any pretense of being a merchant ship out the window. Perhaps future Q-ships would use things like heaters on the hull of their fake freighter and cloth skins for the counter-painting that could be jettisoned as they approach the shipping lanes.

Here's a picture of WWI "dazzle" cameo:
http://en.wikipedia.org/wiki/File:SS_Empress_of_Russia_1918.jpg

Here's a picture of a NASA AD-1:
https://en.wikipedia.org/wiki/File:AD-1_ObliqueWing_60deg_19800701.jpg

Thoughts?

~~
Samantha

Rick said...

Welcome to another new commenter!

I think the specific example of dazzle camouflage, or similar, would probably be defeated by computerized sensors, more patient than the human eye and less prone to visual illusions.

But taking it a little more broadly, this is all about 'noise' - creating a welter of irrelevant information that makes it harder to tease out the relevant information.

And there's another kind of 'dazzle' that could be relevant - zapping sensors. A very bright light will saturate the sensors, and a much brighter light will burn them out. I don't know the required energy levels, but surely a lot lower than burning through armor, etc.

Laser armed ships mutually blinding each other is a scenario that makes things very odd and puzzling!

Anonymous said...

The main problem with dazzling enemy sensors with a laser is that doing so is an overt act of aggression. Part of stealth could also be concealment of your intentions. Consider commerce raiders: If you're a privateer (a pretty good analogue to a space fairing U-boat, actually) even if your target knows you're a privateer and that you're at the location you are you want them to be unsure about your intentions as long as possible. A convoy won't, for example, take evasive maneuvers until they're sure you're attacking them due to the cost of the reaction mass ect... Another example would be a laserstar or other orbital defense: If you're defending against an approaching armada you want that armada to put off evasive maneuvers for as long as possible so when you make your opening salvo they you have the best chance to hit. Camouflaging your laserstars so it's difficult for the enemy telescopes to tell which direction they're pointing is a great way to do that.

Also, there's more to this than signal processing. Consider that at extreme range a ship might only show up as a few pixels on most sensors if that. Just look at most NASA asteroid photographs that are nothing more than a bright cluster of maybe five pixels or so. So, if I'm looking at a ship who's entire hull covers only a handful of pixels how do I tell the difference between sensor noise and a camouflaged hull with regards to that one very dim pixel on the side of the couple bright pixels? I suppose if you watch it for a while you might get a pattern but hey, stalling my enemy for a few minutes is a pretty good use for a can of paint!

Even when you consider signal processing it gives you an advantage. Camouflage paint on missile buses, for example, could seriously increase the processing overhead required by the point defense system!

Speaking of signal processing the problem with computers as we know them is that they're very reliable in their response to a given stimuli which also means that they're not flexible which is a serious impediment when it comes to analyzing complex images. Consider the performance of current top-of-the-line OCR equipment which isn't very good even when their signal-to-noise ratio and available resolution of the characters is quite a bit better than would be available from a grainy, low resolution, potentially very low contrast sensor image.

This is assuming a computer as we understand them today by which I mean they operate by preprogrammed algorithms. True, it's possible that a future computer can learn and reason like we can. There really isn't any reason they can't assuming they're based on a technology that can evolve and adapt new connections like our brains can. The problem with this is that reliability and self-adaptability is a zero sum game because learning requires unorthodoxy and you can't have unorthodoxy in a reliable system. In other words "strong" AIs probably aren't any better than humans at seeing through sensor tricks other than the fact that a "strong" sensor AI might be more practiced at it.

I actually would love to talk to you about the possibilities of "strong" AI in the plausible midfuture.

I think the biggest advantage of "dazzle" camouflage over "active" stealth like blinding lasers is that being camouflaged doesn't require you to shoot at the enemy. This would be a *huge* advantage for everyone from privateers to blockade runners to navies engaged in a cold war or a tepid peace.

I could even see it being used by legitimate merchantmen where the threat of raiders is present due to the difficulty it would give the raiders in setting up an intercept solution.

Thoughts?

~~
Samantha

Rick said...

Playing catch up:

The main problem with dazzling enemy sensors with a laser is that doing so is an overt act of aggression.

Uh, well ... yeah. I was thinking of essentially tactical situations, where you are already beyond diplomatic incidents. But zapping sensors is essentially like shooting out searchlights. (Shades of Bogart in 'To Have and Have Not.')

Your mention of images only a few pixels across reminds me of an experiment I once did - taking images of naval ships and reducing them to that level. I don't remember my conclusions, but I think they supported your argument.

Of course, underlying this are whatever 'natural' differences of configurations between civil and military craft. In the age of sail, or between midcentury airliners and bombers, not that much. For steam era naval craft, a lot.

Orbit choice is also a factor in 'stealth' of this sort. Looking like a merchant ship isn't so helpful if you're on a track no merchie would actually take.

All of which can plausibly be finessed for desired effect in a work of fiction!

Byron said...

Hydroxide:
I think Daniel's objections have not really been adequately adressed in the discussion. The distance at which something can be detected is rather meaningless - unless you have reason to expect that anything moving in the area is a potential threat. Given the delay in measurement on interplanetary distances, you might not even note a divergence from known "traffic lanes", if such exist at all, until it's already too late. A telescope network is a nice idea, but in reality introduces further delays, because the information has to go from the source to the telescope first, and then from the telescope to the defender, which is a larger distance than straight from the source to the defender.
I think you're getting your timescales confused. At the detection ranges expected, closing times will be on the order of weeks if not months. Signal processing delays are minimal compared to that.

I think the discussion has been too focused on the basic detectability. But stealth is not about the "invisible ships" someone mentione. When an F-117 flies by on a bright day, you can easily see it. When you stand next to its landing strip, you can see it at night. Stealth is not a Klingon cloaking device, which is what the arguments presented really rule out. Stealth is making it hard to identify you as what you are and making it harder to target you. In the end, that is perfectly possible in space. Will there always be a way to detect a ship? Sure. Will there be a way to identify a ship as a threat and target it and take it out IN TIME? Much more difficult. Especially when you're an unsuspecting target either as a sitting duck (by space standards) or travelling on a consistent vector.
This is exactly the point. If you can cut detection range from 1 light-minute to 10 light-seconds, good for you. However, assuming that you're doing 100 km/s, that's still 8 hours warning. Depending on the mission, that might or might not be significant. For an attack on a major enemy base, it won't matter at all. They'll go to full alert, and you're worse off because of the money spent on the stealth system instead of more ships. For an attack on a minor outpost, it might make a significant difference, as the enemy can't intercept. However, in no case do I see a chance of achieving tactical surprise, which is what people are really looking for.

Yes, I can see a whole lot of things if I put up a telescope and take a 30 minutes exposure. In battle, that's just not going to cut it. It is meaningless that I see that there's something moving, unless I can reliably pinpoint the characteristics. To do that, it's not enough that a handful of photons hit my sensors. I need the confidence interval for the deduced characteristics to be reasonably small for them to be meaningful.
This is only relevant for tactical timescales and ranges. Those are the light-second scales where you will be detected by the fact that you're occluding a couple of stars, if nothing else. And, as noted below, stealth behavior is abnormal, and therefore suspicious. Which will tend to attract attention from whoever you're trying to hide from. Even burning your engine without a beacon will be suspicious. After all, if the burn is going to be detected, hiding your identity is going to be taken as a sign you have something else to hide.

Byron said...

I wonder how easy it would actually separate an IR decoy "lamp" drone from an actual warship. I've heard the usual argument about emission lines, but that usually involves doing spectrography on it, something that takes hours in real life to get the proper exposure.
That would depend on what you're doing. We're not trying to do highly precise science experiments here. If we have an unknown ship, we know its mass and the drive power, and we have a database of engines, then all we have to do is match the drive signature to one of those engines. I'd be surprised if there were more then a couple dozen candidates, and who cares if it's power X Type 3 Mark 4 or Type 3 Mark 5? We know it's power X.

Any stealth engine is by definition of limited time-averaged delta-V capability. By that I mean that no matter what you do (low power mass driver/ion thruster, cold gas, etc) the power limitations will keep the delta-V available for a mission down to modern-day levels. This means missions take a long time, or require a different system to start.
And therein lies the real problem with running cold. Even if you have a near-perfect "running cold" stealth, the system is still effectively useless.
Situation:
Power X detects a vessel of power Y's burning for a while. Then the vessel suddenly disappears. No reactor, no nothing. That's obviously abnormal. They then take the last known position and velocity, and crank in the best delta-Vs available from an undetectable thruster. The result forms a cone. This is where the ship could be. The cone passes right through Power Xs planet. I wonder what's going on?
At this point, Power X sends a couple of conventional warships towards the last known location of the stealthed ship. They locate it at close range, and follow it until it realizes that the operation is blown and heads home.

The other advantage of that is that any small unmanned sensor platform unlucky enough to detect you just got destroyed by the same beam of heat that it detected. And if a larger ship detected you, you just damaged it enough to be at an advantage in the fight that just broke out
Not even close. Even a tightly-focusd radiator is not going to form a beam capable of doing damage at significant range. That's not to say I advocate standing right next to the radiators, but it will spread in such a way that by the time it reaches anything that couldn't detect you already, it's not going to damage them. If it could, you'd use it as a weapon. And there are thermo problems with that, not to mention engineering ones.

We discussed Q-ships elsewhere, about a year ago. I can't remember which thread, though.

Tony said...

Also, big problem with "big damn scope", interplanetary-range detection is that for a given resolution, the farther out you try to look, the smaller the field of view gets, and (because the scope is getting physiaclly bigger and bigger) the harder the intrument is to accurately steer.

TokyoTrainStyle said...

I have a couple of suggestions for stealth scenarios.

The first one isn't really stealth, more of a "exploitation of uncertainty". Basically during battle you'd set off a nuke between you and the enemy. This would up the background radiation so as to camouflage your ship. Might be useful if you wanted to pull an unexpected maneuver, and appear suddenly out of the radiation cloud. This form of blinding may in some situations be superior to the more efficient and accurate 'laser their sensors' method, as you could never be sure you had got them all, while a massive nuclear blast is almost guaranteed to give you cover. Unless of course they manage to get sensors behind you.

My other suggestion is to redirect an asteroid. You would have to do all of your thrust and correction outside their sensor range, which would require a lot of accuracy. Also you'd have to come in on a believable trajectory, ideally from the opposite direction to what they expect. A sufficiently large asteroid could easily hide the heat signature of an AI or skeleton crew ship. At the closest point you'd suddenly pivot and hit them where it hurts with a laser. Maybe one of those x-ray single shot ones? Ouch. This trick would only work once though, they'd learn to blow up all near flying asteroids just in case.

A different asteroid scenario is one where you catapault a whole bunch of rocks at the enemy, and hide ships in only a few of them. They'd know they were being attacked, but wouldn't know which rocks to destroy. If they targeted an asteroid with a ship inside it, it would return fire and evade, turning into a normal attack. They wouldn't know how many ships were attacking however until they'd destroyed every rock. Their tactic may then be to give each only a little fire to see if it responds; I can imagine a tense scene with a commander accepting enemy fire in an attempt to fool them into believing his ship is only an asteroid. Then of course you'd turn and hit them with everything. More boringly, but with less casualties, this kind of attack could be done with droids.

That kind of attack perhaps isn't really stealth however.

Unknown said...

I realize this is a week or two old but for a personal writing project I've been doing some research into this. A new development in the field of metamaterial cloaking opens up a possibility. For most frequencies of light (visual, IR ultraviolet, Gama rays, xrays, terrahertz) and even sonic ranges scientists have figured out how to use metamaterials to construct a cloak.

Honestly the description of the paper for IR cloaking reads like classic technobabble but these are real scientists and their work is posted on arvix. But basically the metamaterial allows light/sound of specific frequency to flow through the material, allowing you to build a shell that acts as a cloak for whatever is inside. You even get a heat concentrator that allows you to control the temperature the material gives off, so if the background temp changes you remain invisible.

Now having this cloak survive space travel intact is another story, but if your ship has a low refractive index (also possible with metamaterials) and is effectively invisible on IR if you aren't burning/adjusting course, I'd say the detection chances have just dropped dramatically.

Thoughts?

Rick said...

Welcome (slightly belated) to new commenters!

Nuclear explosions are about as un-stealthy as you can get! :-) Also, in space the debris cloud expands and dissipates very quickly. But the initial flash might blind sensors.

I used to be a big fan of fragmentation clutter (on a smaller scale), but I've come to think that numerous small target seekers are likely to be cheaper and more effective.


The article sounds interesting - is there a link? My instinct is that metamaterials like this would be more useful in a ground clutter environment than in space. But that is without reading the paper.

Unknown said...

Rick,

Having trouble tracking down the original article (it was in a BBC article but I'm only finding an older, related article regarding a Tank looking like an SUV in IR; this was NOT the article I recently read).

This is the same paper I believe, albeit from a different source:

http://www.opticsinfobase.org/oe/fulltext.cfm?uri=oe-20-7-8207&id=231514

Some similar reading on meta material cloaking:

http://arxiv.org/ftp/arxiv/papers/0709/0709.0363.pdf
http://etd.fcla.edu/CF/CFE0003201/Shelton_David_J_201008_PhD.pdf

Unknown said...

Found it! Helps if I search for thermal vs infrared: http://www.bbc.co.uk/news/science-environment-17518210

Anonymous said...

Sorry, this is a bit outside of my area of expertise. However if the basis of detection and classification is predicated on the detection of radiation emissions why not simply crank out extra radiation. If the whole point of this is that they'll know you're there, why give them a freebie of WHAT you are. A given IR/radiation output might match up to a 100 ton object accelerating at 10 m/s using a particular type of drive. Throw on some over grown heat lamps (for IR), radiation emitters, etc. that are spectrum matched to your drive output and blaze away. Now you appear as a larger object accelerating at the same rate.

If all your contacts appear as say a "battleship" (for lack of a better term) accelerating at full thrust, vs a bunch of "scout ships" accelerating at the same rate your tactical response will be much different. Getting into visual range to tell the difference between different types of vessel is required.

As many of the posters have mentioned, tactical and even strategic surprise don't require "stealth" just misdirection. As I prefaced this, this is not my area of expertise so if I'm off base please let me know.

Unknown said...

Hey, sorry for possible noob question (Also this thread is rather old now so please feel free to ignore). What about autonomous drones? They wouldn't need life support, obviously, so they could run VERY cold when not burning their engines, and furthermore, they could be very light, so their burns would be a lot smaller. Wouldn't that reduce effective detection distance by a lot?

Charles A said...

Sorry to dredge up an old post, but I often see people stating "you can't hide in space" as a fact, but I haven't been able to piece together the evidence to reach that conclusion on my own.

If someone could point me to some resources where I could do the math on IR output, sensor sensitivity, etc. so I could see the numbers for myself, I would really appreciate it!

Like I said, I've had a hard time finding this data, and the fact that it's so hard to find 1km wide asteroids (which are around room temperature, I believe) makes me think that it would be essentially impossible to find a 30m long spaceship specifically designed to be hard to detect. But everyone seems so adamant about it, there must be some evidence somewhere!

I just can't find it for the life of me.

Anonymous said...

I wont start a flame war by making a blanket statement as those are often wrong anyway. that having been said, this artical and the subsiquent posts don't take into acount every possible senario so making a blanket statement such as stealth is imposible in space is also wrong. while perfect stealth (if achived) would render the stealth craft undetectable it would also render every thing else undetectable to the stealth craft as every form of EM radiation and matter emition would be blocked from entering or exiting an area around said craft. Therefor if one could bend the EM state around the craft and prevent any EM state from leaking out save for a selected band used (presumably) by the craft in question it would be as stealthy as any stealth aircraft (Since they are not actually completely undetectable in themselves). Thus you would reduce the posibility of detection to a minimum, while not eliminating it completely. Also as long as the thermal storage method and the power used to bend the EM state around the craft held up it could maintain said stealth. that being the case, nothing could maintain stealth indefinatly, but thats not really the question here.
Ok so, for example say you have two waring ships and one is equiped with a cloak that bent EM states around it. If the enemy ship (the one with no cloak) sent sensor beam A at the cloaked ship it would presumably bend around the ship and continue on its marry way out into space making the enemy vessle belive nothing was there since it recieved no report from beam A. Now if the field bent only say 95% of beam A around the cloaked ship and the other 5% was small enough amount for the ship to absorb it could presumably still go undetected but have a passive sensor system of sorts. this however would fail if beam A was being intersepted by a reciever as it would register said 5% drop in beam A (presumably it could compensate for the divergance of beam A in its calculation of how much of beam A arrived at the reciever).

In summory while being completely undetectable might be possible it would be utterly useless as the undetectable ship would not in itself be able to detect anything else. however stealth and undetectable are not of neccesity the same thing. Thus stealth may be possible but making it absolute would be useless.

BTW if one were able to bend the EM states around a ship you would not nessisarily violate physics, such as thermodynamics and relitivity (which is looking more and more like its inacurate anyway), as we do not know enough about the EM states to claim that we know every thing.

also i know i cant spell, but one need not be good at spelling to be a scientist. lol. MS word spell check is a lifesaver.

Jim2B said...

The arguments stating "there is no stealth" run along the lines of the total power emitted from a 100 yard long ship even at room temperate is enough to be viewable from the HST for 500,000,000 miles or so.

The counter argument to that is that the HST must *know* where to look to spot something that dim. As Space Watch keeps demonstrating (when it finds hunks of rock much larger only after they fly by the Earth.

Obviously you won't be able to "Stealth" any sort of "torch" ship. However, something with ion thrusters or other much lower power engine won't be terribly obvious. You will likely not see it coming unless you are looking for it.

Once someone does spot you though, there will be no place you can hide as long as you are not near a major body.

Unknown said...

Im sure I am missing something but I don't get the obsession with the need for massive radiators that stick out the side of the space ship. Why cant we have millions of very small spherical radiators that radiate infrared inside parabolic reflectors aimed at infrared versions of optical fibers. The optical fibers coalesces into to a single optical fiber that itself emits into a parabolic reflector that focus Infrared radiation away from the spaceship. it wont be lazer like cone but it will likely be better than 60 degrees mentioned in the article. Infrared reflective materials are very effiant but obviously this system will absorb some of the IR released from the small spherical radiators. However due to the fact that you can use layers of these inside the volume of the spaceship you can have many times the surface area of a simple external radiator. No doubt I am missing an obvious reason this would not work, or have misunderstood a basic scientific principle.

Martin M said...

I don't think that's necessarily impossible, it just seems like it has practical issues. It would be awfully heavy. Your mirrors and fiber optics better not absorb too much of that infrared either. And I take issue with the idea of a nice neat point source radiator in a parabolic mirror. Everything on your ship radiates, including the pipes that get the fluid to your spherical radiators.

Saint Michael said...

Necromancy rules!

Leaving aside various means of camouflage, any vessel capable of significant delta-V will be easily thermally visible *at distances close enough to do damage*, unless you posit an interplanetary superlaser, which would suffer lightspeed lag anyway. A one-pixel target spotted across the Solar System is not a target, it's a background detail.

One factor about thermal sensors is, they need to be cold to operate. Present day tech needs cooling often down to cryo temperatures, so anything that heats the ship or stops it from radiating heat, could lead to possible sensor degradation. Heat pumps will help (and hinder, of course).