You kind of can really. You could send a drone ship to drop off a replacement or even just fire off the sensor station like a missile, with only enough fuel in a simple engine to decelerate it to take up a preplanned position with an acceptable rate of drift. Real satellites deploy in a largely autonomous process already after all. Repairs are highly unlikely for the simple reason that any damage inflicted is highly likely to be catastrophic.

I go back to the DEW line example. No defensives, absurdly remote to the point that maintenance just can’t even be conducted for part of the winter at any price, and highly exposed to a first strike. None the less the death of the DEW line was critical to give the Mid Canada line and interceptor bases behind it time to get ready and avoid being taken by strategic surprise deep in friendly territory.

Even with no long range sensor coverage the situation is not necessarily dire anyway. We fought naval wars in the age of sail across the globe with no sensors past the visual horizon, and communications only slightly faster then the heaviest enemy warships could travel after all.

Should a comprehensive attack destroy long range coverage over a wide area, then craft within that sector will concentrate in bastions around key manned installations until such a time as a fleet can be brought up with mobile sensors (likely to be larger and individually much more powerful then the expendable boarder post if less numerous). A space fleet is quite likely to have dedicated ships for this purpose, something like the USNS Observation IslandICBM tracking ship or Sea Based X-band Radar except with at least a token self defence armament. The massive Soviet space tracking and communications relay ship Kosmonaut Yuri Gagarin would be another example of what can be done.

The bastions meanwhile would have there own spherical perimeters and sensors nets, within the overall defensive system which is primarily oriented only in the direction of the enemy. This was more or less how Blitzkrieg encirclements were countered in real life by forming strong points within a defensive belt, and waiting for the enemy to quite simply run out of fuel and be forced to pause his offensive. All the while you probe outward to hit his logistic lines and determine his intentions while building up for counter attack, ideally while he is stuck refueling his tank hoard.

Of course the maximum speed of the enemy, and just as importantly his effective striking radius (not just fuel but lack of food, spare parts, ammo and the like will all bring an attack to a halt) will determine the details of scale required. Its just not likely to be possible to defeat the enemy in a far forward area… and no real reason exists why it should be so. I think that’s one thing many writers could stand to learn. When you have powers that are something even remotely like equal you have to expect campaigns to see-saw. You can’t expect to be strong at every point and just repel all incursions and accept and decisively win all battles

In some respects space war could mirror land war as much as naval or air war on earth. This is because while space is mostly empty and open to movement like the ocean or sky, you can pretty much claim any point in it as a basis for a base or defensive position as you could on land.

---

"This cult of special forces is as sensible as to form a Royal Corps of Tree Climbers and say that no soldier who does not wear its green hat with a bunch of oak leaves stuck in it should be expected to climb a tree"
— Field Marshal William Slim 1956

Flecks of dust and such have two items mitigating against them being determined to be a valid detection event. First, it's very likely to be crossing the field of view rather fast. Thus, it'll show up on the detector as a streak, rather than a point. Second a closely located second sensor will show the streak in a different location, allowing a determination that it's far to close to be a valid target.

The estimate of required energy to be detectable is off by multiple orders of magnitude. The actual threshold with modern technology is femtojoule, not picojoule. This the ranges are around 20x longer than earlier calculated

---

"preemptive killing of cops might not be such a bad idea from a personal saftey[sic] standpoint..." --Keevan Colton
"There's a word for bias you can't see: Yours." -- William Saletan

I imagine that this depends on what kind of propulsion technology you have in whatever sci-fi verse we're writing for. If you have magic-tech, then sure. Warp out there and replace the observation post. If you don't, it could take months or years to get a ship out there.


Keep in mind, however, that the interceptors themselves (and the nuclear deterrent that they served as a tripwire for) had a much, much shorter deployment time. Armageddon was (and is) only minutes away. If it took months or years to get vehicles out there or launch any kind of counterstrike on that basis, they would not be nearly as useful in that role.


True. Mind you, if the propulsion technology in this sci-fi verse is somewhat realistic, then light-speed communications would still be vastly faster than any ship, so it would be more like the Victorian British Empire but with radio and satellite communication.


Oh yes, I'd agree that there would have to be a lot of give-and-take, especially in a presumed "border area" or contested zone between the two nation-states. The idea that one side sets up a defense which is strong everywhere is silly, just as is the idea that the other side can easily attack anywhere.


I'd agree, although with the caveat that everything is so much more expensive. A forward land base could, after all, employ local assistance for food, entertainment, etc. Even expeditionary naval vessels could lay in at local ports for provisions. But in space, one can assume there's nothing to draw on for such resources: virtually all supplies must be delivered from home in some form or other. Warfare far from home thus becomes an extremely expensive proposition, unless it is almost entirely conducted by robots (which may be the only logical method for a realistic sci-fi universe).

Of course, that takes us to the question of why such a war would be worth it for either side at all. There would have to be a very valuable resource to fight over, or some kind of very cheap space propulsion method to make it worthwhile.

Over what timeframe?

If drive technology is expensive to use and maintain, then juicy asteroids full of rare, technologically useful metals that are located on "cheap" orbits would be one candidate for warfare. Even one modestly sized space rock would contain incredible quantities of material, and a good location may make mining orders of magnitude less expensive in time, fuel or both. Same goes for juicy moons, or even small atmosphereless planets (which are basically giant asteroids).

There's plenty of stuff necessary for high technology which isn't all that easy to find in orbit.

As stated earlier, Pan-STARRS resolves down to magnitude 24 with a 30-60 second exposure time using a 1.8m telescope. Also, my knowledge of sqrts fails: ~31x longer range (sqrt(1000)).

---

"preemptive killing of cops might not be such a bad idea from a personal saftey[sic] standpoint..." --Keevan Colton
"There's a word for bias you can't see: Yours." -- William Saletan

Would you actually see a clear streak at that range, or a smudge?

BTW, for anyone reading this thread who doesn't already know, the "Magnitude" Beowulf speaks of is a peculiar astronomy thing. It's actually one of those legacy unit systems that's always really annoyed me, but let's leave that aside for the moment.

It's based on a star called Vega, which (if a bit of quick web searching suffices) is 25.3 light years away and has 37 times the luminosity of Sol. Therefore, its light intensity at our location would be roughly (37·3.8E26)/(4π(25.3·3E8·3600·24·365)²) = 5E-7 W/m².

The magnitude system is apparently scaled logarithmic, so that 5 magnitudes = 1E2, so a magnitude 25 star would be 1E10 times dimmer than Vega. In order to calculate the light we would receive by looking at it, we would have to multiply that figure by the exposure time and the light collecting area of the telescope or array.

Note: astronomy is not really my forté. Corrections or clarifications to the above, if necessary, are welcomed.

It's actually easier to calculate based off the sun's apparent magnitude and output at earth orbit. Sun is -26.73, and puts out ~1400 W/m2. The system is no longer based on Vega, but rather is more strictly defined now, resulting Vega now having a magnitude of 0.03 or something close to that.

So using my calculations: 1400 W/M2 * (100.2)-26.73 = 2.8e-8 W/M2. One of us has obviously done our math wrong.

I don't know enough to be able to tell you the answer. I do know that the quoted resolution of the sensor on Pan-STARRS is 1 px = .26 arc seconds. It's unlikely that the sensors will actually become more sensitive than the current state of the art, since a quoted read noise of 5 electrons implies that they are pretty close to 1 photon = 1 electron = 1 detection event. (there's also a quoted 7 electrons from sky noise, which is another limiting factor in S/N ratio).

---

"preemptive killing of cops might not be such a bad idea from a personal saftey[sic] standpoint..." --Keevan Colton
"There's a word for bias you can't see: Yours." -- William Saletan

I must have gotten sloppy with the calculator. I plugged the figures into the calculator again, and this time I got 2E-8 instead of 5E-7: a difference of exactly 4, which suggests that I didn't enter the "4" in the denominator when I plugged it into the calculator the first time. Mind you, that's still not precisely the same as your figure, but I was calculating it based on Vega which, as you say, is no longer considered precisely magnitude 0.

<insert obligatory rant about arbitrary unit systems here>

BTW, I think it's common in sci-fi to simply assume that future technology will allow us to greatly increase our sensitivity. But if we're already at the point of counting individual photons, it looks like we're pretty much into a brick wall and it's just not going to get any better. Unfortunately, I doubt sci-fi authors will change their ways any time soon. Star Trek has established a pattern of being able to easily take clearly detailed real-time video of small objects from millions of km away, and sci-fi authors follow those kinds of precedents without thinking.

Well then that means the enemy will also take months or years to reach anything important behind the sensor net, in which case the sensor net has done a very good job of providing early warning. You have time to literally build a fleet and convert to a war economy. The sensors themselves are irrelevant; they only exist to protect other things.

This would be more relevant in something less then all out war, a border dispute perhaps in which all the enemy ever does is try to blind you’re sensors, but such wars usually do not go on long without some kind of settlement. It’s hard to see people fighting like that over deep space anyway, in fact it might just be treated as a no mans land in which both sides feel free to place sensor station at well, even if they are within the boundaries of the other sides own sensor grid. That’d pretty much how the open ocean works in real life with things like SOSUS.



Yeah it would be, and that kind of scenario would pretty heavily favor the defence operating on interior lines. An serious attack just can’t really develop faster then you could react to defend your key areas.



A very high level of automation would be essential. I think even if you had some low speed FTL and could travel say twenty times the speed of light you’d still end up deemphasizing humans. In the age of sail 3-5 year voyages were typical, but even that wouldn’t get you very far and I can’t see people standing up to much longer periods even with a luxury liner level of accommodation.

In the vacuum of space everything has to come from home, but if you have planets and asteroids around its not hard to see a naval force being able to extract its own hydrogen fuel, and potentially even do more complicated things like mine and enrich uranium or iron from local resources. A refinery and foundry could go into a fairly reasonable sized ship (real naval repair ships already have foundries, able to handle at least a couple hundred pounds at a time) and one would presume we would have perfected the technology to do these things in zero-g long before we fought an interstellar space war.



It seems to me the only thing that would ever be worth fighting over would be habitable planets. Anything else would just be too common to matter, barring the arbitrary introduction of unobtainium based compounds. In terms of normal raw material resources I would imagine the earth’s solar system could easily be shared by several civilizations with tens of billions of people apiece if only we had decent means of space travel to get at them all.

I suppose one might also just encounter some kind of xenophobic warrior race or giant space ants that attack for the sake of attacking, but such species would probably be too dumb to be a serious threat.

---

"This cult of special forces is as sensible as to form a Royal Corps of Tree Climbers and say that no soldier who does not wear its green hat with a bunch of oak leaves stuck in it should be expected to climb a tree"
— Field Marshal William Slim 1956

Actually that brings up something that's been bothering me about the 'geography' of an in-system war, in that with orbiting territory it won't look at all like your traditional war. If you took a case of inner system vs outer system, well the outer system is a little boned for scrambling defenses if the inner system waits for a good time.

Different question: how much more difficult does background noise from a planet make detection of ships defending it? Are attackers going to be stuck flying in blind?

Interesting and informative discussion! I am learning new things!

So we should be able to detect a 60 megawatt light out to 3.6 AU and a 4 gigawatt light out to 31 AU? Of course, as I understand at this point you're seeing a flash of a handful of photons, so it wouldn't tell you much at all. Something like a few hundred or thousand of these flashes happening very close to a known enemy base, however, might be suspicious. And a drive that weak would require at least hours or days of thrusting (i.e. tens or hundreds of thousands of 1 second burns) to get up to any decent velocity, so once it's been tagged as suspicious there's plenty of time to work out stuff like the flashes are all happening in a straight line, and whatever it is it's accelerating, so it's pretty obviously a ship. Sound about right?

Of course, as I understand the Pan-STARRS takes 30-60 seconds to get this resolution, whereas a ship trying to be sneaky would probably use much shorter duration burns. I imagine this could be compensated for by using a larger lens though. A 14.4 meter lens sounds like it ought to do it if a 1 second exposure time is the target. Thankfully these sensor platforms will probably be in space, where you could build fairly large structures like telescopes with huge lenses relatively easily. I imagine there is also the option of an array of smaller telescopes, if a single lens that big is problematic.


I've always suspected it would realistically most likely work that way myself. Trying to set up and enforce borders around cubic AUs of dead space always struck me as something that would almost certainly be an exercise in futility. Let's not even get into the silliness of the Star Trek style idea of setting up borders that span entire portions of the galaxy.


There is the option of artificially induced hibernation or suspended animation for the really long trips. I agree with you though; realistically you're probably going to want a crew as small as you can get away with. A human operator is probably going to be very mass intensive compared to a computer capable of doing the same job.


Another thing I second: I imagine a space navy is fairly like to devote considerable effort to perfecting food, water, and air recycling and utilization of local asteroid and comet resources. You'd at least want your distant manned outposts to be pretty self-sufficient, if not your ships.

How the fuck are you going to cast a ~15m lens blank? We have trouble even doing 1m blanks. A mirror blank is much easier to cast*, but still, 8.4m is the largest single blank we have ever done (two of them for the LBT). You could do a segmented mirror approach if you want a single very large dish, but even so, the precision required to position the damn thing is going to be extreme. Not to mention using a very large element like that, even at f/1 or below, is like looking through a pinhole. To effectively scan the sky, you need small, very short focal length telescopes**, but then you sacrifice resolution and sensitivity, as has been mentioned several times already.

What you would need is very good communication between the sensor drones in this kind of system. If several of them see something in the same place, that's good enough to warrant another look, once the object has been cross-referenced against all known astronomical objects. So to make the system effective, you basically much better coverage and redundancy than the minimum necessary.

*This is because a lens blank must be optically perfect, but a mirror blank does not have to be. Astronomical mirrors are front-coated, not back-coated like your bathroom mirror, so they do not pass light through them. This means that there can all manner of bubbles and other imperfections in them so long as none of them intersect the eventual polished surface of the mirror.

**This results in a wide field of view because the magnification an optical system provides, and thus its true field, depends on the focal length of both the camera/eyepiece and the instrument. The shorter the camera's focal length, the smaller the true field is. Make the telescope shorter, and the true field becomes larger. However, a shorter focal length telescope objective is much, much harder to make than a longer one, so we end up needing to make the telescope aperture smaller in order to get a larger field.

So as I understand it you'll want an array of many smaller telescopes, rather than one large one?

Good to know.

I don't know if there is a way around it or not. It depends on the implementation.


The New Horizons LORRI telescope has an aperture diameter of 208 mm, which gives us an aperture of 3.3979e-2 m². A sphere with radius of 1 AU (149.598e9 m) has a surface area of 7.0307e22 m². To get 100 pJ in one second requires the power coming into the aperture to be 100 pW, and at 1 AU, this requires a source power of 206.91 terawatts, and that's in the visual band. By your figs, not even Project Daedalus pumps out this power, and that's over the entire spectrum in which it's radiating. Ralph is even smaller arpeture, 6 cm (60 mm), so it's not going to be any better.

Furthermore, with the LORRI telescope, you have a pixel resolution of 4.94 µrad. At 1 AU, 4.94 µrad corresponds to 369.5 kilometers. Its field of view, on the other hand, is 0.29 degrees squared (1024 x 1024 px²), which corresponds at that distance to a grid 378,300 km x 378,300 km. A ship of similar engine parameters as Project Daedalus can spend 38 seconds thrusting and still remain within one pixel of your resolution if its mass is greater than a tenth that of Daedalus. Unless you know that mass, you don't know if this was 2 km/s delta-V or 0.2 km/s delta-V or even 20 km/s delta-V. In a day's time, it could've gone anywhere from 17,280 km (same view) to 1,728,000 km (four views over). (Actually, we can go all the way down to zero, but let's ignore that.) This gives you approximately 65 frames to scan after a day to catch another burn (58% chance of catching it). After two days between burns, 263 frames (14.5%); three days, 589 frames (6.45%). That assumes that shift to the new field is instantaneous. If not — if it takes a second or longer to acquire a new field to start observing, then the chances go way down.

But even if you're lucky enough to catch the burn, you only have the ship's proper motion (if you were lucky enough to get the triangulation first off — see below). Even if you deduce the exhaust velocity, you can't detect the radial velocity of the craft because the velocity of the exhaust washes any difference out, even if you assume that the exhaust is moving the same velocity as last time. So by the second burn, you only have the proper motion of the ship (but not radial velocity) and the radial velocity of the exhaust (but not proper motion). Dispite what Dr. John Shilling says, while you do have the mass flow (a scalar), you only have one component of the exhaust velocity (a vector), and therefore only one component of the thrust (also a vector). Any mathematician can tell you this problem is unidentified.

Triangulation? Forget it. First, you have to contact another post and tell them that you need its help. But then you have to figure out if the other post is free or if it's observing something else that's interesting, so you have to wait for its reply. Assuming that you get a reply in the affirmative, you then have to tell the other post where to look. That's a three step communication which means that over the burn, light has to go three ways. This means that your partner has to be within 3.8 million km (~13 light-seconds), otherwise the burn disappears before the second post can draw bead on it. Of course, that assumes you know where the next bead needs to be. If it's not, your partner will have to scan along your own line of sight until you acquire the interesting object. The preceding are predicating on the posts catching the burn promptly. If it has to look for the next burn, then there will be less time to triangulate even if it's caught.

I'm going to leave things there for you to mull over.


See above. If the staging area is far enough away, not only will you have trouble detecting the burn and not be boggled by a billion and one other things that could distract you (unless it's a torchship), for short burns you will not be able to determine its trajectory during the burn with any reasonable accuracy. You'll have the mass flow and (maybe) the radial component of the exhaust velocity, but nothing else. That's not nearly enough to determine a trajectory, and especially not in a solar system.

---

Darth Wong on Strollers vs. Assholes: "There were days when I wished that my stroller had weapons on it."
wilfulton on Bible genetics: "If two screaming lunatics copulate in front of another screaming lunatic, the result will be yet another screaming lunatic. "
SirNitram: "The nation of France is a theory, not a fact. It should therefore be approached with an open mind, and critically debated and considered."

Cornivore! | BAN-WATCH CANE: XVII | WWJDFAKB? - What Would Jesus Do... For a Klondike Bar? | Evil Bayesian Conspiracy

Hmm, doing some quick calculations based on the area of a circle, I got a 200.88 meter diameter aperture to be able to image a 206 megawatt light at a distance of 1 AU. Obviously, this would have to take the form of an array of linked telescopes, rather than a single giant scope. The inverse square law would then suggest that being able to detect at longer ranges would be a matter of multiplying the effective radius of the scope by 2 (and hence the effective area by 4) if you want to double the range. So monitoring the entire solar system up to around Pluto orbit would require an effective aperture around 6 km across (or, rather, a truly massive array of smaller telescopes). You'll probably want a rosette of such arrays, of course, so you can monitor the entire solar system.

That sound about right?


If the problem is light lag the solution seems relatively simple. You just have two redundant posts within a short distance of each other (say, a few thousand or tens of thousands of kilometers).


How short is short? A low thrust drive system would need a burn lasting hours or days at least to get any significant velocity. I'd think that would leave you plenty of times to do stuff like, say, track the changing distance between the ship and several of your own platforms with parallax and use that to calculate the probable trajectory.

The enemy can do a long series of tiny burns, of course, but that still leaves the necessity for many thousands of tiny flashes. A few hundred or thousand such flashes in close proximity to a known enemy base could be marked as suspicious, and from there observation could begin, with an eye toward putting together a cumulative profile from thousands of events rather than trying to determine everything from just one sighting.

The problem is still one of collecting area. Even if you can make a 200m diameter array, you will not get the light gathering ability of a single 200m mirror because you will not be able to fill the entire area with collecting surface. And how do you propose to add the signals together? They would have to all be focused on one CCD*, which introduces more positioning and precision problems. As Wyrm pointed out, getting multiple sensors to look at the same thing can be rather difficult, but it is still a better alternative to a close-packed, single array, unless you can afford to literally flood you territory with sensors going every which way.

*This is necessary because alone, each segment and camera isn't sensitive enough to pick up the signal.

I meant an array with a cumulative mirror area equal to that of a 200 meter telescope. I didn't know about the CCD problem though, thanks.

I am somewhat confused as to what you mean by "getting multiple sensors to look at the same thing can be rather difficult, but it is still a better alternative to a close-packed, single array, unless you can afford to literally flood you territory with sensors going every which way." Are you saying have many telescopes focus on one point and put together their images, and how is this different from an array (i.e. how does it solve the common CCD problem)? Or are you saying that it would be a better idea to spread very many small sensors throughout the solar system in order to cover the solar system with many, many short-range sensors?

Edit: I'm also afraid I don't quite understand what the CCD problem refers to. Quickly looking up what a CCD does, it sounds like you're saying the light from the telescopes would have to be focused on one point where it gets translated into electrical signals. Is that it, or am I misinterpreting you? Sorry, optics is another thing that really isn't my field.

Your confusion is well warranted; as written, my statement makes no sense to me either. Consider it retracted until I figure out what the hell I was trying to say. In any case, think of a CCD like digital film (your digital camera has one in it), except that CCDs aren't used up in the process of taking a picture. They are more sensitive and convenient than photographic film, thus their use for anything astronomy related today. That's really all you need to know for this discussion.

The CCD problem refers to the fact that if you use multiple small sensors, each with its own CCD, they will each generate an image using their lower sensitivity, rather than what you wanted, which was the sensitivity of the much larger effective aperture. This is because you have already used the light coming into each telescope to make an image before combining it, so each one acts as if the others never existed. For example, say that you saw 10 photons come in from a source over a previously mentioned ~200m aperture. If you focus this light onto a single CCD, you will see ten photons hit one spot on the CCD. This is good enough to warrant another look at least, once you've eliminated all known objects. But if you have an array of smaller scopes each with their own CCD, you will see fewer photons on more CCDs. This likely isn't enough to register above background, and will be discarded. So for a large close-packed array, they must focus their light on a single CCD to be viable. If you can't do that, you're better off just building fewer rich-field sensors, because you'll get the same sensitivity either way. If you can flood the system with sensors, one of them will almost invariably be close enough to register a hit, rendering the need for such high sensitivity moot. As a SWAG, though, that would require trillions of them, no small undertaking.

Why is a three-step process necessary? Instead, just send a request for help along with any location information in the first message. The second site can act on it immediately if they so choose. Of course, that's only half an order of magnitude improvement in speed or baseline size, so it's still SOL time.

Why are we assuming we need 100 pJ in 1 second? Which figures are correct, yours or GMTs? To quickly requote the pertinent entry:



Basically saying that you can have detection several orders of magnitude less than the 100 pJ you're asking for.

Here's the info on the "detect pluto in 1 second at 4.2 billion km" http://www.spacedaily.com/reports/Pluto ... y_999.html They actually tracked it down with 6 one second exposures on 2 different days, plus they knew where to look, but balancing that with respect to our sci-fi scenario: that telescope was built years and years ago, it's small, and there's only one of it.

GMT's figure for sensitivity of the CCD is relatively similar to that of Pan-STARRS. Second, why are we using figures from LORRI instead of Pan-STARRS? Because one's flown, and the other is a ground based telescope? Pan-STARRS not only has a greater field of view, but also a greater resolution, at .26 arcseconds or 1.26 µrad. This is approximately 4 times better resolution, and it's still possible for it to increase resolution with better technology. One of the primary problems with current technology is pulling off and processing the 1.4 Gigapixel image. It's literally terabytes of information that they get each night, that they have to compare with earlier observations.

As for cueing other sensors to look at a given patch of sky so that you can triangulate: in a plausibly large system, you won't need to. Sensors would be staring, so it's a matter of asking another sensor that's also staring in the right direction whether is saw something at the right time. This also solves the problem of posts needing to be looking in the right direction at the right time to pick up something.

Determining the trajectory of the ship, if you can get a sufficient number of observations from widely enough seperated points, should be doable. The motion of the ship in each of the 2D fields of view should resolve to a single trajectory track.

---

"preemptive killing of cops might not be such a bad idea from a personal saftey[sic] standpoint..." --Keevan Colton
"There's a word for bias you can't see: Yours." -- William Saletan

Mostly because I know about LORRI, whereas I didn't know about Pan-STARRS. Also, LORRI performed a quantifiable demonstration with its one second detection of Pluto at 4.2 billion kilometers, simplifying the math I had to do for my example.


I extrapolated from LORRI to Kepler, though, which is a 0.95 meter effective aperture telescope with a mind-boggling ten degree FOV, which seemed to be the ideal instrument to carry aboard a spaceship, because a couple of these could scan the whole sky in a very short period of time. Though to take advantage of the theoretical resolution limit of a 1 meter telescope (0.25 arcseconds) across Kepler's 105 deg2 field requires a 22 gigapixel camera generating over 68 terabytes of information per all-sky (well, half a sky, broken up into 196 areas to image,) assuming 8-bit pixels and single exposures. Multiply by however many images you're planning on stacking to eliminate sensor noise.