Taking it a step further, so far we have:
Flux density of 1GW reactor at 1 AU: 3.54e-15 W/m²
Flux density of 1GW reactor at 100,000 km: 7.96e-9 W/m²
Flux density of G V star at 4.3 ly: 1.87e-8 W/m²
Now, how do sensors work? They detect photons. To detect something, your detector needs to be able to absorb enough photons from the source to get a meaningful signal.
How many photons per square metre (per second) do these numbers translate to?
Watts are Joules per second. We can find the energy of a single photon easily enough, using E= hc/lambda where h = Planck's constant (6.626068e-34 m².kg/s), c = speed of light (3e8 m/s) and lambda = wavelength of the photon (assuming it's all IR, let's call it 500 nm, or 5e-7 m).
So the energy of an IR photon is 3.976e-19 J.
Taking the 1GW reactor, each square metre of detector at 1 AU distance receives 3.54e-15 Joules per second. Divide that by the energy of a photon and we'll get the
number of photons striking the detector. That turns out to be about 8900 photons per square metre per second.
For the 1GW reactor with a sensor at 100,000 km, that goes up to... rather a lot. 2e10 photons per square metre per second.
For Alpha Centauri, it's even higher of course - 4.7e10 photons per square metre per second.
Putting it that way, I suspect that Alpha Centauri, and a ship with a 1 GW reactor at 100,000 km distance should show up pretty well on a metre-scale IR sensor.
I'm not so convinced about a 1 GW reactor at 1 AU distance though. This is where my understanding of sensors breaks down though. 8900 photons per second per m² isn't a lot, and spreading that out over a square metre would mean that each individual pixel of the CCD sensor would be getting very few (if any) photons.
Heck, I'm not even sure that 2e10 photons/s/m² is that high when it comes to detectability. There are all sorts of factors involved - diffraction, dark current, signal to noise, conversion of photons to electrical signal to register on the sensor, etc. I did find a very technical document about how sensors work at
http://www.nap.edu/openbook.php?record_id=12896&page=23 which may be of some use to people with more background in that topic.
Detectability aside, there are clearly other concerns here. The sensor would need to be kept supercool to be able to detect anything. The detector would have to be rather large too (and probably have some means of focussing photons, like a big telescope). As it happens, the
NASA Spitzer telescope is essentially a big IR detector, and that's four metres tall, kept supercool, and has a mirror about 85 cm in radius. But that's on a stable platform in orbit, with very few stresses on it. Put that on a spacecraft that's accelerating all over the place and it's probably not going to last that long.
And again, it's a matter of knowing where to look for your IR-emitting target in the first place. A telescope has a limited field of view, and a full sweep of the sky won't be possible to do on a mobile spacecraft in a short time at the required resolution.
It may not be
impossible to detect spacecraft via IR (and I never claimed that it was), but I think this and my other calculations show that they aren't necessarily blindingly obvious, and that more often than not at long ranges it certainly will be very difficult to do from a sensor mounted on a spacecraft in Traveller.
.
