No reason

[james_davis_nicoll]

How early would we have been able to notice an object of roughly sunlike luminousity with a surface temperature of about 300 K, somewhere within 100 light years of the Sun?

Comments

[doc-lemming.livejournal.com]

How do you have roughly sunlike luminosity and a surface temperature of 300K?

And are you really asking when did we have the technology to notice such, or is the answer (distance in light-years) years after it was lit?

[james-nicoll.livejournal.com]

It's very big, roughly one AU in diameter, so the low power output per unit area adds up to about the same total for the smaller but brighter-per-unit-area Sun.

[james-nicoll.livejournal.com]

I mean "one AU in radius."

[rosefox]

About 100 years after it appeared out of nowhere.

[mckitterick.livejournal.com]

Ba-da-boom, tsch!

I was going to suggest the same *g*

Seriously, though, it would be plenty bright - more than nebula-bright, I'd estimate - at that luminosity and size. I'm just curious how it would get such luminosity at such a low temperature. What is this thing? A giant comet? A dead, outgassing star? A relic of interplanetary war?

[dd-b.livejournal.com]

This issue is detecting the energy -- if the surface is at 300k, it's not radiating much in the visible spectrum, and ground-based IR observatories aren't so hot.

Do we get bunches of "radio" in black-body radiation from 300K? I'm not too expert with this stuff. We might notice that before the actual IR, in many years in the past.

[rfmcdpei.livejournal.com]

Perhaps a Dyson sphere.

[beamjockey.livejournal.com]

I'd guess 1970s at the earliest, maybe IRAS in the 1980s if you have to survey the infrared sky to notice it. The thing can't be putting out much energy at visible wavelengths.

[dd-b.livejournal.com]

What about the other direction, radio? (No, I don't know the black-body radiation equations off the top of my head). That might be noticed more readily than IR, depending on the relative power.

[autopope.livejournal.com]

If it's a black body, then it's going to be near-as-dammit invisible at visible wavelengths. Right?

But what about microwave emissions?

(Hint: is James contemplating Matrioshka brains? (Checks Wikipedia, bounces off entry with yet another reference to self, found Robert Bradbury's web server is down, but this paper may be relevant.)

[(Anonymous)]

It sounds like a Dyson sphere -- 1 au radius, radiating a Sun's worth of energy but stepped down to room temperature.


Doug M.

The 1940s

[(Anonymous)]

Radio astronomy started in 1932, when Karl Jansky discovered radio emissions from the center of the Milky Way. But it really took off after WWII, using military spinoff technologies to build the first specialized radio telescopes.

300K gives a peak output around 10 micrometers, which is IR rather than radio. However, it would still be a respectable radio source too -- far below solar luminosity, but bright enough to detect at <100 LY.

So, around 1950 it will be spotted as a faint radio source By the middle of the decade parallax measurements will show that it's nearby. IR astronomy will arrive in the 1960s, at which point everyone will suddenly sit up.

N.B., detection will not need to wait for IRAS -- there were perfectly good IR scopes a generation earlier. They just had to squint through our dense, IR-opaque atmosphere, is all. But they were good enough to spot IR sources all over the place, and they'd be good enough to spot a nearby Dyson sphere.

So, I'd say 1968 +/- three years. Sometime during the Apollo Program, if you like.


Doug M.

Re: The 1940s

[trogon.livejournal.com]

IR sources pre-IRAS, sure.

IR sources with the same temperature as the atmosphere? I'm not convinced.

According to http://astro.uchicago.edu/cara/research/site_testing/brightness.html the sky brightness at 9-11 microns at the South Pole in winter -- absolutely optimal ground-based conditions -- is about 30 Jy/square arcsec. The site says that's about ten times better than other sites, and in reality space-based IR satellites are easier than south-pole-in-winter sites, so let's take the "other sources" (Mauna Kea, etc.) For the sake of a back-of-the-envelope calculation let's assume the source emits all its radiation in a one-micron bandwidth between 10 and 11 um, which at 30 pc gives us about 1000 Jy/square arcsec -- three times brighter than the sky. (You can do the integral if you want to get a real /Hz into the equation, but I'm not going to bother). I don't know offhand what the resolution of the early IR scopes was, but it will probably be an issue as well in smearing out the source flux.

You may be able to do it; it's on the hairy edge, and if you have a radio detection previously to convince people there really is something there it starts becoming plausible. But it isn't a slam-dunk pre-IRAS.

(IOW, I'd have a lot of trouble with such a thing suddenly being discovered now in our timeline, since IRAS should have seen it. I could suspend my disbelief enough to have early balloon-borne IR experiments see something tantalizing and IRAS getting a slam-dunk, in an alternate timeline.)

Re: The 1940s

[carloshasanax.livejournal.com]

I'm having some fun looking through the 1970s AFCRL Sky Survey papers on adsabs. I expected the resolution to be considerably worse.

But I think this case might be a little too dim. Not sure though.

Re: The 1940s

[carloshasanax.livejournal.com]

Found this:

Two new survey instruments were built by Hughes Aircraft Co. in 1970 with ARPA funds under the Air Force Space and Missile Systems Organization (SAMSO) management. These instruments, designated as the HISTAR sensors, were patterned after previously developed sensors for use in an exo-atmospheric environment. The doubly folded f/2.2 Gregorian telescope had beryllium optics with a 16.5-cm aperture and an effective collecting area of about 150 cm^2. The 1.2 degree cross scan field was covered by three linear staggered arrays of doped Ge detectors; each array contained eight 3.3' x 10.5' elements and was filtered to cover the 3-5-um, 8-14-um, and 16-24-um spectral regions. The sensor was cooled from a supercritical helium reservoir. Seven experiments were flown on Aerobee 170 rockets from WSMR between April 1971 and December 1972.

This is from Price, "The Infrared Sky: A survey of Surveys", Publications of the Astronomical Society of the Pacific, 100: 171-186, 1988, and Price was one of the guys who worked on it.

The elements are too big to resolve it -- I am pretty sure that they were originally designed to detect ICBM flares or anti-satellite weapons -- but is that going to matter too much? A lot of follow-up work seems to have been figuring out what smear matched what dot on a plate.

On the other hand, it looks a little too dim (or the effective aperture too small). Price reports:

... although the average detection level in the AFGL catalog of Price and Walker was estimated to be about 2.5 x 10^-16 W cm^-2 um^-1 or 100 Jy (Jy = 10^-26 W m^-2 Hz^-1) at 11 and 20 um based on log (no. of sources) vs. log (flux) plot it was subsequently discovered that about 10% of the sources were spurious.

But, you know, NMF.

Re: The 1940s

[jamiam.livejournal.com]

Also, IRAS was a proper all-sky survey, you know? You'd have to be integrating and sky-subtracting for a loooong time on just the right part of the sky to see it... which I assume you'd only do if you were making a targeted observation of something else.

Did anybody calculate the radio luminosity for a 300K bb?

Re: The 1940s

[jamiam.livejournal.com]

Ooops, which T already said below.

Re: The 1940s

[(Anonymous)]

>So, I'd say 1968 +/- three years.

Good guess - I think around that time they started getting first useful
IR spectra of circumstellar dust in the 8-13 micron window.

Andreas Morlok

[trogon.livejournal.com]

In other words, a Dyson sphere?

Not until we got space-based infrared telescopes. There's rather a lot of local 300K blackbody to deal with otherwise. IRAS's 12 micron survey would have found it; you'd need a survey since there'd be no reason for pointed observations to look at a blank bit of sky, and early IR detectors were *tiny*.

[carloshasanax.livejournal.com]

There's a window in atmospheric opacity around the 300 K blackbody peak, about 10 microns -- 10000 nm -- so not a problem there.

The real problem was instrument sensitivity. Bolometers are strange devices. The Mount Wilson and Mount John surveys of the late 1960s used liquid nitrogen cooled lead sulfide detectors, with a detection peak around another window at 2.2 microns -- 2200 nm -- or about 1500 K. That's well on the exponential decay side of the blackbody spectrum for 300 K, so unless the source is close in, probably not.

On the other hand, it's going to light up liquid helium germanium detectors, which could detect longer wavelengths, corresponding to lower blackbody temperatures -- also around the same time, but with different search strategies, since they used brief rocket flights and balloons.

Something interesting will be making the rounds in 1971-3, is my guess.

[carloshasanax.livejournal.com]

Let me amend: something will be cataloged then. How interesting they'll find it, I dunno.

[webbob.livejournal.com]

Wouldn't this partly depend on the mass of the thing? Or do you mean "notice" solely in the sense of detecting EM radiation from it?

[carloshasanax.livejournal.com]

Hah! found the paper: Sagan (of course) and Russell Walker, the AFCRL/AFGL/IRAS guy, who wondered about this problem back in 1966, Astrophysical Journal 144, 1216.

Problem is, I can't figure out how the early Air Force sounding rocket-based survey instrumentation relates to this data. (Hm, why would the Air Force not make the specs of fast precise infrared detectors available in the early 1970s in the scientific literature. Or maybe my search-fu is weak.)

[trogon.livejournal.com]

Hah, I didn't even think to look at ADS (probably because I forgot the early stuff is publicly available and having left the field no longer have access to new stuff).

So it looks like I was overly pessimistic.

[carloshasanax.livejournal.com]

It does look like the Air Force sensors weren't quite sensitive enough, though from what I have been able to gather, they used fairly loose criteria in their data reduction, so maybe.

But I think your hunch was right: it's probably IRAS, unless it's close in.

[(Anonymous)]

This has been an interesting discussion.

So, if I have this right, the consensus is:

1) Faint shortwave radio source visible in the 1950s, down near the limits of sensitivity; parallax suggests it's nearby, but it's just a curiosity.

2) IR telescopy in the late '60s shows that there's something there, but unless the the beast is pretty close it's right at the bleeding edge of what can be resolved.

At this point I think we enter "buzz of speculation" territory. What looks like a Dyson sphere from ~50 ly away? They'll know it's a blackbody. A brown dwarf would be the right temperature but much too dim. Protostars don't usually appear in isolation, and anyway it's off the temperature-luminosity curve for those too. I'm guessing the conservative consensus will be "small dense nebula being warmed by [handwave]."

3) Sometime in the Carter administration, Carl Sagan points out that "object NCL 1080761 *could* be a Dyson Sphere, but ha ha that's wildly unlikely."

4) January 1983: IRAS goes up. NCL 1080761 is its first target.

Six hours later, all hell breaks loose. At 10 micrometers this is one of the brightest objects in the sky. There's no natural object with this signature. People will have been speculating for a decade by then, so it won't take long for them to reach the obvious conclusion.

N.B., IR astronomy is now good enough that we can say there are no Sol-type Dyson spheres within several hundred light years. We might miss little ones around red dwarfs, but there are no Sol-luminosity 300k sources in our neighborhood. Alas.


Doug M.

For Fuck's Sake Technorati; This Is Ridiculous

[(Anonymous)]

And then, there's the vital role of sci-fi.

So back in the late 40s/early 50s, the chaps from Manchester University were playing around with those old German Riesen Wurzburg air defence radars; the hydrogen line was discovered using the receiver side of one as a radio telescope. They note a curious result; no-one notices very much expect that Bernard Lovell pushes even harder for funding for Jodrell Bank.

But Arthur C. Clarke does, perhaps through one of his old Radar Research Establishment/TRE pals now working at the Cambridge Maths Lab with Maurice Wilkes, who of course knows Turing's former colleagues at Manchester. He works the idea into a short story, which appears in New Worlds in 1962; rapidly recognised as a classic.

Now, during the Cuban crisis, the Lovell telescope was wired up by the RAF as an early-warning sensor; I don't know what it was meant to detect (telemetry? COMINT? backscatter from radars?), but for our purposes it puts people in the US-UK air/nuclear establishment in touch with the idea. In the late 60s, UMIST students are working on an IR scope, and of course reading Clarke.

How do they get it up there? I haven't decided whether this saves the UK space programme, with Ariel getting an actual payload, or whether they do something hackerish like getting it taken up on a high-altitude Canberra aircraft - one reached 70,000ft in 1957 and NASA still uses two for Shuttle-tracking missions. Naturally, the first target is straight at the radio coordinates...and all hell breaks loose.

Boosted by an unexpected brush with the space age, Harold Wilson is re-elected in 1970...

[carloshasanax.livejournal.com]

Much too dim for radio, I think the consensus is. (Think about it. The human body is about 310 K -- how much radio do _we_ emit?)

Barely detectable in the early 1970s, but it will look like a blackbody with a peak around 10 microns, which is what one of the primary researchers is specifically looking for. The Air Force program will improve through the early 1980s, but limited coverage. Call it a one in three chance.

Sagan will say a lot of things.

In any case, IRAS will spot it pretty quickly. It won't be one of the brightest 10 micron objects in the sky, or even in the top thousand. But it'll be solidly in the second tier. No associated object in the visual spectrum, even at the limits of ground-based photographic detection (I think, my calculations are rusty). They'll search the plates and not find anything.

Then things start picking up. It's not *unnatural* per se, but it's funny looking: no dust, no nebula. They'll get parallax: it's near not far. There won't be any measurable radial velocity. No variability. Interferometry will show something that looks more like a cueball than a star.

Mid-1990s for it to really seem like a big black monolith.

[(Anonymous)]

"It won't be one of the brightest 10 micron objects in the sky, or even in the top thousand."

It won't?

James said within 100 LY; split the difference and say 50. One solar luminosity at 50 LY is sixth magnitude.

At 10 microns, what's brighter than that? The bolometric correction for most nearby stars drops them pretty far down; at 10 microns, most of the sky is dark. (Though Venus is damn impressive.)

What am I missing?


Doug M.

Doh! "The Technorati monster escaped" is not funny

[(Anonymous)]

Answering own question, btw, the use of the Lovell dish as a missile early-warning system; they hacked it for use as a honking great radar and had it pointed towards Russia.

[carloshasanax.livejournal.com]

Um. I looked up papers analyzing the original list. I do that. You know.

The stars in the visible night sky are mainly giants, and the same is true for the infrared sky. Just different ones.

Agree with the consensus...

[prof-brotherton.livejournal.com]

Late to the party, but I'll pile on.

IRAS in the 1980s would have been the first reliable way of noticing a "relatively nearby" Dyson sphere, both due to the sensitivity at the right wavelengths and the all sky survey. There would have been potentially a big problem with confusion, however, in that the spatial resolution was on order of an arcminute, and there are a lot of sources in the Galactic plane.

Higher resolution work more recently, with greater sensitivities, like Spitzer's GLIMPSE survey (in conjunction with 2MASS), could turn up a Dyson sphere at much greater distances. I wonder if anyone has looked? It would be easy to do...a day's work probably. Hmmm...could be some complicating issues I've overlooked, but it would be a fun thing to do.

Re: Agree with the consensus...

[prof-brotherton.livejournal.com]

Protostars would be an issue, yes, but perhaps you could exclude them easily enough by avoiding known regions of star formation in the search. There are plenty of people who work to identify these areas in GLIPMSE. Distance vs. intrinsic luminosity is always going to be a concern, however, and it may not be easy at all to figure out you've got a Dyson sphere or not.