Identifying nearby planets with spectroscopy?

Factual Questions
1

This is another question regarding a story I’m writing.

The situation runs thusly: a couple of extraterrestrials have just arrived in or near our solar system. (We’ll say they just dropped out of hyperspace; that’s the extent of the hand-waving I’m willing to do to allow FTL travel.) They’re expecting to find Earth, but aside from its approximate mean orbital distance they have no solid information about it or the rest of the system. So I had to give some thought to how they might detect and identify the various planets in the system.

My first thought was that parallax measurements would be the obvious way to do it; continue accelerating towards the sun and see which dots move against the relatively fixed backdrop of the rest of the galaxy. But then it occurred to me that at relatively close range (i.e., a couple billion miles or so), there might be a detectable difference between the light reflected from a planet (terrestrial or otherwise) and the light you’d expect to receive from a distant star.

I’ve done some searching, and I’ve found a few things (like this abstract) that seem to indicate that there is a difference between typical planetary and stellar emission spectra, though understandably it seems that all the research in this direction is aimed at identifying planets at distances of many light-years instead of a few paltry billions of miles. I couldn’t seem to find anything specifically saying that this would be the most obvious way of discovering nearby planets, and that’s more or less what I’d like to know.

Or, for that matter, how bright would the Earth appear from a distance comparable to the orbit of Pluto? My initial feeling and some rough calculations seem to indicate that at that kind of distance, none of the planets would be bright enough to suggest their proximity. For example, Jupiter’s maximum brightness as seen from Earth is magnitude -2.9, and at its closest Jupiter is about 588 million kilometers from Earth. Brightness falls off with the square of the distance, so if we were viewing Jupiter from, say, 5.88 billion kilometers away, we would expect it to be 100 times less bright, which I think indicates an apparent magnitude of about -2.9 + log2.5(100) ~= 2.1, which is visible with the naked eye, but still less bright than many stars.

So to summarize, what would be the quickest and easiest way to find the planets in a system you just dropped into, without any prior knowledge? Would the emission spectra be such an obvious indicator at the outset that taking parallax measurements would be redundant? Or is it a sufficiently tricky thing that parallax would still be useful? Of course, they’ll need the parallax data to measure the distances involved anyway, but I’d be much obliged if any Doper astronomers could weigh in.

(For all purposes, we can assume that they have advanced optical systems and computers capable of photographing and scanning very large sections of the “sky” very quickly.)

2

They could have a catalog of stars, and know which “stars” they’re seeing aren’t in the catalog. Then, yeah, they could look at the spectra they’re seeing on those few. As your link says, the presence of Oxygen is a good indication of life, since it’s so reactive, it won’t remain in a planet’s atmosphere for long.

3

If they’re going to Earth as we know it, is there any specific reason they couldn’t just pick up radio waves and use those as a directional guide?

4

From what range are they observing, how long do you want to let them watch before figuring it out, and how far ahead of ours is their telescope technology?

5

If you’re looking at the system from the distance of Pluto, all the rocky planets will be within 3° of the sun.
Model the view here:JPL’s Solar System Simulator
A 3° circle works out to about 7 square degrees.
Assuming you haven’t hypered in so as to put the milky way or an open cluster directly behind the local star, you should see an average of 1 to 2 background stars in that 3° circle.
Polar Project:

Anything beyond that number is likely a planet, but unless you’re unlucky, you will not have to look at more than a hundred or so lights in the sky to decide whether or not they are planets.
If you look at the spectra of those objects, you should see nice hydrogen absorption lines from everything that is not a planet.

6

Well, they already know that the life is located on the third rock from the sun, around 150 million kilometers out. The main goal is to find which point of light corresponds to that planet (which would require the parallax to measure the distance), and incidentally to chart the other planets in the system while they’re at it.

True, they might be able to look for artificial-looking signals on a wide range of frequencies (given a decent amount of time to sit and listen), but that would only give directional information which might not be terribly specific at long range. And it wouldn’t tell them anything about the other planets, which they might be interested in jotting down just for future reference. :wink:

I was thinking something on the order of a billion miles or so. My goal was to make the initial rough plotting process take place during a stretch of conversational dialogue, so maybe half an hour, give or take? They’ll have a pretty high velocity on entering the system; on the order of several million meters per second, which I’m hoping should be enough to get a pretty decent displacement for parallax measurements in short order. As for their telescope technology, I’m assuming they’re capable of considerably more precision; I guess it would have to be at least whatever level of precision is required to accurately measure the distances to the inner planets over a displacement of one or two million kilometers. :smiley:
Squink: Thanks very much for that info, and for the link to the solar system simulator! That will be most useful. So I guess the spectra would be a bit of a giveaway once you’ve looked at each object of interest. I think I might have them jump in a bit closer in that case, to give them a bigger range of sky to search; it would also be convenient to reduce the sublight transit time from a couple of weeks to a only a few days.

7

Keep in mind, I think it’s an equation of Kepler’s, lets us know pretty accurately exactly where in orbit a planet/body will be, once we know the size of the star in question.

Thus, assuming the aliens have a similiar forumla, they will know that at such and such distance, there should be a planet. Then it’s just a matter of scanning the right band of space to find it.

Unless I’m REALLY misunderstanding what it is you’re asking…?

8

Planets can orbit at any radius; Kepler’s Third Law relates the radius of the planet’s orbit to its orbital period. You might be thinking of Bode’s Law, but that was never more than an empirical rule and failed to correctly predict the orbit of Neptune.

9

Bode’s law does pretty damned good for Neptune. It’s Pluto that it fails for.

A more recent reworking, the Blagg-Richardson Law, does better than Titius-Bode (Titius never gets the recognition he deserves), and fit satellite data as well.

10

Titius-Bode is off by about 30% for Neptune (39 AU predicted vs. 30 AU observed), compared to no more than 5% or so for the inner seven planets plus Ceres. Whether or not that’s “pretty damned good” is, I suppose, up for debate. Certainly it’s a much better prediction than it gives for Pluto, but that’s not exactly high praise.

11

the problem with spectroscopy is that planets do not emit light. They only reflect it.

Currently we can use the process on planets as they pass in front of their stars from our perspective. (We can determine the contents of the planets atmosphere as light passes through). In that sense we can find planets with certain atmospheres if we catch the planet in front of the star. (We see a slight change in the spectrum from the star as the planet passes)

Otherwise the problem is knowing where to look. Planets are so small and do not emit enough light, even through reflection, to be seen from great distances.

12

It’s better than the fir Kepler got for the planets with his Nested Platonic Solids model.

13

Astronomers are now confident that most stars have planets.

Yet, we cannot see them at our nearest neighbor - Alpha Centauri (conclusive calcuations will be complete on a suspected planet around apha centauri B after 2011 observations.)

14

If you’re at the orbit of Pluto, the question of where the planets are is obvious. They’re right next to the Sun.

If they want to find Earth, it would be easy for them to just go to region around the Sun where they predict liquid water would be possible, then go up to the observation lounge and look around for bright stars with their naked eyes. Assuming they have eyes. Anything really bright is bound to be a planet. Look at how bright Mars is from Earth. Well, from around Earth orbit, Earth is going to be much brighter than Mars, although not as bright as Venus, although it could be if Earth turns out to be pretty close in orbit. So you look around, you find a couple of really bright objects, and there you go. You’ve found your planets. Venus, Earth, Mars and Jupiter would be trivial to find.

If the question is, “How can aliens find Earth?”, it won’t be hard at all. Or is the question, “At what distance could the aliens reliably find Earth?”

15

From the orbit of Jupiter, everything out to Mars is within 15 degrees of the sun. That’d give you enough sky to scan to make searching tough, plus easy acess to the planets once you found one. Of course, Jupiter is at least 30 light minutes inside Sol’s hyper limit, so you’re going to have to spend some time getting there in any event.