Say we find extraterrestrial life, how to prove it?

Originally posted by FabianFnas
"Say we find extraterrestrial life, how to prove it?"

If you have a strong oxygen O2 line in the spectrum from a star that cannot be accounted for from the star itself, then this star has probably a oxygen rich planet. There is no way that O2 can have a natural cause, other than a biological life. That's a proof of life.

As far as I know, no spectrum from a star has this O2 line. Other than our solar system.

That is true, but that is only if you find a planet with life that evolved to a level comparable to our own planet and that is like hitting the jackpot so to speak. No planet or moon in this solar system other than Earth has that much O2 so if life were on Europa (for example) that would not be the case. You would likely have to send an advanced robot to probe under the ice to know. Any planet in another solar system cannot be seen from here.

Originally posted by FabianFnas
...You can easily distinguish a O2 line from a planet from that of a star (does stars have O2 lines at all...?) because of the periodic doppler shift.

Oh yes, I didn't think of the periodic doppler shift.
The periodic doppler shift might satisfies my said condition of:

"you will need to somehow have a way to greatly narrow down which photons come from reflected star light off the planet and which come directly from the star ( i.e. without reflecting off the planet ) before you can get good evidence of a O2 rich planetary atmosphere. "

providing the light direct from the star is not so bright compared to the star light reflected from the planet that any spectrum properties of the star light reflected from the planet would be undetectable because of that massive relative difference in brightness.
If for every photon detected comes from that reflected off the planet, there is a billion photons that come directly from the star, you would have to first detect many trillions of photons before you can statistically detect any periodic doppler shift coming from that reflected off the planet else any such shift would be statistically 'shined out' (if that is the right term ) by the photons coming directly from the star.
I am not sure if that has been done or if that can be done yet or if it could be done any time soon. Not sure but I think that would be difficult.

Originally posted by humy
Oh yes, I didn't think of the periodic doppler shift.
The periodic doppler shift might satisfies my said condition of:

"you will need to somehow have a way to greatly narrow down which photons come from reflected star light off the planet and which come directly from the star ( i.e. without reflecting off the planet ) before you can get good evidence of a ...[text shortened]... be done yet or if it could be done any time soon. Not sure but I think that would be difficult.

Yes, a star outshines a planet if you count photons of every wavelength.
But if you just look at a particular wavelength, the one produced or absorbed by the O2, then you can see that even a planet can outshine a star.

Are there any O2 in a star? Perhaps only in a cold enough star where oxygen can combine into O2, but I think oxygene would rather combine with hydrogen into H2O. In the atmosphere of hotter stars oxygen cannot combine into O2. In an planetary atmosphere, however, if there are oxygen, it gladly combine into O2.

Originally posted by FabianFnas
Yes, a star outshines a planet if you count photons of every wavelength.
But if you just look at a particular wavelength, the one produced or absorbed by the O2, then you can see that even a planet can outshine a star.

Are there any O2 in a star? Perhaps only in a cold enough star where oxygen can combine into O2, but I think oxygene would rather comb ...[text shortened]... ne into O2. In an planetary atmosphere, however, if there are oxygen, it gladly combine into O2.

But if you just look at a particular wavelength, the one produced or absorbed by the O2, then you can see that even a planet can outshine a star.

I could be wrong and, if I am, please will someone put me right here, but, my line of reasoning on this is as follows:

lets say we have got a very narrow O2 absorption band between 750nm and 751nm wavelengths in any O2-rich planetary atmosphere (I believe there is an actual O2 absorption band somewhere very near there but probably not exactly there ).
lets also say the photons coming directly coming from the star outnumber those reflected off a very O2-rich-atmosphere-planet by a billion fold (it is likely to be even greater than that! ).
Then if you look at the spectrum of the direct star light and the reflected planetary light combined, the photons of wavelengths detected in that spectrum in the two arbitrary defined narrow ranges of equal magnitude either side of that O2 absorption band, i.e. the ranges from 749nm to 750 and the range from 751nm to 752nm, will appear to be about equal in number as to the number of photons exactly in the range from 750nm to 751nm. That is because for every photon that came from the planet either side of that O2 observation range, there will be a billion photons coming directly from the star. While, even if you get 100% absorption of that O2 absorption range from the planet, the resulting reduction in the number of photons from that absorption band (from 750nm to 751nm ) from the total light coming from the star and planet combined will be about one-billionth! Thus, even if you detected, say, 100 million photons from each of those three wavelength bands (which I personally would intuitively assume unlikely unless the star is relatively extremely 'close'. Any one: is that right? If my argument is flawed, this is where I think it is most likely to be wrong ) , taking into account random statistical error, (or perhaps I should say “error of measurement”? ) , you will not detect a statistically significant (i.e. conclusive ) signature for that O2 absorption.

I assume a similar argument applies to a very narrow O2 emission band

But, as I said, I am not an expert on this so I could be wrong. So:

Anyone:
Am I right or wrong here?

Originally posted by humy

But if you just look at a particular wavelength, the one produced or absorbed by the O2, then you can see that even a planet can outshine a star.

I could be wrong and, if I am, please will someone put me right here, but, my line of reasoning on this is as follows:

lets say we have got a very narrow O2 absorption band between 750nm an ...[text shortened]... said, I am not an expert on this so I could be wrong. So:

Anyone:
Am I right or wrong here?

Using ordinary telescopes that might be true, but the newest generation have tricks that lower the light by the star a billion to one and thus leaves the captured planet with those photons untouched so now it is like blocking out the light from the star and you can now discount the glare from that star. Some of them use tricky circular polarization techniques and others use a simple block in front of the star, way in front of the scope and just about the same angular diameter as the star, hopefully leaving the planet unmodified so those photons can now be counted. BTW, I may be wrong, but I don't think filters are so good as to capture say 750 nm light and reject 751. I may have missed the latest generation of tools but that's what I remember. I worked in photonics for years and we had filters and optical signal analyzers that attempted to do just that but I think the numbers were more like say 750 nm vs 760 nm.
Of course that could have been just because of the level of technology used at Lucent at the time, maybe better tools are out there now.

Well I looked at Edmond Scientific and found this ad for a notch filter and I was right, this one has two notches, one at 355 and the other at 532, UV and green, but the full width/half max wavelength is 10 nm. I would think Edmond would have the best but I may be wrong:

http://www.edmundoptics.com/optics/optical-filters/notch-filters/od-6-multi-notch-filters-for-nd-yag-lasers/3579/