[matteocantiello]

I thought the idea of searching for industrial pollution was actually a pretty bad one. We've been polluting at detectable levels for just ~100 years. And if we keep doing it we'll likely wipe us out in a few hundred more. Overall it seems very likely industrial pollution is a very short-lived phase of any civilizations, so chances of detecting this signature are tiny.

[naikrovek]

Pollution capable of completely killing the dominant species of a planet would last a long time after the creation of that pollution ended.

The pollution would vastly outlive the polluters, so it would be detectable for some time after their demise, especially if anything they automated also polluted. That stuff would continue to pollute after its creators all died, at least a bit.

Also, it seems that the chance of detecting pollution would increase linearly as we continue to discover more planets.

[vladTheInhaler]

The phrases "a long time", "some time", and "at least a bit" are doing a lot of heavy lifting there. Oxygen has been in the Earth's atmosphere for over 2 billion years. How long, roughly, is "some time"?

[naikrovek]

How big, roughly is the planet in question? How thick, roughly, is the atmosphere? What, roughly, is the composition of the atmosphere? What is the median temperature of the planet? List all species capable of metabolizing pollution back into harmless compounds. What are their lifespans? How do they reproduce?

...hopefully my point is made, now. I can't give details because they don't exist.

If a species produces sufficient pollution that it kills that species, it is a reasonable assumption that it was also produced at a rate far higher than the planets natural ecosystem could absorb it. It is also reasonable that the pollution also kills one or more species which were consuming or absorbing the pollution, slowing the rate at which it can be removed, or maybe even stopping that process completely. Pollution does not disappear at the same time pollutants stop being introduced into the atmosphere.

There will be a time after which no pollutants are produced and before no pollutants are detectable. That is a non-zero amount of time. That is why I say that pollution could one day be detected on a planet that has not had a living civilization for "a long time." "A long time" being "more than five minutes" if you really want me to nail it down.

[vladTheInhaler]

> I can't give details because they don't exist.

Right. And if the argument depends on those details, then it's not a very convincing argument.

[naikrovek]

My argument is that pollution outlives a society that it kills by a non-zero amount of time.

[vladTheInhaler]

I understand. And my argument is that nonzero isn't a good enough lower bound to justify investing in something like that. Oxygen is produced by even very simple organisms, and thus has been present in our atmosphere in significant concentrations for billions of years, currently makes up 20+ percent, and its residence time is 4,500 years [1]. On the other hand, many CFC concentrations are measured in parts per trillion, and have a residence time measured in years or decades [2], and have only existed at all for a century or so at most. Even if every single living thing on this earth died tomorrow, oxygen would be detectible long, long after many pollutants were gone. And chances are, if anything were to survive the kind of disaster that would wipe out humanity, it would be something like the single-celled organisms that excrete oxygen. I'm prepared to believe that you're correct, but you have to present some more compelling evidence.

[1] https://en.wikipedia.org/wiki/Oxygen_cycle#Capacities_and_fl...

[2] https://cdiac.ess-dive.lbl.gov/pns/current_ghg.html

[matteocantiello]

Why would the pollution vastly outlive the polluters? It's not obvious to me. Mixing and recycling timescale of an atmosphere seem very short compared to other relevant timescales of the problem. Impactors, volcanic eruptions... the memory of the atmospheric impact of those events is in the sediments, not the air we breathe (e.g. KT boundary). Finally, civilizations that do not destroy themselves in the short-lived polluting era should move to a clean nuclear fusion era. We are probably ~50-100 yrs away if we don't fuck up.

[naikrovek]

If the pollution is severe enough to KILL the species creating it, it is already being produced far faster than nature can absorb it or counteract it. It also seems feasible that other species are killed as well; species which may have helped clean the pollution from the planet, which would slow the cleaning process or, if enough species die off, stopping the cleansing process almost entirely.

Pollution doesn't vanish when it isn't being made anymore.

So, if there is so much pollution that it both kills life on the planet, it will take a good amount of time (millions of years? How big of a planet are we talking about?) to be cleansed back to undetectable levels.

[GoblinSlayer]

So far volcanic eruptions were the most dangerous for life.

[stefanmichael]

I thought this might be the case as well but at least for CFCs the lifespan without replenishment from human pollution is only about 140 years. so "a bit" for this particular circumstance is not very long at all in the eyes of the universe

[naikrovek]

CFCs are not the only pollutants.

An alien society may produce pollutants which last a million years. Plastics can last millions of years.

Sufficient pollution could cause climate change severe enough to keep all kinds of things airborne for who knows how long. We are constantly surprised by what we learn about other worlds. It is silly to me to presume that anything we know about Earth pollution will be a given on a planet from another star system.

[newsbinator]

I wish we could develop a technology to detect plastics across space.

[zelon88]

I believe that we've probably altered the geography of Earth to an extent that it would be trivially easy to determine from great distance that it was once inhabited by intelligent beings. Even if all of our cities are gone, the mineral deposits left behind would be so concentrated. Nevermind the nuclear endeavors.

We were able to detect dinosaurs on our own planet 65 million years after they went extinct, and they didn't even pollute anything.

[sjy]

I recommend Schmidt and Frank’s article on the Silurian hypothesis [1] for a surprisingly different take on this. The full text is on arXiv [2]. “We summarize the likely geological fingerprint of the Anthropocene, and demonstrate that while clear, it will not differ greatly in many respects from other known events in the geological record” – that is, the record we can observe from a distance of zero!

[1] https://en.wikipedia.org/wiki/Silurian_hypothesis

[2] https://arxiv.org/abs/1804.03748

[stefanmichael]

I think that you might be right but it would be more conclusive as evidence than detecting just oxygen. CFCs are not something that occurs spontaneously in nature, so essentially if we did get a positive result it would be meaningful.

To your point getting strictly negative results would not necessarily mean anything one way or the other.

[matteocantiello]

Yes, I agree with the point that Oxygen is not answering the question definitively. The issue is that say, you build a very, very expensive space telescope and you have to decide if you want it to look for Oxygen or CFC. CFC is a shot in the dark: likely you won't see anything (see my previous comment). Which means likely no data, no PhD thesis, nothing. You pretty much wasted a few billion dollars and you're not ruling out much because what if civilizations only use chemical energy sources for ~200 yrs, then they all switch to nuclear fusion? On the other hand is likely Oxygen will show up as a signature in some exoplanets, and a lot of new understanding about exoplanet atmosphere and geology would follow. Which, among many other things, might help us narrowing down planets where to look for life and how. And next iteration of instruments we might be able to really discover life. So this is why review panels do opt for choices that are more likely to benefit the scientific community at large. It might sound a bit incremental at first, but it's how progress is made. This said, it's not like astrophysicists are not thinking about detecting technosignatures in exoplanetary spectra, see e.g. 3.2 in https://www.liebertpub.com/doi/10.1089/ast.2009.0371 and e.g. https://elib.dlr.de/119683/1/1803.05179.pdf