Posted by Ask an astronomer what searching for alien life actually looks like on a Tuesday afternoon, and you will not get a story about antennas waiting for a message. You will get a plot. A jagged line of brightness against wavelength, error bars big enough to argue about, made from starlight that grazed the edge of a planet a hundred light-years away one night. That line is where most of the field now lives. The question has narrowed from “is anybody out there” into something a telescope can actually be pointed at: what molecules are in the air of that world, and can anything other than biology put them there?
It is a chemistry problem wearing an astronomy costume.
A Rainbow With Slices Missing
The workhorse method is transmission spectroscopy, and the idea behind it is almost embarrassingly simple. When a planet crosses in front of its star, a thin ring of starlight filters through whatever atmosphere the planet has. Molecules up there absorb at their own particular wavelengths, so the planet looks very slightly bigger at the colors its air is opaque to and slightly smaller everywhere else. Subtract the out-of-transit spectrum from the in-transit one and the leftovers are, in principle, a fingerprint of the gas.
In practice the signal is brutal. For a puffed-up hot Jupiter the extra dimming might be a few hundred parts per million. For a small rocky world you are hunting tens of parts per million, and if an alien observer tried this on Earth crossing the Sun, our entire atmosphere would register as roughly one part in a million of the star’s light. Kepler taught the field to trust dips of a few hundredths of a percent. Atmospheres demand far better, from a signal you cannot repeat on demand, because there is one transit per orbit and no more.
There is a second trick for hotter planets. Watch the system as the planet slips behind its star and back out, and the difference between those two states is the planet’s own thermal glow, carrying its own absorption features. JWST has been doing this since 2022, with loud results: carbon dioxide in the giant WASP-39b, then sulfur dioxide that looks manufactured by photochemistry rather than inherited at birth. Genuinely new capabilities, on the easy cases.
The star itself is often the real adversary. Cool red dwarfs are blotched with spots and bright faculae, and if the planet crosses a spotted patch, that contamination stamps itself onto the spectrum in ways that impersonate molecular absorption. A large fraction of the argument in any given paper is not about the planet at all. It is about whether the team modeled the star correctly, and which of a dozen defensible reduction choices the pipeline made at three in the morning.
Nothing on the Shopping List Is Proof
The gases people talk about are oxygen, ozone, methane, water vapor, carbon dioxide, nitrous oxide, and a small zoo of sulfur compounds. Oxygen is the celebrity, because Earth’s oxygen is almost entirely the exhaust of photosynthesis. Find it somewhere else and you may have found a biosphere.
Or you may have found a planet losing its ocean. Ultraviolet light from an active red dwarf splits water vapor, the hydrogen escapes because it is light and the planet’s grip is weak, and the abandoned oxygen accumulates with nothing alive anywhere near it. Methane has abiotic sources too, most obviously serpentinization, where water reacts with certain minerals and releases it. Every item on the list has a boring explanation available. That is the central difficulty of biosignature science, and why the field talks in probabilities rather than announcements.
So the strategy shifted from single molecules to disequilibrium. James Lovelock made the argument in 1965, while thinking about how to test Mars: a dead atmosphere relaxes toward chemical equilibrium, and life holds it away from equilibrium by pumping things out faster than chemistry can clean up. Oxygen and methane cannot comfortably coexist. Methane oxidizes away on a timescale of about a decade. Seeing both in quantity means something is resupplying the methane fast enough to keep losing that fight, continuously, for as long as anyone watches.
The best sanity check ever run on this was aimed at us. In 1990 the Galileo spacecraft swung past Earth for a gravity assist, and Carl Sagan’s team used the flyby as a blind control experiment: could the instruments detect life on a planet already known to be crawling with it? They found the oxygen-methane disequilibrium, a strange jump in reflectance near 700 nanometers that turned out to be chlorophyll, and narrowband radio. It worked. It also required flying through the neighborhood, which nobody gets to do at a hundred light-years.
K2-18b and How Quickly a Claim Wobbles
K2-18b deserves a walk-through, because it shows the whole machine running in public. About 8.6 Earth masses, roughly 120 light-years away, orbiting a red dwarf at a distance where liquid water is at least conceivable. Too large to be a rock, too small to be Neptune. One popular reading calls it a hycean world, a deep ocean under a thick hydrogen envelope, though that interpretation was contested from the start.
Hubble reported a water signature in 2019, later reinterpreted by some groups as methane instead. JWST looked in 2023 and found methane and carbon dioxide with real confidence, plus a faint suggestion of dimethyl sulfide, a molecule that on Earth comes overwhelmingly from marine plankton. That last item is what reached the headlines. A follow-up in 2025 argued the sulfur signal had firmed up.
Independent teams then reran the reductions under different assumptions and watched most of the significance drain out. Dimethyl sulfide has meanwhile turned up in data from comet 67P and in an interstellar cloud, which rather damages the premise that only plankton can make it. Nobody was caught doing anything improper here. This is what a three-sigma hint looks like while it is being fought over, and the next candidate will go through the same grinder.
The Same Argument, Where We Can Fly
Searching for alien life inside our own system runs on identical logic with far better data, and the results are humbling. Phosphine at Venus in 2020 is the clean example: a team led by Jane Greaves reported about twenty parts per billion in the cloud deck using two telescopes, and phosphine on a rocky planet is hard to make without biology. Reprocessing the ALMA data shrank the signal toward nothing. Later observations have gone both ways, and six years on it is still unsettled. Mars has its own open file, where a trace of methane measured from the surface has never been confirmed from orbit.
Then there is Bennu. OSIRIS-REx dropped its capsule into the Utah desert in September 2023, and the sample turned out to be loaded with amino acids and nucleobases, the letters of the genetic alphabet, delivered by a lump of rock that was never alive. Prebiotic chemistry is apparently cheap. That is encouraging and inconvenient in the same breath, because it raises the bar for calling any organic molecule a sign of biology.
The Planets We Want Are the Hardest
Since 51 Pegasi b turned up around a Sun-like star in 1995, found by Michel Mayor and Didier Queloz, the catalog has grown into the thousands, mostly courtesy of Kepler and now TESS. Only a handful of those worlds are useful for atmospheric work. Kepler-186f and Kepler-452b are landmarks in the tally and hopeless as spectroscopy targets, too far away, orbiting stars too faint to give up a clean spectrum.
TRAPPIST-1 was supposed to be the answer. Seven Earth-sized planets around a dim red dwarf forty light-years off, announced in 2017, several in a plausible temperature range and transiting often, which matters when you need to stack dozens of events. JWST has since measured the daysides of the two innermost worlds and found them consistent with bare rock. Flares may have stripped them long ago. The outer planets remain open, and each costs an enormous amount of telescope time to settle.
Proxima b, detected in 2016 around the nearest star to the Sun, does not transit from our viewing angle. No transit, no transmission spectrum. For that world the road runs through direct imaging and high-resolution spectrographs on the Extremely Large Telescope, whose thirty-nine-metre mirror is going up on a Chilean ridge, with the aim of prying planet light away from a star that outshines it by a factor of millions.
Nobody yet has an instrument capable of proving a distant planet is inhabited. What exists are instruments capable of producing a claim that other instruments cannot immediately kill. Lower bar, still extremely hard to clear.
What a Real Detection Would Look Like
Not one line in one spectrum. Several gases that clash chemically, seen by two different telescopes, reduced independently by teams that dislike each other’s methods, holding steady across repeated observations, on a planet whose mass, radius, star and history let the story hang together. Ideally something that varies with the seasons, since a biosphere breathes and geology mostly does not. It would take years and be tedious for long stretches. That is a feature.
Searching for alien life has become an argument conducted in absorption lines, and it is wide open right now, with several contested candidate signals sitting in data from telescopes already on the ground and in orbit. If you would rather follow the chemistry than the headlines about it, SETIworld tracks these results as they land and the arguments that follow them, and there is room for more people in that conversation.
