NASA is building a new space telescope to search for life on nearby planets. What would it have seen on ancient Earth?

a cylindrical telescope in space looks at earth, both on a starry black background — Artist's illustration of NASA's Habitable Worlds Observatory, with an inset photo of Earth taken by the Deep Space Climate Observatory (DSCOVR) spacecraft. (Image credit: NASA)

NASA's Habitable Worlds Observatory is the agency's next flagship space telescope, designed to do something no previous instrument has managed: directly image Earth-like planets around nearby stars and analyze the light reflecting off their atmospheres for signs of life.

The mission is still years from launch. But the design choices being made right now will determine what it can actually detect. A new paper posted to the arXiv preprint server tackles one of the most consequential of those choices: spectral resolution.

The authors of the study ran a careful analysis of how finely the Habitable Worlds Observatory (HWO) would need to slice up the light from a distant Earth to confidently spot biosignatures in its atmosphere. The question matters more than you might think.

Spectral resolution is how well a telescope can distinguish between adjacent colors of light. Higher resolution means a more detailed atmospheric fingerprint, but it also means longer exposure times, more detector noise, and trickier engineering. Push too high and you blow the mission's observing schedule. Push too low and you can't tell the difference between an inhabited planet and a barren one.

To estimate what type of spectral resolution would be needed to detect biological signatures on our planet in its infancy, the team modeled what HWO would see when staring at versions of Earth through geological time.

Earth's atmosphere has changed dramatically over its history. The Archean Earth, before plants and cyanobacteria got going, had almost no oxygen. The Proterozoic Earth had some, but not much. The Phanerozoic Earth, the one we know, hit roughly 20 percent oxygen once complex life took hold. Each leaves a different spectral signature, and HWO will need to recognize all three.

The headline numbers are surprisingly modest. To detect molecular oxygen, the gold-standard biosignature on a planet like our own, HWO needs a visible-light resolving power of about 140. Ozone shows up at a much lower resolving power of around 7 in the ultraviolet. Those numbers are well within what current optical designs can deliver.

The infrared is harder. Carbon dioxide and carbon monoxide have spectral features that overlap, and if HWO can't tell them apart, it could mistake a volcanically active dead planet for a living one. The team found that a near-infrared resolving power of at least 40 is the minimum needed to break that degeneracy. To characterize an atmosphere through Earth's entire geological history, they recommend a nominal infrared resolving power of about 70.

a blue and green planet dotted with white clouds — An illustration of ancient Earth. (Image credit: Mark Garlick/Science Photo Library/Getty Images)

How did they arrive at those numbers? They generated synthetic HWO observations across resolving powers from 20 to 5,000, then ran each simulated spectrum through retrieval algorithms to see what could actually be inferred about the underlying atmosphere. They factored in detector noise, exposure time, and anti-biosignatures (atmospheric features that would argue against the presence of life).

There are real engineering limits in play. The dark current of HWO's detectors, the tiny background hum of electrons that any detector generates even with no light hitting it, sets a hard floor on what fine resolution can buy you. Pushing oxygen detection significantly further than the baseline would require reducing that dark current by roughly a factor of ten. And pushing to higher resolution for oxygen would roughly double the exposure time needed for water vapor.

The authors are careful about the limits of their analysis. Their absolute exposure times could be off by around 20 percent. And the more philosophical caveat is the one that has always shadowed this work: even a confident detection of oxygen, ozone, methane, and water in an exoplanet atmosphere is not the same as a confident detection of life.

The universe has non-biological ways to make any one of those gases. HWO's job isn't to declare victory on its own: it's to find the candidates worth following up on.

What this paper provides is a clear, quantitative target for the engineers building the instrument. A resolving power of 140 in the visible, 7 in the ultraviolet, and 70 in the near-infrared, with low enough dark current to make oxygen detection routine. That is the spec sheet for a telescope that could, in principle, find signs of life on another world.

Now we just have to build it.

Paul M. Sutter is a cosmologist at Johns Hopkins University, host of Ask a Spaceman, and author of How to Die in Space.