NASA's Next Giant Space Telescope Could Search for Alien Worlds Beyond the Solar System
Why imaging Earth-like exoplanets is so hard, what the proposed Habitable Worlds Observatory would measure, and why biosignature claims demand caution and independent confirmation.
The next large NASA space telescope that astronomers hope will search for Earth-like planets is, as of this writing, a proposed mission rather than a built one. It is called the Habitable Worlds Observatory (HWO), and it follows a recommendation of the 2020 astronomy and astrophysics decadal survey from the US National Academies. Its goal is to image rocky planets around Sun-like stars directly and look for chemical evidence of life in their atmospheres. This article explains why that is difficult, what a successful result would and would not demonstrate, and what remains uncertain.
Where the field stands
More than 5,500 exoplanets have been confirmed, according to the NASA Exoplanet Archive, most found indirectly by the transit method (the small dip in starlight when a planet crosses its star) or by the radial-velocity method (the star's wobble). These methods tell us a planet's size or minimum mass and its orbit. They say little about its atmosphere unless the planet happens to be favourably placed.
The James Webb Space Telescope, launched in December 2021, has changed this for some planets by measuring transmission spectra, the tiny spectral imprint of starlight filtered through a planet's atmosphere during transit. It has detected carbon dioxide and other molecules on gas giants and has begun probing rocky planets around small, cool red dwarf stars, such as those in the TRAPPIST-1 system. But Webb was not designed to study Earth twins around Sun-like stars, and for those the signal is too small.
The technical problem: contrast
An Earth-like planet in a Sun-like system is roughly ten billion times fainter than its star in visible light, and sits very close to it on the sky. Seeing it is like seeing a firefly next to a searchlight from thousands of kilometres away. The proposed solution is a coronagraph or a starshade, instruments that suppress starlight before it reaches the detector. A coronagraph blocks the star inside the telescope and has to cope with any tiny imperfection in the optics, which scatters light and produces speckles that look like planets. This demands mirror surfaces stable to a fraction of the width of an atom, active wavefront control and thermal stability. NASA's Nancy Grace Roman Space Telescope carries a coronagraph instrument intended as a technology demonstration for exactly these capabilities.
HWO is envisioned as a roughly six-metre-class telescope working from the ultraviolet through the visible to the near-infrared, designed to survey on the order of a few dozen nearby Sun-like stars for habitable-zone rocky planets. Its precursor concepts, HabEx and LUVOIR, were studied for the decadal survey. Design details, launch date and cost are not fixed, and the 2030s to 2040s is the timeframe usually discussed.
Why visible and near-infrared light
Infrared astronomy is essential: infrared wavelengths pass through dust and carry thermal emission, and Webb works there. For HWO, however, the key spectral features include molecular oxygen (a band near 0.76 micrometres, in the visible), water, and ozone's ultraviolet absorption, which is why the observatory is planned to cover wavelengths from the ultraviolet to the near-infrared. Different wavelength ranges answer different questions; there is no single "best" band.
Reading an atmosphere
A spectrum shows which molecules are present and in what proportions. Interest focuses on gases that, together, would be hard to maintain without a source. The classic argument, advanced by James Lovelock in the 1960s, is chemical disequilibrium: oxygen and methane react with each other, so finding both in abundance implies something keeps replenishing them. On Earth that something is life. Carbon dioxide and water vapour describe a planet's climate and habitability but are not biosignatures on their own.
A single gas is rarely conclusive. Oxygen can accumulate through non-biological processes, for example when a planet around an M dwarf loses much of its water to space and the leftover oxygen builds up, as modelled by Luger and Barnes (2015). In 2023 and 2025, claims of dimethyl sulfide in the atmosphere of the sub-Neptune K2-18 b, a molecule that on Earth is produced mainly by marine life, drew great attention. Independent reanalyses did not confirm the significance of the signal, and the community regards the result as unresolved. This episode is a practical lesson about the evidence needed before anyone claims "signs of life".
Where machine learning fits
Machine learning is a useful tool in astronomy for specific, well-defined tasks. Shallue and Vanderburg (2018) trained a neural network on labelled Kepler light curves to rank transit candidates and identified a new planet in the Kepler-90 system, which became an eighth known planet there. Classifiers also help sort galaxies in large surveys and flag transient events. In direct imaging, algorithms are used to model and subtract speckle noise. In each case, performance depends on the quality of the labelled training data and on how well it represents the real observations, and false positives still have to be vetted by other methods. Applying AI to biosignature interpretation is far less mature, since there are no labelled examples of alien life to train on.
Ocean worlds within the Solar System
The search is not limited to distant stars. Europa and Enceladus are believed to hold subsurface liquid water oceans, supported by magnetic field measurements and by plumes sampled by the Cassini spacecraft at Enceladus, which contained organic molecules and hydrogen. Titan has a thick nitrogen atmosphere, surface lakes of methane and ethane, and probably a subsurface water ocean. Whether any of these environments is inhabited is unknown. NASA's Europa Clipper and Dragonfly missions are designed to assess habitability rather than to detect life directly. Telescopes and spacecraft therefore address different questions: one looks for statistical evidence of life on many distant planets, the other samples a few nearby places in detail.
Limitations and open questions
- Few targets. Even a large telescope can survey only nearby stars. If Earth-like planets are rare, the sample may yield few or no candidates.
- Habitable-zone definition. The zone is based on the possibility of surface liquid water under particular atmospheric assumptions. Presence in the zone does not imply habitability.
- Clouds and hazes. They mute spectral features and complicate interpretation.
- False positives and negatives. Abiotic oxygen and methane pathways exist, while a living world might not show an obvious signature.
- Cost and schedule. Large observatories have historically exceeded initial estimates and slipped in launch date.
Questions worth testing
- How should a biosignature claim be quantified in a Bayesian framework that accounts for the prior probability of abiotic explanations?
- Which combination of gases and which context measurements (stellar ultraviolet, planet mass, surface pressure) best discriminates biological from abiotic origins?
- How much observing time per target is needed for a confident oxygen detection at a given contrast?
- Can speckle subtraction methods developed with machine learning be validated well enough that they do not create or remove planet-like signals?
The honest summary is that detecting a habitable planet is plausible within a few decades if the planet population cooperates, but claiming life would require several independent lines of evidence and prolonged scrutiny.
References and further reading
- National Academies of Sciences, Engineering, and Medicine. Pathways to Discovery in Astronomy and Astrophysics for the 2020s. National Academies Press, 2021.
- NASA Exoplanet Archive, exoplanetarchive.ipac.caltech.edu (planet counts, updated regularly).
- Shallue, C. J., Vanderburg, A. "Identifying Exoplanets with Deep Learning: A Five-planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90." The Astronomical Journal, 155, 94, 2018.
- Luger, R., Barnes, R. "Extreme Water Loss and Abiotic O2 Buildup on Planets throughout the Habitable Zones of M Dwarfs." Astrobiology, 15, 119, 2015.
- Lovelock, J. E. "A physical basis for life detection experiments." Nature, 207, 568, 1965.
- Seager, S. Exoplanet Atmospheres: Physical Processes. Princeton University Press, 2010.