In a nutshell
- With 40-80 observations of Earth-like planets showing no signs of life, scientists could conclude with 99.9% confidence that life is rare (present on fewer than 10-20% of suitable planets).
- Planned space telescopes like LIFE and HWO will examine enough planets (16-38) to draw meaningful statistical conclusions, even if they detect nothing.
- The accuracy of observations is more critical than sample size—if scientists can’t be highly certain about individual detections, even hundreds of observations won’t yield conclusive results.
ZURICH, Switzerland — In our quest to answer whether we’re alone in the universe, astronomers are preparing for a major advancement: new space telescopes capable of scanning distant worlds for signs of habitability and possibly life itself. But what happens if these ambitious missions find nothing?
A recent study published in The Astronomical Journal explores this scenario, revealing that even a series of negative results from upcoming exoplanet surveys could yield profound insights about life’s rarity in the universe.
The Statistical Power of Zero
The research, led by Daniel Angerhausen of ETH Zurich, tackles a fundamental question in the search for extraterrestrial life: How many planets must we examine before the absence of detections becomes statistically meaningful?
This question isn’t merely academic. With NASA’s Habitable Worlds Observatory (HWO) and the proposed Large Interferometer for Exoplanets (LIFE) on the horizon, scientists are planning missions that will examine dozens of nearby Earth-like worlds. These telescopes represent billions in investment and decades of scientific effort, making the interpretation of their results—even null results—critically important.
The research suggests that absence of evidence can become evidence of absence, given enough observations. Through Bayesian statistical analysis, the researchers found that observing 20-50 terrestrial exoplanets with no signs of habitability would allow scientists to draw significant conclusions about how rare habitable environments might be, regardless of their initial assumptions.
Researchers simulated observational scenarios where scientists survey up to 100 planets. They focused specifically on scenarios with no detections of habitability or life—the “null result” case. Their analysis revealed how many planets would need observation before scientists could make statistically sound inferences about the prevalence of habitable or life-bearing worlds.
If future missions survey 40 Earth-like planets and find no biosignatures, astronomers could conclude with 99.9% confidence that the fraction of terrestrial planets with detectable life is less than 20%. This would reshape our understanding of life’s rarity and potentially point to cosmic “filters” that make the emergence or persistence of life exceptionally uncommon.
For context, while thousands of exoplanets have been cataloged since the first discovery three decades ago, scientists haven’t yet confirmed a truly Earth-like world with conditions suitable for life as we know it. The next generation of space telescopes aims to change that by directly analyzing the atmospheres of nearby rocky planets orbiting in their stars’ habitable zones.
This scientific effort connects to one of humanity’s deepest questions, addressed in the famous Drake Equation, which attempts to estimate the number of communicative civilizations in our galaxy. Two critical terms in this equation are the fraction of habitable planets that develop life and the fraction where intelligence evolves. The missions discussed in this paper may provide the first empirical constraints on these previously speculative values.
Future Missions and Potential Discoveries
Two upcoming missions feature prominently in this research: NASA’s Habitable Worlds Observatory (HWO) and the proposed Large Interferometer for Exoplanets (LIFE). The findings align with the planned capabilities of these future missions. The LIFE telescope concept, in its extended configuration, could potentially characterize around 38 terrestrial planets in the habitable zones of nearby stars. HWO might examine about 20 Earth-like worlds for signs of oxygen or water.
The study confirms these sample sizes, while modest compared to the billions of planets in our galaxy, should be sufficient to provide meaningful statistical constraints on the prevalence of habitable environments and life.
However, these calculations assume scientists can definitively confirm or rule out habitability markers with complete confidence. In practice, interpretation is never so straightforward.
The research reveals that interpretation uncertainty poses a more significant limitation than sample size. If scientists are only 80% confident in each negative observation (a 20% false-negative rate), then even with hundreds of observations, they couldn’t confidently conclude that the frequency of habitable or life-bearing planets is less than 20%.
Life As We Don’t Know It
The findings connect directly to the “Great Filter” hypothesis, which proposes that some developmental stage between simple chemistry and advanced civilization must be extraordinarily difficult, explaining why we haven’t detected alien civilizations despite the apparent abundance of potentially habitable planets.
If future missions survey dozens of suitable planets but find no signs of life, it would suggest that abiogenesis—the emergence of life from non-living matter—might be this filter, an extremely rare event despite suitable conditions. Conversely, if simple life appears common but complex multicellular life or intelligence remains absent, it would indicate the filter operates at a later evolutionary stage.
Such discoveries would have profound implications for humanity’s future. If simple life proves common throughout the galaxy, but technological civilization is unique to Earth, it could suggest that civilization-ending catastrophes are nearly inevitable, painting a sobering picture of our species’ long-term survival chances.
Paper Summary
Methodology
The researchers employed a statistical approach called Bayesian analysis, which allows scientists to update their beliefs about a hypothesis as new evidence emerges. This method is particularly useful when dealing with binary questions like “Does this planet have detectable biosignatures or not?”
The team started with different “prior beliefs” about how common habitable or life-bearing planets might be, ranging from optimistic to pessimistic. They then simulated how these beliefs would change as more and more planets were observed without detecting signs of habitability or life.
The analysis used a mathematical model called the beta-binomial distribution. While that sounds complex, the concept is straightforward: if you observe a series of planets and don’t detect the feature you’re looking for (habitability or biosignatures), how does that change your estimate of how common that feature is across all planets?
The researchers also modeled two types of uncertainties that could affect real observations: sample uncertainty (the possibility that some planets in your survey aren’t representative of the population you’re trying to study) and interpretation uncertainty (the possibility of false negative results due to limitations in detection technology or understanding).
Results
The study found that with approximately 20-50 “perfect” observations (where scientists are 100% confident they can detect or exclude the presence of habitability markers or biosignatures), astronomers could draw statistically meaningful conclusions about the prevalence of these features, regardless of their initial assumptions.
Specifically, about 40 observations would be needed to establish with 99.9% confidence that the fraction of terrestrial planets with the sought-after feature is less than 20%. For a more stringent threshold of less than 10%, approximately 80 observations would be required.
When accounting for uncertainties, the picture changes dramatically. If scientists are only 80% confident in each negative observation (meaning there’s a 20% chance they missed something that was actually there), then even hundreds of observations wouldn’t allow them to conclude with high confidence that the feature occurs in less than 20% of planets.
Importantly, the planned capabilities of future missions like LIFE and HWO align well with these requirements. LIFE could potentially characterize 16-38 Earth-like planets (depending on configuration), while HWO might examine about 17-20 planets for signs of oxygen or water.
Limitations
The research operates under several important assumptions and limitations. First, it assumes that each observation is independent, which might not be true if planets within the same stellar system share evolutionary histories or if mechanisms like panspermia (the hypothesis that life can transfer between planets via meteorites) operate across nearby systems.
The study also simplifies the observational process to a binary outcome (detection or non-detection), whereas real observations will likely involve complex spectroscopic data with various levels of confidence and multiple potential interpretations.
Additionally, the analysis doesn’t account for the possibility that life might take forms radically different from Earth life, with biosignatures unlike anything scientists currently know to look for. The focus on Earth-like conditions might miss evidence of exotic biochemistries.
Finally, the mathematical models assume that either all planets have the feature or none do, when in reality, the likelihood of habitability or life developing might vary continuously based on many planetary and stellar factors.
Funding and Disclosures
This research was carried out within the framework of the National Centre of Competence in Research PlanetS supported by the Swiss National Science Foundation under grant 51NF40_205606. Several authors received additional funding from institutions including the Swiss National Science Foundation, the University of Belgrade—Faculty of Mathematics through grants by the Ministry of Science, Technological Development, and Innovation of the Republic of Serbia.
Publication Information
The paper “What if we Find Nothing? Bayesian Analysis of the Statistical Information of Null Results in Future Exoplanet Habitability and Biosignature Surveys” was published in The Astronomical Journal (Volume 169, Number 238) on April 7, 2025. The research was conducted by Daniel Angerhausen, Amedeo Balbi, Andjelka B. Kovačević, Emily O. Garvin, and Sascha P. Quanz, representing institutions including ETH Zurich, Blue Marble Space Institute of Science, SETI Institute, Università di Roma “Tor Vergata,” and the University of Belgrade.