Current estimates suggest the Milky Way contains somewhere between several billion and tens of billions of rocky, roughly Earth-sized planets orbiting in the habitable zones of their stars. The exact number depends heavily on how you define “Earth-like” and which star types you include, and the spread among published estimates is wide enough that honest researchers will tell you we still do not know the answer to better than an order of magnitude. The most careful analyses of NASA’s Kepler mission data place the frequency of habitable-zone rocky planets around Sun-like stars at roughly 0.37 to 0.60 per star for conservative definitions, climbing toward one per star under more generous assumptions. Since the galaxy holds on the order of 100 to 400 billion stars, even conservative estimates yield staggering numbers.
What Astronomers Mean by “Earth-Like”
The phrase “Earth-like planet” does a lot of heavy lifting in popular science, and the vagueness causes real confusion. In the research literature, the standard measure is called eta-Earth, defined as the fraction of stars that host at least one rocky planet in the habitable zone. But within that seemingly straightforward definition, every term is contested. “Rocky” usually means a planet with a radius between about 0.5 and 1.5 times Earth’s, though some studies stretch the upper boundary to 2 Earth radii, which dramatically changes the count. A planet twice Earth’s size could be a scaled-up rocky world or a mini-Neptune wrapped in a thick gaseous envelope, and we often cannot tell the difference from transit data alone.
The habitable zone itself is not a fixed ring. It depends on the star’s temperature and brightness, but also on atmospheric assumptions for the planet. Under conservative definitions, the habitable zone for our Sun runs from about 0.99 to 1.70 astronomical units, placing Earth right near the inner edge.1The Astrophysical Journal. HABITABLE ZONES AROUND MAIN-SEQUENCE STARS: NEW ESTIMATES Optimistic boundaries push both edges wider, allowing Venus-like insolation at the inner edge and early-Mars conditions at the outer edge.2PubMed Central. Remote life-detection criteria, habitable zone boundaries, and the frequency of Earth-like planets around M and late K stars The choice between conservative and optimistic boundaries alone can double the estimated number of habitable-zone planets.
What the Kepler Data Actually Show
Nearly everything we know about the frequency of small planets comes from the Kepler space telescope, which spent four years staring at a patch of roughly 150,000 stars, watching for the tiny dips in brightness caused by a planet crossing in front of its star. Kepler was not designed to give a perfect census. It could only detect planets whose orbits happened to be aligned edge-on from our perspective, and it was less sensitive to small, long-period planets in the habitable zones of Sun-like stars. Every published estimate of eta-Earth therefore involves substantial statistical corrections to account for the planets Kepler missed.
An influential 2013 analysis of about 42,000 Sun-like stars in the Kepler field found that roughly 11 percent of them harbor an Earth-sized planet receiving between one and four times Earth’s stellar energy. That translates to about one in nine Sun-like stars hosting a potentially rocky world in or near the habitable zone.3PubMed Central. Prevalence of Earth-size planets orbiting Sun-like stars More recent work using the final Kepler data release and improved stellar properties from the Gaia spacecraft found higher numbers. For planets between 0.5 and 1.5 Earth radii orbiting stars with temperatures between 4,800 and 6,300 Kelvin, the conservative habitable-zone occurrence rate came in between about 0.37 and 0.60 planets per star, with the optimistic habitable zone pushing that to 0.58 to 0.88 per star.4The Astronomical Journal. The Occurrence of Rocky Habitable Zone Planets Around Solar-Like Stars from Kepler Data
Not everyone agrees those numbers are right. A separate search of the full Kepler dataset, focusing specifically on planets between 0.75 and 1.5 Earth radii in the conservative habitable zone around G-type stars like the Sun, could only place an upper limit of fewer than 0.18 planets per star.5The Astronomical Journal. Searching the Entirety of Kepler Data. II. Occurrence Rate Estimates for FGK Stars That is roughly three to four times lower than the more generous estimates, and the gap has not been fully resolved. The difference comes down to how researchers handle the completeness corrections, which planet candidates they trust, and where they draw the boundaries of the habitable zone. One systematic comparison found that eta-Earth estimates can vary by a factor of 4.5 simply by using different thresholds for vetting planet candidates from the same underlying data.6Publications of the Astronomical Society of the Pacific. Are We There Yet? Challenges in Quantifying the Frequency of Earth Analogs in the Habitable Zone
Why the Estimates Disagree So Much
The transit method is inherently biased. A planet has to pass directly between us and its star to be detected, and the geometric probability of that alignment drops sharply for planets on wider orbits. For an Earth twin orbiting a Sun-like star at 1 AU, the chance of a transit is less than half a percent.7Monthly Notices of the Royal Astronomical Society. Observational biases for transiting planets That means researchers are extrapolating from a handful of detections to the whole population, and small changes in assumptions ripple through to the final count.
Even when a transit occurs, detecting an Earth-sized planet around a Sun-like star pushes Kepler’s sensitivity to its limit. Transit-injection experiments, where synthetic signals are buried in real Kepler light curves and then searched for, show that the recovery rate for Earth-sized planets varies depending on the search algorithm used, ranging from about 76 percent for older methods to around 93 percent for optimized approaches.8Astronomy & Astrophysics. Optimized transit detection algorithm to search for periodic transits of small planets Those differences in assumed completeness feed directly into the final occurrence rates. On top of that, the stellar properties used to calculate planet sizes have been revised significantly since Kepler launched, thanks to more precise distance measurements from the Gaia mission. Older stellar catalogues led to eta-Earth values nearly twice as high as those calculated using updated stellar data.6Publications of the Astronomical Society of the Pacific. Are We There Yet? Challenges in Quantifying the Frequency of Earth Analogs in the Habitable Zone
Red Dwarf Stars Change the Picture
Most of the headline estimates of eta-Earth focus on Sun-like stars, which are the types easiest to compare to our own system. But Sun-like stars make up a small fraction of the galaxy. Red dwarfs, the cool, dim stars classified as M-dwarfs, account for roughly 70 percent of all stars in the Milky Way. If rocky habitable-zone planets are common around red dwarfs, the total number of potentially habitable worlds shoots up dramatically.
The evidence suggests they are indeed common. A reanalysis of Kepler data for M-dwarf stars found about 0.48 to 0.53 terrestrial planets per star in the conservative habitable zone, depending on the size range used.9The Astrophysical Journal Letters. A revised estimate of the occurrence rate of terrestrial planets in the habitable zones around kepler m-dwarfs That is roughly one habitable-zone rocky planet for every two red dwarfs, give or take. Since there may be 200 billion or more red dwarfs in the galaxy, the arithmetic produces tens of billions of candidates from those stars alone.
The catch is that “habitable zone” around a red dwarf means something quite different from the habitable zone around the Sun. Because red dwarfs are so faint, the zone sits much closer to the star. Planets there are likely tidally locked, permanently showing one face to their star. They are also exposed to intense stellar flares and energetic particle events that could strip their atmospheres or drive dramatic chemical changes.10The Astronomical Journal. Effects of Transient Stellar Emissions on Planetary Climates of Tidally Locked Exo-Earths Whether these worlds can maintain conditions suitable for life remains genuinely uncertain.
Being the Right Size Is Not Enough
A planet can be the right size and sit at the right distance from its star and still be nothing like Earth. Several additional factors determine whether a rocky planet could actually support liquid water on its surface for geological timescales.
One of the most important is the carbon cycle. On Earth, the long-term thermostat depends on carbon dioxide cycling between the atmosphere and the interior through volcanic outgassing and the chemical weathering of rocks. This process requires tectonically active geology, ideally something resembling plate tectonics, to supply fresh rock to the surface and to recycle carbon back into the mantle through subduction.11The Astrophysical Journal. The Role of Plate Tectonic–Climate Coupling and Exposed Land Area in the Development of Habitable Climates on Rocky Planets Without it, a planet’s climate can run away in either direction, freezing over permanently or cooking under a runaway greenhouse. We have no reliable way to detect plate tectonics on exoplanets, so this remains a major unknown.12PubMed. Long-Term Planetary Habitability and the Carbonate-Silicate Cycle
Water content matters too, but in a counterintuitive way. Too little water and you get a desert world with no oceans. Too much and you may end up with a waterworld where the ocean is so deep that high-pressure ice forms at the bottom, sealing off the rocky interior and potentially shutting down the geochemical cycles that regulate climate. Modeling suggests that planets can follow divergent evolutionary paths depending on their early water inventory, ending up as land-dominated, ocean-dominated, or balanced worlds like Earth.13PubMed. Land Fraction Diversity on Earth-like Planets and Implications for Their Habitability Several known habitable-zone exoplanets, including Kepler-62f and Kepler-452b, may contain so much water that they qualify as ocean worlds rather than Earth analogues.14Astronomy & Astrophysics. Water content of rocky exoplanets in the habitable zone
The architecture of the planetary system also plays a role. Simulations show that the orbit of a Jupiter-like planet in the outer system affects how much water gets delivered to the inner planets during formation. An eccentric giant planet tends to produce drier terrestrial planets with more elongated orbits.15Icarus. Making other earths: dynamical simulations of terrestrial planet formation and water delivery On the other hand, Jupiter’s growth and migration in our own solar system may have transported tens of times Earth’s current ocean mass into the inner system.16The Planetary Science Journal. Early Water Delivery to Terrestrial Planet Regions during the Stages of Jupiter’s Formation and Migration in the Grand Tack Model Whether other systems have the right kind of giant planet architecture to enable habitable terrestrial worlds is largely unknown.
The Rocky-to-Gaseous Boundary
One underappreciated complication is that many planets counted as “rocky” in occurrence rate studies may not be rocky at all. The observed distribution of exoplanet radii shows a gap, often called the radius valley, separating smaller super-Earths from larger sub-Neptunes. The valley sits at roughly 1.5 to 2 Earth radii, and its origin tells us something important about what these planets are made of.
Formation models suggest the valley marks the divide between genuinely rocky planets that formed close to their star and water-rich planets that migrated inward from beyond the snow line, where volatile ices are abundant. The smaller planets below the valley are rocky cores that lost any primordial gaseous envelope to photoevaporation, while the larger ones above it are essentially steam worlds with thick water-vapor atmospheres mixed with hydrogen and helium.17PubMed Central. A radius valley between migrated steam worlds and evaporated rocky cores The location of this valley shifts depending on the mass of the host star, moving to smaller radii around lower-mass stars.18Astronomy & Astrophysics. A fading radius valley towards M dwarfs, a persistent density valley across stellar types This means that a 1.5-Earth-radius planet orbiting a Sun-like star may be rocky, while the same-sized planet around a red dwarf is more likely to be volatile-rich. Occurrence rate estimates that use a fixed radius cutoff across all star types are blending genuinely rocky worlds with gas-enveloped ones.
TRAPPIST-1 and What JWST Can Tell Us
The TRAPPIST-1 system, with seven Earth-sized planets orbiting a nearby red dwarf, has become the primary laboratory for testing whether small planets around cool stars can hold atmospheres. Three of those planets sit in or near the habitable zone. Early Hubble observations ruled out cloud-free hydrogen-rich atmospheres for several of the planets, which was encouraging because it suggested they might be rocky rather than mini-Neptunes.19PubMed Central. A Review of Possible Planetary Atmospheres in the TRAPPIST-1 System
The James Webb Space Telescope was expected to push this further, and pre-launch simulations suggested that JWST could detect carbon-dioxide-rich atmospheres on all seven TRAPPIST-1 planets in fewer than ten transits each, assuming the atmospheres lack thick high-altitude hazes.20The Astronomical Journal. The Detectability and Characterization of the TRAPPIST-1 Exoplanet Atmospheres with JWST Early JWST results for the innermost planets have been discouraging, showing little to no evidence of substantial atmospheres on TRAPPIST-1 b and c. The habitable-zone planets remain to be thoroughly characterized, but the emerging picture raises the possibility that close-in planets around active red dwarfs simply cannot hold onto their air. If confirmed, that would significantly reduce the number of genuinely habitable worlds in the galaxy, since so many of the statistical candidates orbit red dwarfs.
Detecting biosignature gases like ozone would require considerably more observing time. Simulations estimated that ozone at Earth-like levels could be identified on TRAPPIST-1d, the most promising candidate, after roughly 30 transits each with two different JWST instruments.21Monthly Notices of the Royal Astronomical Society: Letters. Habitable worlds with JWST: transit spectroscopy of the TRAPPIST-1 system? That is a major investment of telescope time, and it only applies to the most favorable known system. For the typical habitable-zone planet, JWST simply does not have the capability to detect biosignatures.
How Long Does Habitability Last
Even if a planet starts out habitable, it does not necessarily stay that way. Stars brighten as they age, and the habitable zone migrates outward over time. Earth has been in the Sun’s habitable zone for roughly four billion years, but it will exit the inner edge in perhaps another billion years as the Sun continues to brighten. The concept of a “habitable zone lifetime,” the total time a planet spends within the habitable zone, sets an upper limit on how long any biosphere could persist.22PubMed. Habitable zone lifetimes of exoplanets around main sequence stars
This temporal dimension matters when thinking about the number of Earth-like planets. A planet that spends only a few hundred million years in the habitable zone might develop microbial life but would be unlikely to evolve anything complex. If you are asking “how many planets could support complex life right now,” the number shrinks considerably compared to “how many planets could support any life at some point.” Lower-mass stars evolve more slowly and have longer-lived habitable zones, which is another point in favor of red dwarfs, if their planets can survive the flare activity.
Where in the Galaxy to Look
Not all regions of the Milky Way are equally hospitable. The concept of a galactic habitable zone narrows the search to a ring-shaped region where the chemistry is right for building rocky planets and the local stellar environment is not too violent. Stars need enough heavy elements, what astronomers call metals, to form rocky planets in the first place. The inner galaxy is metal-rich but crowded with supernovae that could periodically sterilize nearby planets. The outer galaxy is quieter but metal-poor, making rocky planet formation less likely.23Astronomy & Astrophysics. Galactic habitable zone around M and FGK stars with chemical evolution models that include dust Our Sun sits comfortably within this zone, roughly two-thirds of the way out from the galactic center.
Future Missions and the Search for Oxygen
JWST can characterize atmospheres of transiting planets, but it cannot image Earth-like planets directly. That requires a different kind of mission: a large space telescope equipped with a coronagraph or starshade to block the host star’s light and see the planet as a separate dot. Two proposed mission concepts, LUVOIR and HabEx, were designed specifically for this task. Simulations of these missions explored how well they could constrain the fraction of habitable-zone rocky planets that actually have Earth-like atmospheres with oxygen or ozone. A 15-meter LUVOIR-class telescope, if it found no oxygen on any observed candidates, could rule out more than about 9 percent of habitable-zone planets being truly Earth-like, assuming roughly one in four Sun-like stars has a rocky habitable-zone planet. A smaller 4-meter HabEx with a starshade could constrain that fraction to about 56 percent.24The Astronomical Journal. Probing the Capability of Future Direct-imaging Missions to Spectrally Constrain the Frequency of Earth-like Planets Neither mission has been approved for construction, but NASA’s Habitable Worlds Observatory concept inherits much of this design heritage.
The point is that we are still at least a couple of decades away from being able to directly survey even nearby stars for biosignature gases. The current number of confirmed planets that could plausibly be called Earth-like, rocky, temperate, and potentially habitable, can be counted on two hands. Everything beyond that is statistical inference from incomplete data.
Habitable Environments Beyond the Habitable Zone
The traditional habitable zone assumes a planet needs direct starlight to maintain surface liquid water. But our own solar system offers examples of potentially habitable environments far outside that zone. Europa and Enceladus almost certainly host subsurface oceans kept liquid by tidal heating from their giant planet hosts. Tidal forces from a nearby massive planet can pump enough energy into a moon’s interior to maintain liquid water indefinitely, even in the frozen outer reaches of a stellar system.25Monthly Notices of the Royal Astronomical Society. Tidal heating of terrestrial extrasolar planets and implications for their habitability
Modeling of small, icy exomoons suggests that subsurface habitable environments could exist essentially anywhere beyond the snow line of a planetary system, largely independent of the distance to the host star.26Astronomy & Astrophysics. The subsurface habitability of small, icy exomoons If subsurface oceans count, the number of potentially habitable environments in the galaxy grows enormously, but with the caveat that we have no idea how to detect life trapped under kilometers of ice on a world orbiting a distant star. Even more speculatively, when a Sun-like star eventually becomes a red giant, its luminosity increases by thousands of times, temporarily making the outer solar system balmy enough for surface liquid water on worlds that spent billions of years frozen. Objects rich in water and organic compounds, like Kuiper Belt bodies, could briefly become sites for prebiotic chemistry during that late stellar phase.27PubMed. Delayed gratification habitable zones: when deep outer solar system regions become balmy during post-main sequence stellar evolution These scenarios are fun to think about but essentially untestable with current or near-future technology.