Exoplanet Detection: The Five Methods That Have Revealed 5,500+ Worlds — And Their Limits

In 1992, the first confirmed exoplanets were announced — and they turned out to orbit a pulsar, a spinning neutron star, rather than anything resembling a Sun-like star. The discovery was simultaneously remarkable and deeply strange: worlds irradiated by pulsars seemed to offer no prospect of life and hinted at exotic formation mechanisms. The first planet confirmed around a Sun-like star, 51 Pegasi b, came in 1995, and it was also strange — a "hot Jupiter," a gas giant half the mass of Jupiter orbiting its star every four days at a distance of 0.05 astronomical units, far inside where Mercury sits in our own solar system.

Three decades later, the confirmed exoplanet catalog has passed 5,500 entries and is growing at roughly 500 new worlds per year. The methods used to detect these planets are elegantly varied, each with its own physical basis, its own detection biases, and its own range of what it can and cannot see. Understanding these biases is essential to interpreting what the catalog actually tells us about the distribution of planets in the galaxy — and what it is almost certainly missing.

## Transit Photometry: The Kepler Revolution

The transit method detects planets by monitoring the brightness of a star and looking for periodic dips caused by a planet passing across the stellar disk. When a planet transits its star, it blocks a fraction of the star's light equal to the ratio of the planet's cross-sectional area to the star's: ΔF/F = (R_planet/R_star)². For an Earth-size planet transiting a Sun-size star, this produces a brightness decrease of roughly 84 parts per million — a signal that requires space-based photometry to detect reliably.

The Kepler Space Telescope, launched in 2009, stared continuously at a fixed field of approximately 150,000 stars in the constellations Cygnus and Lyra for four years before its reaction wheels failed, then continued in a modified "K2" mission until 2018. Its yield was extraordinary: over 2,600 confirmed planets and more than 2,000 planet candidates from a single field of view. TESS (Transiting Exoplanet Survey Satellite), launched in 2018, surveyed the entire sky, focusing on the nearest and brightest stars most amenable to follow-up study.

The transit method's biases are systematic and well-understood. It preferentially detects large planets (because larger planets block more light) and planets in short orbits (because they transit more frequently, increasing the probability of being in a favorable geometric alignment and providing multiple transits within the observing window). A Jupiter orbiting at 1 AU has only a 0.5% probability of transiting its star from any random viewing angle; for a planet at 0.05 AU, the probability rises to roughly 10%. The transit method's catalog is therefore dominated by short-period planets — hot Jupiters, hot Neptunes, and super-Earths in tight orbits — and represents the distribution of planets in close-in orbits far better than it represents planets at habitable-zone distances from Sun-like stars.

## Radial Velocity: Precision Doppler Spectroscopy

The radial velocity method exploits Newton's third law. A planet does not orbit a stationary star; both the planet and the star orbit their common center of mass. As the star moves in this reflex orbit, its spectral lines shift slightly toward shorter wavelengths (blueshift) when it moves toward Earth and toward longer wavelengths (redshift) when it moves away. By measuring these Doppler shifts in high-resolution stellar spectra with extraordinary precision, astronomers can detect the gravitational tug of orbiting planets.

The HARPS spectrograph at the European Southern Observatory achieved velocity precision of approximately 1 meter per second in the 2000s — sufficient to detect Earth-mass planets in close-in orbits around quiet stars. The ESPRESSO spectrograph, also at ESO, has pushed this to approximately 10 centimeters per second precision — approaching the threshold needed to detect Earth-mass planets in habitable-zone orbits around Sun-like stars (which require better than 10 cm/s sensitivity). This remains technically challenging because stellar surface activity — convection, pulsations, sunspot activity — introduces intrinsic stellar radial velocity noise at the few meters-per-second level, comparable to or larger than the planet signal.

Radial velocity measurements yield the planet's orbital period and the minimum mass (M_planet × sin i, where i is the orbital inclination, which is unknown without a transit detection). The method is sensitive to planets with high masses and short periods — similar biases to the transit method, though less extreme, since it does not require a favorable viewing geometry.

## Direct Imaging: Seeing the Light

The most conceptually straightforward method — take a picture of the planet — is also the most technically demanding. The challenge is contrast: a Jupiter-size planet reflects about a billion times less light than its host star at visible wavelengths, and is separated from it by an angular distance that, for a planet at 5 AU around a star 10 parsecs away, is only 0.5 arcseconds. Blocking the overwhelming starlight to reveal the planet beside it requires either a coronagraph (a mask that suppresses the stellar diffraction pattern) or extreme adaptive optics to achieve near-diffraction-limited resolution that separates the stellar point spread function from the planetary signal.

Current direct imaging instruments — the Gemini Planet Imager, SPHERE at the VLT, SCExAO at Subaru — have successfully imaged a handful of directly detected planets, most of them young, recently formed giant planets at large orbital separations (tens to hundreds of AU) from their host stars. These planets are self-luminous: they are still radiating the heat of their formation and are detectable in the near-infrared. Mature, cold planets at Earth-like orbital separations around Sun-like stars remain beyond current ground-based capabilities.

The Nancy Grace Roman Space Telescope, scheduled for launch in 2027, will carry a coronagraphic instrument designed to demonstrate direct imaging of mature Jupiter-like planets at several AU from nearby stars. The proposed Habitable Worlds Observatory (a future large flagship mission) aims to directly image and obtain spectra of Earth-like planets around nearby stars — a scientific goal that will require coronagraph performance roughly 100 times better than current instruments.

## Gravitational Microlensing

When two stars happen to align along the line of sight, the gravitational field of the foreground star acts as a lens, bending and amplifying the light from the background star in a characteristic light curve. If the foreground star has a planet, the planet's gravitational contribution produces a brief additional spike in the light curve — lasting hours to days — riding on top of the broader microlensing event that lasts weeks to months.

Microlensing is sensitive to planets at orbital separations of 1-10 AU around stellar lenses at distances of several kiloparsecs — the region of the galaxy between Earth and the Galactic bulge. It is the only method sensitive to cold planets at large orbital separations and to free-floating planets that are not bound to any star. Statistical analyses of microlensing survey data suggest that free-floating Jupiter-mass objects may be common in the galaxy — potentially outnumbering stars — though the interpretation of this result remains contested.

The critical limitation of microlensing is that each event is unique and unrepeatable: the particular alignment of lensing star, planet, and background star will never recur, and no follow-up observations of the detected planet system are possible. The Roman Space Telescope will conduct a dedicated microlensing survey toward the Galactic bulge that is expected to detect thousands of planets and substantially improve the census of planets in the cool outer regions of stellar systems.

## JWST Atmospheric Characterization and the TRAPPIST-1 Frontier

Perhaps the most exciting capability to emerge from recent years is not detection but characterization: using the James Webb Space Telescope's infrared sensitivity to analyze the atmospheres of transiting planets during and after transits.

When a planet transits, starlight filters through the planet's atmosphere, where molecules absorb specific wavelengths. By comparing the apparent size of the planet at different wavelengths — transmission spectroscopy — the composition of the atmosphere can be inferred. JWST has already detected CO2 absorption in the atmosphere of the hot gas giant WASP-39b, detected dimethyl sulfide candidate absorption around K2-18b (though interpretation remains contested), and begun systematic study of rocky planet atmospheres.

The TRAPPIST-1 system — seven Earth-size planets orbiting an ultracool M-dwarf star 12 light-years away, three in the habitable zone — represents the current frontier. TRAPPIST-1b and 1c have been studied with JWST, with results suggesting TRAPPIST-1c lacks a thick Venus-like CO2 atmosphere. The inner planets appear airless or with only thin atmospheres, though the definitive characterization of TRAPPIST-1e, f, and g — the potentially habitable outer planets — will require many more JWST transit and eclipse observations.

All TRAPPIST-1 planets are tidally locked: one face permanently toward the star, one permanently away. They also receive intense ultraviolet and X-ray irradiation from stellar flares that may strip atmospheres over geological timescales. Whether any maintain atmospheres dense enough to moderate surface temperatures into a habitable range remains an open empirical question — and JWST is uniquely positioned to begin answering it.

The combined picture painted by five decades of exoplanet detection is one of extraordinary diversity: super-Earths and mini-Neptunes (with no solar system analog) are apparently the most common planet types in the galaxy. Hot Jupiters, rare in the solar neighborhood, exist in measurable numbers. Multiplanet systems in closely packed orbital configurations appear to be common around M-dwarf stars. What the galaxy has in uncommon abundance, at least in the census so far, are planets quite like Earth — at habitable-zone distances from Sun-like stars. Whether that rarity is real or an artifact of detection bias remains the organizing scientific question of the field.

Exoplanet Detection: The Five Methods That Have Revealed 5,500+ Worlds — And Their Limits

// COMMENTS

ON THIS PAGE