Exoplanet in transit in front of its star, revealing its atmosphere through the transmission spectrum where absorption lines of H₂O, CO₂, CH₄, Na, and K appear.
Image source: astronoo.com
This article presents the methods for detecting exoplanets in the Milky Way. Five main techniques are described: the radial velocity method (Doppler), transit, direct imaging, gravitational microlensing, and astrometry. Each has its own limitations: radial velocities favor massive, nearby planets; transits require perfect alignment; direct imaging is reserved for young, distant planets; microlensing is a one-time event; and astrometry requires extreme precision. The article also highlights the importance of protoplanetary disks in the formation of gas giants and multiple planetary systems.
The question of whether other habitable worlds exist is one of the driving forces of modern astronomy. But how can we detect an exoplanet, located light-years away, when it emits no light and its star blinds us? Astronomers have developed several clever techniques to meet this challenge. The radial velocity method (or Doppler) measures the oscillations of a star caused by the gravity of a planet: as the star and planet orbit their common center of mass, the star's spectrum shifts red or blue depending on whether it is moving away from or toward us. This technique is particularly effective for massive and close-in planets, such as "Hot Jupiters." The transit method detects the drop in a star's brightness when a planet passes in front of it, allowing its size to be estimated and its atmosphere to be studied via spectroscopy, but it requires a perfect alignment between the star, planet, and observer. Direct imaging, although very difficult, uses coronagraphs to block the star's light and directly photograph the planet, which is mostly feasible for young, massive, and distant planets. Gravitational microlensing exploits the bending of light from a background star by a planet, but this event is unique and non-reproducible. Finally, astrometry measures the movements of a star in the plane of the sky, a complementary technique that requires extreme precision. These methods, often combined, have allowed the discovery of thousands of exoplanets, revealing the diversity of planetary systems and the complexity of planetary formation from protoplanetary disks.
“Are we alone in the Universe?” This question has long been on the minds of astronomers. Answering it requires knowing whether other planets exist where other forms of life could develop, hence the search for exoplanets.
Exoplanets, by definition, are located outside our solar system. With several hundred billion stars in the Milky Way alone, the search for these extrasolar planets is far from over, even though it remains confined to our own galaxy: with the exception of a few punctual detections by gravitational microlensing in nearby galaxies, almost all known exoplanets orbit stars in the Milky Way, as intergalactic distances make their individual detection beyond the reach of current instruments.
Detecting such planets is not easy. They are very far from us, emit no light of their own, and the brightness of the star they orbit literally drowns them out. The Hubble Space Telescope can only observe Pluto, a member of our own solar system, as a barely resolved point. How, then, can we hope to distinguish an extrasolar planet orbiting a star sometimes thousands of times farther away than Pluto?
Earth orbits the Sun, but in reality, if a star is accompanied by a single planet, both orbit around the center of mass of the gravitational system they form. Given the star's much greater mass, this center of mass is much closer to the star than to the planet. It is even most often inside the star, without, however, coinciding exactly with its geometric center.
The star thus undergoes a slight periodic displacement caused by the presence of the planet. This is precisely what the radial velocity method reveals: it consists of detecting these variations in the spectrum of the light emitted by the star, via the shift of its spectral lines.
This spectral shift is explained by the Doppler-Fizeau effect: the star's light appears slightly redshifted when it moves away from the observer and blueshifted when it approaches. As the star is alternately pulled in one direction and then the other by the orbiting planet, its spectrum periodically oscillates between a slight excess of red and a slight excess of blue, at the rhythm of the planet's orbital period.
This alternation, although tiny—on the order of a few meters per second for an Earth-sized planet—can be measured with very high-precision spectrographs like HARPS or ESPRESSO. It enabled the very first detection of an exoplanet around a solar-type star, 51 Pegasi b, in 1995, and remains today one of the reference methods for confirming and characterizing worlds discovered by other techniques.
Since the discovery of the first exoplanet around a solar-type star in 1995 (51 Pegasi b), more than 8,000 worlds have been confirmed beyond the Solar System (as of February 2026). Five major methods, often complementary, allow astronomers to detect and characterize them.
Limitation: Sensitivity decreases with distance from the star and the planet's mass; the method thus favors the detection of massive planets on short orbits ("Hot Jupiters").
Limitation: Only works if the planet's orbital plane is almost aligned with the line of sight; only a fraction of planetary systems are favorably oriented.
Limitation: Reserved for young, massive, hot planets widely separated from their star; in 2025, JWST made its first unprecedented discovery by coronagraphy, a low-mass planet in the debris disk surrounding the young star TWA 7.
Limitation: Transient and non-reproducible event, which never repeats for the same system; in July 2026, the TESS telescope—designed for transits—made its first microlensing detection (Gaia23bra b) by re-examining its archives following a Gaia alert, proving that the method can also be applied a posteriori to data collected for another purpose.
Limitation: Requires extreme measurement precision, on the order of microarcseconds; the ESA's Gaia mission (2013-2025), whose DR4 catalog is expected in 2026, should provide the first major astrometric survey of exoplanets.
These methods, often used in combination, have enabled the discovery of more than 8,000 confirmed exoplanets and reveal a diversity of planetary systems much greater than what the sole example of the Solar System suggested.
The radial velocity method, which enabled the very first detection of an exoplanet around a solar-type star in 1995 (51 Pegasi b), consists of detecting the periodic motion of a star around the center of mass it shares with its planet. It is not the star itself that "oscillates" in the proper sense, but its position varies slightly under the effect of the gravitational attraction of the orbiting body, a displacement revealed by the Doppler shift of its spectrum.
This method does not "miss" small rocky planets by chance: it suffers from an intrinsic detection bias that makes it much more sensitive to gas giants in close orbits than to terrestrial planets.
The amount of dust and gas available in the protoplanetary disk surrounding a young star largely determines the size and number of planets that will form there. According to the core accretion model, gas giants form preferentially beyond the snow line, where the condensation of volatile compounds provides enough solid material to rapidly assemble a massive core, capable of later capturing a gas envelope before the disk dissipates.
When the protoplanetary disk is particularly rich in gas and dust, several giants can form successively, giving rise to multi-giant systems like that of HR 8799, which has four giant planets.
NASA Exoplanet Archive – California Institute of Technology / IPAC.
The Extrasolar Planets Encyclopaedia – Observatoire de Paris.
ESA – Gaia Mission.
NASA – TESS (Transiting Exoplanet Survey Satellite).
NASA/ESA/CSA – James Webb Space Telescope.
Mayor & Queloz (1995), Nature – A Jupiter-mass companion to a solar-type star.
Charbonneau et al. (2000), ApJ – Detection of Planetary Transits Across a Sun-like Star.
Marois et al. (2008), Science – Direct Imaging of Multiple Planets Orbiting HR 8799.
Lagrange et al. (2025), Nature – Evidence for a sub-Jovian planet in the young TWA 7 disk.
Gould & Loeb (1992), ApJ – Discovering planetary systems through gravitational microlenses.
Harris et al. (2026), ApJL – TESS's First Bound Microlensing Planet (Gaia23bra b).
Gaia Collaboration, Prusti et al. (2016), A&A – The Gaia mission.
The transit method is currently the most productive: it has enabled the discovery of more than 70% of known exoplanets, notably thanks to NASA's Kepler mission. It consists of measuring the drop in brightness of a star when a planet passes in front of it. This drop, even tiny (less than 1%), is detectable with great precision. The transit not only confirms the planet's existence but also estimates its size (from the transit depth) and, combined with the radial velocity method, determines its mass and thus its density. Its main limitation is that it requires a specific orbital alignment, meaning only a fraction of planetary systems are detectable by this method.
Hot Jupiters are planets with the mass of Jupiter (about 300 times the mass of Earth) that orbit at a very close distance from their star (less than 0.1 AU, or about one-tenth the Earth-Sun distance). Their high mass produces significant stellar oscillations, easily measurable by the radial velocity method (significant Doppler shift). Their close orbit also increases the probability of transit and the frequency of transits, making them detectable by the transit method. Finally, since they are close to their star, the microlensing effect is more pronounced. These planets were thus the first to be discovered (e.g., 51 Pegasi b in 1995) and remain privileged targets for atmospheric studies.
A protoplanetary disk is a ring of gas and dust that surrounds a young star. It is the birthplace of planets. The mass of the disk determines the size and number of planets that will form: a massive disk can give rise to gas giants (like Jupiter) in its outer regions, while a more modest disk forms terrestrial planets (like Earth) in the inner regions. If the disk is very rich, several gas giants can form, creating multiple planetary systems (like the 55 Cancri system). These giants then influence the system's architecture: their gravitational interactions can eject planets or modify their orbits. The article emphasizes that the amount of dust and gas in the protosolar nebula directly determines the size and composition of the planets that will form.
Each exoplanet detection method has observational biases: radial velocity is sensitive to massive, nearby planets but only provides the minimum mass (as the orbit's inclination is unknown); transit requires perfect alignment (low probability, especially for wide orbits) and only provides the radius; direct imaging is limited to young, distant planets (as they are still hot and bright); microlensing is a one-time event (non-reproducible) and does not allow re-observation of the planet; astrometry requires extreme instrumental stability, difficult to achieve from the ground. This is why a combination of several methods is often necessary to fully characterize a planetary system (mass, radius, density, atmospheric composition).
The radial velocity method, also called the Doppler method, relies on measuring the spectral shift of a star's light caused by the orbital motion induced by a planet. When a planet orbits a star, both bodies orbit around their common center of mass. The star, although massive, is slightly displaced by the planet's gravity. This motion produces a Doppler effect: if the star is moving toward Earth, its spectrum is blueshifted (shorter wavelengths); if it is moving away, it is redshifted (longer wavelengths). By measuring these periodic variations in the star's radial velocity (up to a few meters per second), we can deduce the planet's orbital period and minimum mass. This method enabled the discovery of the first exoplanet around a solar-type star (51 Pegasi b) in 1995 by Michel Mayor and Didier Queloz, winners of the 2019 Nobel Prize in Physics.