Posted by The portion of the universe explored well enough to say anything concrete about its planets is small, and every bit of it has been won by instruments rather than by argument. Thirty years ago the count of confirmed worlds outside the Solar System was zero. It is now in the thousands, and for a growing handful astronomers can quote a radius, a mass, a density and a partial list of the gases in the atmosphere. That progression came from a set of techniques worth understanding, because each one has a blind spot and none of them works alone.
Very few of those planets have ever been seen. Almost all were inferred.
Nobody Actually Sees the Planet
An exoplanet is small, dark and sitting next to something billions of times brighter. Direct detection is the exception. The workhorse method is the transit: if the orbit happens to be edge-on from our point of view, the planet crosses the face of its star once per orbit and blocks a fraction of its light.
For an Earth-sized planet crossing a Sun-like star that fraction is about 0.008 percent — eighty parts per million. Kepler was built to catch exactly that, and from 2009 it monitored roughly 150,000 stars in a single field spanning Cygnus and Lyra, waiting for repeated dips. TESS, launched in 2018, does the opposite: it sweeps almost the whole sky in sectors, trading depth for coverage, hunting bright nearby stars whose planets can be followed up by other instruments.
The geometry is the limitation. Transits only work for the small fraction of systems aligned edge-on, so every transit survey is sampling a biased slice and everybody knows it.
The Star Gives Itself Away
A planet and its star both orbit their common centre of mass, so the star traces a small circle and its light shifts blue as it swings toward us, red as it swings away. Measure that shift precisely enough and you get the planet’s orbital period and a minimum mass.
This is how the field started. Michel Mayor and Didier Queloz used the ELODIE spectrograph at Haute-Provence to find 51 Pegasi b in 1995 — a Jupiter-mass planet on a four-day orbit, which nobody’s planet-formation theory had predicted. HARPS at La Silla pushed the precision down to about a metre per second; ESPRESSO at the Very Large Telescope is chasing tens of centimetres per second.
For scale: Earth makes the Sun wobble by roughly nine centimetres per second. Detecting a true Earth analogue around a Sun-like star is right at the edge of what the best spectrographs can do, and stellar surface activity — spots, granulation, magnetic cycles — produces signals of the same size that are not planets at all.
Two Measurements Make a Density
Transits give a radius. Radial velocity gives a mass. Put them together and you get a density, which is the first honest clue about what a planet is made of. A world with the density of rock is a different object from one with the density of a small Neptune, even at identical size.
This is where the habitable zone comes in as a filing tool rather than a conclusion. The zone is the orbital band where surface liquid water is possible given some reasonable atmosphere, and its distance from the star depends on the star’s output — close in for a red dwarf, far out for a hot F-type star. Venus and Mars sit at its edges. Neither is habitable. Orbital distance sorts the target list; it does not describe a planet.
Reading the Air
Atmospheric work is where the universe explored by telescopes gets genuinely thin. During a transit a sliver of starlight filters through the planet’s atmosphere before reaching us, and molecules there absorb specific wavelengths. Separate the light into a spectrum, subtract the star, and what remains is chemistry.
JWST made this practical for a small set of targets, especially in the infrared where water, carbon dioxide, methane and sulfur compounds have their strongest features. Ground-based high-resolution spectrographs attack the same problem differently, resolving individual molecular lines and using the planet’s orbital motion to separate them from Earth’s own atmosphere.
Detailed atmospheric measurements exist for a few dozen planets. The catalogue of confirmed planets runs to thousands. That gap is the honest state of the field, and it is why null results — a planet with no detectable atmosphere at all — get published rather than buried.
Blocking the Star to See the Planet
Direct imaging is the technique that does what everyone imagines astronomy does. A coronagraph inside the instrument masks the star’s light so that anything faint beside it can be picked out, and adaptive optics — deformable mirrors adjusting hundreds of times a second, sometimes guided by an artificial laser star — cancels the blurring caused by Earth’s atmosphere.
It works, within limits. In 2008 astronomers using Keck and Gemini imaged multiple giant planets orbiting HR 8799, and follow-up observations have since watched them move along their orbits. Instruments such as SPHERE on the VLT have added more.
Every one of those planets is young, massive, self-luminous and far from its star. An Earth-sized world in a temperate orbit around a Sun-like star is roughly ten billion times fainter than the star and separated from it by a hair’s breadth on the sky. Nothing currently operating can do it.
The Mirrors Being Built Now
The next step is mostly a matter of aperture and stability. The Extremely Large Telescope, a 39-metre segmented mirror rising on Cerro Armazones in Chile, is designed to combine that collecting area with extreme adaptive optics, and it should reach smaller and cooler planets than anything on the ground today.
In space, ESA’s PLATO is built to hunt transiting Earth-sized planets around bright Sun-like stars — the population Kepler could only reach statistically. NASA’s Roman Space Telescope will add a microlensing survey that finds planets in a completely different part of parameter space, including cold worlds far from their stars and free-floating ones with no star at all. And the Square Kilometre Array, though built for radio astronomy generally, will give technosignature searches a scale of coverage that current facilities cannot approach.
Beyond those sits the idea of a dedicated observatory that could directly image an Earth-like planet around a Sun-like star and take its spectrum. NASA has been sketching such a mission for years. It is a project measured in decades, and today’s target lists — including the disappointing entries — are what it will be pointed at.
One World at a Time
The pattern that has emerged is not a single decisive instrument but a chain. A survey telescope finds a candidate. A spectrograph confirms it and supplies a mass. A big infrared telescope tries the atmosphere. Ground-based facilities characterise the star, because a planet around a flaring red dwarf is a different proposition from the same planet around a quiet one. SETI programmes use the resulting maps to choose which stars to listen to, rather than treating every point of light as equally unknown.
Nothing in that chain proves a world is inhabited. What it does is convert a vague question into a stack of measurable ones, and the universe explored this way grows a few planets at a time rather than in leaps.
SETIworld follows the surveys, the instrument commissioning and the results that quietly overturn last year’s assumptions — which happens more often than the press releases let on.
