Michael Paycer - How exoplanets are found
Astronomy · Exoplanets · Michael Paycer

How Are Exoplanets Found?

A planet is a dark speck next to a blinding star, trillions of miles away. You can't just point a telescope and see it. So how do we know thousands of them exist? The answer is a set of beautifully indirect tricks — measuring a planet not by its light, but by the mark it leaves on its star.

Method 1 — The Workhorse

The transit method: watching for a wink

Most known exoplanets — the large majority — were found this way. If a planet's orbit happens to be edge-on to us, the planet passes in front of its star once per orbit, blocking a tiny sliver of light. The star dims by a fraction of a percent, then brightens again. Plot that brightness over time and you get a light curve with a distinctive dip. The depth of the dip tells you the planet's size; the spacing between dips tells you the length of its year.

Planet crosses the star's face Brightness Time → the dip = a transit deeper dip = bigger planet

Diagram by Michael Paycer. NASA's Kepler and TESS missions stared at hundreds of thousands of stars at once, watching for exactly these dips.

The transit method is powerful because it scales — a space telescope can monitor a whole field of stars simultaneously — and because a transit lets us do something remarkable: when starlight filters through a planet's atmosphere on its way to us, the gases leave chemical fingerprints. That's how the James Webb Space Telescope now sniffs the air of distant worlds. The catch is geometry: only planets whose orbits line up edge-on ever transit, so this method misses most planets that are really there.

Method 2 — The Original

Radial velocity: the star's tiny wobble

A planet doesn't simply orbit its star — the two bodies pull on each other, and both circle their shared center of mass. The star, being far heavier, makes only a tiny circle, but it does move: it wobbles. As it swings toward us its light is squeezed slightly bluer, and as it swings away, slightly redder — the same Doppler effect that raises and lowers the pitch of a passing siren. Measure that rhythmic color-shift and you've found a planet, and gauged its mass.

center of mass star wobbles planet approaches → bluer recedes → redder the color-shift rhythm reveals the planet

Diagram by Michael Paycer. This was the method behind the very first discovery, 51 Pegasi b, in 1995 — and it remains the best way to weigh a planet.

Methods 3 & 4 — The Specialists

Direct imaging and gravitational lensing

Direct imaging is the rare, hard-won case where we actually photograph the planet — by blotting out the star's overwhelming glare with a mask called a coronagraph, the way you'd raise a hand against the Sun to see something beside it. It only works for young, hot, giant planets in wide orbits, so fewer than a hundred planets have been caught this way. But it's the only method that gives us real light from the planet itself.

Gravitational microlensing uses Einstein's insight that mass bends light. When one star drifts precisely in front of a more distant star, the nearer star's gravity briefly magnifies the far star's light — and if the nearer star has a planet, the planet adds its own tiny extra blip to the brightening. It's a one-time alignment that never repeats, but it's uniquely good at finding planets far from their stars, and even rogue planets with no star at all.

MethodWhat it measuresBest at finding
TransitPlanet's size (dip depth)Planets in edge-on orbits; atmospheres
Radial velocityPlanet's mass (star's wobble)Massive planets close to their star
Direct imagingThe planet's own lightYoung giants in wide orbits
MicrolensingA one-time brightness blipDistant and rogue planets
Misconceptions

Reading the fine print

"We photograph exoplanets." — Almost never; the field runs on indirect detection, and the images you see in the news are artists' impressions. "If we haven't found a planet there, there isn't one." — Not so. Each method has blind spots: transits need a lucky edge-on alignment, radial velocity favors big close-in planets, and small cool worlds slip through both. Absence of detection is not absence of planets. And "the dip proves a planet" — on its own, no: a dip can be a companion star or instrument noise, which is why a discovery isn't "confirmed" until a second method or repeat observation backs it up.

Return to Michael Paycer

Explore Michael Paycer's professional SQL Server, cloud, ETL, API, automation, and consulting pages, or continue browsing the personal interests section.

Return to Michael Paycer