From Zero to Thousands in a Single Generation

Pick any astronomy textbook printed before 1992 and look up the planets. You will find nine, or eight if the author had strong opinions about Pluto, and every one of them orbits our own Sun. That was the entire confirmed catalogue. People assumed other stars had planets too, the same way you assume other houses have kitchens, but assuming is not the same as knowing, and for most of human history nobody could close that gap. Then the gap closed, and it closed fast. Today the count runs past five thousand confirmed worlds with thousands more candidates waiting in line, and the number climbs almost every week.

What happened in between is not really a story about a clever idea. The ideas had been around for decades. It is a story about machinery. Specific telescopes, specific detectors, specific missions launched on specific dates, each one built to measure starlight more precisely than anything before it. The leap from zero to thousands happened because astronomers stopped relying on the human eye and the ground-bound observatory and started flying purpose-built instruments above the atmosphere. This article is about that machinery, the surveys that drove the boom, and how raw measurements turn into confirmed planets at industrial scale.

The First Cracks in the Wall

The wall came down in stages. In 1992, two radio astronomers, Aleksander Wolszczan and Dale Frail, found planets orbiting a pulsar called PSR B1257+12, the burnt-out core of a dead star. It was a strange place to find the first confirmed worlds beyond the Sun, and not the kind of place anyone expected to look for life, but it proved planets existed elsewhere. Three years later came the one that really mattered. Michel Mayor and Didier Queloz, working at the Haute-Provence Observatory in France, announced 51 Pegasi b, a planet roughly the mass of Jupiter whipping around a Sun-like star every four days. That detection won a Nobel Prize in 2019, and it kicked off everything that followed.

51 Pegasi b was found from the ground, by watching its star wobble. A planet tugs on its parent star as it orbits, and that tug shows up as a tiny rhythmic shift in the star’s spectrum. The shifts involved are absurdly small, equivalent to the star moving toward and away from us at walking pace, yet the spectrographs of the mid-1990s could just barely measure them. For the next decade, ground-based wobble-hunting was the main game in town, and it found mostly big planets on tight orbits, because those produce the strongest tug. The field needed a different approach to find smaller worlds, and to find them by the hundreds. That approach was already being designed, and it would fly.

Kepler and the Great Haul

If one mission defines the modern era of planet-finding, it is Kepler. NASA launched it in 2009 with a deceptively simple plan. Stare at one patch of sky in the constellation Cygnus, roughly 150,000 stars at once, and watch for the faint, repeating dip in brightness that happens when a planet crosses in front of its star. The dip is small, often a fraction of a percent, and it lasts hours. Doing this from the ground is nearly hopeless because the atmosphere makes starlight shimmer and flicker. From space, with no air in the way, Kepler could measure a star’s brightness steadily enough to catch a planet the size of Earth blocking a sliver of its light.

The haul was staggering. Kepler is credited with somewhere north of 2,600 confirmed planets, more than half of everything we know, plus thousands of additional candidates. It did not just count worlds; it took a census. Before Kepler, nobody knew whether planets were common or rare. After Kepler, we could say with confidence that planets outnumber stars in our galaxy, that small rocky worlds are abundant, and that a meaningful fraction of stars host a planet in the temperate zone where liquid water could survive. That single statistical result reshaped how astronomers think about the Milky Way.

Then, in 2013, two of the spacecraft’s reaction wheels failed, and Kepler lost the ability to hold its steady stare. Most missions would have ended there. Instead, engineers worked out a way to balance the telescope using pressure from sunlight itself, and Kepler came back as a second mission called K2, hopping from one field to another along the plane of the sky. It kept finding planets for years after it should have been dead, until it finally ran out of fuel in 2018. Few instruments have ever earned a second life so completely.

TESS and the All-Sky Sweep

Kepler’s weakness was the flip side of its strength. By staring at one distant patch of sky, it found enormous numbers of planets, but most of them orbit stars hundreds or thousands of light-years away, too faint for detailed follow-up. The next step was to trade depth for breadth and look at the bright, nearby stars instead. That is the job of TESS, the Transiting Exoplanet Survey Satellite, launched in 2018.

TESS does not stare at one field. It tiles the whole sky into sectors and works through them a month at a time, covering both hemispheres over its mission. The planets it finds tend to orbit stars close enough and bright enough that other instruments can study them in detail afterward. That is the real point of TESS: it is a finder, a scout that hands off promising targets to bigger telescopes for the hard work of measuring masses and sniffing atmospheres. It has already turned up thousands of candidates and confirmed hundreds, including small rocky worlds around nearby red dwarf stars that are now prime targets for atmospheric study. Where Kepler asked “how common are planets,” TESS asks “which nearby planets are worth a closer look,” and those are different questions answered by different machines.

Gaia and the Astrometric Map

Not every great planet-finding mission was built to find planets. Gaia, launched by the European Space Agency in 2013, was designed to map the positions, distances, and motions of more than a billion stars with almost unbelievable precision. That map is one of the most valuable datasets in all of astronomy, and it doubles as a planet-detection engine through a method called astrometry. When a planet orbits a star, it does not just make the star wobble toward and away from us; it makes the star trace a tiny loop on the sky. Gaia is precise enough to measure those loops for the right stars.

This matters because astrometry is sensitive to planets that the other methods often miss, particularly massive planets on wide orbits, the Jupiter-like worlds far from their stars. Gaia’s measurements also underpin almost everything else in the field. When TESS or a ground telescope finds a planet, Gaia’s data tells us how far away the host star is, how big and bright it really is, and therefore how big the planet really is. A planet measurement is only as good as the star measurement behind it, and Gaia gave us the star measurements. Its later data releases are expected to deliver thousands of new planet detections on their own, a quiet contribution from a mission that was never marketed as a planet hunter.

The Ground Strikes Back: HARPS, ESPRESSO, and the Big Dishes

Space telescopes get the headlines, but the ground never stopped competing, especially in the wobble game. The atmosphere blurs a star’s brightness, which is why transit surveys went to space, but it does far less damage to the precise color measurements that radial velocity depends on. So the great spectrographs stayed on the ground and simply got better. HARPS, installed on a 3.6-meter telescope at the European Southern Observatory in Chile, set the standard for years, stable enough to detect the gentle tug of small planets. It was HARPS data that helped pin down Proxima b, a roughly Earth-mass planet in the temperate zone of Proxima Centauri, the nearest star to the Sun at just over four light-years.

Its successor, ESPRESSO, went further still, feeding light from the four giant 8-meter units of the Very Large Telescope into a single instrument engineered to measure stellar motion down to a few tens of centimeters per second. That is roughly the pace of a person strolling, measured across light-years of empty space. These spectrographs do something the transit surveys cannot: they weigh planets. A transit tells you how wide a planet is by how much light it blocks; radial velocity tells you how heavy it is by how hard it pulls. Put the two together and you get density, the single number that separates a dense rocky planet from a bloated gas giant. The biggest dishes and arrays on Earth keep this side of the field alive, and the future Extremely Large Telescope, with a mirror nearly forty meters across, is built to push it harder.

Why Space Beats the Atmosphere

It is worth pausing on why so much of this work moved off the planet. Earth’s atmosphere is wonderful for breathing and terrible for precision photometry. Air is turbulent, so starlight arriving at a ground telescope shimmers and wavers, the same effect that makes stars appear to twinkle. For catching a planetary transit that dims a star by a hundredth of a percent, that twinkle is fatal noise. The atmosphere also glows faintly, soaks up whole bands of infrared light, and shifts with weather and temperature. A telescope above all of it sees a steady, dark, transparent sky.

That stability is the whole reason Kepler and TESS could do what no ground survey had managed. A space telescope can watch the same star for weeks without a cloud, a sunrise, or a gust of wind interrupting the measurement, and it can reach infrared wavelengths that never make it through the air at all. The trade is brutal cost and the fact that you cannot send a technician up to fix a broken part, as Kepler’s failed reaction wheels showed. But for the specific job of measuring tiny, steady changes in starlight, getting above the atmosphere is not a luxury. It is the difference between seeing the signal and drowning in noise.

CHEOPS, JWST, and the Shift Toward Atmospheres

Once you have thousands of planets, the question changes from “are they there” to “what are they like.” The newest missions reflect that shift. CHEOPS, a small European telescope launched in 2019, does not hunt for unknown planets across the sky. It revisits stars already known to host planets and measures their transits with extra precision, nailing down sizes and catching additional worlds that the discovery survey missed. It is a characterization instrument, a follow-up specialist rather than a scout.

The James Webb Space Telescope took the field somewhere genuinely new. JWST was not built primarily to find planets, but its enormous mirror and infrared instruments are exquisitely suited to studying their atmospheres. When a planet transits its star, a thin sliver of starlight filters through the planet’s atmosphere on its way to us, and the gases there imprint their fingerprints on that light. JWST can read those fingerprints. It has already detected carbon dioxide, water vapor, and other molecules in the air of distant worlds, including planets in the famous TRAPPIST-1 system, a cluster of seven Earth-sized worlds around a small red star roughly forty light-years away. This is the frontier the whole enterprise was building toward: not just counting planets, but chemically inspecting them for the kinds of imbalances that, on Earth, are produced by life.

From Raw Data to a Confirmed World

None of these instruments hands an astronomer a finished planet. What they produce is data, oceans of it, and the planet has to be dug out. A mission like TESS downloads brightness measurements for tens of thousands of stars, and software pipelines comb through every one of them looking for the periodic dips that signal a transit. The volume is far beyond what any human team could read by hand, which is why machine learning has become a working part of the field, trained to flag the handful of real signals hiding among millions of wiggles caused by starspots, instrument glitches, and pure noise.

A flagged candidate is not yet a planet, only a suspect. Plenty of things mimic a transit, most commonly a pair of stars eclipsing each other in the background and bleeding their dimming into the target’s measurement. So candidates get cross-checked. The radial velocity spectrographs weigh the suspect to confirm it has a planet’s mass and not a star’s. Astrometric and brightness data from Gaia rule out background impostors. Sometimes a second telescope watches an independent transit to make sure the signal is real and tied to the right star. Only after a candidate survives this gauntlet does it graduate to the confirmed list. The pipeline from a faint dip in a data file to an entry in the catalogue is long, automated at the front and skeptical at the back, and it is what lets modern surveys turn raw starlight into thousands of known worlds without flooding the record with mistakes.

The Machinery Keeps Growing

Stand back and the pattern is clear. The explosion in known planets was not one discovery but a chain of instruments, each handing off to the next. Ground spectrographs cracked the wall open. Kepler proved planets are everywhere and took the census. TESS scouts the bright nearby stars. Gaia maps the host stars and catches wide-orbit giants. HARPS and ESPRESSO weigh the worlds. CHEOPS sharpens the measurements. JWST reads the atmospheres. Each machine does one part of a job too big for any single design, and together they built the modern catalogue from a starting point of literally nothing within one human lifetime.

What comes next is more of the same, only larger. The European Space Agency’s PLATO mission, built to find Earth-sized planets in temperate orbits around Sun-like stars, is on the way. The Extremely Large Telescope and other giant ground observatories are rising. Each one narrows the haystack a little more, and each one moves the field closer to the planet everyone wants: a small rocky world in the right place around a nearby star, with an atmosphere we can actually inspect. The instruments are the story, and the story is far from finished.

Register on our portal to follow the missions, instruments, and discoveries that are turning the search for distant worlds into one of the great ongoing projects of modern science.