June 2024

Introduction

An exoplanet (also known as an extrasolar planet) is a planet outside of our Solar System. Exoplanets hold fascination since an exoplanet is the most likely location for life elsewhere in the universe. For millennia, one of the most fundamental questions posed by humans has been "Are we alone?" By studying exoplanets, we can learn how solar systems form and we can determine how common, or rare, are the conditions of our Solar System and our planet Earth. The primary goal for a future NASA flagship astrophysics mission is to find biosignatures in the atmospheres of Earth-like exoplanets orbiting Sun-like stars within 200 light-years of Earth. We will discuss that mission - the Habitable Worlds Observatory - in the July 2024 article. This article focuses on the main methods by which we discover and measure the basic properties of exoplanets.

The First Exoplanet Discovery - Pulsar Timing

While news stories about exoplanets appear on a weekly and sometimes daily basis, it is important to remember that exoplanet science is a very recent field of research. The first exoplanet discovery was announced just 32 years ago, in 1992, by Aleksander Wolszczan, a Polish astronomer working at Pennsylvania State University, and Dale Frail, a Canadian astronomer working at the National Radio Astronomy Observatory (New Mexico).

In 1990, Wolszczan and Frail used the Arecibo Radio Observatory in Puerto Rico to discover a pulsar that is named PSR B1257+12. A pulsar, which is the small, ultra dense remnant of a star that exploded as a supernova, emits radio waves extremely regularly as it rotates. If a planet orbits the pulsar, the planet will cause slight anomalies in the timing of the pulsar's observed radio pulses. In 1992, Wolszczan and Frail announced that the pulsar was orbited by two planets; the exoplanets have been named Poltergeist and Phobetor. Pulsars do not provide hospitable environments for life on their exoplanets and pulsar timing has not been a productive method to find exoplanets. To date, only 7 exoplanets have been found orbiting pulsars. This may be because when the star went supernova (and produced the pulsar) the supernova explosion would likely have destroyed any planets orbiting the progenitor star. Thus, any exoplanets orbiting pulsars are probably "second generation" that formed after the supernova explosions.

The quest for discovery of exoplanets that may be hospitable to life has focused on stars that are on the "main sequence". To understand exoplanets and the habitable zone of exoplanets, it is useful to review the evolution of stars.

The Radial Velocity Method for Detecting Exoplanets

Advancements made in spectrometer technologies and observational techniques during the 1980s enabled a new technique to discover exoplanets around main sequence stars. Astronomers began to precisely measure the wavelengths emitted by a star to determine if an exoplanet was causing the star to "wobble" about the barycenter (center of mass) of the star-planet system. This technique is based on measuring the light of the star. The planet is a hidden companion which is indirectly detected by measuring the gravitational effect of the planet on the star's motion. This approach, called the radial velocity method, is shown in the animation below.

To pursue discovery of exoplanets, the Observatoire de Haute-Provence installed the ELODIE echelle spectrograph at their 1.93 meter telescope in southeastern France, achieving first light in 1993. The spectrograph was placed in a temperature controlled room to minimize flexure due to gravity and expansion/contraction due to temperature changes. Visible light (389.5 to 681.5 nm) from stars was supplied to the ELODIE spectrograph using optical fibers mounted at the Cassegrain focus of the telescope. The ELODIE instrument was able to measure radial velocities to an accuracy of ±7 m/sec.

On October 6, 1995, Swiss scientists Michel Mayor and his graduate student Didier Queloz at the University of Geneva, announced at a conference in Florence, Italy that the ELODIE spectrograph was able to discover the first exoplanet orbiting a main sequence star located 50 light-years from Earth, 51 Pegasi.

For this discovery, Mayor and Queloz were awarded half of the 2019 Nobel prize in Physics. Their Nobel prize was shared with James Peebles, a Canadian-American astrophysicist who is widely regarded as one of the world's leading theoretical cosmologists.

The exoplanet, named 51 Pegasi b, is a "hot Jupiter" (large gas planet) orbiting very close to its host star. The orbit of 51 Pegasi b is only 4.3 days. On the west coast of the United States, two American astronomers, Geoff Marcy and Paul Butler, were also using the radial velocity method to search for exoplanets. One week after Mayor and Queloz made their announcement, Marcy and Butler confirmed the 51 Pegasi b discovery using an echelle spectrograph at the Lick Observatory 3-meter telescope in California.

The race was on to find more exoplanets, and the floodgates of discovery were opened. Marcy and Butler announced their first exoplanet discoveries in January 1996, around the stars 70 Viginis and 47 Ursae Majoris. These exoplanets had orbits closer to what we have in our Solar System, with orbital periods of 116 days and 2.5 years. During the next few years, Marcy and Butler used spectrographs at Lick Observatory and the Keck Observatory (in Hawai'i) to discover at least 70 of the 100 exoplanets found by the radial velocity method in the 1990's.

Telescopes that Geoff Marcy and Paul Butler used to discover exoplanets with the radial velocity method
Left: Lick Observatory 3-meter Shane Telescope (California) - 1,283-m (4,209 ft) altitude
Right: Keck I and Keck II 10-meter telescopes (summit of Mauna Kea, Hawai'i) - 4,145-m (13,600 ft) altitude
Photos courtesy of University of California Regents / Ethan Tweedie Photography

During the last 30 years, nearly every major ground-based telescope has installed a high precision spectrograph for searching for exoplanets with the radial velocity method. Through 2023, the radial velocity method has discovered over 1,000 exoplanets, about 20% of the >5,000 exoplanets confirmed.

Most of the radial velocity spectrographs operate in the visible (a.k.a. "optical") using high performance CCDs supplied by Teledyne e2v Space Imaging (Chelmsford, UK) to detect the light. Precise Doppler spectroscopy is most easily done in the optical since:

• wavelength measurement is well calibrated,
• optical detectors are better behaved and more easily calibrated than infrared detectors, and
• absorption and emission from the Earth's atmosphere is easier to correct in the optical than it is in the infrared.

Over the past 10 years, an increasing number of radial velocity spectrographs have been developed for the infrared due to the following reasons:

• The vast majority of stars in the Milky Way - about 73% - are M-type stars which emit their maximum light flux in the infrared.
• It is easier to find planets that are in the habitable zone around cool M-dwarf stars. The Habitable Zone is the range of orbital radii within which the exoplanet surface can support liquid water given sufficient atmospheric pressure.
  • Sometimes the habitable zone is referred to as the Goldilocks Zone, an allusion to the children's fairy tale of "Goldilocks and the Three Bears", in which a little girl chooses from sets of three items. When choosing porridge, she chose the porridge that was "not too hot and not too cold".
• Since M-stars are cool, the habitable zone is close to the star, and exoplanets in the habitable zone around M-stars have shorter periods than G-type (Sun-like) and K-type stars.
• Exoplanets around M-stars also can have a larger exoplanet to star mass ratio. Combined with the closer orbits of the habitable zone, the star's wobble is much larger, and the precision of an infrared radial velocity spectrograph does not have to be as good as the optical spectrographs.

Most of the infrared radial velocity spectrographs use the H2RG or H4RG-15 infrared focal plane arrays produced by Teledyne Imaging Sensors (Camarillo, California).

Today, the state-of-the-art in high precision radial velocity spectrographs is represented by the HARPS and ESPRESSO spectrographs of the European Southern Observatory (ESO). Both HARPS and ESPRESSO are stable with precision below 1 m/sec, with ESPRESSO aspiring to 0.1 m/sec precision. This is a large improvement over the 7 m/sec precision achieved by the ELOIDE spectrograph that discovered the first exoplanet orbiting a main sequence star in 1995.

The High Accuracy Radial-velocity Planet Searcher (HARPS) spectrograph was designed with every possible feature to enable it to measure radial velocities with high accuracies over long term observations. To achieve the goal of 1 meter/sec accuracy over extended periods, the HARPS spectrograph is housed within a vacuum vessel buried within the building of the 3.6-meter telescope at the La Silla Observatory that is operated by the European Southern Observatory (see image below). The vacuum vessel enables the spectrograph to avoid spectral drift due to temperature and air pressure variations. The temperature is further stabilized by placing the vacuum vessel in a temperature controlled room with human access only during maintenance periods. Starlight is fed to the spectrograph by optical fibers mounted at Cassegrain focus of the 3.6-meter telescope.

The HARPS spectrograph uses two Teledyne e2v 2048x4102 pixel CCDs (15 µm pitch) to sense light over the 378 - 691 nm band (near UV through red). An example of this CCD, named EEV44-82, is shown below in the CCD test dewar at ESO Headquarters (Garching, Germany). The HARPS vacuum vessel is shown while being tested at the Geneva Observatory (Switzerland).

There are advantages and drawbacks for the radial velocity method for discovery of exoplanets.

• Advantages
  • The radial velocity method is ideal for ground-based telescopes since observations do not need to be made continuously as required for transit photometry (discussed in the next section).
  • The radial velocity measurement is not dependent on the distance to the star (with the caveat that stars get fainter the farther the stars are from Earth). Radial velocity could be used for stars across the Milky Way galaxy if those stars were sufficiently bright.
• Drawbacks
  • A fundamental feature of the radial velocity method is that it cannot accurately measure the mass of an exoplanet; it only provides an estimate of the exoplanet's minimum mass. This is because radial velocity only measures the motion of stars towards and away from Earth. This is not a problem if the orbital plane of the exoplanet appears edge-on when observed from Earth. But most exoplanet orbits are tilted relative to the direction to the Earth and the "wobble" of the star is decreased by the sine of the angle between direction toward the Earth and the orbital plane of the exoplanet. This sin(i) term, where i denotes the orbit inclination, reduces the observed wobble and decreases the derived exoplanet mass.
  • A second drawback of the radial velocity method is that it is biased toward finding large exoplanets orbiting smaller M-type stars. These exoplanets are unlikely abodes for life.
  • Lastly, there is a practical limit to the sensitivity of radial velocities since stars have features and convective cells that produce noise on the velocity measurements of the star. Depending on the type of star, this noise can be more than 1 meter/sec. Our Sun is a relatively quiet star but it still produces radial velocity scatter due to sunspots at the level of 50 cm/sec (0.5 m/sec). While that may seem small, the radial velocity signal that the Earth has on our Sun is only 10 cm/sec (0.1 m/sec). No matter what advancements are made during the next 20 years, astronomers do not expect to be able to use the radial velocity method to detect Earth analogue exoplanets.

Radial velocity measurements opened the field of exoplanet discovery and, through 2023, the radial velocity method has discovered over 1,000 exoplanets. However, the greatest number of exoplanets have been discovered using a method termed "transit photometry" which is explained in the next section.

The Transit Photometry Method for Detecting Exoplanets

Viewed from Earth, if the orbital plane of exoplanets are seen edge on, the exoplanet can be detected by measuring the brightness of the star over time to see a dip in brightness when the exoplanet transits its host star. A larger exoplanet will create a larger dip in brightness. This is shown in the animation below.

There is a dip in brightness when the exoplanet passes in front of the star, and also when the exoplanet passes behind the star, i.e. the exoplanet is occulted by the star. This is shown in the graphic below.

The transit photometry approach for finding exoplanets measures the total brightness of the star + exoplanet. If the light from star + exoplanet is fed to a spectrograph, this approach can be expanded to transit spectroscopy. Accurate measurements of the spectra from star alone (planet behind the star), star plus exoplanet (planet to the side of the star), and exoplanet in front of the star can be processed to isolate the exoplanet spectrum from the host star spectrum. Transit spectroscopy is being attempted with the James Webb Space Telescope; preliminary results have been mixed.

Ariel: In 2029, the European Space Agency (ESA) will launch a mission that is optimized for transit spectroscopy: the Atmospheric Remote-Sensing Infrared Large-survey (Ariel) mission. Ariel will be located at Lagrange Point L2 and will perform a chemical census of the atmospheres of 1000 transiting exoplanets. The Ariel mission will focus on exceptionally hot planets in our galaxy, with temperatures greater than 600 degrees Fahrenheit (320 degrees Celsius). Such planets are more likely to transit their star than planets orbiting farther out, and their short orbital periods provide more opportunities to observe transits in a given period of time. More transits give astronomers more data, allowing them to reveal the weak chemical fingerprint of a planet's atmosphere.

Ariel's hot exoplanet population will include gas giants like Jupiter, as well as smaller gaseous planets called mini-Neptunes and rocky worlds bigger than our planet called super-Earths. While these planets are too hot to host life as we know it, they will tell us a lot about how planets and planetary systems form and evolve. Additionally, the techniques and insights learned in studying exoplanets with Ariel will be useful when scientists use future telescopes to look toward smaller, colder, rockier worlds with conditions that more closely resemble Earth's.

Ariel will perform observations in three photometric bands (0.5-0.6 µm, 0.6-0.8 µm, and 0.8-1.1 µm) and three spectroscopic bands: 1.1-1.95 µm, 1.95-3.9 µm, and 3.9-7.8 µm. Ariel will simultaneously measure all wavelengths in the 0.5-7.8 µm spectral range.

Ariel will use infrared focal plane arrays that were produced by Teledyne Imaging Sensors (Camarillo, California). Ariel will use three substrate-removed SWIR (2.3 µm λ_co) H2RG focal plane arrays (FPAs) for the photometric bands and the shortest wavelength spectroscopic band. The SWIR H2RG FPAs will be operated by SIDECAR ASIC modules. These SWIR H2RG FPAs and SIDECAR ASIC modules are residual from ESA's Euclid mission. Ariel will use two H1RG FPAs, with cutoffs of 4.8 and 8.2 µm, for the 1.95-7.8 µm spectroscopic bands. The H1RG FPAs will be actively cooled by a Joule-Thomson system to an operating temperature of 42K.

The transit photometry approach has discovered most of the exoplanets found to date. This success is primarily due to two NASA space missions - Kepler and TESS - which are explained below.

Kepler: The Kepler mission was NASA's first planet-hunting mission to use transit photometry. Launched on March 6, 2009, Kepler combined cutting-edge observational techniques for measuring stellar brightness with the largest digital camera outfitted for space at that time. Kepler used a large focal plane mosaic (shown below) that was outfitted with 25 modules mounted on a curved surface to optimize the instrument focus. (Although it is hard to see in the image, each module is flat.) There are 21 modules with image arrays used for measuring the brightness of stars, and four corner modules used for measuring guide stars to maintain accurate telescope pointing. The focal plane mosaic was assembled by Ball Aerospace in Boulder, Colorado. (Ball Aerospace was acquired by BAE Systems in February 2024.)

Each of the 21 science modules uses two charge coupled devices (CCDs) produced by Teledyne e2v Space Imaging in Chelmsford, UK. Each CCD has 2200x1024 pixels. The total focal plane mosaic contains nearly 95 million pixels.

The original Kepler mission stared continuously for 3.5 years at over 100,000 stars in the constellation Cygnus. The Kepler mission was extended multiple times and was able to observe over 500,000 stars during a nine-year operation that ended in 2018. With a relatively large primary mirror of 1.4-meter diameter and sensitive CCDs, Kepler (and its follow-on mission, dubbed K2) achieved a high signal-to-noise ratio that enabled the discovery of over half of all confirmed exoplanets discovered through April 2024. Kepler and K2 are credited with 3,322 confirmed exoplanets and 2,959 exoplanet candidates yet to be confirmed.

NASA's Kepler mission was a statistical transit survey designed to determine the frequency of Earth-sized planets around other stars. Kepler concentrated on a relatively small region of the sky - 115 square degrees, only about 0.25% of the full sky - searching for exoplanets orbiting stars up to a distance of 3,000 light years. While Kepler was revolutionary in its finding that Earth-to-Neptune-sized planets are common, the bulk of the stars in the Kepler field lie at distances of hundreds to thousands of light-years, making it difficult to obtain ground-based follow-up observations for many systems.

TESS: NASA's second planet-finding mission based on the transit photometry approach, named TESS, has a different goal: observe most of the sky but only find exoplanets relatively close to the Earth. The Transiting Exoplanet Survey Satellite (TESS) mission was designed to discover exoplanets in orbit around the brightest dwarf stars in the sky, searching within 200 light-years of the Earth. This design trade was implemented by four wide-field of view cameras, each with a relatively small 10.5 cm (0.105-m) aperture. The combined field of view of the four cameras is 24x96 degrees. The TESS mission imaged about 75% of the sky during its two-year primary mission. Launched in 2018, the TESS mission has been extended twice and is still operational. As of November 2023, TESS has discovered at least 450 confirmed exoplanets with more than 4,500 candidate exoplanets awaiting confirmation (or rejection) as exoplanets. (The TESS mission is operated by the Massachusetts Institute of Technology. The four TESS cameras use a total of 16 2Kx2K frame transfer CCDs - 2Kx4K pixels each - produced by MIT Lincoln Laboratory.)

PLATO: Not to be left behind in the hunt for exoplanets, the European Space Agency (ESA) will launch an ambitious transit photometry mission in 2026. The PLAnetary Transits and Oscillations (PLATO) mission will have 26 cameras to find and study terrestrial exoplanets in orbits in the habitable zones of Sun-like stars. PLATO will be launched to Lagrange Point L2 and will observe more than 200,000 stars during its nominal four-year mission. Each of the 26 PLATO cameras uses four Teledyne e2v CCDs; each CCD has 4510x4510 pixels (18 µm pitch). Each camera has more than 81 million pixels, and there are a total of more than 2 billion CCD pixels in the 26 cameras of the PLATO mission, with images taken every 25 seconds. ESA officially designated the PLATO CCDs as "CCD Bruyères" in honor of the late Teledyne e2v employee Jean-Francois Bruyères. Jean-Francois was a long-serving and highly respected colleague in the field of space imaging who was instrumental in the development of CCDs for many ESA missions.

PLATO CCD Cameras under test and single CCD
Left: The 26 PLATO cameras integrated and under test in the cleanroom of OHB Systems AG (Bremen, Germany).
Right: One of the 104 Bruyères CCDs (4510x4510 pixels) under test at Teledyne e2v (Chelmsford, UK).
Credit: OHB, Teledyne e2v

The Holy Grail for Exoplanet Discovery and Measurement - Direct Imaging

The discovery of exoplanets using radial velocity spectroscopy was a very significant event in astronomy. Although today we hear news stories about exoplanets on a weekly basis, before the 1990's, the existence of exoplanets was in the realm of theory and speculation. Radial velocity spectroscopy showed us that planets can be readily found orbiting main sequence stars. The large number of exoplanets found via transit photometry, which relies on unique alignment of the exoplanet orbit with direction to the Earth, demonstrated that exoplanets are common. Through April 2024, over 5,600 exoplanets have been confirmed with nearly 10,000 additional exoplanet candidates identified (and awaiting confirmation or rejection). Based on studies to date, astronomers now think there are on average one to two exoplanets per star, with some stars having several exoplanets. That amounts to about 1 billion exoplanets in our Milky Way galaxy. The consensus is that 20% of Sun-like stars have an Earth-sized planet in its habitable zone.

While the radial velocity and transit photometry techniques have been very useful for birthing the new field of exoplanet science, both techniques are indirect measurements of exoplanets, requiring only measurement of the host star light. Missions like Ariel will use transit spectroscopy, a technique that measures the light of both the star and the exoplanet. Ariel will study hot exoplanets (gas giant, mini-Neptunes, super-Earths) but cannot study Earth-like exoplanets orbiting Sun-like stars. For an Earth-like planet orbiting in the habitable zone of a Sun-like star, the host star's light is 100 million (10⁸) to 10 billion (10¹⁰) times brighter than the exoplanet's light. The techniques of radial velocity and transit photometry will not work for Earth analogues since the faint light from an Earth-like exoplanet will be swamped by the "photon noise" of the much brighter Sun-like star.

• All light has statistical fluctuations. Even a "constant intensity" light source has fluctuations in the number of photons received at Earth during a given time period. "Photon shot noise" is equal to the square root of the number of photons.
• Since a Sun-like star is 10⁸ to 10¹⁰ times brighter than an Earth-like exoplanet, the photon noise of the Sun-like star is 10,000 (10⁴) to 100,000 (10⁵) times the signal of the exoplanet.
• Transit spectroscopy will never be able to disentangle the light from an Earth analogue exoplanet from the light of its host star.

The only way to study an Earth-like exoplanet for the biosignatures of life is to fully separate the exoplanet's light, measuring it independently from its host star. The only approach that can do this is direct imaging of the exoplanet while suppressing the light of the star with a "coronagraph". Named after optics used to study the Sun's corona, a coronagraph is a set of optics that blocks out the direct light from a star or other bright object, enabling the light from a faint nearby object (the exoplanet) to be measured. When used to study exoplanets, this set of optics is sometimes called a stellar coronagraph. The concept of a stellar coronagraph is presented in the NASA animation (see below) that explains the coronagraph in the Roman Space Telescope (RST).

The RST coronagraph will reduce an exoplanet's host star light by a factor of 200 million (2x10⁸). The coronagraph achieves this contrast ratio by using a specialized coronagraphic mask and "adaptive optics" to remove any wavefront distortions. The Roman Space Telescope is scheduled to launch by May 2027.

For the past 20 years, coronagraphic direct imaging has been pursued by several ground-based observatories that developed coronagraphs outfitted with adaptive optics to correct for the blurring of the Earth's atmosphere.

The first discovery of an exoplanet by direct imaging was announced in 2008 by Christian Marois of the Herzberg Institute of Astrophysics (Canada). The Marois group used a coronagraphic instrument at the Keck Observatory (Hawai'i) to discover three exoplanets orbiting a young star named HR 8799. The movie below, generated from data taken over a period of 7 years at the Keck Observatory, shows four planets more massive than Jupiter orbiting HR 8799. The movie doesn't show full orbits since only partial orbits were completed over 7 years. The closest-in exoplanet orbits the star in about 40 years; the farthest-out exoplanet takes more than 400 years. The black circle in the center of the image is from the coronagraph blocking the blinding light of the star, making the exoplanets visible. Star light leaking past the coronagraph mask appears as the speckle pattern outside the central black area. The images were initially captured by Dr. Christian Marois of the National Research Council of Canada's Herzberg Institute of Astrophysics. The movie animation was constructed by Jason Wang of the University of California, Berkeley. The movie is based on eight observations of the exoplanets since 2009; a motion interpolation algorithm was used to draw the orbit between those points.

HR 8799 is about 130 light-years away in the constellation of Pegasus. By coincidence, it is quite close to the star 51 Pegasi, where the first exoplanet was detected in 1995. HR 8799 is about 30 million years old and is almost five times brighter than the Sun.

The exoplanets of HR 8799 were the first exoplanets to be directly imaged. During the past 15 years, 40 additional exoplanets have been directly imaged by ground-based telescopes. However, it has only been possible to image large gas exoplanets orbiting far from their host stars. The existing coronagraphs are not good enough to image Earth-like exoplanets around Sun-like stars.

But these observations demonstrate that direct imaging of exoplanets is feasible and, in 2027, the Roman Space Telescope coronagraph will take the next big step in this technology. The world's astronomical community is anxiously awaiting more advanced space-based coronagraph imagers, operating above the Earth's blurring atmosphere, to be able to directly image and spectroscopically study the atmospheres of small exoplanets orbiting close to their host star. The July 2024 article will present the Habitable Worlds Observatory (HWO), NASA's future flagship astrophysics mission that will launch in the late 2030's. HWO aims to achieve 10¹⁰ suppression of the star's light to enable imaging and spectroscopy of Earth-like exoplanets. HWO will attempt to answer the age old question "Are we alone?".

The Earth has been in existence for about 4.6 billion years with the Earth's history roughly divided into four different eras during which the Earth's atmosphere evolved:

• 600 million years in the "Hadean" era
• 1,500 million years in the "Archean" era
• 2,000 million years in the "Proterozoic" era
• 540 million years in the "Modern" era

Postscript

This article discussed only a few of the methods by which exoplanets are discovered, with focus on:

• Radial Velocity
• Transit Photometry
• Direct Imaging

Several other methods to find exoplanets were not presented. But one method that deserves mentioning is gravitational microlensing. Gravitational microlensing occurs when the gravitational field of a star acts like a lens, magnifying the light of a more distant background star. This effect can occur when the two stars are almost exactly aligned. Lensing events are brief, lasting for weeks or days, as the two stars and Earth are all moving relative to each other. Over the past ten years, 217 exoplanets have been discovered by gravitational microlensing.

If the foreground lensing star has an exoplanet, then the exoplanet's own gravitational field can make a detectable contribution to the lensing effect. Since gravitational microlensing from an exoplanet requires a highly unlikely alignment, a very large number of distant stars must be continuously monitored in order to detect an exoplanetary microlensing event. Gravitational microlensing is most fruitful for exoplanets that are located between the Earth and the center of the Milky Way galaxy, as the galactic center provides a large number of background stars.

The Roman Space Telescope has an observational program labeled the "Galactic Bulge Time-Domain Survey" that will scour the center of the Milky Way for microlensing events. This survey will provide one of the deepest views ever into the heart of our galaxy, making observations every 15 minutes for over a year. Tracking how hundreds of millions of stars change in brightness is predicted to discover up to 100,000 exoplanets and find isolated black holes.