Data & analysis
Every number on gravityfinder is one of two things: a measurement someone published, or a value we compute ourselves from public data using a known formula. Nothing is invented. This page is the single place that explains where the data comes from, how the analysis works, and how the pictures are made, so the rest of the site can stay focused on the worlds themselves.
Every source below permits commercial use. Individual photographs are credited on the page they appear.
The confirmed exoplanets, their host stars, and the systems they live in: masses, radii, orbits, temperatures, discovery details.
The galaxy catalog: positions, types, magnitudes, and redshifts for roughly ten thousand galaxies beyond the Milky Way.
The black holes caught merging by LIGO, Virgo, and KAGRA: remnant masses, the two masses that combined, and how far away each collision was.
Curated facts for the supermassive and stellar black holes, and for the famous individual stars in other galaxies. Each of those is also cited to a primary paper on its own page.
Real photographs from ESO, ESA/Hubble, NASA, the Event Horizon Telescope, and Wikimedia Commons, used only where the license allows and credited to the observatory beside each image.
The coastlines used to draw the Earth in the size-comparison illustrations.
These values are not in any catalog. We calculate them from the published parameters above using peer-reviewed formulas, and label the result “computed by gravityfinder.”
From a planet's mass and radius compared with Earth's (gravity = mass ÷ radius², with Earth = 1). Only about a third of exoplanet masses are measured directly, so we are honest about it: a directly measured mass is stated plainly, a minimum mass gives a lower bound (“at least”), and a mass estimated from the planet's size is labelled as an estimate. We never imply more precision than the data allows.
A 0-to-1 score for how Earth-like a world is, built from its radius and temperature (Schulze-Makuch et al., 2011). We suppress it when either input is missing.
Whether a planet orbits where liquid water could exist, from the star's temperature and luminosity using the flux boundaries of Kopparapu et al. (2013).
The length of a year (the orbital period), how large the star looks in the sky compared with our Sun, the colour of daylight (from the star's temperature treated as a blackbody, not its spectral letter), and how likely the world is to be tidally locked (from its orbit, the star's mass, and the system's age).
How long the journey would take at five speeds, from a passenger jet to light itself, and what would happen to an unprotected traveler along the way. Distance ÷ speed, nothing more exotic than that.
For galaxies without a directly measured distance, we derive one from the redshift using Hubble's law, but only where the redshift is large enough to be reliable. Where it is not, we show no distance rather than a wrong one.
The Schwarzschild radius, the edge past which nothing returns, computed from the measured mass. The mass is a published measurement, or, for a merger, comes from the gravitational-wave signal.
How far back in time you are seeing an object. Its distance in light-years is exactly how many years its light has been travelling to reach us.
The “roam” links. Nearest neighbours come from each object's position in three dimensions; similar worlds come from matching size, gravity, sunlight, and host-star temperature. Both are precomputed across the whole catalog.
Every planet, star, and galaxy without a real photo is drawn procedurally from its own measured properties: size, temperature, type, host-star colour. The same inputs always produce the same image, so each one is reproducible rather than a one-off.
Where we have a real telescope image, we show it and credit the observatory that took it. Everything else is clearly marked as an illustration, never passed off as a photograph: the planets, stars, and galaxies are drawn procedurally from their own measurements, and a small number of objects we cannot yet photograph, such as most black holes, carry an artist’s impression labelled “illustration, not a photograph.” Diagrams are schematic and not to scale.
We never show a number we cannot stand behind. Where a value is unknown, disputed, or only a lower bound, we say so or leave it blank rather than guess. The formulas above are all published, peer-reviewed work, and the underlying data is public, openly licensed, and reusable from the original sources above. The way we compile, compute, and illustrate it is our own. If you find something that looks wrong, it is a bug, and we want to fix it.