The close-approach desk · read live
Every near miss.
Counted down honestly.
The board lists every known object that will pass inside 0.05 astronomical units of Earth over the coming weeks — the designation, the moment of closest approach in your own timezone, how far it misses in lunar distances, how fast it is going, and how bright it is. The table is JPL’s own Close Approach Data, read through this folder’s relay without a key or an account.
Below it, the fireball log: the small bodies that did not miss, as U.S. Government sensors recorded them entering the atmosphere — when, how much energy, and where — each one placed on a world map drawn by this page’s own hand.
JPL Solar System Dynamics · approaches re-read every ten minutes, fireballs every thirty · 1 lunar distance = 384,400 km = 0.00257 au
The board
Inside 0.05 au.
Designation, miss distance, velocity and absolute magnitude, in the order they arrive. A countdown here is only as fine as JPL’s own word: where the time of closest approach is known to the minute it counts seconds, where it is known to the hour it counts minutes, and where JPL only knows the day it says the day — with the stated uncertainty printed beside every one.
Approaches · —
waiting for JPL
Units. JPL states every miss distance in astronomical units; this page divides it by 0.0025696 au — one lunar distance, 384,400 km, the mean Earth–Moon distance CNEOS itself uses — so 0.05 au = 19.46 LD. The range in brackets is JPL’s own minimum and maximum. Velocity is the relative velocity at closest approach, in km/s. H is absolute magnitude: smaller is brighter and, for a given surface, larger. No diameter is printed because none is measured — a size from H needs an assumed albedo, and this desk does not assume.
Not read yet.
Fireballs
What the sky already threw.
The thirty most recent fireballs in JPL’s log — atmospheric impacts bright enough for U.S. Government sensors to record, with their total radiated energy and the impact energy derived from it, in kilotons of TNT. Where the sensors reported a position it is on the map, sized by that energy; where they did not, the row says so and the map stays honest.
Fireball log
waiting for JPL
Not read yet.
About
A desk, not an observatory.
Both tables are the Solar System Dynamics group’s at NASA’s
Jet Propulsion Laboratory, ssd-api.jpl.nasa.gov: the
Close Approach Data API for the board, the Fireball Data API for the
log. Neither sends the header a browser needs to read it from another
site, so this folder carries a relay of its own — one hardcoded
host, two hardcoded routes, three numbers a caller may vary and every
one of them range-checked before it is printed into the request. It
is not an open proxy. The relay keeps JPL’s answer for ten
minutes (approaches) or thirty (fireballs), so a page of visitors is
one call to JPL, not one each; the status line says whether you are
looking at a fresh read or the relay’s last good copy, and when
JPL last answered.
A countdown is only as honest as its target. JPL publishes a three-sigma uncertainty for the time of every close approach, and this page prints it under every countdown and lets it set the countdown’s own grain: seconds under an hour of uncertainty, minutes under a day, days and hours beyond that. The times are said in your own timezone because that is the clock on your wall; the UTC figure JPL gave is a hover away.
Inside 0.05 au is JPL’s own default window, and the one within which an object of enough size is called potentially hazardous; the two sieves narrow the same read to five lunar distances and to one. The coastlines are Natural Earth’s public-domain 1:110 million set, carried whole and drawn by the page. What neither table says, this page does not invent: no diameters, no risk, no adjectives. A table that will not answer is reported as exactly that. Nothing is stored anywhere but this browser and the relay’s cache file. Questions and corrections: labs@labs.llc.