This month
The whole calendar month. Four lunar phases, whatever the planets are doing, and any eclipse or shower that falls inside it.
Times are UTC throughout, because an instant is the same everywhere and a local rendering would hide which day an event actually falls on for you. Nothing here knows or asks where you are. Every computed instant carries an uncertainty, and most of those are measurements rather than estimates: every build checks the lunar phases against NASA's own published table, the equinoxes, solstices and Earth's apsides against the US Naval Observatory's, and the planetary positions the remaining events are derived from against JPL Horizons. The Moon's perigee and apogee are the one family with no external table to check against, and their cards say so.
11 events
Last Quarter Moon
ComputedHalf the disc is lit again, on the other side. The Moon rises around midnight, so the evening sky is dark.
How exact: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the Moon's geocentric ecliptic longitude exceeds the Sun's by the phase angle, both referred to the mean equinox of date, located by bisection on the low-precision lunar and solar series. (lunar-phase-elongation-crossing, version 1.0.0)
Uncertainty: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where it applies: The instant is the same everywhere; only the local clock time and whether the Moon is above your horizon differ.
Mercury at inferior conjunction
ComputedMercury passes between Earth and the Sun and is not observable. It moves from the evening sky into the morning sky around this date.
How exact: Within a few hours, from planetary positions checked against JPL Horizons on every build to under five arcminutes.
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the planet's geocentric ecliptic longitude equals the Sun's, both referred to the mean equinox of date; the geocentric distance separates an inferior conjunction from a superior one. (planet-solar-longitude-conjunction, version 1.0.0)
Uncertainty: Within a few hours, from planetary positions checked against JPL Horizons on every build to under five arcminutes.
Where it applies: Not observable from anywhere: the planet is in the same direction as the Sun. Never attempt to look.
New Moon
ComputedThe Moon passes between Earth and the Sun and its near side is unlit. The darkest skies of the month fall around this date.
How exact: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the Moon's geocentric ecliptic longitude exceeds the Sun's by the phase angle, both referred to the mean equinox of date, located by bisection on the low-precision lunar and solar series. (lunar-phase-elongation-crossing, version 1.0.0)
Uncertainty: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where it applies: The instant is the same everywhere; only the local clock time and whether the Moon is above your horizon differ.
Moon at apogee
ComputedThe Moon reaches the farthest point of its orbit this month and appears marginally smaller than average.
How exact: The date is reliable; the hour is not. The distance curve is almost flat for a day either side, and unlike the phases and the equinoxes this one is not checked against an external table — no comparable published series is connected — so treat the time as indicative.
- Distance
- about 405,300 km, centre to centre
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: A turning point of the Earth–Moon distance from the low-precision lunar series, located by bisecting the derivative after a coarse scan. (lunar-apsis-distance-extremum, version 1.0.0)
Uncertainty: The date is reliable; the hour is not. The distance curve is almost flat for a day either side, and unlike the phases and the equinoxes this one is not checked against an external table — no comparable published series is connected — so treat the time as indicative.
Where it applies: A property of the orbit, the same for every observer.
Mars and Jupiter in conjunction
ComputedThe two planets pass 1.2° apart — about 2 full-Moon widths — and are easily taken in at a glance.
How exact: The separation is good to about ten arcminutes — each position is checked against JPL Horizons on every build and agrees to under five — and the instant to a few hours. The pair looks equally close for a night either side.
- Closest separation
- 1.19°
- Elongation from the Sun
- 88° west (morning sky)
- Mars magnitude
- 0.7
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: A minimum of the apparent angular separation of the two planets, located by bisecting the derivative after a daily scan of the geocentric positions. (planet-pair-minimum-separation, version 1.0.0)
Uncertainty: The separation is good to about ten arcminutes — each position is checked against JPL Horizons on every build and agrees to under five — and the instant to a few hours. The pair looks equally close for a night either side.
Where it applies: The separation is geocentric; from the ground it changes by a fraction of a degree with your position, and whether the pair is above the horizon at that hour depends on where you are.
Leonids peak
Annual forecastUsually modest, but famous for spectacular storms roughly every 33 years when Earth crosses dense filaments of Tempel–Tuttle's debris. Meteors are the fastest of any major shower. The Moon is 55% lit on the peak night and will brighten part of the sky.
How exact: The peak night recurs annually and is reliable to about a day. The hour of maximum and the rate are not: observed rates routinely land at a fraction or a multiple of the nominal figure.
- Activity window
- 6 – 30 November
- Nominal ZHR
- 15 — the standardised rate for a radiant overhead in a perfectly dark sky, which no real site achieves. Expect fewer.
- Entry speed
- 71 km/s
- Parent body
- Comet 55P/Tempel–Tuttle
- Moon on the peak night
- 55% lit (computed)
Where this date comes from
Recurs every year at approximately the same point in Earth's orbit. The date shown is the expected one; the exact hour and the strength vary from year to year.
Source: IMO Meteor Shower Calendar working list
Uncertainty: The peak night recurs annually and is reliable to about a day. The hour of maximum and the rate are not: observed rates routinely land at a fraction or a multiple of the nominal figure.
Where it applies: Seen from both hemispheres, though the radiant's altitude — and so the rate — still depends on your latitude.
First Quarter Moon
ComputedHalf the Moon's disc is lit. It stands highest in the early evening and sets around midnight, leaving the second half of the night dark.
How exact: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the Moon's geocentric ecliptic longitude exceeds the Sun's by the phase angle, both referred to the mean equinox of date, located by bisection on the low-precision lunar and solar series. (lunar-phase-elongation-crossing, version 1.0.0)
Uncertainty: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where it applies: The instant is the same everywhere; only the local clock time and whether the Moon is above your horizon differ.
Mercury at greatest western elongation
ComputedMercury stands 20° from the Sun, its widest separation of this apparition, and is best placed in the morning sky before sunrise.
How exact: Within a few hours; the elongation is nearly stationary for days around the maximum, so the instant is much softer than the angle. The underlying positions are checked against JPL Horizons on every build and agree to under five arcminutes.
- Elongation
- 19.6° west of the Sun
- Approximate magnitude
- -0.5
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: A turning point of the true Sun–Earth–planet angle, located by bisecting the derivative after a daily scan; the sign of the ecliptic-longitude difference decides whether the apparition is eastern or western. (inferior-planet-elongation-extremum, version 1.0.0)
Uncertainty: Within a few hours; the elongation is nearly stationary for days around the maximum, so the instant is much softer than the angle. The underlying positions are checked against JPL Horizons on every build and agree to under five arcminutes.
Where it applies: How high the planet actually gets depends strongly on latitude and the season, because it follows the angle the ecliptic makes with the horizon.
Full Moon
ComputedThe Moon is opposite the Sun and fully lit, above the horizon all night. Faint objects are washed out for several nights either side.
How exact: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the Moon's geocentric ecliptic longitude exceeds the Sun's by the phase angle, both referred to the mean equinox of date, located by bisection on the low-precision lunar and solar series. (lunar-phase-elongation-crossing, version 1.0.0)
Uncertainty: Within about forty minutes of NASA’s published instant — measured, not asserted: the build checks every phase of 2026–2028 against NASA’s own table and the largest disagreement is thirty-eight minutes. The limit is the truncated lunar theory the whole platform uses, not the search.
Where it applies: The instant is the same everywhere; only the local clock time and whether the Moon is above your horizon differ.
Uranus at opposition
ComputedUranus lies opposite the Sun, rising as the Sun sets and visible all night. It is near its closest to Earth for this apparition and at its brightest.
How exact: Within a few hours. The platform’s planetary positions are checked on every build against JPL Horizons and agree to under five arcminutes, which at the rate a planet separates from the Sun is a few hours in the date. Opposition is a broad event in any case: the planet is barely changed in brightness or size for a week either side.
- Distance from Earth
- 18.439 au
- Approximate magnitude
- 5.6
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: The instant the planet's geocentric ecliptic longitude differs from the Sun's by exactly 180°, both referred to the mean equinox of date, from the JPL approximate planetary elements. (planet-solar-longitude-opposition, version 1.0.0)
Uncertainty: Within a few hours. The platform’s planetary positions are checked on every build against JPL Horizons and agree to under five arcminutes, which at the rate a planet separates from the Sun is a few hours in the date. Opposition is a broad event in any case: the planet is barely changed in brightness or size for a week either side.
Where it applies: Observable from anywhere the planet rises, which for these planets is most of the inhabited world.
Moon at perigee
ComputedThe Moon reaches the closest point of its orbit this month, appearing marginally larger than average — a difference of a few per cent in diameter, not something the eye registers without a side-by-side comparison.
How exact: The date is reliable; the hour is not. The distance curve is almost flat for a day either side, and unlike the phases and the equinoxes this one is not checked against an external table — no comparable published series is connected — so treat the time as indicative.
- Distance
- about 359,500 km, centre to centre
Where this date comes from
Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
Method: A turning point of the Earth–Moon distance from the low-precision lunar series, located by bisecting the derivative after a coarse scan. (lunar-apsis-distance-extremum, version 1.0.0)
Uncertainty: The date is reliable; the hour is not. The distance curve is almost flat for a day either side, and unlike the phases and the equinoxes this one is not checked against an external table — no comparable published series is connected — so treat the time as indicative.
Where it applies: A property of the orbit, the same for every observer.
Four kinds of date
- Computed
- Derived on this platform from published position series. The algorithm and its version are recorded, and the stated uncertainty is the real one.
- Published prediction
- Taken from an authority's own published prediction. AsteriaStar reproduces it and links the document; it does not recompute it.
- Annual forecast
- Recurs every year at approximately the same point in Earth's orbit. The date shown is the expected one; the exact hour and the strength vary from year to year.
- Planned
- A date somebody intends to meet. Planned dates move, often by months. The timestamp of the last confirmation is shown so you can judge how much the date is worth.
Take this away as a calendar file: subscribe or download the next year of events, or read it as JSON from the Open Data API. Both carry the basis, the method or source, and the uncertainty for every entry.
Provenance — gsfc:solar-eclipses
Forecast- Status
- Forecast
- Kind
- Forecast
- Provider
- NASA Eclipse Web Site
- Organisation
- NASA Goddard Space Flight Center
- Fetched
- 2026-11-03 10:30 UTC
- Cached for
- 7 d
- Treated as stale after
- 90 d
- Licence
- Public domain (US Government work), and NASA states that permission to reproduce the data is granted WHEN ACCOMPANIED BY an acknowledgment — “Eclipse Predictions by Fred Espenak (NASA’s GSFC)” for the solar catalogue and “Eclipse Predictions by Fred Espenak and Jean Meeus (NASA’s GSFC)” for the lunar one. The right acknowledgment travels with every eclipse AsteriaStar shows.
Source file: https://eclipse.gsfc.nasa.gov/SEcat5/SE2001-2100.html
Solar eclipses of the twenty-first century from NASA Eclipse Web Site, fetched from https://eclipse.gsfc.nasa.gov/SEcat5/SE2001-2100.html. The catalogue was computed once and published in 2007. A week is not a compromise between freshness and load — there is no freshness to trade, and re-parsing two hundred and twenty-four eclipses on every page view would be waste with no upside.
Circumstances of GREATEST eclipse only: the instant, the type, the Saros series, gamma, the magnitude, and the point on Earth the shadow axis passes closest to. It does not say what an eclipse looks like from any particular place, and neither does AsteriaStar — local circumstances need the Besselian elements, which NASA publishes separately and which are not read here.
Provenance — gsfc:lunar-eclipses
Forecast- Status
- Forecast
- Kind
- Forecast
- Provider
- NASA Eclipse Web Site
- Organisation
- NASA Goddard Space Flight Center
- Fetched
- 2026-11-03 10:30 UTC
- Cached for
- 7 d
- Treated as stale after
- 90 d
- Licence
- Public domain (US Government work), and NASA states that permission to reproduce the data is granted WHEN ACCOMPANIED BY an acknowledgment — “Eclipse Predictions by Fred Espenak (NASA’s GSFC)” for the solar catalogue and “Eclipse Predictions by Fred Espenak and Jean Meeus (NASA’s GSFC)” for the lunar one. The right acknowledgment travels with every eclipse AsteriaStar shows.
Source file: https://eclipse.gsfc.nasa.gov/LEcat5/LE2001-2100.html
Lunar eclipses of the twenty-first century from NASA Eclipse Web Site, fetched from https://eclipse.gsfc.nasa.gov/LEcat5/LE2001-2100.html. The same fixed catalogue as the solar table, on the same terms.
Circumstances of greatest eclipse and the published durations of the penumbral, partial and total phases. Contact times for a specific location are not in the table and are not derived here.
Provenance — ll2:upcoming-launches
Unavailable- Status
- Unavailable
- Kind
- Forecast
- Provider
- Launch Library 2
- Organisation
- The Space Devs
- Provider cadence
- 30 min
- Cached for
- 30 min
- Treated as stale after
- 6 h
- Licence
- Free public access as stated by the provider; attribution given on every surface that uses it.
Source file: https://ll.thespacedevs.com/2.2.0/launch/upcoming/?limit=40&mode=list
Upcoming orbital launches from Launch Library 2, fetched from https://ll.thespacedevs.com/2.2.0/launch/upcoming/?limit=40&mode=list. The provider allows fifteen requests an hour. Thirty minutes puts AsteriaStar at two, leaving the rest of the allowance for everyone else, and a launch schedule does not change meaningfully inside half an hour — when it does change, it changes by days.
Intentions, not appointments. Every date is a No Earlier Than value whose precision the provider states explicitly, from the second down to the year, and which moves without notice. The feed is an aggregation of operator announcements maintained by volunteers, not a schedule published by any agency.
Why there is no value: Launch Library 2 could not be reached: The operation was aborted due to timeout