Where the Moon Is in Its Apsidal and Nodal Precession Cycles
The Moon’s orbit doesn’t hold still. Its long axis swings around once every 8.85 years, and the plane it orbits in swings around once every 18.6 years. Here’s how to find exactly where both cycles stand right now, and the formulas behind the numbers.
Every source on the Moon’s orbit will tell you the two precession periods: about 8.85 years for the line of apsides (the axis running through perigee and apogee) and about 18.6 years for the line of nodes (the line where the plane of the Moon’s orbit intersects the ecliptic; the ascending node is the point on that line where the Moon crosses from south to north). What’s harder to find is where in those cycles the Moon actually is on a given day — the phase, not just the period.
The Moon’s mean longitude of perigee and mean longitude of ascending node change every day. The calculator below computes both for any date, including today.
Why “mean” elements, not the Moon’s actual position
If you look up the Moon’s instantaneous orbital elements — from JPL Horizons, for instance — you’ll find the perigee and node angles jumping around noticeably from week to week. That’s not measurement error. It’s the Sun’s gravity constantly tugging the Moon’s orbit into a slightly different shape, on top of the slow, steady precession. Those instant-by-instant values are called osculating elements, and they’re the ellipse that best matches the Moon’s motion at that exact moment.
Mean elements strip that short-term wobble out. They’re smooth, predictable functions of time — the average path the perigee and node are steadily tracing, ignoring the day-to-day noise. If you’re modeling the precession cycles themselves, rather than the Moon’s exact position on a specific date, mean elements are what you want: a clean angle for any input date, with no jitter to fight.
Where the cycles begin
Neither precession cycle has a physically meaningful “beginning” — these are angles on a circle, not oscillations with a defined phase zero. What astronomers actually specify is the current orientation of the node and perigee, as mean longitudes measured from the mean equinox of the date — the reference direction shifts slightly over time as Earth’s own axis precesses, independent of anything the Moon is doing. The formulas below use J2000.0 (January 1, 2000, 12:00 Terrestrial Time) as the reference epoch, T = 0, for the polynomial — not as a fixed reference frame the angles stay pinned to. A model seeded with today’s angle, using today’s T, will be correctly phased relative to the mean-element model; comparing it directly against a live ephemeris like JPL Horizons will still show the small periodic wobble described above.
Reference periods
| Nodal precession period | 18.60 years (6798.38 days) |
| Apsidal precession period | 8.85 years (3232.61 days) |
Sources: Apsidal precession, Lunar precession — both citing the Explanatory Supplement to the Astronomical Almanac.
Live calculator
Pick any date to get the mean node and perigee longitudes for that day, plus roughly how many days remain until each next crosses 0°.
Mean ascending node (Ω)
–
Mean longitude of perigee (Π)
–
These two angles define the current positions within the nodal and apsidal precession cycles, measured from the mean equinox of the date.
Julian centuries since J2000.0 (T): –
The formulas
These are Jean Meeus’s mean-element polynomials (Astronomical Algorithms, 2nd ed.), the same ones behind the calculator above. T is Julian centuries measured from J2000.0 (JDE 2451545.0):
L' = 218.3164477 + 481267.88123421*T - 0.0015786*T^2 + T^3/538841 - T^4/65194000 (mean lunar longitude)
M' = 134.9633964 + 477198.8675055*T + 0.0087414*T^2 + T^3/69699 - T^4/14712000 (mean anomaly)
Om = 125.0445479 - 1934.1362891*T + 0.0020754*T^2 + T^3/467441 - T^4/60616000 (mean ascending node)
Perigee longitude: Pi = L' - M' (mod 360)
Node longitude: Omega = Om (mod 360)These were checked against the PyMeeus library implementation and agree to within a few ten-thousandths of a degree — a few arcseconds, or well under a second of timing error translated into angle. The small remaining gap comes from this calculator using plain calendar UT for T, rather than dynamical time (JDE), which includes a small correction that’s currently a little over a minute and grows slowly. That’s negligible for any physical build, but worth knowing if you need to match an ephemeris to sub-arcsecond precision.
A caveat for physical models
If you’re building something that has to run on its own — a clockwork orrery in the tradition of the Antikythera mechanism, a fixed-rate simulation, anything without a way to recalibrate — note that the true precession rate isn’t perfectly constant. It drifts slowly over centuries, which is what the T² and higher-order terms above capture. A mechanism using only the linear (T¹) rate will track the real cycle closely for a decade or two, but will drift steadily beyond that. If the model needs to stay accurate indefinitely, it needs either the higher-order terms built in or a periodic recalibration point.