A Moon that rises further north than the Sun

The Sun rises in the north-east in summer and the south-east in winter, and the extremes are the solstices. They are the same every year, to within the slow drift of the Earth’s tilt, because the Sun moves along the ecliptic and the ecliptic is tilted to the equator by the same twenty-three and a half degrees each year. The Sun’s greatest declination north and south of the equator is the obliquity, and every midsummer sunrise is in the same place.

The Moon moves close to the ecliptic but not on it. Its orbit is tilted by a little over five degrees, so in the course of a month it goes up to five degrees above the ecliptic and five below, and its declination swings north and south through a range that differs from the Sun’s by up to those five degrees either way. Which way depends on where the tilt points. And the tilt’s direction is not fixed: the line along which the Moon’s orbit crosses the ecliptic, the line of nodes, turns backward round the ecliptic once every 18.61 years, dragged by the Sun’s pull on the Moon’s tilted orbit. The same regression drives the largest wobble of the Earth’s axis, and it shows up on the horizon as a slow opening and closing of the Moon’s range.

A range that opens and closes #

When the Moon’s ascending node lies at the vernal equinox, the Moon’s orbit tilts its northern half towards the part of the sky where the ecliptic is already furthest north, and the two tilts add. Each month the Moon then reaches a declination of about twenty-three and a half plus five: 28.6 degrees, north and then, a fortnight later, south. When the node has turned half way round, nine years later, the tilts subtract, and the Moon’s monthly extremes are only 18.3 degrees.

The two extremes are the lunar standstills: major, when the Moon’s range is widest, and minor, when it is narrowest. The name comes from the shape of the curve near its turning points. The monthly extreme follows the cosine of the node’s longitude, and a cosine is flat at its peaks, so for about two years either side of each standstill the Moon’s range hardly changes from month to month: its northernmost rising point seems to stand still on the horizon, as the Sun’s does for a few days around a solstice. The word is the same and so is the mathematics, on a timescale three hundred times longer.

The last major standstill fell in early 2025, and the Moon’s range was close to its widest from the middle of 2024 to the end of 2025. The next is in 2043–44. The minor standstill falls between them, in 2034.

A standstill that flutters #

The node’s regression is not the only thing that changes the Moon’s orbit. The Sun’s tide also changes its inclination, by about a tenth of a degree either way, on a cycle tied to the Sun’s position relative to the node: half an eclipse year, 173 days. Near a standstill, where the slow cycle is flat, that small fast one becomes visible.

The greatest reach of the 2024–25 standstill did not come at one moment. It came as a series of peaks half an eclipse year apart, each a little higher or lower than the smooth curve, and the very largest monthly extreme was set by where the fast cycle happened to peak relative to the slow one. That is why the standstill is properly described as a season of a year and more rather than a date, and why the Moon’s most extreme declination of a given standstill can be predicted only with the full theory of its motion, not from the node alone. The 173-day period is the same one that spaces eclipse seasons, for the same reason: both are set by when the Sun lines up with the Moon’s nodes.

How far north the Moon can rise #

On the horizon, the Moon’s declination becomes a rising direction. The further north a body’s declination, the further north of east it rises, and the effect grows with the observer’s latitude, because the horizon is more steeply tilted to the celestial equator.

At the latitude of southern England, around fifty-two degrees, the midsummer Sun rises about fifty degrees east of north. At the major standstill the Moon at its northern extreme rises nearly forty degrees from north — eleven degrees further round the horizon than the Sun ever reaches — and at its southern extreme a fortnight later it rises the same distance south of the Sun’s midwinter rising point. At the minor standstill the Moon’s extremes fall nine degrees inside the Sun’s. Over the 18.6 years the Moon’s monthly rising points sweep out a band on the horizon twenty degrees wide beyond each of the Sun’s limits.

At higher latitudes the separation grows. Beyond 61.4 degrees north, at a major standstill, the Moon at its northern extreme does not set at all and at its southern extreme does not rise.

The full Moon makes the cycle easy to see without a compass. A full Moon is opposite the Sun, so in midwinter it sits where the Sun stands at midsummer, high in the northern part of the sky, and in midsummer it sits low where the midwinter Sun does. Near a major standstill the midwinter full Moon goes further: five degrees higher than the midsummer Sun ever climbs, the highest full Moon of the generation, riding through the sky for most of a sixteen-hour winter night. The midsummer full Moon of the same years is correspondingly the lowest, barely clearing the southern horizon in northern Europe for a few hours of the short night. At the minor standstill both are muted: the winter full Moon climbs less high than the summer Sun and the summer one sits higher than the winter Sun. Anyone who notices the height of the full Moon on the longest nights over twenty years has observed the node’s regression with no instrument at all.

A Moon that skims the hills #

At Callanish, on the Isle of Lewis at fifty-eight degrees north, the effect becomes a spectacle.

At the southern extreme of a major standstill month, the Moon seen from Callanish crosses the meridian only three degrees above the horizon — six Moon-widths. It rises in the south-south-east, stays above the horizon for under four hours, and sets in the south-south-west, rolling along the line of hills to the south. The stones of Callanish are aligned so that, from within the circle, the Moon at the major standstill appears to set into the hills and re-emerge among the stones, and whether that was intended by their builders five thousand years ago is one of the most argued questions in the archaeology of astronomy. The alignment is real; the intention is an inference. At the minor standstill the same Moon culminates at fourteen degrees, unremarkably.

The argument about intention is old and has not been settled. In the 1960s and 1970s the engineer Alexander Thom surveyed hundreds of British megalithic sites and proposed that many were laid out to mark the Moon’s extreme rising and setting points to an accuracy of minutes of arc, implying an observational programme spanning generations. Later statistical re-analyses found the high-precision claims unsupported: the sites’ horizons, the Moon’s parallax and refraction, and the choice of which of many possible alignments to count leave too much freedom. Low-precision orientations towards the standstill directions, a few degrees wide, survive at some sites, and the major standstill of 2024–25 was the occasion for new, systematic observations at Callanish and Stonehenge designed to test them.

Stonehenge, seven degrees further south #

At Stonehenge, seven degrees south of Callanish, the same cycle is less dramatic and has been argued about longer.

The southernmost Moon of a major standstill never gets far above the horizon even here: ten degrees at its highest, up for a little over six hours, rising and setting close to south-east and south-west. The monument’s principal axis points to the midsummer sunrise and the midwinter sunset. Four stones outside the main circle, the Station Stones, form a rectangle whose sides have been read as pointing to the Moon’s extreme rising and setting directions at the major standstill, and the rectangle’s shape — nearly, but not exactly, square — is what those directions and the solstitial ones would produce at Stonehenge’s latitude and no other. Whether that is design or coincidence turns on the precision with which the stones were placed and on which of the many possible lines are counted, and it was among the questions that the observations of 2024 and 2025 were organised to revisit.

The difference between the two sites is a matter of latitude alone: the same Moon, at the same declination, culminates seven degrees higher seven degrees further south. A monument that marks the Moon skimming the horizon has to be far north, and Callanish is.

Where the Moon’s tilt comes from #

The five-degree tilt of the Moon’s orbit that makes the whole cycle possible is itself a puzzle. A Moon formed from a disc of debris round the young Earth, as the giant-impact picture has it, would have formed in the plane of the Earth’s equator and should have stayed close to the ecliptic as tides pushed it outward; the inclination it has is larger than that history naturally gives, and proposals to supply it range from resonances crossed early in the Moon’s migration to encounters with leftover planetesimals. Without the tilt there would be no standstills, no eclipse seasons and no eclipses at only some new and full Moons rather than every one — the reason, as phases are not shadows, why most full Moons pass above or below the Earth’s shadow.

The eclipse connection runs both ways. Eclipses happen only when the Sun is near one of the Moon’s nodes, and the node’s regression makes those seasons come about nineteen days earlier each year, so that the eclipse year is 346.6 days long; the eclipse that repeats a third of a world away after a saros is the same arithmetic of node, month and year. The standstill cycle is the eclipse cycle’s slow companion: the eclipse seasons mark where the node is; the standstills mark which way it points.

A month with days of no Moon #

In the Arctic the standstill cycle produces a lunar counterpart of the midnight sun.

North of the Arctic Circle the Sun stays above the horizon for weeks in summer and below it for weeks in winter. The Moon does the same within each month, because it covers the same range of declination in four weeks that the Sun covers in a year: when its declination is far enough south it does not rise, and a fortnight later it does not set. How long those stretches last depends on the standstill cycle. At seventy degrees north, at a major standstill, the Moon stays below the horizon for about a week of every month and above it for another week; at a minor standstill its range never reaches far enough to do either. A town in northern Norway sees, once a generation, Moons that stay up for a week, and nine years later sees none.

The same regression, read elsewhere #

The node’s regression is one of the most pervasive rhythms in the Earth–Moon system, and the standstill is only its most visible form. It modulates the Earth’s axis in the 18.6-year nutation, which makes the celestial pole trace a small ellipse round its precessional path and shifts every star’s catalogued position by up to seventeen arcseconds. It modulates the tides: the diurnal tides are strongest when the Moon’s declination is largest, so at the major standstill they are larger by tens of per cent and the semidiurnal ones slightly smaller — the correction that every harmonic prediction has to carry, as a tide table is astronomy with the harbour fitted. It sets the eclipse cycles, since eclipses occur only near the nodes, and the regression is why eclipse seasons arrive about nineteen days earlier each year.

And it reaches down to the climate record. The 18.6-year tidal cycle has been sought in sea levels, in Arctic sea ice and in ocean mixing, because the strength of the tides that stir the ocean varies with it; the reported signals are small and their interpretation is debated, but the forcing is real and exactly periodic. The precession of the Earth’s own axis over twenty-six thousand years slowly moves the pole star; the precession of the Moon’s orbit over eighteen and a half years slowly moves the Moon’s horizon. Both are the Sun and Moon pulling on tilted, spinning or orbiting things.

Why the Moon’s orbit turns #

The regression itself has a simple cause. The Moon’s orbit is tilted to the plane of the Earth’s orbit, so for most of the month the Moon is above or below the line joining the Earth to the Sun. The Sun’s tidal pull on the Moon has a component that tends to pull the Moon’s orbit towards the ecliptic, a torque on a tilted ring. A spinning top responds to a torque by precessing rather than falling, and an orbit, which carries angular momentum, does the same: its plane turns about the ecliptic’s pole rather than flattening into it. The rate is set by the ratio of the Moon’s month to the year — the Sun’s tide acting over a year on an orbit that turns over in a month — and comes out at one revolution in about eighteen and a half years, backwards, because the torque on an orbit tilted this way drives the node westward.

The same calculation, applied to the Earth’s equatorial bulge rather than to the Moon’s orbit, gives the twenty-six-thousand-year precession of the equinoxes. The bulge is a ring of extra mass tilted to the ecliptic; the Sun and Moon pull on it; it precesses. The standstill and the slow march of the pole are the same physics applied to two different rings.

Parallax, refraction and a horizon that is not flat #

The figures use a simplified Moon: its extreme declination as obliquity plus inclination times the cosine of the node, a single fast term for the solar perturbation of the inclination, and a flat horizon with no refraction or parallax. The Moon’s parallax lowers its apparent position by up to a degree near the horizon, and refraction, which lifts the setting Sun by half a degree, raises it again by about half a degree, leaving a net lowering of about a third of a degree, which moves its rising point by a degree or two at northern latitudes — a correction as large as the 173-day flutter and essential for any claim about a monument’s alignment. The obliquity itself is drifting by about half an arcsecond a year. And the Moon is a disc half a degree across: “rising” may mean the first gleam of its upper edge or the moment its centre clears the horizon, and those differ by minutes of time and a fraction of a degree of azimuth. The cycle is exact; its appearance on a real horizon needs all of these.

Still open: what the builders of the stones were watching #

The astronomy of the standstill is settled; what it meant to the people who raised stones aligned to it is not. The cycle is long — a person in a society without writing might see three or four major standstills in a lifetime — and an alignment accurate enough to mark it implies observations sustained and transmitted across generations. The test that would settle the question for a particular site is statistical and archaeological rather than astronomical: whether the monument’s orientations are concentrated on the standstill directions more than chance and the local topography would produce, and whether the dates of its construction cluster near standstills. The observations made during the 2024–25 standstill, at sites from Scotland to the American Southwest, are the first to record the Moon’s actual behaviour on each horizon with modern precision. If the alignments hold at the precision the horizon allows, the eighteen-and-a-half-year cycle was known long before anyone could write it down. If they hold only loosely, the stones mark the Moon’s general direction and the precision was added later, by the people who surveyed them.