Why the Earliest Sunset Is Not on the Shortest Day
In London it is about eight days early. Near the equator, closer to three weeks. The size of the gap depends on where you stand.
Every December, somebody notices that the evenings started drawing out before the shortest day arrived, and assumes they have made a mistake. They have not. The earliest sunset genuinely comes first, the latest sunrise comes later, and the solstice sits between them.
The reason is well covered elsewhere. What is not covered anywhere is that the size of the gap depends on your latitude, and it varies enormously.
| City | Latitude | Earliest sunset | At | vs 21 December |
|---|---|---|---|---|
| Tromsø | 69.6°N | 27 Nov | 11:42 | 24 days earlier |
| Reykjavík | 64.1°N | 18 Dec | 15:28 | 3 days earlier |
| Stockholm | 59.3°N | 16 Dec | 14:46 | 5 days earlier |
| London | 51.5°N | 12 Dec | 15:51 | 9 days earlier |
| New York City | 40.7°N | 7 Dec | 4:28 pm | 14 days earlier |
| Cairo | 30.1°N | 2 Dec | 16:54 | 19 days earlier |
| Mumbai | 19.1°N | 24 Nov | 5:58 pm | 27 days earlier |
| Bogotá | 4.6°N | 8 Nov | 17:37 | 43 days earlier |
| Nairobi | 1.3°S | 6 Nov | 18:21 | 45 days earlier |
What is going on
Two clocks are running, and they disagree.
Apparent solar time is what a sundial reads. Noon is when the sun crosses your meridian, and the interval between one solar noon and the next is a real solar day.
Mean solar time is what your watch reads. It assumes every day is exactly 24 hours long.
They disagree because the real solar day is not constant. Two effects cause this:
The Earth's orbit is elliptical. We move faster when closer to the sun, in early January, and slower in July. Faster orbital motion means the Earth has to turn a little further each day to bring the sun back to the meridian, so solar days near perihelion are longer than 24 hours.
The Earth's axis is tilted. The sun's apparent motion is along the ecliptic, not the equator, and only the component parallel to the equator affects when it crosses your meridian. Near the solstices that component is at its largest, which again lengthens the solar day.
In early December both effects push the same way, and solar days run about 30 seconds longer than 24 hours. That means solar noon drifts about 30 seconds later each day — and so does everything tied to it, including sunset.
The tug of war
Around the December solstice in the northern hemisphere, two things happen at once:
- The day is still getting shorter, which pulls sunset earlier and sunrise later.
- Solar noon is drifting later, which pushes both sunset and sunrise later.
Sunset is the sum of the two. In early December, the drift in solar noon is winning, so sunset stops getting earlier and starts getting later — while the day is still shortening, because sunrise is being pushed later by both effects at once.
The solstice is the day the first effect reverses. Sunrise keeps getting later for another two weeks after it, until the drift finally loses.
This is why the winter solstice feels like a turning point that has already happened. In London the evenings have been drawing out for over a week by the time the shortest day arrives, and it is the mornings that keep getting worse into January.
Why latitude changes the gap
Here is the part the other explanations leave out.
The equation of time is the same everywhere on Earth — it is a property of the orbit and the tilt, not of your location. In early December it shifts solar noon by about 30 seconds a day regardless of where you are.
What does depend on latitude is how fast the day length is changing. Near the solstice, at high latitude, day length changes rapidly; near the equator it barely changes at all.
So the tug of war is uneven. At 60 degrees north, the day-length effect is strong and quickly overwhelms the 30-second drift, so the earliest sunset falls only a few days before the solstice. Near the equator the day-length effect is feeble, the drift dominates for weeks, and the earliest sunset can precede the solstice by a fortnight or more.
Read the table from the bottom up and you can watch the gap close as latitude rises.
The southern mirror
Everything above has a June counterpart in the southern hemisphere, with one asymmetry: the equation of time behaves differently in June than in December, because perihelion is in January.
Around the June solstice the two effects work against each other rather than together, so the drift is smaller. The result is that the southern hemisphere's version of this — earliest sunset before the June solstice — is real but weaker.
There is also a northern-summer version. The latest sunset of the year in the northern hemisphere comes a few days after the June solstice, for the same reason in reverse. In London it falls around 25 June.
Checking it for your own city
Every city page on this site carries the earliest and latest sunset of the year with their dates, alongside the solstices, in the year-overview section. The dates are computed rather than assumed, by solving each day and finding the extreme, so they are correct for that specific place rather than inherited from a reference city.
If you want to watch the effect happen, the monthly calendar table on each city page shows sunset for every day of the month with the daily change. Through early December the change column runs to zero and reverses while day length is still shrinking — the whole phenomenon in one column of numbers.
The name for it
The graph of the difference between apparent and mean solar time is called the equation of time, in the old sense of "equation" meaning a correction to be applied rather than a statement of equality. Plot the sun's position at the same clock time every day for a year and you trace a figure-eight in the sky called an analemma, which is the same information drawn differently.
Both are worth looking up if you want the full picture. The practical version fits in a sentence: the sun runs a little fast or slow against your watch depending on the season, and in December that is enough to move sunset before the day itself has stopped shrinking.