The shortest day of the year is known as the winter solstice, and it falls on (or very close to) December 21. (For those in the southern hemisphere, the shortest day comes on or near June 21.)

But anyone who keeps a close eye on the time at which the sun rises and sets will notice something a bit perplexing: Neither the earliest sunset nor the latest sunrise falls on the day of the winter solstice. In fact, the earliest sunset can fall more than a week before the solstice, and the latest sunrise can fall more than a week after the solstice.

Of course, astronomers are well aware of this phenomenon, but it remains very difficult to explain in simple terms (and I admit I have some trouble wrapping my head around it myself, in spite of having written a book on “time”!).

The short answer is that the discrepancy is due to the earth’s orbital axis being tilted (this is what gives us the seasons), and also to the earth’s orbit being elliptical rather than circular.

Now for a more detailed analysis: The solstice, by definition, is the day on which the sun reaches its most southerly declination in the sky. If we measured time with a sundial – which gives “local solar time” (or “apparent solar time”) – the day of the solstice would also be the day of the earliest sunset and the latest sunrise. But we don’t -- we set our clocks to “mean solar time.” (Actually, we use UTC – Coordinated Universal Time – but for the purposes of this discussion, we can ignore the difference between this and mean solar time). Mean solar time is an average: Think of it as the time that a sundial would read, if the Earth’s axis wasn’t tilted with respect to the ecliptic (the earth’s orbit around the sun), and if its orbit was a circle rather than an ellipse. But of course it is tilted, and the orbit is an ellipse. The result is that clock time differs from local solar time. The difference varies over the course of a year, and at its maximum it can be quite large – as much as 15 minutes. The difference between local and mean solar time is called the “equation of time,” and it depends on one’s location. As a result of these two effects, the time of sunset as read by a clock (i.e. as measured via mean solar time) is at its earliest more than a week before the solstice, and, similarly, the day with the latest sunrise comes more than a week after solstice.

Although less immediately noticeable, the time at which the sun reaches its greatest altitude in the sky also varies over the course of the year. If we kept to local solar time, of course, “noon” would just be half-way between sunrise and sunset. But, since we keep mean solar time, the sun may reach its high point anywhere from 15 minutes before, or after, 12:00.

Thanks to physicist David Tindall of Dalhousie University for some helpful comments, and to my friend Maria for getting the discussion started.