It’s ten past four, nearly sunset. Leonard Tremiel informed me the other day that the earliest sunset is actually two weeks before the solstice, and the latest sunrise two weeks after it. He recommended Earth and Sky’s explanation.
It seems paradoxical. At middle latitudes in the U.S. – and throughout the Northern Hemisphere – the earliest sunsets of the year come about two weeks before the solstice and the shortest day of the year.
Why isn’t the earliest sunset on the year’s shortest day? It’s because of the discrepancy between the clock and the sun. A clock ticks off exactly 24 hours from one noon to the next. But an actual day – as measured by the spin of the Earth, from what is called one “solar noon” to the next – rarely equals 24 hours exactly.
Solar noon is also called simply “midday.” It refers to that instant when the sun reaches its highest point for the day. At this time of year, the time period from one solar noon to the next is actually half a minute longer than 24 hours. Today, on December 7, the sun reaches its noontime position at 11:52 a.m. local standard time. Two weeks later – on the winter solstice – the sun will reach its noontime position around 11:59 a.m. That’s 7 minutes later than today.
The later clock time for solar noon also means a later clock time for sunrise and sunset. The table below helps to explain.
For Philadelphia, Pennsylvania
Date Sunrise Solar Noon (Midday) Sunset Daylight Hours December 7 7:09 a.m. 11:52 a.m. 4:35 p.m. 9 hours 26 minutes December 21 7:19 a.m. 11:59 a.m. 4:39 p.m. 9 hours 20 minutes
Cool huh? Read the rest there, complete with sunset photos.
This is pure sloppiness on somebody’s part, probably the head of the Naval Observatory or the Astronomer Royal.
Which is why every sundial worth having has a chart of the equation of time attached to it so you can better calculate clock time.
Nobody notices because sunrise and sunset change daily, gradually at lower lattitudes (less than 45°N or S), about a minute a day or less. At far north lattitudes (more than 60°N or S), it’s more than two minutes per day, and the variations are really noticeable, including the one you mentioned about the solstice.
A year isn’t even a year around the sun, it’s 365.241 (5h 47m) days, hence why there’s no leap year per century and there *is* a leap year every 4 and 400 (241 per thousand years). But near enough is good enough, both on the clock and the calendar. It’s happenstance that the Earth’s rotation and orbit around the sun are similar enough.
The table helps me feel it: 1) December 7 and 21, sunrise and sunset occur at solar noon -/+ 283 and 280 minutes respectively. I can pause and feel the symmetry of sunrise and sunset around solar noon makes sense. 2) Solar noon moves relative to our clocks. That point in isolation feels fine with me too.
You might like this page about sunsets.
And Lofty, thank you for the animated GIF in your link! I love how the colormap relates the four curves (crossfading RGB from red to green to blue and back to red). I’ve used that colormap on surfaces, but not on curves; I’ll remember it.
I like this line: the French defined legal time as Paris Mean Time minus nine minutes and 21 seconds.
It reminds me of a Woody Allen line: Somewhere in the universe there’s a civilization more advanced than ours by fifteen minutes, and this gives them a great advantage because they don’t need to rush to make appointments.