On Mercury, Sunrise to Sunrise Takes Two Mercury Years
NASA lists three numbers for Mercury. It goes around the Sun once every 88 Earth days. It spins once on its axis in about 59 Earth days. And one “solar day”, a full cycle from one sunrise to the next, equals 176 Earth days [2]. The more exact values in NASA’s fact sheet are 87.969 days for the orbit and 4,222.6 hours (about 175.9 days) for the day [3].
So if you could stand on the surface and wait for the Sun to rise, then rise again, the planet would have gone around the Sun twice in between. A “year” is how long a planet takes to circle the Sun. A “day”, for someone standing on the surface, is how long the Sun takes to come back to the same place in the sky. On Mercury the second takes about twice as long as the first.
Notice there are two different clocks here. The 59-day spin is measured against the distant stars. The 176-day day is measured against the Sun. They differ on Earth too, but only by about four minutes. On Mercury the gap is enormous, and a later section shows why.
For Decades, Astronomers Thought Mercury Always Shows the Sun One Face
The Moon always keeps the same face toward Earth, which is why we never see its far side. Many astronomers believed Mercury did the same thing with the Sun: one side in permanent daylight, the other in permanent night. The astronomer Kenneth Kellermann, who worked at the Parkes telescope in Australia, remembers that every astronomy textbook of the early 1960s said so. One of them described the sunlit side as scorching and the other side as “in eternal darkness”, with a temperature “not far from absolute zero” [1].
He traces the idea back to 1888, to drawings of faint surface markings made by the Italian astronomer Giovanni Schiaparelli, who judged from a few observations that Mercury’s spin and orbit matched [1]. Mercury is hard to study. It never strays far from the Sun in our sky, so it is only well placed for viewing for short spells.
There is a neat reason the mistake lasted. From Earth, Mercury is best seen about once every 116 days (its “synodic period”, 115.88 days in NASA’s tables [3]). Mercury spins just under twice in that time, by my own arithmetic from the 58.6-day spin. Each good viewing therefore shows nearly the same face to us, which looked like a planet that never turned. Kellermann suggests this is how the early observers were fooled [1].
A Warm Night Side and a Radar Echo Showed That Mercury Does Spin
Kellermann’s measurement at Parkes was the first clue. He listened to the faint radio waves that every warm surface gives off. When only the supposedly frozen dark side faced Earth, the signal came in “loud and clear”, as if the surface were about 300 kelvin (roughly 27 °C or 80 °F), close to room temperature. As the planet moved on and more of the sunlit side showed, the reading hardly changed. Day side and night side seemed to have about the same temperature [1].
He tried to explain it with an atmosphere that carried heat to the dark side, since the textbooks were so clear that Mercury did not rotate. He later wrote that the true explanation arrived a few months afterward: radar measurements showed that Mercury does rotate [1].
Those measurements were made in 1965 by Gordon Pettengill and Rolf Dyce at the Arecibo radio telescope in Puerto Rico. They bounced radar pulses off Mercury. If a planet is turning, the edge moving toward us returns echoes at a slightly higher frequency and the edge moving away at a slightly lower one, so the spread of frequencies reveals the spin. They found a period of about 59 days, not 88 [4]. That is two thirds of the orbital period [1].
Once the number was known, theorists, including Giuseppe Colombo, were quick to point out that it was close to what theory allowed. Kellermann notes that, until the radar results, they had apparently been comfortable with an 88-day match [1].
Three Spins in Two Orbits Add Up to One Day Lasting Two Years
Here is the key fact: Mercury turns exactly three times for every two trips around the Sun. Scientists call this a 3:2 spin-orbit resonance. Later spacecraft measurements, including data from NASA’s MESSENGER mission, confirm that Mercury is in this 3:2 state [5].
To see why this stretches the day, count sunrises. A planet that spins three times has had three chances for the Sun to come around. But each trip around the Sun makes the Sun appear to slide backward in the sky by one full turn. Over two orbits, that cancels two of the three. Three spins minus two orbits leaves one sunrise. One sunrise per two orbits means one day lasts two years: 2 × 88 is 176 [2].
The same rule works on Earth, where the spin is a little over 365 turns a year but the day count is one fewer. On Mercury, the spin is so slow that the subtraction takes away most of it.
Why Mercury Did Not Settle Into Facing the Sun Like the Moon Faces Earth
If the Moon can end up showing one face, why not Mercury? A big part of the answer is the shape of Mercury’s orbit. It is a stretched circle: Mercury swings from about 46 million kilometers (29 million miles) from the Sun to about 70 million (43 million) [3]. NASA lists its orbital eccentricity, a measure of how far an orbit is from a perfect circle, as 0.2056 [3].
One research team simulated how tides from the Sun slow Mercury’s spin. With an eccentricity of 0.206, they report, being caught in the 3:2 state is the most probable final outcome, and it happens for good once it has happened. They also found that the capture can happen within about 10 to 20 million years, much faster than earlier estimates [6].
A simple way to picture it is that near the closest point of the orbit, Mercury races past the Sun and the Sun’s pull on it is strongest, so the pull can’t settle on matching a single steady orbital pace. This picture is mine, not a quotation from the research. The research itself is about the full tidal mathematics.
Try Mercury’s 3:2 Walk Around a Lamp or Chair in Your Own Room
You can feel the difference between the Moon’s way and Mercury’s way with your own feet. You need a clear, open patch of floor. Put a chair in the middle (or a lamp on a table, as long as you do not touch the bulb and nothing is in your way).
The Moon way. Walk slowly in one circle around the chair, keeping your nose pointed at it the whole time. Count how many times you turned around to face the same wall of the room. You will find you turned exactly once, yet the chair never left your view.
Mercury’s way. Walk the circle in eight stops, four for each lap, two laps in total. At each stop turn on the spot, in the same direction that you are walking, by about three eighths of a full turn (like moving a clock hand from 12 to about 4:30). After eight stops you have circled twice and turned three times. Watch where the chair appears in front of you. At the start it is straight ahead. It slides a bit further around at each stop, and it takes all eight stops, two full laps, before it is straight ahead again. That is one Mercury day. Move slowly and keep the chair at arm’s length or farther; if you feel dizzy from turning, stop and sit down. This is a simplified model, not Mercury’s real motion.
Sources
I read Kellermann’s account and the two NASA pages in full. The Nature papers by Pettengill and Dyce (1965) and by Colombo (1965) are behind a paywall and I could not open them, so the 1965 radar result is described from Kellermann’s paper and a Wikipedia biography page, and Colombo’s work only as Kellermann reports it. For the two research papers on the 3:2 state I read only the published abstracts. The explanation of why Mercury was not locked is therefore stated at abstract level only. The sunrise counting, the “just under twice” figure and the walk-around-a-chair activity are my own arithmetic and pictures, not claims from the sources.
- K. I. Kellermann, “Early Parkes Observations of Planets and Cosmic Radio Sources,” conference paper, Science with Parkes @ 50 Years Young, 2011. https://arxiv.org/abs/1210.0984 (textbook belief, Schiaparelli 1888, night-side radio temperature near 300 K, the 59-day radar result, theorists’ response, and the 3:2 reason for the mistaken view from Earth; first-hand account, read in full)
- NASA Science, “Mercury Facts.” https://science.nasa.gov/mercury/facts/ (88-day orbit, 59-day rotation, 176-day solar day)
- NASA Space Science Data Coordinated Archive, “Mercury Fact Sheet.” https://nssdc.gsfc.nasa.gov/planetary/factsheet/mercuryfact.html (87.969-day orbit, 1,407.6-hour rotation, 4,222.6-hour day, 115.88-day synodic period, eccentricity 0.2056, distances from the Sun)
- G. H. Pettengill and R. B. Dyce, “A Radar Determination of the Rotation of the Planet Mercury,” Nature 206, 1240 (1965). https://www.nature.com/articles/2061240a0 and Wikipedia, “Gordon Pettengill.” https://en.wikipedia.org/wiki/Gordon_Pettengill (Arecibo radar pulses, 59 days rather than 88; the Nature paper itself could not be opened, the Wikipedia page does not give the year, and the explanation of the frequency spread is the standard Doppler idea, not taken from these pages)
- A. Stark, J. Oberst and H. Hussmann, “Mercury’s Resonant Rotation from Secular Orbital Elements,” arXiv:1506.00008 (2015). https://arxiv.org/abs/1506.00008 (Mercury in a stable 3:2 spin-orbit resonance, using ephemerides that include MESSENGER observations; abstract only)
- B. Noyelles, J. Frouard, V. Makarov and M. Efroimsky, “Spin-orbit evolution of Mercury revisited,” Icarus 241, 26-44 (2014). https://arxiv.org/abs/1307.0136 (3:2 capture most probable at eccentricity 0.206, 10 to 20 million years; abstract only)
Update history
- First published.