Mars Time

Start with the day, because everything else hangs off it. A Martian solar day, the interval from one local noon to the next, averages 24 hours 39 minutes and 35.244 seconds. It has its own word, sol, precisely so that nobody writes "day" and leaves the reader guessing which planet's day is meant. That is not the same as the time Mars takes to turn once on its axis, which is 24 hours 37 minutes 22.663 seconds, the sidereal day. The two differ by 132.58 seconds, and the reason is the same on Mars as on Earth: while the planet turns, it is also moving along its orbit, about 0.524 degrees a day, so a single full rotation leaves the Sun slightly short of the meridian and the planet has to turn a bit further to catch it. On Earth that gap is just under four minutes, which is why the stars rise four minutes earlier each night. Both Martian numbers come from the same place, Allison and McEwen's 2000 paper in Planetary and Space Science, which took the Mars pole and spin from radio tracking of the Viking landers and Mars Pathfinder and turned them into timing recipes precise to a fraction of a second. NASA's own SPICE navigation constants encode the spin as 350.891982443297 degrees per day, which reproduces the sidereal figure to a thousandth of a second. If you want a single number to carry away, carry the sol: 24 h 39 m 35 s.

Now the harder half. There is no Martian UTC. Mean solar time on the Martian prime meridian has a name, Airy Mean Time, and the name Coordinated Mars Time circulates for a civil scale built on it, but no body has defined or adopted one and nobody has defined Martian time zones, so in practice each place on Mars keeps its own local mean solar time: Airy Mean Time plus one Mars hour for every 15 degrees of east longitude. Gale crater, where Curiosity works, is at 137.4 east, about 9 h 10 m ahead of the prime meridian. Jezero, where Perseverance works, is at 77.45 east, about 5 h 10 m ahead. The two rovers are four Martian hours apart, and a rover clock says nothing useful about conditions at the other site. Worse, a mission clock is not even exactly LMST at the place the spacecraft ended up: each was set to the longitude the mission planned to reach. NASA GISS lists the leftovers. Perseverance and Curiosity are out by about 5 seconds, InSight by about 83 seconds, and Phoenix, retargeted in flight, by about three and a half minutes. Spirit and Opportunity are odder still: their clocks were deliberately aligned with true solar time at landing while ticking at the mean rate, which left each of them running roughly 40 minutes behind local mean solar time at its own site for the rest of its life. Even the counting differs. Viking, Phoenix, Curiosity, InSight and Perseverance call the landing sol Sol 0; Pathfinder, Spirit and Opportunity call it Sol 1.

For anything that has to be compared across missions and decades, mission sols are useless, so there is a second, quieter clock. The Mars Sol Date is a plain running count of sols with no spacecraft attached, exactly as the Julian Date is a plain running count of days for astronomers on Earth. Sol zero of the MSD falls on 29 December 1873. That epoch is not sentimental. It was chosen because Earth's and Mars's mean solar midnights nearly coincide there, and because the stretch from then to 6 January 2000 is very close to both a whole number of sols and 67 Mars years, which makes the arithmetic behave. The formula fits on one line: divide the elapsed Julian Date by 1.02749125, add 44796, and subtract a correction of about a thousandth of a sol for the imprecisely known position of Airy-0. That last term is the honest part. Mars's zero of longitude is a 500 metre crater picked out of Mariner 9 imagery on 14 August 1972 by Masursky, de Vaucouleurs and Davies, sitting on a meridian first used by Beer and Mädler in the 1830s, and the modern IAU definition has since been re-anchored to the tracked position of the Viking 1 lander instead, because a lander whose radio you have tracked for years is easier to locate than a small crater. Martian mean solar time is therefore known to a few seconds, and no better, because the planet's prime meridian itself is only known that well.

The date is where Mars stops resembling Earth altogether. Because the seasons are what matter scientifically, the Martian calendar is an angle: Ls, the solar longitude, the direction of Mars as seen from the Sun, measured from the northern spring equinox. Ls 0 is northern spring, Ls 90 northern midsummer, Ls 180 northern autumn, Ls 270 northern midwinter. Papers date their observations in Ls because Ls tells you the season and a month name would not. And the seasons are not close to equal, because Mars's orbit has an eccentricity of 0.0934, about five and a half times Earth's. Northern spring runs 193.3 sols and northern autumn only 142.7, a difference of more than 50 sols, and Mars gets roughly 45 per cent more sunlight at perihelion than at aphelion. Perihelion falls at Ls 251, in southern spring, which is why the global dust storms cluster in the southern summer half of the year. On top of that angle sits a year number. Clancy and colleagues proposed in 2000 that Mars Year 1 start at the equinox of 11 April 1955, an arbitrary zero whose convenience is that the great dust storm of 1956 falls inside it; Piqueux and colleagues extended the count in 2015 back to MY -184 in 1607 and forward to MY 100 in 2141. By that reckoning Mars Year 39 begins on 30 September 2026 and ends on 17 August 2028. Actual calendars with month names, of which the Darian calendar published by Thomas Gangale in 1986 is much the most developed, remain proposals. Nobody has adopted one.

Then there is the part that happens to people. A landed solar-powered spacecraft is a wasting asset, and the fastest way to use it is to have the humans awake when the rover is awake, so mission teams live on Mars time: 39 minutes 35 seconds later every day, all the way round the Earth clock and back again in about five weeks. The Mars Exploration Rover teams did it for the 90 sols of each prime mission in 2004. Bass, Wales and Shalin's post-mission analysis is unusually frank about the result: one team member walked into a wall after a run of shifts, another reported falling asleep at a freeway on-ramp, at least one operator rented a flat near JPL and lived apart from his family, meals became a problem because there was nothing fresh to eat at 3 a.m., and working Mars time carried a badge-of-honour status that made it socially hard to decline. Barger and colleagues then measured it properly during the Phoenix mission: 19 ground crew, 78 days, sleep diaries, wrist actigraphy and performance tests. Most of them did entrain, 13 of 15 assessable subjects showing rhythms locked to the 24.65 hour day, but main sleep averaged 6.2 hours a sol and, where phase could be measured, fell from 6.0 hours in phase to 4.9 hours out of it, and 64 per cent reported increased fatigue. Blue light panels at the workstations helped. Since then the dose has been capped: Curiosity's team worked Mars time for the first three months, Perseverance's for about 90 sols, and then both went back to Earth days.

Several things here are genuinely unsettled, and the honest answer to a few of them is that we do not know. There is no agreed Coordinated Mars Time, no time zones and no adopted calendar, and the recent work is about what a standard would even have to contain. Ashby and Patla at NIST published the relativistic groundwork in The Astronomical Journal on 1 December 2025: a clock on Mars gains 477 microseconds a day on a clock on Earth, oscillating by 226 microseconds a day over the Martian year with a slower modulation of about 40 microseconds a day over roughly 15.8 years, and their own residual uncertainty is around 100 nanoseconds a day. A framework paper by Turyshev in June 2026 sets out Areocentric Coordinate Time for Mars orbit and surface and says plainly that what it offers is a reference-system framework and not a finished Mars time ephemeris, with the choice of planetary ephemeris, seasonal gravity from carbon dioxide moving between the caps and the atmosphere, and calibrated link delays all still open. And the sol itself is not fixed: the RISE experiment on InSight found the planet's spin accelerating by about 4 milliarcseconds per year squared, so the Martian day is getting very slightly shorter, and the cause has not been established: the Nature paper puts the trend down to a long-term change either in the internal dynamics of Mars or in its atmosphere and ice caps, while NASA's release on it names ice loading on the polar caps and post-glacial rebound. The clock our own globe of Mars runs on is, in other words, still being written.