How to Wind a Martian Watch

Keeping time on Mars might seem easy. Like adding leap-years, some standard time offset might be imagined to take care of the longer martian day. Naively put, how to keep track of martian time might seem only to require adding a daily drift. Each day on Mars needs a good Earth watch that can tick 2375.244 seconds longer.

Mars Time

But at a deeper level, time is tied intimately to a planet's orbit. As Michael Allison, a Mars24 clock designer for the current Mars Exploration Rovers, put the problem, one cannot escape "pervasive forcing by sunlight"--even when resident on another planet. The Sun is indeed the master watchmaker.

Sundown on Mars, Pathfinder mission.
Credit: NASA/JPL

On Mars today, the consequences of time drifting incorrectly would be devastating. For power, the rover depends on its solar arrays. Biright daylight hours help warm its electronics for full thermal control during a frigid night spell (falling lower than -70 C). Mission managers often refer to their nightly low-power mode, as a rover "going to sleep" before the sun goes down. The rover "wakes up" late in the morning (usually after 10 AM locally), a daily ritual triggered always after sunrise.

These time problems become greater when mission managers try to synchronize two terrestrial teams' circadian rhythms, while working twin rovers on opposite sides of another planet. As principal investigator Steve Squyres explained: "It gets really confusing." If such a predicament can be considered possible in an interplanetary sense, both Mars and Pasadena are on-call just slightly more than twenty-four-seven. The Pasadena team has to give its best for "~24.6/7".

Astrobiology Magazine had the chance to talk with Michael Allison about the Mars24 Clock. Allison and his colleague, Robert Schmunk, both of NASA's Goddard Institute for Space Studies (GISS) in New York City, have taken on the task of understanding interplanetary time. That way, no matter where we go, time continues ticking safely, as nature's way of keeping everything from happening at once.


Astrobiology Magazine (AM): How did you first get interested in Mars' time keeping?

Michael Allison (MA): I have a long-standing interest in the seasonal-solar forcing of the other planets, including Saturn's satellite Titan, in connection with my research on extraterrestrial atmospheres. But a few years ago, as a member of the science team on the ill-fated Mars Climate Orbiter (MCO), my preparations to map the daily Martian weather from a sun-synchronous spacecraft suddenly made the Mars solar timing issue a matter of practical importance to my own work for the project.

It was apparent to me then that it would be valuable to have an accurate and efficient algorithm for the calculation of the planet's variable solar illumination. For one thing, the extreme eccentricity of the Mars orbit, some five times as large as the Earth's, along with the approximate alignment of its northern winter solstice with its perihelion, makes it a very seasonally responsive planet. And equally important, the very rapid cooling-time of its thin, cold atmosphere also gives Mars an exaggerated thermal response to the local time of day on its surface. (There are some indications from Mars lander data that you could almost set your watch there by the wind direction!)

viking_map
Viking Lander-1 (1976) showed dramatic changes in daily sunlight during dust storm activity. Although the colors shown here are processed, not real, they do show relative changes over nearly 600 Sols on the surface.
Credit: JPL/NASA

While the calculation of the seasonal longitude and solar "hour angles" can be extracted from any of several existing ephemeris programs, these are computationally elaborate and therefore not so easily adapted to offline atmospheric models or data processing routines. I discovered, however, that I could compute both the solar season and "time of day" on Mars to high accuracy with fairly simple nested strings of trigonometric functions, referenced to certain "areocentric" (Mars-centered) coordinate angles which I derived as a fit to an accurate ephemeris calculation.

Although the loss of MCO prevented me from purusing my anticipated Mars weather mapping, it is gratifying to know that my solar timing algorithms are now in use by several Mars experimenters and data archivists. And thanks to the programming efforts of my GISS colleague, Robert Schmunk, our Mars24 Sunclock is now in wide use by the Mars Exploration Rover (MER) team.

AM: How was the term 'Sol' first introduced to account for the approximately 39 minute longer day on Mars and the use of solar time?

MA: To be precise, the average solar day on Mars is 24 hr 39 min 35.244 sec, and given the small but significant difference between this and a solar day on the Earth, some new terminology was clearly needed to provide a clear distinction between the two.

Although I've been asked many times who first coined the term "sol" for a Mars solar day, I still don't know exactly where and when it originated. But it became a common usage of the Viking project, with the first successful landed missions to Mars back in 1976.

AM: The history of keeping time on other planets goes back-- at least on Mars-- to the 1976 Viking lander, where they sort of kicked off a countdown clock after landing, correct?

mars_watch
A local Pasadena watchmaker's design for the MER team.
Credit: JPL/NASA

MA: The Vikings on Mars provided the first occasion for the routine usage of hour and minute read-outs of the "Local Lander Time," as the project referred to it. And yes, lander data were tagged with both an hour-minute clock-time and a sol-number count of Mars days elapsed post-touchdown for each of the two spacecraft.

AM: Then in the 1997 Pathfinder, there was an attempt to kick off Sols by some method almost like New Years Eve, where a new Mars clock was started at midnight on the first day?

MA: As with the 1976 landers, the Mars Pathfinder project kept a mission time log tagged with a running count of sol numbers, but started with "Sol 1" at touchdown instead of the "Sol 0" reference adopted by each of the Vikings. And whereas the Viking Local Lander Time was a variation of an evenly advancing "Mean Solar Time," Pathfinder used the variably advancing "Local True Solar Time," accounting somewhat more accurately for the Sun's location in the sky as affected by both the planet's eccentric orbit and the tilt of its spin axis.

pancam
The Mars Pancam design has a camera bar that contains Pancam and Navcam (navigation camera) heads. A "visor" changes the elevation of the cameras so the rover can look up or down. "NASA is going to get the vehicle's orientation by pointing the Pancam to find the sun in the martian sky. So ironically, they are going to use some of the same information that the sundial relies on for its shadow." --Woody Sullivan, U. Wash, Mars Sundial team
Credit: Cornell University

AM: Why was a decision made for the current round of NASA landers to have another way of calculating Sols?

MA: The MER project adopted yet a third standard for the referenced local solar time, reverting to a kind of evenly advancing mean time, but calibrated so that this would coincide with the Local True Solar Time at the midpoint of the nominal, 91-sol mission for each lander. I wasn't involved in the choice of these standards, but can understand why an evenly advancing time system might seem desirable from the standpoint of experiment planning.

AM: Would the European Beagle 2 lander have followed the same method?

MA: Shortly after the MER project asked us to adapt our Mars24 utility to the project's adopted timing standards, my colleague Rob Schmunk made inquiries with a Beagle scientist as to the project's plans for Mars solar timing. As far as we learned, the Beagle mission had not defined any new solar timing reference for their lander. So Rob coded our Mars24 Sunclock to read out Beagle mission time in terms of the standard definition of Local True Solar Time.

AM: The calculation of Sols is not as straightforward as just adding a half hour to an Earth day, or dividing a longer day by a different unit of seconds. It depends on alot of orbital parameters. Can you summarize in layperson terms a couple of the key factors in constructing a good martian watch?

MA: Well as your question anticipates this can be a long story! My published papers on this subject provide a fairly accurate definition of the so-called "Fictitious Mean Sun" at Mars, as adopted by the MER project, and this can be used to define a "mean solar day" on Mars in analogy to the adopted conventions for terrestrial time-keeping. One can then design a Martian watch to run at the implied slightly slower rate of advance of the Mean Solar Time on the planet.

Once synchronized to the time appropriate to a particular longitude on Mars, such a watch would give an approximate trace of the Sun's position in the sky so that 12 noon, for example, would provide a rough indication of the time it could be expected to cross the overhead meridian. As a result, however, of both the eccentricity of the orbit of Mars about the Sun and the 25°.19 tilt or "obliquity" of its spin axis with respect to the plane of its orbit, the actual instant of "Local True Solar" noon can vary by as much as 51 minutes from the Mean Solar Time, depending on the season.

The discrepancy is referred to as the "equation of time," as needed for the accurate interpretation of a sundial measurement. There is a similar but considerably smaller effect on the Earth, as indicated by the Figure-8 "analemma" printed on some globes (typically somewhere in the south Pacific ocean) showing a plot of the seasonal variation of the equation of time on our planet with the declination of the Sun. Because the Mars winter solstice comes after, rather than before, its perihelion passage, at its closest approach to the Sun, the Martian analemma assumes the rather different shape of raindrop!

So a really good Martian watch would need to be "programmed" with the seasonally variable equation of time effect to give its user an accurate position of the actual Sun time.

AM: Cornell University, Bill Nye and The Planetary Society are sponsoring the Mars Sundial project as an educational project on the camera calibration target. Is that a project that you were contacted about?

In an earlier version of the sundial, the gnomon was gold, but the bright color interfered with callibration
Credit: NASA/JPL/Cornell

MA: I only wish I had been! I think it's a wonderful vehicle for public education, and I happen to be the father of two children who love "Bill Nye the Science Guy" on public television. While attending a Mars conference a few years ago, I met Jim Bell, the principal investigator at Cornell for the experiment that carried the sundial and gave him a reprint of my first published paper on Mars solar timing algorithms. I remember he was sufficiently interested to ask me for a second copy for one of his colleagues.

My recent attention to my current involvement in the Cassini mission to Saturn and Titan has kept me moving in different circles I guess and unfortunately I haven't been to another major Mars conference since. But of course we would be happy to support the Mars Sundial project in any way that might be helpful to them.

AM: Do you have a reaction to the JPL team members using a local jeweler to make custom Mars watches, presumably based on the Mars24 calculations?

MA: I think this is great! It just shows that for all the high-tech sophistication of modern digital devices and computer software, we humans still have an affinity for analog-style implements. We don't just read numbers. We want to touch and watch real things move. I presume these watches were made for the Rover team members adapting to a major adjustment in their daily schedules. As much as I of course like our Mars24 software package, it's not something to get you out of bed in the (well Earthside) evening hours of sunrise at Gusev!

AM: Bill Nye told us that historically, 'time keeping has been more important to civilization than the wheel'. With both the roving laboratory and its way of finding position (and time) by camera-location of the Sun, do you have a perspective on this?

Spirit finds the sun, Sol 4 on Mars. Click image for larger view. "If it's 3:00 AM in Los Angeles, what time is it at Gusev Crater? That's the kind of problem we're wrestling with this week.It gets really confusing."--Steve Squyres Credit: NASA/JPL/ Cornell

MA: I think this is a telling observation. Astronomical timing provided the foundation of data that led Kepler, Newton, and their successors to the discovery of physical laws which serve as the basis of all our technology. Without accurate timing, modern commerce would quickly grind to a halt. It's easy to take for granted the apparently easy global synchronization of our culture. But the standardization of world-wide time zones and their reference to a "prime meridian" was realized only as recently as 1884, largely as result of the tireless persuasion of Sir Sanford Fleming, a Canadian railroad engineer.

AM: There are various proposals (Zubrin calendar, Mars zodiacal calendars, and metric clocks) in which the time units of 12 and 60 take on different importance. Do you favor some familiar earth-like division of Mars time, or a new version entirely?

MA: All of us who like to think about the other planets are naturally given to novelty and as with most academic types tend to be somewhat independently minded, so it's not surprising that there are already so many different proposals for Mars calendars and timekeeping. There's obviously some practical utility to decimal divisions.

Mars24 Applet. Click images for larger view of day-night Mars projections.Credit: NASA/GISS


Dr. Who: 'The Earth isn't my home, Sarah. I'm a Time Lord.'
Sarah: 'Oh, I know you're a Time Lord.'
Dr. Who: 'Deactivating a generator loop without the correct key is like repairing a watch with a hammer and chisel. One false move and you'll never know the time again.'
--Dr. Who, TV Show episode, Mars Pyramids

The French Revolution attempted to reform daily time reckoning in terms of decimal subdivisions, but obviously it didn't catch on. Astronomers, however, frequently use a sequentially reckoned chronology of "Julian Day numbers" with decimal subdivions of the corresponding date. But I think it's sensible to retain the 24 hour - 60 min - 60 sec reckoning of solar time on Mars, as adopted so far by each lander mission.

Someday it will be important to devise a coherent solar day chronology for Mars, in place of one after another separate sol-number series for each landed mission, and someday a humanly reckoned calendar. A few years ago, I devised one myself, combining 12 months with the familiar Gregorian calendar names, but interspersed with 10 new months making a total of 22 subdivisions of 30/31 solar days as needed for a total of 668.6 sols in a Mars tropical year. But there are hundreds of different suggestions for this out there.

Although we have not attempted to support any of these with Mars24, it does include as one of its read-outs a "Mars Sol Date" or MSD reckoned from a particular conjunction of Mars-Earth seasons and meridian mean times in the late-nineteenth century ... along with a decimal, thousandth-part subdivision of each solar day!

AM: Are there any proposals to have a more absolute universal time based on star rotations or something that wouldn't change much from planet to planet in our solar system?

MA: Atomic clocks already provide a sufficiently accurate universal time standard for the measured variation of planetary rotation, including the lunar-tidal spin-down of the Earth. But yes, some have proposed that an appropriate ensemble average of pulsar spin rates could provide a still more accurate reference. But I think there will still be a need to devise systems of time measurement adapted to the temporal cycles unique to the different planets, largely as a result of their pervasive forcing by sunlight.

AM: Any suggestions for human adaptation to such non-circadian rhythms?

MA: I don't know. Get plenty of exercise and start every sol with a good breakfast. I understand that some long-term studies of sleep patterns suggest that in the absence of light, humans may tend to drift to a diurnal repetition period slightly different from the 24 hours we mostly live by.

I like to think that humans can be a very adaptable species . . . and look forward to the time when we are living on other planets.


Related Web Pages

Mars24: NASA GISS
Living on Mars Time
The official U.S. time
United States Naval Observatory - What is Universal Time?
Primary Atomic Reference Clock in Space, NIST
Our Time in the Sun
Two Worlds, One Sun
Night Lights: Interview with Woody Sullivan
Interview with Bill Nye: The Sundial Guy