Tuesday, 16 September 2014

Leap Seconds and Blue Moons


The phrase 'once in a blue moon' is commonly used to refer to something that doesn't happen very often. I have long assumed that the phrase originates from a second full moon occurring in a single calendar month, this is backed up by Wikipedia although why 'blue' should be the colour of this moon I don't really know. Wikipedia has a fairly convoluted explanation as to why this might be the case but it sounds a bit dubious to me. Craig Bohren, in his book 'Clouds in a Glass of Beer' seems pretty sceptical of the double full moon origin as well, describing an event witnessed in the 50's in Edinburgh in which an actual blue Moon and Sun were observed. This was analysed by Robert Wilson, a staff astronomer at the Royal Observatory, who concluded that the event was due to ash particles in the air originating from a wildfire in Canada. This is a far more rare event than the now commonly accepted double full moon, which occurs every two or three years.

Despite some dissention in the camp regarding the origination of the phrase, it seems fairly well accepted now that a 'blue moon' is the second full moon occuring in a calendar month. While this is at best only a moderately interesting fact, it does reveal something deeper about the way in which most of the world has decided to live. Our calendar and general method of timekeeping is, at the most rudimentary level, based upon the movements of the celestial bodies which make up our most local universe. The Earth spinning on its axis has given us the day, the Earth revolving around the Sun has given us our year and the Moon revolving around the Earth has defined the month, at least, that's the common explanation. The fact that we can see two full moons in a single calendar month every once in a while kind of undermines that explanation. If you have a particularly good grasp of celestial dynamics, you might realise that the Moon is revolving around the Earth at the same time that the Earth is revolving around the Sun. This means that, relative to the Earth, the Moon might be turning 12 times in a year, while relative to the Sun itself (or a distant star) it is turning more (13 times in fact). I'm in danger of a digression here so I'll leave that where it is and let you figure it out or google it if it's not obvious to you.

So... back to the original thread, if we can see two full moons in a single calendar month, even if only rarely, it should be fairly obvious that, even if the length of the month was originally tied to the rotation of the Moon, it isn't any more. As for where weeks, hours and minutes come from, I'm not even going to begin to try and address that. Although I will say that it's less than obvious (or even accepted) that our current system is optimal or natural in origin.

I'm bringing all this up because there is a formal proposal on the books, at the ITU (International Telecommunications Union) World Radiocommunication Conference next year to scrap the leap second. This ambiguously* named unit of time is added (or occasionally subtracted) from our clocks and calendars on our behalf by various organisations around the world on a not-very-regular basis. I played a very small role in implementing one such unit of time when working at a telescope in the US. The prospect that this would bring the entire observatory system crashing down around us induced a mild state of panic but the transition was smooth in the end. A significant, ever-increasing, number of organisations that rely on precise timing go through similar contortions of their computer systems every time a leap second is implemented. No doubt bringing great stress to the workers at those organisations but remarkably little anxiety to the rest of the world.

The truth is that very few people have ever noticed, let alone worried about leap seconds. If you're not involved with satellite communications in some way (even tangentially) then I wouldn't be at all surprised if you weren't even aware that they existed. However, the proposal to scrap them still has some far-reaching implications. This might seem a bit dramatic, after all, do you really care if you are a second early or late? Even if you were a full 25 seconds late for a meeting (the cumulative total of all leap seconds since their inauguration in 1972), would anyone notice or care? There is a more philosophical point underlying this though - for the first time in history there are serious suggestions that we should cease to follow the motion of the Earth around the Sun and in its celestial path when setting our clocks.

The way in which we would define time, should the proposal be accepted, would likely rely upon atomic clocks, which are no less 'natural' as a timekeeper, although their nature is somewhat different and of a massively different scale to planetary orbits. This would be almost imperceptible for some time, at least a year or two, when we would inevitably become out of sync with our planetary orbit by a second or so. Give it a century or so and we might be a full minute behind our garden sundial.

This still seems like a fairly trifling matter, that is of course unless you're interested in increasing the length of time over which we consider human affairs. The Long Now Foundation has as its mission to 'creatively foster long-term thinking and responsibility in the framework of the next 10,000 years'. They are therefore understandably interested in how we keep time over long periods and are working to come up with a mechanical solution to that problem.

Personally, you may feel that we could let our clocks slip for a century or so before updating them with a full minute. After all, many of us cope with an hour more or less each time daylight savings is imposed upon us (to the chagrin of some). After all, as has been pointed out by the Long Now blog linked above, by the time it's really important, we may well have found a new, more accurate way of telling time. Possibilities abound when you think that far ahead, so why worry about it now?

The problem with this of course is that we need to define time accurately in some way. Even if you or I don't feel so in our day to day lives, an air traffic controller would probably have rather strong feelings about how precisely their systems are calibrated in time. The practical need for accurate timekeeping is undeniable.

Accepting the real practical need for precise time-keeping, is there a more philosophical angle? Certainly the National Measurement Office seems to thinks so; looking at the guest list for their recent meeting on a public dialogue about leap seconds reveals what must be one of the most varied audiences ever assembled.Representatives from national air traffic control services, navigational institutes and the National Physics Laboratory were in attendance, as were the British Bankers Association who likely had more financial than academic concerns that led to them being invited. There are those for whom the issue would be central to their work, although ultimately unlikely to change things too much, like the timekeeper for Big Ben, the British Horological Institute or the intriguingly named Worshipful Company of Clockmakers. Then there are those less involved in the technical implications of the leap second and more concerned with the human side of things, such as the Board of Deputies of British Jews, the British Muslim Forum and the Hindu Forum of Britain. In fact, there are only slightly more representatives invited from the technical and academic sector than from the loosely defined 'faith' sector.

Clearly, the way in which we track the passing of our days is seen as more than formal timekeeping but also something central to our humanity. I can appreciate this, there is a sense that we live our lives by a natural clock, that waking with the dawn and so on is how things should be. There is also a feeling of continuity in knowing that our ancestors were connected to a natural calendar, possibly more so than our current society. After all, the construction of Stonehenge may well indicate that we were using the Sun for our calendar thousands of years ago. The winter and summer solstices have been recognised as cultural events since ancient times, with the winter solstice likely setting the timing of our Christmas celebrations.

What is lacking from this ancient connection is any precision to our circadian rhythms, many experiments have been run to determine what the 'natural' length of the human day is and, while there have been a multitude of answers, the most conclusive would appear to be '24 hours or so'. Left to our own devices, without mobile phone alarm clocks, we tend to gravitate towards a regular schedule, rising and sleeping around the same time each day. However, this certainly will not be the exact same time each day. If you've had the discipline, good fortune, or whatever to rise whenever you feel like after waking naturally you probably found that this time varied by 15 minutes or so around a regular time. Maybe more on occasion, maybe less. If you're anything like me I doubt that you found this lack of precision in any way affected your day-to-day life.

Just as the occasional double full moon in a calendar month doesn't really affect you (and I'd be very surprised if more than 5% of the population even noticed it!) I suggest that measuring our civic timekeeping by an atomic clock rather than by the orbit of the Earth won't really affect anyone. Let those who need to worry about the precision of their computer systems and let the rest of us carry on regardless. I, for one, would be very happy if our society paid a little less attention to the clock on the wall, watch on our wrist and phone in our pocket. Measuring our days in ever-smaller chunks is vital when it comes to guiding aeroplanes and financial transactions, it is far less so when it comes to what time I should be waking up or be getting to work. After all, regardless of your actual beliefs, it is worth remembering that while God used to be in the details, that position has now been taken over by the Devil.

*leap years are longer by a day than other years, leap seconds are the same length as a normal second

No comments:

Post a Comment