Welcome to the NavList Message Boards.


A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

Compose Your Message

Add Images & Files
    Re: Navigation without Leap Seconds
    From: Richard B. Langley
    Date: 2008 Apr 19, 15:14 -0300

    A couple of relevant items:
    -- Richard Langley
    On Sat, 19 Apr 2008, George Huxtable wrote:
    >Geoffrey Kolbe has written-
    >"Well, my Long Term Almanac is nominally good up to the year 2050 and will
    >actually work OK up until 2060. So this is a very real question for me. I am
    >assuming that if leap seconds are abolished, anyone using my tables - which
    >assume GMT, or UT1 as it is called these days - will know the difference
    >between broadcast time and GMT and use this as a correction to broadcast
    >And following a question from Lu Abel, elaborated-
    >"Actually, the theory used to generate my tables includes an estimation for
    >delta T, the difference between ephemeris time and Universal Time. How delta
    >T will vary in the future is not that well understood, of course, which
    >means that by 2050 my tables will probably be out by a second or so. "
    >Response from George-
    >I wonder how confident Geoffrey really is about that projection into the
    >future, to within a second "or so".
    >Let me propose a simple test. If we take the date on which his tables were
    >based, and arrive at the number of years, to 2050, that the prediction of
    >delta-t was to be valid for (so, 42 years if the tables were issued in 2008,
    >for example). And then take whatever projection Geoffrey uses to predict
    >delta-T to 2050, and instead apply it backwards, for the corresponding
    >number of years, to "predict" changes in delta-T in the past. Then, even
    >though we are now enabled to use a certain amount of hindsight in making
    >that projection, I ask him what's the maximum difference, over those years,
    >between delta-t, "predicted" that way, and delta-t as it actually happened.
    >It's an interesting question, what should be done in our measurement of
    >time, as the rotation of the Earth slows more and more into the distant
    >future. A second defined as a fraction of a day (which I will call an
    >Earth-second) will diverge more and more from the "scientific", constant,
    >second, to which physical laws conform. That is a fact of life, quite
    >inevitable, and we can do nothing about it except adapt to it as best we
    >can. To me, the answer isn't clear-cut.
    >It's a problem that can only get worse, faster, and faster, into the future,
    >and we need to bequeath to posterity a system that is practical and
    >applicable into the far distance, not one that will call for some major
    >upheaval at some future date long after we're all dead and gone.
    >The second was defined to correspond to the rotation of the Earth around
    >1900, and delta-t was set roughly to zero about then, and it's been growing
    >ever since, at a steadily increasing rate.
    >There are two main causes at work here. One is rather well understood, now,
    >and can be readily predicted. It's due to slowing of the Earth's rotation,
    >due to the action of the tides, caused mainly by the gravity-gradient of the
    >Moon (and, to a much lesser extent, the Sun). This component can be
    >measured, rather well, because the same forces have a corresponding effect
    >on the Moon, driving it out to a larger radius from Earth, which can be
    >measured by radar ranging. Over the long term, that is the major effect,
    >that will cause delta-t to grow from its present 38 seconds or so to about
    >an hour, in a thousand years time. By then, leap seconds would have to be
    >inserted, not at intervals of a couple of years as at present, but every
    >couple of months. Another thousand years, and it will have reached four
    >hours, because it changes according to a square law. 5,000 years from now,
    >the times will have diverged to put the two dates one whole day different,
    >and by then there will be a leap second needed at fortnightly intervals! So,
    >in the far future, will leap-seconds be a viable proposition? I only ask.
    >The other main cause of the variation in the Earth's rotation is due to
    >fluctuating motion in the fluid core. This is combined with smaller faster
    >changes that result from winds and ocean currents, and much slower changes
    >due to continental drift, which we can ignore for now. But its those effects
    >of the fluid motion that are unpredictable, and mask, to a large extent, the
    >predictable changes in rotation rate over time-scales of decades at a time.
    >It's these fluctuations that I predict will cause Geoffrey Kolbe a bit of
    >So what would be the practical effect of arresting  change in delta-t,
    >presumanly at its current value, so that Earth-time is forced to follow the
    >constant seconds that scientists use? For navigators, not a lot, I predict.
    >Almanac-makers, including Geoffrey Kolbe, would be able to produce their
    >wares for dates far into the future, instead of them having to anticipate
    >changes in the value of delta-t, for a few years ahead. Instead of using a
    >special time-scale (ephemeris time, now called terrestrial dynamical time)
    >to compute their dynamics, differing unpredictably from the UT (same as GMT)
    >that ordinary mortals use, then everyone would be forced to use the same
    >rational time as astronomers do.
    >The snag comes when converting sky-positions to a geographical position with
    >respect to the Earth's surface, in calculating Greenwich Hour Angles (GHA).
    >In doing that, the increasing  and unpredictable discrepancy corresponding
    >to delta-t will have to be allowed for. If the almanac is to remain valid
    >for many years ahead, that can't be done within its pages, but only by the
    >navigator, knowing what the correction happens to be at his current date.
    >Any difficulties resulting from such a change, in freezing leap-seconds,
    >would be felt by the ordinary person in real-life, way into the far future,
    >who would become aware that the time given by his clock on the wall no
    >longer corresponded with his time-by-the-Sun. Would that matter? We already
    >have got so used to tinkering with our clocks these days, first with mean
    >time, then with daylight saving and time-zones, that it's no longer sacred
    >any more for the Sun to be overhead at noon.
    >Perhaps an all-change in the time zones by an hour, a millenium or so from
    >now, would be called for. From then on, the interval between such changes
    >would get less and less.
    >Should such things matter? We should ponder hard about the consequences of
    >our actions before making changes. What's a few thousand years in the future
    >when we consider how long civilisation has been going? How far ahead was
    >Julius Caesar thinking, when he reformed the calendar? And then, Pope
    >Freezing leap-seconds at some date would create a bit of a mess, anyway. We
    >would end up with three different "scientific" time-scales, running in
    >parallel, with constant differences between them. First, there's Ephemeris
    >Time, (or TDT), a constant measure of time frozen from Greenwich Time as it
    >was in 1900, . Then there's GPS time, which was effectively Greenwich Time
    >but frozen at some date around 1980. And then we would have a new measure of
    >time, frozen from Greenwich time at some date a few years from now. There
    >would be a range of several tens of seconds spanning these three
    >time-scales. What chaos!
    >I have tried to offer a balanced view of this matter, being somewhat
    >undecided about it myself. I would resist any hurry to change.
    >A curious fact has struck me, in thinking about all this. Nobody seems to
    >have coined a word for delta-T, except "delta-T", sometimes expressed as a
    >Greek symbol. Isn't that a bit of a surprise? It doesn't really describe
    >well what it represents, and it certainly gives no clue as to which
    >time-scale is lagging on which. Delta-T , as defined, is normally positive,
    >but can be negative, as it was for a few years near 1900. Is there another
    >word for this difference, used in any other language? In many ways, it's
    >similar to the confusion that surrounds another time-difference, that
    >between mean and apparent time, which is so confusingly expressed by the
    >words "equation of time".
    >contact George Huxtable at george@huxtable.u-net.com
    >or at +44 1865 820222 (from UK, 01865 820222)
    >or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
     Richard B. Langley                            E-mail: lang@unb.ca
     Geodetic Research Laboratory                  Web: http://www.unb.ca/GGE/
     Dept. of Geodesy and Geomatics Engineering    Phone:    +1 506 453-5142
     University of New Brunswick                   Fax:      +1 506 453-4943
     Fredericton, N.B., Canada  E3B 5A3
         Fredericton?  Where's that?  See: http://www.city.fredericton.nb.ca/
    Navigation List archive: www.fer3.com/arc
    To post, email NavList@fer3.com
    To unsubscribe, email NavList-unsubscribe@fer3.com

    Browse Files

    Drop Files


    What is NavList?

    Join NavList

    (please, no nicknames or handles)
    Do you want to receive all group messages by email?
    Yes No

    You can also join by posting. Your first on-topic post automatically makes you a member.

    Posting Code

    Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.

    Email Settings

    Posting Code:

    Custom Index

    Start date: (yyyymm dd)
    End date: (yyyymm dd)

    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site