NavList:

   

Reply

I suspect Peter Hakel and I are at cross-purposes here. There seems to be
misunderstanding, rather than disagreement, between us.

He wrote, about taking a series of Sun altitudes to determine Local Apparent
Noon- "However, according to the procedure you were supposed to do these
noon-curve measurements anyway, so you're not doing any extra observational
work beyond what is already mandated."

Presumably, Peter here is thinking of the "procedure", "mandated" by the
instructions quoted by byronink@netzero.com, in "Noon Observe LAN. Recommend
observations be started at 10 minutes before computed time of LAN, and for a
couple of minutes after. Reduce sighting and determine ship's latitude."

But observing over such a short period as 12 minutes would provide no useful
information about time of meridiam passage. To obtain that  requires a
number of observations, to maximum precision, spread over a period of an
hour or so, symmetrically about the peak altitude, as has been discussed at
some length on this list. That operation has never been part of a mariner's
repertoire, as far as I know. I've not seen such a procedure recommended in
a textbook, nor come across it referred to in use, in any navigator's
logbook. If Frank's perusal of old ship's logs has shown up any such record,
no doubt he will tell us about it.

Near noon, if a mariner knew his GMT, and could estimate his longitude
reasonably well (by dead-reckoning from a morning Sun position line, say),
he could predict the moment of LAN and measure Sun altitude at that moment.
He might take a few observations, closely-spaced, and average them to
increase accuracy, but that's all. From that observation at that moment , he
could derive latitude, without having to allow for North-South speed
component. That appears to be the intention of the precedure quoted by
byronink@netzero.com.

On the other hand, if, as in early navigation, no chronometer was carried,
or if it was distrusted, or if the longitude couldn't be estimated, the
operation would have to be somewhat different. A series of altitudes would
be taken near noon, increasing then decreasing, and the maximum altitude
noted. This was done solely in order to acquire the VALUE of that maximum
altitude to determine latitude; not its TIME.  For precision, that latitude
might call for a correction for North-South speed component, though in
mid-ocean that would often be neglected.

This question, about finding time of meridian passage from a prolonged
series of altitude observations around noon, has become something of an
obsession among some, here on Navlist, but this quest has never been of
importance to the wider community of practical navigators. There were better
methods, and they were aware of them.

As for the "running fix", that;s a technique about which an aura of
mysticism and distrust has recently been woven here by John Karl. But it can
be a precise business, and is used a lot in celestial navigation, to link
observations made at different times. The dead reckoning between
observations is affected by unknown local currents, but in ocean travel,
those are seldom greater than half a knot, and where they are, mariners are
usually aware of them and can estimate them. So errors from that source,
over a 6-hour interval , should not be greater than 3 miles or so. With
care, a ship's speed through the waterso , even under square rig, could be
estimated to within 10%, or thereabouts. Even in my own small sailing craft,
which on a good day could do all of 5 knots, in the swirling waters of the
English Channel with tidal streams which were often well over half that
speed, my dead-reckonings would seldom be out by more than 4 or 5 miles,
over a 6-hour "tide".

However, the procedures being referred to were for a modern warship
proceeding under power in well-charted waters, of which ocean currents would
be predictably known, and dead reckoning of her progress should be highly
predictable. With Sun altitudes morning , noon, and afternoon, linked
together by a running fix, precise postions would result.

George.

contact George Huxtable, at  george@hux.me.uk
or at +44 1865 820222 (from UK, 01865 820222)
or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
----- Original Message -----
From: "P H" 
To: 
Sent: Saturday, January 09, 2010 7:20 PM
Subject: [NavList] Re: QMOW Days work in Navigation


| Perhaps I should clarify why I found this detail so interesting. There are
two reasons, in fact.
|
|
| First, it is the extensive use of the running fix which has been both
criticized and defended quite recently on this list.
|
| Second, less recently some of us (Jeremy, Antoine, Andres,...) went
through some real-life meridian transit data that resulted in longitudes
that were better than mediocre, in my opinion.  The construction of the noon
or any other meridian transit curve indeed "require extended and precise
observational data" as George said.  However, according to the procedure you
were supposed to do these noon-curve measurements anyway, so you're not
doing any extra observational work beyond what is already mandated.  Before
computers the "noon curve longitude" determination would result in
non-trivial amount of extra computational work, which I suspect as the main
reason why this was not done traditionally.  This led to my hypothesis that
accuracy may not have been the issue; the extra labor required was, hence
the running fix instead.
|
| Now in 2010 we do have powerful computers and automatization of these
calculational chores by software, essentially eliminating the extra labor
problem caused by vessel's motion, for instance.  This brings me back to the
question of accuracy and also to my reason #1.  If this "day's work
procedures" were to be followed, wouldn't it be better to recommend
"longitude by LAN" (which is a true fix) instead of doing the running fix
using the DR advanced mid-morning Sun LOP and LAN latitude?
|
|
| Peter Hakel
|
|
|
|
| ________________________________
| From: George Huxtable 
| To: NavList@fer3.com
| Sent: Sat, January 9, 2010 5:30:43 AM
| Subject: [NavList] Re: QMOW Days work in Navigation
|
| Peter Hakel wrote-
|
| " ...experience has apparently shown that doing a running fix is
preferable
| to getting longitude from the time of LAN.  This is very interesting.
Does
| anyone know how this rule was established?  Has anyone tried both methods
| and compared their accuracy?  It is also conceivable that nobody has
really
| tested this and the reasons are historical: i.e. the LAN is used to give
| latitude ONLY and thus the running fix is the next best thing to establish
| position."
|
| ==================================
|
| Peter has it right. Although much attention has been given on this list to
| the determination of longitude from observations around Local Apparent
Noon,
| mentions of its use at sea in actual chronometer-navigation are rare. That
| is because its defects were recognised, as a method which would require
| extended and precise observational data, to achieve a mediocre result;
there
| were better ways of doing the job. The moment of Local Apparent Noon was
| indeed calculated, but that was to provide the best monent to take an
| altitude of the Sun to determine latitude, which would (unlike the maximum
| observed altitude) be unaffected by the vessels course and speed.
|
| [rest deleted by PH]
|
|
|
|

   

Reply