NavList:

   

Reply

> -----Original Message-----
> From: Navigation Mailing List
> [mailto:NAVIGATION-L@LISTSERV.WEBKAHUNA.COM]On Behalf Of George Huxtable
>
> Thanks to Stacy Hanna for an interesting contribution about US Navy
> practice. It's good to know that astronavigation is still being practised.

And the extent of the current standard of practice.  Suddenly I don't feel
that I am wasting my time taking what seemed to be a totally esoteric course
(CN)!

> Here is an extract from Stacy's posting on the Day's Work, which is
> relevant to our discussions about noon sights-
>
> >8-12: QMOW compute watch time of LAN (Local Apparent Noon).
> >......
> >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.
>
> You will note that there is no mention at all of measuring the MAXIMUM
> altitude of the Sun around Noon. What is needed is the altitude AT the
> moment of LAN (Local  Apparent Noon), whether that is or isn't the maximum
> observed altitude. If the Navy vessel (generally, a fast ship) is making
> significant Northing or Southing, then at that moment of LAN the Sun will
> still be observed as rising or falling. Only by observing the Sun's
> altitude at the predicted moment of LAN will the exact latitude be
> obtained. But to predict that moment requires a reasonably good knowledge
> of the DR longitude and the ships speed in the E-W direction.

That is the nub of the issue we have been discussing.

> The resulting error varies as the square of the ship's speed, so is
> generally negligible for our small sailing craft, but becomes increasingly
> important as ship-speeds increase. As there are hydrofoil vessels at sea
> nowadays doing 40 knots and more, then if they happen to be observing noon
> latitudes (which may be unlikely except for Navy vessels) it will be
> important for them to follow Dutton's recommendations, below.
>
> Reading Stacy's comments sent me back to my ageing copy of Dutton's
> "Navigation and Piloting", 1969, which was presumably the "bible" of US
> Navy practice at that time. In art. 2803 he explains how to predict the
> moment of LAN, taking account of the dead-reckoning longitude and the
> ship's speed in Easting/Westing.
>
> Dutton says (with my comments in square brackets)-
>
> "Precision in determining LAN is especially necessary when a ship is
> steaming  on a generally northerly or southerly course at speed.
> For a ship
> proceeding towards the south, the sun will continue to increase its
> altitude for a considerable period AFTER it has actually crossed
> the ship's
> meridian, and an observation made at the moment when it reaches
> its maximum
> altitude will yield a latitude which may be considerably in error."
>
> [For Dutton, usually very precise, this is a bit slapdash. What he
> describes above applies only where the vessel is North of the Sun, as it
> would always be in US home waters. When South of the Sun, the situation is
> reversed.]
>
> [Dutton's paragraph quoted above, however, is a bit inconsistent with what
> he says in the next paragraph, as follows-]
>
> "Under most conditions, however, the sun will appear to "hang" for an
> appreciable period of time at LAN. That is, it will not change perceptibly
> in altitude. To obtain a latitude line at LAN, the navigator
> usually starts
> observing about two minutes before the time of transit, and continues to
> obtain sights until the altitude begins to decrease. The average of the
> times of the three or four highest altitudes will be used for
> reduction. On
> a northerly or southerly course, several altitudes may be taken before and
> after transit, to ensure that an unacceptable random error did
> not occur in
> the sight taken at the instant of transit."
>
> It seems, then, that the navigator is advised to adopt quite different
> procedures, depending on his course and speed. In the second procedure
> "under most conditions" he can stop taking sights "when the
> altitude begins
> to decrease". But if a significant component of his speed is pointing away
> from the Sun, LAN will not occur until several minutes after the moment
> when the altitude starts to decrease. So in that case, if he plans to
> measure altitude at LAN, he has to stick to his job for some time after
> that maximum.
>
> And what is Dutton getting at when he says "The average of the
> times of the
> three or four highest altitudes will be used for reduction"? Note that he
> refers to averaging the TIMES, not the ALTITUDES, here: are the altitudes
> to be averaged also? Reduction of an altitude at LAN doesn't normally need
> any times, in any case; that's the whole point of observing altitude then.
> Is he proposing a further correction, using "ex-meridian" tables, such as
> tables 29 and 30 in Bowditch (1981)? I am failing to understand this
> passage, and wonder if any listmember can explain it.

I don't know.

> Dutton's recoomendation, to choose to observe at LAN rather than maximum
> altitude, applies, he says to a ship "steering a generally Northerly ot
> Southerly course at speed." But he doesn't set any limit on how Northerly
> or Southerly that course has to be. Because the Northerly component is
> proportional to cos(course), a vessel steering 45 deg off, at say
> NW, still
> retains 70% of her speed as a Northerly component, so it must apply to a
> rather wide band either side of North and South.
>
> Can anyone else confirm whether a more recent edition of Dutton's has
> altered those comments?

Dutton's 15th Edition (2004) appears to contain exactly the same statements
(article 2802, page 344).

Dutton's 15th edition (2004) gives these methods for determining LAN.  I do
not have access to an older version to see if this is any different:

Article 2801: Describes a method for predicting LAN by comparing the ship's
rate of change of longitude with the Sun's, and determining when the two
will meet (GMT); accurate to within 4 seconds.
1. During the morning, for your DR position, determine the sun's meridian
angle east(tE).
2. Determine the rates of change of longitude for the sun (900'/hour) and
for the ship (predict from DR plot, course and speed).
3. Combine the rates of change of longitude: add if vessel steaming east,
subtract the ship's from the sun's.
4. Divide tE by the combined rates of change of longitude.
5. Add this number to the GMT hour from the Almanac.

Article 2802: Calculation based on Mer. Pass. entry in the Almanac; accurate
to nearest minute.
1. Obtain Mer. Pass. time, assume it represents LMT.
2. Plot vessel's DR for that time in ZT (= LMT for zone meridian).
3. Calculate DLo between the DR position in (2) and the longitude of the ZM.
4. Apply DLo expressed as time units to the LMT for LAN: add if DR is west
of ZM, subtract if east: gives first estimate of ZT transit.
5. If vessel is moving:
6. Plot new DR position for the first estimate of ZT transit.
7. Using the longitude for the DR position in (6), compute a new ZT
correction as in (3) and (4): gives the second estimate of ZT transit.
8. Repeat procedure in (6) and (7) for a third estimate in certain
situations [higher vessel speeds?? -- JT].

> Dutton's is getting at the resulting errors in the latitude deduced from a
> near-noon sight, resulting from the time shift between maximum
> altitude and
> LAN. For small slow vessels such as many of us sail, those errors in
> latitude are quite negligible. He is not dealing with  errors in longitude
> caused by that time-shift, which would be by no means negligible for our
> craft. In fact, he doesn't even refer to a noon sight as providing any
> useful information other than latitude; and rightly so, in my opinion.

Article 2809: Observations for Longitude:
- Longitude before St. Hilaire was based on a latitude obtained from LAN
sun, carried forward or backward to DR plot, so longitude could only be
accurate if the DR was accurate.
- Observations of bodies on their prime vertical give the most accurate
longitudes, [which is the most special case of observing a body toward the
west or east when the navigator wants to be more certain of longitude --
JT].  But predicting the time of prime vertical is even more tedious than
predicting LAN for the best estimate of latitude, so it is often not done.
"Nonetheless, to make observations of bodies on the prime vertical [azimuth
090 or 270] as the opportunities present themselves is good practice; a
reliable longitude line is always valuable."

> Indeed, we might as well accept that noon is the VERY WORST moment for
> attempting to determine longitude. Particularly as it only involves a wait
> of an hour or two, to make a much better job of it.

Dutton's 15th, article 2809: "Bodies not on the prime vertical can also be
used by sight reduction to yield computed longitude, but with decreasing
accuracy as their departure from the prime vertical increases."  [But I
presume that it is rarely possible to get the sun on its prime vertical,
since its altitude would be too low to the horizon for a good sextant
reading? -- moot point anyway.]

This all makes great geometric sense to me:

A. At one end of the spectrum -- noon -- the sun is due south or north at
LAN (meridian transit, aziumth 180 or 360); the LOP being perpendicular to
azimuth is parallel to latitude; and so the error in estimating latitude
from a single LOP at that time is much LESS than the error in estimating
longitude.

B. At the other end of the spectrum -- sun on prime verticle -- the sun is
due east or west (azimuth 090 or 270); the LOP being perpendicular to
azimuth is parallel to lines of longitude; and so the error in estimating
latitude from a single LOP at that time is much MORE than the error in
estimating longitude.

C. Sights taken between these two extremes -- ie. mid-morning or
mid-afternoon -- produce LOPs that lie at an angle to both parallels of
latitude and longitude, so these sights are characterized by INTERMEDIATE
errors in both latitude and longitude.

Today I made up and posted some images to demonstrate what I mean:
See "Is the Noon Sun Sight Perfect?" on the page:
http://jimthompson.net/boating/CelestialNav/NoonSunSight.htm

Jim Thompson
jim2@jimthompson.net
www.jimthompson.net
Outgoing mail scanned by Norton Antivirus
-----------------------------------------

   

Reply