NavList:

   

Reply

This time I try to find out what's happened
in Point Venus in August 1773, that is about a year
befre, during Cook's first visit of Point Venus
in his SECOND voyage.

Something very interesting has happened indeed:-)

The set consists of 12 shots of Sun-Moon distances.
Unlike in May 1774, they use astronomical quadrant
this time for the altitude of Moon, and some
unspecified quadrant/sextant for the altitude of Sun.
This time they say what the height above sea level was,
and again I prefer to use the Sun
altitude to derive the time
of observations.

So the scheme of my computations remains the same.
1. I reduce the Sun alt by adding SD, refraction and dip
corrections (from modern almanac).
2. Using terraserver position and Sun corrected alt
I derive the time (GMT) of observation.
3. Using terraserver coordinates and this GMT
I derive Lunar distance error and Moon atl error,
and Clock error. (Having certainly corrected their
observed Moon altitudes with the modern almanac).

Here are the results.
Lunar distances are perfect.
Just outstanding (by my standard, which may differ from
Frank's much higher standard.)
The average error in the first series (6 observations)
is -0.2' and sigma is 0.65.
The average error in the second series (6 observations)
is -0.2' and sigma is 0.1' (!!)

The most amazing thing about the whole business is
WHAT HAS HAPPENED BETWEEN THE FIRST and the SECOND series.

The time interval between them is 5 minutes.
During this time, they re-checked IC for all three
instruments. And the thing I find most puzzling is
that IC of their sextant used to take the Lunars
changed from +1' to -3'42" (This is what they
recorded !!!)

Was it dropped??
More likely is that the observer noticed a substantial
side error/lack of parallelism or whatever while
shooting the first series. So he adjusted the sextant
and re-checked the IC.
Other instruments IC were re-checked at the same time,
and they did not change.

Another possibility is that they used another sextant
for this second series (and did not record this fact).
Anyway, the Lunars of both series seem outstanding to me.

2. Moon's altitude. There are still problems with
Moon's altitude, though not so severe as on
May 1774.
The average error is -3.8' and sigma is 2'.
(I recall that this time this was a stationary
"astronomical quadrant", of hudge radius
and using artificial horizon (a plumb line).
The stated error/correction is 0.5'.

I can only conjecture that taking the Moon altitude
was considered not very important business
which was delegated to some apprentice midshipman
(or perhaps even a boy) who was not really thinking
of high precision, but rather about Thaitian girls
passing by:-)

Unlike Frank's XX century apprentices who shoot
everythig with less than 0.2' error, without any
preliminary experience:-)

3. Derived longitude.
This time the "residual error" of shooting
was of the opposite sign to the almanac error.
The almanac error on this day was 0.9'
and should had contributed 25' in the longitude error.
The shooting error, as I said before was only about -0.2,
and its contribution to longitude was 5'5.

The actual errors in their longitude are
14.6' for the first series and 2.6'
for the second.

Which is just within best possible theoretical
precision of the Lunars method, using ordinary tools
almanacs and reduction methods.

I may add that WHEN NEEDED, they COULD shot altitudes
to 0.5'. This is seen from their latitude observations.
Their latitude deviates from Terraserver by only 0.5'.
Perhaps by even less, because I am not 100% sure
in the Terraserver position of their observation
place.

Alex.



--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to NavList@fer3.com
To , send email to NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---

   

Reply