|
|
|
|
|||||
|
||||||
From: Frank Reed (no email)
Date: Mon Jul 05 2004 - 00:48:52 EDT
I've created a new web tool which can provide detailed, accurate online
'Nautical Almanac' data. It uses the same engine as the lunars prediction page
which I've described recently on the list. It includes all (? have I missed
anything ?) of the information in Reis's Online Nautical Almanac except the "v" and
"d" interpolation figures, though not in the same format. It's valid over a
much longer time period, and I believe that the accuracy is excellent. As
relatively new software, there are likely to be bugs popping up. Let me know when
they bug you!
Access the online almanac via the link on my lunars page here:
http://www.HistoricalAtlas.com/lunars
or directly here:
http://www.HistoricalAtlas.com/lunars/nadata_v5.html
On this page, you can select the object that interests you (including
navigational stars will generate a rather long list) at the top. Next choose any date
from 1750 to 2050. For most purposes, you would select "Gr. Mean Time" as the
time basis for the almanac. If you're interested in historical almanac data,
you might want to use Greenwich Apparent Time. The prefered modern angle
format is degrees and tenths of minutes (dd mm.m) but you can also get data in
degrees, minutes, and seconds (dd mm ss) or in decimal degrees (dd.dddd).
If you select the option "Visible at DR only" and enter your latitude and
longitude, you will get almanac data only when the objects are in the sky from
your location and visible (the Sun has to be below 5 degrees for the stars to
show up in the table (don't complain that Venus is visible in daylight
sometimes... you can always DE-select this option if you're interested in daytime
navigation sights).
There is also an option to display SHA instead of GHA. This was really just
something for me while I was experimenting and it may disappear or change
later.
This web tool is doing relatively little calculation. The positions of the
Sun, Moon, and planets are stored in a ten megabyte database to the nearest
arcsecond. These positions are originally from the JPL numerical integrations, so
they are intrinsically very good. Note that this approach sidesteps the issue
of coding algorithms to generate the positions. The Moon positional data here
is considerably more accurate than the almanac data found in most online
tools. The positions of the stars are precessed, nutated, and advanced for proper
motion using the usual algorithms (see Meeus "Astronomical Algorithms", for
example) and should be accurate to a fraction of an arcsecond. The "delta-T"
correction for dynamical time (ephemeris time) is tabulated internally at ten year
intervals based on the list in Meeus and interpolated for years between those
dates.
When calculating positions for "Greenwich Apparent Time", the software
iterates until the GHA of the Sun is exactly equal to the Greenwich time. So if it's
noon at Greenwich, the Sun's GHA is zero degrees exactly.
NOTE ALSO: I've added a "Greenwich Apparent Time" option to the lunars
prediction page here: http://www.HistoricalAtlas.com/lunars/lunars_pre_v5.html
Using this option, you can generate data which can be compared directly with
the old (pre 1834) Nautical Almanacs. Also, if you have old lunars sights
recorded with Local Apparent Time, it is a much simpler matter now to reduce
those. You don't have to locate an old N.A. or convert times to mean time. I've got
a couple of pages from old almanacs on my web site here:
http://www.HistoricalAtlas.com/lunars/na.html
It's interesting to see how the distances on those old pages compare with
modern calculations. I get an average error for the August 1843 lunars of about
0.24 minutes of arc (that is, the published data has an average absolute error
of about that amount compared with the calculations generated by my web site
tool). This sounds about right since Lecky commented that published lunars
(before c.1880) were inaccurate to an extent of about 6 to 8 minutes in the
longitude and multiplying that 0.24 by the usual factor of 30 yields 7.2 --right in
the range Lecky described.
Frank R
[ ] Mystic, Connecticut
[X] Chicago, Illinois
|