Re: flare/mapblast/gps

From: Chris Olsson (
Date: Sun Sep 02 2001 - 19:45:14 EDT

As Bruno points out,
>SGP4/SDP4 use the equinox of date which for the next
>few years is close enough to J2000.

RA/Dec can, in the absence of contrary information, reasonably be presumed to
refer to the 2000 epoch, despite the fact that there are still some star
atlases and other datasets which refer to 1950.

Times, unless stated otherwise, can reasonably be presumed to be UTC, despite
the fact that there are several other time reference frames -- such as
Terrestrial Time, GPS time, etc.

For the IAU/Cospar convention of four decimals of degrees, one might presume
that there is a common datum for all listed co-ordinates, but that would be
wrong.  The level of precision which is shown by many astronomers, amateurs as
well as those who ought to know better, is hugely finer than the disparity
between the numerous spheroids and geodetic datums.

Although there may be one or two people who would prefer to have us dumb down
positional accuracy and render geodetic positions meaningless by removing their
underlying basis, no good can be served by reducing rather than improving the
quality of people's observations and their stated positions.

Several people have written asking about the most common datum in the (lower
48) States and wondering what the most likely spheroid and datum is when none
is specifically mentioned.

There are two commonly used systems in the contiguous States of the US.  NAD83
is functionally identical to WGS84 for most practical purposes. It is now the
datum which is preferred by the USGS and is the one which you see used on such
online services as and one or two others.

The other system in the lower 48 is its predecessor which predates satellite
positioning by several decades.  That is NAD27.  It is based upon a quite
different spheroid and uses the Meade's Ranch datum.  For astronomical
purposes, it is quite sufficient to simply label NAD27 co-ords as such.  NAD27
is neither more nor less "correct" than NAD83, merely different.

Last week Japan formally abandoned its own national mapping datum in favour of
WGS84 and promulgated that fact widely among its user base.  The difference is
about 400 metres.  For many years to come, there will still be plenty of
"legacy" datasets which are based on the "old" system, just as there are in
Hawaii, for example.  That is not a problem because the responsible course of
action is to label any geodetic co-ordinates with their geodetic basis to make
them useable.

If visual observations are made to a precision of of an arc-second of
Declination and someone wants to match that observation to another made
elsewhere on another datum, then a conversion to a common geodetic framework
can easily be computed to get the best out of any observational accuracy.  Such
commonality is only possible if the geodetic basis of a position is explicitly

Those who are familiar with the SeeSat/VSO FAQ will be aware that the custom is
to measure time with reference to UTC.  It is also requested that the
"observer's exact location" should be measured by means of "geodetic maps".  A
geodetic map which omits the geodetic basis of its plotted data is not a
geodetic map at all.  In order for such co-ordinates to make any sense, they
must refer to a recognisable spheroid and geodetic datum.  Otherwise the
apparent precision, which may or may not be commensurate with its actual
accuracy, is rendered effectively meaningless.

For all practical purposes, NAD83 may be regarded as synonymous with WGS84, as
may Euref89, Sweref99 and all of the ITRFs of all epochs.  Other datums should
not be presumed to be similar to eachother or to WGS84.  Differences of 400
metres or more are not at all uncommon and any such differences might become
additive if two or more different observation locations are used in a single
computation.  The efforts of those who strive for single arc-second (30 metres)
precision  in their reported observations are blown away by an immediate error
of a couple of stations each having 400 metres disparity, summing to a
systematic error which is something like half a mile.

Most observers who take the time and trouble to measure their observation
locations to a precision of a tenth of a mile or better do so because they care
about standards and about doing the best they can to meet the IAU/Cospar
conventional standards.  It seems a pity to make a mockery of their efforts by
omitting the fundamental basis of those measurements and leaving the reader to
flounder in guesswork when trying to make sense of a declared position which
has an apparent precision of 11 metres or less.

I, for one, am glad that the IAU and Cospar standards are not being eroded by
those who would try to dumb down accuracy and precision and drag us all down to
the level of the lowest common denominator.

I applaud those who take the time and trouble to make sense of otherwise
insensible terrestrial co-ordinates by the simple expedient of openly declaring
the basis of stated Lat/Longs.  They have nothing to hide and nothing to gain
from hiding the basis of their published co-ordinates.

Cheers,       Chris Olsson
All generalisations are false

Unsubscribe from SeeSat-L by sending a message with 'unsubscribe'
in the SUBJECT to

This archive was generated by hypermail 2b29 : Thu Sep 20 2001 - 17:55:52 EDT