On Fri, Mar 02, 2018 at 05:57:26PM +0100, Radek via Therion wrote:
> Here's more accurate
That uses the same model as Therion does, so I can't see why it should
be more accurate (unless you're using a version of therion more than a
year old, as before https://github.com/therion/therion/pull/52 Therion
calculated the fraction of the year to use in a slightly odd way, though
even then the difference is tiny if you're specifying a full date).
Also, if you look up values by hand and specify them explicitly to
therion you are hard coding the answers for the current version of the
IGRF model. But each generation of the IGRF model is necessarily
predictive for dates after it was issued and is only declared to be
definitive for dates more than 5 years before it was issued. So each
new generation of the model can change the declination figures reported
for the last 10 years compared to the previous generation (and the new
answers should be expected to be better).
As Xavier suggests, the table of reported declinations in the log is for
the start of each year, but declinations used for surveys are calculated
for the actual survey dates.
Maybe this needs making clearer somehow, as this isn't the first time
this summary table has led someone to think that the declination was
only being calculated for the start of each year:
Therion mailing list