Astrometric solutions: value of PV1_2

[fredvanner]

Last year I noted in a General Forum thread, that I believe the value produced by PixInsight for the FITS header keyword PV1_2 (the native latitude of the fiducial point), at least for zenithal projections such as TAN, is not correct.
"M. R. Calabretta and E. W. Greisen - Representations of celestial coordinates in FITS" states:

Since all zenithal projections are constructed with the pole of the native coordinate system at the reference point we set (0; 0)_zenithal = (0; 90)

So we should usually find:
PV1_2 = 90
The incorrect value may cause problems with FITS export to other applications (as in the thread cited above).

 

[Andres.Pozo]

Hi Fred,

the WCS specifications do not say that PV1_2 has to be 90 for all zenithal projections. In fact, the value of PV1_2 has different meanings for each zenithal projection. You can read the description of the zenithal projections in subsections 5.1.x of https://www.aanda.org/articles/aa/pdf/2002/45/aah3860.pdf. The table 13 on page 1119 provides a summary of the parameters of each projection.

For the gnomonic (TAN) projection, the value of PV1_2 has no meaning. Maybe PI should not write it, but:
1) the value PV1_2=0 should not be harmful.
2) at https://www.atnf.csiro.au/people/mcalabre/WCS/example_data.html by Mark R. Calabretta (one of the authors of the specifications) the TAN projection example has the keyword PV1_2=0 !!

Because of this, I think this is not a PixInsight bug.

Andrés.