[fitswcs] Adaptive mesh refinement

[William Thompson]

Folks:

I was asked how to store data from model calculations using adaptive mesh 
refinement in FITS files.  My first thought was to use the -TAB projection, but 
I wanted to verify that I understand how it works.

Suppose that we have a model with N points in it, each of which has coordinates 
(x_i, y_i, z_i).  My belief is that I could store this using the following keywords:

NAXIS	= 1
NAXIS1	= 1000
WCSAXES	= 3

CTYPE1  = 'HEQX-TAB'
CRPIX1	= 1
CRVAL1	= 1
CDELT1	= 1
CTYPE2  = 'HEQY-TAB'
CRPIX2	= 1
CRVAL2	= 1
CDELT2	= 1
CTYPE3  = 'HEQZ-TAB'
CRPIX3	= 1
CRVAL3	= 1
CDELT3	= 1

PC1_1	= 1			/Use PC matrix to tie coordinates together.
PC1_2	= 0
PC1_3	= 0
PC2_1	= 1
PC2_2	= 1
PC2_3	= 0
PC3_1	= 1
PC3_2	= 0
PC3_3	= 1

PS1_0	= 'WCS-TAB'
PV1_1	= 1
PV1_2	= 1
PS1_1	= 'COORDS'
PV1_3	= 1

PS2_0	= 'WCS-TAB'
PV2_1	= 1
PV2_2	= 1
PS2_1	= 'COORDS'
PV2_3	= 2

PS3_0	= 'WCS-TAB'
PV3_1	= 1
PV3_2	= 1
PS3_1	= 'COORDS'
PV3_3	= 3

The coordinate array in the binary table would have the dimensions 
(3,1000,1000,1000).  Is this correct?

Thank you,

Bill Thompson


-- 
William Thompson
NASA Goddard Space Flight Center
Code 612.1
Greenbelt, MD  20771
USA

301-286-2040
William.T.Thompson.1 at gsfc.nasa.gov