Using reproject_from_healpix produces an array with 3 dimensions

[Kimmo1]

Hi

I am using reproject_from_healpix with a Healpix map of hydrogen HI data (LAB_fullvel.fits), which is available at
https://lambda.gsfc.nasa.gov/product/foreground/fg_LAB_HI_Survey_get.html
Also the Fits header of the file is available on the same page.

I am trying to reproject part of the Healpix map into tangential projection, with RA and Dec coordinates. The Python program I used is at the end of this message. The program works without errors, but the array which the reproject_from_healpix produces has three dimensions, like (100,100,1024)

What is the reason for this, where is that axis with a length of 1024 coming from ?

As far as I understand, the Healpix data is not a data cube, having a third axis for velocity.

The Fits header has
TFORM1 = '1024E ' / data format of field: 4-byte REAL
Does this has something to do with the extra dimension with a length of 1024 ?

Thanks for any help, Kimmo Lehtinen

---

from astropy.io import fits
from astropy.wcs import WCS
from reproject import reproject_from_healpix

hdu = fits.open('LAB_fullvel.fits')[1]

target_header = fits.Header.fromstring("""
NAXIS = 2
NAXIS1 = 100
NAXIS2 = 100
CTYPE1 = 'RA---TAN'
CRPIX1 = 50.5
CRVAL1 = 239.2
CDELT1 = -0.3
CUNIT1 = 'deg '
CTYPE2 = 'DEC--TAN'
CRPIX2 = 50.5
CRVAL2 = -53.1
CDELT2 = 0.3
CUNIT2 = 'deg '
COORDSYS= 'icrs '
""", sep='\n')

data = hdu.data['TEMPERATURE']

array, footprint = reproject_from_healpix((data, 'G'), target_header, nested=False)