Units of aperture photometry output table

[hcferguson]

I expected that aperture_photometry -- when fed data in units of MJy/sr and aperture radii in arc seconds -- would output in some units that integrate over area. Instead, the output is in MJy/sr.

In my case I fed it an NDData array drawn from a JWST level 3 data file:

hdu=fits.open('/Volumes/CEERS/ceers/CEERSJune2022/v0.07/ceers_nircam1_f200w_v0.07_i2d.fits')
flux_unit = u.Unit(hdu[1].header['BUNIT'])
data = NDData(hdu[1].data*flux_unit,StdDevUncertainty(hdu[2].data*flux_unit),wcs=wcs)

phot_table = aperture_photometry(data,aperturelist[0])

print(phot_table)

id      xcenter            ycenter            sky_center          aperture_sum     aperture_sum_err 
           pix                pix               deg,deg              MJy / sr           MJy / sr     
--- ------------------ ------------------ -------------------- ------------------- ------------------
  1 399.90599357353494 3526.4940214727003 214.943153,52.942449 0.01577787901579053 0.1537752456576211

[larrybradley]

As you noticed, aperture_photometry simply returns the aperture_sum as the sum of the (weighted) pixel values within the aperture. I'll consider this a feature request to handle images in surface brightness units. The tricky part here is that I don't know of a good (non-hacky) way to automatically determine if the input data are in units of SB. This feature may need a new keyword flag to turn on this mode.

EDIT: This feature would only work for non-geometrically distorted data.

[larrybradley]

I'm sure you know this, but you can simply multiply the data by the pixel area before inputing it into aperture_photometry. That's what I do in practice.

Given the wcs object, you can use this to get the pixel area:
wcs.proj_plane_pixel_area().to(u.steradian)

or for JWST i2d data, get the value of the 'PIXAR_SR' header value.