About WISE IR data documented in zcatalog

Question

Hi,

I find that in your zcatalog, there are a few rows related to WISE, like FLUX_W1 and FLUX_W2 which I suppose these should be the photometry of the target corssmatching to NEOWISE (since only W1 and W2 but, in fact, it doesn't matter which WISE catalog exactly) catalog. I cannot find detailed description to these rows but I find it disagree with the results if one directly searching from IRSA. I haven't thoroughly check the full catalog but for the sample which I'm concerning, this photometry is badly wrong. The same problem also occurs in some value-added catalogs.

There's an easy way to confirm this problem. Since it seems the legacy survey sky browser system adopted the same value, you can randomly find one spectra in sky browser system and check the data. Then put that location into NASA/IPAC WISE catalog and see the photometry therein. 

I just give one random target: https://www.legacysurvey.org/viewer/desi-spectrum/edr/targetid39627782233263814 

As you can see the RA and DEC is 211.4448 and -0.3605 respectively. And the magnitude of W1 and W2 is 18.31 and 18.04.

You can put that coordinate into this website, matching within 1 arcsec, and see what returns. https://irsa.ipac.caltech.edu/cgi-bin/Gator/nph-dd

Only one matched sources with a distance less than 0.2 arcsec. The W1 magnitude is 15.665 and W2 magnitude is 14.732. The difference is so large that it can not be correct evenif you are using some other WISE observation epochs. Since this is a QSO target from DESI spectra, it is natually expected that the W1-W2 would be around 1.

I'm using the galaxy mass from the SED fitting results. Thus, I really want to know if I can believe the results from those value-added catalog. Please let me know if I have any misunderstand to these data.

Hope to hear from you soon.

Answer 1

The most likely explanation is that Legacy Survey WISE data (which is what’s used in the zcatalog) is in AB magnitudes, while AllWISE is in Vega magnitudes. This should explain the difference in the quoted numbers nearly exactly. See the LS DR10 photometry discussion for further details.

The second possibility would only be an option for extended sources where you do see some differences between LS and unWISE or AllWISE magnitudes. Since this example is a QSO, that is unlikely.

One other thing to note is that Legacy Survey photometry is forced-photometry based on the optical positions, not a cross-match, so the resolution of pixels into sources can differ.  That’s only really significant when blending is an issue.

Comments

The difference in magnitude system should be the reason here. I didn't realize that the system difference would be such huge in IR band. Thanks for your prompt reply.