In a previous comment thread and the Bosch et al 2018 paper it is made clear that all flux measurements are relative to the photometric reference catalogue. Given this is the case which measurement is chosen in the current PanSTARRS reference catalogue?
Going forward to Rubin, will this decision be repeated? Presumably it is chosen based on the performance of the fitted colour terms in predicting HSC fluxes. Is it for instance the ‘$bandMeanPSFMag’ column from the public release? If we consider point sources as determined by PanSTARRS processing would these therefore be the most tightly correlated with HSC PSF fluxes?
I am constructing a photometric reference for VISTA by positionally cross matching to the PanSTARRS reference. I am currently using ‘curve of growth corrected’ aperture magnitudes but could also use Kron magnitudes. I am bootstrapping to the previous VISTA catalogues in order to produce catalogues with photometry that is consistent with the previous releases but I suppose this isn’t necessarily relevant. I am also experimenting with using the public 2MASS catalogues and colour terms that were calculated from the public VISTA releases.
The Pan-STARRS reference catalog in LSST format that I distributed merges the single-visit, warp and stack PSF magnitudes in an attempt to span a large dynamic range in magnitude (because HSC saturates at 18th magnitude). @StephenGwyn has pointed out to me that this approach results in asymmetric residuals at the faint end.
Going forward to Rubin, they cannot rely completely on the Pan-STARRS catalog, since it cuts off around Dec ~ -30 deg, and LSST will survey much further south. I expect that they will generate a reference catalog directly from LSST observations, probably using Gaia as an anchor (or, at least as a constraint), although I wouldn’t be surprised to learn that @CStubbs and @rhl have plans to calibrate everything directly to physical units by observing BD +17°4708 and knowing everything about the system.
If you want to merge NIR measurements with the Pan-STARRS catalog, I suggest you check that the colors you get make astrophysical sense, as there could be an offset between the two systems due to different definitions and assumptions (aperture corrections, airmasses, filter curves).
Thank you for your answer. I am seeing asymmetric residuals at the faint end which may be due to me using aperture mags. I will try using PSF mags to see what impact that has.
The declination limit of PanSTARRS doesn’t affect me yet as I am currently attempting to run on the HSC overlap only. However, I am keen to see what the eventual LSST solution will be as we are preparing for that using the HSC public data as a test.