Photometry of the MUSE UDF fields

The fitting algorithms in the imphot module were tested using MUSE observations of 10 MUSE fields in the Hubble UDF, denoted UDF01 to UDF10. The image fitting algorithm worked well for MUSE fields UDF02, UDF03, UDF08, UDF09, UDF10, where there are no bright point sources. However bright point sources in fields UDF01, UDF04, UDF05, UDF06 and UDF07 caused problems until they were excluded. This was due to two effects.

  1. The stars in the UDF fields have sufficient proper motion that they had moved significantly in the 11 years that elapsed between the original HST UDF observations and the MUSE UDF observations.

  2. The HST WCS suffered from a CTE charge loss effect, and this had a much larger effect on point sources against a dark background, than other sources. The result is that if one tries to scale the fluxes of an HST image to match a MUSE image, a different scale factor is needed for unresolved point sources, such as stars, than for the resolved sources, such as galaxies. Stars need to be scaled by 10% to 20% more than galaxies.

Both of these issues break the assumption that the only differences between an HST and Hubble image of a field are caused by instrumental differences and atmospheric conditions. To resolve this problem, stars and QSOs were excluded from the fits, by passing the fitting procedure a ds9 region file that excludes small circular regions centered on each star.

The following pages discuss the procedures used to fit the photometric parameters of each of the MUSE UDF fields. The goal of these pages is to guide the processing of future observations of these fields, and to provide real-world examples for processing other fields.

The UDF01 field contains both a star and a QSO, so it provides examples of the problems noted above.