It’s not an issue of depth or exposure time – MzLS is much deeper than SDSS z band – it’s an issue related to estimating the background around bright galaxies.
DR10 will not change this - DR10 only processes the DECam data (south).
Yes, I know DR10 is a DECaLS expansion and it may not happen with the northern survey, but is there a way to add extra Z-band (or edit the z-band data from MzLS) exposure and make the images a lot more equal?