Known Problems in frames

 

Throughout this section, we have listed various known problems with the processing, many of which will be fixed in future releases of the data. We now list problems that were not otherwise mentioned earlier.

• There are several quantities that have not yet been finalized, and are currently place-holders in the pipeline outputs. These include error estimates for PetroR50 and PetroR90, all errors associated with isophotal quantities, a texture parameter to measure the small-scale roughness of an object, and fracPSF, the fraction of light included in the PSF. Quantities which are not calculated are designated -9999 in the database, while errors which are not calculated are designated -1000. Note that some quantities are not calculated, e.g., for objects very close to the edge of a frame.
• We have not done a thorough testing of every one of the outputs of the pipeline. Quantities such as psfMag and petroMag, which are used in target selection (§ 4.8), are very extensively tested, while quantities such as the parameters of the elliptical isophote fit, or the errors in the model fit parameters, have not been as carefully vetted.
• Due to difficulties in cosmic-ray rejection, one should be suspicious of objects detected in only one band. This is particularly problematic in z, where the occasional cosmic ray can splatter in the thick chip to be indistinguishable from the PSF (see the discussion in fan01b). Future releases of frames will set a flag MAYBE_CR to flag ambiguous cases.
• Electronics cross-talk can give faint ghosts of saturated stars in z. Draw a line down the central column of the CCD; the ghost will appear at the mirror position relative to this line. These will be flagged in future releases.
• The code does not subtract diffraction spikes from stars (although bleed trails are interpolated over). These will be subtracted in future releases.
• There are a variety of faint ghosts due to reflection off the edges of the filters. Thus one should be suspicious of the reality of very low surface brightness (fainter than 24 mag per square arcsec in r) extended features close to the edges of frames, especially in the u band.
• Similarly, the deblender occasionally pulls off low surface brightness features from large galaxies. This problem is much reduced with improvements to the deblender in subsequent versions of the pipeline.
• The proper motion errors are occasionally underestimated (yielding a few quasars with nominally significant proper motions!). This is due to uncertainties in the astrometric offsets between bands, and will be fixed in future releases. However, overlapping scans have shown that the proper motion errors for the majority of asteroids are overestimated by roughly a factor of two.
• Ghosting in the u filter causes systematic errors in the flat-fields in some of the runs, affecting the photometry by up to 10%. This is discussed in detail in § 4.5.3 below.

Finally, a disclaimer: we have tested the outputs of the pipeline to the best of our ability, but there are no doubt subtle problems not listed here that we are currently unaware of. For example, many of the quoted error estimates have not yet been tested adequately. We are eager for feedback from the community if any problems are found.