Data Product Types
PTF uses a nomenclature for types of data products common to NASA missions, grouping data into “Levels” of processing. This nomenclature is introduced here because the data retrieval user interface uses it. Each level number is in some way derived from the level that immediately precedes it. Data prodcuts are additionally split into "primary" and "ancillary" data products. The archive searches for data by identifying the primary products; ancillary products are then located by association with the primary.
- Level-0 (raw data) – these are multi-extension FITS files (one per CCD) as produced by the camera. These are usually just referred to as "raw" frames.
- Level-1 (single exposure) – these are processed data products corresponding to the level-0 data. The PTF level-1 primary data are images. The ancillary data are mask files and catalog files. This data is usually referred to as "epochal" or "frameproc" data.
- Level-2 (ensemble processed) – these data are derived from level-1 data. These are known variously as "coadds" or "reference images". Data Release 1 included no Level-2 products. Data Release 2 will include reference images and reference catalogs derived from these images. The coadded image is the primary products. Ancillary products include psf and aperture-photometry catalogs, coverage maps, and uncertainty maps,
File Naming Conventions
PTF data products use a uniform naming convention designed for easy identification and sorting. All files of a given type have the same fields, and the fields are of fixed length. Underscores are used as delimiters.
- PTF – the first field indicates the survey, and is always the same for all PTF data.
- 200905021538 – this is the datestamp of when the exposure was taken. Of the form YYYMMDD.DDDD.
- i_p_scie – this string indicates the file type.
- t034128 – timestamp of the hour when the data was taken, of the form HHMMSS.
- u005371541 – this is a unique serial number for this specific data product. This is a random non-repeating number assigned by the database.
- f02 – this is the filter id. 01=g-band, 02=R-band, 11=H-alpha (6560), 12=H-alpha (6630). At the current time, only g- and R-band data are publicly available.
- p003787 – this is the PTF field id. This references the specific location on the sky in the PTF tiling system.
- c11 – CCD id, ranging from 0 to 11.
- .fits - the suffix ".fits" indicates an image. ".ctlg" indicates a catalog, which is actually a binary fits table.
This convention has the effect that the natural sorting order will list all images in a single exposure together, as a function of time. Since each field is prefixed by a unique letter, it is also very easy to use wildcards to select subsets of data.
The reference image filenames are slightly different, and are designed for sorting by field (tile) id, filter, and chip. For example:
- PTF - the first field indicates the survey, and is the same for all PTF data.
- d003787 - the PTF field id, i.e. tile number.
- f01 - the filter (1=g-band, 2=R-band).
- c11 - the CCD id.
- u000131960 - the reference image ID number, a unique serial number assigned to each image.
- p12 - the pipeline ID number. Several versions of the reference image pipeline exist for various purposes, and this number is not very relevant to end users. The archive will always retrieve the "best" image.
- refimg.fits - this indicates the filetype. "refimg.fits" is the actual flux image. Other files include "depcov.fits" (depth of coverage in units of input images), "uncert.fits" (uncertainty image), "psfcat.ctlg" (DAOPhot-based psf extraction catalog), and "sexcat.ctlg" (Sextractor-based aperture extraction catalog).