Next: BAT Data Up: Data Previous: Data Organization   Contents

HEASARC Data Levels

Swift data conforms to the HEASARC standards. There are three levels of data that are archived.

Level 1--Raw Data
Level 1 data are telemetry (also known as Level 0 data) that have been converted to FITS format files. No data processing is done in the conversion from telemetry format to FITS format. Examples of Level 1 data are the raw images, uncalibrated event lists, and detector plane histograms.

Scientists generally need to start their analysis from Level 1 products only if there have been improvements in the telescope calibration since the most recent data reprocessing, or if non-standard data screening is intended. Images are stored in raw detector coordinates. Event data are unscreened and recorded in raw detector coordinates. Level 1 data are merely the telemetry and TDRSS data after it has been converted to FITS format files.

Level 2--Reduced Data
Level 2 data have been calibrated and have had some data processing done on it. Examples of Level 2 data are detector and sky images, and calibrated event lists.

It is envisaged that most scientists will start their analysis from the Level 2 products. The data reduction pipeline has been performed on images and event tables. Images are stored in J2000 sky coordinates (RA,Dec), although grism data are also stored in corrected detector coordinates. Exposure maps for each individual image are available. Columns containing corrected detector and sky coordinates have been populated. Event data have been screened for standard bad times, e.g. SAA passage and Earth limb avoidance.

Level 3--High-Level Data Products
Level 3 data are source level products. They are science data that can be used by the community. Examples of Level 3 data are source catalogs, calibrated spectra, and light curves.

Level 3 data products are intended to be quick-and-dirty representations of many of the products that the scientist would routinely create during their analysis; e.g. time-averaged images, light curves, spectra, etc. Since they are created non-interactively they are not intended as publishable products but as useful guides to the scientist. The number and type of Level 3 products in the archive will be determined by the confidence in any non-interactive burst detection and/or the time since the burst. More Level 3 products will populate the archive at an early stage if the burst has been identified during standard pipeline processing. Products will take longer to arrive if interactive analysis is required to identify the burst.

Next: BAT Data Up: Data Previous: Data Organization   Contents
Eleonora Troja 2013-09-03