The output event list is a clone of the input event list in which the tables EVENTS, STDGTI0n, EXPOSU0n and REJPIXnn have been destructively filtered and the node-specific exposure maps have been deleted or replaced. The STDGTI0n tables and the REJFLAGS attribute of the EVENTS table indicate what filtering has been performed. The specification of the node-specific exposure maps follows:
A node-specific exposure map is an image (FITS array) on a channel grid that is commensurate with the channel coordinate system defined in the EVENTS table, but covering only the exposed surface of a single node. Array values are seconds of exposure in single-precision, floating-point format. The two dimensions of the array are dispersion and cross-dispersion, arranged such that values adjacent in dispersion are adjacent in memory. The following attributes are provided:
The two exposure maps for a chip are not constructed unless the EXPOSU0n table for that chip is present in the input event list. However, if this table does exist and contains no frames, two null exposure maps are added to the output. A null map has zero extent in both dimensions. In this case the WCS, the coordinate transform, and the data range attributes above cannot be defined and are omitted.
If enabled, the combined exposure map is placed in this output dataset as the primary image (FITS array). It is formatted in the same way as a node-specific exposure map, except that the channel grid covers the full range specified in the EVENTS table. Coordinates are identical between this map and the event list. The following attributes are provided:
Attributes TELESCOP, INSTRUME, OBS_ID, EXP_ID, EXPIDSTR, DATE-OBS and DATE-END are also copied from the primary header of the input event list. In the unusual case that there is no exposure on any of the chips (all of the node-specific maps are null) the combined map is zero everywhere and the attributes T_BETA_N and T_ORIG_N, which cannot be defined, are omitted.
