Count rates can be calculated for a source list either by evaluating the exposure image at the source positions or by evaluating the TIM_INT and the LIF_TIM descriptors of the background image. The first method makes it necessary to use the command
READ/EXPOSURE_OF_SOURCES sou_tab exp_ima [lab_rad] [sel_reg]
Note:For sel_reg=CIRCLE slightly different values for the mean exposure are determined compared to the result from INTEGRATE/RING as slightly different methods are used.The exposure is written to a column :EXPOSURE. The exposure image in general contains the exposure time the vignetting. The source count rates and the error in the source count rates are calculated by the table column operations:
:RATE = :CTS / :EXPOSURE :ERATE = :CERR / :EXPOSUREThe second method needs no exposure map. It is automatically applied in the maximum likelihood and upper limit counts calculation command in case the command COMPUTE /LIFE_TIME has been applied to the background image
COMPUTE/LIFE_TIME image1 eventrates bacmp1The image image1.bdf (in this example) is used to get the accepted time intervals and the eventrates table eventrates.tbl to determine the life time in these time intervals.
Note:Choose this second method if no exposure map is available. The values of :CTS and :CERR depend on the chosen radius of the extract circle. Such an extract circle radius is specified in the parameter file for the commands DETECT/MAXLIK and COMPUTE/UPPER_LIMITS in the parameter CUT_RADIUS. As no point spread function correction is applied to the counts determined by the maximum likelihood algorithm, the count rates may be underestimated. It is not at all recommendable to use a much larger value for CUT_RADIUS than the default value as then close by sources may be contributing and give unreliable results. Sometimes it may even be appropriate to choose a CUT_RADIUS of about 1.0 (in the pointing case) to get reliable source positions. A second run of the command DETECT/MAXLIK or COMPUTE/UPPER_LIMITS with the parameter settings FIT_POSITION=fixed and CUT_RADIUS=2.5 may then be recommendable in order to get reliable counts.