The FTOOLS provide a suite of programs that can be used to clean, merge, and background subtract a lightcurve.
A method for extracting background-subtracted lightcurves from the 1-3 Å Copernicus data is described below.
Merging and Cleaning the Raw data file : There are typically many files associated with a Copernicus observation. The FTOOL fmerge creates a single file from a given input file list. When running fmerge, the parameter lastkey should be set to TSTOP and all the columns should be included.
There are two basic parameters used to clean the data: the high tension (set to 1 to represent detector on) and the observation constraints (set to 1 to represent no constraint). In the FITS file, the column name containing the observation constraints is OBS_CNSTR, and for the 1-3 Å data the high tension column name is HTF_1_3. However, it is good practice to always plot, using fplot , the values stored in the columns SEN_1_3, BKGMODE, LOWVLTFL, and ANTELGRP to insure that the settings of these parameters do not change during the observation. The definition of these parameters can be found within the header of each FITS file.
In the simplest case, the basic cleaning of the merged file is obtained using the FTOOL fselect with the following expression:
OBS_CNSTR == 1 && HTF_1_3 == 1Example :
fselect Name of FITS file and [ext#] copernicus.fits Name of output FITS file clean.fits Selection Expression OBS_CNSTR == 1 && HTF_1_3 == 1
Calculating the Background Model : The method to calculate the background is to find the parameters that fit the correlation between the Guard counts (GRD_1_3) and the background counts (CNT_1_3). These parameters can then be used to subtract the background from the source counts. To obtain the correlation parameters, it is recommended to use observations nearby in time and location to the observation of interest. Copernicus observed many UV sources that have no significant X-ray emission (the sensitivity limit was about 3 mCrab) and these data are suitable to determine such a correlation. A good target is Tau Sco, which was observed many times during the mission lifetime. The plot below shows the Guard counts (GRD_1_3) vs. Counts (CNT_1_3) correlation obtained for an observation of the UV source Tau Sco. The fit here was obtained using a linear correlation plus a constant, with values 2.5 x 10-2 and 3.4 for the slope and constant, respectively. These parameters then can be used to background subtract the source lightcurve.
fcalc Name of FITS file and [ext#] clean.fits Name of output FITS file clean_bs.fits Resultant column name CNT_1_3SB Arithmetic expression CNT_1_3 - (3.4 + 0.025*GRD_1_3)The figure below shows the results obtained applying this procedure to an observation of X Per. The original data are shown in the upper panel and the background-subtracted data in the lower.
This background-subtracted lightcurve reproduces Fig. 1 of the White et al. 1976 MNRAS 176 201 paper, where the detection of the X Per periodicity was first reported.