Oberon Version

Ian M George: Temporal Analysis

These pages are unofficial in all respects. They are primarily simply notes to myself. The information is not guaranteed to be correct or up-to-date.

Last Updated: 2004 Jul 10

Make Standard Light Curves

Here we are going to use the dispersed spectra to make a light curve of the target in the 0.7 - 8.0 keV band. Before making the light curve, we need to get some ancillary input files.

cp /data/oberon5/george/data/axaf/scripts/*wi* ./

Most of the light curves we are after can be made using:

/data/oberon5/george/data/axaf/scripts/mk_hetgs_1024_lcurves.csh

However, the script does not make one we require later, so lets make it by hand, first by filtering the event file (actually using the wavelength [from the position of each event], rather than the energy [as detected by ACIS-S]):

fselect meg_clnd.evts meg_clnd_o1_src_0780.evts "((TG_M.EQ.1).OR.(TG_M.EQ.-1))AND.(TG_D.LE.0.0005).AND.(TG_D.GE.-0.0005).AND.(tg_lam.GE.1.55).AND.(tg_lam.LE.17.71)" clobber =yes
fselect heg_clnd.evts heg_clnd_o1_src_0780.evts "((TG_M.EQ.1).OR.(TG_M.EQ.-1))AND.(TG_D.LE.0.0005).AND.(TG_D.GE.-0.0005).AND.(tg_lam.GE.1.55).AND.(tg_lam.LE.17.71)" clobber =yes

So now we can make the light curve using a couple of FTOOLS:

lcurve nser=2 cfile1=meg_clnd_o1_src_0780.evts  cfile2=heg_clnd_o1_src_0780.evts window=2files050050.wi dtnb=1024 nbint=450000 tunits=2 clobber=yes outfile=megheg_clnd_o1_src_0780_1024  plot=no 
fselect megheg_clnd_o1_src_0780_1024.flc+1 megheg_clnd_o1_src_0780_1024_snip.flc expr='FRACEXP >= 0.95' clobber=yes

The above steps will result in a set of files:

     megheg_clnd_o1_src_0710_1024_snip.flc
     megheg_clnd_o1_src_1030_1024_snip.flc
     megheg_clnd_o1_src_3080_1024_snip.flc
     megheg_clnd_o1_src_0780_1024_snip.flc

These can be plotted using the FTOOL fplot, eg:

fplot infile="megheg_clnd_o1_src_0780_1024_snip.flc+1" xparm="TIME[XAX_E]" yparm="SUM12[SUM_E]" rows ="-" device ="/XW" pltcmd=" "

VISUAL CHECK: What we are checking here is (i) that there are no unexpected breaks or drop-outs in the data train, (ii) that the light curve is 'reasonable', and (iii) that the source does not exhibit variability characteristics such as to make 'meaningless' some/all of the standard spectral analysis we will be performing later.

HARDCOPY: Scott, please make a hardcopy of the plot resulting from at least the

fplot infile="megheg_clnd_o1_src_0780_1024_snip.flc+1"...{blah}..{blah}..

step above (ie printout of at least the light curve over the full 0.7 - 8.0 keV band).

Make Hardness Datasets

[New 04jun21; Updated 2004 Jul 10]
OK, so we have made the FITS light curves, now we want to see if/how the "hardness ratio" changes with time.
We have some old (crude) code to do this, but unfortuantely the code takes ASCII files as input.
Thus the first step to to dump the data to the ASCII files. This can be achived by repeating the above fplot
command for each file:

fplot infile="megheg_clnd_o1_src_0780_1024_snip.flc+1" xparm="TIME[XAX_E]" yparm="SUM12[SUM_E]" rows ="-" device ="/XW" pltcmd="we megheg_clnd_o1_src_0780_1024_snip.qdp "
fplot infile="megheg_clnd_o1_src_0710_1024_snip.flc+1" xparm="TIME[XAX_E]" yparm="SUM12[SUM_E]" rows ="-" device ="/XW" pltcmd="we megheg_clnd_o1_src_0710_1024_snip.qdp "
fplot infile="megheg_clnd_o1_src_1030_1024_snip.flc+1" xparm="TIME[XAX_E]" yparm="SUM12[SUM_E]" rows ="-" device ="/XW" pltcmd="we megheg_clnd_o1_src_1030_1024_snip.qdp "
fplot infile="megheg_clnd_o1_src_3080_1024_snip.flc+1" xparm="TIME[XAX_E]" yparm="SUM12[SUM_E]" rows ="-" device ="/XW" pltcmd="we megheg_clnd_o1_src_3080_1024_snip.qdp "

typing quit at the PLT> prompt each time. [Sorry this is so crude!]

[Code & text Updated 2004 Jul 10]
OK, so now we are ready to use the {personal} task lc_hardness.e. This is not in any way sophisticated piece of code -- it merely takes 3 ASCII files as input and calculates the ratio of the first two files, and makes two 'qdp' files as output. All 3 input files simply simply list the count-rate detected in a given (standard) energy band as a function of time (& are produced in the previous step). The first output file gives the above ratio (file1/file2) as a function of time. The second output file is this same ratio as a function of the simultaneous count-rate in the 3^rd file (ie. count rate in the 3^rd energy band).

Standard Softness vs Hard

[Basically New 2004 Jul 10]
So, the first thing we are going to do is to look at the ratio of the counts observed in the 0.7-1.0 keV band to those observed in the 1.0-3.0 keV band. We are going to look at this ratio as a function of time, and as a function of the counts observed in the 3.0-8.0keV band. This standard combination can be performed using (the new, improved version of lc_hardness.e via:

/local/home/oberon/george/bin/lc_hardness.e
  LC_HARDNESS version 2.0 (2004 July 10)
  Mode (S, H, or I)
  S

This response to the Mode (S, H, or I) prompt with make assumptions regarding the i/p filenames, and assign names to the o/p files. Thus, on successful completion, this will produce two files

     megheg_0710on1030vs3080_1024_vs_time.qdp
     megheg_0710on1030vs3080_1024_vs_flx.qdp

which can be viewed using qdp (for eg)

qdp megheg_0710on1030vs3080_1024_vs_flx.qdp
PLT> log on
PLT> res .01 1 3e-3 .3
PLT> hard
PLT> q
cp pgplot.ps megheg_0710on1030vs3080_1024_vs_flx.ps

Update Note: [Code & text Updated 2004 Jul 10] The code used to be terrible, and often failed. I hope the new version is much more robust (But of course, if you have problems, please let me know!!)

VISUAL CHECK: What we are looking at here is how the overall shape of the source spectrum changes as the overall intensity of the source changes. This will again help us understand whether of not the standard spectral analysis (we will be performing later) is meaningful.

Standard Softness vs Total

[this come soon!]

Possibly More Steps to Come...

Onto the Next Step