HEASARC Staff Scientist Position - Applications are now being accepted for a Staff Scientist with significant experience and interest in the technical aspects of astrophysics research, to work in the High Energy Astrophysics Science Archive Research Center (HEASARC) at NASA Goddard Space Flight Center (GSFC) in Greenbelt, MD. Refer to the AAS Job register for full details.

Search:

[Advanced Search]



Creating NICER Light Curves

Overview

Producing light curves is the first step of temporal X-ray data analysis. This thread introduces how to make NICER light curves using the standard HEASoft astronomical tools.

Read this thread if you want to: Produce light curves from NICER data using Xselect

Last update: 2022-06-08

Introduction

NICER scientific analysis ultimately leads to spectral analysis and light curve (timing). This thread provides an introduction to generation of light curves from NICER data.

After running the standard processing pipeline nicerl2, a cleaned, MPU-merged event file is produced. This output product is the "cl" file, and this is the event file which you will use for light curve analysis. We will use the tool "xselect" for light curve extraction.

The tool we use to bin the data is the HEASoft XSELECT package and the output data type is a light curve file.

This thread describes a hands-on example of producing light curves in various time bins or energy bands using the Xselect HEASoft tool. For the purposes of this tutorial we will assume that you wish to make a light curve from a single, MPU-merged, cleaned event data file from a single OBSID. This example takes the NICER data of the nearby sun-like star kappa1 Cet on December 10, 2019 (ObsID 2300020114) and produces a day-long light curve. It does not provide a pulsation search or other advanced timing analysis.

Prerequisites

Here is what is needed:

  • The HEASoft software package including XSELECT installed (see Setting Up a NICER Analysis Environment thread for more information)
  • NICER/XTI cleaned event data (in obsid/xti/event_cl/niNNNNNNNNNN_0mpu7_cl.evt), which is the so-called "cl" file.

Initial Steps

We recommend that you reprocess and rescreen NICER data with the latest NICER calibration using nicerl2 before starting the analysis (please see the analysis thread "nicerl2" for more information). This example uses an additional conservative screening criterion excluding high cut-off rigidity intervals with COR_SAX > 1.5, as the target has a moderate X-ray flux.

First, change the working directory to the location of your observation data files. cd /NICER/data/top/directory In this example, we will be working with Observation ID 2300020114, so you can verify that this directory exists. ls 2300020114/ The output of the "ls" command produces one directory which is our observation's data directory.

In this example we will reprocess to only include data in the COR range of >1.5 GeV/c, so we will use the nicerl2 command, nicerl2 2300020114 clobber=YES cor_range="1.5-*" Here,

  • 2300020114 is the observation ID (substitute your own observation ID)
  • cor_range gives the range 1.5-**, which means 1.5 and above
For more discussion of options to nicerl2, please see the "nicerl2" analysis thread.

After running nicerl2, you find reprocessed event files under obsid/xti/event_cl/. In this example, we can run the "ls" command to reveal the contents of that directory. term> ls 2300020114/xti/event_cl/ ni2300020114_0mpu0_ufa.evt ni2300020114_0mpu2_ufa.evt ni2300020114_0mpu4_ufa.evt ni2300020114_0mpu6_ufa.evt ni2300020114_0mpu7_ufa.evt ni2300020114_0mpu1_ufa.evt ni2300020114_0mpu3_ufa.evt ni2300020114_0mpu5_ufa.evt ni2300020114_0mpu7_cl.evt You can see that the file ni2300020114_0mpu7_cl.evt is the cleaned event file (or "cl" file). This is the file which we will use for subsequent analysis. It is important that you use the "cl" file and not another file.

Producing Light Curves

Now we will use the xselect environment to extract a light curve. We move to a working directory and launch Xselect.

term> xselect ** XSELECT V2.5a ** > Enter session name >[xsel5471] lc

Type any session name you like. This name will be the header of the Xselect's temporary files. You can relaunch the session with this name if you save it at the exit.

You can read a cleaned event file (*_cl.evt) by typing "read events". Xselect asks you to enter a relative or absolute path to the event_cl directory and event file name. lc:SUZAKU > read events > Enter the Event file dir >[./] 2300020114/xti/event_cl > Enter Event file list >[] ni2300020114_0mpu7_cl.evt Got new mission: NICER > Reset the mission ? >[yes]

Xselect will prompt you for the event file's working directory and the name of the file. You can also specify the event file name directly in the command read events 2300020114/xti/event_cl/ni2300020114_0mpu7_cl.evt and the results will be the same.

Xselect will print some diagnostic information about the event file to the console. Notes: XSELECT set up for NICER Time keyword is TIME in units of s Default timing binsize = 16.000 Setting... Image keywords = RAWX RAWY with binning = 1 Energy keyword = PI with binning = 1 Getting Min and Max for Energy Column... Got min and max for PI: 0 1500 Got the minimum time resolution of the read data: 0.40000E-07 MJDREF = 5.6658000777593E+04 with TIMESYS = TT Number of files read in: 1 ******************** Observation Catalogue ******************** Data Directory is: /local/data/puffin1/kenji/Instrument/NICER/data/2300020114/xti/event_cl/ HK Directory is: /local/data/puffin1/kenji/Instrument/NICER/data/2300020114/xti/event_cl/ OBJECT OBS_ID DATE-OBS 1 kap01_Cet 2300020114 2019-12-10T17:35:52 lc:NICER-XTI-PHOTON > Xselect finds the mission name in the event file and asks you to change it if NICER is not your default mission in Xselect. You type "yes" or just return, and then Xselect optimizes some analysis parameters for NICER data and changes the prompt mission name to "NICER-XTI-PHOTON."

Once we have loaded event data into Xselect we can produce a light curve. Type the "extract curve" command to generate a light curve with the default 16 sec time bins.

lc:NICER-XTI-PHOTON > extract curve

Xselect will do some work and produce a light curve. The console output indicates some statistics about the events that xselect processed. extractor v6.07 8 Apr 2022 Getting FITS WCS Keywords Doing file: /local/data/puffin1/kenji/Instrument/NICER/data/2300020114/xti/event_cl/ni2300020114_0mpu7_cl.evt 100% completed Total Good Bad: Time Phase Grade Cut 71630 71630 0 0 0 0 =============================================================================== Grand Total Good Bad: Time Phase Grade Cut 71630 71630 0 0 0 0 in 4153.0 seconds Fits light curve has 7.1630E+04 counts for 17.25 counts/sec Thresholding removed significant counts Try exposure=0.0 on the extract command in xselect or lcthresh=0.0 if running extractor stand-alone Thresholding removed more than half the bins Try exposure=0.0 on the extract command in xselect or lcthresh=0.0 if running extractor stand-alone

You can see the light curve on a PLT/QDP window. lc:NICER-XTI-PHOTON > plot curve PLT> rescale y 0 50 NICER X-ray Light curve with 16 second bins

Choosing A Different Time Bin Size

You can change the time bins by the "set binsize" command. By default the bin size is 16 seconds. We can make the time bin smaller, which allows one to see more fine details (but with larger statistical errors), or make the time bin larger which allows one to see coarser details with more statistical accuracy.

In this example, we will choose 100 sec bins.

lc:NICER-XTI-PHOTON > set binsize 100 and then run "extract curve" again. lc:NICER-XTI-PHOTON > extract curve We can again plot the light curve with the "plot curve" command. lc:NICER-XTI-PHOTON > plot curve PLT> r y 0 50 NICER X-ray Light curve with 100 second bins

Choosing A Different Energy Range

The above light curves include all cleaned event data between 20-1501 PI channels, equivalent to 0.2-15 keV. Since in this example the target is a soft source, we limit the energy range between 0.3-2 keV (30-200 PI channels) with the "filter pha_cutoff" command.

For this filter, we give the lower and upper pulse height range. In our example, we are looking for the energy range 0.3-2 keV, so we want to use pulse height bins 30 - 200. (see Gain Calibration for more information about NICER energy channel binning). lc:NICER-XTI-PHOTON > filter pha_cutoff 30 200 We can again plot the light curve with the "plot curve" command. lc:NICER-XTI-PHOTON > plot curve PLT> r y 0 50 NICER X-ray Light curve between 0.3-2 keV with 100 second bins

The light curve does not change remarkably though the total counts reduce slightly from 71630 counts to 63372 counts.

By contrast, we can examine the 5-10 keV range and see that the light curve is quite different. The 5-10 keV range corresponds to PI channels 500-1000. In this higher energy range, most events originate from particle background instead of from the target. To make this second lighr curve, first we reset the PHA filter and then set the filter between 500 and 1000 and then extract a new light curve.

lc:NICER-XTI-PHOTON > clear pha_cutoff lc:NICER-XTI-PHOTON > filter pha_cutoff 500 1000 lc:NICER-XTI-PHOTON > extract curve lc:NICER-XTI-PHOTON > plot curve PLT> r y 0 50 NICER X-ray Light curve between 0.3-2 keV with 100 second bins

Saving Light Curve

If you quit xselect, your results will not be saved by default. You can save the light curve in a FITS file for further analysis. Use the "save curve" command in xselect, which will prompt you for a file name.

lc:NICER-XTI-PHOTON > save curve > Give output file name >[lckese] lc03020.fits Wrote FITS light curve to file lc03020.fits Or, you can specify the light curve file name on the command line, like this, save curve lc03020.fits

Producing a Light Curve from Multiple Datasets

You can make light curves from multiple NICER datasets in the same way by simply listing up event files when loading event files with "read events."

Next Steps: Related Topics

You can feed the saved FITS light curve into downstream tools for further analysis. For example, HEASARC provides the Xronos package for plotting multiple light curves, making a hardness ratio time-series, and so on.

Modifications

  • 2022-06-08 - proofreading
  • 2022-04-27 - initial draft