Skip banner and navigation tools.

 |  site map

Swift-XRT light curve Repository — Online Documentation

Contents

Overview

This web site provides a repository of Swift-XRT light curves, hardness ratios and images for every GRB this instrument has detected. There are four ways provided of selecting a GRB to view: a page of thumbnail images for each GRB, a search box (where you can enter a GRB number, or BAT trigger number), a year/month menu and a panel showing the 5 most recently observed GRBs. Each of these links to the products page for that burst.

Bursts which the XRT did not observe or did not detect do not have thumbnail images, even if they are Swift-detected bursts.

Back to contents | Back to repository

Usage

The data on this repository are available for use in research and publications, provided that this is acknowledged. We request two forms of acknowledgement. Where the data are first introduced, please cite Evans et al. (2007, A&A, 469, 379) and Evans et al. (2009, MNRAS, 397, 1177) which describe how the data were produced. In the acknowledgements section, please use the following text: This work made use of data supplied by the UK Swift Science Data Centre at the University of Leicester.

Back to contents | Back to repository

Products Pages

When you select a GRB, you will be taken to the products page, which is described here. (For details of how these products are created see Evans et al [2007, 2009], and the changelog on this page).

Overview

The products page begins with a title, and then indicates which dataset triggered the most recent update of the light curve. This is followed by a series of links to any other products which exist for this burst, to the rebin interface and this documentation page; and a warning if data for this GRB are still on the quicklook site (i.e. not yet final).

The products follow, presented first in graphical format. There will be (up to) five images, the first 3 of which link to the ASCII data from which they were created. These images are:

  1. Flux light curve (where available) -- the count-rate light curve (described next) converted to observed flux using a constant conversion factor taken from the spectrum repository.
  2. Count-rate light curve -- showing the source count rate, after the background has been subtracted and corrections for pileup, bad columns, and the size of the source-data extraction region have been applied.
  3. "Detailed light curve" -- as above, but with the background level shown on the same plot (note that at low count rates, the net source count rate can lie below the background rate but still be a 3-σ detection). The fractional exposure is also shown, in the bottom panel.
  4. Hardness ratio -- similar to the basic light curve, but in two bands: 1.5—10 keV and 0.3—1.5 keV. The bottom panel is the ratio: hard/soft.
  5. 3-colour deep image (where available) -- an image of the total exposure so far, split into three energy bands to provide colours and smoothed for aesthetic reasons. See note on colour images

Finally, there are a series of links to the data products in formats which cannot be embedded in a web page. i.e. postscript versions of the light curves, the ASCII data from which the images were created, and the source and background event lists created for the light curve. The format of the ASCII data files are detailed below.

Rescaling light curves

Before each of the light curve images is a link inviting you to "rescale" the light curve. If you have Javascript enabled (recommended) this opens a new panel (non-javascript users are redirected to a new page containing the panel). Here you can choose the axis limits, and whether they are linear or logarithmic, and then replot the data. Note that, unlike rebinning this does not change the data, only the visualisation.

The Flux Light Curve

The flux light curve, when available is almost identical to the Basic Light Curve, except that the count-rate column has been multiplied by a single conversion factor to yield the observed (i.e. absorbed) flux. This conversion factor is ordinarily taken from the XRT spectrum repository, however a link is always provided. Note that a single conversion factor like this does not account for spectral evolution as the GRB evolves. For light curves which include spectral evolution, see the Burst Analyser.

The Basic Light Curve

The Basic Light Curve is just that. It shows the count-rate against the time. The related ASCII file is ready for use in QDP, thus begins with READ SERR/TERR lines (to tell qdp which columns contain errors) There are 5 columns in this file, which are:

  1. Time
  2. Time -- positive error
  3. Time -- negative error
  4. Source count rate (1-σ)
  5. Error in count rate (1-σ)

There are up to five datasets in the file, separated by a line of NO NO NO NO NO, (again, for compatibility with QDP). The datasets appear in the following order: Windowed Timing (WT) settling mode data, WT data; WT upper limit; Photon Counting (PC) data; PC upper limit. Depending on the object, some of these may be missing, however each dataset begins with a comment (denoted by a leading !) identifying it.

The Detailed Light Curve

The Detailed Light Curve contains the information from the Basic Light Curve and also the background level in the top panel. The lower panel shows the fractional exposure in each bin. Note that the background level is the expected background rate in the source region, however as the source region is dynamic (it is smaller when the source is fainter) some correlation between source and background count-rates is expected.

The columns in the ASCII file are the same as for the basic light curve, however there are more datasets (again separated by lines of NO entries): the count-rate, upper limit (if applicable), background level and fractional exposure; first for WT mode then PC mode. Note that the WT settling mode data do not appear in here since all settling mode data points have a fractional exposure of 1 and the background rate is negligible.

The Hardness Ratio Light Curve

The hardness ratio plot contains 3 panels, showing the hard-band data in the top panel, the soft-band data in the middle, and the hardness ratio in the bottom panel. The hardness ratio is defined as Hard/Soft. By default the hard and soft bands are 1.5—10 keV and 0.3—1.5 keV respectively, although if you rebin the light curve you can change these values.

The hardness ratio ASCII file again contains the same basic 5 columns as in the basic light curve, with three datasets: hard band data, soft band data and hardness ratio, per instrument mode (WT settling mode, WT mode, PC mode).

Detailed data downloads

Among the list of files available to download at the bottom of the page are the "Detailed Photon Counting data" and a Windowed Timing equivalent. These are the raw output of the light curve generator, from which all the above files were parsed. Using something like awk or Perl, one can extract the data to create new plots. There is one file per mode, and one dataset per file. Where there is an upper limit, the file ends with a comment line announcing this fact. The columns are as follows:

  1. Time
  2. Time -- positive error
  3. Time -- negative error
  4. Net source count rate
  5. Error in the above (1-σ)
  6. Fractional exposure
  7. Background rate
  8. Error in background
  9. Total correction factor
  10. Exposure time in the bin (i.e. bin width * fractional exposure)
  11. Detection significance in this bin

The "total correction factor" is the ratio of the corrected count rate to observed count rate in this bin. The corrected count rate has been amended to account for: bad pixels/columns, pile-up and source counts landing outside the extraction region.
Sigma is the detection significance of the source in this bin (=net counts/error in bg counts). For an upper limit, the value written is the detection significance of the data which were replaced with the limit. The upper limit is always at the 3-σ level, and is calculated using the Bayesian method (see Kraft, Burrows & Nousek, 1991, ApJ, 374, 344).

Event lists

The event lists (source and background for each mode), contain all source or background events selected during the light curve creation. As well as the standard columns (TIME, position information, energy etc), the following information has been added:

  1. SRCRAD - Source extraction region radius (pixels).
  2. PUPRAD - Radius excluded due to pileup (arcseconds) - source event lists only.
  3. SRCAREA - Area of the source extraction region (square pixels) - background event lists only.
  4. BGAREA - Area of the background extraction region (square pixels) - background event lists only.

A note on the colour images. To create the colour image, the data are split into three energy bands: 0.3—1.2 keV, 1.2—1.8 keV and 1.8—10 keV. These are treated as the red, green and blue channels respectively and combined to give the colour image, which is then smoothed. The energy bands used were chosen based on a typical GRB spectrum, to give equal numbers of counts in the three channels. Thus, as one may intuitively expect, a comparatively soft burst will appear redder, and a hard burst bluer, in these images.

Back to contents | Back to repository

Multi-source Web Pages

For most GRBs it is immediately obvious from the XRT data which X-ray object is the afterglow. However, in some cases this is not true. If the afterglow is faint (either intrinsically, or because of a delay in observing it with the XRT) it may be difficult to tell whether it is fading, for example. Also, for some GRBs with large error circles (e.g. Fermi-LAT bursts) Swift may have to perform a tiled series of observations to cover the error circle. It is inevitable in such cases that a number of serendipitous sources will be found.

In cases with more than one candidate afterglow, the products page will show thumbnail light curves of each candidate (which serve as links to the standard products page for that afterglow), and then an image of the field, with each of the sources marked on it (for fields which are part of a tiled campaign, the sources in the image will contain a letter which refers to the field, e.g. "A1"). Should one of the sources be confirmed as the afterglow by the XRT Team, the main products page for this GRB will revert to being the light curve page for that candidate.

Note that, for sources which are subject to a tiled campaign, the image shown on this page contains only data for the field with the target ID of that URL. To see an image of the entire tiled dataset, you will need to visit the results page for that tiled campaign. This will be linked to from the light curve page.

Back to contents | Back to repository

How to rebin the data

During light curve generation, there are 4 criteria which must be met before a bin is considered complete. These are:

  1. There must be at least X counts in the source region
  2. The bin must be at least Y seconds long
  3. The bin must use all events from a given CCD frame
  4. The source count rate must be at least 3-σ above the background

Criterion 3 is a little obscure. Basically, since the data are read out at discrete times, one can get many events with the same timestamp. Criterion 3 ensures that these events are all put in the same bin, to prevent overlapping error bars.

In the standard products, the data are binned dynamically; that is, the binning criteria vary with count rate.
X, the minimum number of counts in a bin, is valid at a count rate of 1 count per second (cps); when the count rate changes by some factor (the rate factor), X changes by some other factor (the bin factor). The default values for these factors are 10 and 1.5 respectively, thus an order-of-magnitude change in count rate produces a factor of 1.5 change in bin size. This is done discretely rather than continuously, so if X=20 then where the count rate lies in the range 1-9.99 cps a bin must contain at least 20 counts. However, when the count rate is 10-99.99 cps, there must be 30 counts for a bin to be full. At lower rates of 0.1-0.99 cps, only 13 (=20/1.5) counts are needed to complete a bin, although this is rounded to 15: no matter how low the count rate goes, a bin will always need at least 15 counts, to ensure that using Gaussian methods to propagate errors statistics is valid.

While the above approach gives useful, usable results for almost all GRBs, it may not give the best possible results for a given burst. We have thus provided an interface to allow the user to rebin the light curves for themselves.

From the products page, follow the "Rebin this light curve" link. The form with which you are presented contains the fields detailed below. If you have javascript enabled, clicking on the field name will open a help box with more information about that field.

Counts per bin:
The minimum number of counts needed for a bin to be complete (criterion 1 above). Must be specified for WT and PC mode.
Minimum bin length:
The minimum duration required for a bin to be complete (criterion 2 above). Must be specified for WT and PC mode.
Dynamic binning:
Whether to use dynamic binning
Rate/bin factor:
These two boxes control the dynamic binning (if selected), full details are above.
Minimum counts per bin:
Dynamic binning cannot reduce the minimum counts per bin indefinitely. This defines the "hard" minimum, i.e. regardless of dynamic binning there must always be at least this many counts. We recommend that this is at least 15, to ensure that the Gaussian error propagation is valid.
Minimum sigma:
The minimum significance a bin must have to be considered complete (criterion 4 above).
Energy and grade selection:
The default is to use the same energy and grade ranges which were used for the light curve you are rebinning (the standard light curves all use 0.3-10 keV for the main light curve, 0.3-1.5 keV and 1.5-10 keV as the soft and hard bands respetively, and grades 0-12). You can select "User-specified" here, allowing you to set these fields yourself. If you do not have javascript, this option will not work. Instead, you should follow the link at the top of the page for advanced rebinning.

The following fields are only available if you have chosen "User specified" for the energy and grade selection.

Energy range for main curve:
Only events with energies in this range will be included in the light curves.
Grade range for main curve:
Only events with grades in this range will be included in the light curves.
Soft energy band:
The range of energies to use in the soft band for the hardness ratio. Note, these energies must lie within the range for the main curve.
Hard energy band:
The range of energies to use in the hard band for the hardness ratio. Note, these energies must lie within the range for the main curve.

Once you are happy, click "Rebin data". The rebinning process may take a while — the duration scales roughly with number of events — so for particularly bright or long lasting bursts this can take over five minutes. You will be redirected to a "holder page" which will auto-reload every 30 seconds until the process is finished at which point it will be replaced with the products page. Note: your rebinned light curve will be stored at a temporary URL, and deleted after 24 hours. There is no link to it, so it is a good idea to bookmark the holder page when the rebin process starts, so that you can find the page again.

Back to contents | Back to repository

New Bursts

When a GRB is detected by the Swift-BAT, or a ToO for a GRB detected by another mission is uploaded, the UKSSDC servers automatically register the object for analysis. The arrival of data on our quicklook site automatically triggers the creation of products including the light curve. First, however, the XRT data are reprocessed locally with the most recent version of the Swift software. (There will be a short delay between a new software release and its incorporation into our system. Note also that we will not as standard rebuild all historical data products after each software release).

Back to contents | Back to repository

Updating the data

The update procedure is almost identical to the new burst procedure. It is triggered by the arrival of new data, and again runs after those data have been locally reprocessed. When a light curve is updated it is not ordinarily completely rebuilt. Instead the time of the start of the final hardness ratio bin is determined, all existing data after this point are deleted, and the update begins at this time. There are two exceptions to this: if older data are also received (sometimes data are not delivered in "order"), or if the light curve was previously built before any PC mode data had arrived, but PC mode data are now available. The reason for the latter is that, without PC mode data, we cannot determine the most accurate XRT position of the GRB so the light curve is built using the position from the prompt data analysis. As soon as PC mode data are available a better position is available and should be used.

Back to contents | Back to repository

Changelog

The light-curve generation software is occasionally modified. Any changes are listed here.

Back to contents | Back to repository