Skip banner and navigation tools.

 |  site map

Skip sidebars of related page links.

How to access the Swift archive

The UK Swift Science Data Centre provides two ways of accessing the Swift archive: a simple download portal and a more complex catalogue query tool. Although there is some overlap, these interfaces are designed to provide different services. Neither of these provide access to data less than a week old. This still resides on the quick-look site.

Which interface should I use?

This depends on what you want to do. If you want to download all the data for a target, then you probably want to use the download portal (described in more detail below). However, this only allows you to find observations by the name of the target, or by searching through a target list for a given year/month. If this is not sufficient, for example you may want to do a cone search, you need the more advanced LEDAS interface. The lists below show briefly the pros and cons of the two interfaces.

Download Portal

Advantages

Disadvantages

LEDAS

Advantages

Disadvantages


How to use the download portal

1. Choose Search Method

There are two ways of obtaining data through the download portal. The first is to search by name. In the text box, enter the name of the object you are looking for. This may contain wildcards: _ or ? indicate a single wildcard character, * or % indicate 0 or more wildcards. When you're happy, select "Search Names".

Alternatively, you can use the drop-down menus. Choose whether you're looking for a GRB, anything except GRBs, or everything. Then you can select the year and/or month you're looking for. If you chose both year and month, and you have Javascript enabled in your browser, the fourth menu will briefly display "Working...", and then change to a list of all objects which match your criteria. You may select a single object and then "Get data" and skip to step 3. You may want to leave the selection as "All" for now, or you may not want to specify the year and month (or you may not have Javascript). In which case, select the "Find objects" button.

2. Select the object

You will now be presented with a list of all (if any) objects which matched your search, along with their target IDs and how many observations there are of this object. When you've found the object you were looking for, select the relevant "Get data" button.

(Note that one object can have multiple target IDs, in which case only one is shown here, but when you choose to download the data all observations of the object are included by default.)

3. Choose your products

On the next page, use the checkboxes to choose which of the TDRSS, BAT, UVOT and XRT data products you want. The auxilliary files are included regardless, as you will need them. If you have Javascript enabled, you are also given an estimate of the size of the file you'll be downloading (this takes a few seconds to update when you change your selection). When you're happy, select "Get data" and you'll be provided with a tar file of your data.

By default, you receive every observation of this object which could be found in the archive. You may not want all of these. In that case, select the "choose which observations you want" link, and you'll be given a list of every observation of this target, along with the start and stop times, and the ontime (= the XRT exposure). Select the observations you want, and they will be all you receive.

A few caveats

While we have done everything we can to make this interface as helpful as possible, some things are beyond our control. The database itself, and the data files, are provided to us by the Swift Data Center in the US. Occasionally, there are mistakes in these data, such as non-GRB objects which have a target ID which identifies them as a GRB. Also, some targets are simply named "Burst" or "Automatic Target." Some of these are false-triggers, others are genuine GRBs which slipped through the net and their names never got updated. As far as possible we have tried to correct for these, so that objects are correctly classified and named. We apologise for times where we have not managed this.

Due to the size of the data files (often many gigabytes), data are packaged up and shipped "on the fly." No local cache is provided. While this enables speedy delivery (and keeps our disks managable), it does mean that if the download is interrupted, it is lost and must be started again from scratch. If you have a particularly unreliable internet connection, we strongly recommend that you download large datasets in several chunks, to minimise the risk of interruption. This can be done at Step 3 above, by following the "choose which observations you want" link, and only selecting some of the observations you want. Selecting "Get data" should not take you away from the page, so once your data has arrived, you can change which observations are selected and download again, repeating until you have everything you want.