How to Access SkyMapper Data Products

We currently provide three ways of accessing SkyMapper Southern Sky Survey data — a simple cone search of the DR1 (or EDR) distilled master catalogues around a given sky position, arbitrary queries of the full DR1 database and external tables using ADQL, and reduced image cutouts with WCS around a given sky position. All three methods can also be accessed through Virtual Observatory-compatible tools such as TOPCATAladin and custom scripts.

Cone Search^ Back to top

Performing a Cone Search is the simplest way to query the SkyMapper distilled photometry table (dr1.master) and requires only a sky position and a search radius. You may optionally specify how many columns are returned by the service using the metadata verbosity parameter. No other search criteria are allowed. The website currently supports three coordinate options:

Cone Searches from the website are currently limited to a search radius of 1 degree (3,600 arcseconds) and/or 2,000 rows, though larger queries (up to 1,000,000 rows) are possible from other clients. For more complex queries (including cross-matching), we suggest using the SkyMapper TAP service described below

The query page will validate your inputs before performing the search to avoid any errors. By default, Cone Search results are sorted by the angular distance from the query centre. You may re-sort the columns by clicking on the headers, or filter the rows using a variety of operators (including < and >) to select only those objects meeting your criteria. The table can then be saved as a VOTable (XML), CSV or tab separated file, or the entire table sent to TOPCAT (if it is running) or another Virtual Observatory client for further processing.

Virtual Observatory Access (Simple Cone Search)^ Back to top

You may also directly submit requests to this Cone Search service via TOPCAT, your web browser or other favourite tool (e.g. wget, Python) using the Virtual Observatory Simple Cone Search standard. This is especially useful if you would like to call a Cone Search from within a script. Instructions for using TOPCAT to load the results of a Cone Search can be found here

For manual queries, the URL to submit your request to is:

http://skymapper.anu.edu.au/sm-cone/query?

If constructing the URL yourself, use the required parameters  RA, DEC (both ICRS decimal degrees) and SR (search radius in degrees), and the optional parameter VERB (1=minimum to 3=maximum), for example:

http://skymapper.anu.edu.au/sm-cone/query?RA=100&DEC=-30&SR=0.1&VERB=3

By default, queries are made on the dr1.master catalog. You may search the object detection catalog by setting CATALOG=dr1.fs_photometry (or similarly edr.[master|fs_photometry] for the EDR tables).

You can also change the table format returned by the service using the optional RESPONSEFORMAT parameter and choosing either VOTABLE (VOTable XML file with full metadata), CSV (comma separated file with header row), TSV (tab separated file with header row) or their associated MIME-types (e.g. application/xml or text/tab-separated-values).

For example:

http://skymapper.anu.edu.au/sm-cone/query?RA=100&DEC=-30&SR=0.1&RESPONSEFORMAT=CSV

will download a comma separated file (called result.csv) of default verbosity.

Some basic error checking will be performed by the server. More details on the Simple Cone Search standard and interpreting the VOTable results can be found in the IVOA documentation.

 


Image Cutout Service^ Back to top

This website also provides an interface to the DR1 Image Cutout service, which conforms to Version 1.0 of the Virtual Observatory Simple Image Access protocol (SIAP). At a minimum, an image cutout request requires:

Note: the image size is calculated along the x and y image axes, which for DR1 images (PA = 0 deg) will align with the N-S,E-W World Coordinate System axes. 

Further query options are also available:

The website will validate your inputs to avoid any errors. The default output is to return a PNG preview, as well as download links for the FITS image and bitmask.

The intersection keyword allows you to narrow your query to select only those images that completely cover the requested cutout region (COVERS), at least contain the centre of the area (CENTER, default behaviour), or overlap any part of the area (OVERLAPS).  This may give cutouts that are smaller than the dimensions requested. Note that the image centre given in the results table (ra_cntr, dec_cntr) and image name is that of the actual cutout, not the requested centre.

On the results page you can sort the columns by clicking on the headers, or filter the rows to select only those images meeting further criteria. The table can then be saved as a VOTable (XML), CSV or tab separated file, or the entire table sent to TOPCAT (if it is running) or another Virtual Observatory client for further processing. The individual FITS files can be sent to Aladin or saved to your computer along with the image bitmasks. You can also select multiple images and download them as a ZIP file. Note that this feature sometimes does not work properly on earlier versions of Safari, or results in a file called 'Unknown' with no .zip extension. 

Virtual Observatory Access (Simple Image Access v1)^ Back to top

Instead of using the website you may also submit requests directly to the image cutout service via TOPCAT, CDS Aladin or your favourite Virtual Observatory Simple Image Access tool. Instructions for using TOPCAT are available here.  More details on the Simple Image Access standard and interpreting the results can be found in the SIAP v1.0 documentation, available here.  

For manual queries, the URL to submit your requests to is:

http://skymappersiap.asvo.nci.org.au/dr1_cutout/query?

If constructing the URL yourself, use the required parameters POS ('RA,Dec' in ICRS degrees), SIZE ('X[,Y]' in degrees) and the optional parameters VERB (0=minimum to 3=maximum), BAND (comma separated list of SkyMapper filters), FORMAT (image/fits and/or image/png MIME types, or the special values ALL: all formats, METADATA: service metadata only, GRAPHIC: PNG output), INTERSECT (one of COVERS, CENTER, OVERLAPS), MJD_START (Modified Julian Date) and/or MJD_END

For example:

http://skymappersiap.asvo.nci.org.au/dr1_cutout/query?POS=189.99763,-11.62305&SIZE=0.05&BAND=g,r,i&FORMAT=GRAPHIC&VERB=3

will return a list of nine 3'x3' maximum PNG images around the Sombrero galaxy where the central coordinates are inside the cutout region (INTERSECT=CENTER). Three of these images do not cover the full region and could be removed by specifying INTERSECT=COVERS). All available image metadata columns will be returned.

Another example:

http://skymappersiap.asvo.nci.org.au/dr1_cutout/query?POS=189.99763,-11.62305&SIZE=0.05,0.1&FORMAT=image/fits&INTERSECT=covers&MJD_END=56970

will generate a list of six 3'x6' rectangular FITS cutouts in all available filters (default BAND=all) from images taken before 2014 November 9 which completely cover the cutout region (INTERSECT=covers)

You may also change the table format returned by the service using the optional RESPONSEFORMAT parameter and choosing either HTML (sortable, filterable table for viewing in a browser), VOTABLE (VOTable XML file with full metadata), CSV (comma separated file), TSV (tab separated file) or their associated MIME-types (e.g. application/xml or text/tab-separated-values). For instance, adding RESPONSEFORMAT=CSV to the above query will download the image list as a comma separated file.     

Note that the service does not directly return images, only a list of cutouts meeting your requirements. Once downloaded, the table can be further filtered by your client (e.g. TOPCAT or a Python script) and the images themselves downloaded at the URL in the get_image column for the requested FORMAT, or equivalently get_fits or get_mask for the FITS images and bitmasks.

If you know the image ID and CCD number (from an ADQL query against the fs_photometry or ccds tables, for instance), you can directly generate images by querying the service /get_image endpoint:

http://skymappersiap.asvo.nci.org.au/dr1_cutout/get_image?IMAGE=20140425124821-10&SIZE=0.0833&POS=189.99763,-11.62305&FORMAT=png

This will generate a 5'x5' PNG cutout around the Sombrero galaxy from CCD 10 of the g-band image wth ID 20140425124821. Using FORMAT=fits will download the corresponding FITS file and FORMAT=mask will download the FITS bitmask, which can be used to identify bad or saturated pixels. 

For bandwith reasons we do not offer full CCD images (38 x 19 arcmin, 4096 x 2048 pix) for download as part of DR1. If these would be useful, please contact us and we will consider offering them as part of later data releases, or as a bulk download.

 


Full Catalogue Search (ADQL)^ Back to top

The most powerful way to access SkyMapper and co-hosted external catalogues is through a Full Catalogue Search using the Astronomical Data Query Language (ADQL), a Virtual Observatory standard almost identical to the SQL used by relational databases such as MySQL, PostgreSQL and the SDSS CasJobs service. Requests of this type are processed by the VO-standard Table Access Protocol (TAP). A typical ADQL query might be:

SELECT * FROM dr1.master WHERE dej2000<-80 AND g_psf<16 AND FLAGS=0

which would select all columns from the SkyMapper dr1.master catalogue for stars below a declination of -80 degrees with g-band magnitudes < 16 and good flags.

For simple queries (fast-running, <2,000 rows) you can use the website to perform your search. Several example ADQL queries are provided with the form to get you started – simply click on the name of an example to populate the query box. You can find further information on ADQL and detailed tutorials herehere and here. To help you formulate your queries you can also browse the available tables and columns using the Table Browser.

To avoid long queries tying up resources on the server and crashing your browser, results are limited to 100 rows by default for queries from the website. You may override this limit by specifying SELECT TOP nnn in your request or changing the value in the drop-down box. However, if you attempt to request over 2,000 results, the system will automatically scale your request back to that limit, with no guarantee of the same rows each time the query is run. This is only a hard limit on access from the website and larger requests (up to 1 million rows) are possible using other TAP clients (e.g. TOPCAT).

Requests submitted via the website will be validated for both ADQL syntax errors and compatibility with the tables and columns available in database.  For reference, your ADQL query is displayed on the results page and saved to the header of the VOTable returned by the service.  You can sort the columns on the results page by clicking on the headers, or filter the output to select only those rows meeting further criteria. The table can then be saved as a VOTable (XML), CSV or tab separated file, or the entire table sent to TOPCAT (if it is running) or another Virtual Observatory client for further processing.

Virtual Observatory Access (TAP)^ Back to top

You may also submit requests directly to the TAP service via tools such as TOPCAT or TAPsh. The service is discoverable by TOPCAT and can be found by searching for "SkyMapper" in the Select Service panel of the TAP Query window.

For large, complex or long-running requests we recommend using the 'asynchronous' query option if your client permits it. This allows your query to run in the background without your client blocking or timing out, and will save the results on the server for a period of time (currently one week) for you to collect. In TOPCAT this is accomplished by selecting the 'Asynchronous' mode in the Enter Query window.

For manual queries, the TAP endpoint is:

http://skymappertap.asvo.nci.org.au/ncitap/tap/

and has the IVOA identifier  ivo://nci.org.au/skymapper/tapSynchronous and asynchonous queries of up to 1 million rows can be made to the tap/sync/ and tap/async/ endpoints, respectively. The results of asynchronous queries are currently held on the server for one week unless deleted earlier by the client. A full description of manipulating asynchronous TAP jobs through the Universal Worker Service (UWS) is beyond the scope of this documentation, but is discusssed in Section 5 of the TAP standard

You can view all available table metadata (in XML format) at the tables/ endpoint:

http://skymappertap.asvo.nci.org.au/ncitap/tap/tables

or directly query the TAP_SCHEMA.tables and TAP_SCHEMA.columns tables within the TAP service itself. 

Please contact us if extra metadata, columns, tables, stored procedures or other catalogues would be useful for your work with SkyMapper and we will consider offering them as part of upcoming data releases.