Introduction
The SDSS is the most ambitious sky survey to have been carried out to date and includes both imaging and spectroscopy of a very large number of objects. More than 106 spectra and hundreds of millions of individual objects have been detected in the imaging. The data has been made publicly available through dedicated data-releases - the current, and last, data-release is DR7.
There are a number of papers providing an overview of SDSS, see the list of technical papers for instance, remember also that you need to cite these papers (see the publication rules for details.). However the best reference is often the web page of the appropriate data-release.
To make optimal use of the massive spectroscopic database of the SDSS, we started back in 2000/2001 a project to re-analyse all SDSS spectra using an optimized pipeline. This work resulted in a number of papers and our MPA-JHU datarelease which has been widely used in publications.
The current version of the MPA-JHU catalogues is for DR7 and can be found here.
This page will complement the MPA-JHU page by including more up-to-date information, various software to access our files as well asmore complete information on our fits, such as the full PDFs etc.
SpecObjId and PhotoObjId
It can often be useful to know the SpecObjId and the PhotoObjId of the objects. To this end we have collated all the information in the CAS table SpecObjAll for the objects in our sample and stored it in SpecObjAll.fits. Note that this is updated (15/08/2011) thanks to Jörg Dietrich who pointed out the shape of the table was 90 degrees away from ok. This table does however not have the BestObjId.
To get the object id for each galaxy, you can use the CAS ID file, which is small and contains Plate, MJD, fiberID and ObjID (as well as SpecObjID for some). Alternatively all the data in the PhotoObj table for the objects is available in the PhotoAllSpectra.fits
Improved stellar masses
The standard stellar mass catalogue on the MPA site has failures, particularly at z >0.3 because of poor u, and sometimes z, photometry. This was poorly handled in earlier versions, here is an improved file:
- Updated stellar masses totlgm_dr7_v5_2b.fit.
Emission line classifications
We have classified the sample using the precepts discussed in Brinchmann et al (2004). This resulted in the following table:
| SF | 203630 | 21.95% |
| AGN | 91477 | 9.86% |
| COMP | 47704 | 5.14% |
| LOWSNLIN | 68198 | 7.35% |
| BADSN | 652939 | 70.39% |
| LOWSNSF | 180969 | 19.51% |
| UNCLASS | 403772 | 43.53% |
You can download this classifications here, or as a FITS binary table here. The former file gives the classification as a single number according to the following translation scheme:
- -1 : Unclassifiable
- 0 : Not used
- 1 : SF
- 2 : low S/N SF
- 3 : Composite
- 4 : AGN non-Liner
- 5 : Low S/N Liner
The latter file has the same information encoded in the I_CLASS key and in addition identification data such as plateID (PLATE), MJD, fiberid, specobjid and objid.
Absolute magnitudes from kcorrect
We have absolute magnitudes and stellar masses from running kcorrect as well. Please download this file to get this.
Matches to other catalogues
The main SDSS pipeline contains matches to the FIRST and ROSAT survey data - in our catalogues this is kept in the gal_photo table.
The NYU SDSS group maintain an excellent resource for SDSS galaxy studies at their Value Added Galaxy Catalogue and this dataset includes matches to some other catalogues, including 2MASS.
The AKARI survey have released mid-IR data from their all-sky survey and I have matched these to our DR7 data. To do this I adopted a maximum separation of 7 arcsec (which is what the AKARI consortium used to do their matching between multiple filters (as descibed in the Release notes. This is a conservative approach and leads to 337 matches.
The matched files have the columns from the AKARI catalogues described in the release notes linked above. I have provided two different files:
- AKARI-SDSS-DR7-matches.fits.gz contains just the data for the matches and is hence a small catalogue. In many situations this is what you want.
- gal_AKARI_dr7_v5_2.fits.gz contains the same but now matched 1-1 with the gal_info table and its 927,552 rows. Obviously this corresponds to a waste of space in general but it does make matched plotting easier.