GOGREEN and GCLASS First Data Release at NOIRLab's Astro Data Lab
Release date: Aug 11, 2020
Data Lab primer and disclaimer
Hosting of the GOGREEN and GCLASS Data Release 1 products at Astro Data Lab is the result of a pilot project between the GOGREEN survey team, the Gemini Large-and-Long-Program (LLP), and NOIRLab's Astro Data Lab at the Community Science and Data Center (CSDC).
This first section of the GOGREEN and GCLASS Data Release 1 survey landing page was added to reflect the data access modes at NOIRLab's Astro Data Lab science platform. The original GOGREEN survey landing page is also reproduced below (mostly unchanged).
While the original survey page describes access to the data products on local disk, at Data Lab the catalogs are also accessible through a TAP service (via SQL queries), while all survey data products (with all files unchanged) are accessible through a file service (aka a "public VOSpace"). Data Lab also provides access to the survey image files through a Simple Image Access (SIA) protocol.
The biggest benefit of a science platform such as Data Lab is that you don't have to download any data or install any software on your local computer. All data and necessary code are installed remotely on our science platform, and all you need to start working with the data is a web browser. Using a Jupyter notebook framework you have the power of Python at your fingertips, with all software packages installed, and coupled with co-located access to enormous data sets for joint analysis. And to the GOGREEN and GCLASS Data Release 1 data products, of course.Back to Top
Jupyter notebooks for access to GOGREEN and GCLASS data at Data Lab
Data Lab has adapted the GOGREEN DR1 data access Jupyter notebook;
you can find in on Github,
and as a Data Lab user in
your Jupyter notebook directory, under
This adapted notebook additionally demonstrates TAP access to the
catalogs, and file service access to all survey data products, but all
the scientific content is unchanged from the original notebook found
Image access via SIA is demonstrated in another notebook, available both on
and in a Data Lab
user's Jupyter notebook
Description and Executive Summary
This release includes photometry (imaging, catalogues and derived products) and spectroscopy for all systems in GOGREEN and GCLASS, except SpARCS1033 for which most of the photometric imaging and catalogues are not available. We include the available, reduced HST images for all GOGREEN clusters. The Ultravista photometric catalogues (Muzzin et al. 2013) are also included, as these are the source of photometry for the COSMOS- systems in the sample. The SXDF catalogue of Mehta et al. (2018) must be downloaded separately, from http://homepages.spa.umn.edu/~mehta074/splash/
Finally we provide two python3 Jupyter notebooks for reading, manipulating and plotting the data.
Please report problems and questions to firstname.lastname@example.org.
- Aug 13, 2020: The WFC3/F140W images that were originally provided were not the most recent reductions. These have been removed and replaced (PHOTOMETRY/IMAGES/HST/*/F140W/) with the images (and weight/exposure maps) used in Matharu et al. (2019).
The whole data release is ~24Gb in size. This is dominated by the images in the PHOTOMETRY/IMAGES directory. If you don’t need access to those you can save a lot of download time.
- CADC (https://www.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/en/community/gogreen)
- NSF’s NOIRLab Data Labs (coming soon, to https://datalab.noao.edu/gogreendr1/). In addition to the raw data directory, Data Labs will soon provide an integrated file service with Simple Image Access and other features being developed.
If you make use of these data in your research, please cite the Data Release paper, Balogh et al. (2020). In addition, depending on your usage:
- The GOGREEN survey description paper, Balogh et al. (2017) , where the survey strategy, sample selection and parts of the data reduction are described in more detail.
- Muzzin et al. (2012), for more details about the GCLASS survey and data reduction
- Research that makes significant use of the GOGREEN photometric catalogues should cite van der Burg et al. (2020) for details of data reduction and measurements.
- Any research that makes use of the reduced HST/WFC3 F160W imaging of the GOGREEN clusters should cite Chan et al. (2020, in prep). Research that uses the HST/WFC3 F140W imaging of the GCLASS clusters should cite Matharu et al. (2019).
- Spectroscopy in the SXDF and COSMOS fields is coupled with photometry from the SPLASH and Ultravista surveys, respectively. Please cite Mehta et al. (2018) and/or Muzzin et al. (2013) if you make use of these data.
This data release includes data on all GOGREEN and GCLASS clusters. From Table 1 in the Data Release Paper:
Back to Top
For many applications, this is all you will need.
In directory CATS/
- Clusters.fits : is a FITS table with information about each group or cluster in the sample. Described in Table 2 of the Data Release paper.
- Redshift_catalogue.fits : This is a catalogue of all unique objects (mostly galaxies) with either GCLASS or GOGREEN spectroscopy. Includes redshifts, line indices and other information as described in Table 4 of the Data Release paper.
- Photo.fits : Contains all sources with good photometry in all available filters, and a subset of relevant columns extracted from the photometric catalogues. Matches with the spectroscopic sample are identified. It is described in Table 3 of the Data Release paper. It is conservative in the sense that objects are excluded if they are missing data in even one filter (totmask=0 in original catalogues).
We provide two Jupyter notebooks with some code to read and manipulate the data tables. We strongly suggest using these as a starting point for your analysis to avoid common mistakes and ensure you are using the right source of the data.
- DR1_Notebook: Provides simple code to open and read the three main catalogues into DataFrames. Provides some sample code to reproduce plots from the Data Release Paper. Also includes some sample code to access the 1D and 2D spectra, and associated imaging, to examine objects and make calculations.
- build_Table3: This is a script to build Photo.fits (Table 3 in the Data release paper) from the various associated photometric files. This is useful if you want to customize Photo.fits further (e.g. adding other information that is not currently included), or for examples of where to find relevant information and how to work with it.
If you want the 1D and 2D spectra, you will need this directory as well.
For each cluster there is a file called SPECTROSCOPY/OneD/CLUSTER_final.fits. This is a multiextension FITS (MEF) file containing 1D spectra for all unique objects in CLUSTER, from GOGREEN and GCLASS.
|[SCI,i] for i=1,N||science data. Absolute flux calibrated 1D spectra for each target i of N.|
|[VAR,i] for i=1,N||Corresponding variance array.|
|[DQ,i] for i=1,N||data quality array (GOGREEN spectra only). Here this corresponds to the number of pixels extracted in each column. So DQ=0 means no data and in general DQ<5 or so means a badly contaminated column for one reason or another. Shouldn’t usually need to worry about this as the VAR array contains what you need |
|[MDF]||A FITS table with information about the target. |
The corresponding 2D spectra are in SPECTROSCOPY/TwoD/CLUSTER_twod.fits. Important notes about these:
- These only include GOGREEN spectra. So the dimension of the MEF does not always match that of the 1D file. In the latter, the GCLASS spectra are appended to the GOGREEN spectra, so the entries in the 2D file should align with the first entries in the 1D.
- No relative or absolute flux calibration has been applied to these spectra
- The spatial dimension is in pixels, with a pixel scale of 0.16″
Warning: the photometry directory is large (>20Gb), mostly due to the IMAGES subdirectory. Unless you want access to the images and the input photometric catalogues (information not contained in CATS/Photo.fits), you may not need this.
The photometry is arranged in different subdirectories. Note there are some differences between the data formats and availability for the five GCLASS clusters that are not part of GOGREEN. Please read the README and README.gclass documentation carefully.
The Photo.fits file provided in the CATS/ directory provides the most commonly used photometric information, for all systems, in a homogeneous way. This is the safest way to access the data. We provide the python script that generates this file, which would be a good starting point if you want to access other information.
This is the largest subdirectory. We provide reduced images in all available filters for all SpARCS and SPT systems except SpARCS1033.
- The GOGREEN images are resampled to 3000×3000 pixels with pixel scale 0.200″/pix in the center (~ 10′ on a side). All filters are aligned in x and y.
- The two northern GCLASS-only images are resampled to 5000×5000 pixels with pixel scale 0.185″/pix in the center (approx 15′ on a side)
- The three southern GCLASS-only images are resampled to 3000×3000 pixels with pixel scale 0.185″/pix in the center (approx 10′ on a side)
Note the zeropoints of the images vary, and are described in an associated file.
These are treated differently and are available in a separate subdirectory (12Gb). Reduced images only are provided. See the data release paper and Chan et al. (in prep) for a description.
Ks-band selected catalogs, where each detected object is required to have 5 adjacent pixels with at least 1.5 sigma in the original (unconvolved) Ks-band image. All flux values have an AB magnitude zeropoint of 25 (equivalent to a flux scale of 0.3631 uJy per count). Therefore m_filter = -2.5*log10(flux_filter) + 25.
The column totmask is a conservative mask, equal to zero only if good photometric data is available in all filters. Otherwise it is set to 1, and not include in the compilation catalogue Photo.fits. Many of these data will still be useful, to those who don’t require access to all filters.
The five GCLASS-only clusters have a slightly different structure to the catalogue. See the README.gclass file provided with the data for details.Back to Top
ID-matched to the main photometric catalogue.
photoz-cats using EAZY code. (extension .zout)
id – same identifier as PHOTOM_CAT
z_peak – probably best estimate for photo-z
A quadratic function is fit to z_peak(z_spec) for the GOGREEN clusters. This correction has been applied to the redshifts in *_zphot.dat files.Back to Top
For the GOGREEN clusters we provide SDSS, FUV, NUV, U, B, V, J rest-frame colours estimated with EAZY. For each source, two different set of rest-frame colours are estimated:
- The first set assumes the cluster mean redshift for each source.
- The second set is based on the individually-measured spectroscopic redshifts, or, if there is no spec-z measured, the peak of the posterior P(z).
- Offsets to U-V and V-J colours have been applied to make the quiescent loci match those of UltraVISTA. Corrections are in UVJcorrections.dat
The five GCLASS-only clusters have only U-V and V-J colours provided, in a differently formatted file.Back to Top
ID-matched to the main photometric catalogue. Simple shifts are applied to bring the spectroscopic catalogue to the same astrometric reference as photometric catalogue prior to matching. Matching was done within 1.0”, with priority:
- FORS2 (SpARCS-0335)
- NED general search for spectroscopic redshifts
When there are two (or more) matches to spectroscopic catalogues, the second match is also reported (following the same priority list). To help cross-matching with GOGREEN spectra, the spectroscopic GOGREEN ID is reported in the last column.Back to Top
ID-matched to the main photometric catalogue.
Stellar masses and other SED-fitting parameters from FAST. There are two versions:
*Ks.fout takes the z_peak from the photo-z catalogue, or the spectroscopic redshift when available.
*Ks_fixz.fout fixes the redshift of all objects to the cluster mean redshift. This may come in handy when performing a statistical background subtraction.
All relevant parameters for the grid of template models are listed in the header of the files.
Best-fitting SED models to the photometry based on the BC03 library and the best spectroscopic redshift for each source. Includes the aperture correction to MAG_AUTO.Back to Top
RGB colour images of the clusters, constructed from B, I and Ks images.
Spectroscopic targets are indicated, using a very simple division between green (within 0.01 of mean cluster redshift) and red (rest).
We provide the DR1 Ultravista catalogue v4.1 here, for convenience. We use this as the source of photometry for the relevant GOGREEN systems.
The catalogue was obtained from https://www.strw.leidenuniv.nl/galaxyevolution/ULTRAVISTA/Ultravista/UltraVISTA_Catalog_Home.html
If you use this catalog in your research directly, or if you use the photometric information associated with the COSMOS groups or field sample in GOGREEN, please acknowledge it by citing the catalog paper, Muzzin et al. (2013), ApJS, 206, 8.Back to Top
We use the SPLASH v1.6 catalogue as the source of photometry for the GOGREEN systems in the SXDF field. The catalogue can be obtained from http://splash.caltech.edu/public/catalogs.html#sxds
If you use this catalog in your research directly, or if you use the photometric information associated with the SXDF groups or field sample in GOGREEN, please acknowledge it by citing the catalog paper, Mehta et al. (2018).
In this directory we provide a .csv file with a subset of columns and restframe UVJ colours computed consistently with the rest of our sample.Back to Top
Related Surveys & CollaborationsBack to Top
Michael Balogh (GOGREEN survey): mbalogh [at] uwaterloo [dot] ca Oliver Oberdorf (Gemini LLP): ooberdorf [at] gemini [dot] edu Robert Nikutta (Data Lab): nikutta [at] noao [dot] eduBack to Top