zCluster is a code for measuring galaxy cluster photometric redshifts. For details of the algorithm, its performance, and the output of the code, refer to Hilton et al. (2018). It has built-in support for querying large photometric surveys - currently:
- SDSS (DR7 - DR12)
- SDSS Stripe 82 (from SDSS DR7)
- CFHTLenS
- PS1 (DR2)
- DECaLS (DR7)
- DES (DR1 and Y3)
- KiDS (DR4)
The published paper shows results based on SDSS, S82, and CFHTLenS only. The other surveys listed above are work in progress (so use with caution; PS1 in particular is problematic).
zCluster can also run on user-supplied .fits table photometric catalogs, provided that they have columns
named ID, RADeg, decDeg, and magnitude column names in the form u_MAG_AUTO, u_MAGERR_AUTO etc..
zCluster itself is written in pure Python (developed on Python 3.6; it may still run using Python 2.7 but this is not supported). It requires the following additional Python modules (current versions used by the author are given in brackets, earlier and later versions also probably work):
- numpy (1.13.3)
- scipy (0.19.1)
- matplotlib (2.0.2)
- astLib (0.10.2)
- astropy (3.0.5)
- IPython (7.2.0)
- mastcasjobs (for PS1; https://github.com/rlwastro/mastcasjobs)
- casjobs (for PS1; https://github.com/dfm/casjobs)
IPython isn't really required, but is used for debugging.
To run on DES photometry, there is an additional dependency:
- easyaccess (1.4.5)
If you want to run the code in parallel, you will also need:
- mpi4py (2.0.0)
Note that if you want to run the code on a cluster, the bottleneck will be fetching the photometric catalogs over the internet. The MPI mode is still useful though on any machine with multiple cores.
All of the dependencies can be installed using pip.
As root:
sudo python setup.py install
Or, in your home directory:
python setup.py install --prefix=$HOME/local
Then add $HOME/local/bin to $PATH, and e.g., $HOME/local/lib/python3.6/site-packages to $PYTHONPATH.
export PATH=$HOME/local/bin:$PATH export PYTHONPATH=$HOME/local/lib/python3.6/site-packages:$PYTHONPATH
Alternatively,
python setup.py install --user
will install zCluster under $HOME/.local (on Ubuntu), and in some other default location on Mac.
See the examples/ dir. Here you will find two cluster catalog files (in .csv format rather than .fits),
400SDAll.csv and 400SDSmall.csv. The latter contains just the first 20 rows of the former, and is enough
to check that the code is working.
This example runs zCluster using SDSS DR12 photometry on part of the Burenin et al. (2007) 400SD cluster catalog.
zCluster 400SDSmall.csv SDSSDR12
This example will result in the creation of a catalog file called zCluster_400SDSmall_SDSSDR12.fits,
and a directory (where the results for individual clusters are cached) called 400SDSmall_SDSSDR12/.
If you want to check how well zCluster is doing versus spectroscopic redshifts, you can use the zClusterComparisonPlot code.
zClusterComparisonPlot 400SDSmall.csv zCluster_400SDSmall_SDSSDR12.fits
This will produce a plot named comparison_400SDSmall_vs_zCluster_400SDSmall_SDSSDR12.pdf, as well
outputting relevant statistics to the console.
You can find out about other options for both codes using the -h flag.
Please contact Matt Hilton matt.hilton@mykolab.com.