Large Scale Structure

Overview

 

The Large Scale Structure galaxy catalogs combine the list of cosmological tracer targets (typically galaxies or QSOs) with the results of the spectroscopic data reduction to create data and random catalogs that allow the user to estimate the cosmological tracer density fluctuations at any point within the survey footprint.  Weights are assigned to tracers to account for observational imperfections such as a failure of the spectroscopic pipeline to obtain a redshift, fiber collisions that preclude simultaneously assigning spectroscopic fibers to targets closer than 62”, and non-cosmological fluctuations imprinted the target catalog, such as the correlation between targets and stellar density described in  Ross et al. 2011,  Ho et al. 2012,  Ross et al. 2012.  The random catalogs are designed to randomly sample the survey footprint with a density proportional to the map of the survey completeness.  Pairs of data and random catalogues are generated for samples with distinct selection functions. In BOSS, we generate LSS catalogues for the LOWZ and CMASS samples separately. The LOWZ target class captures objects primarily at z < 0.4, while the CMASS class selects objects at 0.4 < z < 0.8. With DR12 we also provide a combined sample catalogue, where these two samples are optimally combined (details in Reid et al. 2016). In eBOSS, we generate LSS catalogues for the LRGs, ELGs and QSOs independently. Finally, we continue to generate catalogs in the north galactic cap and south galactic cap separately, for the reasons detailed in Ross et al. 2011Ho et al. 2012Ross et al. 2012.

 

Galaxy Catalogs

If you want to work directly with the catalogs used in our analyses, download the following files directly from the SAS for DR10, DR11, and DR12 . DR14 catalogues for QSOs and LRGs will appear in due course. The contents of each relevant file is described by a data model:

where X, SAMPLE, and NS can change depending on the data release, the sample and the galactic cap.

Mock Galaxy Catalogs

We created a set of mock galaxy catalogs with the same survey footprint as the BOSS survey.  We use these to validate our methodology and estimate the covariance matrix associated with our observables, which is necessary to compute the uncertainty on any derived quantity from our galaxy catalogs, such as the BAO scale.  We used two different methodologies, QPM and PATCHY, both described in Alam et al. 2017.  Galaxy and random catalogs are available for each set of mocks and described by a data model: mock_galaxy_DRX_SAMPLE_NS_QPM_IDNUMBER.fits.gz The data model for QPM and PATCHY mocks is the same.

Galaxy LSS catalog creation code [mksample/mkEsample]

The code used by the BOSS galaxy clustering working group to produce the DR10, DR11, and DR12 catalogs is called mksample and is available to download on the SAS.  The DR10 and DR11 catalogs were produced using the same algorithm but with different input files.  The DR12 catalogs were produced with an updated version of mksample.  For details of the DR10/DR11 algorithms, consult Anderson et al. 2014.  The DR12 catalogs and updated mksample are described in Reid et al. 2016.  The code can be used to create new sets of random catalogs or to generate new catalogs for a subsample of galaxies in the CMASS or LOWZ target classes.  The necessary input files and algorithms are described in more detail in the tutorial.

eBOSS DR14 LSS catalogues for QSOs and LRGs are computed with a slightly updated version of the code, called mkEsample. mkEsample will be released in due course.