The SLACS strong lens sample, debiased

Datasets from the article "The SLACS strong lens sample, debiased" (Sonnenfeld 2024, A&A in press).

Content

SLACS_table.cat: table with data of the SLACS lenses. parent_sample.fits: .fits file with data of the parent sample. full_inference.hdf5: chain of samples from the posterior probability distribution of the full model. nofpprior_inference.hdf5: chain of samples from the posterior probability distribution of the model with no fundamental plane prior. slonly_inference.hdf5: chain of samples from the posterior probability distribution of the lens-only model, with no selection function correction. full_pp_samples.hdf5: posterior predicted samples from the full model. full_nopfind_pp_samples.hdf5: posterior predicted samples from the full model, with no lens finding probability term. This is used to predict the population of detectable lenses.

Chain files

File full_inference.hdf5 contains the chain with samples from the posterior probability. Samples in each of the parameters are stored in separate datasets. The shape of each dataset is (100, 2000), with the first dimension being the number of walkers in the MCMC. The datasets are the following:

Name
Description

mu_m5
Mean m5 at log-stellar mass 11.3 and average size

mu_gamma
Mean gamma at log-stellar mass 11.3 and average size

beta_m5
Dependence of m5 on stellar mass

xi_m5
Dependence of m5 on excess size

beta_gamma
Dependence of gamma on stellar mass

xi_gamma
Dependence of gamma on excess size

sigma_m5
Intrinsic scatter in m5 around the mean

sigma_gamma
Intrinsic scatter in gamma around the mean

t_find
Parameter theta_0 of the lens finding probability

la_find
Log-10 of parameter a of the lens finding probability

fpfit
Parameters of the fundamental plane fit

logp
log of posterior probability

The file slonly_inference.hdf5 is similarly structured.

Posterior predicted samples

The file full_pp.hdf5 contains posterior predicted samples of properties of parent sample galaxies and strong lenses. It contains 1000 draws from the posterior. The data are organised in groups, as follows:

hyperpars: values of the model parameters.

subset: properties of a random sample of 100 galaxies (different seed for each sample).

lenses: properties of a sample of 59 lenses.

pop_sigma_bin: average properties of parent sample galaxies, in bins of velocity dispersion. Bin edges are defined in sigma_bins.

pop_ms_bin: average properties of parent sample galaxies, in bins of log-stellar mass. Bin edges are defined in ms_bins.

lens_sigma_bin: average properties of parent sample galaxies, in bins of velocity dispersion.

lens_ms_bin: average properties of parent sample galaxies, in bins of log-stellar mass.

The file full_nopfind_pp.hdf5 is similarly structured.