The dependence of the intracluster light fraction on galaxy cluster properties

TL;DR

The paper investigates how the intracluster light (ICL) fraction depends on cluster properties by measuring ICL fractions for 177 groups and clusters identified in HSC-SSP imaging. It deploys MICL, an ML-based regression framework, trained on synthetic data and fine-tuned on CAMIRA clusters, to obtain ICL fractions within a fixed 300 kpc aperture and corrected for observational effects, enabling large-scale, homogeneous comparisons with redshift $z$, halo mass $M_{200}$, and magnitude gap $ riangle m_{12}$. The results show a strong negative correlation between ICL fraction and redshift $r_S=-0.604$ (p $=9\times10^{-10}$), a weak negative trend with halo mass after controlling for redshift $r_S=-0.330$ (p $=8\times10^{-5}$), and a marginal positive correlation with magnitude gap $r_S=0.226$ (p $=0.01$), with observational effects and subsampling used to interpret these trends. The authors conclude that galaxy–galaxy interactions, such as tidal stripping, likely dominate ICL production and that the large sample size clarifies large-scale ICL trends, though they emphasize the need to account for measurement methods and redshift effects in interpreting evolution.

Abstract

We use machine learning to measure the intracluster light (ICL) fractions of 177 galaxy groups and clusters identified from Hyper Suprime-Cam Subaru Strategic Program imaging to explore how the ICL varies with the properties of its host cluster. We study the variation in ICL fraction with host cluster redshift, halo mass, and magnitude gap to investigate how the ICL develops over time, in various cluster environments, and with cluster relaxation. We find that there is a decreasing correlation with redshift (Spearman correlation $r_S=-0.604$, p-value $=9\times10^{-10}$), however this can be plausibly accounted for by including the effects of cosmological surface brightness dimming and the passive aging of stellar populations. There is a weak negative correlation with halo mass ($r_S=-0.330$, p-value $=8\times 10^{-5}$) where ICL fractions are higher in lower halo mass groups than higher halo mass clusters. We also find that there is a marginal positive correlation with magnitude gap ($r_S=0.226$, p-value = 0.01), indicating that relaxed clusters are more likely to host higher ICL fractions. These results are consistent with a scenario where the dominant formation mechanism of the ICL is galaxy-galaxy interactions such as tidal stripping, and demonstrates the capability of the method to easily construct large samples and study large-scale trends in the ICL fraction.

Figures (13)

• Figure 1: Distribution of redshift (left), halo mass (centre), and magnitude gap (right) for CAMIRA cluster and GAMA group sample.
• Figure 2: Halo masses for clusters identified by both CAMIRA and GAMA calculated with the CAMIRA murata_mass-richness_2019 and GAMA viola_dark_2015 scaling relations. The mean difference in halo masses is -0.25 dex.
• Figure 3: Correlation between redshift, halo mass, and magnitude gap parameters for the CAMIRA and GAMA samples. Spearman correlation coefficients and p-values for the combined sample are shown on each panel. Each of these parameters have significant correlations at the $>3\sigma$ level with each other.
• Figure 4: Model predicted ICL fractions for 10 GAMA groups and the CAMIRA sample of 101 clusters, compared to manually measured ICL fractions. The dashed line indicates a one-to-one-relationship. Vertical error bars are taken from the $1\sigma$ confidence intervals calculated from the model's output probability distribution. The model does not exhibit any particular bias in its predictions on either sample, and we find that it provides an accurate estimation of its uncertainties on its predicted fractions.
• Figure 5: Model residuals as a function of redshift (left), halo mass (centre), and magnitude gap (right). Red boxes denote the five bins with equal numbers of clusters, with the vertical extent of each bin showing one standard deviation and the centre line showing the median. For all bins, the residuals are consistent with 0 within one standard deviation, indicating that the model's performance is not dependent on these parameters.