Euclid Quick Data Release (Q1), A first look at the fraction of bars in massive galaxies at $z<1$

TL;DR

This study uses Euclid data and deep learning to measure the fraction of barred galaxies at redshifts below 1, revealing mass-dependent evolution and comparing observations with cosmological simulations.

Contribution

First application of Euclid Q1 data combined with deep learning to quantify bar fractions in massive galaxies at z<1, providing new insights into galaxy evolution.

Findings

Massive galaxies have higher bar fractions at fixed redshift.

Lower-mass galaxies show a steeper decline in bar fraction with redshift.

Simulations broadly agree but overpredict bar fractions in high-mass systems.

Abstract

Stellar bars are key structures in disc galaxies, driving angular momentum redistribution and influencing processes such as bulge growth and star formation. Quantifying the bar fraction as a function of redshift and stellar mass is therefore important for constraining the physical processes that drive disc formation and evolution across the history of the Universe. Leveraging the unprecedented resolution and survey area of the Euclid Q1 data release combined with the Zoobot deep-learning model trained on citizen-science labels, we identify 7711 barred galaxies with $M_* \gtrsim 10^{10}M_\odot$ in a magnitude-selected sample $I_E < 20.5$ spanning $63.1 deg^2$. We measure a mean bar fraction of $0.2-0.4$, consistent with prior studies. At fixed redshift, massive galaxies exhibit higher bar fractions, while lower-mass systems show a steeper decline with redshift, suggesting earlier disc assembly in massive galaxies. Comparisons with cosmological simulations (e.g., TNG50, Auriga) reveal a broadly consistent bar fraction, but highlight overpredictions for high-mass systems, pointing to potential over-efficiency in central stellar mass build-up in simulations. These findings demonstrate Euclid's transformative potential for galaxy morphology studies and underscore the importance of refining theoretical models to better reproduce observed trends. Future work will explore finer mass bins, environmental correlations, and additional morphological indicators.