-
A morphological segmentation approach to determining bar lengths
Authors:
Mitchell K. Cavanagh,
Kenji Bekki,
Brent A. Groves
Abstract:
Bars are important drivers of galaxy evolution, influencing many physical processes and properties. Characterising bars is a difficult task, especially in large-scale surveys. In this work, we propose a novel morphological segmentation technique for determining bar lengths based on deep learning. We develop U-Nets capable of decomposing galaxy images into pixel masks highlighting the regions corre…
▽ More
Bars are important drivers of galaxy evolution, influencing many physical processes and properties. Characterising bars is a difficult task, especially in large-scale surveys. In this work, we propose a novel morphological segmentation technique for determining bar lengths based on deep learning. We develop U-Nets capable of decomposing galaxy images into pixel masks highlighting the regions corresponding to bars and spiral arms. We demonstrate the versatility of this technique through applying our models to galaxy images from two different observational datasets with different source imagery, and to RGB colour and monochromatic galaxy imaging. We apply our models to analyse SDSS and Subaru HSC imaging of barred galaxies from the NA10 and SAMI catalogues in order to determine the dependence of bar length on stellar mass, morphology, redshift and the spin parameter proxy $λ_{R_e}$. Based on the predicted bar masks, we show that the relative bar scale length varies with morphology, with early type galaxies hosting longer bars. While bars are longer in more massive galaxies in absolute terms, relative to the galaxy disc they are actually shorter. We also find that the normalised bar length decreases with increasing redshift, with bars in early-type galaxies exhibiting the strongest rate of decline. We show that it is possible to distinguish spiral arms and bars in monochrome imaging, although for a given galaxy the estimated length in monochrome tends to be longer than in colour imaging. Our morphological segmentation technique can be efficiently applied to study bars in large-scale surveys and even in cosmological simulations.
△ Less
Submitted 5 September, 2023;
originally announced September 2023.
-
The redshift evolution of the S0 fraction for $z<1$ in COSMOS
Authors:
Mitchell K. Cavanagh,
Kenji Bekki,
Brent A. Groves
Abstract:
Lenticular (S0) galaxies are galaxies that exhibit a bulge and disk component, yet lack any clear spiral features. With features considered intermediary between spirals and ellipticals, S0s have been proposed to be a transitional morphology, however their exact origin and nature is still debated. In this work, we study the redshift evolution of the S0 fraction out to $z \sim 1$ using deep learning…
▽ More
Lenticular (S0) galaxies are galaxies that exhibit a bulge and disk component, yet lack any clear spiral features. With features considered intermediary between spirals and ellipticals, S0s have been proposed to be a transitional morphology, however their exact origin and nature is still debated. In this work, we study the redshift evolution of the S0 fraction out to $z \sim 1$ using deep learning to classify F814W ($i$-band) HST-ACS images of 85,378 galaxies in the Cosmological Evolution Survey (COSMOS). We classify galaxies into four morphological categories: elliptical (E), S0, spiral (Sp), and irregular/miscellaneous (IrrM). Our deep learning models, initially trained to classify SDSS images with known morphologies, have been successfully adapted to classify high-redshift COSMOS images via transfer learning and data augmentation, enabling us to classify S0s with superior accuracy. We find that there is an increase in the fraction of S0 galaxies with decreasing redshift, along with a corresponding reduction in the fraction of spirals. We find a bimodality in the mass distribution of our classified S0s, from which we find two separate S0s populations: high-mass S0s, which are mostly red and quiescent; and low-mass S0s, which are generally bluer and include both passive and star-forming S0s, the latter of which cannot solely be explained via the faded spiral formation pathway. We also find that the S0 fraction in high-mass galaxies begins rising at higher $z$ than in low-mass galaxies, implying that high-mass S0s evolved earlier.
△ Less
Submitted 9 February, 2023;
originally announced February 2023.
-
The galaxy morphology-density relation in the EAGLE simulation
Authors:
Joel Pfeffer,
Mitchell K. Cavanagh,
Kenji Bekki,
Warrick J. Couch,
Michael J. Drinkwater,
Duncan A. Forbes,
Bärbel S. Koribalski
Abstract:
The optical morphology of galaxies is strongly related to galactic environment, with the fraction of early-type galaxies increasing with local galaxy density. In this work we present the first analysis of the galaxy morphology-density relation in a cosmological hydrodynamical simulation. We use a convolutional neural network, trained on observed galaxies, to perform visual morphological classifica…
▽ More
The optical morphology of galaxies is strongly related to galactic environment, with the fraction of early-type galaxies increasing with local galaxy density. In this work we present the first analysis of the galaxy morphology-density relation in a cosmological hydrodynamical simulation. We use a convolutional neural network, trained on observed galaxies, to perform visual morphological classification of galaxies with stellar masses $M_\ast > 10^{10} \, \mathrm{M}_\odot$ in the EAGLE simulation into elliptical, lenticular and late-type (spiral/irregular) classes. We find that EAGLE reproduces both the galaxy morphology-density and morphology-mass relations. Using the simulations, we find three key processes that result in the observed morphology-density relation: (i) transformation of disc-dominated galaxies from late-type (spiral) to lenticular galaxies through gas stripping in high-density environments, (ii) formation of lenticular galaxies by merger-induced black hole feedback in low-density environments, and (iii) an increasing fraction of high-mass galaxies, which are more often elliptical galaxies, at higher galactic densities.
△ Less
Submitted 16 December, 2022;
originally announced December 2022.
-
The evolution of barred galaxies in the EAGLE simulations
Authors:
Mitchell K. Cavanagh,
Kenji Bekki,
Brent A. Groves,
Joel Pfeffer
Abstract:
We study the morphologies of 3,964 galaxies and their progenitors with $M_\star > 10^{10} M_\odot$ in the reference EAGLE hydrodynamical simulation from redshifts $z=1$ to $z=0$, concentrating on the redshift evolution of the bar fraction. We apply two convolutional neural networks (CNNs) to classify 35,082 synthetic g-band images across 10 snapshots in redshift. We identify galaxies as either bar…
▽ More
We study the morphologies of 3,964 galaxies and their progenitors with $M_\star > 10^{10} M_\odot$ in the reference EAGLE hydrodynamical simulation from redshifts $z=1$ to $z=0$, concentrating on the redshift evolution of the bar fraction. We apply two convolutional neural networks (CNNs) to classify 35,082 synthetic g-band images across 10 snapshots in redshift. We identify galaxies as either barred or unbarred, while also classifying each sample into one of four morphological types: elliptical (E), lenticular (S0), spiral (Sp), and irregular/miscellaneous (IrrM). We find that the bar fraction is roughly constant between $z = 0.0$ to $z = 0.5$ (32% to 33%), before exhibiting a general decline to 26% out to $z = 1$. The bar fraction is highest in spiral galaxies, from 49% at $z = 0$ to 39% at $z = 1$. The bar fraction in S0s is lower, ranging from 22% to 18%, with similar values for the miscellaneous category. Under 5% of ellipticals were classified as barred. We find that the bar fraction is highest in low mass galaxies ($M_\star \leq 10^{10.5} M_\odot$). Through tracking the evolution of galaxies across each snapshot, we find that some barred galaxies undergo episodes of bar creation, destruction and regeneration, with a mean bar lifetime of 2.24 Gyr. We further find that incidences of bar destruction are more commonly linked to major merging, while minor merging and accretion is linked to both bar creation and destruction.
△ Less
Submitted 23 December, 2021;
originally announced December 2021.
-
Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
Authors:
Mitchell K. Cavanagh,
Kenji Bekki,
Brent A. Groves
Abstract:
Classifying the morphologies of galaxies is an important step in understanding their physical properties and evolutionary histories. The advent of large-scale surveys has hastened the need to develop techniques for automated morphological classification. We train and test several convolutional neural network architectures to classify the morphologies of galaxies in both a 3-class (elliptical, lent…
▽ More
Classifying the morphologies of galaxies is an important step in understanding their physical properties and evolutionary histories. The advent of large-scale surveys has hastened the need to develop techniques for automated morphological classification. We train and test several convolutional neural network architectures to classify the morphologies of galaxies in both a 3-class (elliptical, lenticular, spiral) and 4-class (+irregular/miscellaneous) schema with a dataset of 14034 visually-classified SDSS images. We develop a new CNN architecture that outperforms existing models in both 3 and 4-way classification, with overall classification accuracies of 83% and 81% respectively. We also compare the accuracies of 2-way / binary classifications between all four classes, showing that ellipticals and spirals are most easily distinguished (>98% accuracy), while spirals and irregulars are hardest to differentiate (78% accuracy). Through an analysis of all classified samples, we find tentative evidence that misclassifications are physically meaningful, with lenticulars misclassified as ellipticals tending to be more massive, among other trends. We further combine our binary CNN classifiers to perform a hierarchical classification of samples, obtaining comparable accuracies (81%) to the direct 3-class CNN, but considerably worse accuracies in the 4-way case (65%). As an additional verification, we apply our networks to a small sample of Galaxy Zoo images, obtaining accuracies of 92%, 82% and 77% for the binary, 3-way and 4-way classifications respectively.
△ Less
Submitted 2 June, 2021;
originally announced June 2021.
