Spin parity of spiral galaxies II: a catalogue of 80 k spiral galaxies using big data from the Subaru Hyper Suprime-Cam survey and deep learning

Tadaki, Ken-ichi; Iye, Masanori; Fukumoto, Hideya; Hayashi, Masao; Rusu, Cristian E.; Shimakawa, Rhythm; Tosaki, Tomoka

doi:10.1093/mnras/staa1880

Cited by 20 publications

(18 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First of all, we apply the well-tuned deep learning model to the main targets, the 54871 bright galaxies at z = 0.01-0.3, to choose spiral galaxies among them. Because our final sample consists of sufficiently bright (rpetro < 18) and massive (M > 10 10 M ) galaxies at z < 0.3, we assume that the redshift bias of the spiral classification is negligible (see also Tadaki et al 2020). Automatic and efficient deep learning classification enabled us to categorize such a large amount of dataset in only 330 sec on the GPU environment.…”

Section: Selection Of Passive Spiral Galaxiesmentioning

confidence: 99%

“…With such high-quality data, we can expect to capture fainter spiral arms not detected in previous surveys and those with smaller pitch angles mistaken for ringed galaxies. Many studies have demonstrated the utility of the HSC-SSP data, such as galaxy morphological classifications (Tadaki et al 2020;Martin et al 2020;Shimakawa et al 2021), weak lensing measurements (Oguri et al 2018;Oguri et al 2021), and anomaly detection (Storey-Fisher et al 2021;Tanaka et al 2021). Therefore, the HSC-SSP data will enable us to significantly expand upon existing passive spiral samples, offering a unique opportunity to understand the transition phase from spiral galaxies to early-type galaxies statistically.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Passive spiral galaxies deeply captured by Subaru Hyper Suprime-Cam

Shimakawa,

Tanaka,

Bottrell

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents a thousand passive spiral galaxy samples at z = 0.01-0.3 based on a combined analysis of the Third Public Data Release of the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP PDR3) and the GALEX-SDSS-WISE Legacy Catalog (GSWLC-2). Among 54871 gri galaxy cutouts taken from the HSC-SSP PDR3 over 1072 deg 2 , we conducted a search with deep-learning morphological classification for candidates of passive spirals below the star-forming main sequence derived by UV to mid-IR SED fitting in the GSWLC-2. We then classified the candidates into 1100 passive spirals and 1141 secondary samples based on visual inspections. Most of the latter cases are considered to be passive ringed S0 or pseudo-ringed galaxies. The remainder of these secondary samples has ambiguous morphologies, including two peculiar objects with diamond-shaped stellar wings. The selected passive spirals have a similar distribution to the general quiescent galaxies on the EW Hδ -D n 4000 diagram and concentration indices. Moreover, we detected an enhanced passive fraction of spiral galaxies in X-ray clusters. Passive spirals in galaxy clusters are preferentially located in the midterm or late infall phase on the phase-space diagram, supporting the ram pressure scenario, which has been widely advocated in previous studies. The source catalog and gri-composite images are available on the HSC-SSP PDR3 website (https://hsc.mtk.nao.ac.jp/ssp/data-release/). Future updates, including integration with a citizen science project dedicated to the HSC data, will achieve more effective and comprehensive classifications.

show abstract

Section: Selection Of Passive Spiral Galaxiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Passive spiral galaxies deeply captured by Subaru Hyper Suprime-Cam

Shimakawa,

Tanaka,

Bottrell

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In such applications, the CNN is trained on labelled observations and then applied to mock images. In addition to global morphology, Tadaki et al (2020) used also a similar supervised CNN setting to classify spiral galaxies based on their resolved properties (i.e. type of spiral arms).…”

Section: Cnns As State-of-the-art Approachmentioning

confidence: 99%

The Dawes Review 10: The impact of deep learning for the analysis of galaxy surveys

Huertas-Company

Lanusse²

2023

Publ. Astron. Soc. Aust.

View full text Add to dashboard Cite

The amount and complexity of data delivered by modern galaxy surveys has been steadily increasing over the past years. New facilities will soon provide imaging and spectra of hundreds of millions of galaxies. Extracting coherent scientific information from these large and multi-modal data sets remains an open issue for the community and data-driven approaches such as deep learning have rapidly emerged as a potentially powerful solution to some long lasting challenges. This enthusiasm is reflected in an unprecedented exponential growth of publications using neural networks, which have gone from a handful of works in 2015 to an average of one paper per week in 2021 in the area of galaxy surveys. Half a decade after the first published work in astronomy mentioning deep learning, and shortly before new big data sets such as Euclid and LSST start becoming available, we believe it is timely to review what has been the real impact of this new technology in the field and its potential to solve key challenges raised by the size and complexity of the new datasets. The purpose of this review is thus two-fold. We first aim at summarising, in a common document, the main applications of deep learning for galaxy surveys that have emerged so far. We then extract the major achievements and lessons learned and highlight key open questions and limitations, which in our opinion, will require particular attention in the coming years. Overall, state-of-the-art deep learning methods are rapidly adopted by the astronomical community, reflecting a democratisation of these methods. This review shows that the majority of works using deep learning up to date are oriented to computer vision tasks (e.g. classification, segmentation). This is also the domain of application where deep learning has brought the most important breakthroughs so far. However, we also report that the applications are becoming more diverse and deep learning is used for estimating galaxy properties, identifying outliers or constraining the cosmological model. Most of these works remain at the exploratory level though which could partially explain the limited impact in terms of citations. Some common challenges will most likely need to be addressed before moving to the next phase of massive deployment of deep learning in the processing of future surveys; for example, uncertainty quantification, interpretability, data labelling and domain shift issues from training with simulations, which constitutes a common practice in astronomy.

show abstract

“…However, this techniques has not achieved a high accuracy (e.g., Huertas-Company et al 2014;Mager et al 2017). More recently, in efforts to find an automatic approach scalable to large numbers of objects (> 10 4 ) and applicable to future surveys such as the Large Survey of Space and Time (LSST) some works have used machine learning and convolutional neural networks to provide automated classification (e.g., Dieleman et al 2015;Aniyan & Thorat 2017;Ghosh 2019;Tadaki et al 2020;Cheng et al;Walmsley et al 2020). However, these networks are not applicable to all data sets as they are each trained on different data sets with different qualities with varying degrees of success.…”

Section: Morphological Classification Of Large Samplesmentioning

confidence: 99%

The Emergence of Bulges and Disks in the Universe

Hashemizadeh¹

2022

Preprint

View full text Add to dashboard Cite

This thesis makes use of the imaging data from the Advanced Camera for Surveys (ACS) of the Hubble Space Telescope (HST) in the Cosmic Evolution Survey (COSMOS) and the Deep Extragalactic VIsible Legacy Survey (DEV-ILS) field. We provide visual morphological classifications of 44,000 galaxies out to redshift z = 1 and above a stellar mass of 10 9.5 M (D10/ACS sample). We perform a robust Bayesian bulge-disk decomposition analysis of the D10/ACS sample. This study forms one of the largest morphological classification and structural analyses catalogues in this field to date. Using these catalogues, we explore the evolution of the stellar mass function (SMF) and the stellar mass density (SMD) together with the stellar mass-size relations (M * − R e ) of galaxies as a function of morphological type as well as for disks and bulges, separately. We quantify that one-third of the current stellar mass of the Universe was formed during the last 8 Gyr. We find that the moderate growth of the high-mass end of the SMF is dominated by the growth of elliptical systems and that the vast majority of the stellar mass of the Universe is locked up in disk+bulge systems at all epochs and that they increase their contribution to the total SMD with time. The contribution of the pure-disk morphology gradually decreases with time (∼ 40%), while ellipticals increase their contribution by a factor of 1.7 since z = 1. By decomposing galaxies into disks and bulges we quantify that on average ∼ 50% of the total stellar mass of the Universe at all epochs is in disk structures with this contribution relatively unchanged since z ∼ 0.6. With this comes more rapid growth of pseudo-bulges and spheroids (bulges and ellipticals) in mass. Furthermore, while the cosmic star-formation history is declining the Universe is transitioning from a disk dominated era to an epoch when pseudo-and classical-bulges are emerging. Finally, the M * − R e relations of galaxies and their components show that different components follow distinct relations which may reflect their distinct evolutionary pathways through which these systems have been built.The evolution of the M * − R e relations of galaxies and their components shows minimal variation. The unchanged scaling relations since z = 1 is consistent with the evolution along the M * − R e trends.

show abstract

Spin parity of spiral galaxies II: a catalogue of 80 k spiral galaxies using big data from the Subaru Hyper Suprime-Cam survey and deep learning

Cited by 20 publications

References 51 publications

Passive spiral galaxies deeply captured by Subaru Hyper Suprime-Cam

Passive spiral galaxies deeply captured by Subaru Hyper Suprime-Cam

The Dawes Review 10: The impact of deep learning for the analysis of galaxy surveys

The Emergence of Bulges and Disks in the Universe

Contact Info

Product

Resources

About