The Edge-on Galaxies in the Pan-STARRS survey (EGIPS)

TL;DR

This paper presents a large, high-quality catalogue of over 16,500 edge-on galaxies from the Pan-STARRS survey, created using a neural network for selection and providing detailed morphological and photometric data.

Contribution

The study introduces a new catalogue of edge-on galaxies with improved selection accuracy using deep learning, covering three-quarters of the sky and including detailed morphological and photometric information.

Findings

Red sequence galaxies are redder and thicker than blue cloud galaxies.

Galaxy thickness varies with color, with redder galaxies being thicker.

Blue cloud galaxies tend to be thinner and bluer.

Abstract

We present a catalogue of 16551 edge-on galaxies created using the public DR2 data of the Pan-STARRS survey. The catalogue covers the three quarters of the sky above Dec.=-30 degrees. The galaxies were selected using a convolutional neural network, trained on a sample of edge-on galaxies identified earlier in the SDSS survey. This approach allows us to dramatically improve the quality of the candidate selection and perform a thorough visual inspection in a reasonable amount of time. The catalogue provides homogeneous information on astrometry, SExtractor photometry, and non-parametric morphological statistics of the galaxies. The photometry is reliably for objects in the 13.8-17.4 r-band magnitude range. According to the HyperLeda database, redshifts are known for about 63 percent of the galaxies in the catalogue. Our sample is well separated into the red sequence and blue cloud galaxy populations. The edge-on galaxies of the red sequence are systematically Delta(g-i)~0.1 mag redder than galaxies oriented at an arbitrary angle to the observer. We found a variation of the galaxy thickness with the galaxy colour. The red sequence galaxies are thicker than the galaxies of the blue cloud. In the blue cloud, on average, thinner galaxies turn out to be bluer. In the future, based on this catalogue it is intended to explore the three-dimensional structure of galaxies of different morphologies, as well as to study the scaling relations for discs and bulges.