A new way to constrain the densities of intra-group medium in groups of galaxies with convolutional neural networks

Abstract

Ram pressure (RP) can influence the evolution of cold gas content and star formation rates of galaxies. One of the key parameters for the strength of RP is the density of intra-group medium (${\rho }_{\mathrm{igm}}$), which is difficult to estimate if the X-ray emission from it is too weak to be observed. We propose a new way to constrain ${\rho }_{\mathrm{igm}}$ through an application of convolutional neural networks (CNNs) to simulated gas density and kinematic maps galaxies under strong RP. We train CNNs using $9\times 10^4$ 2D images of galaxies under various RP conditions, then validate performance with $10^4$ new test images. This new method can be applied to real observational data from ongoing WALLABY and SKA surveys to quickly obtain estimates of ${\rho }_{\mathrm{igm}}$. Simulated galaxy images have $1.0$ kpc resolution, which is consistent with that expected from the future WALLABY survey. The trained CNN models predict the normalised IGM density, ${\hat{\rho }}_{\mathrm{igm}}$ where $0.0\le {\hat{\rho }}_{\mathrm{igm},\mathrm{n}}<10.0$, accurately with root mean squared error values ($\mathrm{RMSE}$) of $0.72$, $0.83$ and $0.74$ for the density, kinematic and joined 2D maps, respectively. Trained models are unable to predict the relative velocity of galaxies with respect to the IGM ($v_{\mathrm{rel}}$) precisely, and struggle to generalise for different RP conditions. We apply our CNNs to the observed HI column density map of NGC 1566 in the Dorado group to estimate its IGM density.

15 pages, 12 figures

References in corpus (15)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Improving neural networks by preventing co-adaptation of feature detectors
• ADADELTA: An Adaptive Learning Rate Method
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Strangulation in Galaxy Groups
• Classifying Radio Galaxies with Convolutional Neural Network
• Emulation of reionization simulations for Bayesian inference of astrophysics parameters using neural networks
• Deep learning for galaxy surface brightness profile fitting
• Dust-regulated galaxy formation and evolution:A new chemodynamical model with live dust particles
• The Journey Counts: The Importance of Including Orbits when Simulating Ram Pressure Stripping
• WALLABY Early Science - III. An HI Study of the Spiral Galaxy NGC 1566
• Satellite dwarf galaxies: Stripped but not quenched
• The effects of ram-pressure stripping on the internal kinematics of simulated spiral galaxies
• Explaining the enhanced star formation rate of Jellyfish galaxies in galaxy clusters
• Classifying the formation processes of S0 galaxies using Convolutional Neural Networks

Cited by in corpus (3)

• Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
• Quantifying the fine structures of disk galaxies with deep learning:Segmentation of spiral arms in different Hubble types
• A self-supervised, physics-aware, Bayesian neural network architecture for modelling galaxy emission-line kinematics