DeepVel: deep learning for the estimation of horizontal velocities at the solar surface

Abstract

Many phenomena taking place in the solar photosphere are controlled by plasma motions. Although the line-of-sight component of the velocity can be estimated using the Doppler effect, we do not have direct spectroscopic access to the components that are perpendicular to the line-of-sight. These components are typically estimated using methods based on local correlation tracking. We have designed DeepVel, an end-to-end deep neural network that produces an estimation of the velocity at every single pixel and at every time step and at three different heights in the atmosphere from just two consecutive continuum images. We confront DeepVel with local correlation tracking, pointing out that they give very similar results in the time- and spatially-averaged cases. We use the network to study the evolution in height of the horizontal velocity field in fragmenting granules, supporting the buoyancy-braking mechanism for the formation of integranular lanes in these granules. We also show that DeepVel can capture very small vortices, so that we can potentially expand the scaling cascade of vortices to very small sizes and durations.

9 pages, 5 figures, accepted for publication in A&A

References in corpus (6)

• A solar surface dynamo
• Convectively driven vortex flows in the Sun
• The second flight of the SUNRISE balloon-borne solar observatory: overview of instrument updates, the flight, the data and first results
• On the Formation of Active Regions
• The history of a quiet-Sun magnetic element revealed by IMaX/SUNRISE
• Analysing the effects of apodizing windows on local correlation tracking using Nirvana simulations of convection

Cited by in corpus (35)

• Enhancing SDO/HMI images using deep learning
• Stokes Inversion based on Convolutional Neural Networks
• Real-time multiframe blind deconvolution of solar images
• Segmentation of coronal holes in solar disk images with a convolutional neural network
• Classification of Magnetohydrodynamic Simulations using Wavelet Scattering Transforms
• Fast Solar Image Classification Using Deep Learning and its Importance for Automation in Solar Physics
• Persistent magnetic vortex flow at a supergranular vertex
• Solar image denoising with convolutional neural networks
• Combining magneto-hydrostatic constraints with Stokes profiles inversions
• Mimicking spectropolarimetric inversions using convolutional neural networks
• Combining magneto-hydrostatic constraints with Stokes profile inversions. II. Application to Hinode/SP observations
• Chromospheric swirls I. Automated detection in H$\alpha$ observations and their statistical properties
• Coronal hole picoflare jets are progenitors of both fast and Alfvénic slow solar wind
• High-Cadence Imaging and Imaging Spectroscopy at the GREGOR Solar Telescope - A Collaborative Research Environment for High-Resolution Solar Physics
• Accurately constraining velocity information from spectral imaging observations using machine learning techniques
• Multi-Scale Deep Learning for Estimating Horizontal Velocity Fields on the Solar Surface
• The importance of horizontal Poynting flux in the solar photosphere
• Average radial structures of gas convection in the solar granulation
• Using the infrared iron lines to probe solar subsurface convection
• Dynamics of Trees of Fragmenting Granules in the Quiet Sun: Hinode/SOT Observations Compared to Numerical Simulation
• Supervised Neural Networks for Helioseismic Ring-Diagram Inversions
• New approach for analysing dynamical processes on the surface of photospheric vortex tubes
• Power spectrum of turbulent convection in the solar photosphere
• Spectropolarimetric Inversion in Four Dimensions with Deep Learning (SPIn4D): I. Overview, Magnetohydrodynamic Modeling, and Stokes Profile Synthesis
• Photospheric plasma and magnetic field dynamics during the formation of solar AR 11190
• Validation of the PDFI_SS method for electric field inversions using a magnetic flux emergence simulation
• A nonlinear solar magnetic field calibration method for the filter-based magnetograph by the residual network
• Coronal kink oscillations and photospheric driving: combining SolO/EUI and SST/CRISP high-resolution observations
• Machine learning spatio-temporal epidemiological model to evaluate Germany-county-level COVID-19 risk
• Proper Motions of Sunspots' Umbral Dots at High Temporal and Spatial Resolution
• Multi-height probing of horizontal flows in the solar photosphere
• Inference of horizontal velocity fields from the induction equation in the solar atmosphere. I. Analytical and numerical solutions in 2D
• Emergence of magnetic flux sheets in the quiet Sun. I. Statistical properties
• Structure and dynamics of erupting solar prominences using the Rolling Hough Transform: Toward a feature-oriented classification
• Diagnosing Turbulence in the Neutral and Molecular Interstellar Medium of Galaxies