Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Modelling solar radial velocities and photometric variability with SOAPv4

Alba Barka, Eduardo Cristo, Ângela R. G. Santos, Nuno C. Santos, Khaled Al Moulla, Teresa Barata, Ricardo Gafeira, Michael Cretignier

Abstract

Stellar activity remains one of the main limitations in the detection of Earth-like planets using radial velocity (RV) measurements. The Sun, as the only star for which surface features can be spatially resolved, offers a unique testbed for studying the impact of active regions on RV and photometric variability. Using SOAPv4, we modelled solar RV and photometric variability induced by spots and faculae over long timescales. Our goal is to verify whether present-day, state-of-the-art models of the cross-correlation function correctly reproduce the observed variability. Moreover, we aim to assess how the choice of input data and identification technique influences the agreement between simulated and observed signals. To simulate solar RV and photometric time series, we first identified active regions in SDO images. This was done using mathematical morphological (MM) transforms applied to SDO/HMI and AIA images. MM identification was validated against other state-of-the-art identification methods. Using these inputs, we ran SOAPv4 to simulate solar RVs and photometry, and we validated the results with HARPS-N RV observations, as well as with VIRGO/SPM photometric measurements. The simulations that use MM identification achieved the best match with the observed RV time series, yielding residuals with a measured standard deviation of ~0.91 m/s. Other state-of-the-art methods produced higher filling factors and, consequently, larger discrepancies. The photometric simulations reproduced the overall variability trends. We demonstrate that MM transforms accurately identify solar active regions. Using these inputs, SOAPv4 reproduces the observed solar RV variability with a measured standard deviation of the residuals of ~0.91 m/s. Photometric simulations capture the overall variability trends, confirming that SOAP can reliably model the impact of both spots and faculae on solar RVs and photometry.

Modelling solar radial velocities and photometric variability with SOAPv4

Abstract

Stellar activity remains one of the main limitations in the detection of Earth-like planets using radial velocity (RV) measurements. The Sun, as the only star for which surface features can be spatially resolved, offers a unique testbed for studying the impact of active regions on RV and photometric variability. Using SOAPv4, we modelled solar RV and photometric variability induced by spots and faculae over long timescales. Our goal is to verify whether present-day, state-of-the-art models of the cross-correlation function correctly reproduce the observed variability. Moreover, we aim to assess how the choice of input data and identification technique influences the agreement between simulated and observed signals. To simulate solar RV and photometric time series, we first identified active regions in SDO images. This was done using mathematical morphological (MM) transforms applied to SDO/HMI and AIA images. MM identification was validated against other state-of-the-art identification methods. Using these inputs, we ran SOAPv4 to simulate solar RVs and photometry, and we validated the results with HARPS-N RV observations, as well as with VIRGO/SPM photometric measurements. The simulations that use MM identification achieved the best match with the observed RV time series, yielding residuals with a measured standard deviation of ~0.91 m/s. Other state-of-the-art methods produced higher filling factors and, consequently, larger discrepancies. The photometric simulations reproduced the overall variability trends. We demonstrate that MM transforms accurately identify solar active regions. Using these inputs, SOAPv4 reproduces the observed solar RV variability with a measured standard deviation of the residuals of ~0.91 m/s. Photometric simulations capture the overall variability trends, confirming that SOAP can reliably model the impact of both spots and faculae on solar RVs and photometry.
Paper Structure (19 sections, 4 equations, 12 figures, 4 tables)

This paper contains 19 sections, 4 equations, 12 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Data
  3. SDO images
  4. The Debrecen catalogue
  5. HARPS-N RVs
  6. VIRGO/SPM photometric data
  7. Active-region identification
  8. Correction and normalisation of images
  9. Spot identification from HMI
  10. Plage identification from AIA 1700 Å
  11. Facula identification from HMI magnetograms
  12. SOAPv4 simulations
  13. Results and discussion
  14. Filling factor comparison
  15. RV comparison
  16. ...and 4 more sections

Figures (12)

  • Figure 1: Left column: Example of original SDO images from 11 August 2016: (a) HMI intensitygram, (c) HMI magnetogram, and (e) AIA UV image. Right column: Corresponding active-region identifications for the same observation: (b) HMI intensitygram, (d) HMI magnetogram, and (f) AIA UV image. The contours mark the detected active regions.
  • Figure 2: Left: Larger facular regions identified using the H16 method. Right: Plages detected using the MM method. The example is from 11 August 2016.
  • Figure 3: Comparison of the spot filling factors measured from MM on HMI images (orange), from the Debrecen catalogue (blue), and following H16 (brown). Panel (a) shows the temporal evolution of each, while panels (b) and (c) show, respectively, the latter two as a function of the former, which we take as reference in our sample.
  • Figure 4: Comparison of the facula filling factors measured from MM on AIA images (orange), considering H16+MM (blue), and following H16 (dark red). Panel (a) shows the temporal evolution of each, while panels (b) and (c) show, respectively, the latter two as a function of the former, which we take as reference in our sample.
  • Figure 5: Left: Comparison of RVs from HARPS-N DRS and RV SOAPv4 simulations based on the MM method for both spots and plages (SOAP with MM). Middle: RV time series from HARPS-N DRS (black) and from SOAP with MM (top), together with the residuals, calculated as the difference between observed and simulated RVs and their std value (bottom). Right: Histogram of the residuals between the two datasets.
  • ...and 7 more figures