Solar photospheric velocities measured in space: a comparison between SO/PHI-HRT and SDO/HMI

TL;DR

This study demonstrates that LoS photospheric velocities measured by SO/PHI-HRT and SDO/HMI are in strong agreement when remapped to a common frame and corrected for large-scale solar motions. Using a 4-hour coeval dataset from 2023-03-29, the authors perform pixel-by-pixel comparisons in time and Fourier space, finding a linear relation with slope $0.94$ and a correlation of $0.90$, and estimate a formation-height separation of $10\pm12$ km. The penumbra and Evershed flow are analyzed, revealing very high concordance and highlighting residual calibration effects and algorithmic differences as the main sources of offsets and rms variations. These results support the combined use of PHI-HRT and HMI data for multi-view, 2-D flow studies and active-region tracking across solar vantage points.

Abstract

The Polarimetric and Helioseismic Imager (SO/PHI) onboard Solar Orbiter is a spectropolarimeter scanning the Fe I line at 617.3 nm, providing data of the solar photosphere. The same line is sampled by the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO) and many other on-ground instruments. In this paper, we aim at assessing the consistency between line-of-sight (LoS) velocity measurements from the two instruments. Reliable measurements of up and down flows from SO/PHI are crucial and unique when Solar Orbiter is facing the far side of the Sun. Also, a combination of measurements from two vantage points to study horizontal flows must rely on consistent observations. For this purpose, we compare the LoS velocity measured by SO/PHI's High Resolution Telescope (SO/PHI-HRT) and SDO/HMI on 29 March 2023, when Solar Orbiter was crossing the Sun-Earth line at a distance of 0.39 au from the Sun. Because such co-alignments are rare, this configuration offered an almost unique opportunity to directly compare data products from both telescopes. The data are aligned and remapped to allow a pixel-by-pixel comparison of the whole time series of 4 hours length. Temporal and spatial variations are considered for a direct combination of the measurements. The LoS velocity distributions are evaluated and a clear linear relation is found between the two instruments with a slope of 0.94 and a correlation of 90%. We find that the signals form at similar heights, with a separation of 10$\pm$12 km, which is larger than previous estimates. A close-up look at the penumbra of a sunspot and its Evershed flow is presented. We conclude that the signals inferred by SO/PHI-HRT and SDO/HMI show very good agreement and high correlation when instrumental effects and large-scale velocities on the Sun are properly accounted for.

Figures (8)

• Figure 1: Top left panel: intensity continuum as measured by SO/PHI-HRT. A 10 Mm scale is given as reference. Top right panel: LoS velocity as measured by SO/PHI-HRT. Bottom left panel: LoS magnetic field as measured by SO/PHI-HRT. Bottom right panel: LoS magnetic field as measured by SDO/HMI. The yellow box shows the SO/PHI-HRT FoV. The black arrows in the top left panel indicate the solar north (N) and east (E). Note that all the panels of this figure are shown in the SDO/HMI detector frame.
• Figure 2: k-$\omega$ diagrams. Top left panel: Power spectrum of the remapped SO/PHI-HRT time series (units: $m^2/s^2$). Top right panel: Power spectrum of the SDO/HMI time series interpolated to 60 s cadence (units: $m^2/s^2$). Bottom left panel: phase difference between SO/PHI-HRT and SDO/HMI where positive values means that SO/PHI-HRT is lagging behind SDO/HMI (units: degrees). The red dashed box shows the area used to compute the height difference between the two signals. Bottom right panel: coherence spectrum between the two signals. The blue and red contours show the 50% and 80% coherence respectively.
• Figure 3: Scatter plot distribution of the velocity field values as measured by the remapped SO/PHI-HRT ($x$ axis) and SDO/HMI ($y$ axis). The color scale shows the number of points in each bin on a logarithmic scale. The black line shows an orthogonal least square linear fit to the data points, while the contours depict the results of a 2-D Gaussian distribution fit. The slope and offset obtained by the linear fit, the Pearson correlation coefficient, and the 1$\sigma$ spread of the Gaussian fit are indicated in the top left corner of the figure.
• Figure 4: Temporal variation of the fitting results. The dots represent the results of the fits obtained frame by frame, whereas the dashed lines show the linear trends of these results. The error bars are not shown because they are comparable with the marker size. Top panel: slope (left axis, blue) and offset (right axis, orange) obtained by the linear fit. Middle panel: $x$ (left axis, green) and $y$ (right axis, red) coordinates of the mean of the 2-D Gaussian distribution. Bottom panel: root mean square (rms) of the remapped SO/PHI-HRT (left axis, purple) and of the SDO/HMI (right axis, brown) time series.
• Figure 5: Spatial dependence of the temporal correlation between the remapped SO/PHI-HRT and SDO/HMI signals (top panel), and of the slope (middle panel) and offset (bottom panel) obtained by the linear fit. The blue and red contours show the -200 and +200 G values of the SO/PHI-HRT LoS magnetic field respectively. The FoV is shown as a yellow box in Fig. \ref{['fig:data']}. The arrows in each panel indicate the north (N) and east (E) solar limb as in Fig. \ref{['fig:data']}.