betaFeedback

WorksPhysics and Astronomy

High-Resolution DKIST/ViSP Observation Reveals Small-scale Photospheric Magnetic Flux beneath a Coronal Hole Region

Abstract

The magnetic unipolarity of photospheric regions beneath coronal holes is well established at spatial resolutions of about 1 arcsecond, but the effect of smaller-scale magnetic structures remains uncertain. We compare co-aligned observations from the Helioseismic and Magnetic Imager (HMI) aboard the Solar Dynamics Observatory and the Visible Spectro-Polarimeter (ViSP) at the Daniel K. Inouye Solar Telescope (DKIST). HMI observes the Fe I 6173 Å line, while ViSP observes the Fe I 6302.5 Å line; both form at similar photospheric heights. We derive the line-of-sight magnetic field (B_LOS) from ViSP spectropolarimetric data using a Milne-Eddington inversion and compare it with the HMI field derived from vector magnetograms over a common coronal-hole region. ViSP detects smaller magnetic structures with stronger fields than HMI. The absolute mean field and mean absolute field are higher by factors of 1.30 and 1.37, respectively. HMI measures approximately 28% less unsigned magnetic flux and 24% less net signed magnetic flux than ViSP. A comparison of spatially degraded ViSP data with HMI indicates that differences in effective resolution and associated spatial averaging contribute to these discrepancies. Nevertheless, the magnetic flux imbalance remains similar: 69% for HMI and 65% for ViSP. Thus, the additional small-scale flux detected by ViSP does not substantially change the measured unipolarity in this region. Because ViSP has a limited field of view, observations covering entire coronal holes are needed to test whether this result holds more generally.

Introduction

Coronal holes exhibit distinct observational signatures across different layers of the solar atmosphere. In the upper atmosphere, including the corona and upper transition region, they display reduced emission in extreme-ultraviolet (EUV) and soft X-ray (SXR) wavelengths [47, 6, 43]. This lower intensity is caused by plasma densities and temperatures that are significantly lower than those of the surrounding quiet Sun (QS) and active regions (AR). The reduced electron density strongly contributes to the lower EUV and X-ray intensity of coronal holes, since optically thin coronal emission scales approximately as ne2G(T)n_{\mathrm{e}}^{2}G(T) [8], where nen_{\mathrm{e}} is the electron density and G(T)G(T) is the temperature-dependent contribution function. Thus, even a modest decrease in density can produce a substantial reduction in emission. The temperature difference between coronal holes and the quiet Sun is comparatively smaller, but it is sufficient to shift the plasma away from the temperature range in which commonly observed coronal EUV lines are formed, further reducing the observed intensity. EUV bands centered at 193 Å, 171 Å, and 211 Å provide the highest contrast between coronal holes and non-coronal hole regions [33, 13] making these wavelengths most useful for identifying coronal holes.

In the chromosphere, coronal holes are observed as features of enhanced intensity in He I 10830 Å filtergrams, resulting from reduced chromospheric absorption of the underlying continuum emission (Brajša et al. 1996). However, their weak contrast with the surrounding chromosphere makes them difficult to identify reliably ([14]). In the photosphere, coronal holes do not exhibit a distinct intensity signature and therefore cannot be directly identified in continuum intensity observations ([9]).

Coronal holes are characterized by a predominantly unipolar magnetic field configuration (e.g. [9, 11, 15 b]). In them, magnetic field lines originate below the solar surface and extend outward through the solar atmosphere into interplanetary space. An open magnetic field topology enables the solar plasma to escape more efficiently from the solar atmosphere at a speed of approximately 400400–800800 km s−1^{-1} ([45]). As a result, coronal holes are widely recognized as the primary sources of the fast solar wind ([22, 29, 50, 26]). Coronal holes are also the primary source of open magnetic flux in the upper solar atmosphere and play a crucial role in shaping the heliospheric magnetic field ([49, 48]). Direct measurements of these coronal magnetic fields remain observationally challenging due to the low plasma density, high temperature, and weak magnetic-field strength of the solar corona ([39, 40, 2]). As an alternative approach, magnetic field extrapolation techniques are often employed to infer the coronal and heliospheric magnetic fields from photospheric measurements ([51]). These extrapolations use photospheric magnetograms as the lower boundary condition. Potential-field extrapolations use the normal or radial magnetic-field component of the photospheric magnetic field, which can be estimated from line-of-sight magnetograms ([1]), whereas nonlinear force-free field (NLFFF) extrapolations typically use all three components of the photospheric magnetic field vector ([52]). The accuracy and reliability of such extrapolations therefore critically depend on the accuracy and consistency of the underlying photospheric magnetic field measurements.

In this context, the high spatial and spectral resolution of the Visible Spectro-Polarimeter (ViSP) ([7]) detects small-scale photospheric magnetic structures that may remain unresolved in lower-resolution observations, providing better input for magnetic-field extrapolations. To date, magnetic-field extrapolations have been primarily used on photospheric magnetograms from instruments such as the Michelson Doppler Imager (MDI; pixel scale ≈2′′\approx2'', effective resolution ≈4′′\approx4''; [41]), the Helioseismic and Magnetic Imager (HMI; pixel scale ≈0.5′′\approx0.5'', effective resolution ≈1′′\approx1''; [42]), and the Global Oscillation Network Group (GONG; spatial resolution ≈2.5′′\approx2.5''–5′′5''; [10]). Among currently available full-disk instruments, HMI provides the highest spatial resolution. However, even this resolution remains insufficient to resolve the smallest relevant magnetic structure. Numerical simulations suggest that individual photospheric magnetic structures may be as small as 0.05′′0.05'', corresponding to approximately 35 km ([31, 34]). To reliably infer photospheric magnetic fields, an instrument must have sufficient spatial resolution to resolve magnetic structures at fundamental physical scales. Because these magnetic fields are commonly used as lower boundary conditions for coronal and heliospheric magnetic-field models, resolution-related spatial averaging can introduce uncertainties into the modeled fields.

The identification of coronal holes typically begins with visual inspection of coronal images, where they are usually identified as contiguous dark regions. However, not all dark coronal regions are coronal holes, since filaments [32] and coronal voids [28] can also exhibit reduced emission. Magnetic unipolarity is therefore a key follow-up criterion for validating dark regions as coronal holes, since filaments and coronal voids generally exhibit bipolar magnetic field distributions. This distinction in polarity is well established at a spatial resolution of approximately 1′′1^{\prime\prime} using HMI photospheric magnetograms. The key question is whether ViSP’s higher spatial resolution alters the inferred magnetic properties of the coronal-hole region. These properties include the total unsigned flux (Φu\Phi_{\mathrm{u}}), total signed flux (Φs\Phi_{\mathrm{s}}) absolute means field (∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle|), mean absolute field (⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle), the polarity ratio (Φr\Phi_{\mathrm{r}}), the skewness (SS), the flux imbalance (Φimb\Phi_{\mathrm{imb}}). Their mathematical definitions are presented in the second column of Table 1. Comparing these quantities between ViSP BlosB_{\mathrm{los}} and HMI BlosB_{\mathrm{los}} allows us to determine how small-scale magnetic features that are unresolved or spatially averaged in HMI affect the inferred field strength, the BlosB_{\mathrm{los}} distribution, and the magnetic unipolarity of the region.

PropertyDefinitionHMIViSP ME Inversion
Magnetic Flux Properties
Total unsigned flux, Φu\Phi_u [Mx]Φu=∑∣Blos∣A\Phi_u = \sum|B_{\mathrm{los}}|A1.189×10201.189 \times10^{20}1.656×10201.656 \times10^{20}
Total signed flux, Φs\Phi_s [Mx]Φs=∑BlosA\Phi_s = \sum B_{\mathrm{los}}A8.162×10198.162 \times10^{19}1.073×10201.073 \times10^{20}
Positive flux, Φ+\Phi^+ [Mx]∑Blos>0BlosA\sum_{B_{\mathrm{los}}>0} B_{\mathrm{los}}A1.003×10201.003 \times10^{20}1.364×10201.364 \times10^{20}
Negative-flux magnitude, ∣Φ−∣|\Phi^-| [Mx]∑Blos<0BlosA\sum_{B_{\mathrm{los}}<0} B_{\mathrm{los}}A1.866×10191.866 \times10^{19}2.913×10192.913 \times10^{19}
Polarity-flux ratio, ϕr\phi_rΦ+∣Φ−∣\frac{\Phi^+}{|\Phi^-|}5.45.44.74.7
Absolute mean field [G]∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle|4.94.96.36.3
Mean absolute field [G]⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle7.17.19.79.7
Flux Enhancement (ViSP/HMI)
Total unsigned-flux enhancementΦuViSPΦuHMI\frac{\Phi_u^{\mathrm{ViSP}}}{\Phi_u^{\mathrm{HMI}}}—1.41.4
Positive-flux enhancementΦViSP+ΦHMI+\frac{\Phi^+_{\mathrm{ViSP}}}{\Phi^+_{\mathrm{HMI}}}—1.41.4
Negative-flux enhancement∣ΦViSP−∣∣ΦHMI−∣\frac{|\Phi^-_{\mathrm{ViSP}}|}{|\Phi^-_{\mathrm{HMI}}|}—1.61.6
Distribution Properties
Flux imbalance, Φimb\Phi_{\mathrm{imb}} [%]100∣Φs∣Φu100\frac{|\Phi_s|}{\Phi_u}69696565
Skewness, SS3[⟨Blos⟩−median⁡(Blos)]σBlos\frac{3[\langle B_{\mathrm{los}}\rangle-\operatorname{median}(B_{\mathrm{los}})]}{\sigma_{B_{\mathrm{los}}}}0.500.500.460.46
Missing Flux in HMI Relative to ViSP
Missing unsigned flux, Φm,unsigned\Phi_{\mathrm{m,unsigned}} [%](1−ΦuHMIΦuViSP)×100\left(1-\frac{\Phi_u^{\mathrm{HMI}}}{\Phi_u^{\mathrm{ViSP}}}\right)\times100—2828
Missing signed flux, Φm,signed\Phi_{\mathrm{m,signed}} [%](1−∣ΦsHMI∣∣ΦsViSP∣)×100\left(1-\frac{|\Phi_s^{\mathrm{HMI}}|}{|\Phi_s^{\mathrm{ViSP}}|}\right)\times100—2424

Table 1. Magnetic flux properties derived from the HMI BlosB_{\mathrm{los}} and ViSP BlosB_{\mathrm{los}} maps over the common field of view within the coronal hole. The listed quantities include the total unsigned flux Φu\Phi_u, total signed flux Φs\Phi_s, positive flux Φ+\Phi^+, negative-flux magnitude ∣Φ−∣|\Phi^-|, polarity-flux ratio ϕr=Φ+/∣Φ−∣\phi_r = \Phi^+/|\Phi^-|, absolute mean field ∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle|, and mean absolute field ⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle. The table also presents the ViSP-to-HMI flux-enhancement ratios, flux imbalance Φimb\Phi_{\mathrm{imb}}, skewness SS, and the percentages of missing unsigned and signed magnetic flux, Φm,unsigned\Phi_{\mathrm{m,unsigned}} and Φm,signed\Phi_{\mathrm{m,signed}}, respectively, in HMI relative to the ViSP ME inversion.

In this work, high-resolution spectropolarimetric observations obtained with ViSP and retrieved from the Daniel K. Inouye Solar Telescope (DKIST; [36]) Data Center were used to investigate the photospheric magnetic field. ViSP measures wavelength-dependent Stokes profiles, which contain information about the magnetic field in the solar atmosphere. Because the magnetic field cannot be measured directly, it is inferred from its signatures in polarized spectral lines using diagnostic models, techniques, and assumptions. We infer BlosB_{\mathrm{los}} from the Fe I 6302.5 Å line using a Milne–Eddington (ME) inversion and compare the resulting magnetic-field properties with those derived from the corresponding HMI vector magnetograms.

The paper is organized as follows: Section 2 describes the observational data and setup; Section 3 outlines the methodology; Section 4 presents the results and analysis; and Section 5 summarizes the main conclusions.

Datasets

This study combines space-based and ground-based observations to investigate BlosB_{\mathrm{los}} in a common photospheric region beneath a coronal hole. The space-based observations were obtained with the Atmospheric Imaging Assembly (AIA; [24]) and the Helioseismic and Magnetic Imager (HMI; [42]) onboard the Solar Dynamics Observatory (SDO; [30]). The AIA data were used to identify the coronal hole in the solar corona, while the HMI data were selected to investigate the magnetic properties of the underlying photospheric region. Ground-based observations acquired with ViSP at DKIST were also used to study the photospheric magnetic properties at higher spatial resolution. The coronal hole region analyzed in this study was located at approximately −30∘-30^\circ solar latitude and near the central meridian. The observations were obtained on 2024 April 24 between 19:42 UT and 20:29 UT using both DKIST and SDO.

SDO

SDO carries three instruments. Data from two instruments, AIA and HMI are used in this study. AIA provides full-disk EUV images with a plate scale of 0′′.60^{\prime\prime}.6 pixels−1^{-1}, an effective spatial resolution of 1′′.51^{\prime\prime}.5, and a cadence of 12 s [24]. The AIA 193 Å channel is used to identify the coronal hole. Figure 1a shows an AIA 193 Å intensity map of the solar corona, in which the coronal hole appears as a region of reduced intensity. The red contour outlines the coronal-hole boundary, identified as described in § 3. HMI provides full-disk BlosB_{\mathrm{los}} measurements every ∼ 45 s, with a plate scale of 0′′.50^{\prime\prime}.5 pixels−1^{-1} and an effective spatial resolution of ∼ 0′′.910^{\prime\prime}.91 in the photosphere [42]. These measurements were derived from spectropolarimetric observations of the Fe I 6173 Å spectral line. HMI also provides full-disk vector magnetograms, derived from Stokes inversions of the same Fe I 6173 Å spectral line, at a cadence of ∼ 720 s [12]. The BlosB_{\mathrm{los}} derived from the HMI vector magnetograms is used in this study because it provides a higher signal-to-noise ratio than the HMI 45 s line-of-sight magnetograms. Figure 1b shows the HMI BlosB_{\mathrm{los}} map of the photospheric magnetic field beneath the coronal hole region shown in Figure 1a. The red contour is the same coronal hole boundary shown in Figure 1a. Science-ready Level 1.5 AIA and HMI data were obtained via the Virtual Solar Observatory1 using SunPy [46]. An example of the data-acquisition procedure can be found at the Zenodo link (Katwual 2025)

AIA 193 Å intensity map and HMI 6173 Å magnetogram of a coronal hole

Figure 1. Panel ‘a’: AIA 193 Å intensity map showing the coronal hole as a region of reduced EUV emission. Panel ‘b’: Co-temporal HMI magnetogram (720 s cadence) of the same region. The red contours in both panels outline the coronal-hole boundary. The dashed white box marks the high-resolution field of view observed by Visible spectro-polarimetry (ViSP) in the Fe I 6302.5 Å spectral line, which is the region analyzed in this study. The observations were obtained on 2024 April 24 from 19:42 to 20:29 UT.

DKIST

ViSP at DKIST is a slit-scanning spectro-polarimeter designed to provide high spatial and spectral resolution observations. ViSP records the full Stokes vector (II, QQ, UU, VV), where II is the intensity, QQ and UU are linear polarizations, and VV is the circular polarization, over the wavelength range 3800–9000 Å. The instrument can simultaneously observe up to three spectral regions using separate detectors.

In this observation, spectral windows containing the Fe I 6301.5 Å and Fe I 6302.5 Å lines, the Na I D1_1 5896 Å line, and the Ca II 8542 Å line were recorded. These lines probe different layers of the solar atmosphere. The Fe I 6301.5 Å and Fe I 6302.5 Å lines form in the photosphere, the Na I D1_1 5896 Å samples the upper photosphere to the lower chromosphere, and the Ca II 8542 Å primarily forms in the chromosphere. Since this work focuses on the BlosB_{\rm los} at photospheric heights, we analyzed only the Fe I 6302.5 Å spectral line. The inferred ViSP BlosB_{\rm los} is compared with the HMI BlosB_{\rm los} derived from an HMI vector magnetogram, as both diagnostics probe similar photospheric layers [37].

The white dotted box shown in both panels of Figure 1 indicates the ViSP field of view. This observation is obtained under Proposal ID pid_2_28. The data corresponds to the Level 1 product L1-WKCSV, processed with the 10_to_l1_visp calibration workflow version 5.4.02. The dataset consists of a 1000-step spectro-polarimetric raster with four Stokes parameters (II, QQ, UU, VV). The observed spectral window spans 6296.49–6307.29 Å, covering the Fe I 6301.5 Å and Fe I 6302.5 Å lines, with a

spectral sampling of 0.012 Å pixel−1^{-1}. The spatial sampling is 0″.024 along the slit and 0″.052 in the raster direction, with a slit width of 0″.054. With 1000 raster steps and 2547 spatial pixels along the slit, the observations cover a field of view of approximately 62″ × 52″. The ViSP field of view spans helioprojective coordinates from (−5.40″, −391.75″) to (−57.16″, −453.51″) relative to disk center, with a center located at (X,Y)=(−31′′.28,−422′′.63)(X,Y) = (-31''.28,-422''.63). The average temporal cadence between successive raster steps is 2.82 s. The Level 1 data values are normalized relative to the average solar disk-center intensity following the standard ViSP calibration procedure. These observations target the photospheric magnetic field underlying a coronal hole region. They are used to investigate the small-scale magnetic flux distribution and its contribution to the open magnetic field. The high spatial, spectral, and polarimetric precision of ViSP enables detailed measurements of fine-scale magnetic elements that remain unresolved in lower-resolution observations.

Data Reduction

The Level 1 data used in this study are publicly available through the National Solar Observatory (NSO)/DKIST Archive3, and the basic calibration steps have already been applied. These include dark and background subtraction, gain calibration, geometric-distortion correction, polarimetric demodulation, beam combination, and correction for telescope-induced polarization. Details of the ViSP Level 1 calibration pipeline are provided in the DKIST documentation.2 The continuum-normalized Stokes intensity (I/IcI/I_c), and circular-polarization (V/IcV/I_c), maps from this dataset are shown in Figure 2a and Figure 2b, respectively. The I/IcI/I_c map shows spatial variations in the continuum-normalized Fe I 6302.5 Å line intensity, whereas the V/IcV/I_c map shows the circular-polarization signal produced primarily by the longitudinal Zeeman effect. It is therefore sensitive to the strength and polarity of BlosB_{\mathrm{los}}. Our objective is to derive BlosB_{\mathrm{los}} from the observed circular polarization profiles. We need additional processing of these data before performing the inversions. This processing includes polarization crosstalk correction and wavelength calibration of the Fe I 6302.5 Å spectral line; further details are available at the Zenodo link (katuwal et al. 2026a). The blue crosses in both panels of Figure 2 mark the selected location at raster step 270 and spatial pixel 450. This location is used to demonstrate the effect of the crosstalk correction. The correction procedure and its effect on the representative Stokes profiles are presented and discussed in detail in Appendix 7. The absolute wavelength calibration is obtained by matching the mean quiet-Sun spectrum from the ViSP observations to the disk-center reference spectrum from the BASS 2000 solar atlas,3 as described in Yadav et al. [53]. The data are then rebinned along the slit to improve the signal-to-noise ratio and produce approximately square spatial pixels. The resulting processed Stokes profiles are used as input to the ME inversions. The data preprocessing code is available on Zenodo link (katuwal et al. 2026a)

Figure 2. Spatial maps of the Fe I 6302.5 Å line obtained from the ViSP raster corresponding to the region outlined by the white dotted box in Figure 1. Panel ‘a’ shows the Stokes I/IcI/I_c intensity, while panel ‘b’ shows the corresponding Stokes V/IcV/I_c signal. Both quantities were averaged over wavelength indices 590–607 (corresponding to the wavelength range 6302.392–6302.595 Å) around the line core (Fe I 6302.5 Å). The horizontal axis represents the spatial position along the slit, while the vertical axis represents the raster-scan step. The blue cross marks the selected location at slit pixel 450 and raster step 270, which was used to examine the effect of the crosstalk correction on representative Stokes profiles. See Appendix § 7 for details.

Methodology

The observational target is a coronal hole region. Because coronal holes are difficult to identify directly at photospheric heights, we use the AIA 193 Å channel as a reference, as shown in Figure 1a. This wavelength provides a strong contrast between coronal holes and the surrounding corona ([33]) and is commonly used to identify coronal hole boundaries. The boundary is determined using an intensity threshold of 100 DN, following [19], and is indicated by the red contour outlining the region of reduced emission. The red contour is then overlaid on the BlosB_{\mathrm{los}} map derived from the HMI vector magnetogram to investigate the magnetic properties of the photospheric region underlying the coronal hole, as shown in Figure 1b. HMI data products are ready to use for magnetic-field analysis. However, ViSP data require further processing and inversion to obtain BlosB_{\mathrm{los}}.

Magnetic Field Inversions

The ViSP instrument does not directly provide BlosB_{\mathrm{los}}. Instead, it records the full Stokes profiles (II, QQ, UU, VV), which requires further postprocessing to infer the magnetic field vector. We derive BlosB_{\mathrm{los}} from the Fe I 6302.5 Å spectral line using ME inversion of the Stokes profiles using the SPIN inversion code ([54]), which solves the radiative transfer equation under a ME atmospheric framework. This model assumes that all physical parameters remain constant with height, except for the source function, which varies linearly with optical depth. After retrieving BlosB_{\mathrm{los}}, the resulting ViSP BlosB_{\mathrm{los}} map is co-aligned with the HMI BlosB_{\mathrm{los}} map. For reference, the HMI magnetogram has a noise level of approximately 5–10 G ([42, 44, 5]), while the ViSP measurements have an estimated noise level of a few gauss. These values provide an approximate indication of the weakest magnetic fields detectable by the two instruments. Within the identified coronal-hole region, we calculate magnetic flux properties, flux-enhancement ratios, distribution properties, and the fractions of unsigned and signed flux that HMI misses. These quantities are then compared between the two instruments. A complete summary of the comparison of the derived magnetic characteristics is presented in Table 1.

Spatial Co-alignment of HMI and ViSP

To co-align the high-resolution ViSP Fe I 6302.5 Å observations with the HMI magnetograms, we constructed a reference BlosB_{\mathrm{los}} map from the HMI 720 s vector data product, hmi.b_720s. Using the native helioprojective World Coordinate System (WCS) preserved in the HMI FITS header, the line-of-sight magnetic field was computed as

Blos=Bcos⁡γ,(1)B_{\mathrm{los}} = B \cos\gamma, \tag*{(1)}

where BB is the magnetic-field strength and γ\gamma is the field inclination. The corresponding ViSP BlosB_{\mathrm{los}} map was derived from the Level-1 Stokes VV FITS file, with the spatial plate scales along the slit and raster directions extracted from the DKIST header keywords CDELT1 and CDELT2.

The ViSP field of view, indicated by the white dotted box in Figure 1, was defined using two helioprojective coordinate pairs, (−5.40′′,−391.75′′)(-5.40^{\prime\prime}, -391.75^{\prime\prime}) and (−57.16′′,−453.51′′)(-57.16^{\prime\prime}, -453.51^{\prime\prime}), measured relative to the center of the solar disk. These coordinates were converted into the native Helioprojective coordinate frame using the SunPy software package to extract the corresponding spatial submap from the full-disk HMI data. To account for the ViSP slit-scanning geometry, the ViSP raster was rotated by 90∘90^\circ and flipped horizontally to align its spatial axes with helioprojective longitude and latitude. This procedure ensured that the resolved magnetic polarities were physically consistent with the reference HMI BlosB_{\mathrm{los}} map. The final alignment was verified by plotting the ViSP BlosB_{\mathrm{los}} (Figure 3b) and HMI BlosB_{\mathrm{los}} (Figure 3a) images side by side on the same helioprojective arcsecond scale within the shared WCS framework. The two panels cover an identical field of view.

Results and Analysis

Visual Comparison of the HMI and ViSP Magnetograms

Figure 3a and Figure 3b show the co-aligned HMI and ViSP magnetograms, respectively. The main positive-polarity features near the center and the lower-left edge appear at the same locations and have similar shapes in both maps. The ViSP BlosB_{\mathrm{los}} map (Figure 3b) displays finer spatial features and resolves additional small-scale magnetic features, particularly within the weak background field and around the main magnetic concentrations. Most of the field of view contains weak BlosB_{\mathrm{los}} values near zero, together with small patches of both polarities. The positive-polarity features are generally larger and more elongated, whereas the negative-polarity features are smaller and more widely scattered. In the HMI BlosB_{\mathrm{los}} (Figure 3a), the positive-polarity features appear more blurred, and many of the small negative-polarity features are less clearly visible. Overall, the region contains weak, mixed-polarity photospheric fields and no strong sunspot-like magnetic concentrations. This visual comparison provides context for the quantitative analysis in which we examine whether the similarities and differences seen in the two maps are also reflected in their magnetic field distributions and derived magnetic properties.

Distribution of B-los

fig:3c compares the BlosB_{\mathrm{los}} distributions measured by HMI (green curve) and ViSP (blue curve) over the same field of view. Both curves peak strongly near Blos≈0B_{\mathrm{los}} \approx0 G, indicating that relatively weak magnetic fields occupy most of the region. To quantify this concentration, we calculate the fraction of pixels satisfying ∣Blos∣<∣Blos∣max⁡/10\lvert B_{\mathrm{los}}\rvert< \lvert B_{\mathrm{los}}\rvert_{\max}/10 for each distribution and determine the unsigned magnetic flux contributed by those pixels relative to the total unsigned flux. For ViSP, ∣Blos∣max⁡/10=79.8\lvert B_{\mathrm{los}}\rvert_{\max}/10 = 79.8 G, and 98.3% of the pixels fall within this range, contributing approximately 68.2% of the total unsigned magnetic flux. For HMI, ∣Blos∣max⁡/10=55.5\lvert B_{\mathrm{los}}\rvert_{\max}/10 = 55.5 G, and 98.1% of the pixels fall within this range, contributing approximately 68.0% of the total unsigned magnetic flux. The HMI BlosB_{\mathrm{los}} distribution is relatively narrow, whereas the ViSP distribution is substantially broader and spans a wider range of field strengths. The ViSP distribution extends to approximately 800 G on the positive side and −297-297 G on the negative side, compared with approximately 555 G and −57-57 G, respectively, for HMI. An isolated ViSP value near −400-400 G is treated as an artifact and is therefore not considered representative of the distribution. This value is spatially isolated, and the corresponding Stokes VV signal does not show a sufficiently strong circular-polarization signal to support a magnetic field of approximately −400-400 G. Including or excluding this outlier does not significantly affect the calculated statistics. For HMI, however, the high-field pixels with Blos>250B_{\mathrm{los}} > 250 G occur mainly in a compact group of neighboring pixels, with field strengths extending up to approximately 555 G. Although some of these values appear in histogram bins containing only one or two pixels, they are associated with a coherent magnetic concentration and are therefore not considered artifacts. Both BlosB_{\mathrm{los}} distributions are positively asymmetric, consistent with the dominance of positive-polarity magnetic fields in the coronal-hole region. The broader ViSP distribution indicates that ViSP detects stronger small-scale magnetic concentrations that are spatially averaged in HMI, supporting the visual evidence that ViSP reveals additional small-scale magnetic flux that is less clearly resolved in the HMI map.

Magnetic unipolarity

Magnetic unipolarity refers to the dominance of one magnetic polarity, either positive or negative, in the region under study. The unsigned magnetic flux (Φu\Phi_{\mathrm{u}}) represents the total magnetic flux contributed by both positive and negative magnetic polarities, regardless of their signs. In contrast, the signed magnetic flux (Φs\Phi_{\mathrm{s}}) represents the net magnetic flux within a region, accounting for the contributions of both polarities. The unsigned magnetic flux is calculated as

Φu=∑i∣Bi∣Apixel,(2)\Phi_{\mathrm{u}} = \sum_i \lvert B_i\rvert A_{\mathrm{pixel}}, \tag*{(2)}

and the signed magnetic flux is calculated as

Φs=∑iBiApixel,(3)\Phi_{\mathrm{s}} = \sum_i B_i A_{\mathrm{pixel}}, \tag*{(3)}

where BiB_i is the BlosB_{\mathrm{los}} in pixel ii, and ApixelA_{\mathrm{pixel}} is the physical area of a single pixel. The summation includes all valid pixels.

We characterize the magnetic-field distribution using the skewness (SS) and quantify the degree of magnetic unipolarity using the flux imbalance (Φimb\Phi_{\mathrm{imb}}), defined as

S=3[⟨Blos⟩−median⁡(Blos)]σBlos,(4)S = \frac{3\left[\langle B_{\mathrm{los}}\rangle- \operatorname{median}(B_{\mathrm{los}})\right]}{\sigma_{B_{\mathrm{los}}}}, \tag*{(4)}

and

Φimb=(ΦsΦu)×100%,(5)\Phi_{\mathrm{imb}} = \left(\frac{\Phi_{\mathrm{s}}}{\Phi_{\mathrm{u}}}\right) \times100\%, \tag*{(5)}

respectively, where σBlos\sigma_{B_{\mathrm{los}}} is the standard deviation of the BlosB_{\mathrm{los}} distribution. Skewness measures the asymmetry in the shape of the distribution, whereas flux imbalance directly quantifies the predominance of one magnetic polarity over the other. These quantities therefore define key magnetic characteristics of coronal holes, which are generally associated with predominantly open and unipolar magnetic-field regions.

In our previous work [19], we partitioned the coronal hole into smaller subregions and investigated the degree of magnetic unipolarity in each region. Most of the analyzed subregions showed a clear dominance of one magnetic polarity. Figure 3c compares the BlosB_{\mathrm{los}} distributions measured by HMI and ViSP. Both datasets show positive skewness. The HMI map has a skewness of 0.50 and a magnetic flux imbalance of 69%, while the ViSP map has a skewness of 0.46 and a magnetic flux imbalance of 65%. These similar skewness values indicate comparable asymmetry in the two BlosB_{\mathrm{los}} distributions, while the flux imbalance values indicate strong positive-polarity dominance in both datasets. The positive-to-negative flux ratios are likewise similar, with values of 5.4 for HMI and 4.7 for ViSP, confirming strong positive-polarity dominance in both measurements. The somewhat lower flux imbalance and flux ratio in ViSP indicate a relatively larger contribution from negative-polarity flux than in HMI. The positive flux measured by ViSP is 1.4 times that measured by HMI, while the corresponding factor for negative flux is 1.6. Thus, ViSP measures 40% more positive flux and 60% more negative flux than HMI, indicating a substantial increase in detected magnetic flux despite the two instruments exhibiting similar overall polarity dominance. The larger relative increase in negative flux is consistent with ViSP detecting additional small-scale negative-polarity flux, leading to a modest reduction in the flux imbalance while preserving the region’s overall positive-polarity dominance. However, high-resolution observations covering the full coronal-hole field of view are needed to confirm whether this behavior holds across the entire coronal hole.

Figure 3. Panels ‘a’ and ‘b’ show co-aligned images of the same photospheric field of view observed by HMI and ViSP, respectively. This region corresponds to the area marked by the dotted rectangle in Figure 1. Panel ‘c’ shows histograms comparing the BlosB_{\mathrm{los}} distributions in panels ‘a’ and ‘b’ and the vertical dashed red line indicates Blos=0B_{\mathrm{los}} = 0 G.

Missing Flux

Using the definition from equation (2), we find that the unsigned magnetic flux measured by ViSP is ΦuViSP=1.656×1020\Phi_{\mathrm{u}}^{\mathrm{ViSP}} = 1.656 \times10^{20} Mx, whereas the corresponding HMI value is ΦuHMI=1.189×1020\Phi_{\mathrm{u}}^{\mathrm{HMI}} = 1.189 \times10^{20} Mx.

The percentage difference in unsigned magnetic flux relative to ViSP is quantified as

Φm,unsigned=(1−ΦuHMIΦuViSP)×100%.(6)\Phi_{\mathrm{m,unsigned}} = \left(1 - \frac{\Phi_{\mathrm{u}}^{\mathrm{HMI}}}{\Phi_{\mathrm{u}}^{\mathrm{ViSP}}}\right) \times100\%. \tag*{(6)}

The difference in unsigned magnetic flux between the two measurements is ΔΦu=4.670×1019\Delta\Phi_{\mathrm{u}} = 4.670 \times10^{19} Mx, indicating that HMI measures approximately 28% less unsigned magnetic flux than ViSP. This difference includes contributions of 22% from positive polarity and 6% from negative polarity, relative to the total unsigned magnetic flux measured by ViSP. Thus, HMI measures less flux from both polarities, with the larger absolute difference associated with the dominant positive polarity.

Consistent with the larger unsigned flux measured by ViSP, the mean absolute line-of-sight magnetic field, ⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle, is 9.7 G for ViSP and 7.1 G for HMI. Thus, the ViSP ⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle is larger than the HMI ⟨∣Blos∣⟩\langle|B_{\mathrm{los}}|\rangle by a factor of 1.37.

Using the definition from equation (3), we find that the total signed magnetic flux measured by ViSP is ΦsViSP=1.073×1020\Phi_s^{\mathrm{ViSP}} = 1.073 \times10^{20} Mx, whereas the corresponding HMI value is ΦsHMI=8.162×1019\Phi_s^{\mathrm{HMI}} = 8.162 \times10^{19} Mx. The percentage difference in signed magnetic flux magnitude relative to ViSP is

quantified as

Φm,signed=(1−∣ΦsHMI∣∣ΦsViSP∣)×100%.(7)\Phi_{\mathrm{m,signed}} = \left(1-\frac{|\Phi_s^{\mathrm{HMI}}|}{|\Phi_s^{\mathrm{ViSP}}|}\right)\times100\%. \tag*{(7)}

The difference in the signed-flux magnitude between the two measurements is ΔΦs=2.568×1019\Delta\Phi_s = 2.568 \times10^{19} Mx, indicating that the magnitude of the signed magnetic flux measured by HMI is approximately 24% lower than that measured by ViSP. Consistent with the larger signed flux measured by ViSP, the absolute mean line-of-sight magnetic field, ∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle|, is 6.3 G for ViSP and 4.9 G for HMI. Thus, the ViSP ∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle| is larger than the HMI ∣⟨Blos⟩∣|\langle B_{\mathrm{los}}\rangle| by a factor of 1.30.

Effect of Spatial Resolution on the ViSP–HMI Comparison

Figure 4a shows the degraded ViSP BlosB_{\mathrm{los}} map derived from panel ‘b’ of Figure 3, which presents the original ViSP BlosB_{\mathrm{los}} map. The original ViSP BlosB_{\mathrm{los}} map was convolved with a Gaussian kernel chosen to approximate the HMI spatial resolution and was then resampled onto the native HMI pixel grid. The resulting degraded ViSP BlosB_{\mathrm{los}} map is compared with the original HMI BlosB_{\mathrm{los}} map in Figure 4b. These two maps show broadly similar spatial distributions.

Figure 4. Comparison of BlosB_{\mathrm{los}} measurements from ViSP and HMI. Panel ‘a’ shows the ViSP BlosB_{\mathrm{los}} map degraded to the HMI spatial resolution and pixel grid, while panel ‘b’ shows the original HMI BlosB_{\mathrm{los}} map. Panel ‘c’ compares the BlosB_{\mathrm{los}} distributions from the original ViSP BlosB_{\mathrm{los}} map (dotted blue curve), the degraded ViSP BlosB_{\mathrm{los}} map (solid blue curve), and HMI BlosB_{\mathrm{los}} (green curve).

In Figure 4c, the dotted blue curve represents the original ViSP BlosB_{\mathrm{los}} distribution, the solid blue curve represents the degraded ViSP BlosB_{\mathrm{los}} distribution, and the green curve represents the HMI BlosB_{\mathrm{los}} distribution. The degraded ViSP distribution is narrower than the original ViSP distribution and is more similar to the HMI distribution near the central peak and along the negative-field tail, while noticeable differences remain in the positive-field tail.

To assess quantitatively how closely the degraded ViSP and HMI BlosB_{\mathrm{los}} distributions agree overall, we calculated the mean, median, standard deviation, and skewness. Table 2 summarizes these statistical measures for the degraded ViSP and HMI BlosB_{\mathrm{los}} distributions. The degraded ViSP and HMI BlosB_{\mathrm{los}} distributions have mean values of 6.3 and 4.9 G, median values of 1.7 and 1.3 G, standard deviations of 22 and 21 G, and skewness values of 0.63 and 0.50, respectively. These statistical measures indicate broadly similar distributions, although some differences remain, consistent with the discrepancies visible in their histograms.

StatisticDegraded ViSP MEHMI
Mean [G]6.34.9
Median [G]1.71.3
Standard deviation [G]2221
Skewness0.630.50

Table 2. Statistical comparison of the degraded ViSP BlosB_{\mathrm{los}} and HMI BlosB_{\mathrm{los}} distributions.

The close similarity is expected because both datasets sample the same solar region, and degrading the ViSP data reduces the fine-scale structure and brings its effective spatial resolution closer to that of HMI. However, exact overlap is not expected because the two instruments differ in their effective spatial resolution, instrumental profiles, and observing techniques. The ground-based ViSP observations may also be affected by atmospheric seeing. In addition, HMI is an imaging filtergraph, whereas ViSP is a slit-based spectropolarimeter. These instrumental differences may contribute to the remaining discrepancies between the two BlosB_{\mathrm{los}} distributions, particularly in the positive-field tail.

In contrast, the broader distribution of the original ViSP BlosB_{\mathrm{los}} demonstrates that ViSP detects small-scale magnetic features that are partially averaged out in the HMI observations.

Conclusions

Over the past few decades, systematic comparisons of BlosB_{\mathrm{los}} measurements have shown that spatial resolution strongly affects inferred photospheric magnetic fields. Earlier Solar and Heliospheric Observatory (SOHO)/MDI, Advanced Stokes Polarimeter (ASP), SDO/HMI, and multi-observatory comparisons showed that weak-field measurements are especially sensitive to instrumental resolution and unresolved mixed-polarity flux cancellation [3, 25, 12, 35]. High-resolution SUNRISE/IMaX observations further demonstrated that small-scale photospheric magnetic features can be resolved at sub-arcsecond spatial scales [23]. More recently, Nóbrega-Siverio et al. [27] compared SDO/HMI magnetograms with high-resolution observations obtained using the CRisp Imaging SpectroPolarimeter (CRISP) at the Swedish 1-m Solar Telescope (SST), demonstrating that small-scale magnetic flux emergence episodes detected by SST/CRISP can remain indiscernible in HMI observations, leading to an underestimation of the associated magnetic flux. The distinctive aspect of this work is the use of ViSP spectropolarimetric observations, which provide among the highest-resolution measurements of the photospheric magnetic field currently available, to examine a selected field of view within a coronal hole. This high-resolution view allows us to investigate weak, fine-scale magnetic flux that may be spatially averaged or unresolved in HMI observations.

We analyze the BlosB_{\mathrm{los}} derived from ViSP spectropolarimetric observations of the Fe I 6302 Å line using the ME inversion. We then compare the ViSP ME-derived BlosB_{\mathrm{los}} map with the corresponding HMI Fe I 6173 Å vector magnetic-field measurement over the common field of view to examine how much small-scale photospheric magnetic flux is resolved by ViSP but remains unresolved in HMI observations. Since both the ViSP and HMI vector magnetic-field maps are obtained through inversion procedures based on ME atmospheric assumptions, this comparison provides a consistent framework for evaluating spatial-resolution effects on the inferred photospheric magnetic field. The main conclusions of this study are summarized below:

  • ViSP detects stronger and smaller scale BlosB_{\mathrm{los}} features than HMI within a common coronal hole region. This leads to a larger ⟨∣Blos∣⟩\langle|B_{\mathrm{los}}| \rangle in ViSP, indicating that HMI systematically underestimates the photospheric magnetic-field strength due to the spatial dilution of small-scale features within larger pixels.

  • ViSP measures approximately 1.4 times the total unsigned magnetic flux measured by HMI. This corresponds to HMI measuring approximately 28% less unsigned magnetic flux than ViSP within the common field of view.

  • HMI measures approximately 24% less net signed magnetic flux than ViSP, indicating a difference in the measured dominant-polarity magnetic flux within the observed coronal-hole region.

  • Despite their different spatial resolutions, ViSP and HMI exhibit remarkably similar magnetic flux imbalance and skewness. The flux imbalance is 65% for ViSP and 69% for HMI, while the corresponding skewness values are 0.46 and 0.50, respectively.

  • The observed coronal-hole region remains strongly unipolar in both HMI and ViSP. ViSP reveals additional mixed-polarity magnetic elements at smaller spatial scales; however, these do not change the overall unipolar character of the region.

  • The ViSP magnetogram provides more detailed photospheric BlosB_{\mathrm{los}} information than the HMI magnetogram. This may improve photospheric boundary conditions for magnetic-field extrapolations. One caveat is the limited ViSP field of view so full coverage of the coronal-hole is needed to determine whether these results apply to the full coronal hole structure.

  • The degraded-ViSP BlosB_{\mathrm{los}} comparison with HMI BlosB_{\mathrm{los}} shows that the difference between the two measurements is influenced by finite spatial resolution and the associated spatial averaging. When the ViSP map is degraded to HMI-like resolution, its BlosB_{\mathrm{los}} distribution approaches the HMI distribution. Residual differences may also reflect differences in the spectral lines, instrumental characteristics, and magnetic-field retrieval methods used by ViSP and HMI.

Overall, the higher-resolution ViSP measurements recover additional small-scale magnetic flux while preserving the dominant polarity and overall unipolar character of the observed coronal-hole region. Such measurements can provide improved photospheric boundary conditions for magnetic-field extrapolations and for studying the connection between small-scale photospheric fields and the upper solar atmosphere. However, the limited ViSP field of view prevents us from determining the total open magnetic flux of the coronal hole; future observations with broader coronal-hole coverage will be required to do so.

Acknowledgments

This research is supported by the DKIST Ambassadors program, funded by the National Solar Observatory through NSF Cooperative Support Agreement AST-1400450. This work is directly related to DKIST observations, and we are grateful for the program’s financial support throughout this study. We sincerely thank the anonymous referee for their careful evaluation of our manuscript and for providing insightful and constructive comments and suggestions. Their thoughtful feedback has helped improve the clarity, scientific interpretation, and overall quality of this work.

Data and Code Availability

The DKIST data used in this study were obtained through the DKIST Data Center (https://dkist.data.nso.edu). The SDO/AIA and SDO/HMI data used in this study were obtained through the Virtual Solar Observatory (VSO; https://sdac.virtualsolar.org/cgi/search) and the Joint Science Operations Center (JSOC; https://jsoc.stanford.edu).

The three Python computational notebooks developed for this study are publicly available through Zenodo ((Katuwal 2025; katuwal et al. 2026b; katuwal & mcateer 2026)). These notebooks document the computational procedures used in the analysis and support the reproducibility of the results presented in this work.

Appendix

The leakage of an unwanted signal from one Stokes parameter into another during polarimetric measurements is known as crosstalk. Crosstalk in the Fe I 6302.5 Å spectra is removed following the method described by [38], where the leakage of Stokes II into QQ, UU, and VV is estimated from the continuum and subtracted from the corresponding Stokes profiles. Figure 5 illustrates an example of a single raster scan. This specific scan corresponds to raster step 270 and captures the region marked by the blue cross at spatial pixel 450 in both panels of Figure 2. The first three panels show the Stokes Q/IcQ/I_c, U/IcU/I_c, and V/IcV/I_c signals before the I→Q,U,VI \to Q,U,V crosstalk correction, while the next three panels show the corresponding signals after the correction. The horizontal axis represents the wavelength-pixel index, and the vertical axis represents the spatial pixels along the slit. The red dashed line (same blue cross in Figure 2) marks the reference spatial pixel along the slit used to examine the spectra in more detail. In this work, we focus on BlosB_{\mathrm{los}} rather than the full vector magnetic field; however, the complete Stokes information is retained for potential future analysis.

Stokes polarization signals before and after crosstalk removal

Figure 5. The panels show the Stokes polarization signals before and after removal of the intensity-to-polarization (I→Q,U,VI \to Q, U, V) crosstalk for one scan step of the ViSP raster observation. The left three panels show the original Stokes Q/IcQ/I_c, U/IcU/I_c, and V/IcV/I_c wavelength–slit maps before correction, while the right three panels display the corresponding maps after applying the crosstalk correction. The horizontal axis represents the wavelength-pixel index, and the vertical axis represents the spatial pixels along the slit. The red dashed line indicates the spatial location of the reference pixel used in the analysis.

To further examine the effect of the crosstalk correction, we extract the Stokes spectra at the reference spatial pixel marked by the red dashed line in Figure 5. The resulting spectra are shown in Figure 6. The blue curves represent the Stokes profiles before the I→Q,U,VI \to Q,U,V crosstalk correction, while the orange curves show the profiles after the correction. The Stokes II profile remains nearly unchanged, while the largest visible modification occurs in Stokes QQ. In comparison, only relatively small differences are seen in Stokes UU and VV. Since the present study focuses on BlosB_{\mathrm{los}}, the comparatively small change in Stokes VV, which is most directly sensitive to BlosB_{\mathrm{los}}, is particularly relevant. To improve the signal-to-noise ratio, we rebinned the spatial pixels and obtained approximately square pixels. The full crosstalk-corrected Stokes I,Q,U,VI,Q,U,V profiles are subsequently used in the Milne-Eddington inversion.

Representative Stokes spectra before and after crosstalk removal

Figure 6. Representative Stokes I/IcI/I_c, Q/IcQ/I_c, U/IcU/I_c, and V/IcV/I_c spectra at the reference spatial pixel along the slit indicated by the dashed line in Figure 5. The blue and orange curves show the profiles before and after removal of the intensity-to-polarization (I→Q,U,VI \to Q, U, V) crosstalk, respectively.

References

  1. [1]Altschuler, M. D., & Newkirk, Jr., G. 1969, SoPh, 9, 131, doi: 10.1007/BF00145734DOI
  2. [2]Baweja, U., Pant, V., & Arregui, I. 2024, ApJ, 963, 69, doi: 10.3847/1538-4357/ad1b57DOI
  3. [3]Berger, T. E., & Lites, B. W. 2003, SoPh, 213, 213, doi: 10.1023/A:1023953716633DOI
  4. [5]Couvidat, S., Schou, J., Hoeksema, J. T., et al. 2016, SoPh, 291, 1887, doi: 10.1007/s11207-016-0957-3DOI
  5. [6]Cranmer, S. 2002, in COSPAR Colloquia Series, Vol. 13, Multi-wavelength Observations of Coronal Structure and Dynamics, ed. P. C. Martens & D. P. Cauffman (Pergamon), 3–12, doi: https://doi.org/10.1016/S0964-2749(02)80003-8
  6. [7]de Wijn, A. G., Casini, R., Carlile, A., et al. 2022, SoPh, 297, 22, doi: 10.1007/s11207-022-01954-1DOI
  7. [8]Del Zanna, G., & Mason, H. E. 2018, Living Reviews in Solar Physics, 15, 5, doi: 10.1007/s41116-018-0015-3
  8. [9]Harvey, J. W., & Sheeley, N. R., J. 1979, SSRv, 23, 139, doi: 10.1007/BF00173808DOI
  9. [10]Harvey, J. W., Hill, F., Hubbard, R. P., et al. 1996, Science, 272, 1284, doi: 10.1126/science.272.5266.1284DOI
  10. [11]Harvey, K. L., Sheeley, Jr., N. R., & Harvey, J. W. 1982, SoPh, 79, 149, doi: 10.1007/BF00146979DOI
  11. [12]Hoeksema, J. T., Liu, Y., Hayashi, K., et al. 2014, SoPh, 289, 3483, doi: 10.1007/s11207-014-0516-8DOI
  12. [13]Hofmeister, S. J., Utz, D., Heinemann, S. G., Veronig, A., & Temmer, M. 2019, Astronomy & Astrophysics, 629, A22
  13. [14]Kahler, S. W., Davis, J. M., & Harvey, J. W. 1983, SoPh, 87, 47, doi: 10.1007/BF00151159DOI
  14. [15]Katuwal, K. 2023a, in American Astronomical Society Meeting Abstracts, Vol. 242, American Astronomical Society Meeting Abstracts #242, 214.04
  15. [19]Katuwal, K., & McAteer, R. J. 2026, The Astrophysical Journal, 999, 63
  16. [22]Krieger, A. S., Timothy, A. F., & Roelof, E. C. 1973, SoPh, 29, 505, doi: 10.1007/BF00150828DOI
  17. [23]Lagg, A., Solanki, S. K., Riethmüller, T. L., et al. 2010, ApJL, 723, L164, doi: 10.1088/2041-8205/723/2/L164DOI
  18. [24]Lemen, J. R., Title, A. M., Akin, D. J., et al. 2012, SoPh, 275, 17, doi: 10.1007/s11207-011-9776-8DOI
  19. [25]Liu, Y., Hoeksema, J. T., Scherrer, P. H., et al. 2012, SoPh, 279, 295, doi: 10.1007/s11207-012-9976-xDOI
  20. [26]McComas, D., Velli, M., Lewis, W., et al. 2007, Reviews of Geophysics, 45
  21. [27]Nóbrega-Siverio, D., Cabello, I., Bose, S., et al. 2024, A&A, 686, A218, doi: 10.1051/0004-6361/202348894
  22. [28]Nölke, J. D., Solanki, S. K., Hirzberger, J., et al. 2023, A&A, 678, A196, doi: 10.1051/0004-6361/202346040DOI
  23. [29]Nolte, J. T., Krieger, A. S., Timothy, A. F., et al. 1976, SoPh, 46, 303, doi: 10.1007/BF00149859DOI
  24. [30]Pesnell, W. D., Thompson, B. J., & Chamberlin, P. C. 2012, SoPh, 275, 3, doi: 10.1007/s11207-011-9841-3DOI
  25. [31]Pietarila Graham, J., Cameron, R., & Schüssler, M. 2010, ApJ, 714, 1606, doi: 10.1088/0004-637X/714/2/1606DOI
  26. [32]Reiss, M., Temmer, M., Rott, T., Hofmeister, S. J., & Veronig, A. M. 2014, Central European Astrophysical Bulletin, 38, 95, doi: 10.48550/arXiv.1408.2777
  27. [33]Reiss, M. A., Muglach, K., Mason, E., et al. 2024, ApJS, 271, 6, doi: 10.3847/1538-4365/ad1408DOI
  28. [34]Rempel, M. 2014, ApJ, 789, 132, doi: 10.1088/0004-637X/789/2/132DOI
  29. [35]Riley, P., Ben-Nun, M., Linker, J. A., et al. 2014, SoPh, 289, 769, doi: 10.1007/s11207-013-0353-1DOI
  30. [36]Rimmele, T. R., Warner, M., Keil, S. L., et al. 2020, SoPh, 295, 172, doi: 10.1007/s11207-020-01736-7DOI
  31. [37]Sainz Dalda, A. 2017, ApJ, 851, 111, doi: 10.3847/1538-4357/aa97e3DOI
  32. [38]Sanchez Almeida, J., & Lites, B. W. 1992, ApJ, 398, 359, doi: 10.1086/171861DOI
  33. [39]Schad, T. A., Kuhn, J. R., Fehlmann, A., et al. 2023, The Astrophysical Journal, 943, 59, doi: 10.3847/1538-4357/acabbd
  34. [40]Schad, T. A., Petrie, G., Kuhn, J., et al. 2024, Science Advances, 10, eadq1604, doi: 10.1126/sciadv.adq1604
  35. [41]Scherrer, P. H., Bogart, R. S., Bush, R. I., et al. 1995, SoPh, 162, 129, doi: 10.1007/BF00733429DOI
  36. [42]Scherrer, P. H., Schou, J., Bush, R. I., et al. 2012, SoPh, 275, 207, doi: 10.1007/s11207-011-9834-2DOI
  37. [43]Schonfeld, S. J., Henney, C. J., Jones, S. I., & Arge, C. N. 2022, ApJ, 932, 115, doi: 10.3847/1538-4357/ac6ba1DOI
  38. [44]Schou, J., Scherrer, P. H., Bush, R. I., et al. 2012, SoPh, 275, 229, doi: 10.1007/s11207-011-9842-2DOI
  39. [45]Schwenn, R. 2006, SSRv, 124, 51, doi: 10.1007/s11214-006-9099-5DOI
  40. [46]The SunPy Community, Barnes, W. T., Bobra, M. G., et al. 2020, The Astrophysical Journal, 890, 68, doi: 10.3847/1538-4357/ab4f7a
  41. [47]Vaiana, G. S., Krieger, A. S., Timothy, A. F., & Zombeck, M. 1976, Ap&SS, 39, 75, doi: 10.1007/BF00640511DOI
  42. [48]Wang, Y.-M. 2009, SSRv, 144, 383, doi: 10.1007/s11214-008-9434-0DOI
  43. [49]Wang, Y.-M., Hawley, S. H., & Sheeley, Jr., N. R. 1996, Science, 271, 464, doi: 10.1126/science.271.5248.464DOI
  44. [50]Wang, Y.-M., & Sheeley Jr, N. 1990, Astrophysical Journal, Part 1 (ISSN 0004-637X), vol. 355, June 1, 1990, p. 726-732. Research supported by the US Navy., 355, 726
  45. [51]Wiegelmann, T. 2008, Journal of Geophysical Research: Space Physics, 113, doi: https://doi.org/10.1029/2007JA012432
  46. [52]Wiegelmann, T., & Sakurai, T. 2021, Living Reviews in Solar Physics, 18, 1, doi: 10.1007/s41116-020-00027-4DOI
  47. [53]Yadav, R., Kazachenko, M. D., Afanasyev, A. N., Cauzzi, G., & Reardon, K. 2024, ApJL, 973, L10, doi: 10.3847/2041-8213/ad758dDOI
  48. [54]Yadav, R., Mathew, S. K., & Tiwary, A. R. 2017, SoPh, 292, 105, doi: 10.1007/s11207-017-1131-2DOI

Paper details

vlnc:W_5kehx7kg
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 2

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents