The Source Surface Height Through the Solar Cycle: A Path to Better Solar Wind Forecasts

From HelioWiki Home Page
Jump to navigation Jump to search


Nugget
Number: 531
1st Author: Sandeep KUMAR
2nd Author: Nandita SRIVASTAVA
Published: July 20, 2026
Next Nugget:
Previous Nugget: Filament Eruptions as seen in the Sun-as-a-star H-alpha Spectrum



Introduction

The potential field source surface (PFSS) model serves as the basis of many state-of-the-art space weather forecasting frameworks, as described in our Ref. [1]. Essentially PFSS takes the photospheric magnetic field, observable in detail by Zeeman effect spectroscopy, and uses it to extrapolate into the solar corona. This requires the assumption of just a single parameter: the radial distance of the "source surface", a fictitious sphere separating radial field of the solar wind, from the structured field of the photosphere. Mathematically any field induced by external sources can be represented by a fitting spherical harmonic functions at its boundaries. This is tractable but physically it may seem terribly wrong, since it ignores all coronal field sources (currents) within the volume itself. But it works surprisingly well, perhaps because the currents that we know to be present only make a minor perturbation of the basic potential field.

The nature of the source surface

The PFSS model contains only a single free parameter, the source surface height, at which magnetic field lines are assumed to be exactly radial. A value of 2.5 R for the source-surface height has traditionally been adopted in most studies. Optimizing the source-surface value significantly improves model forecasts of the solar-wind speed as observed at the L1 Lagrangian point, just upstream of Earth in the solar wind (Ref. [1]). This study employed two types ("HU STD" and "HU ZPC"] of ([https://en.wikipedia.org/wiki/Global_Oscillations_Network_Group GONG) synoptic magnetic field maps. The fidelity of these data was confirmed by comparing the extrapolated global magnetic field structures with the large-scale corona observed in the extended field of view of the PROBA2/SWAP images as shown in Figure 1.

Figure 1: PFSS-extrapolated magnetic field lines overlaid on the SWAP (top and bottom panel) image of 20 August 2017, 10:45 UT. The middle panel shows the SWAP mosaic without PFSS extrapolation. The top left panel shows PFSS extrapolation with HU STD map, and the bottom left panel with HU ZPC map (on 20 August 2017, 12:14 UT). The right column shows the zoomed-in version of the white rectangle on the left panels.

The solar cycle

How must the PFSS model evolve through the solar cycle, during which the solar wind obviously changes substantially? In Ref. [3] we carried out one of the most comprehensive and long-term studies of SS height optimisation for solar wind prediction at L1, analyzing synoptic magnetograms from both space-based (SDO/HMI) and ground-based (GONG) observatories, across nearly three solar cycles, SC23-SC25. This optimization used the PFSS parameters in the model of Ref. [2] to compare with the wind-speed observations. Our main finding is that the optimal SS height varies systematically with the solar cycle: higher SS heights (≥2.5 R) provide better solar wind speed predictions during solar minimum, whereas lower SS heights (<2.5 R) perform better during solar maximum, suggesting a relationship between SS height and SC phase over long timescales. We also found that the optimized SS height depends on the choice of magnetogram, whereas the pattern remains similar for a given choice of magnetograms. The HMI and GONG "ZPC" magnetograms perform similarly, and both provide better forecasting than GONG "STD" maps, as shown in Figure 2. The results provide strong evidence that solar-cycle-dependent optimization of the PFSS source surface height is a practical way to improve solar wind forecasting, which serves as the foundation for heliospheric models.

Figure 2: Top panel: the source-surface radial height, in black for the standard fixed value of 2.5 R and in blue and gold for two alternative optimizations. The time range covers the 100 Carrington rotations from January 2018 (solar minimum) to July 2025 (high activity). The lower panel shows Pearson correlation coefficients for predictions of the solar-wind speed.

Conclusions

Our adjusted PFSS radii clearly improve forecasts of the solar-wind speed. The study highlights an inherent limitation of the PFSS model, which assumes a spherical source surface. This simplifying assumption restricts further improvement in the correlation coefficient. We anticipate much greater improvements from future modeling frameworks outside that of spherical symmetry of the source surface.

References

[1] "On the role of source surface height and magnetograms in solar wind forecast accuracy"

[2] "On the role played by magnetic expansion factor in the prediction of solar wind speed"

[3] "Source Surface Height Optimization for Improved Solar Wind Velocity Forecasting across Solar Cycles 23, 24, and 25"