Ultra-fast simulations of the solar dipole and open flux

Ismo T\"ahtinen; Kalevi Mursula; Timo Asikainen

arxiv: 2604.11342 · v1 · submitted 2026-04-13 · 🌌 astro-ph.SR

Ultra-fast simulations of the solar dipole and open flux

Ismo T\"ahtinen , Timo Asikainen , Kalevi Mursula This is my paper

Pith reviewed 2026-05-10 16:01 UTC · model grok-4.3

classification 🌌 astro-ph.SR

keywords solar dipolesurface flux transportopen solar fluxpropagator matricesvector sum compressionfast simulationsolar magnetic fieldsolar cycle

0 comments

The pith

A compressed matrix method simulates the solar dipole up to 1000 times faster than surface flux transport models while producing equivalent results.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The dipole flux transport method evolves the solar dipole using precomputed compressed propagator matrices rather than simulating the full magnetic field evolution. It delivers results equivalent to traditional surface flux transport models but 80 to 1000 times faster, depending on the time resolution and source complexity. This matters for studying open solar flux because the dipole magnitude matches the open flux from potential field source surface models, enabling rapid exploration of many activity scenarios on ordinary computers.

Core claim

Propagator matrices are built by evolving basis vectors of a synoptic map in the surface flux transport model. These matrices are compressed to less than one ten-thousandth of their size using the vector sum method, allowing the dipole vector to be advanced in time through matrix multiplication. The resulting evolution matches full surface flux transport simulations while running far faster, and the dipole magnitude provides a close proxy for open solar flux from the potential field source surface model.

What carries the argument

Compressed propagator matrices from basis vector simulations that evolve the dipole vector representation by matrix multiplication.

Load-bearing premise

The vector sum compression accurately maintains the dipole evolution to within 1 percent for any combination of source regions and time resolutions without accumulating systematic errors over multiple cycles.

What would settle it

A multi-cycle simulation in which the DFT and full SFT dipole magnitudes differ by more than 1 percent in a manner that increases with simulation length or varies with source distribution.

Figures

Figures reproduced from arXiv: 2604.11342 by Ismo T\"ahtinen, Kalevi Mursula, Timo Asikainen.

**Figure 1.** Figure 1: Comparison of SFT and DFT simulations. Upper panel: Scatter plot showing the dipole vector magnitude from the DFT (y-axis) and from the SFT (x-axis) simulation. Lower panel: Relative (black) and absolute (right) error of the DFT dipole vector magnitude. Article number, page 3 of 5 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

read the original abstract

Context. Solar dipole captures important information about the large-scale solar magnetic field. The evolution of the solar magnetic field including the solar dipole can be simulated with a surface flux transport (SFT) model, but these simulations are more extensive than is necessary to produce the evolution of the dipole alone. Aims. We present a dipole flux transport (DFT), matrix method that combines the classic SFT model with dipole vector representation of the solar magnetic field, allowing significantly faster simulations of the solar dipole. Methods. By simulating the evolution of basis vectors of a synoptic map, we constructed propagator matrices that produce the time evolution of the solar magnetic field by means of matrix multiplication. The computational speedup is achieved by compressing the propagator matrices to very small fraction $(< 10^{-4}$) of their original size with a recent vector sum method. Results. Depending on time resolution, the DFT performs 100-1000 times faster than a 4-year SFT simulation of a single active region while producing equivalent results. For multiple source regions, daily propagation matrices are sufficient to produce results that agree within 1\% with the SFT simulation of solar cycle 24, while performing 80 times faster. If the evolution of individual active regions is needed, the DFT performs 50000 times faster than the SFT model. Conclusions. DFT makes solar dipole simulations extremely fast, making it possible to run thousands of simulations in a few minutes with a basic laptop setup. As the magnitude of the dipole vector closely matches with open solar flux (OSF) from the potential field source surface model, the DFT can be used to study the development of OSF in various scenarios extremely efficiently.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper gives a fast compressed-matrix method for solar dipole evolution that matches full SFT results in the tested cases but leaves long-term compression error growth unexamined.

read the letter

The main thing to know is that the authors built propagator matrices from basis-vector runs of the standard surface flux transport equations, then applied vector-sum compression to shrink them to under 10^{-4} of original size. This produces a dipole-only evolution method that runs 80-1000 times faster than full SFT while staying within 1% for solar cycle 24 with daily matrices and for single active regions. The dipole magnitude also tracks open flux from PFSS models in their checks. That is the practical advance here, and the direct numerical comparisons are what make the speedup claim credible rather than just theoretical. The work stays grounded in the existing SFT framework with no circular fitting or new free parameters beyond time resolution and compression threshold. They earn credit for showing the method reproduces the dipole component accurately enough in the reported runs and for noting the open-flux connection. The soft spot is the compression step itself. Reducing the matrices that aggressively works for the single-cycle and basic-region tests shown, but the paper does not quantify whether small truncation errors accumulate in the dipole vector sum over multiple cycles or for more varied source patterns. The stress-test concern about possible systematic drift is reasonable given what is presented; more explicit long-integration checks would close that gap. This is aimed at solar physicists and space-weather modelers who need to run thousands of dipole or open-flux scenarios quickly for ensembles. Readers doing parameter sweeps or heliospheric modeling will get the most use from it. The numerical backing and clear reformulation are enough to send it to a serious referee, even if the compression stability needs more attention in revision.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces a Dipole Flux Transport (DFT) method that derives propagator matrices from surface flux transport (SFT) basis-vector simulations and compresses them via a vector-sum technique to enable rapid computation of solar dipole evolution. It reports speedups of 100-1000 times versus 4-year SFT runs for single active regions (with equivalent results) and 80 times faster for solar cycle 24 using daily matrices (agreeing within 1%), while noting close matching of dipole magnitude to PFSS open flux.

Significance. If the accuracy of the compressed matrices holds without accumulating bias, the DFT approach would enable thousands of dipole and open-flux simulations in minutes on modest hardware, substantially expanding the scope of ensemble and parameter-space studies in solar magnetic field evolution.

major comments (2)

[Results] Results section (multiple source regions paragraph): the 1% agreement with full SFT for solar cycle 24 is stated for daily matrices, but no explicit test or bound is given for error accumulation or systematic drift in the dipole vector sum arising from vector-sum compression over multiple cycles or arbitrary source configurations.
[Methods] Methods section (propagator matrix construction and compression): the reduction to <10^{-4} of original size is load-bearing for the speedup claim, yet the choice of compression threshold and its quantitative effect on preserving the dipole components (beyond the single-cycle and single-region cases) is not accompanied by sensitivity metrics or error budgets.

minor comments (2)

[Abstract] Abstract: the phrase 'producing equivalent results' should specify the precise metric (e.g., dipole moment magnitude, vector difference, or open-flux proxy) used for the 1% agreement.
The manuscript would benefit from a short statement on the range of time resolutions tested and any observed trade-off between matrix update frequency and dipole accuracy.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments on our manuscript. We provide point-by-point responses to the major comments below and indicate where revisions will be made.

read point-by-point responses

Referee: [Results] Results section (multiple source regions paragraph): the 1% agreement with full SFT for solar cycle 24 is stated for daily matrices, but no explicit test or bound is given for error accumulation or systematic drift in the dipole vector sum arising from vector-sum compression over multiple cycles or arbitrary source configurations.

Authors: We agree that an explicit test for error accumulation over multiple cycles would strengthen the claims. Although the 1% agreement for a full solar cycle (cycle 24) with numerous active regions already provides evidence against significant systematic drift, we will add a new test simulating an extended period covering two solar cycles using the DFT method and compare the dipole evolution to a reference SFT run where feasible, or quantify the per-step error propagation analytically. This will include bounds on the dipole vector sum drift. revision: yes
Referee: [Methods] Methods section (propagator matrix construction and compression): the reduction to <10^{-4} of original size is load-bearing for the speedup claim, yet the choice of compression threshold and its quantitative effect on preserving the dipole components (beyond the single-cycle and single-region cases) is not accompanied by sensitivity metrics or error budgets.

Authors: The compression threshold was determined empirically to balance computational efficiency with accuracy in the dipole components for the tested scenarios. To address this, we will include additional sensitivity analysis in the Methods section, such as varying the compression threshold and reporting the resulting errors in the dipole magnitude and components for the solar cycle 24 simulation, along with an error budget. revision: yes

Circularity Check

0 steps flagged

No circularity detected in DFT derivation chain

full rationale

The DFT method is constructed by simulating basis-vector evolutions under the standard SFT equations to obtain propagator matrices, followed by an independent vector-sum compression step whose accuracy is validated through direct numerical comparisons against full SFT runs for single active regions and solar cycle 24. These comparisons constitute external checks rather than tautological re-derivations; the reported speedups (100-1000x, 80x, 50000x) arise from dimensionality reduction to the dipole vector and matrix compression, not from any fitted parameter or self-referential definition. No load-bearing self-citations, ansatzes smuggled via prior work, or renaming of known results appear in the core chain. The derivation remains self-contained as a linear-algebra reformulation of an established model with explicit approximation-error quantification.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The method rests on the standard surface flux transport equations plus a new compression step whose accuracy is asserted but not derived from first principles.

free parameters (2)

time resolution of propagation matrices
Daily versus coarser matrices are chosen to balance speed and 1% accuracy; the choice is tuned to the solar-cycle-24 test case.
compression threshold in vector-sum method
The factor <10^{-4} size reduction is presented as achieved but the exact tolerance parameter is not stated.

axioms (1)

domain assumption The solar magnetic field evolution can be accurately represented by linear propagation of a small set of basis vectors.
Invoked when constructing the propagator matrices from synoptic maps.

pith-pipeline@v0.9.0 · 5608 in / 1279 out tokens · 33005 ms · 2026-05-10T16:01:48.752629+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

Altschuler, M. D. & Newkirk, G. 1969, Sol. Phys., 9, 131 Bobra, M. G., Sun, X., Hoeksema, J. T., et al. 2014, Sol. Phys., 289, 3549 Cameron, R. & Schüssler, M. 2007, ApJ, 659, 801 Cameron, R. H., Jiang, J., & Schüssler, M. 2016, ApJ, 823, L22 DeV ore, C. R., Boris, J. P., & Sheeley, Jr., N. R. 1984, Sol. Phys., 92, 1 Iijima, H., Hotta, H., Imada, S., Kusa...

work page 1969
[2]

The explicit form of the governing equation is ∂Br ∂t =− 1 R⊙ sinθ ∂ ∂θ (sinθu θBr) −Ω(θ) ∂Br ∂ϕ + η R2 ⊙ sinθ ∂ ∂θ sinθ ∂Br ∂θ ! + η R2 ⊙ sin2 θ ∂2Br ∂ϕ2 +S

describes the evolution of the radial magnetic field subject to differential rotationΩ(θ), meridional flowu θ(θ), and turbulent diffusionη with the induction equation ∂Br ∂t +∇ h ·(u hBr)=η∇ 2 hBr +S,(A.1) whereu h describes the horizontal flow velocity (due toΩ(θ) andu θ(θ)) and the magnetic flux emergence is modeled with the source term S. The explicit ...

work page 1990

[1] [1]

Altschuler, M. D. & Newkirk, G. 1969, Sol. Phys., 9, 131 Bobra, M. G., Sun, X., Hoeksema, J. T., et al. 2014, Sol. Phys., 289, 3549 Cameron, R. & Schüssler, M. 2007, ApJ, 659, 801 Cameron, R. H., Jiang, J., & Schüssler, M. 2016, ApJ, 823, L22 DeV ore, C. R., Boris, J. P., & Sheeley, Jr., N. R. 1984, Sol. Phys., 92, 1 Iijima, H., Hotta, H., Imada, S., Kusa...

work page 1969

[2] [2]

The explicit form of the governing equation is ∂Br ∂t =− 1 R⊙ sinθ ∂ ∂θ (sinθu θBr) −Ω(θ) ∂Br ∂ϕ + η R2 ⊙ sinθ ∂ ∂θ sinθ ∂Br ∂θ ! + η R2 ⊙ sin2 θ ∂2Br ∂ϕ2 +S

describes the evolution of the radial magnetic field subject to differential rotationΩ(θ), meridional flowu θ(θ), and turbulent diffusionη with the induction equation ∂Br ∂t +∇ h ·(u hBr)=η∇ 2 hBr +S,(A.1) whereu h describes the horizontal flow velocity (due toΩ(θ) andu θ(θ)) and the magnetic flux emergence is modeled with the source term S. The explicit ...

work page 1990