Global MHD simulations of the solar convective zone using a volleyball mesh decomposition. I. Pilot
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.
Vera: We're looking at the paper: "Global MHD simulations of the solar convective zone using a volleyball mesh decomposition. I. Pilot".
Jocelyn: Here's the gist of the paper. Global MHD simulations of the solar convective zone using a volleyball mesh decomposition address outstanding bottlenecks in connecting internal convection zone and dynamo simulations to the surface of the Sun and conduct a proof-of-concept high resolution global simulation using the task-based DISPATCH code framework Simulation Framework and Mesh Decomposition The study employs a new ‘volleyball’ mesh decomposition, which consists of Cartesian patches tessellated on a sphere with no singularities <ref:2211.09564#pg5>.
Vera: This decomposition involves arranging Cartesian patches in experiment-specific geometries on a sphere, where each patch contains N cells (generally N = 24, plus 3 guard zone cells on each side) <ref:2211.09564#pg2>.
Jocelyn: The radial extent of the simulation domain is set from 0.655 to 0.995 R <ref:2211.09564#pg8>.
Vera: The mesh arrangement is generated by stepping angularly and then applying symmetry and flipping operations to cover the whole surface of a layer <ref:2211.09564#pg5>.
Title and authors: Vera: So we’re looking at this paper called "Global MHD simulations of the solar convective zone using a volleyball mesh decomposition. I. Pilot". It sounds like they're tackling that big problem of linking the deep inside convection zone stuff with what actually happens on the surface of the Sun, which is usually really hard to do because those scales and environments are so different.
Jocelyn: Yeah, and they’re using this specific method called a volleyball mesh decomposition. It sounds like they're trying to solve that connection issue by making the simulation setup much more flexible for handling those different physical layers.
Subrahmanyan: And the authors are using a task-based code framework called DISPATCH, which is interesting because it lets them mix and match different solvers and physics on the fly, which is often necessary when you're dealing with complex phenomena like this.
Vera: Exactly. So what they’re proposing here is that by using this specific mesh decomposition on a sphere, they can do a proof of concept high resolution global simulation of the solar convection zone. It’s about showing that it’s possible to get a realistic picture across all those scales without hitting those major geometric problems we usually run into.
Jocelyn: And what I found in the summary is that they're using this framework to test how they can actually connect internal convection zone and dynamo simulations to the surface of the Sun, which is a really important connection for understanding solar activity.
Subrahmanyan: It’s about testing if this setup works as a proof of concept for what’s possible in terms of high resolution global simulation, which is usually computationally demanding. I wonder how they're handling the computational cost with this approach.
Vera: Well, they are using a new 'volleyball' mesh decomposition where it consists of Cartesian patches tessellated on a sphere with no singularities at all, which is a big deal because standard spherical coordinates have those kinds of problems. Plus, each patch has twenty-four cells plus three guard zone cells on each side.
Jocelyn: That sounds like they’ve built a way to keep the geometry clean while still covering the whole surface of the layer they are simulating. So, it’s about having this specific geometric structure that avoids those nasty coordinate singularities.
Subrahmanyan: The radial extent they set for this simulation is from zero point six five five R to nine nine five R, which means they're getting a pretty good look at the entire convection zone radius without running into issues at the very center or the very edge.
Vera: And they generate this mesh arrangement by stepping angularly and then applying symmetry and flipping operations to cover the whole surface of a layer, which sounds like a clever way to tile the sphere properly. It’s about making sure every part of that volume is covered correctly.
Title and authors: Jocelyn: Then they use this task-based DISPATCH code framework, which gives them local time-steps for each patch and makes it solver agnostic so they can swap things out if needed. That flexibility is what lets them handle the different physical effects involved in the simulation without getting stuck with one specific math method.
Subrahmanyan: The framework also supports hybrid MPI and OpenMP parallelism with only nearest-neighbour MPI communications, which is designed to give them theoretically unlimited scaling on supercomputers, which is crucial for these kinds of global simulations.
Vera: And for the magnetohydrodynamics part, they’re using a new entropy based HLLD Riemann solver called HLLS. It’s supposed to give them the accuracy of a Godunov solver but it avoids that negative pressure problem that sometimes trips up other methods.
Jocelyn: That means they can model the thermal energy evolution in these turbulent flows more accurately because they aren't dealing with those numerical glitches caused by that negative pressure issue. It’s about getting better physics out of the math.
Subrahmanyan: The fluid equations are written as a system of conservation laws, and the magnetic field is evolved using a two-D Riemann method corresponding to the equation for how the magnetic field changes over time with velocity, which is an important detail for coupling things together.
Vera: They’ve included several key physical effects like gravity, Newton cooling in the top few cell layers to drive convection, Coriolis force, and differential rotation throughout the domain. These are all factors that really influence how the Sun behaves physically.
Jocelyn: For gravity they model it using a radially dependent g of r where fg is negative g of r times the unit vector in radial direction, which keeps that physics realistic as you move away from the center.
Subrahmanyan: Newton cooling is used in those top few cell layers to drive convection, and heating per unit mass is defined by a specific term involving epsilon zero psi minus delta epsilon s minus epsilon x comma y comma z times eta. That’s how they are modeling that energy source.
Vera: The initial magnetic field they set up is a uniform field aligned with the vertical axis in the global frame of reference, and it’s set to zero point one Gauss. This gives them a clean starting point for their simulation run-time.
Jocelyn: So, looking at the results, they found that these simulations develop convective motions with complex structures where small scale dynamo action amplifies magnetic field magnitudes by more than two orders of magnitude within the initial run-time. That’s a pretty big amplification effect they are seeing early on.
Subrahmanyan: The relaxation process shows that those convective motions appear around two hours of solar time, and by six hours convection is essentially disordered, which suggests a certain timescale for how quickly the system settles into its state. It’s about the dynamics evolving over time.
Title and authors: Vera: And they found that this new HLLS solver proves effective at catching numerical glitches because guard zone interpolations do not lead to noticeable numerical artefacts or artificial discontinuities in the results. That’s a win for their stability and accuracy.
Jocelyn: The framework also demonstrates excellent scaling properties for both weak and strong scaling tests on supercomputers, which means it’s computationally efficient enough for large-scale production runs when you need to run this kind of global simulation.
Subrahmanyan: I think the main implication here is that they’ve found a way to get resolution that can be varied with radius while avoiding coordinate singularities, which is a huge step for studying plasma dynamics across all those scales. This could significantly reduce simulation costs when combined with local time-stepping.
Vera: The structural design of the volleyball setup allows them to explore with zoom in type of simulations for studying specific features in space and or time, and it can be applied to any problem with spherical geometry without needing to handle coordinate singularities and always having one axis pointing in the vertical direction.
Jocelyn: They also mentioned that future work involves adding a photosphere, replacing Newton cooling with full radiative heat transport, and including the convective stable central part of the Sun for global helioseismological studies. That shows they have a clear path forward to make this simulation even more realistic.
Subrahmanyan: It sounds like they are setting up a solid foundation here that lets them explore specific features in space and time using zoom in techniques, which is very useful when you want to study things like active regions or flux emergence with sufficient resolution for detailed analysis.
Vera: So, to wrap up on this paper, the main thing they’ve shown is that this method provides a way to simulate the solar convective zone with a mesh decomposition that lets them control resolution by radius without hitting coordinate singularities.
Jocelyn: And it proves how flexible the DISPATCH framework is, showing you can build up experiments by manipulating the mesh decomposition, adding layers, refining things, or zooming in.
Subrahmanyan: Ultimately, they’re showing that this volleyball setup can be applied to any problem with spherical geometry without needing to handle those coordinate singularities and always having one axis pointing in the vertical direction.
Vera: This technique could significantly reduce simulation costs when combined with local time-stepping, which is a key part of the whole framework. It’s about making high resolution simulations more accessible.
Jocelyn: And it really lays out how to connect those internal and surface simulations in a way that's feasible right now. So, they’ve given us a solid proof of concept for this kind of global simulation.
The paper's summary: Vera: So, this paper is basically showing how they’re using a new way to tile space on a sphere to simulate the solar convection zone globally, which is usually super tricky because of all those geometric headaches we have with spherical coordinates.
Jocelyn: Right, and they call it the volleyball mesh decomposition. It’s about creating these Cartesian patches that cover the whole surface without any singularities popping up, which is a big deal for making sure the simulation stays physically sound across different scales.
Vera: Yeah, and they are using this framework with their DISPATCH code to do a proof of concept run of a high resolution global simulation. It’s not just about running one thing; it’s about proving the setup works for connecting that deep convection zone stuff to what we see on the surface.
Jocelyn: They're testing how they can link those internal dynamo simulations with surface observations, which is exactly what we need to really understand solar activity. It’s a way to bridge that gap between the inside and outside of the Sun.
Vera: The key thing they found in this pilot run is that they got these complex convective motions developing, and their small scale dynamo action—that’s the magnetic field getting amplified on small scales—it boosted those magnetic field magnitudes by more than two orders of magnitude really fast.
Jocelyn: That massive early amplification is interesting because it shows how quickly things can change in this turbulent environment, and they noted that convection starts appearing around two hours of solar time before it gets quite disordered by six hours.
Vera: They also confirmed that their new solver, the entropy based HLLD Riemann solver, is actually working well; the guard zone interpolations they use don't cause those weird numerical glitches or artificial breaks in the data.
Jocelyn: And from a computational standpoint, they showed that this whole volleyball setup and framework scales really well on supercomputers for both weak and strong scaling tests, which means it’s efficient enough for big production runs.
Vera: So what this means is that we have a new geometric tool to simulate solar physics without running into those coordinate problems, and the flexibility of the DISPATCH framework lets them build up the simulation layer by layer.
Jocelyn: They are still planning some next steps, like adding a real photosphere and using full radiative heat transport instead of just Newton cooling. It shows they’re serious about pushing this toward even more realistic models.
Vera: And they mentioned zoom in techniques for studying active regions or flux emergence without having to impose artificial initial conditions, which is super useful for targeted research.
Jocelyn: So this pilot run gives us a solid foundation to see if we can actually build up these global simulations systematically, and it sets the stage for looking at more specific features in space and time.
The paper's improvements: Vera: So, we’re looking at how they are suggesting ways to make this simulation even better, moving beyond just this pilot run setup. It’s about taking what they did and making it more robust for future work.
Jocelyn: Right, and they are talking about adding a photosphere to the model later on. That means moving from just simulating the convection zone interior to actually modeling what happens right at the surface of the Sun where we observe things directly.
Vera: They’re also looking at replacing Newton cooling with full radiative heat transport, which is a much more complete way to describe how energy moves through that layer than just using that simplified cooling term they used in this run.
Jocelyn: That would give us a more accurate picture of the thermal dynamics inside the convection zone, which is key because thermal structure drives all those convective motions we saw earlier.
Vera: And they’re considering including the central part of the Sun that is stable for convection, which would be really important for doing detailed helioseismological studies later on.
Jocelyn: Plus, they are talking about zoom in techniques to focus on specific active regions or flux emergence without having to set up those artificial boundary conditions from scratch. That lets researchers target specific events without the simulation getting bogged down in setting up the whole Sun first.
Vera: And they’re suggesting using this volleyball setup for other problems with spherical geometry, like simulating planet growth or even supernova explosions, which shows how versatile this mesh decomposition is.
Jocelyn: So it’s not just about the Sun anymore; it’s a general method for any problem that has a spherical shape, which is a really useful piece of software engineering here.
Vera: The authors also highlight that they can use this setup to explore specific features in space and time using those zoom in methods, which is how you get high resolution for things like magnetic field structures.
Jocelyn: And they emphasize that the structural design of the volleyball setup lets you always have one axis pointing vertically, so it simplifies things when you’re thinking about solar geometry.
Vera: Overall, these improvements show a clear path to making this a more comprehensive tool for studying solar physics, moving from this initial proof of concept toward a full physical model.
Jocelyn: It shows they are thinking about how to actually use the data and the simulation results in more advanced ways down the line.
Conclusion: Tom: So, to wrap up this paper titled "Global MHD simulations of the solar convective zone using a volleyball mesh decomposition. I. Pilot," they’ve shown how this specific way of tiling space on a sphere can handle complex global magnetohydrodynamics without running into those coordinate singularities we always worry about.
Vera: Yeah, they proved that you can vary the resolution across different radii while keeping the math clean, which is a big win for getting a realistic picture of the convection zone.
Jocelyn: It really demonstrates how flexible that DISPATCH framework is; you can build up experiments by just changing how you arrange those patches or adding more layers as needed.
Tom: And they confirmed that this setup works well for scaling on supercomputers, which means we can trust it for large, computationally heavy production runs.
Vera: What this means is that we’ve got a new geometric tool to simulate solar physics without hitting those coordinate problems and it lets us explore specific features like active regions with enough resolution for detailed analysis.
Jocelyn: They also showed how this volleyball setup works for any problem with spherical geometry, which is super handy because it always has one axis pointing vertically, simplifying things when we’re thinking about solar structure.
Tom: So this technique could really help cut down simulation costs when you pair it with local time-stepping, which is a crucial part of the whole framework.
Vera: It’s about making high resolution simulations more accessible to more researchers who want to tackle these problems.
Jocelyn: They also pointed out that future work will involve adding a photosphere and using full radiative heat transport instead of just Newton cooling, which is where they’re headed next.
Tom: So this pilot run gives us a solid foundation for connecting internal dynamo simulations with surface observations in a way that’s feasible right now.
Vera: It really lays out how to build up these experiments systematically, and it sets the stage for looking at more specific features in space and time later on.
Jocelyn: We can see this as a major step forward for modeling solar dynamics across all those scales that we struggle to connect right now.
Tom: Subrahmanyan, what do you think about this geometric approach to solving global MHD problems?
Subrahmanyan: From my side, the ability to vary resolution with radius without coordinate singularities is significant because it directly addresses a major bottleneck in modeling the entire convection zone structure.
Vera: I agree, it’s a clean way to handle geometry that we just don't have easily right now.
Jocelyn: It shows that engineering solutions for these complex physics problems are becoming more sophisticated.
Rosseland Centre for Solar Physics, University of Oslo · Institute of Theoretical Astrophysics, University of Oslo · Niels Bohr Institute, University of Copenhagen
astro-ph.SR, physics.comp-ph
Submitted: 2022-11-17
Updated: 2026-11-09
Comments: 14 pages, 10 figures, 3 appendices (5 pages), submitted to A&A
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 79/100
The gist: Global MHD simulations of the solar convective zone using a volleyball mesh decomposition address outstanding bottlenecks in connecting internal convection zone and dynamo simulations to the surface…
Key concepts
- Volleyball Mesh Decomposition
- This is a new way to divide the solar convection zone into patches on a sphere. Instead of traditional grids that cause problems at certain points, this method uses Cartesian patches arranged on a sphere. This allows the simulation to cover the entire surface without creating mathematical singularities, enabling resolution changes across different regions.
- Task-based DISPATCH Code Framework
- This is a flexible software structure that manages the simulation. It allows researchers to easily select and combine different numerical solvers and physical models, such as non-ideal MHD or radiative heat transfer. This modularity lets scientists gradually add complex physics to study phenomena like local dynamo effects.
- Entropy-based HLLD Riemann Solver (HLLS)
- This is a specific mathematical tool used to solve the fluid dynamics equations in the simulation. It is designed to be accurate like a Godunov solver but avoids numerical issues, specifically preventing the 'negative pressure problem.' This ensures that the simulation correctly models how thermal energy evolves within turbulent flows.
- Small-Scale Dynamo Action
- This refers to a process where small magnetic field fluctuations within the solar convection zone grow significantly. The simulations showed this action amplified the magnetic field strength by more than two orders of magnitude early in the run, indicating a powerful mechanism for generating and sustaining solar magnetism.
Terminology
Summary
Global MHD simulations of the solar convective zone using a volleyball mesh decomposition address outstanding bottlenecks in connecting internal convection zone and dynamo simulations to the surface of the Sun and conduct a proof-of-concept high resolution global simulation using the task-based DISPATCH code framework
Simulation Framework and Mesh Decomposition
The study employs a new ‘volleyball’ mesh decomposition, which consists of Cartesian patches tessellated on a sphere with no singularities <ref:2211.09564#pg5>. This decomposition involves arranging Cartesian patches in experiment-specific geometries on a sphere, where each patch contains N cells (generally N = 24, plus 3 guard zone cells on each side) <ref:2211. p. 3 Domain decomposition
In DISPATCH, we use a set of Cartesian patches arranged
in experiment-specific geometries. Each patch contains N
3
cells (in the current work, we generally use N = 24, plus 3
guard zone cells on each side, thus 303
cells in total. Experimental setup
We use the ‘volleyball’ domain decomposition described below to conduct the experiment. The radial extent of the
simulation domain is from 0. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. (11)
This stepping, interleaved with steps northward,
θ = θold + ∆θ, (12)
ends up covering a quarter of a face (1/24 or the entire
surface). The rest of the surface is covered by executing a
sequence of symmetry and flipping operations—we spawn
the identical points on the other 6 volleyball faces by simply applying a symmetry operation, which involves three
permutation operations (going from front to right, to top,
returning to the front) and two reflection operations (going from front to back).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Numerical Methods and Solver Selection
The simulation utilizes the task-based DISPATCH code framework, which features local time-steps for each patch, solver agnosticism, hybrid MPI/OpenMP parallelism with nearest-neighbour MPI communications, and support for curvilinear meshes <ref:2211.09564#pg3>. For magnetohydrodynamics (MHD), the work uses a new entropy based HLLD Riemann solver (HLLS) which provides the accuracy of a Godunov solver while avoiding the negative pressure problem <ref:2211. p. 1 2022),
which do not necessarily resolve the problems, but at least
make them less visible. In this work, we use a new, entropy-based HLLD Riemann solver (HLLS; Popovas 2022), which gives the accuracy and performance of a Godunov solver while avoiding
the negative pressure problem and providing correct convective flow velocities. These procedures are explained in detail in
Popovas (2022)—here we only briefly summarise the partial
differential equations the solver employs. Partial Differential Equations
The fluid equations for the total mass density, momentum
density, entropy per unit mass and the magnetic field of the
system can be written as a system of conservation laws,
∂
∂t
ρ
ρu
ρS
B
+∇·
ρu
ρ(u ⊗ u) + Ptot − B ⊗ B
ρuS
B ⊗ u − u ⊗ B
=
0
Φ
Q/Tgas + Sgen
0
,
(1)
with
∇ · B = 0, (2)
where ρ, u, S , Ptot, and B are density, velocity, entropy per
unit mass, total pressure and magnetic field respectively;
Φ is force per unit volume, which includes force of gravity,
Coriolis force, etc; Q is heating per unit volume, which includes Newton cooling, radiative heat transfer, etc; Tgas is
gas temperature and Sgen is entropy generation from converting kinetic energy to heat.Where the paper says it
Where the paper says it
Physical Modeling and Initial Conditions
The simulation incorporates several key physical effects, including gravity, Newton cooling, Coriolis force, and differential rotation <ref:2211.09564#pg2>. The force of gravity is modeled using a radially dependent g(r), where fg = −g(r) ∗ rˆ <ref:2211. p. 6 Additional micro- and macro-physics
effects can easily be added later. Force of gravity
For smaller 3D Cartesian simulations, it is common to
have a globally constant gravitational acceleration g, acting along one of the axes, but since our simulation extends
through the whole convective zone we must have a more
realistic, radially dependent, g(r). We interpolate thermal energy per unit mass,
ε0, from our hydrostatic equilibrium data and obtain its
slope, ∆ε. Heating per unit mass is then
QN = [(ε0ψ − ∆εs) − εx,y,z]η, (22)
where ψ is a parameter that controls the strength of the
Newton cooling; s = p − r, with p being the position of a
patch’s centre in a global frame of reference; η = ξ/(1+ξ)τ0,
is an exponential decay function, where ξ = e
−s and τ0 is
the cooling timescale, which we set to 200 s. Thus the magnetic field can be initialised
either as a thorus, or as a uniform field in the global frame
of reference. We initialise a uniform magnetic field, aligned
with the vertical axis in the global frame of reference, i. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Results and Validation
The simulations develop convective motions with complex structures where small-scale dynamo action amplifies magnetic field magnitudes by more than two orders of magnitude within the initial run-time <ref:2211.09564#pg4>. The relaxation process shows that convective motions appear around 2 hours of solar time, and by 6 hours convection is essentially disordered <ref:2211. p. 7 However, this repetition is very quickly broken
down by numerical rounding errors, and at around 6 hours
of solar time the convection pattern is essentially disordered. Figure 5 shows entropy per unit mass in the experiment. Using our new HLLS solver provides us with a unique
bug-catching tool. Guard zone interpolation glitches create
artificial discontinuities. Weak and strong scaling
The DISPATCH framework has demonstrated excellent
scaling results past 150,000 cores on a fluid dynamics
problem (Nordlund et al. p. 1 Incompressible or anelastic models are not applicable
close to the photosphere of stars, Cartesian boxes are not
very suitable for maintaining a spherical hydrostatic equilibrium, and spherical coordinates have singularities at the
poles, to mention just a few. To address outstanding bottlenecks and connect internal
convection zone and dynamo simulations to the surface of
the Sun, we employ the task-based DISPATCH code framework (Nordlund et al. Incompressible or anelastic models are not applicable
close to the photosphere of stars, Cartesian boxes are not
very suitable for maintaining a spherical hydrostatic equilibrium, and spherical coordinates have singularities at the
poles, to mention just a few. To address outstanding bottlenecks and connect internal
convection zone and dynamo simulations to the surface of
the Sun, we employ the task-based DISPATCH code framework (Nordlund et al.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Future Directions
Future work involves adding a photosphere, replacing Newton cooling with full radiative heat transport, and including the convective stable central part of the Sun for global helioseismological studies <ref:2211.09564#pg3>. Zoom-in techniques will be used to focus on individual active regions without imposing artificial initial and boundary conditions <ref:2211.09564#pg2>. The volleyball setup is also applicable to problems with spherical geometry, such as simulating planet growth or supernova explosions <ref:2211.09564#pg2>. The structural design facilitates exploration with zoom-in type of simulations for studying specific features in space and/or time <ref:2211. p. 0 ) but at the cost of
having to neglect the surface physics and the existence of
sunspots. To resolve this disconnect many bottlenecks must be addressed: Highly disparate spatial and temporal scales, complex microphysics, the interplay between local and global
effects, very wide dynamic range, and required resolution
in both space and time to resolve various features at different depths of the atmosphere.Where the paper says it
Conclusion
The paper describes a method to simulate the solar convective zone with a new ‘volleyball’ mesh decomposition that provides resolution that can be varied with radius while avoiding coordinate singularities <ref:2211.09564#pg2>. The proof of concept runs demonstrate how the DISPATCH framework’s flexibility allows for building up experiments by manipulating the mesh decomposition, adding layers, refining, or zooming in <ref:2211. p. 0 We address the outstanding bottlenecks in connecting internal convection zone and dynamo simulations to the
surface of the Sun, and conduct a proof-of-concept high resolution global simulation of the convection zone of the Sun,
using the task-based DISPATCH code framework. We present a new ‘volleyball’ mesh decomposition, which has Cartesian patches tessellated on a sphere with
no singularities. (17)
Since the transformation from source to target can also be
achieved by first translating the source coordinates to global
coordinates and then go from there to target coordinates,
it follows that the relative transformation can be obtained
as the product of the transposed target rotation matrix and
the source rotation matrix:
R = R
T
target
· R
source
(18)
3. Guard zone interpolations
DISPATCH utilises local time-stepping for individual
patches, thus each patch needs guard zone data from neighbour (nbor) patches interpolated in both space and time.Where the paper says it
Where the paper says it
References
Bezanson, J., Edelman, A., Karpinski, S., & Shah, V. B. 2014, arXiv e-prints, arXiv:1411. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Brchnelova, M. p. 7 This is very encouraging, as smooth transitions
between interfaces are rather challenging (e. Brchnelova
et al. This is very encouraging, as smooth transitions
between interfaces are rather challenging (e. Brchnelova
et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Carlsson, Mats et al. p. 0 For
instance, surface models, however complex they have become (e. Carlsson, Mats et al. For
instance, surface models, however complex they have become (e. Carlsson, Mats et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Chen, F., Rempel, M., & Fan, Y. p. 2 Indeed,
accounting for this entropy increase is the defining feature
of this new solver. The generated entropy S gen depends on
the state of the Riemann fan and is characterised by the
change of density, normal velocity and total pressure, when
crossing a given discontinuity. Indeed,
accounting for this entropy increase is the defining feature
of this new solver. The generated entropy S gen depends on
the state of the Riemann fan and is characterised by the
change of density, normal velocity and total pressure, when
crossing a given discontinuity. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Christensen-Dalsgaard et al. p. 6 This small perturbation speeds
up the development of convective motions. Hydrostatic equilibrium
We start the experiment with a hydrostatic equilibrium,
iterated using our EOS tables together with the 1D ModelS from Christensen-Dalsgaard et al. This small perturbation speeds
up the development of convective motions. Hydrostatic equilibrium
We start the experiment with a hydrostatic equilibrium,
iterated using our EOS tables together with the 1D ModelS from Christensen-Dalsgaard et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Druett et al. p. 6 2016;
Kohutova & Popovas 2021; Finley et al. p. 6 2022; Carlsson, Mats et al. 2016;
Kohutova & Popovas 2021; Finley et al. 2016;
Kohutova & Popovas 2021; Finley et al. p. 6 2022; Carlsson, Mats et al. 2016;
Kohutova & Popovas 2021; Finley et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Evans & Hawley 1988, ApJ, 332, 659 <ref:2211. p. 1 Maintaining an accurate hydrostatic
equilibrium is also more manageable with entropy as the
main energy variable. We use the MUSCL-Hancock algorithm with constrained transport (CT; Evans & Hawley 1988) for the induction equation, as well as a positivity preserving 3D unsplit TVD slope limiter; see Fromang et al. p. 1 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al. 2006; Fromang et al. 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al. 2006; Fromang et al. Maintaining an accurate hydrostatic
equilibrium is also more manageable with entropy as the
main energy variable. We use the MUSCL-Hancock algorithm with constrained transport (CT; Evans & Hawley 1988) for the induction equation, as well as a positivity preserving 3D unsplit TVD slope limiter; see Fromang et al.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Finley et al. 2022, A&A, 665, A118 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Fromang et al. 2006, A&A, 457, 371 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Gallice et al. p. 1 Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al. Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Gudiksen et al. p. 1 Basic equations and set-up
The DISPATCH framework can use several different solvers
for magnetohydrodynamics (MHD); an entropy based
STAGGER solver, used in Popovas et al. p. 0 Box 1029, Blindern, NO-0315 Oslo, Norway
e-mail: andrius. popovas@astro. Box 1029, Blindern, NO-0315 Oslo, Norway
e-mail: andrius. popovas@astro. Basic equations and set-up
The DISPATCH framework can use several different solvers
for magnetohydrodynamics (MHD); an entropy based
STAGGER solver, used in Popovas et al. (2018, 2019), an
internal energy based Stagger solver taken straight from the
Bifrost (Gudiksen et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Guerrero et al. p. 0 They used spherical geometry
and a Yin-Yang grid (Kageyama & Sato 2004) with the
sphere spanning 0. Guerrero et al. They used spherical geometry
and a Yin-Yang grid (Kageyama & Sato 2004) with the
sphere spanning 0. Guerrero et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Hotta & Kusano 2021, Nature Astronomy, 5, 1100 <ref:2211. p. 0 Tremendous efforts have
been made recently to resolve such disconnects for the Sun
and other stars. Hotta & Kusano (2021); Hotta et al. p. 0 Tremendous efforts have
been made recently to resolve such disconnects for the Sun
and other stars. Hotta & Kusano (2021); Hotta et al. Tremendous efforts have
been made recently to resolve such disconnects for the Sun
and other stars. Hotta & Kusano (2021); Hotta et al. Tremendous efforts have
been made recently to resolve such disconnects for the Sun
and other stars. Hotta & Kusano (2021); Hotta et al.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Hotta et al. 2022, ApJ, 933, 199 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Ismail & Roe 2009, Journal of Computational Physics, 228, 5410 <ref:2211. p. 1 Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al. Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al.Where the paper says it
Where the paper says it
Kageyama & Sato 2004, Geochemistry, Geophysics, Geosystems, 5 <ref:2211. p. 0 (2022)
used global simulations of the solar convection zone to study
solar differential rotation. They used spherical geometry
and a Yin-Yang grid (Kageyama & Sato 2004) with the
sphere spanning 0.Where the paper says it
Kapylä et al. 2021, A&A, 651, A66 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Kohutova & Popovas 2021, A&A, 647, A81 <ref:2211. p. 6 2022; Carlsson, Mats et al. 2016;
Kohutova & Popovas 2021; Finley et al. 2022; Carlsson, Mats et al. 2016;
Kohutova & Popovas 2021; Finley et al.Where the paper says it
Where the paper says it
Kuffmeier et al. 2017, ApJ, 846, 7 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Londrillo & Del Zanna 2000, ApJ, 530, 508 <ref:2211. p. 1 1 can be written in vectorial form,
similar to e. Londrillo & Del Zanna (2000),
∂U
∂t
+
∂F
∂x
+
∂G
∂y
+
∂H
∂z
= Ψ, (3)
where
U = (ρ, ρux, ρuy, ρuz, ρS)T
, (4)
Ψ = (0, Φx, Φy, Φz
, Q/Tgas + S gen)
T
, (5)
and
F =
ρux
ρu
2
x + Ptot − B
2
x
ρuxuy − BxBy
ρuxuz − BxBz
ρuxS
(6)
is the flux function.Where the paper says it
Nordlund et al. (2018), MNRAS, 477, 624 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Padoan et al. (2020), ApJ, 900, 82 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Pan et al. (2018), ApJ, 866, L17 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Pan et al. (2019), ApJ, 881, 155 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Parenti 2014, Living Reviews in Solar Physics, 11, 1 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Peter et al. p. 0 In addition, certain features in the atmosphere of the Sun can span large distances, e. coronal
loops ∼ 100 Mm (Peter, H. In addition, certain features in the atmosphere of the Sun can span large distances, e. coronal
loops ∼ 100 Mm (Peter, H. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Popovas (2022), arXiv e-prints, arXiv:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. 09564v1 [astro-ph. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. 09564v1 [astro-ph. 09564v1 [astro-ph. SR] 17 Nov 2022 09564v1 [astro-ph. SR] 17 Nov 2022Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Popovas & Nordlund (preliminary), A&A <ref:2211. p. 2 Popovas, Nordlund & Szydlarski: Volleyball Sun
The divergence-free magnetic field condition (Eq. 2) inside patches is maintained via the very accurate CT algorithm (see e. Popovas, Nordlund & Szydlarski: Volleyball Sun
The divergence-free magnetic field condition (Eq. 2) inside patches is maintained via the very accurate CT algorithm (see e.Where the paper says it
Where the paper says it
Popovas, Nordlund & Ramsey, J. p. 2 Popovas, Nordlund & Szydlarski: Volleyball Sun
The divergence-free magnetic field condition (Eq. 2) inside patches is maintained via the very accurate CT algorithm (see e. Popovas, Nordlund & Szydlarski: Volleyball Sun
The divergence-free magnetic field condition (Eq. 2) inside patches is maintained via the very accurate CT algorithm (see e. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Popovas et al. 2018, MNRAS, 479, 5136 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Przybylski et al. p. 0 (2016); Chen et al. (2022);
Przybylski et al. (2016); Chen et al. (2022);
Przybylski et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Reale 2010, Living Reviews in Solar Physics, 7, 5 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Rieutord & Rincon 2010, Living Reviews in Solar Physics, 7, 2 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Ronchi et al. p. 3 The tilt between adjacent patches is very small, with exceptions at the ‘seams’
between the six ‘faces’ of the volleyball. Volleyball patch arrangement
This mesh decomposition may, from a first glance, appear
similar to a cubed-sphere (Ronchi et al. The tilt between adjacent patches is very small, with exceptions at the ‘seams’
between the six ‘faces’ of the volleyball. Volleyball patch arrangement
This mesh decomposition may, from a first glance, appear
similar to a cubed-sphere (Ronchi et al. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Teyssier (2002), A&A, 385, 337 <ref:2211. p. 1 (2018, 2019), an
internal energy based Stagger solver taken straight from the
Bifrost (Gudiksen et al. 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al. p. 1 (2018, 2019), an
internal energy based Stagger solver taken straight from the
Bifrost (Gudiksen et al. 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al. (2018, 2019), an
internal energy based Stagger solver taken straight from the
Bifrost (Gudiksen et al. 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al. (2018, 2019), an
internal energy based Stagger solver taken straight from the
Bifrost (Gudiksen et al. 2011) code, and several Godunovtype Riemann solvers based on the HLLC and HLLD solvers
from the public domain RAMSES code (Teyssier 2002;
Teyssier et al.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Teyssier et al. 2006, Journal of Computational Physics, 218, 44 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Tomida & Hori (2016), personal communication <ref:2211. p. 2 Equation of state
We use a tabular EOS (Tomida et al. p. 2 A more detailed
procedure description is beyond the scope of this work and
is deferred to Popovas (2022). Equation of state
We use a tabular EOS (Tomida et al. A more detailed
procedure description is beyond the scope of this work and
is deferred to Popovas (2022). Equation of state
We use a tabular EOS (Tomida et al. Equation of state
We use a tabular EOS (Tomida et al. 2013, updated by
Tomida & Hori 2016), which uses logarithmic density and
temperature as independent variables.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Tomida et al. 2013, ApJ, 763, 6 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211. These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Where the paper says it
Winters & Gassner 2016, Journal of Computational Physics, 304, 72 <ref:2211. p. 1 Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al. Many different ad hoc solutions have been suggested (e. Ismail &
Roe 2009; Winters & Gassner 2016; Gallice et al.Where the paper says it
Where the paper says it
Acknowledgments
This research was supported by the Research Council of Norway through its Centres of Excellence scheme, project number 262622, and through grants of computing time from the Programme for Supercomputing, as well as through the Synergy Grant number 810218 (ERC-2018-SyG) of the European Research Council <ref:2211.09564#pg10>.
Bezanson, J., Edelman, A., Karpinski, S., & Shah, V. B. 2014, arXiv e-prints, arXiv:1411.1607 <ref:2211.09564#pg10>.
Brchnelova, M., Zhang, F., Leitner, P., et al. 2022, Journal of Plasma Physics, 88, 905880205 <ref:2211.09564#pg5>.
Carlsson, Mats et al. (2016), A&A, 585, A4 <ref:2211.09564#pg4>.
Chen, F., Rempel, M., & Fan, Y. 2022, ApJ, 937, 91 <ref:2211.09564#pg8>.
Christensen-Dalsgaard et al. 1996, Science, 272, 1286 <ref:2211.09564#pg5>.
Druett et al. 2022, A&A, 665, A6 <ref:2211.09564#pg4>.
Evans & Hawley 1988, ApJ, 332, 659 <ref:2211.09564#pg2>.
Finley et al. 2022, A&A, 665, A118 <ref:2211.09564#pg5>.
Fromang et al. 2006, A&A, 457, 371 <ref:2211.09564#pg4>.
Gallice et al. 2022, Journal of Computational Physics, 468, 111493 <ref:2211.09564#pg9>.
Gudiksen et al. (2011), A&A, 531, A154 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Guerrero et al. 2022, arXiv e-prints, arXiv:2208.05738 <ref:2211.09564#pg8>.
Hotta & Kusano 2021, Nature Astronomy, 5, 1100 <ref:2211.09564#pg2>.
Hotta et al. 2022, ApJ, 933, 199 <ref:2211.09564#pg5>.
Ismail & Roe 2009, Journal of Computational Physics, 228, 5410 <ref:2211.09564#pg2>.
Kageyama & Sato (2004), Geochemistry, Geophysics, Geosystems, 5 <ref:2211. p. 0 (2022)
used global simulations of the solar convection zone to study
solar differential rotation. They used spherical geometry
and a Yin-Yang grid (Kageyama & Sato 2004) with the
sphere spanning 0.Where the paper says it
Kapylä (2021), A&A, 651, A66 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Kohutova & Popovas 2021, A&A, 647, A81 <ref:2211.09564#pg2>.
Kuffmeier et al. (2017), ApJ, 846, 7 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Londrillo & Del Zanna (2000), ApJ, 530, 508 <ref:2211. p. 1 1 can be written in vectorial form,
similar to e. Londrillo & Del Zanna (2000),
∂U
∂t
+
∂F
∂x
+
∂G
∂y
+
∂H
∂z
= Ψ, (3)
where
U = (ρ, ρux, ρuy, ρuz, ρS)T
, (4)
Ψ = (0, Φx, Φy, Φz
, Q/Tgas + S gen)
T
, (5)
and
F =
ρux
ρu
2
x + Ptot − B
2
x
ρuxuy − BxBy
ρuxuz − BxBz
ρuxS
(6)
is the flux function.Where the paper says it
Nordlund et al. (2018), MNRAS, 477, 624 <ref:2211.09564#pg10>.
Padoan et al. (2020), ApJ, 900, 82 <ref:2211.09564#pg4>.
Pan et al. (2018), ApJ, 866, L17 <ref:2211.09564#pg5>.
Pan et al. (2019), ApJ, 881, 155 <ref:2211.09564#pg5>.
Parenti (2014), Living Reviews in Solar Physics, 11, 1 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Peter et al. (2013), A&A, 556, A104 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Popovas (2022), arXiv e-prints, arXiv:2211.09564v1 [astro-ph.SR] 17 Nov 2022 <ref:2211.09564#pg4>.
Popovas & Nordlund (preliminary), A&A <ref:2211.09564#pg4>.
Popovas, Nordlund & Ramsey, J. P. 2019, MNRAS, 482, L107 <ref:2211.09564#pg10>.
Popovas et al. 2018, MNRAS, 479, 5136 <ref:2211.09564#pg5>.
Przybylski et al. (2022), A&A, 664, A91 <ref:2211.09564#pg2>.
Reale (2010), Living Reviews in Solar Physics, 7, 5 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Rieutord & Rincon (2010), Living Reviews in Solar Physics, 7, 2 <ref:2211. p. 0 coronal
loops ∼ 100 Mm (Peter, H. 2013; Reale 2010), prominences ∼ 10–100 Mm (Parenti 2014), and supergranulation
∼ 30–50 Mm (Rieutord & Rincon 2010).Where the paper says it
Ronchi et al. 1996, Journal of Computational Physics, 124, 93 <ref:2211.09564#pg2>.
Teyssier (2002), A&A, 385, 337 <ref:2211.09564#pg5>.
Teyssier et al. (2006), Journal of Computational Physics, 218, 44 <ref:2211. p. 0 These works are just a few examples, using rather different mesh decomposition, physics involved, spatial and time resolution, etc. What they have
in common is an effort to connect the deep interiors with
Article number, page 1 of 12
arXiv:2211.Where the paper says it
Tomida & Hori (2016), personal communication <ref:2211.09564#pg3>.
Tomida et al. 2013, ApJ, 763, 6 <ref:2211.09564#pg5>.
Winters & Gassner 2016, Journal of Computational Physics, 304, 72 <ref:2211.09564#pg9>.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved system can conduct global MHD simulations of the solar convective zone using a new volleyball mesh decomposition that avoids coordinate singularities while maintaining resolution variation with radius. This allows for studying plasma dynamics across scales from deep interiors to surface layers without sacrificing realism due to geometric constraints.
-
Improved system can employ a new entropy based HLLD Riemann solver, which
gives the accuracy and performance of a Godunov solver while avoiding the negative pressure problem and providing correct convective flow velocities.
This enhances the simulation's ability to accurately model thermal energy evolution in turbulent flows. -
Improved system can handle complex boundary conditions by using radial boundary conditions for the entire surface of the Sun, simplifying setup compared to plane-parallel models. It can also use
diminishing boundary condition for the vertical momentum component
at the bottom and acork-like
condition at the top to emulate solar physics accurately. -
Improved system can perform advanced mesh management through static mesh refinement, allowing resolution to increase as convection develops, with refinement levels added during routine restarts or based on simulation conditions. This enables researchers to study features like
active regions and flux emergence
with sufficient resolution for detailed analysis. -
Improved system can utilize a flexible task-based DISPATCH code framework that allows for modular selection of solvers and physics, including
non-ideal MHD, heat conduction, and radiative energy transfer.
This flexibility enables the gradual inclusion of more complex microphysics to study phenomena likelocal and global dynamo effects.
-
Improved system can perform robust parallel scaling tests across different supercomputers by demonstrating
nearly perfect weak scaling
on diverse hardware configurations. This confirms that thevolleyball mesh decomposition together with the experimental setup has very good strong scaling properties,
ensuring the simulation is computationally efficient for large-scale production runs.
Abstract
Solar modelling has long been split into `internal' and `surface' modelling because of the lack of tools to connect the very different scales in space and time, as well as the widely different environments and the dominating physical effects involved. Significant efforts have recently been put into resolving this disconnect. We address the outstanding bottlenecks in connecting internal convection zone and dynamo simulations to the surface of the Sun, and conduct a proof-of-concept high-resolution global simulation of the convection zone of the Sun, using the task-based DISPATCH code framework. We present a new `volleyball' mesh decomposition, which has Cartesian patches tessellated on a sphere with no singularities. We use our new entropy based HLLS approximate Riemann solver to model magneto-hydrodynamics in a global simulation, ranging between 0.655---0.998 R, with an initial ambient magnetic field set to 0.1 Gauss. This paper describes the methods and the pilot runs of the setup. The mesh decomposition exhibits neither unwanted artefacts, nor noise build-up. The simulations develop convective motions with complex, turbulent structures. Small-scale dynamo action twists the ambient magnetic field and locally amplifies magnetic field magnitudes by more than two orders of magnitude within the initial run-time.
Sources
- Julia: A Fresh Approach to Numerical Computing
- Implicit large eddy simulations of global solar convection: effects of numerical resolution in non-rotating and rotating cases
- DISPATCH methods: an approximate, entropy-based Riemann solver for ideal magnetohydrodynamics
Related papers
- He-star donor pathway for the hypervelocity star D6-2
- The ASPIICS solar coronagraph aboard the Proba-3 formation flying mission. Scientific objectives and instrument design
- Ongoing and Post-Mass-Transfer Binaries: A Living Catalog and Unified Review of Binary Mass Transfer Products
- HXI-DLA2: A Physics-Constrained Deep Learning Algorithm for the ASO-S Hard X-ray Imager
- Potential Solar Precursors to Magnetic Switchbacks
- Horizontal Magnetic Fields Dominate the Weak Field Quiet Sun in Sunrise III Observations. Evidence from Traditional and Transformer-Based Inversions