Parkers Physics is a real-time space weather simulation platform that runs entirely in your browser. It pulls live data — L1 solar wind, Kyoto Dst, GOES X-ray flux — and feeds it through an actual physics model of Earth’s ring current that I built with Claude. What you see on screen isn’t an animation of space weather. It is real-time space weather, computed as it happens with NASA data.
Parkers Physics is a real-time space weather simulation platform that runs entirely in your browser. It pulls live data — L1 solar wind, Kyoto Dst, GOES X-ray flux — and feeds it through an actual physics model of Earth’s ring current, so what you see on screen isn’t an animation of space weather. It is space weather, computed as it happens.
▶ View the Ring Current Simulation live on ParkersPhysics.com
This post covers how the simulation works, the physics running under the hood, and where the project is headed — including EarthView, the weather side of the app, and the Sun digital twin that’s coming next.
My goal with Parkers Physics is to be agile and productive with the AI tools now available to us and build as much high-fidelity simulation software as possible. Quality improves iteratively, and I believe strongly in transparency — so I’ll always be honest when the models or simulations fall short.
What Is Earth’s Magnetosphere?
Earth’s magnetosphere is the giant magnetic bubble generated by our planet’s molten iron-nickel core dynamo. The solar wind — a constant stream of charged particles from the Sun — slams into that bubble and shapes it, compressing the dayside and stretching the nightside into a long tail. Where the wind and field interact, you get distinct regions: the magnetosheath, the ionosphere, and, circling the planet at a few Earth radii, the ring current — a river of trapped ionic particles drifting around Earth.
The ring current matters far beyond academic curiosity. Its behavior during geomagnetic storms affects power grids, GPS accuracy, satellite operations, and communications — and it’s one of the key things we need to understand well before humans spend serious time in space.
A Real-Time Ring Current Simulation, Not a Visualization
The Parkers Physics ring current simulation was built to show particle populations and their spectroscopic properties in real time. The one-line version: it’s a physics pipeline that happens to be beautiful, not a visualization that happens to cite physics.
Everything on screen traces back to one unbroken causal chain, computed end to end:
Solar wind → convection → anisotropic injection → wave growth → proton aurora + He⁺ erosion + GEO electron dropouts
Nothing in that chain is scripted. Every arrow is a model step you can inspect, and cause and effect share colors across the page so you can follow the story visually.
There are three default camera angles. The Sun view centers on the Sun and its output — a shader corona with shearing streamers, breathing with the live GOES X-ray flux, fast wind fanning off coronal holes at today’s actual positions, and the Parker-spiral line connecting it all back to Earth. The Earth view gives you the full distribution of particle life cycles, with 4,700 GPU-resident particles whose kinematics are the vertex shader. And there’s a draggable ENA imager that does Roelof-style line-of-sight integration in a shader — the same technique real satellites use to photograph the ring current in energetic neutral atoms.
The Physics Under the Hood
For readers who want the deeper detail, here’s what’s actually being computed. (If equations aren’t your thing, feel free to skip to the next section — the short version is: it’s a real transport model, implemented twice, and tested hard.)
A dual-implementation transport core. The heart of the simulation is a bounce-averaged ring current model tracking particles across L-shell, magnetic local time, energy, and species. It’s implemented twice — once in JavaScript as a reference oracle, once as a Rust→WASM kernel — and an automated parity test pins the two to agree exactly. The model includes corotation and shielded Volland–Stern convection drift, Schulz–Lanzerotti radial diffusion, charge exchange against the Rairden geocorona, and Kp-gated plasma-sheet injection. The Dst* index isn’t fit to data — it emerges from the modeled energy via the Dessler–Parker–Sckopke relation.
A self-limiting wave feedback loop. Pitch-angle anisotropy rides through the same transport operators as an evolving state: particles inject anisotropic and isotropize over about twelve hours. EMIC waves ignite only where the locally evolved anisotropy crosses threshold, with enough hot-proton pressure, inside the cold-plasma overlap near the plasmapause. The scattering that drains protons then relaxes the anisotropy back down — the waves regulate themselves.
Three consequences from one wave field. That single computed wave-activity map gates proton precipitation into the atmosphere, He⁺-band scattering, and MeV radiation-belt electron dropouts — the belt layer runs Brautigam–Albert radial diffusion with both real dropout mechanisms, magnetopause shadowing and EMIC scattering.
The satisfying part is what emerges. The waves localize to the dusk sector on their own. Storm-time gating shows a 46× contrast over quiet conditions. The model finds its activity peak at L 4.2, MLT 19.8 — the textbook proton-arc location — without ever being told where it should be.
Research-Grade Instruments in a Web Page
Alongside the 3D scene, the simulation carries a set of analysis panels you’d normally expect in mission science software:
- A drift-path tracer and Alfvén layer showing both trapping regimes, switching live as you move through the 5–300 keV energy selector
- Phase-space density profiles — the RBSP-style f(μ; L) analysis, with an automatic verdict of inward diffusion versus local acceleration
- A storm-phase auto-detector (quiet / initial / main / recovery) with fitted recovery time and an energy ledger: peak energy, energy shed, energy built
- A Dst budget decomposition with live signed bars for the ring, magnetopause, and baseline contributions — plus the honest unmodeled residual, and a live correlation score against the Kyoto Dst index
That last point matters to me. The residual bar exists precisely because I’d rather show you what the model doesn’t capture than pretend it captures everything.
On the verification side: roughly 80 test pins across 7 suites, exact e-folding replication tests, twin-run controls, and two independent language implementations of the core physics held byte-equal. That’s a standard most research codes don’t meet — and it all runs at 60 fps in a browser, no bundler, no framework.
EarthView: The Weather Side of Parkers Physics
The project actually started as a pure solar wind simulator and recorder, and it evolved fast. As I moved the simulation inward from space toward Earth, EarthView was born — enter your location and it gives you a forecast that, in my testing, beats most weather apps on accuracy.
EarthView currently includes:
- Location-specific temperature analysis and forecasting
- Live storm watch tracking storms as they emerge (Elida, most recently), including how pressure systems shape storm evolution
- A 24-hour time scrubber for barometric pressure, with wind and clouds catching up soon (the clouds are a long-term visual project, LOL)
- Sunrise, sunset, and solar noon for your location
- Kp index, aurora alerts, and moon phase
- Air quality (AQI) and UV index
- Fire tracking, with an aerosol pollution analyzer already in the works
Wind analysis works well, though there’s a long road ahead on data fidelity — especially for regions without public data sources. I’m hoping to release EarthView on mobile alongside a couple of other Parkers Physics simulations, most notably the AurOracle aurora simulator.
What’s Next: A Digital Twin of the Sun
The next big step for Parkers Physics is the Sun digital twin. The current version already has active region analysis and a new far-side analyzer, but the solar wind model needs much more work before it approaches anything predictive. The Sun is, in many ways, still a mystery to us — which is exactly what makes it worth simulating.
Moving up through the upper atmosphere and magnetosphere made one thing obvious to me: these are the spaces where AI becomes invaluable. Real-time magnetospheric measurement and simulation is now possible for a solo developer, thanks to tools like Claude and Codex — the “digital twin” idea is becoming more real in my work every month.
Frequently Asked Questions
Is the simulation using real data? Yes. It ingests live L1 solar wind measurements, the Kyoto Dst index, GOES X-ray flux, HEK coronal hole positions, and DONKI event data. The physics model runs on that live input — nothing is pre-rendered or scripted.
What is the ring current? The ring current is a population of charged particles (mostly protons and He⁺ ions, tens to hundreds of keV) trapped in Earth’s magnetic field, drifting in a ring around the planet. During geomagnetic storms it intensifies dramatically, which is what depresses the Dst index and drives many storm effects on grids and satellites.
Do I need to install anything? No — it runs in any modern browser at parkersphysics.com/ring-current.html, with adaptive quality shedding for slower machines.
Where does the physics come from? Published, peer-reviewed models: Volland–Stern convection, Schulz–Lanzerotti diffusion, Dessler–Parker–Sckopke energy coupling, Brautigam–Albert electron diffusion, and more. The references are linked in the app, and if you’re interested in the physics, I can’t recommend the reference reading enough.
Feedback on the simulation is more than welcome, and I’m looking forward to developing it further. If you have questions, reach out to me on X/Twitter.
Earth’s Magnetosphere – A River of Ionic Particles
View the Simulation Live on Parker’s Physics.com
Surrounding planet Earth, there is a giant magnetic barrier called the magnetosphere that corresponds to the magnetic qualities of the iron and nickel core dynamo of the Earth. This barrier is created by the solar wind and creates what is called the magnetosphere, ionosphere and magneto-sheath, which are different functional aspects of how the wind interacts with the planet.
The Parkers Physics Ring Current UI was built to understand the particle populations as well as their spectroscopic properties in real time. This gives us a good look into the layered and complex behaviors of the current current and how its behavior can be measured as we get more advanced tools and more satellite data. This is an evolving field of study and is extremely important for space travel, not to mention Earthbound grids, GPS, and communications during storms. This should provide a good baseline for measurements as we move forward.
ENA Imager and Spectroscopy in the Parkers Physics Ring Current Simulation
The Earth camera view of the simulation is pretty awesome, we get a full distribution of particle life cycles and activity and we also get to analyze their individual behaviors according the the equations given by the references. There are many of them, so if you are interested in the physics, I can’t recommend the reference reading enough.
Primary Reference:
- THE TERRESTRIAL RING CURRENT: FROM IN SITU MEASUREMENTS TO GLOBAL IMAGES USING ENERGETIC NEUTRAL ATOMS – EDMOND C. ROELOF and DONALD 1. WILLIAMS
