EME Report
Read all

Interactive MCE Simulations

Run the MCE Theory numerical simulations directly in your browser — no installation required. Each simulation executes real Python/NumPy/Matplotlib code via Pyodide (Python compiled to WebAssembly). All computation is local to your device; nothing is sent to a server.

First run: The Python runtime (NumPy + Matplotlib, ~25 MB) downloads once per browser session. Expect 15–40 seconds on a typical connection. All subsequent simulations on this page are instant as the runtime is cached.

Available Simulations

Observable Phase DiagramCore prediction
Ready
Colour-coded (r, ρ) map of the MCE suppression function S(r,ρ). Shows why MICROSCOPE finds a null result (deep blue) whilst the atom interferometry target at r ≈ 1 µm sits in the red detectable zone. The green band marks the current working range λ_c ∈ [1, 10] µm.
Runs entirely in your browser via Pyodide — no server, no timeout.
Suppression Function ProfilesTheory validation
Ready
Slice plots of S(r,ρ) vs distance r (left) and vs density ρ (right) for a range of parameter values. Demonstrates the seven-orders-of-magnitude dynamic range that makes MCE simultaneously compatible with MICROSCOPE and predictive at the micrometre scale.
Runs entirely in your browser via Pyodide — no server, no timeout.
RG Running & Lattice QCD Error BudgetQuantitative prediction
Ready
One-loop QCD renormalisation group running of C_QFT from the EFT UV cutoff (10 GeV) to the hadronic scale (200 MeV), with full error propagation from FLAG 2024 lattice inputs. Yields the conservative pre-registerable benchmark Δa/a = (6.0 ± 0.7) × 10⁻⁹ together with the explicit λ_c scan up to ≈ 1.5 × 10⁻⁸ at r = 1 µm.
Runs entirely in your browser via Pyodide — no server, no timeout.
GRACE-FO Toroidal Coupling ForecastAppendix J prediction
Ready
Simulates the predicted MCE geomagnetic-QVP coupling as a GRACE-FO gravity residual map. Models the toroidal B-field proxy and predicted Δg pattern, then adds a realistic GRACE-FO noise floor to produce a simulated observable. Includes cross-correlation significance test.
Runs entirely in your browser via Pyodide — no server, no timeout.

Technical Notes

  • Source code: The browser simulations above are functionally equivalent to the standalone Python scripts in the scripts/ directory, which can be run locally with NumPy and Matplotlib installed.
  • Reproducibility: All random seeds are fixed (np.random.default_rng(42)). Results are deterministic and identical across runs.
  • Figure download: Click the "Download figure" link beneath each plot to save a high-resolution PNG.
  • Parameters: λc benchmark = 1 µm with current theory band [1, 10] µm, ρc = 1.1 × 10³ kg/m³, ΔδAl–Au = 1.9 × 10⁻⁸. See Appendix L for the RG-improved values.