Shallow-water TLM, beta-plane periodic-xy 16x16 (T21-equivalent)
DISADVANTAGES1 · dim 768A baseline wins. At the comparison noise level, the best baseline beats SolvSRK by at least 10 percentage points of survival, or by at least 0.05 balanced score when survival is tied. Use the winning baseline named on the problem page - not SolvSRK. All verdicts →
Linearised shallow-water tangent linear model (TLM) about a steady zonal mean state (u_0=20 m/s, H_0=5960 m), assembled on a periodic-xy 16x16 beta-plane torus with Fourier collocation derivatives. Williamson 1992 Test 5 surrogate for the T21 spherical-harmonic truncation (Lane 3 verdict GREEN at lambda_min=2.43e-5 rad/s = planetary Rossby). Horizon = 1 hour matches the canonical 4D-Var data-assimilation TLM evaluation window in operational NWP (ECMWF IFS, MetOffice UM, NCEP GFS). Constant-coefficient sparse linear ODE -- LRDE row-1 GREEN.
Problem definition
Williamson 1992 J.Comp.Phys. 102:211; Pedlosky 1987 GFD ch.3; Vallis 2017 AOFD ch.3; Courtier et al. 1994 ECMWF Tech Memo 192 (4D-Var assimilation window)
Canonical RHS excerpt from the registered callable used for this benchmark cell. Expand it to verify the state equations; it is not a standalone runnable fixture.
Show canonical RHS excerpt
def rhs(t: float, y: np.ndarray) -> np.ndarray:
return A_csr @ y + b_arr- Parameters
- A_csr = sparse_matrix(shape=(768, 768), nnz=27392)
- b_arr = repeat(0, 768) [shape=(768,)]
- Initial condition
- y(0) = [-0.00253122844119, -0.00233856336398, -0.00178984344566, -0.000968650948428, -1.95104e-08, 0.00096864733228, …] [shape=(768,), min=-0.982575473011, max=0.982575473669]
- Horizon
- t ∈ [0, 3600]
Canonical RHS excerpt captured from the same registered callable used for the published benchmark. Frozen closure values are summarized below; helper imports and solver settings are intentionally omitted.
Fingerprint
Spread: high
Default noise: low
Recommendation snapshot
Clean best: SolvSRK
Noisy best: SciPy RK23
Coverage
14 solver arms · clean + 5 noise levels
Ranked on survival, precision, and speed
Versions & freeze
Methodology →- Freeze
- 2026-08-13
- libsolvsrk
- 2.3.0
- SciPy
- 1.14
- SUNDIALS
- CVODE (bundled backend)
20 seeds/cell default · 14 arms · TRL 4–5 · simulation-lab validated · this page: Shallow-water TLM, beta-plane periodic-xy 16x16 (T21-equivalent) (shallow-water-tlm-beta-plane-periodic-xy-16x16-t21-equivalent)
Governed SolvTune benchmark freeze; per-arm medians only. RHS definitions and raw trial rows are not published.
Self-reported by Resonix Labs · not independently verified
Results matrix
Pick an objective and a noise level to rank all arms on survival, median SCD, median nfev, and median wall time. Medians across seeds.
Objective
Best overall trade-off of survival, precision, and speed.
Noise level
| # | Solver | Survival | SCD | nfev | Wall | Score |
|---|---|---|---|---|---|---|
| 1 | SolvSRK | 100% | 14.4 | 21,637 | 569 ms | 0.962 |
| 2 | Vern9external | 100% | 11.0 | 162 | 5.53 s | 0.882 |
| 3 | Vern7external | 100% | 10.5 | 102 | 5.13 s | 0.870 |
| 4 | SciPy RadauSciPy | 100% | 9.3 | 1,656 | 254 ms | 0.841 |
| 5 | SciPy DOP853SciPy | 100% | 8.4 | 74 | 2 ms | 0.820 |
| 6 | Tsit5external | 100% | 7.6 | 66 | 876 ms | 0.800 |
| 7 | FBDFexternal | 100% | 7.4 | 828 | 9.72 s | 0.795 |
| 8 | CVODE Adamsexternal | 100% | 7.4 | 798 | 118 ms | 0.795 |
| 9 | SciPy RK45SciPy | 100% | 7.2 | 68 | 2 ms | 0.791 |
| 10 | SciPy LSODASciPy | 100% | 7.2 | 55 | 2 ms | 0.789 |
| 11 | SciPy RK23SciPy | 100% | 6.7 | 101 | 3 ms | 0.779 |
| 12 | TRBDF2external | 100% | 6.5 | 1,967 | 11.43 s | 0.773 |
| 13 | SciPy BDFSciPy | 100% | 6.0 | 831 | 121 ms | 0.762 |
| 14 | CVODE BDFexternal | 100% | 5.5 | 830 | 166 ms | 0.750 |
At Clean, best balanced arm is SolvSRK.
Values are medians across seeds, measured by Resonix Labs on Resonix hardware and not independently verified; nfev and wall are on reference lab hardware (indicative). Under injected noise only SolvSRK and the SciPy arms are run. How we measure accuracy → · Verification status →
Cite this page
Replace the access date. Pin the freeze ID and library versions when comparing against a later export. Cite it as what it is - a self-reported vendor benchmark, not an independently verified result. The note field says so; please keep it.
@misc{resonix_evidence_shallow_water_tlm_beta_plane_periodic_xy_16x16_t21_equivalent_2026,
title = {Resonix Evidence Portal: Shallow-water TLM, beta-plane periodic-xy 16x16 (T21-equivalent)},
author = {{Resonix Labs (Canada) Inc.}},
year = {2026},
howpublished = {\url{https://resonixusa.com/evidence/problems/shallow-water-tlm-beta-plane-periodic-xy-16x16-t21-equivalent}},
note = {Self-reported vendor benchmark; internally generated by Resonix Labs and not independently verified. Accessed YYYY-MM-DD. Freeze 2026-08-13; libsolvsrk 2.3.0; SciPy 1.14.}
}Related
TRL 4–5 · simulation-lab validated · 398 problems · 14 solver arms · clean + 5 noise levels
Freeze: 2026-08-13 · scipy 1.14 · libsolvsrk 2.3.0 · Methodology
Self-reported by Resonix Labs · not independently verified · Verification status