STEAM Model Proposes Scale-Invariant Turbulons to Replace Navier-Stokes in Atmospheric Simulation

DeWitt argues that atmospheric scale invariance permits emergent laws built on a new primitive called the turbulon, a buoyancy-adjusted generalization of turbulence theory. The STEAM implementation superposes these objects to generate synthetic volumes whose cloud visualizations and selected statistical moments match output from state-of-the-art large-eddy simulations. Computational savings arise because explicit resolution of small-scale contortions is replaced by statistical superposition rules derived from symmetry principles.

The approach directly challenges the assumption that Navier-Stokes integration remains the only viable path for weather and climate prediction. If scale invariance holds across the observed inertial range, many sub-grid processes currently parameterized through closure schemes could instead be generated from the same turbulon statistics. This reframes longstanding difficulties with cumulus and boundary-layer schemes as symptoms of an overly granular modeling ontology rather than missing physics.

Existing literature on atmospheric scaling (e.g., Lovejoy 2019 on multiplicative cascades and Schertzer & Lovejoy 1987 on generalized scale invariance) provides partial precedent, yet STEAM’s explicit construction of a discrete turbulon basis distinguishes it. Operational adoption would require systematic verification against reanalysis and satellite radiances; the preprint already notes superior reproduction of certain cloud statistics but acknowledges deficiencies in others.

Next steps include controlled intercomparison with CMIP-class models and targeted field campaigns to test whether turbulon statistics remain invariant under changing thermodynamic regimes. Independent replication on open-source hydrodynamic benchmarks is the immediate falsification threshold.