Self-organizing snow drifting over flat terrain: Two distinct pathways to dynamic equilibrium in Eulerian–Eulerian numerical models

Wind-induced snow drifting is a self-organizing process: the transport rate does not grow indefinitely but saturates to a dynamic equilibrium through feedback between the wind field and the snow phase. Different Eulerian–Eulerian frameworks all reproduce this equilibrium, yet the mechanisms by which they do so remain unexplored, which obscures the true sources of model accuracy. We compare five representative sub-methods (spanning scalar transport, mixture, and two-fluid formulations) in a single Reynolds-averaged Navier–Stokes environment with identical mesh, boundary conditions, turbulence closure, and physical parameters. All five are validated against an uncertainty envelope of measured drift-density and snow-flux profiles, with the mixture model showing the most consistent quantitative agreement and the two-fluid model close behind. Despite reaching the same equilibrium, the models follow two fundamentally distinct pathways, rooted in the density scale that each framework uses to define the surface friction velocity. In the scalar transport models, a shear-stress-dominated negative feedback drives the bottom friction velocity toward its threshold, suppressing further erosion. In the mixture and two-fluid models, the near-surface drift density rises by orders of magnitude and the mixture density by tens of percent, while the bottom shear stress barely changes and interphase momentum exchange appears not to be dominant under the present closures. This dichotomy, not visible in steady-state profiles alone, is confined to the saltation layer, while in the dilute suspension above all models follow ordinary turbulent diffusion. These findings challenge a common assumption: that adding interphase momentum coupling is the key to improving snow transport models. Instead, in this benchmark the accuracy of the mixture and two-fluid models stems primarily from how they represent the bulk properties of the wind–snow mixture, not from the momentum exchange itself.