Answers are grounded in this paper and cite their sources.
Complex-terrain wind resource assessment usually relies on commercial CFD tools. We validate an open-source OpenFOAM RANS workflow under production constraints (10 m grid, public land-cover input, neutral stratification) against the Perdigão 2017 campaign, alongside Meteodyn WT (neutral and stability-stratified) on the same terrain inputs. A numerical verification came first. Converged solutions over-predict flow separation, more so with canopy drag and a second-order momentum scheme, and customary residual thresholds stop the solver before convergence. Sector averaging over three inflow directions was the only measure that reduced the error. Over all mast pairs at ≥20 m it gives the open-source chain lower errors than neutral Meteodyn (speedup log-RMSE 0.40 against 0.49) and a lower time-series transfer error than stratified Meteodyn (2.54 against 2.75 m/s); with one direction per sector the two neutral chains are equivalent. At hub height and between ridge-top masts the three configurations are equivalent on speed; on turbulence intensity Meteodyn is better in both configurations, and stratification improves only climatological quantities. All configurations under-predict the downstream-ridge speed by 7–30 % for cross-ridge flow, independently of mesh resolution, closure, forest treatment and stability classes. We interpret this as a limit of steady RANS rather than of either implementation.
No takes yet. Share an insight, caveat, or question.
Ingo Meirold‐Mautner (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: