FLOW-Alaiz benchmark for coupled terrain and array interaction flow models. Baseline Results

The FLOW-Alaiz benchmark is established to enhance flow models in complex terrain, building on insights and datasets from previous IEA Task 31 benchmarks around the same test site. The new benchmark aims to create a composite validation dataset by gathering measurements from different periods, notably, the New European Wind Atlas (NEWA) ALEX17 experiment and the Alaiz Site Calibration (SC) campaign. The overall study focuses on wind conditions relevant to siting, power performance, and energy yield prediction. Initial results use industry-standard flow models as baselines such as SiteFlow and PyWAsP/PyWAsP-CFD, all in neutral steady-state conditions. Public datasets, documentation, and evaluation scripts are available in the FLOW-Hub repository to provide transparency on the evaluation methodology while private data is kept on SGRE premises for blind testing. Baseline results focus on cross-predictions and wind profiles for the North and South sectors. Cross-predictions reveal challenges in the NEWA case due to large elevation differences and complex flow interactions in the valley, while SC campaigns show better performance in more realistic siting conditions. Wind profiles for neutral conditions generally align with expectations with main limitations in the wake of flow-separation and forest canopy zones. A comparison with vertical profiles from the ALEX17 Diurnal Cycles benchmark highlights potential improvements from high-fidelity mesoscale-to-microscale modeling, revealing distinct variations in wind shear and profile characteristics when stability is used to categorize the atmospheric conditions.