Study on the influence of static terrain refinement correction on flow field simulation of Ulanqab wind power bases

The development of gigawatt-scale wind power bases has become a major trend in wind energy deployment. Yet, accurate assessment of wind resources and wind-farm wakes remains difficult because large uncertainties are introduced by complex surface conditions and sparse mast observations. In the present study, a refined framework for static terrain modification is presented, in which the Weather Research and Forecasting model is integrated with high-resolution Copernicus GLO-30 m elevation dataset elevation data and International Geosphere-Biosphere Programme land-cover data. Applied to the 6 GW Ulanqab wind power base in Inner Mongolia, the framework achieved a 26% reduction in background-flow simulation error relative to the default terrain configuration. This result indicates that sub-grid topographic refinement can capture non-linear terrain–wake interactions without the computational cost of mesoscale–microscale coupling. Quantitative wake analysis further showed that the refined inputs enabled realistic vertical wake stratification, with deficits increasing from 6.23% at rotor top to 16.03% at hub height. Wake recovery was found to be strongly controlled by atmospheric stability. Recovery distances ranged from 123 km (unstable) to 190 km (stable). In densely clustered arrays, cumulative perturbations within the boundary layer were identified, and vertical wake influence extended to 740 m depth. This depth greatly exceeds conventional boundary-layer recovery assumptions. These findings indicate priority should be given to static terrain data quality, not just higher grid resolution. A cost-effective route to engineering-grade accuracy in the mesoscale grey zone is therefore provided. Practical constraints also guide adaptive inter-farm spacing and seasonal operational optimization in large-scale wind power development.