**4. Conclusions**

The capability of the CFD software WindSim to predict the power outputs of the wind turbines of a wind farm over complex terrain was evaluated in this study. The site of the case study featured the co-presence of three complexities: topography, heterogeneous vegetation with a woodland-grassland mosaic, and interactions between wind turbine wakes. Hence, it allowed an in-depth evaluation of CFD models. The outcome of this study can be concluded as follows:


Comparison with more advanced models such as large-eddy simulation together with actuator disk model would help to verify that the good agreemen<sup>t</sup> was not due to the offset of various modeling errors discussed in the paper. Moreover, for forested mountainous regions, high-resolution terrain and vegetation data such as canopy height and density are needed to estimate accurately the relevant parameters for numerical wind energy prediction.

**Author Contributions:** Conceptualization, J.F. and F.P.-A.; Methodology, J.F. and D.T.; Software, D.T. and J.F.; Validation, D.T. and J.F.; Formal Analysis, D.T. and J.F.; Investigation, D.T. and J.F.; Resources, F.P.-A.; Data Curation, D.T. and J.F.; Writing—Original Draft Preparation, D.T. and J.F.; Writing—Review and Editing, J.F. and F.P.-A.; Visualization, D.T. and J.F.; Supervision, J.F.; Project Administration, J.F. and F.P.-A.; Funding Acquisition, F.P.-A. and J.F.

**Funding:** This research was funded by the Swiss Innovation Agency (Innosuisse) gran<sup>t</sup> number 1155002544 within the frame of the Swiss Centre for Competence in Energy Research on the Future Swiss Electrical Infrastructure (SCCER-FURIES).

**Acknowledgments:** The authors thank Johannes Vogel for providing the Juvent wind farm data. The first author wishes to thank Charles Meneveau for providing the scholarship.

**Conflicts of Interest:** The authors declare no conflict of interest.
