Even the most advanced Micro- and Meso-scale flow models depend upon simplified parameterizations derived from horizontally-homogeneous boundary layers - conditions never satisfied at actual wind turbine sites. To develop better models, we must continually support research and improve our understanding of fundamental fluid mechanics. Belcher, Harman, and Finnigan (2012) describe our current understanding of the fluid mechanics of flow through forest canopies in hilly terrain and identifies metrics to assess whether a particular site is likely subject to complex terrain effects where flow models may fail.
The flow dynamics over a forest can be quite different than over a similar rough hill. Whereas a rough surface increases the surface friction, a forest influences the flow via pressure drag, making forested hills more prone to flow separation. The canopy drag length scale, Lc, is a critical parameter to investigate if rough-hill approximations are appropriate for a forested site. Lc is inversely proportional to the canopy density, so sparse forests are more prone to cause complex terrain effects than dense ones. In addition, Belcher, Harman, and Finnigan (2012) also describe how hill slope, tree-type, forest edges and stability may violate the homogeneous assumptions on which most flow models are.
Reference
Belcher, Stephen E., Ian N. Harman, and John J. Finnigan. 2012. “The Wind in the Willows: Flows in Forest Canopies in Complex Terrain.” Annual Review of Fluid Mechanics 44 (1): 479–504. https://doi.org/10.1146/annurev-fluid-120710-101036.