Hybrid physical-statistical modelling of wind farms and complex terrain - CASE project

Modelling the aerodynamics of large wind turbine arrays is a complex multi-scale problem, but is an essential part in the design and optimisation of ever-growing wind power plants or wind farms. The small-scale details of the flow around each turbine can feed back onto the large-scale environment to affect the availability of wind energy across the whole turbine array. This multi-scale nature of wind farm modelling shares much common theoretical ground with the modelling of flow over complex terrain. In weather and climate models, unresolved features near the earth's surface such as small hills, vegetation and buildings impart a drag on the atmosphere. Accurately modelling this drag is important not only for correctly forecasting near-surface conditions of the atmosphere, but also for the overall trajectory of the weather/climate model over a long lead time due to an up-scale transfer of information onto the resolved flow. For both of these modelling problems, the key question is how the large-scale atmospheric flow is modulated by an exchange of momentum with small-scale features near the earth's surface.

This DPhil project will build on an ongoing collaboration between the University of Oxford and the UK Met Office in the area of multi-scale modelling of large wind farms. Our recent study has demonstrated that a relatively simple theoretical model, derived from the law of momentum conservation in a two-scale coupled manner, can predict the key effect of atmospheric interaction with wind turbines satisfactorily under a variety of realistic atmospheric stability conditions (or wind profiles). This motivates an extension of the two-scale coupled flow modelling to more complex problems, including those of flow over complex terrain.

One of the main difficulties in the modelling of flow over complex terrain is to parameterize Turbulent Orographic Form Drag (TOFD) for small-scale terrain that is unresolved in an atmospheric flow model. A recent investigation at the Met Office into the use of new high-resolution orographic data sets suggests that the formulation currently used in the Met Office's weather prediction model (UM) is not capable of handling the much broader spectrum of scales available from the new orographic data sets. A new approach based on the concept of two-scale coupled flow modelling is therefore developed and used in this DPhil project to improve the parameterization of TOFD. Such an extension of the two-scale coupled flow modelling for complex terrain will also help improve the modelling of large wind farms further.

The student will be supervised jointly by Dr Takafumi Nishino (Civil Engineering Research Group in the Department of Engineering Science, University of Oxford) and Dr Thomas Dunstan (Atmospheric Processes and Parameterizations Group, UK Met Office). The student will also have opportunities to work together with scientists at the Met Office during a placement (for approximately 6 months) as well as through regular research meetings. The expected start date of the project is in October 2020.