Marta Gil Bardaji, Vortex FdC / University of Barcelona, Spain

The correct representation of wind conditions and patterns is essential for wind resource assessment, and arguably the most accurate way of generating synthetic wind time series is to use an atmospheric model such as WRF to downscale large-scale reanalysis to local wind flow. To avoid the spurious numerical reflections at the transitions between nests, there has been a trend in recent years towards multiscale models that smoothly transition between scales, like MPAS. The goal of this study is to present a first operative MPAS-powered downscaling solution for wind resource assessment.

This study has focused on designing a mesh for first-stage wind prospecting: long-term simulations at relatively high resolutions over the wind farm area fed by reanalysis boundary conditions. The innovative ideas behind the methodology of building reliable meshes, as well as the limitations still in place, will be discussed in detail.

To assess the performance of MPAS and compare it to the industry-trusted WRF, we have designed a benchmark validation of one-year simulations in wind-energy-relevant locations representing different geographies and flow complexity scenarios. MPAS runs on a regional variable resolution mesh (from 3 to 20km), whereas WRF is set up on two nested domains (3 and 9km) that use one-way nesting (and, optionally, nudging). To focus on identifying real differences between modeling wind time series using MPAS meshes compared to using WRF nested domains, both models share the same settings whenever possible (static datasets, physics schemes, etc) and the post-process is analogous for MPAS and WRF output.

As a result of this analysis, it is now possible to carry out a realistic wind resource assessment project using seamless atmospheric modeling. Further developments in the MPAS dynamical core and, specially, the inclusion of a time-adaptive numerical scheme, will expand the horizons of MPAS capabilities and bring seamless modeling to the microscale.