Recent developments in high-resolution wind field modeling in complex terrain for dispersion simulations using GRAMM-SCI

The rather complex terrain in Austria requires that wind fields with sub-kilometer resolutions have to be generated prior to any kind of dispersion modeling due to the large demand on computation times required by non-hydrostatic prognostic mesoscale models. Proper modeling of highly resolved flows in Alpine regions is still a matter of research, and no harmonized methodology is available at the moment. In order to harmonize the meteorological input for dispersion modeling at the regional level, authorities in Austria aim at providing so-called wind-field libraries for a certain reference year for all stakeholders involved in air quality assessments for regulatory purposes. The region of Styria was the first to establish such a library using the prognostic, non-hydrostatic mesoscale model GRAMM with a horizontal resolution of 300 m in 2015. Over the years, attempts have been made for improving the quality of the wind fields. One of the most challenging issues is the interaction between synoptic-scale flows and local thermally driven winds. In this work, the newly developed mesoscale model GRAMM-SCI is presented, which is driven by ERA5 reanalysis data. Especially, novel nudging techniques allow for nesting and dynamic downscaling wind fields with a horizontal resolution of 100–200 m. Moreover, a methodology called “match-to-observation” (MTO) is presented, which greatly improves the final quality of wind fields. For the first time, a wind-field library for the reference year 2017 has been generated for Styria with this new approach.