Posters

Abstract

Turbulence Intensity (TI) is one of the most important turbulence metrics in the context of wind resource assessment, particularly for site characterization and load estimation. Ideally, industry practice relies on cup-anemometer measurements of TI, stratified by wind speed and direction bins. Offshore, however, the installation of cups is often avoided due to high costs and operational challenges. As a result, alternatives are needed. Numerical weather prediction models such as the Weather Research and Forecasting (WRF) model, coupled with wake parameterizations, are already widely applied for providing mean wind fields. These models are not commonly used to supply turbulence information, yet several of their turbulence closure schemes prognose subgrid-scale turbulent kinetic energy (TKE). Since TKE inherently carries information about turbulent fluctuations within the grid cell, it represents the most promising basis for estimating TI from model output. In this study, we investigate different approaches for converting TKE into TI using high-frequency sonic anemometer data at 55 m height from an extensive onshore  experiment. We benchmark several formulations against the horizontal TI, which serves as the reference. Our results show that the Archer method performs best, yielding a slope of 0.983, R² of 0.793, and RMSE of 0.033. The Larsén method performs nearly as well (slope 0.970, R² 0.792, RMSE 0.037), despite requiring only TKE and mean wind speed. By contrast, the widely used 2TKE and 2/3TKE formulations consistently over- and underestimate TI, respectively, although discrepancies diminish at wind speeds above 2 m s⁻¹. These findings provide a systematic evaluation of existing TKE–TI conversion methods and highlight pathways for exploiting WRF-derived TKE as a practical proxy for offshore TI characterization.