Forecasting Turbulence within the Atmospheric Boundary Layer in support of Unmanned Aerial System Operations

A key aspect to successfully operate emerging unmanned aerial systems (UAS) is to be able to accurately forecast the intensity of relevant scales of turbulence eddies within the lowest few hundred meters above ground level. However, conventional aviation turbulence forecasts are usually provided at much higher altitudes (9-12km), where the nature of turbulence is dramatically different, and the relevant aviation-scale eddies are much larger. Therefore, new methods to forecast aviation turbulence within the atmospheric boundary layer (ABL) need to be developed to support UAS operations. To understand the limitations of upper-level diagnostics and remove the uncertainties associated with forecast errors, we have performed fine-resolution (20m) idealized large-eddy simulations (LES) of the ABL and used these as ‘truth’ to evaluate candidate low-level turbulence diagnostics. The advantage of LES is that it explicitly resolves the largest turbulent eddies, and this can be used to directly compute turbulence intensity as measured by the energy dissipation rate ε or EDR=ε^1/3.  The LES produced EDR fields are then compared to candidate EDR turbulence diagnostic predictions using mean quantities provided by numerical weather prediction (NWP) models. The best-performing EDR diagnostics are employed to forecast ABL turbulence using the Weather Research and Forecasting (WRF) NWP model at 3 km horizontal grid spacing. Verification against EDR estimates from in situ high-frequency sonic anemometer data collected during the XPIA campaign near Boulder for a one-month period is performed (March 2015), encompassing the first 300 m of the atmosphere with 7 measurement levels. The role of upstream terrain and atmospheric ABL stability on EDR is assessed. Finally, challenges and research needs to provide accurate real-time forecasts of near-surface EDRs in the context of UAS operations will be discussed.