Probabilistic estimation of the Dynamic Wake Meandering model parameters using SpinnerLidar-derived wake characteristics

Abstract. We study the calibration of Dynamic Wake Meandering (DWM) model using high-spatial- and high-temporal-resolution SpinnerLidar measurements wake field collected at Scaled Wind Farm Technology (SWiFT) facility located in Lubbock, Texas, USA. derive two-dimensional flow characteristics including deficit, turbulence, meandering from lidar observations under different atmospheric stability conditions, inflow wind speeds, downstream distances up to five rotor diameters. then apply Bayesian inference obtain a probabilistic DWM model, where resulting joint distribution parameters allows for both implementation uncertainty assessment. validate fully resolved predictions against discuss most critical sources uncertainty. The results indicate that can accurately predict mean velocity turbulence fields far-wake region beyond four diameters as long properly calibrated are used, time series replicated. show current IEC standard lead conservative deficit ambient intensities above 12 % SWiFT site. Finally, we provide practical recommendations reliable procedures.