Wind ramp events at an Iowa wind farm: a climatology and evaluation of WRF ensemble forecast skill

Rapid changes in wind speed that lead to extreme changes in wind power output, known as wind ramp events, cause considerable problems for the wind energy industry. Accurate forecasts of the events could be of great benefit to the industry, but because these events occur over short temporal scales, forecasting of them can be difficult. Using 80 m wind data from a meteorological tower at the Pomeroy wind farm in northwestern Iowa, we have classified nearly 300 wind ramp events occurring during a sample of over 100 cases for which we have run an 8 member ensemble of WRF members. An event was considered to be a ramp event if the change in wind power was 50% or more of total capacity in four hours or less, and this was approximated using a typical wind turbine power curve such that any wind speed increase or decrease of more than 3 m/s within the 6-12 m/s window (where power production varies greatly) in four hours or less would be considered a ramp. In addition, ramp down events occurring due to high wind speed shutdowns were also documented. The WRF simulations available during the ramp events used 10 km grid spacing and varied planetary boundary layer schemes with either 00 UTC NAM or 00 UTC GFS output used for initial and lateral boundary conditions. For some cases, additional runs were available with staggered initialization times. We will present a climatology of the wind ramp events, examining their timing and causes. In addition, we will examine model skill at predicting wind ramp events. We will not only examine model forecasts around the times of observed wind ramp events, but also analyze model forecasts to look for false alarms in the models.