Regional wind energy potentials are defined by the kinetic energy budget of the atmospheric boundary layer

Abstract:

Bottom-up estimates of regional wind energy potentials assume that higher wind speeds, larger turbine capacities, and increasing number of turbines lead to greater generation. Better turbine technology and larger wind farms are thus expected to increase capacity factors and generation. However, the increased extraction of kinetic energy is not expected to reduce regional wind speeds. Here we show that at capacity densities envisioned by such estimates (3-10 MW_i km^-2), the limit of generation is defined by the kinetic energy budget of the atmospheric boundary layer (ABL) and mixing within it. Mesoscale simulations (WRF) of hypothetical large-scale wind turbine deployment (~ 10⁵ km²) in Kansas show that higher wind speeds at night do not lead to higher generation and greater capacity densities lead to lower capacity factors, wind speeds and generation. We evaluated the influence of diurnal variation in ABL heights on generation and wind speed reductions, using a conservation of energy-based approach known as Kinetic Energy Budget of the Atmosphere (KEBA). A higher (lower) boundary layer height during the day (night) implies a larger (smaller) energy budget and better (poorer) vertical energy replenishment; and thus greater (lesser) generation. Across a range of capacity densities, daytime generation and mean wind speed reductions modelled by KEBA agreed well with WRF. At night, KEBA overestimated generation by ~50% and underestimated mean wind speed reduction by ~20% because it assumes a well-mixed convective boundary layer condition. However, nighttime boundary layers are stable and stratified. This impedes the downward mixing of energy from the free atmosphere leading to lower regional nighttime generation, despite higher wind speeds. Both methods predicted that day and night-time capacity factors reduce to <10% for the 10 MW_i km^-2 case. Relative to WRF, the bottom-up approach overestimates both day and nighttime generation by 30-180% and 140-600%, respectively. We conclude that regional wind energy potentials are set by the kinetic energy budget of the boundary layer rather than just wind speeds and turbine/wind farm characteristics. Our results imply that diurnal variations in boundary layer must be considered, along with feed backs on wind speeds, for robust estimates of regional wind energy potentials.