Wind-Ramp-Forecast Sensitivity to Closure Parameters in a Boundary-Layer Parametrization Scheme

• Published in: Boundary-layer meteorology Vol. 164; no. 3; pp. 475 - 490
• Main Authors: Jahn, David E., Takle, Eugene S., Gallus, William A.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.09.2017; Springer; Springer Nature B.V
• Subjects: Analysis; Atmospheric boundary layer; Atmospheric Protection/Air Quality Control/Air Pollution; Atmospheric Sciences; Boundary layer parameters; Boundary layer stability; Boundary layers; Computational fluid dynamics; Dissipation; Earth and Environmental Science; Earth Sciences; Electric power generation; Electric utilities; energy; Energy (Physics); equations; Error detection; Failures; Industry forecasts; Kinetic energy; Mathematical models; Meteorology; mixing; Motivation; Numerical models; Parameter sensitivity; Parameterization; Parameters; power generation; Ramps; Research Article; Stable boundary layer; Tests; Turbulence; Turbulent flow; Turbulent mixing; Weather forecasting; Wind power generation; Wind speed; Boundary-layer parametrization; Wind-speed forecasts; Wind ramps
• ISSN: 0006-8314; 1573-1472
• DOI: 10.1007/s10546-017-0250-5

Wind ramps are relatively large changes in wind speed over a period of a few hours and present a challenge for electric utilities to balance power generation and load. Failures of boundary-layer parametrization schemes to represent physical processes limit the ability of numerical models to forecast wind ramps, especially in a stable boundary layer. Herein, the eight “closure parameters” of a widely used boundary-layer parameterization scheme are subject to sensitivity tests for a set of wind-ramp cases. A marked sensitivity of forecast wind speed to closure-parameter values is observed primarily for three parameters that influence in the closure equations the depth of turbulent mixing, dissipation, and the transfer of kinetic energy from the mean to the turbulent flow. Reducing the value of these parameters independently by 25% or by 50% reduces the overall average in forecast wind-speed errors by at least 24% for the first two parameters and increases average forecast error by at least 63% for the third parameter. Doubling any of these three parameters increases average forecast error by at least 67%. Such forecast sensitivity to closure parameter values provides motivation to explore alternative values in the context of a stable boundary layer.