Sensitivity of Turbine Height Wind Speeds to Parameters in Planetary Boundary Layer and Surface Layer Schemes in WRF: Extension to Wintertime Conditions

Our previous work explored the sensitivity of simulations completed with the Weather Research and Forecasting (WRF) model over the Pacific Northwest. These runs were completed using two common parametrizations: the Mellor-Yamada-Nakanishi-Ninno (MYNN) Planetary Boundary Layer and the revised MM5 (Fifth generation Pennsylvania State University-National Center for Atmospheric Research Mesoscale Model) Monin-Obukhov schemes. Our earlier analysis, however, was focused on the spring of 2011 and our conclusions could change if wintertime conditions are considered. For this reason, a second set of simulations were completed for February 2011 and an analysis identical to that used in the previous studies has been carried out using the same model configuration and sensitivity analysis framework for a model domain similar to that used by the Department of Energy supported second Wind Forecast Improvement Project (WFIP 2).

Seasonal differences in the model sensitivity are found and are most evident during daytime. The largest differences in the seasonal sensitivities are associated with parameters used to calculate the length scale in the MYNN scheme and to some extent the value of the Prandlt Number (Pr). These differences are more apparent at the Vansycle Ridge and Hanford locations than at two wind farms located within the domain. The differences in the sensitivity can largely be explained by seasonal changes in the static stability over the region. An increased frequency of stable conditions during daytime would be expected in the winter compared to spring or summer. As the most influential parameters are those applied in the calculation of the length scale during stable conditions, one would expect an increase in the daytime sensitivity with the increase in frequency of occurrence of stable conditions during the day. The increased sensitivity of the simulations to Pr seen during over the basin in February is surprising. In the formulation of the MYNN parameterization, the value of Pr impacts the simulated sensible heat flux regardless of the stability. The sensitivity to this parameter, however, could be related to the differences in temperature gradients seen in stable verses unstable conditions. It is also interesting to note that the sensitivity to Pr is relatively large in areas where the sensitivity to turbulence dissipation rate is relatively small.