Resources Science ›› 2017, Vol. 39 ›› Issue (7): 1349-1360.doi: 10.18402/resci.2017.07.12

• Orginal Article • Previous Articles     Next Articles

Three planetary boundary layer parameterization schemes for the preliminary evaluation of near surface wind simulation accuracy over complex terrain

Qingchen MU1,2(), Yongwei WANG1,2(), Kai SHAO3, Kefei WANG1,2, Yaqi GAO1,2   

  1. 1. Yale-NUIST Center on Atmospheric Environment,Nanjing University of Information Science and Technology,Nanjing 210044,China
    2. Nanjing University of Information Science and Technology,Atmospheric Physics Institute,Nanjing 210044,China
    3. Goldwind Science and Technology Co. Ltd.,Beijing 100176,China
  • Received:2016-12-10 Revised:2017-05-05 Online:2017-07-20 Published:2017-07-20


China is rich in wind energy resources and its wind power development industry has gradually matured. The accuracy of wind speed prediction remains a key problem for the sector. Planetary boundary layer parameterization schemes for numerical models are very important to small scale wind speed forecasting over complex terrain because of rational calculating the atmospheric turbulence characteristics caused by surface thermodynamics and dynamic forces. Here we selected three kinds of planetary boundary layer schemes(MYJ,YSU and ACM2)to simulate Guizhou Jiucaiping in April 2010 for near-ground wind velocity at mountain underlying conditions of complex topography to test simulation performance. We found that under complex mountainous terrain in southwest China the ACM2 scheme simulation of nearly instantaneous wind speed is better than the two other schemes at a height of 70m,due to the calculation mechanism contains the local closure of large scale vortex effect on movement. The simulation results of ACM2 and observations one by one hour root mean square error is 3.56ms-1and the consistency index is 0.94 in April 2010. The wind direction simulation results of the three scheme were very similar,but as height increased the error between simulations and observations of wind direction decreased. The simulation results for wind direction at a height of 70m is better than the result at 10 m. In order to estimate average wind energy density we used the Weibull probability density function to calculate wind energy. Results show that average wind energy density varies at different heights,but at 50m and 70m (the general wind turbine height)the ACM2 scheme performs better.

Key words: WRF model, simulation of wind velocity, PBL schemes, mountain topography