Your Location:  Home Page> Scientific research> Results> Publications>

Analyses of Simulation Result in Loess Plateau by WRF Model with Two Reanalysis Data


Journal

Plateau Meteorology

Authors

Ma Chenchen, Yu Ye, He Jianjun, Chen Xing, Xie Jin

Year

2014

Volume

33

Issue

3

Pages

698-711

Corresponding Author

Yu, Y

Request Full Text

yyu@lzb.ac.cn

Keywords

WRF model; NCEP reanalysis data; ERA-interim reanalysis data; Loess Plateau

Abstract

WRF is one of the most widely used numerical weather prediction models, for which reanalysis data provide initial and lateral boundary conditions. Different reanalysis products may affect WRF simulation results differently. Taking Loess Plateau as research area, two WRF experiments (NCEP/WRF and ERA/WRF), using surface observation data and sounding data, taking NCEP and ERA-Interim reanalysis as initial and boundary conditions, respectively, were conducted to evaluate the influence of different reanalysis products on WRF simulations. The results indicated that both experiments can accurately reproduce the diurnal variation of 2 m temperature, 2 m relative humidity and surface temperature, with ERA/WRF giving better results than NCEP/WRF. Due to the complex terrain in the studied area, model performance for 10 m wind speed was poor. Both experiments cauld approximately simulate the diurnal variation of radiation components and ground heat fluxes. The deviation from observations mainly appeared at noon. Both experiments captured the variation of potential temperature and specific humidity in the atmospheric boundary layer well with correlation coefficient higher than 0.8. However, the performance for wind speed was not as good, with correlation coefficients of 0.63 and 0.60, respectively, for NCEP/WRF and ERA/WRF experiments. NCEP/WRF performed better than ERA/WRF in the atmospheric boundary layer.