Near Surface Atmospheric Turbulence Parameters For Solving The Method Of Comparison And Its Application

Two different coordinate rotation methods including the triple rotation method and the planar fit method used to correct the eddy covariance are presented in term of the data collected by the eddy covariance system at the PBL site on a grass land in Anhui province of china ,during period from June 11 to July 20,2002 during '973' experiment. Furthermore, Fluxes obtained from flux profile method are compared to those derived from the eddy covariance method. Results show that effect for advancing the parameterization schemes of different land processes and helping improving the eddy covariance method presion . The main results are as follows:The triple rotation method depends on single average vertical velocity and horizontal velocity speed. Every single rotation angle is different ,so it can't be considered as the true tilt angle. The average rotation angle in the x-z plane is -1.52 and in the y-z plane is -0.44 during our experiments. The planar fit method utilizes many runs data to get =-1.53 , =-0.40 ,which is close to the angle by the triple rotation method and can be considered as the true tilt angle. The composed velocity corrected by the two methods are almost same ;but the vertical velocity corrected by the two methods are different, and the vertical velocity corrected by triple methods is zero and corrected by planar fit method is close to zero but unequal to zero. It reflects the true vertical velocity after correction by planar fit method. The friction velocity and the heat fluxes corrected by triple rotation method are 92% and 95% of the corresponding result without correction individually, and those derived from Planar fit method are 98% of the corresponding result without correction. The heat flux and the friction velocity are smaller than the corresponding results without correction because the falsity patition caused by horizonal velocity are eliminiated by the two coordinate rotation methods .The relative difference of friction velocity corrected by the two methods is investigated . The result shows that stress correction error caused by triple rotation method increases with decreasing of velocity because of the mean velocity and the lateral stress sampling errors .The flux obtained from the COARE scheme are about 95% of those derived from eddy covariance without correction ,and is more close to those derived from eddy covariance with correction . Comparing with the results obtained from the eddy covariance , we find that Dyer-Hick flux-profile relationship (DH scheme in short ) gives a realistic estimation of flux. Businger flux-profile relationship ( BWIB scheme in short ) underestimates and Pruitt flux-profile relationship ( PML scheme in short ) overestimates. The results show that the relative difference between different scheme is seriously affected by atmosphere stability. The relative difference in neutral contidion is lowest and increases with stability or instability change , and the relative difference in instability condition is lower than that in stability. The difference between the flux-profile relationships is mostly affected by the Von Karman constant k and Prandtl number PtN.The results gived by three different flux-profile relationship is more close to eddy covariance when three flux-profile relationship are changed to the form with k equaling 0.4 and PtN equaling 0.95. The difference between the flux-profile relationships is lower as well. We conclude that the flux-profile relationships with k equaling 0.4 and PtN equaling 0.95 is creditable ,which is consistent with the conclusion from the forerunner.