Abstract: At present, GDAS atmospheric forcing data is widely used as the inputs in the simulation of water and energy cycles. Besides the problem of the model itself, these forcing data determine the accuracy of the simulation. Although there are obvious uncertainties and low spatial resolution for these data, most of researches still used them to perform the simulation of land surface process because no better way can be used. The purpose of this paper is to test the accuracy of the results simulated by NOAH land surface model taking GDAS atmospheric forcing data as the inputs and gives the objective evaluations, which has important reference value for the users of the data and the model. The test results obtained on the basis of the observation data from Yucheng, Xilinhaote and Changbaishan ecological observation sites showed that the accuracy of shortwave net radiation and total net radiation in GDAS dataset are very good. It is reliable in the applications. Large differences were found between precipitation data in GDAS dataset and those from the three sites; however, there is good consistence between their precipitation times. Considering the effect of scale difference between GDAS and measurement on the validation, and the lack of the spatial precipitation data, this data is also worth of application. In addition, there are also good consistence between the change trend of latent heat flux, sensible heat flux and the top soil water content simulated by NOAH and that by measurements, and the simulated results can truly reflect the change process of surface situation although there are large difference between their absolute values. Poor accuracy was found in simulated soil heat flux.

Key words: accuracy test, China, GDAS data, NOAH