The Lund-Potsdam-Jena Dynamic Global Vegetation Model (LPJ), as one of the Dynamic Global Vegetaion Model (DGVM), was developed by Lund University,Potsdam Climate Research Centre and Max-Planck-Institute for Biogeochemistry, Jena. The climatic parameters, soil texture and CO2 concentration data were used as the inputs to simulate carbon and water exchange process between ecosystem and environment. In order to use daily input data, the LPJ code was rewritten in Visual Basic based on the original Fortran version, which was downloaded from the website of PIK. Yingke Oasis Station is located in a corn field (100°25′E, 38°51′N, 1 519 m) and 8 km far from Zhangye City, which belongs to a typical oasis cropland ecosystem in the inland river basin of the arid regions of China. Its observation items include wind speed, air temperature and humidity (3 m & 10 m), wind direction, air pressure precipitation, four components of radiation, land surface temperature, soil temperature and moisture profile (10 cm, 20 cm, 40 cm, 80 cm, 120 cm, and 160 cm), and heat flux (5 cm & 15 cm) and Eddy Covariance. Some items of these observations were used as the inputs of LPJ model to simulate evapotranspiration (ET) and NEE of corn field in Yingke. The latent heat flux and CO2 flux data observed from Eddy Covariance system were used to validate the simulated results. The results indicate that the LPJ model can better simulate water and carbon exchange between vegetation and environment. The R square of ET simulated by LPJ and observed by Eddy Covariance in Yingke is close to 0.8. The ET was also observed by the Micro-Lysimeter, which was also closed to the LPJ simulated results. The R square of NEE simulated by LPJ and observed by Eddy Covariance can reach 0.79.