Using continuous flux data, meteorological data and biological data in 2006(from June 1 to August 9) from Jinzhou agricultural ecosystem research station, based on BATS1e model, the sensitivity of land surface model to dynamic assignment of roughness(Z0), leaf area index(LAI) and fractional vegetation coverage (FVEG) and albedo(α) were investigated. The results show that dynamic assignment of Z0 has effect on simulating surface soil temperature(SST) and sensible heat flux (SH) especially on the time when maize field surface covered changes from bare soil to vegetation. Dynamic LAI plays important role in improving the simulation of SST, net absorbed solar energy flux(Frs), SH and surface soil water content(SWC), at the same time affects simulation of latent heat flux (LE). Dynamic FVEG affects obviously simulations of all abovementioned variables and shows greater sensitivity when they are smaller. Also, dynamical change of α can affect simulations of SST, LE and SH, especially for the latter. Furthermore, the interactions among different dynamic parameters are ignored by the model. The improvement of single land surface parameter might be helpful for simulating one or multivariate but not for all variables. In short, it is necessary to set up a parameterization scheme with the interactions of different land surface parameters.