Leaf area index（LAI）is an important vegetation biophysical variable

and has been widely applied to the estimation of crop yield evapotranspiration

and net photosynthesis.Mean while as an important input or output parameter of some dynamic process models such as crop growth model and common land model

LAI is a connecting bridge to couple the process models with remote sensing models.The Moderate Resolution Imaging Spectroradiometer（MODIS）

carried aboard the Terra and Aqua satellites

can collect the information of LAI change continuously.The MODIS LAI standard product

issued since 2000

have been used in the meteorological and hydrologic field.However

spatio-temporal discontinuity exit in the MODIS standard LAI product

which restricts its application in the crop growth monitoring and yield estimation

land surface process simulation and global change research.The paper analyzed the MODIS standard LAI product in China region

and pointed out that the spatio-temporal discontinuity of the MODIS standard LAI product in China region is caused not only by the MODIS LAI reverse algorithm but also by the MODIS reflectance data quality.Based on the TSF filter algorithm a new improved TSF filter algorithm is put forward in this paper

in which more attention is paid to the relatively serious cloud covering in the crops growth season that results in poor quality of reflection data in China region.The improved TSF filter algorithm makes full use of the data quality control information in the MODIS data product and takes further consideration of the influence of surface reflectance data quality on MODIS LAI standard product.Experimental results show that the improved TSF algorithm can better process MODIS LAI standard product.Further

the improved TSF algorithm has been used to produce a new suit of MODIS LAI product in China region with much better spatio-temporal continuity

which is more reasonable and reliable dataset for integrating the remote sensing data product with application models.