Based on soil data in 1981 and 2011, and AVHRR and MODIS remote sensing data, models for predicting the changes of the topsoil organic matter and total nitrogen were built by studying the relationships between soil organic matter, total nitrogen and these remote sensing parameters, such as normalized difference vegetation index (NDVI), vegetation coverage and net primary productivity (NPP). The study showed that soil organic matter and total nitrogen in the study area had changed significantly during the past 30 years with an increase of 5.43 g/kg and 0.21 g/kg respectively, while the three remote sensing parameters (NDVI, vegetation coverage and NPP) showed a gentle rising trend between 1981 and 2011. Positive correlation was found between the changes of remote sensing parameters and the change of soil organic matter and total nitrogen. A model was built to predict the change of soil organic matter content and the change of total nitrogen respectively by stepwise multiple linear regression analysis by using those parameters. The models can explain 16.9% and 20.3% of their changes respectively. Then the models were employed to map the spatial variations of the changes of soil organic matter content and total nitrogen content in the cultivated land of the region, the results showed that soil organic matter and total nitrogen had an increase of 6.65 g/kg and 0.31 g/kg in the cultivated land. The validation results showed that the models could well reflect the spatial distribution features of changes of soil organic matter and total nitrogen content in the study area.