Spatial pattern of grassland soil organic matter based on TM image in the Maduo County

Soil organic matter is not only an important part of the soil, but also is one of the core indices of soil fertility. The proper bands of the TM image correlating with the content of surface soil organic matter were selected by analyzing the quantitative relationship between field data and the digital number (DN) values of the bands of TM image, and then a remote sensing retrieval model was established to simulate the surface soil organic matter in the Maduo County. The results of this study showed that surface soil organic matter had a significant negative correlation with the DN values of TM3 and TM4 (r=-0.670 and -0.675), and an inverse cubic order polynomial regression with the DN values of TM3 and TM5 (R2=0.813, P0.001). Remote sensing retrieval model could make a reliable prediction of the spatial distribution pattern of surface soil organic matter in the Maduo County. Across the Maduo County, the surface soil organic matter content was low, in which the north was significant lower than that the southern. During the past 15 years, the surface soil organic matter content increased in the most part of Maduo County. But the surface soil organic matter degradation was severe especially in the areas along traffic artery and close to settlement with frequent human activities.