土壤有机质含量的高光谱特性及其反演

The Hyperspectral Characteristics and Retrieval of Soil Organic Matter Content

目的应用高光谱技术阐释土壤有机质光谱规律及对有机质在土壤中的含量进行定量分析,为土壤肥力测定和评价提供指导。方法利用ASD FieldSpec Pro地物光谱仪在自然环境条件下对不同有机质含量的土壤样本进行光谱测量。通过对获取的土壤样品高光谱反射率进行倒数、导数、对数和标准化比值变换,运用统计单相关方法进行分析。结果确定了511 nm波长为诊断土壤有机质含量的敏感波段,采用450～750 nm可见光波段反射率均值对507～516 nm敏感范围反射率均值进行标准化比值处理后获得的有机质诊断指数(OII)对土壤有机质含量的估算精度较高,它们存在着简单的线性相关关系。结论土壤有机质诊断指数(OII)反演模型为土壤肥力的快速测定提供了新的途径。

[ Objective ] The study aims to search for the hyperspectral bands most sensitive to soil organic matter content and their retrieval model. [Method] In the field, the spectra of soil samples were measured by anASD PRO FR2500 in the 350-2 500 nm regions. Then the hyperspectral reflectance data were transformed to several organic matter identification indices （OII）. [Result] According to the correlation analysis, the 506-517 nm bands were regarded as the best detecting regions for soil organic matter content. Moreover, the accuracy of the retrieval model was the highest using the OII （ R / R 450-750 ） as the independent variable in the linear model. [ Conclusion ] The retrieval model of the organic matter identification index （OII） provides a new approach to the soil fertility monitoring.