新疆北部不同类型土壤光谱特征及对有机质含量的预测

对北疆地区淡栗钙土、冷钙土、石灰性黑钙土、石膏灰棕漠土等4种土壤类型的反射光谱进行分析，研究土壤有机质含量与光谱反射率之间的关系。结果表明，石灰性黑钙土的反射率明显低于其它土壤类型。有机质含量高于14 g·kg^-1时光谱反射率与有机质含量呈负相关，有机质含量很低（<8 g·kg^-1）时，土壤的光谱反射率与有机质含量之间呈正相关。分别采用593.6 nm波段的原始光谱反射率、661 nm波段的反射率去除包络线和547.4 nm波段反射率的一阶微分与土壤有机质含量建立回归模型，经检验三种模型均能较好地预测有机质的含量，其中光谱的一阶微分预测精度相对较高，可较好地预测北疆主要类型土壤的有机质含量。

Hyperspectral characteristics and estimation of organic matter contents of different soil types in North Xinjiang

The hyperspectral characteristics of four soil types (Light chesthnut soil, Cold calcic soil, Calcic chernozem, Gypsum gray brown desert soil) in North Xinjiang and their relationship with soil organic matter content (SOM) were studied. The results showed that the spectral reflectance of Calcic chernozem was significantly lower than other soil types. Negative correlation between the spectral reflectance and SOM was detected as SOM was higher than 14 g·kg^-1. By contrast, positive correlation existed with low SOM (<8 g·kg^-1). Correlation analysis of spectrum reflectance with SOM indicated that SOM correlated with the spectrum reflectance in 593.6 nm band, i n 661 nm band by removal of the envelope of reflectance, and in 547.4 nm band through first order differential of reflectance. In addition, the prediction accuracy of the first order differential was relatively high, and thus can be used in the estimation SOM of mainly soil types in North Xinjiang.