新疆玛纳斯河流域不同水分含量盐渍化土壤光谱信息特征研究

Spectral Information Feature of Saline Soils with Different Moisture Contents at Xinjiang Manas River Area

以新疆绿洲玛纳斯地区典型盐渍土为研究对象,运用统计学和可见光—近红外光谱分析方法对不同水分含量盐渍化土壤进行野外和室内测定,探讨了盐渍化土壤的反射光谱特征与土壤含水量的关系。试验结果表明:盐渍土原始光谱反射率与土壤水分含量变化相关性不高,通过对原始光谱反射率的一阶微分、连续统去除处理后与含水量的相关性显著增强,试验得出盐渍化土壤水分含量变化敏感的特征波段为:450 nm、1320 nm、1416 nm、1650 nm、1810 nm、1910 nm、2110nm、2202 nm附近;经检验得出盐渍土含水量的多元线性回归预测模型为y=1.436 x1416 nm-1.4416 x1810 nm-1.152 x1910 nm+0.934 x2202 nm+0.138(R2=0.990*,P=0.019),研究为水盐动态条件下的盐渍化土壤大面积遥感监测提供理论依据。

The objective of this research was to study the relationship between spectral reflectance characteristics and soil water content for typical salinized soil of oasis in Xinjiang Manas, which was measured in field and laboratory with using statistics and visible llght-near Infrared analysis. The results showed that the correlations rang about original spectral reflectance with moisture content was not higher, while the first order differential and continuum removal correlation markedly enhanced. Test got the sensitive characteristics of saline soil moisture content： 450 nm, 1320 nm, 1416 nm, 1650 nm, 1810 nm, 1910 nm, 2110 nm and 2202 nm nearby. The multiple linear regressions forecast model about saline soil water content is： Y=1.436x1416nm-1.4416x1810nm-1.152x1910nm＋0.934x2202nm＋0.138（R^20.990＊,P=0.019）, which provides a theoretical basis for large area remote sensing monitoring under the dynamic conditions of the water and salt.