水分对土壤近红外光谱检测影响的二维相关光谱解析

为了能进一步分析水分对土壤近红外光谱检测的影响，分别从三个地方(桃园堡、 牧场和农大示范田)采集了三种土壤，经过筛、 烘干后分别配制了含水率为20%, 15%和10%的土壤，然后采用ASD公司的FieldSpec3光谱仪在保证其他检测条件不变的情况下，对不同含水率下不同采集地的土壤进行了检测，获得了其不同含水率下的动态光谱图。 最后视水分为外部干扰，应用二维相关光谱的理论对其进行了分析。 结果表明，在350～2 500 nm波段范围内，三个不同采集地土壤的同步二维相关光谱图比较类似，都在2 210和1 929 nm附近处出现较强的自动峰，在1 415 nm附近处有较弱的自动峰;从三处自相关峰的密集程度来看，1 929 nm处对应的官能团对水分最敏感，2 210 nm处次之，1 415 nm处最不敏感。 该研究明确找出了水分对土壤近红外光谱检测影响的敏感波段和敏感程度，为今后消除水分影响建立抗水分干扰模型提供了依据。

Analysis of the Effect of Moisture on Soil Spectra Detection by Using Two-Dimensional Correlation Near Infrared Spectroscopy

In order to analyze the effects of moisture on soil spectra detection, soil samples from three different places (Taoyuanbao, Muchang, and Shanxi agricultural university) were collected. All soil samples were sieved, dried and prepared with different water content (20%, 15%and 10%). The spectra of soil samples with different water content were collected by ASD FieldSpec3. Finally, moisture was looked as the the external disturbance, and two-dimensional correlation near-infrared spectroscopy (2D-NIR) technique was applied to analyze the effect of moisture on soil spectra detection. 2D-NIR analysis was realized by using shige software (2Dshige shigeaki Morita, Kwansei-Gakuin University).The results showed that, the sensitive bands of moisture in 2D-NIR spectra were more obvious than one-dimensional spectra. In the region 350～2 500 nm, the synchronous spectra of soil samples from different places were similar. Two automatic peaks near 2 210 and 1 929 nm were strong, and one automatic peak near 1 415 nm was weak. Moreover, the peak 1 929 nm was most sensitive to moisture content, next was 2 210 nm, and the last one was 1 415 nm. The sensitive bands and sensitive degree were pointed out in this research. It will be helpful for anti moisture model building.