Effect Mechanism of Soil Minerlas on Spectral Characterisitics of Main Soil Classes in Songnen Plain
LIU Huan-jun1,2, WANG Xiang1, LI Hou-xuan1, MENG Xiang-tian1, JIANG Bai-wen1*, ZHANG Xin-le1, YU Zi-yang1
1. College of Resources and Environment, Northeast Agricultural University, Harbin 150030, China
2. Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun 130012, China
摘要： 尽管纯矿物的反射光谱特征分析与数据库建设工作已经开展，但土壤中各原生矿物、粘土矿物的测试主要是定性的，即能测定土壤中含有何种矿物，但难以测定准确的矿物组分含量。土壤矿物是土壤学与地质学的交叉点，易被忽视，特别是已有研究忽略了土壤矿物对土壤反射光谱曲线的影响。探讨了土壤矿物在可见光-近红外光谱部分(400~2 500 nm)对土壤反射光谱特征的影响，明确影响土壤反射光谱特征的主要机理。土壤样本于2014年采集于松嫩平原黑龙江部分，包括4个土类和7个土属，共54个土壤样本。土壤样本通过研磨、过筛后，在室内暗室中测得反射光谱数据，土壤矿物的反射光谱数据在2017美国地质调查局(USGS)最新矿物光谱库Spectral Library Version 7中获得，对反射光谱数据进行九点平滑、10 nm重采样和去包络线处理。土壤矿物含量测试采用荷兰Philip X’ Pert Pro 型X射线衍射仪分析样品的矿物组成，测试了土壤中石英、长石、方解石和闪石等原生矿物和蒙脱石、伊利石和高岭石等粘土矿物的含量。首先分析7个土属的反射光谱特征，明确每个土属反射光谱曲线的形状特征和吸收位置，其次分析土属的矿物含量情况，找出不同土属各矿物含量的共性和差异；再次分析不同粘土矿物和原生矿物的反射光谱特征，确定不同土壤矿物反射光谱曲线的形状特征和特征吸收的位置；最后将不同土属的反射光谱特征、不同土属的矿物含量情况和土壤矿物的反射光谱特征结合，得到如下结论：(1)土壤矿物决定了土壤反射光谱的骨架特征，土壤矿物对土属的反射光谱影响最明显，由于土类存在多种反射光谱特征，土壤矿物对土类的影响不明显。(2)粘土矿物对土壤反射光谱特征的影响大于原生矿物，主要受蒙脱石和伊利石等粘土矿物的影响，但砂性土受部分原生矿物的影响，主要是长石类矿物和高岭石的影响。(3)蒙脱石和伊利石分别决定土壤反射光谱的第一个吸收谷和第二个吸收谷特征，高岭石决定1 400和1 900 nm前的两个小吸收谷特征，钾长石和钠长石决定了砂性土的前两个吸收谷特征。(4)蒙脱石含量足够高时，会完全掩盖高岭石和长石类的反射光谱特征，部分掩盖伊利石的反射光谱特征；随着蒙脱石含量降低，伊利石的反射光谱特征逐渐体现；蒙脱石和伊利石的含量降到很低时，高岭石和长石类矿物的反射光谱特征逐渐体现出来。研究结果揭示了不同土属反射光谱特征差异的原因，可以为土壤反射光谱分类、土壤精细制图和基于高光谱图像的矿物分布研究等提供理论依据。
关键词：高光谱；粘土矿物；土属；去包络线
Abstract：Although the analysis of the reflectance spectral characteristics of pure minerals and the establishment of a database have been carried out, the test of primary minerals and clay minerals in soils is mainly qualitative, which means that it is possible to determine what minerals are contained in soils. However, it is difficult to accurately determine the contents of minerals. Soil minerals are the intersection of pedology and geology, which is easy to be ignored, especially the effects of soil minerals on soil reflectance spectral curves has been neglected in other researches. In this paper, we discussed the effects of soil minerals on the characteristics of soil reflectance spectra in the visible and near infrared region (400~2 500 nm), and clarified the main mechanism about affecting the characteristics of soil reflectance spectra. Soil samples were collected in the Heilongjiang part of Songnen Plain in 2014, including four great groups and seven genera, a total of 54 soil samples. After grinding and sifting, the soil samples were measured in the darkroom to obtain reflectance spectrum data. The reflectance spectrum data of soil minerals were obtained from the 2017 USGS mineral spectral library. We smoothed spectral reflectance data with nine points, resampled at 10-nm intervals and continuum removal. The mineralogical phases of the samples were detected by an X’Pert-Pro XRD (Philip, Holland). The contents of primary minerals, such as quartz, feldspar, calcite and amphibole, and clay minerals such as montmorillonite, illite and kaolinite were measured. First of all, we analyze the reflectance spectral characteristics of seven genera, and determine the shape characteristics and absorption position of spectral curves of each genus. Secondly, we analyze the mineral content of genera, and find out the commonness and difference of each mineral content of different genera. Thirdly, we analyze reflectance spectral characteristics of different primary minerals and clay minerals, and determine the shape characteristics and absorption position of different soil minerals. Finally, we combine the spectral characteristics of different genera, mineral content of different genera and spectral characteristics of soil minerals, the following conclusions are obtained: (1) The skeleton characteristics of soil reflectance spectra are determined by soil minerals, and the effect of soil minerals on reflectance spectra of genera is the most obvious, however, the effect at great group level isn’t obvious due to the existence of various reflectance spectral characteristics of great group. (2) The effect of clay minerals on soil reflectance spectral characteristics is greater than primary minerals, mainly by clay minerals such as montmorillonite and illite, but feldspar and kaolinite affect sandy soils. (3) Montmorillonite determines the characteristics of the first absorption valley, the illite determines the second valley, kaolinite determines the two small absorption valleys before 1 400 and 1 900 nm, and microcline and albite determine the first and second valleys of sandy soils. (4) When the content of montmorillonite is high enough, the spectral characteristics of kaolinite and feldspar will be completely masked, and the spectral characteristics of illite will be partially masked. With the decrease of montmorillonite content, the spectral characteristics of illite will be gradually reflected. When the content of montmorillonite and illite decreases to a very low level, the spectral characteristics of kaolinite and feldspar minerals gradually manifest. The results explain the reasons for the differences in spectral characteristics of different genera, which can provide theoretical basis for soil spectral classification, detailed soil mapping and mineral distribution based on hyperspectral images.
刘焕军，王 翔，李厚萱，孟祥添，姜佰文，张新乐，于滋洋. 土壤矿物对松嫩平原主要土壤类型反射光谱特征的影响机理[J]. 光谱学与光谱分析, 2018, 38(10): 3238-3244.
LIU Huan-jun, WANG Xiang, LI Hou-xuan, MENG Xiang-tian, JIANG Bai-wen, ZHANG Xin-le, YU Zi-yang. Effect Mechanism of Soil Minerlas on Spectral Characterisitics of Main Soil Classes in Songnen Plain. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(10): 3238-3244.
[1] Rossel R A V, Cattle S R, Ortega A, et al. Geoderma, 2009, 150(3-4): 253.
[2] DU Chang-wen, ZHOU Gui-qin, DENG Jing, et al(杜昌文, 周桂勤, 邓 晶, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2009, 40(s1): 154.
[3] GAO Peng, FU Tong-gang, WANG Ke-lin, et al(高 鹏, 付同刚, 王克林, 等). Chinese Journal of Applied Ecology(应用生态学报), 2013, 24(11): 3179.
[4] ZHANG Xiao-meng, ZHANG Xun-yu, CHEN Yuan, et al(张小梦, 张迅与, 陈 媛, 等). Earth Science Frontiers(地学前缘), 2018, 25(1): 314.
[5] PENG Jie, ZHOU Qing, ZHANG Yang-zhu, et al(彭 杰, 周 清, 张杨珠, 等). Acta Pedologica Sinica(土壤学报), 2013 , 50(3): 517.
[6] ZHU Ya-xing, YU Lei, HONG Yong-sheng, et al(朱亚星, 于 雷, 洪永胜, 等). Scientia Agricultura Sinica(中国农业科学), 2017, 50(22): 4325.
[7] Zhang B, Song X F, Zhang Y H, et al. Journal of Integrative Agriculture, 2017, 16(1): 229.
[8] Zhang X K, Liu H J, Zhang X L, et al. Geoderma, 2018, 320：12.
[9] Kokaly, Clark, Swayze, et al. USGS Spectral Library Version 7: U. S. Geological Survey Data Series 1035, 2017. 61.
[10] ZHANG Hua-an, ZHU Yong-hao(张华安, 朱永豪). Remote Sensing for Land & Resources(国土资源遥感), 1994, 6(2): 52.
