若尔盖湿地土壤重金属元素含量的遥感反演  

Inversion on Contents of Heavy Metals in Soils of Wetlands in Zoigê Plateau based on Remote Sensing Data

作　　者：袁中强[1] 曹春香[1,2] 鲍达明[3] 陈伟 田蓉[2] 俄尕 李华 

机构地区：[1]电子科技大学资源与环境学院,四川成都611731 [2]遥感科学国家重点实验室,中国科学院遥感与数字地球研究所,北京100101 [3]国家林业局湿地保护管理中心,北京100714 [4]四川省若尔盖湿地国家级自然保护区保护管理局,四川阿坝624500

出　　处：《湿地科学》2016年第1期113-116,共4页Wetland Science

基　　金：科技基础性工作专项项目(2013FY1118063);国家自然科学基金项目(41171330);林业公益性行业科研专项项目(201504323)资助

摘　　要：土壤中的重金属元素对植物和动物具有一定的毒害作用,测量湿地土壤重金属元素含量,对评估湿地生态系统健康具有重要意义。以光谱分辨率的优势,高光谱遥感可以获取地物的连续光谱信息,可应用于土壤研究。以四川若尔盖国际重要湿地为研究区,在研究区中采集土样,测定土样的5种重金属元素含量;利用偏最小二乘回归(partial least squares regression,PLSR)方法,分析了环境一号卫星上搭载的超光谱成像仪(hyperspectral imaging radiometer,HSI)的地表反射率数据与实测的土壤重金属元素含量的关系;利用原始地表反射率及原始地表反射率曲线的一阶微分值和倒数的对数值,分别建立了预测土壤中5种重金属含量的3个回归模型。结果表明,土壤锌含量的平均值(62.83 mg/kg)最高,镉含量的平均值(0.18 mg/kg)最小;土壤锌含量与铅、铬含量显著相关,土壤铬含量与铜、铅含量显著相关;在3个回归模型中,利用一阶微分值所建立的模型的预测效果最好,土壤5种重金属含量的实测值与预测值都显著相关,其次是利用原始反射率所建立的模型,而利用倒数的对数值所建立的模型的预测结果相对最差,尽管如此,除了镉和铅含量外,其它重金属元素含量的实测值与预测值也都显著相关。利用高光谱遥感技术反演土壤某些重金属元素含量具有可行性,有利于实时掌握湿地土壤重金属含量及其对湿地生态系统健康的影响。Hyperspectral remote sensing technology could be used to obtain continuous spectral features of terrestrial objects, benefit from its advantage of high spectral resolution. In this paper, taking the wetlands in Zoigê Plateau as study area, the contents of 5 kinds of heavy metal elements in the soil of the wetlands were measured, based on which, the relationship between contents of heavy metals and the surface reflectance, first order differential reflectance and logarithm of reciprocal reflectance from HJ-A HSI data were analyzed, and the regression models were established by partial least squares regression（PLSR） method. The results showed that the mean content of zinc（62.83 mg/kg） in the soil was the highest and that of cadmium（0.18 mg/kg） was the lowest. The correlations between zinc content and lead content, and chromium content were both significant. For the 3 regression models, that using first order differential reflectance performed best with significant correlations between the measured and predicted values of all 5 heavy metal contents. The regression model based on the logarithm of reciprocal reflectance gave the worst performances. However, except cadmium and lead contents, significant correlations could still be achieved for the other heavy metal elements. This study demonstrated that it is feasible to obtain the soil heavy metal content using hyperspectral remote sensing technology. It is conducive to acquire the dynamics of soil heavy metal contents of the wetlands and their impacts on the wetland ecosystem health.

关 键 词：高光谱遥感 环境卫星数据 土壤重金属含量 湿地 偏最小二乘回归法 

分 类 号：S153.61[农业科学—土壤学] TP79[农业科学—农业基础科学]