摘要
砷(As)是我国多金属矿区的主要污染物之一,对环境、农业和人类健康构成严重威胁。近地高光谱技术具有快速、动态、无损、光谱分辨率高等优势,对于多金属矿区土壤As污染监测与综合治理具有巨大应用潜力。然而,由于受污染区域、土壤背景以及高光谱质量、光谱输入量等因素影响,高光谱反演模型的适用性和精度差异较大。本研究针对湘南某多金属矿区,基于Pearson相关性分析并结合变量投影重要性(VIP)准则,提取18种变换光谱形式下的单变量特征波段及4种光谱指数算法下的优化光谱指数作为光谱输入量,建立偏最小二乘回归(PLSR)模型,实现了矿区土壤As含量反演。结果表明:倒数(RT)、对数(L)、平方根(Sqrt)、标准正态变量变换二阶导(SNV_SD)等变换后的光谱数据与As含量具有较高的相关性;优化光谱指数能从二维光谱空间揭示As的光谱响应,相较于单变量特征波段,以优化光谱指数为自变量构建的模型性能更优;比值指数(RI)模型的R_(c)^(2)、RMSE_(c)、R_(p)^(2)、RMSE_(p)、RPD分别为0.908、50.8 mg/kg、0.949、35.6 mg/kg、4.45,是研究区土壤As含量反演的最优模型。单变量特征波段结合优化光谱指数预测土壤As含量具有较好的可行性,可为多金属矿区土壤As污染高光谱快速监测提供科学依据。
Arsenic(As)is a prominent contaminant within polymetallic mining areas in China,posing substantial threats to the environment,agriculture,and human health.Near-ground hyperspectral technology,characterized by its rapid,dynamic,non-destructive and high spectral resolution,holds significant potential for the monitoring and integrated management of soil arsenic pollution in polymetallic mining areas.However,the applicability and accuracy of hyperspectral inversion models are subject to variations influenced by factors such as contaminated areas,soil background,hyperspectral quality,and spectral inputs.This study focused on a polymetallic mining area in southern Hunan,utilizing Pearson correlation analysis in conjunction with variable projection importance(VIP)criteria,we extracted univariate spectral bands under 18 transformed spectral forms,as well as optimized spectral indices under 4 spectral index algorithms,as spectral input variables.These variables were then utilized to construct a partial least squares regression(PLSR)model to achieve the inversion of soil As content within the mining area.The results show that,there are high correlations between transformed spectral data(reciprocal(RT),logarithmic(L),square root(Sqrt),second derivative of standard normal variables(SNV_SD),etc)and As content.The optimized spectral indices reveal the spectral response characteristics of As in a two-dimensional spectral space,and the PLSR model constructed with the optimised spectral indices has better performance compared to the univariate characteristic bands.The ratio index(RI)model,whose R_(c)^(2),RMSE_(c),R_(p)^(2),RMSE_(p) and RPD are 0.908,50.8 mg/kg,0.949,35.6 mg/kg and 4.45,respectively,emerges as the optimal model for the inversion of soil As content in the study area in this study.The combination of univariate characteristic bands with optimized spectral indices demonstrates favorable feasibility in predicting soil As content,providing a scientific foundation for the rapid monitoring of soil As pollution in polymetallic mining areas.
作者
周瑶
成永生
王丹平
张泽文
曾德兴
李向阳
毛春旺
ZHOU Yao;CHENG Yongsheng;WANG Danping;ZHANG Zewen;ZENG Dexing;LI Xiangyang;MAO Chunwang(Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring,Ministry of Education,Central South University,Changsha 410083,China;Hunan Key Laboratory of Nonferrous Resources and Geological Hazards Exploration,Changsha 410083,China;School of Geosciences and Info-Physics,Central South University,Changsha 410083,China)
出处
《中国有色金属学报》
EI
CAS
CSCD
北大核心
2024年第2期653-667,共15页
The Chinese Journal of Nonferrous Metals
基金
湖南省重点研发计划资助项目(2023SK2006,2022SK2072)
湖南省自然科学基金资助项目(2023JJ50057)
长沙市自然科学基金资助项目(kq2202090)
广西岩溶动力学重大科技创新基地开放课题资助项目(BL202105)。
关键词
土壤重金属
砷
高光谱遥感
光谱变换
优化光谱指数
偏最小二乘回归
soil heavy metals
arsenic
hyperspectralremote sensing
spectral transform
optimised spectral indices
partial least squares regression(PLSR)
