摘要
基于133个滨海湿地土样的全氮(TN)含量和光谱反射率(R′)及其对数(lgR)、对数的一阶微分((lgR)')、倒数(1/R)、倒数的一阶微分((1/R)')、一阶微分(R')、平方根(R)、一阶微分的倒数(1/(R)')变换,采用偏最小二乘回归(PLSR)、随机森林回归(RFR)和支持向量机回归(SVR)3种算法分别建立土壤TN含量估测模型。结果表明:①土壤TN含量与光谱变换形式相关性由高到低为:(1/R)'>R'>(lgR)'>1/R>lgR>1/(R)'>R>R,经光谱变换,土壤TN含量与变换光谱的相关性均高于R,其中与(1/R)'的Pearson相关系数最大为0.746。②PLSR和SVR基于R'、(1/R)'、(lgR)'和1/(R)'变换构建的模型、RFR方法构建的所有模型R^(2)均大于0.732,均可用于滨海湿地土壤TN含量的估算。③基于1/(R)'建立的SVR模型预测精度最高,其R^(2)为0.987,RMSE为0.057 g/kg,MAE为0.050 g/kg,是预测滨海湿地土壤TN含量的最优模型,可为准确获取滨海湿地土壤TN含量提供稳定方法。
Based on total nitrogen(TN)contents,spectral reflectance(R)of and their logarithm(lgR),logarithm first derivative((lgR)'),reciprocal(1/R),reciprocal first derivative((1/R)'),first derivative(R'),square root(R)and reciprocal first derivative(1/(R)')transformations of 133 coastal wetland soil samples,the predicating models of soil TN contents were established by partial least squares regression(PLSR),random forest regression(RFR)and support vector regression(SVR).The results showed that:Correlations between soil TN contents and spectral forms from high to low were:(1/R)'>R'>(lgR)'>1/R>lgR>1/(R)'>R>R.Correlations between soil TN contents and spectral transformations were higher than those of R,and Pearson correlation coefficient of(1/R)'was highest(0.746).R^(2)of all models established by PLSR and SVR based on R',(1/R)',(lgR)'and 1/(R)'transformations and RFR method were greater than 0.732,indicating their applicable for soil TN content estimation,and SVR model based on 1/(R)'had the highest accuracy,with R^(2)of 0.987,RMSE of 0.057 g/kg and MAE of 0.050 g/kg,which was the optimal model for accurately predicting TN content in coastal wetland soil.
作者
张清文
吴风华
宋敬茹
汪金花
张永彬
刘明月
李孟倩
李春景
郝玉峰
满卫东
ZHANG Qingwen;WU Fenghua;SONG Jingru;WANG Jinhua;ZHANG Yongbin;LIU Mingyue;LI Mengqian;LI Chunjing;HAO Yufeng;MAN Weidong(College of Mining Engineering,North China University of Science and Technology,Tangshan,Hebei 063210,China;Tangshan Key Laboratory of Resources and Environmental Remote Sensing,Tangshan,Hebei 063210,China;Hebei Industrial Technology Institute of Mine Ecological Remediation,Tangshan,Hebei 063210,China;Collaborative Innovation Center of Green Development and Ecological Restoration of Mineral Resources,Tangshan,Hebei 063210,China;College of Geography and Ocean Sciences,Yanbian University,Yanji,Jilin 133002,China;The 8th Geological Brigade of Hebei Bureau of Geology and Mineral Resource Exploration,Qinhuangdao,Hebei 066001,China)
出处
《土壤》
CAS
CSCD
北大核心
2023年第4期880-886,共7页
Soils
基金
国家自然科学基金项目(41901375,42101393)
河北省自然科学基金项目(D2022209005)资助。
关键词
光谱变换
土壤全氮含量
偏最小二乘回归
随机森林回归
支持向量机回归
Spectral transformation
Soil total nitrogen content
Partial least squares regression
Random forest regression
Support vector regression
