首页 | 本学科首页   官方微博 | 高级检索  
     

可见-近红外光谱的土壤养分快速检测
引用本文:李雪莹,范萍萍,侯广利,吕美蓉,王茜,刘岩. 可见-近红外光谱的土壤养分快速检测[J]. 光谱学与光谱分析, 2017, 37(11). DOI: 10.3964/j.issn.1000-0593(2017)11-3562-05
作者姓名:李雪莹  范萍萍  侯广利  吕美蓉  王茜  刘岩
作者单位:山东省科学院海洋仪器仪表研究所,山东 青岛,266061
基金项目:山东省重点研发计划项目,国家自然科学基金项目,山东省自然科学基金项目
摘    要:
在测定土壤养分中,可见-近红外光谱技术具有很大的应用空间。该研究探讨了基于可见-近红外光谱(250~950nm)离线、快速测定土壤总氮(TN)、总磷(TP)、总钾(TK)、总碳(TC)等土壤养分的方法及应用。采集青岛三个不同地区土壤样品(异质性较高的山地土壤与河畔土壤)各60份,总计180份,并测定其TN,TP,TK,TC含量及其可见-近红外反射光谱,利用Kennard-Stone法按2∶1比例划分校正集和检验集,采用遗传算法分别提取TN,TP,TK,TC特征波长,以偏最小二乘法建立定量分析模型。TN,TP,TK,TC含量所建光谱模型的相关系数分别为0.970,0.964,0.680和0.967,检验集的相关系数分别为0.980,0.937,0.717和0.972,检验集的RPD值分别为4.570,2.424,1.411和4.135。结果表明,该方法能够对土壤TN,TP,TC含量进行精确预测,对土壤TK含量进行粗略预测。该研究主要依靠可见光波段,较好的预测了异质性较高的土壤的氮磷钾等养分含量,有望降低未来土壤养分速测的成本。此外,该研究还提供了青岛土壤养分的光谱库,为我国土壤大数据库的建立提供技术支撑。

关 键 词:可见-近红外光谱  土壤养分  遗传算法  快速检测

Rapid Detection of Soil Nutrients Based on Visible and Near Infrared Spectroscopy
LI Xue-ying,FAN Ping-ping,HOU Guang-li,L,#; Mei-rong,WANG Qian,LIU Yan. Rapid Detection of Soil Nutrients Based on Visible and Near Infrared Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2017, 37(11). DOI: 10.3964/j.issn.1000-0593(2017)11-3562-05
Authors:LI Xue-ying,FAN Ping-ping,HOU Guang-li,L&#   Mei-rong,WANG Qian,LIU Yan
Affiliation:LI Xue-ying,FAN Ping-ping,HOU Guang-li,Lü Mei-rong,WANG Qian,LIU Yan
Abstract:
Visible and near infrared spectroscopy analysis technique has many advantages,including simultaneous determination of multiple parameters,non-destructive assay,remote measurement and real-time analysis.And it is simple,fast,low-cost,and needs no sample preparation or simple preprocessing.It will be the preferred technology for soil nutrients determination in fu-ture.This research discussed the method and application of visible and near infrared spectroscopy for the off-line and rapid deter-mination of total nitrogen (TN),total phosphorus (TP),total potassium (TK)and total carbon (TC)in soil.In this paper,60 soil samples from three different and highly heterogeneous regions (2 mountains and 1 riverside)of Qingdao were collected re-spectively.The concentrations of TN,TP,TK,and TC in soils and their visible and near infrared reflectance spectroscopy were determined.First,the calibration set and the test set were divided as 2:1 by Kennard-Stone method.Second,the typical wave-lengths were selected by genetic algorithm.Finally,the quantitative analysis model between soil nutrients and soil spectroscopy was established by partial least square method.The correlation coefficients of the calibration set for TN,TP,TK,and TC were 0.970,0.964,0.680 and 0.967,respectively,and the correlation coefficients of the test set for TN,TP,TK,and TC were 0.980,0.937,0.717 and 0.972,respectively.RPD values for TN,TP,TK,and TC were 4.570,2.424,1.411 and 4.135, respectively.These results showed that the method could accurately predict the content of TN,TC and TP in soils and roughly predict the content of TK in soils.This study mainly used the visible spectroscopy to predict the concentrations of N,P and K in heterogeneous soils,which could reduce the costs for rapid determination of soil nutrients.This study also provided soil nutrient spectroscopy of Qingdao soils,which would support the setup of Chinese large soil database.
Keywords:Visible and near infrared reflectance spectroscopy  Soil nutrient  Genetic algorithm  Rapid detection
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号