江汉平原土壤有机碳含量高光谱预测模型优选

卢延年 刘艳芳 陈奕云 姜庆虎

中国农学通报. 2014, 30(26): 127-133

中国农学通报 ›› 2014, Vol. 30 ›› Issue (26) : 127-133. DOI: 10.11924/j.issn.1000-6850.2014-0439
工程 机械 水利 装备

江汉平原土壤有机碳含量高光谱预测模型优选

  • 卢延年 刘艳芳 陈奕云 姜庆虎
作者信息 +

Optimization of the Hyperspectral Prediction Model of Soil Organic Carbon Contents of Jianghan Plain

Author information +
History +

摘要

研究探讨了贫瘠地区低有机碳含量条件下,不同光谱预处理与建模方法用于土壤有机碳估测的最佳组合,对贫瘠土壤属性信息快速获取和精确农业发展具有重要意义。以江汉平原不同利用条件下的土壤为研究对象,使用可见光/近红外高光谱技术,结合包括Savitzky-Golay平滑(SG)、一阶导数(FD)、多元散射校正(MSC)在内的光谱预处理方法,分别建立用于估测土壤有机碳(SOC)含量的多元线性回归(MLR)、主成分回归(PCR)、偏最小二乘回归(PLSR)和支持向量机回归(SVMR)模型。结果表明:不同建模方法预测精度差异明显,PLSR和SVMR的预测结果优于MLR和PCR;不同预处理方法对模型的预测精度亦有较大影响,表现为MSC>FD>SG;基于FD和MSC组合预处理的SVMR模型的预测能力最好,R2=0.84,RPD=2.50,满足土壤有机碳的预测。有机质含量大于2%并不是建立优质模型的必要条件。

Abstract

The study explored the best combination of spectral transformations and modeling methods under the condition of low organic carbon content in barren regions, and is of great significance for precision agriculture. With soil samples of different land use types collected in Jianghan Plain, visible/near- infrared spectroscopy was used in the estimation of soil organic carbon (SOC). Spectral transformations including Savitzky- Golay smooth (SG), the first derivative (FD) and multiple scatter correction (MSC), coupled with multiple linear regression (MLR), principal component regression (PCR), partial least squares regression (PLSR) and support vector machine regression (SVMR) were used for model calibration. Results showed that the PLSR and SVMR models outperform the MLR and PCR models. In terms of spectral transformations: MSC>FD>SG. The SVMR model coupled with FD and MSC outperformed the other models, with R2=0.84, RPD=2.50, and could be used for SOC prediction in the study area. And soil organic matter above 2% was not necessary for building prediction models of high quality.

关键词

高光谱; 土壤有机碳; 光谱预处理; 偏最小二乘回归; 支持向量机回归

Key words

hyperspectra; soil organic carbon; spectral pretreatment; partial least-squares regression; support vector machine regression

引用本文

导出引用
卢延年 刘艳芳 陈奕云 姜庆虎. 江汉平原土壤有机碳含量高光谱预测模型优选. 中国农学通报. 2014, 30(26): 127-133 https://doi.org/10.11924/j.issn.1000-6850.2014-0439
Optimization of the Hyperspectral Prediction Model of Soil Organic Carbon Contents of Jianghan Plain. Chinese Agricultural Science Bulletin. 2014, 30(26): 127-133 https://doi.org/10.11924/j.issn.1000-6850.2014-0439

参考文献

[1] 史学正,于东升,高鹏,等.中国土壤信息系统(SISChina)及其应用基础研究[J].土壤,2007,39(3):329-333.
[2] 田永超,张娟娟,姚霞,等.基于近红外光声光谱的土壤有机碳含量定量建模方法[J].农业工程学报,2012,28(01):145-152.
[3] 沈润平,丁国香,魏国栓,等.基于人工神经网络的土壤有机碳含量高光谱反演[J].土壤学报,2009,46(3):391-397.
[4] 沙晋明,陈鹏程,陈松林.土壤有机碳光谱响应特性研究[J].水土保持研究,2003,10(2):21-24.
[5] 彭杰,周清,张杨珠,等.有机质对土壤光谱特性的影响研究[J].土壤学报,2013,50(3):517-524.
[6] 李硕.基于 Vis/NIR 光谱不同粒径下土壤碳氮的预测研究[D].武汉:华中农业大学,2010:1-3.
[7] 张淑娟,王凤花,张海红,等.基于主成分分析和BP神经网络的土壤养分近红外光谱检测[J].山西农业大学学报:自然科学版,2009,29 (06):483-487.
[8] 于士凯,姚艳敏,王德营,等.基于高光谱的土壤有机质含量反演研究[J].中国农学通报,2013,29(23):146-152.
[9] Vasques GM, Grunwald S, Sickman JO. Comparison of multivariate methods for inferential modeling of soil carbon using visible/near-infraredspectra[J].Geoderma,2008,146(1/2):14-25.
[10] Rinnan A, Berg FVD, Engelsen SB. Review of the most common pre-processing techniques for near-infrared spectra[J].TrAC Trends in Analytical Chemistry,2009,28(10):1201-1222.
[11] Kooistra L, Wanders J, Epema GF, et al. The potential of field spectroscopy for the assessment of sediment properties in river flood plains[J].Analytica Chimica Acta,2003,484(2):189-200.
[12] 王欣.近红外分析中光谱预处理方法的研究与应用进展[J].科技资讯,2013,15:2.
[13] 申艳,张小平,梁爱珍,等.多元散射校正和逐步回归法建立黑土有机碳近红外光谱定量模型[J].农业系统科学与综合研究,2010,26 (02):174-180.
[14] 宋海燕,秦刚,韩小平,等.基于可见光谱的不同质地土壤有机碳快速测定[J].农业机械学报,2012,43(07):69-72.
[15] 解宪丽,孙波,周慧珍等.中国土壤有机碳密度和储量的估算与空间分布分析[J].土壤学报,2004,41(1):35-43.
[16] Savitzky A, Golay MJE. Smoothing and differentiation data by simplified least square procedure [J].Analytical Chemistry,1964,36 (8):1627-1638.
[17] 陈华舟,潘涛,陈洁梅.多元散射校正与Savitzky-Golay平滑模式的组合优选应用于土壤有机碳的近红外光谱分析[J].计算机与应用化学,2011,28(05):518- 522.
[18] 褚小立,袁洪福,陆婉珍.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4):528-542.
[19] 姚艳敏,魏娜,唐鹏钦,等.黑土土壤水分高光谱特征及反演模型[J].农业工程学报,2011,27(8):95-100.
[20] Rossel RAV, Behrens T. Using data mining to model and interpret soil diffuse reflectance spectra[J].Geoderma,2010,158(1/2):46-54.
[21] 王昌昆,潘贤章,周睿,等.应用基于PLSR的土壤-环境模型预测土壤属性[J].土壤学报,2012,49(2):237-245.
[22] Scholkopf B, Burges JC, Smola AJ. Advances in kernel methods support vector learning[M].Cambridge, MA: MIT Press,1999:169- 184.
[23] Shao Y,He Y. Nitrogen, phosphorus, and potassium prediction in soils, using infrared spectroscopy[J].Soil Research,2011,49(2):166- 172.
[24] Chang CW, Laird DA, Mausbach MJ, et al. Near- infrared reflectance spectroscopy- principal components regression analyses of soil properties[J].Soil Science Society of America Journal,2001, 65(2):480-490.
[25] Shi T, Cui L, Wang J, et al. Comparison of multivariate methods for estimating soil total nitrogen with visible/near infrared spectroscopy [J].Plant and Soil,2013,336(1/2):363-375.
[26] 谢伯承,薛绪掌,刘伟东,等.基于包络线法对土壤光谱特征的提取及其分析[J].土壤学报,2005,42(1):171-175.
[27] Gomez C, Lagacherie P, Coulouma G. Continuum removal versus PLSR method for clay and calcium carbonate content estimation from laboratory and airborne hyperspectral measurements[J]. Geoderma,2008,148(2):141-148.
[28] 曾涛,琚存勇,蔡体久,等.利用变量投影重要性准则筛选郁闭度估测参数[J].北京林业大学学报,2010,32(6):37-41.
[29] 张小超,吴静珠,徐云.近红外光谱分析技术及其在现代农业种的应用[M].北京:电子工业出版社,2012:139.
[30] 张娟娟,田永超,姚霞,等.基于近红外光谱的土壤全氮含量估算模型[J].农业工程学报,2012,28(12):183-188.
[31] 纪文君,史舟,周清,等.几种不同类型土壤的VIS-NIR光谱特性及有机质响应波段[J].红外与毫米波学报,2012,31(3):277-282.
[32] Krishnan P, Alexander J D, Butler B J, et al. Reflectance technique for predicting soil organic matter[J].Soil Science Society of America Journal,1980,44(6):1282-1285.

36

Accesses

0

Citation

Detail

段落导航
相关文章

/