土壤有机质高光谱特征与波长变量优选方法

doi:10.3864/j.issn.0578-1752.2017.22.009

摘要/Abstract

摘要： 【目的】探究土壤有机质的高光谱特征及响应规律，优选土壤有机质的敏感波长，降低土壤有机质高光谱估测模型复杂度，提高模型稳健性，为利用高光谱技术对农田土壤肥力的定量监测提供理论支撑。【方法】采集江汉平原潮土土样130个，将其中40个样本作为训练集，测量其去有机质前、后的土壤有机质含量及光谱数据，计算差值及变化率，分析土壤有机质含量变化对光谱特征的影响，结合无信息变量消除（uninformative variables elimination，UVE）、竞争适应重加权采样（competitive adaptive reweighted sampling，CARS）变量优选方法确定土壤有机质敏感波长；采用45个建模集样本，基于偏最小二乘回归(partial Least Squares Regression，PLSR)和反向传播神经网络（back propagation neural network，BPNN）建立土壤有机质含量的估算模型；利用45个验证集样本检验敏感波长对同类土壤的适用性。【结果】通过有机质去除试验，供试土壤的平均光谱反射率在全波段均有所增加，在可见光波段变化率高于近红外波段;比较UVE、CARS、UVE-CARS、CARS-UVE这4种变量优选方法，得到最佳变量优选方法为UVE-CARS，该方法从2001个波长变量中优选得到84个变量作为土壤有机质的敏感波长，分布于561—721、1 920—2 280 nm波段覆盖范围;基于敏感波长的PLSR、BPNN模型性能均优于全波段模型，其中，基于敏感波长的BPNN模型的估测能力高于PLSR，模型验证集R²、RMSE、RPD、MAE、MRE值分别为0.74、1.33 g·kg^-1、2.02、1.04 g·kg^-1、6.2%，可实现土壤有机质含量的有效估测。【结论】通过训练集获得的土壤有机质敏感波长，能够实现对该试验区同种土壤类型样本土壤有机质含量的有效估测；利用去有机质试验结合变量优选方法确定的敏感波长建模，不仅将输入波长压缩至全波段波长数目的4.2 %，而且提升了模型估测精度，降低了变量维度和模型复杂度，为快速准确评估农田土壤有机质含量提供了新途径。

关键词: 土壤有机质, 高光谱, 变量优选, 偏最小二乘回归, 反向传播神经网络, 潮土

Abstract: 【Objective】The objective of this study is to explore the hyperspectral features and response regularity of the soil organic matter, and to select the sensitive wavelengths of soil organic matter, so as to reduce complexity of hyperspectral estimation model of soil organic matter and improve robustness of the model, which is to provide theoretical support to quantitatively monitor the soil fertility of farmland by using the hyperspectral technology. 【Method】 A total of 130 fluvo-aquic soil samples were collected from Jianghan plain, of which 40 were the training set samples. The soil organic matter content (SOMC_raw) and spectral reflectance (SR_raw) were measured from total samples, and an experiment of removal of organic matter was performed using the training set samples, and then we measured the soil organic matter content (SOMC_rem) and spectral reflectance (SR_rem) from samples of removal of organic matter. By calculating the difference and rate of change between SR_raw and SR_remfrom training set samples, we could analyze how the content changes of soil organic matter itself influence the spectral features. The soil organic matter sensitive wavelengths were determined by the methods of uninformative variables elimination (UVE) and competitive adaptive reweighted sampling (CARS). The calibration set with 45 samples was utilized to build the soil organic matter estimation models base on partial least squares regression (PLSR) and back propagation neural network (BPNN), and the validation set of 45 samples was utilized to test whether sensitive wavelengths were suitable for the same type soil. 【Result】 The experiment of removal of organic matter showed that the average spectral reflectance of test soil samples increased at full-spectrum with removing organic matter content, especially at the visible spectrum; after the comparison of UVE, CARS, UVE-CARS, and CARS-UVE, the optimal method of variables selection was UVE-CARS. The method of UVE-CARS provided 84 selected variables which were the soil organic matter sensitive wavelengths with coverage area of 561-721, 1 920-2 280 nm. Based on soil organic matter sensitive wavelengths, the PLSR and BPNN had better performance than full-spectrum model, and BPNN was better than PLSR in predictive ability with its value of R², RMSE, RPD, MAE, MRE were 0.74, 1.33 g·kg^-1, 2.02, 1.04 g·kg^-1, 6.2%, respectively, so it could effectively estimate soil organic matter. 【Conclusion】 The soil organic matter sensitive wavelengths from training set could effectively estimate soil organic matter content in this test area with the same type samples. In addition, modeling of sensitive wavelengths by obtaining from the experiment of removal of organic matter and variables selection method could not only compress input wavelengths down into 4.2% of full-spectrum, but also enhance the estimation accuracy and reduce the model complexity. In this study, it provided a new approach to quickly and accurately evaluate soil organic matter content in the farmland.

Key words: soil organic matter, hyperspectra, variables selection, partial least squares regression, back propagation neural network, fluvo-aquic

朱亚星，于雷，洪永胜，章涛，朱强，李思缔，郭力，刘家胜. 土壤有机质高光谱特征与波长变量优选方法[J]. 中国农业科学, 2017, 50(22): 4325-4337.

ZHU YaXing, YU Lei, HONG YongSheng, ZHANG Tao, ZHU Qiang, LI SiDi, GUO Li, LIU JiaSheng. Hyperspectral Features and Wavelength Variables Selection Methods of Soil Organic Matter[J]. Scientia Agricultura Sinica, 2017, 50(22): 4325-4337.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://www.chinaagrisci.com/CN/10.3864/j.issn.0578-1752.2017.22.009

https://www.chinaagrisci.com/CN/Y2017/V50/I22/4325

参考文献

[1] 胡克林, 余艳, 张凤荣, 王茹. 北京郊区土壤有机质含量的时空变异及其影响因素. 中国农业科学, 2006, 39(4): 764-771.

HU K L, YU Y, ZHANG F R, WANG R. The spatial-temporal variability of soil organic matter and its influencing factors in suburban area of Beijing. Scientia Agricultura Sinica, 2006, 39(4): 764-771. (in Chinese)

[2] ROSSEL VISCARRA R A, CATTLE S R, ORTEGA A, FOUAD Y. In situ measurements of soil colour, mineral composition and clay content by vis–NIR spectroscopy. Geoderma, 2009, 150(3/4): 253-266.

[3] 史舟, 王乾龙, 彭杰, 纪文君, 刘焕军, 李曦, ROSSEL VISCARRA R.A. 中国主要土壤高光谱反射特性分类与有机质光谱预测模型. 中国科学: 地球科学, 2014, 44(5): 978-988.

SHI Z, WANG Q L, PENG J, JI W J, LIU H J, LI X, ROSSEL VISCARRA R A. Scientia Sinica(Terrae), 2014, 44(5): 978-988. (in Chinese)

[4] ZHENG G H, RYU D, JIAO C X, HONG C Q. Estimation of organic matter content in coastal soil using reflectance spectroscopy. Pedosphere, 2016, 26(1): 130-136.

[5] BAO N S, WU L X, YE B Y, ZHOU W. Assessing soil organic matter of reclaimed soil from a large surface coal mine using a field spectroradiometer in laboratory. Geoderma, 2017, 288: 47-55.

[6] SHI T Z, CHEN Y Y, LIU Y L, WU G F. Visible and near-infrared reflectance spectroscopy—An alternative for monitoring soil contamination by heavy metals. Journal of Hazardous Materials, 2014, 265(30): 166-176.

[7] 卢艳丽, 白由路, 杨俐苹, 王红娟. 基于高光谱的土壤有机质含量预测模型的建立与评价. 中国农业科学, 2007, 40(9): 1989-1995.

LU Y L, BAI Y L, LIU L P, WANG H J. Prediction and validation of soil organic matter content based on Hyperspectrum. Scientia Agricultura Sinica, 2007, 40(9): 1989-1995. (in Chinese)

[8] 于士凯, 姚艳敏, 王德营, 司海青. 基于高光谱的土壤有机质含量反演研究. 中国农学通报, 2013, 29(23): 146-152.

YU S K, YAO Y M, WANG D Y, SI H Q. Studies on the inversion of soil organic matter content based on hyper-spectrum. Chinese Agricultural Science Bulletin, 2013, 29(23): 146-152. (in Chinese)

[9] 刘焕军, 张柏, 赵军, 张兴义, 宋开山, 王宗明, 段洪涛. 黑土有机质含量高光谱模型研究. 土壤学报, 2007, 44(1): 27-32.

LIU H J, ZHANG B, ZHAO J, ZHANG X Y, SONG K S, WANG Z M, DUAN H T. Spectral models for prediction of organic matter in black soil. Acta Pedologica Sinica, 2007, 44(1): 27-32. (in Chinese)

[10] 徐彬彬, 季耿善, 朱永豪. 中国陆地背景和土壤光谱反射特性的地理分区的初步研究. 环境遥感, 1991, 6(2): 142-151.

XU B B, JI G S, ZHU Y H. A preliminary research of geographic regionalization of China land background and Spectral reflectance characteristics of soil. Remote Sensing of Environment China, 1991, 6(2): 142-151. (in Chinese)

[11] 彭杰, 张杨珠, 周清, 刘香伶, 周卫军. 去除有机质对土壤光谱特性的影响. 土壤, 2006, 38(4): 453-458.

PENG J, ZHANG Y Z, ZHOU Q, LIU X L, ZHOU W J. Spectral characteristics of soils in Hunan province as affected by removal of soil organic matter. Soils, 2006, 38(4): 453-458. (in Chinese)

[12] 彭杰, 周清, 张杨珠, 向红英. 有机质对土壤光谱特性的影响研究. 土壤学报, 2013, 50(3): 517-524.

PENG J, ZHOU Q, ZHANG Y Z, XIANG H Y. Effect of soil organic matter on spectral characteristic of soil. Acta Pedologica Sinica, 2013, 50(3): 517-524. (in Chinese)

[13] VOHLAND M, LUDWIG M, THIELE-BRUHN S, LUDWIG B. Determination of soil properties with visible to near- and mid-infrared spectroscopy: Effects of spectral variable selection. Geoderma, 2014, 223-225: 88-96.

[14] XU S X, ZHAO Y C, WANG M Y, SHI X Z. Determination of rice root density from Vis–NIR spectroscopy by support vector machine regression and spectral variable selection techniques. Catena, 2017, 157: 12-23.

[15] YANG H, KUANG B, MOUAZEN A M. Quantitative analysis of soil nitrogen and carbon at a farm scale using visible and near infrared spectroscopy coupled with wavelength reduction. European Journal of Soil Science, 2012, 63(3): 410-420.

[16] 于雷, 洪永胜, 周勇, 朱强, 徐良, 李冀云, 聂艳. 高光谱估算土壤有机质含量的波长变量筛选方法. 农业工程学报, 2016, 32(13): 95-102.

YU L, HONG Y S, ZHOU Y, ZHU Q, XU L, LI J Y, NIE Y. Wavelength variable selection methods for estimation of soil organic matter content using hyperspectral technique. Transactions of the CSAE, 2016, 32(13): 95-102. (in Chinese)

[17] 鲍士旦. 土壤农化分析. 3 版. 北京: 中国农业出版社, 2013: 30-34.

BAO S D. Soil Agricultural Chemistry Analysis. 3rd ed. Beijing: China Agriculture Press, 2013: 30-34. (in Chinese)

[18] 洪永胜, 于雷, 耿雷, 张薇, 聂艳, 周勇. 应用DS算法消除室内几何测试条件对土壤高光谱数据波动性的影响. 华中师范大学学报, 2016, 50(2): 303-308.

HONG Y S, YU L, GENG L, ZHANG W, NIE Y, ZHOU Y. Using direct standardization algorithm to eliminate the effect of laboratory geometric parameters in soil hyperspectral data fluctuate characteristic. Journal of Central Normal University, 2016, 50(2): 303-308. (in Chinese)

[19] CENTNER V, MASSART D L, DE NOORD O E, DE JONG S, VANDEGINSTE B M, STERNA C. Elimination of uninformative variables for multivariate calibration. Analytical Chemistry, 1996, 68(21): 3851-3858.

[20] 杨梅花, 赵小敏. 基于可见-近红外光谱变量选择的土壤全氮含量估测研究. 中国农业科学, 2014, 47(12): 2374-2383.

YANG M H, ZHAO X M. Study on soil total N estimation by Vis-NIR spectra with variable selection. Scientia Agricultura Sinica, 2014, 47(12): 2374-2383. (in Chinese)

[21] CAI W S, LI Y K, SHAO X G. A variable selection method based on uninformative variable elimination for multivariate calibration of near-infrared spectra. Chemometrics and Intelligent Laboratory Systems, 2008, 90(2): 188-194．

[22] 孙通, 吴宜青, 刘秀红, 莫欣欣, 刘木华. 激光诱导击穿光谱联合UVE变量优选检测大豆油中的铬含量. 光谱学与光谱分析, 2016, 36(10): 3341-3345.

SUN T, WU Y Q, LIU X H, MO X X, LIU M H. Detection of chromium content in soybean oil by laser induced breakdown spectroscopy and UVE method. Spectroscopy and Spectral Analysis, 2016, 36(10): 3341-3345. (in Chinese)

[23] 于雷, 朱亚星, 洪永胜, 夏天, 刘目兴, 周勇. 高光谱技术结合CARS算法预测土壤水分含量. 农业工程学报, 2016, 32(22): 138-145.

YU L, ZHU Y X, HONG Y S, XIA T, LIU M X, ZHOU Y. Determination of soil moisture content by hyperspectral technology with CARS algorithm. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(22): 138-145. (in Chinese)

[24] JIANG H, ZHANG H, CHEN Q S, MEI C L, LIU G H. Identification of solid state fermentation degree with FT-NIR spectroscopy: Comparison of wavelength variable selection methods of CARS and SCARS. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2015, 149: 1-7.

[25] 彭小婷, 高文秀, 王俊杰. 基于包络线去除和偏最小二乘的土壤参数光谱反演. 武汉大学学报(信息科学版), 2014, 39(7): 862-866.

PENG X T, GAO W X, WANG J J. Inversion of soil parameters from hyperspectra based on continuum removal and partial least squares regression. Geomatics and Information Science of Wuhan University, 2014, 39(7): 862-866. (in Chinese)

[26] 于雷, 洪永胜, 周勇, 朱强. 连续小波变换高光谱数据的土壤有机质含量反演模型构建. 光谱学与光谱分析, 2016, 36(5): 1428-1433.

YU L, HONG Y S, ZHOU Y, ZHU Q. Inversion of soil organic matter content using hyperspectral data based on continuous wavelet transformation. Spectroscopy and Spectral Analysis, 2016, 36(5): 1428-1433. (in Chinese)

[27] 王凯龙, 熊黑钢, 张芳. 基于PLSR-BP复合模型的绿洲土壤pH高光谱反演. 干旱区研究, 2014, 31(6): 1005-1009.

WANG K L, XIONG H G, ZHANG F. PLSR-BP complex model-based hyper-spectrum retrieval of oasis soil pH. Arid Zone Research, 2014, 31(6): 1005-1009. (in Chinese)

[28] 郑立华, 李民赞, 潘娈, 孙建英, 唐宁. 基于近红外光谱技术的土壤参数BP神经网络预测. 光谱学与光谱分析, 2008, 28(5): 1160-1164.

ZHENG L H, LI M Z, PAN L, SUN J Y, TANG N. Estimation of soil organic matter and soil total nitrogen based on NIR spectroscopy and BP neural network. Spectroscopy and Spectral Analysis, 2008, 28(5): 1160-1164. (in Chinese)

[29] 徐彬彬, 戴昌达. 南疆土壤光谱反射特性与有机质含量的相关分析. 科学通报, 1980(6): 282-284.

XU B B, DAI C D. Relationship between soil re?ectance characteristics and SOM content in south area of Xinjiang Province. Chinese Science Bulletin, 1980(6): 282-284. (in Chinese)

[30] 彭杰, 张杨珠, 庞新安, 王家强. 新疆南部土壤有机质含量的高光谱特征分析. 干旱区地理, 2010, 33(5): 740-746.

PENG J, ZHANG Y Z, PANG X A, WANG J Q. Hyperspectral features of soil organic matter content in South Xinjiang. Arid Land Geography, 2010, 33(5): 740-746. (in Chinese)

[31] 卢艳丽, 白由路, 杨俐苹, 王磊, 王贺. 东北平原不同类型土壤有机质含量高光谱反演模型同质性研究. 植物营养与肥料学报, 2011, 17(2): 456- 463.

LU Y L, BAI Y L, YANG L P, WANG L, WANG H. Homogeneity of retrieval models for soil organic matter of different soil types in Northeast Plain using hyperspectral data. Plant Nutrition and Fertilizer Science, 2011, 17(2): 456- 463. (in Chinese)

[32] BEN-DOR E, BANIN A. Near-infrared analysis as a rapid method to simultaneously evaluate several soil properties. Soil Science Society of America of Journal, 1995, 59(2): 364-372.

[33] LI H D, LIANG Y Z, XU Q S, CAO D S. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. Analytica Chimica Acta, 2009, 648:77-84.

[34] 纪文君, 李曦, 李成学, 周银, 史舟. 基于全谱数据挖掘技术的土壤有机质高光谱预测建模研究. 光谱学与光谱分析, 2012, 32(9): 2393-2398.

JI W J, LI X, LI C X, ZHOU Y, SHI Z. Using different data mining algorithms to predict soil organic matter based on visible-near infrared spectroscopy. Spectroscopy and Spectral Analysis, 2012, 32(9): 2393-2398. (in Chinese)

[35] ASKARI M S, O’ROURKE S M. HOLDEN N M. Evaluation of soil quality for agricultural production using visible near infrared spectroscopy. Geoderma, 2015, 243-144: 80-91.

[36] 武永峰, 董一威, 胡新, 吕国华, 任德超, 宋吉青. 基于近红外漫反射光谱的农田原位表层土壤含水量定量建模方法比较. 光谱学与光谱分析, 2015, 35(12): 3416-3421.

WU Y F, DONG Y W, HU X, LÜ G H, REN D C, SONG J Q. Quantification of agricultural in-situ surface soil moisture content using near infrared diffuse reflectance spectroscopy: a comparison of modeling methods. Spectroscopy and Spectral Analysis, 2015, 35(12): 3416-3421. (in Chinese)