Determination of Soil Organic Matter and Total Nitrogen from Visible Near-Infrared Spectroscopy by Multivariate Models and Variable Selection Techniques

IF 1.4 4区农林科学 Q4 SOIL SCIENCE

Eurasian Soil Science Pub Date : 2024-04-10 DOI:10.1134/s1064229323603505

Hailiang Zhang, Jing Zhang, Zailiang Chen, Chaoyong Xie, Baishao Zhan, Wei Luo, Xuemei Liu

{"title":"Determination of Soil Organic Matter and Total Nitrogen from Visible Near-Infrared Spectroscopy by Multivariate Models and Variable Selection Techniques","authors":"Hailiang Zhang, Jing Zhang, Zailiang Chen, Chaoyong Xie, Baishao Zhan, Wei Luo, Xuemei Liu","doi":"10.1134/s1064229323603505","DOIUrl":null,"url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The status of soil nutrient content is a fundamental factor affecting changes in soil quality, influencing the growth conditions and yield levels of crops. The practicality of combining visible and near-infrared spectroscopy to evaluate soil organic matter (SOM) and total nitrogen (TN) may give an alternative to soil physicochemical examination in the laboratory, which is laborious and contaminative. A total of 394 Ferralsols soil samples were gathered from navel orange orchards located in the province of Jiangxi, China. To enhance the spectrum information, the spectra were preprocessed using five different techniques, including lg(1/<i>R</i>), multiplicative scatter correction, standard normal variate), detrending and Savitzky-Golay smoothing. Four variable selection algorithms—competitive adaptive reweighted sampling, successive projections algorithm, random frog, and genetic algorithm – were combined with three multivariate methods—partial least squares regression, multiple linear regression, and least squares support vector machine. The most efficient strategy combines LSSVM calibration methods with GA and lg(1/<i>R</i>) preprocessing. It generates values for the determination coefficient of prediction, root mean square error of prediction, and residual predictive deviation that are as follows: 0.8948, 0.1597, and 3.0949, respectively, for SOM; and 0.9129, 0.0021, and 3.4014, respectively, for TN. The results indicate that this method can accurately determine the SOM and TN in agricultural land soil, facilitating the timely adjustment of soil management measures.</p>","PeriodicalId":11892,"journal":{"name":"Eurasian Soil Science","volume":"14 1","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Eurasian Soil Science","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.1134/s1064229323603505","RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"SOIL SCIENCE","Score":null,"Total":0}

引用次数: 0

Abstract

The status of soil nutrient content is a fundamental factor affecting changes in soil quality, influencing the growth conditions and yield levels of crops. The practicality of combining visible and near-infrared spectroscopy to evaluate soil organic matter (SOM) and total nitrogen (TN) may give an alternative to soil physicochemical examination in the laboratory, which is laborious and contaminative. A total of 394 Ferralsols soil samples were gathered from navel orange orchards located in the province of Jiangxi, China. To enhance the spectrum information, the spectra were preprocessed using five different techniques, including lg(1/R), multiplicative scatter correction, standard normal variate), detrending and Savitzky-Golay smoothing. Four variable selection algorithms—competitive adaptive reweighted sampling, successive projections algorithm, random frog, and genetic algorithm – were combined with three multivariate methods—partial least squares regression, multiple linear regression, and least squares support vector machine. The most efficient strategy combines LSSVM calibration methods with GA and lg(1/R) preprocessing. It generates values for the determination coefficient of prediction, root mean square error of prediction, and residual predictive deviation that are as follows: 0.8948, 0.1597, and 3.0949, respectively, for SOM; and 0.9129, 0.0021, and 3.4014, respectively, for TN. The results indicate that this method can accurately determine the SOM and TN in agricultural land soil, facilitating the timely adjustment of soil management measures.

Abstract Image

查看原文本刊更多论文

利用多元模型和变量选择技术通过可见近红外光谱测定土壤有机质和总氮

摘要土壤养分含量状况是影响土壤质量变化的基本因素，会影响作物的生长条件和产量水平。结合可见光和近红外光谱评估土壤有机质（SOM）和全氮（TN）的实用性可以替代实验室土壤理化检验，因为实验室检验既费力又易污染。本研究从中国江西省的脐橙果园中采集了 394 份费拉索尔土壤样本。为了增强光谱信息，使用了五种不同的技术对光谱进行预处理，包括 lg(1/R)、乘法散度校正、标准正态变异）、去趋势和萨维茨基-戈莱平滑。四种变量选择算法--竞争性自适应再加权采样、连续投影算法、随机蛙算法和遗传算法--与三种多元方法--部分最小二乘回归法、多元线性回归法和最小二乘支持向量机相结合。最有效的策略是将 LSSVM 校准方法与 GA 和 lg(1/R) 预处理相结合。它产生的预测确定系数、预测均方根误差和残差预测偏差值如下：SOM 分别为 0.8948、0.1597 和 3.0949；TN 分别为 0.9129、0.0021 和 3.4014。结果表明，该方法能准确测定农田土壤中的 SOM 和 TN，便于及时调整土壤管理措施。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Eurasian Soil Science 农林科学-土壤科学

CiteScore

2.70

自引率

35.70%

发文量

137

审稿时长

12-24 weeks

期刊介绍： Eurasian Soil Science publishes original research papers on global and regional studies discussing both theoretical and experimental problems of genesis, geography, physics, chemistry, biology, fertility, management, conservation, and remediation of soils. Special sections are devoted to current news in the life of the International and Russian soil science societies and to the history of soil sciences. Since 2000, the journal Agricultural Chemistry, the English version of the journal of the Russian Academy of Sciences Agrokhimiya, has been merged into the journal Eurasian Soil Science and is no longer published as a separate title.