Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques
Identifieur interne : 001939 ( Istex/Corpus ); précédent : 001938; suivant : 001940Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques
Auteurs : M. Vohland ; C. EmmerlingSource :
- European Journal of Soil Science [ 1351-0754 ] ; 2011-08.
Abstract
The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; n = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, , RPDcv = 2.42; HWE‐Ccv, , RPDcv = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; n = 21 and n = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, , RPDcv = 3.14; Fluvisols/Cambisols, , RPDcv = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.
Url:
DOI: 10.1111/j.1365-2389.2011.01369.x
Links to Exploration step
ISTEX:E31B17583AD68BCE4C308206A32B65C9670A5854Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title><author><name sortKey="Vohland, M" sort="Vohland, M" uniqKey="Vohland M" first="M." last="Vohland">M. Vohland</name><affiliation><mods:affiliation>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: vohland@uni‐trier.de</mods:affiliation></affiliation></author><author><name sortKey="Emmerling, C" sort="Emmerling, C" uniqKey="Emmerling C" first="C." last="Emmerling">C. Emmerling</name><affiliation><mods:affiliation>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:E31B17583AD68BCE4C308206A32B65C9670A5854</idno><date when="2011" year="2011">2011</date><idno type="doi">10.1111/j.1365-2389.2011.01369.x</idno><idno type="url">https://api.istex.fr/document/E31B17583AD68BCE4C308206A32B65C9670A5854/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001939</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001939</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title><author><name sortKey="Vohland, M" sort="Vohland, M" uniqKey="Vohland M" first="M." last="Vohland">M. Vohland</name><affiliation><mods:affiliation>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: vohland@uni‐trier.de</mods:affiliation></affiliation></author><author><name sortKey="Emmerling, C" sort="Emmerling, C" uniqKey="Emmerling C" first="C." last="Emmerling">C. Emmerling</name><affiliation><mods:affiliation>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">European Journal of Soil Science</title><idno type="ISSN">1351-0754</idno><idno type="eISSN">1365-2389</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><date type="published" when="2011-08">2011-08</date><biblScope unit="volume">62</biblScope><biblScope unit="issue">4</biblScope><biblScope unit="page" from="598">598</biblScope><biblScope unit="page" to="606">606</biblScope></imprint><idno type="ISSN">1351-0754</idno></series><idno type="istex">E31B17583AD68BCE4C308206A32B65C9670A5854</idno><idno type="DOI">10.1111/j.1365-2389.2011.01369.x</idno><idno type="ArticleID">EJSS1369</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">1351-0754</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; n = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, , RPDcv = 2.42; HWE‐Ccv, , RPDcv = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; n = 21 and n = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, , RPDcv = 3.14; Fluvisols/Cambisols, , RPDcv = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.</div></front></TEI><istex><corpusName>wiley</corpusName><author><json:item><name>M. Vohland</name><affiliations><json:string>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</json:string><json:string>E-mail: vohland@uni‐trier.de</json:string></affiliations></json:item><json:item><name>C. Emmerling</name><affiliations><json:string>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</json:string></affiliations></json:item></author><articleId><json:string>EJSS1369</json:string></articleId><language><json:string>eng</json:string></language><originalGenre><json:string>article</json:string></originalGenre><abstract>The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; n = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, , RPDcv = 2.42; HWE‐Ccv, , RPDcv = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; n = 21 and n = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, , RPDcv = 3.14; Fluvisols/Cambisols, , RPDcv = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.</abstract><qualityIndicators><score>7.963</score><pdfVersion>1.3</pdfVersion><pdfPageSize>595.276 x 782.362 pts</pdfPageSize><refBibsNative>true</refBibsNative><abstractCharCount>1787</abstractCharCount><pdfWordCount>4963</pdfWordCount><pdfCharCount>31625</pdfCharCount><pdfPageCount>9</pdfPageCount><abstractWordCount>268</abstractWordCount></qualityIndicators><title>Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title><refBibs><json:item><author><json:item><name>M.C.U. Araújo</name></json:item><json:item><name>T.C.B. Saldanha</name></json:item><json:item><name>R.K.H. Galvão</name></json:item><json:item><name>T. Yoneyama</name></json:item><json:item><name>H.C. Chame</name></json:item><json:item><name>V. Visani</name></json:item></author><host><volume>57</volume><pages><last>73.</last><first>65</first></pages><author></author><title>Chemometrics & Intelligent Laboratory Systems</title></host><title>The successive projections algorithm for variable selection in spectroscopic multicomponent analysis.</title></json:item><json:item><author><json:item><name>M.F. Baumgardner</name></json:item><json:item><name>S.J. Kristof</name></json:item><json:item><name>C.J. Johansen</name></json:item><json:item><name>A.L. Zachary</name></json:item></author><host><volume>79</volume><pages><last>422.</last><first>413</first></pages><author></author><title>Proceedings of Indian Academy of Science</title></host><title>Effect of organic matter on the multispectral properties of soils.</title></json:item><json:item><author><json:item><name>E. Ben‐Dor</name></json:item><json:item><name>A. Banin</name></json:item></author><host><volume>59</volume><pages><last>372.</last><first>364</first></pages><author></author><title>Soil Science Society of America Journal</title></host><title>Near‐infrared analysis as a rapid method to simultaneously evaluate several soil properties.</title></json:item><json:item><author><json:item><name>E. Ben‐Dor</name></json:item><json:item><name>A. Banin</name></json:item></author><host><volume>159</volume><pages><last>270.</last><first>259</first></pages><author></author><title>Soil Science</title></host><title>Near infrared analysis (NIR) as a method to simultaneously evaluate spectral featureless constituents in soils.</title></json:item><json:item><author><json:item><name>E. Ben‐Dor</name></json:item><json:item><name>J.A. Irons</name></json:item><json:item><name>A. Epema</name></json:item></author><host><pages><last>188.</last><first>111</first></pages><author></author><title>Manual of Remote Sensing</title></host><title>Soil spectroscopy.</title></json:item><json:item><author><json:item><name>E. Ben‐Dor</name></json:item><json:item><name>R.G. Taylor</name></json:item><json:item><name>J. Hill</name></json:item><json:item><name>J.A.M. Demattê</name></json:item><json:item><name>M.L. Whiting</name></json:item><json:item><name>S. Chabrillat</name></json:item></author><host><pages><last>392.</last><first>321</first></pages><author></author><title>Advances in Agronomy</title></host><title>Imaging spectrometry for soil applications.</title></json:item><json:item><author><json:item><name>S.A. Bowers</name></json:item><json:item><name>R.J. Hanks</name></json:item></author><host><volume>100</volume><pages><last>138.</last><first>130</first></pages><author></author><title>Soil Science</title></host><title>Reflection of radiant energy from soils.</title></json:item><json:item><author><json:item><name>L. Cécillon</name></json:item><json:item><name>B.G. Barthès</name></json:item><json:item><name>C. Gomez</name></json:item><json:item><name>D. Ertlen</name></json:item><json:item><name>V. Genot</name></json:item><json:item><name>M. Hedde</name></json:item></author><host><volume>60</volume><pages><last>784.</last><first>770</first></pages><author></author><title>European Journal of Soil Science</title></host><title>Assessment and monitoring of soil quality using near‐infrared reflectance spectroscopy (NIRS).</title></json:item><json:item><author><json:item><name>M. Chodak</name></json:item><json:item><name>P. Khanna</name></json:item><json:item><name>F. Beese</name></json:item></author><host><volume>39</volume><pages><last>130.</last><first>123</first></pages><author></author><title>Biology & Fertility of Soils</title></host><title>Hot water extractable C and N in relation to microbiological properties of soils under beech forests.</title></json:item><json:item><author><json:item><name>M. Confalonieri</name></json:item><json:item><name>F. Fornasier</name></json:item><json:item><name>A. Ursino</name></json:item><json:item><name>F. Boccardi</name></json:item><json:item><name>B. Pintus</name></json:item><json:item><name>M. Odoardi</name></json:item></author><host><volume>9</volume><pages><last>141.</last><first>123</first></pages><author></author><title>Journal of Near Infrared Spectroscopy</title></host><title>The potential of near infrared reflectance spectroscopy as a tool for the chemical characterisation of agricultural soils.</title></json:item><json:item><author><json:item><name>M.M. Coûteaux</name></json:item><json:item><name>B. Berg</name></json:item><json:item><name>P. Rovira</name></json:item></author><host><volume>35</volume><pages><last>1600.</last><first>1587</first></pages><author></author><title>Soil Biology & Biochemistry</title></host><title>Near infrared reflectance spectroscopy for determination of organic matter fractions including microbial biomass in coniferous forest soils.</title></json:item><json:item><author><json:item><name>R.C. Dalal</name></json:item><json:item><name>R.J. Henry</name></json:item></author><host><volume>50</volume><pages><last>123.</last><first>120</first></pages><author></author><title>Soil Science Society of America Journal</title></host><title>Simultaneous determination of moisture, organic carbon, and total nitrogen by near infrared reflectance spectrophotometry.</title></json:item><json:item><host><author></author><title>Hartge, K.H. & Horn, R. 1989. Die physikalische Untersuchung von Böden.Enke, Stuttgart.</title></host></json:item><json:item><author><json:item><name>T.L. Henderson</name></json:item><json:item><name>M.F. Baumgardner</name></json:item><json:item><name>D.P. Franzmeier</name></json:item><json:item><name>D.E. Stott</name></json:item><json:item><name>D.C. Coster</name></json:item></author><host><volume>56</volume><pages><last>872.</last><first>865</first></pages><author></author><title>Soil Science Society of America Journal</title></host><title>High dimensional reflectance analysis of soil organic matter.</title></json:item><json:item><author><json:item><name>D. Landgraf</name></json:item><json:item><name>C. Böhm</name></json:item><json:item><name>F. Makeschin</name></json:item></author><host><volume>166</volume><pages><last>325.</last><first>319</first></pages><author></author><title>Journal of Plant Nutrition & Soil Science</title></host><title>Dynamic of different C and N fractions in a cambisol under five year succession fallow in Saxony (Germany).</title></json:item><json:item><author><json:item><name>R. Leardi</name></json:item><json:item><name>A.L. González</name></json:item></author><host><volume>41</volume><pages><last>207.</last><first>195</first></pages><author></author><title>Chemometrics & Intelligent Laboratory Systems</title></host><title>Genetic algorithms applied to feature selection in PLS regression: how and when to use them.</title></json:item><json:item><author><json:item><name>D.F. Malley</name></json:item><json:item><name>P.D. Martin</name></json:item><json:item><name>E. Ben‐Dor</name></json:item></author><host><pages><last>783.</last><first>729</first></pages><author></author><title>Near‐Infrared Spectroscopy in Agriculture</title></host><title>Application in analysis of soils.</title></json:item><json:item><author><json:item><name>G.W. McCarty</name></json:item><json:item><name>J.B. Reeves</name></json:item><json:item><name>V.B. Reeves</name></json:item><json:item><name>R.F. Follett</name></json:item><json:item><name>J.M. Kimble</name></json:item></author><host><volume>66</volume><pages><last>646.</last><first>640</first></pages><author></author><title>Soil Science Society of America Journal</title></host><title>Mid‐infrared and near‐infrared diffuse reflectance spectroscopy for soil carbon measurement.</title></json:item><json:item><author><json:item><name>A.M. Mouazen</name></json:item><json:item><name>M.R. Maleki</name></json:item><json:item><name>J. De Baerdemaeker</name></json:item><json:item><name>H. Ramon</name></json:item></author><host><volume>93</volume><pages><last>27.</last><first>13</first></pages><author></author><title>Soil & Tillage Research</title></host><title>On‐line measurement of some selected soil properties using a VIS‐NIR sensor.</title></json:item><json:item><author><json:item><name>J.B. Reeves</name></json:item></author><host><volume>158</volume><pages><last>14.</last><first>3</first></pages><author></author><title>Geoderma</title></host><title>Near‐ versus mid‐infrared diffuse reflectance spectroscopy for soil analysis emphasizing carbon and laboratory versus on‐site analysis: where are we and what needs to be done?</title></json:item><json:item><author><json:item><name>R. Rinnan</name></json:item><json:item><name>A. Rinnan</name></json:item></author><host><volume>39</volume><pages><last>1673.</last><first>1664</first></pages><author></author><title>Soil Biology & Biochemistry</title></host><title>Application of near infrared reflectance (NIR) and fluorescence spectroscopy to analysis of microbiological and chemical properties of arctic soil.</title></json:item><json:item><author><json:item><name>Å. Rinnan</name></json:item><json:item><name>F. Van den Berg</name></json:item><json:item><name>S.B. Engelsen</name></json:item></author><host><volume>28</volume><pages><last>1222.</last><first>1201</first></pages><author></author><title>Trends in Analytical Chemistry</title></host><title>Review of the most common pre‐processing techniques for near‐infrared spectra.</title></json:item><json:item><author><json:item><name>E. Schulz</name></json:item><json:item><name>M. Körschens</name></json:item></author><host><volume>31</volume><pages><last>813.</last><first>809</first></pages><author></author><title>Eurasian Soil Science</title></host><title>Characterization of the decomposable part of soil organic matter (SOM) and transformation processes by hot water extraction.</title></json:item><json:item><author><json:item><name>G.M. Vasques</name></json:item><json:item><name>S. Grunwald</name></json:item><json:item><name>J.O. Sickman</name></json:item></author><host><volume>146</volume><pages><last>25.</last><first>14</first></pages><author></author><title>Geoderma</title></host><title>Comparison of multivariate methods for inferential modeling of soil carbon using visible/near‐infrared spectra.</title></json:item><json:item><author><json:item><name>G.M. Vasques</name></json:item><json:item><name>S. Grunwald</name></json:item><json:item><name>J.O. Sickman</name></json:item></author><host><volume>73</volume><pages><last>184.</last><first>176</first></pages><author></author><title>Soil Science Society of America Journal</title></host><title>Modeling of soil carbon fractions using visible‐near‐infrared spectroscopy.</title></json:item><json:item><author><json:item><name>R.A. Viscarra Rossel</name></json:item><json:item><name>R.M. Lark</name></json:item></author><host><volume>60</volume><pages><last>464.</last><first>453</first></pages><author></author><title>European Journal of Soil Science</title></host><title>Improved analysis and modelling of soil diffuse reflectance spectra using wavelets.</title></json:item><json:item><author><json:item><name>R.A. Viscarra Rossel</name></json:item><json:item><name>D.J.J. Walvoort</name></json:item><json:item><name>A.B. McBratney</name></json:item><json:item><name>L.J. Janik</name></json:item><json:item><name>J.O. Skjemstad</name></json:item></author><host><volume>131</volume><pages><last>75.</last><first>59</first></pages><author></author><title>Geoderma</title></host><title>Visible, near infrared, mid infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties.</title></json:item><json:item><author><json:item><name>S. Wold</name></json:item><json:item><name>M. Sjöström</name></json:item><json:item><name>L. Eriksson</name></json:item></author><host><volume>58</volume><pages><last>130.</last><first>109</first></pages><author></author><title>Chemometrics & Intelligent Laboratory Systems</title></host><title>PLS‐regression: a basic tool of chemometrics.</title></json:item><json:item><author><json:item><name>S. Ye</name></json:item><json:item><name>D. Wang</name></json:item><json:item><name>S. Min</name></json:item></author><host><volume>91</volume><pages><last>199.</last><first>194</first></pages><author></author><title>Chemometrics & Intelligent Laboratory Systems</title></host><title>Successive projections algorithm combined with uninformative variable elimination for spectral variable selection.</title></json:item><json:item><author><json:item><name>H. Yoshido</name></json:item><json:item><name>R. Leardi</name></json:item><json:item><name>K. Funatsu</name></json:item><json:item><name>K. Varmuza</name></json:item></author><host><volume>446</volume><pages><last>494.</last><first>485</first></pages><author></author><title>Analytica Chimica Acta</title></host><title>Feature selection by genetic algorithms for mass spectral classifiers.</title></json:item></refBibs><genre><json:string>article</json:string></genre><host><volume>62</volume><publisherId><json:string>EJSS</json:string></publisherId><pages><total>9</total><last>606</last><first>598</first></pages><issn><json:string>1351-0754</json:string></issn><issue>4</issue><genre><json:string>journal</json:string></genre><language><json:string>unknown</json:string></language><eissn><json:string>1365-2389</json:string></eissn><title>European Journal of Soil Science</title><doi><json:string>10.1111/(ISSN)1365-2389</json:string></doi></host><categories><wos><json:string>science</json:string><json:string>soil science</json:string></wos><scienceMetrix><json:string>applied sciences</json:string><json:string>agriculture, fisheries & forestry</json:string><json:string>agronomy & agriculture</json:string></scienceMetrix></categories><publicationDate>2011</publicationDate><copyrightDate>2011</copyrightDate><doi><json:string>10.1111/j.1365-2389.2011.01369.x</json:string></doi><id>E31B17583AD68BCE4C308206A32B65C9670A5854</id><score>2.1497672</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/E31B17583AD68BCE4C308206A32B65C9670A5854/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/E31B17583AD68BCE4C308206A32B65C9670A5854/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/E31B17583AD68BCE4C308206A32B65C9670A5854/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><availability><p>© 2011 The Authors. Journal compilation © 2011 British Society of Soil Science</p></availability><date>2011</date></publicationStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title><author xml:id="author-1"><persName><forename type="first">M.</forename><surname>Vohland</surname></persName><email>vohland@uni‐trier.de</email><affiliation>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</affiliation></author><author xml:id="author-2"><persName><forename type="first">C.</forename><surname>Emmerling</surname></persName><affiliation>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</affiliation></author></analytic><monogr><title level="j">European Journal of Soil Science</title><idno type="pISSN">1351-0754</idno><idno type="eISSN">1365-2389</idno><idno type="DOI">10.1111/(ISSN)1365-2389</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><date type="published" when="2011-08"></date><biblScope unit="volume">62</biblScope><biblScope unit="issue">4</biblScope><biblScope unit="page" from="598">598</biblScope><biblScope unit="page" to="606">606</biblScope></imprint></monogr><idno type="istex">E31B17583AD68BCE4C308206A32B65C9670A5854</idno><idno type="DOI">10.1111/j.1365-2389.2011.01369.x</idno><idno type="ArticleID">EJSS1369</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>2011</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; n = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, , RPDcv = 2.42; HWE‐Ccv, , RPDcv = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; n = 21 and n = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, , RPDcv = 3.14; Fluvisols/Cambisols, , RPDcv = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.</p></abstract></profileDesc><revisionDesc><change when="2011-08">Published</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/E31B17583AD68BCE4C308206A32B65C9670A5854/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="Wiley, elements deleted: body"><istex:xmlDeclaration>version="1.0" encoding="UTF-8" standalone="yes"</istex:xmlDeclaration><istex:document><component version="2.0" type="serialArticle" xml:lang="en"><header><publicationMeta level="product"><publisherInfo><publisherName>Blackwell Publishing Ltd</publisherName><publisherLoc>Oxford, UK</publisherLoc></publisherInfo><doi origin="wiley" registered="yes">10.1111/(ISSN)1365-2389</doi><issn type="print">1351-0754</issn><issn type="electronic">1365-2389</issn><idGroup><id type="product" value="EJSS"></id><id type="publisherDivision" value="ST"></id></idGroup><titleGroup><title type="main" sort="EUROPEAN JOURNAL OF SOIL SCIENCE">European Journal of Soil Science</title></titleGroup></publicationMeta><publicationMeta level="part" position="08104"><doi origin="wiley">10.1111/ejs.2011.62.issue-4</doi><numberingGroup><numbering type="journalVolume" number="62">62</numbering><numbering type="journalIssue" number="4">4</numbering></numberingGroup><coverDate startDate="2011-08">August 2011</coverDate></publicationMeta><publicationMeta level="unit" type="article" position="12" status="forIssue"><doi origin="wiley">10.1111/j.1365-2389.2011.01369.x</doi><idGroup><id type="unit" value="EJSS1369"></id></idGroup><countGroup><count type="pageTotal" number="9"></count></countGroup><titleGroup><title type="tocHeading1">Chemical processes and functions</title></titleGroup><copyright>© 2011 The Authors. Journal compilation © 2011 British Society of Soil Science</copyright><eventGroup><event type="xmlConverted" agent="Converter:BPG_TO_WML3G version:2.5.2 mode:FullText" date="2011-07-18"></event><event type="publishedOnlineEarlyUnpaginated" date="2011-04-20"></event><event type="firstOnline" date="2011-04-20"></event><event type="publishedOnlineFinalForm" date="2011-07-18"></event><event type="xmlConverted" agent="Converter:WILEY_ML3G_TO_WILEY_ML3GV2 version:4.0.1" date="2014-03-12"></event><event type="xmlConverted" agent="Converter:WML3G_To_WML3G version:4.1.7 mode:FullText,remove_FC" date="2014-10-16"></event></eventGroup><numberingGroup><numbering type="pageFirst" number="598">598</numbering><numbering type="pageLast" number="606">606</numbering></numberingGroup><correspondenceTo>M. Vohland. E‐mail: <email normalForm="vohland@uni-trier.de">vohland@uni‐trier.de</email></correspondenceTo><linkGroup><link type="toTypesetVersion" href="file:EJSS.EJSS1369.pdf"></link></linkGroup></publicationMeta><contentMeta><unparsedEditorialHistory>Received 17 March 2010; revised version accepted 18 January 2011</unparsedEditorialHistory><countGroup><count type="figureTotal" number="5"></count><count type="tableTotal" number="4"></count><count type="formulaTotal" number="0"></count><count type="referenceTotal" number="30"></count></countGroup><titleGroup><title type="main">Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title><title type="shortAuthors"><i>M. Vohland & C. Emmerling</i></title></titleGroup><creators><creator creatorRole="author" xml:id="cr1" affiliationRef="#a1" corresponding="yes"><personName><givenNames>M.</givenNames><familyName>Vohland</familyName></personName></creator><creator creatorRole="author" xml:id="cr2" affiliationRef="#a2"><personName><givenNames>C.</givenNames><familyName>Emmerling</familyName></personName></creator></creators><affiliationGroup><affiliation xml:id="a1" countryCode="DE"><unparsedAffiliation><i>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</i></unparsedAffiliation></affiliation><affiliation xml:id="a2" countryCode="DE"><unparsedAffiliation><i>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</i></unparsedAffiliation></affiliation></affiliationGroup><abstractGroup><abstract type="main" xml:lang="en"><p>The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; <i>n</i> = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, <inlineGraphic alt="inline image" location="equation/EJSS_1369_mu1.gif" href=""></inlineGraphic>, RPD<sub>cv</sub> = 2.42; HWE‐C<sub>cv</sub>, <inlineGraphic alt="inline image" location="equation/EJSS_1369_mu2.gif" href=""></inlineGraphic>, RPD<sub>cv</sub> = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; <i>n</i> = 21 and <i>n</i> = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, <inlineGraphic alt="inline image" location="equation/EJSS_1369_mu3.gif" href=""></inlineGraphic>, RPD<sub>cv</sub> = 3.14; Fluvisols/Cambisols, <inlineGraphic alt="inline image" location="equation/EJSS_1369_mu4.gif" href=""></inlineGraphic>, RPD<sub>cv</sub> = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.</p></abstract></abstractGroup></contentMeta></header></component></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title></titleInfo><titleInfo type="alternative" contentType="CDATA" lang="en"><title>Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques</title></titleInfo><name type="personal"><namePart type="given">M.</namePart><namePart type="family">Vohland</namePart><affiliation>Faculty of Geography and Geosciences, Remote Sensing Department, University of Trier, D‐54286 Trier, Germany</affiliation><affiliation>E-mail: vohland@uni‐trier.de</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">C.</namePart><namePart type="family">Emmerling</namePart><affiliation>Faculty of Geography and Geosciences, Soil Science Department, University of Trier, D‐54286 Trier, Germany</affiliation><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="article" displayLabel="article"></genre><originInfo><publisher>Blackwell Publishing Ltd</publisher><place><placeTerm type="text">Oxford, UK</placeTerm></place><dateIssued encoding="w3cdtf">2011-08</dateIssued><edition>Received 17 March 2010; revised version accepted 18 January 2011</edition><copyrightDate encoding="w3cdtf">2011</copyrightDate></originInfo><language><languageTerm type="code" authority="rfc3066">en</languageTerm><languageTerm type="code" authority="iso639-2b">eng</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType><extent unit="figures">5</extent><extent unit="tables">4</extent><extent unit="references">30</extent></physicalDescription><abstract lang="en">The calibration of soil organic C (SOC) and hot water‐extractable C (HWE‐C) from visible and near‐infrared soil reflectance spectra is hindered by the complex spectral interaction of soil chromophores that usually varies from one soil or soil type to another. The exploitation of spectral variables from spectroradiometer data is further affected by multicollinearity and noise. In this study, a set of soil samples (Fluvisols, Podzols, Cambisols and Chernozems; n = 48) representing a wide range of properties was analysed. Spectral readings with a fibre‐optics visible to near‐infrared instrument were used to estimate SOC and HWE‐C contents by partial least squares regression (PLS). In addition to full‐spectrum PLS, spectral feature selection techniques were applied with PLS (uninformative variable elimination, UVE‐PLS, and a genetic algorithm, GA‐PLS). On the basis of normalized spectra (mean centring + vector normalization), the order of prediction accuracy was GA‐PLS ≫ UVE‐PLS > PLS for SOC; for HWE‐C, it was GA‐PLS > UVE‐PLS, PLS. With GA‐PLS, acceptable cross‐validated (cv) prediction accuracies were obtained for the complete dataset (SOC, , RPDcv = 2.42; HWE‐Ccv, , RPDcv = 2.13). Splitting the soil data into two groups with different basic properties (Podzols compared with Fluvisols/Cambisols; n = 21 and n = 23, respectively) improved SOC predictions with GA‐PLS distinctly (Podzols, , RPDcv = 3.14; Fluvisols/Cambisols, , RPDcv = 3.64). This demonstrates the importance of using stratified models for successful quantitative approaches after an initial rough screening. GA selection frequencies suggest that the spectral region over 1900 nm, and in particular the hydroxyl band at 2200 nm are of great importance for the spectral prediction of both SOC and HWE‐C.</abstract><relatedItem type="host"><titleInfo><title>European Journal of Soil Science</title></titleInfo><genre type="journal">journal</genre><identifier type="ISSN">1351-0754</identifier><identifier type="eISSN">1365-2389</identifier><identifier type="DOI">10.1111/(ISSN)1365-2389</identifier><identifier type="PublisherID">EJSS</identifier><part><date>2011</date><detail type="volume"><caption>vol.</caption><number>62</number></detail><detail type="issue"><caption>no.</caption><number>4</number></detail><extent unit="pages"><start>598</start><end>606</end><total>9</total></extent></part></relatedItem><identifier type="istex">E31B17583AD68BCE4C308206A32B65C9670A5854</identifier><identifier type="DOI">10.1111/j.1365-2389.2011.01369.x</identifier><identifier type="ArticleID">EJSS1369</identifier><accessCondition type="use and reproduction" contentType="copyright">© 2011 The Authors. Journal compilation © 2011 British Society of Soil Science</accessCondition><recordInfo><recordContentSource>WILEY</recordContentSource><recordOrigin>Blackwell Publishing Ltd</recordOrigin></recordInfo></mods></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Rhénanie/explor/UnivTrevesV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001939 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 001939 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Rhénanie
|area= UnivTrevesV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:E31B17583AD68BCE4C308206A32B65C9670A5854
|texte= Determination of total soil organic C and hot water‐extractable C from VIS‐NIR soil reflectance with partial least squares regression and spectral feature selection techniques
}}
| This area was generated with Dilib version V0.6.31. |  |