Estimating forest soil bulk density using boosted regression modelling
Identifieur interne : 000573 ( Main/Exploration ); précédent : 000572; suivant : 000574Estimating forest soil bulk density using boosted regression modelling
Auteurs : S. S. M. Jalabert [France] ; M. P. Martin [France] ; J. Renaud [France] ; L. Boulonne [France] ; C. Jolivet [France] ; L. Montanarella [Italie] ; D. Arrouays [France]Source :
- Soil Use and Management [ 0266-0032 ] ; 2010-12.
English descriptors
Abstract
Soil bulk density (ρ) is an important physical property, but its measurement is frequently lacking in soil surveys due to the time‐consuming nature of making the measurement. As a result pedotransfer functions (PTFs) have been developed to predict ρ from other more easily available soil properties. These functions are generally derived from regression methods that aim to fit a single model. In this study, we use a technique called Generalized Boosted Regression Modelling (GBM; Ridgeway, 2006) which combines two algorithms: regression trees and boosting. We built two models and compared their predictive performance with published PTFs. All the functions were fitted based on the French forest soil dataset for the European demonstration Biosoil project. The two GBM models were Model G3 which involved the three most frequent quantitative predictors used to estimate soil bulk density (organic carbon, clay and silt), and Model G10, which included ten qualitative and quantitative input variables such as parent material or tree species. Based on the full dataset, Models G3 and G10 gave R2 values of 0.45 and 0.86, respectively. Model G3 did not significantly outperform the best published model. Even when fitted from an external dataset, it explained only 29% of the variation of ρ with a root mean square error of 0.244 g/cm3. In contrast, the more complex Model G10 outperformed the other models during external validation, with a R2 of 0.67 and a predictive deviation of ±0.168 g/cm3. The variation in forest soil bulk densities was mainly explained by five input variables: organic carbon content, tree species, the coarse fragment content, parent material and sampling depth.
Url:
DOI: 10.1111/j.1475-2743.2010.00305.x
Affiliations:
- France, Italie
- Aquitaine, Centre-Val de Loire, Grand Est, Lorraine (région), Nouvelle-Aquitaine, Région Centre
- Gradignan, Orléans, Velaine‐en‐Haye
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 001752
- to stream Istex, to step Curation: 001744
- to stream Istex, to step Checkpoint: 000288
- to stream Main, to step Merge: 000575
- to stream Main, to step Curation: 000573
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Estimating forest soil bulk density using boosted regression modelling</title><author><name sortKey="Jalabert, S S M" sort="Jalabert, S S M" uniqKey="Jalabert S" first="S. S. M." last="Jalabert">S. S. M. Jalabert</name></author><author><name sortKey="Martin, M P" sort="Martin, M P" uniqKey="Martin M" first="M. P." last="Martin">M. P. Martin</name></author><author><name sortKey="Renaud, J" sort="Renaud, J" uniqKey="Renaud J" first="J." last="Renaud">J. Renaud</name></author><author><name sortKey="Boulonne, L" sort="Boulonne, L" uniqKey="Boulonne L" first="L." last="Boulonne">L. Boulonne</name></author><author><name sortKey="Jolivet, C" sort="Jolivet, C" uniqKey="Jolivet C" first="C." last="Jolivet">C. Jolivet</name></author><author><name sortKey="Montanarella, L" sort="Montanarella, L" uniqKey="Montanarella L" first="L." last="Montanarella">L. Montanarella</name></author><author><name sortKey="Arrouays, D" sort="Arrouays, D" uniqKey="Arrouays D" first="D." last="Arrouays">D. Arrouays</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:293CD78BBF76B3551D9B631970FF655B4B3DB475</idno><date when="2010" year="2010">2010</date><idno type="doi">10.1111/j.1475-2743.2010.00305.x</idno><idno type="url">https://api.istex.fr/document/293CD78BBF76B3551D9B631970FF655B4B3DB475/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001752</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001752</idno><idno type="wicri:Area/Istex/Curation">001744</idno><idno type="wicri:Area/Istex/Checkpoint">000288</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000288</idno><idno type="wicri:doubleKey">0266-0032:2010:Jalabert S:estimating:forest:soil</idno><idno type="wicri:Area/Main/Merge">000575</idno><idno type="wicri:Area/Main/Curation">000573</idno><idno type="wicri:Area/Main/Exploration">000573</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Estimating forest soil bulk density using boosted regression modelling</title><author><name sortKey="Jalabert, S S M" sort="Jalabert, S S M" uniqKey="Jalabert S" first="S. S. M." last="Jalabert">S. S. M. Jalabert</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>ENITA de Bordeaux, UF Agrosystèmes et Forêts, EA Géoressources et Environnement, 1 Cours du Général de Gaulle, CS 40201, 33175 Gradignan CEDEX</wicri:regionArea><placeName><region type="region" nuts="2">Nouvelle-Aquitaine</region><region type="old region" nuts="2">Aquitaine</region><settlement type="city">Gradignan</settlement></placeName></affiliation><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>INRA, US 1106 INFOSOL, 2163 Avenue de la Pomme de Pin, CS 40001 ARDON, 45000 Orléans CEDEX 2</wicri:regionArea><placeName><region type="region" nuts="2">Centre-Val de Loire</region><region type="old region" nuts="2">Région Centre</region><settlement type="city">Orléans</settlement></placeName></affiliation></author><author><name sortKey="Martin, M P" sort="Martin, M P" uniqKey="Martin M" first="M. P." last="Martin">M. P. Martin</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>INRA, US 1106 INFOSOL, 2163 Avenue de la Pomme de Pin, CS 40001 ARDON, 45000 Orléans CEDEX 2</wicri:regionArea><placeName><region type="region" nuts="2">Centre-Val de Loire</region><region type="old region" nuts="2">Région Centre</region><settlement type="city">Orléans</settlement></placeName></affiliation></author><author><name sortKey="Renaud, J" sort="Renaud, J" uniqKey="Renaud J" first="J." last="Renaud">J. Renaud</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>Office National des Forêts, Mission Expérimentations et Méthodes, Les Merises, Parc de Haye, 54840 Velaine‐en‐Haye</wicri:regionArea><placeName><region type="region" nuts="2">Grand Est</region><region type="old region" nuts="2">Lorraine (région)</region><settlement type="city">Velaine‐en‐Haye</settlement></placeName></affiliation></author><author><name sortKey="Boulonne, L" sort="Boulonne, L" uniqKey="Boulonne L" first="L." last="Boulonne">L. Boulonne</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>INRA, US 1106 INFOSOL, 2163 Avenue de la Pomme de Pin, CS 40001 ARDON, 45000 Orléans CEDEX 2</wicri:regionArea><placeName><region type="region" nuts="2">Centre-Val de Loire</region><region type="old region" nuts="2">Région Centre</region><settlement type="city">Orléans</settlement></placeName></affiliation></author><author><name sortKey="Jolivet, C" sort="Jolivet, C" uniqKey="Jolivet C" first="C." last="Jolivet">C. Jolivet</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>INRA, US 1106 INFOSOL, 2163 Avenue de la Pomme de Pin, CS 40001 ARDON, 45000 Orléans CEDEX 2</wicri:regionArea><placeName><region type="region" nuts="2">Centre-Val de Loire</region><region type="old region" nuts="2">Région Centre</region><settlement type="city">Orléans</settlement></placeName></affiliation></author><author><name sortKey="Montanarella, L" sort="Montanarella, L" uniqKey="Montanarella L" first="L." last="Montanarella">L. Montanarella</name><affiliation wicri:level="1"><country xml:lang="fr">Italie</country><wicri:regionArea>European Commission DG Joint Research Centre, Institute for Environment and Sustainability, Land Management and Natural Hazards Unit, Via Fermi, ISPRA (VA) I‐21020</wicri:regionArea><wicri:noRegion>ISPRA (VA) I‐21020</wicri:noRegion></affiliation></author><author><name sortKey="Arrouays, D" sort="Arrouays, D" uniqKey="Arrouays D" first="D." last="Arrouays">D. Arrouays</name><affiliation wicri:level="3"><country xml:lang="fr">France</country><wicri:regionArea>INRA, US 1106 INFOSOL, 2163 Avenue de la Pomme de Pin, CS 40001 ARDON, 45000 Orléans CEDEX 2</wicri:regionArea><placeName><region type="region" nuts="2">Centre-Val de Loire</region><region type="old region" nuts="2">Région Centre</region><settlement type="city">Orléans</settlement></placeName></affiliation></author></analytic><monogr></monogr><series><title level="j">Soil Use and Management</title><idno type="ISSN">0266-0032</idno><idno type="eISSN">1475-2743</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><date type="published" when="2010-12">2010-12</date><biblScope unit="volume">26</biblScope><biblScope unit="issue">4</biblScope><biblScope unit="page" from="516">516</biblScope><biblScope unit="page" to="528">528</biblScope></imprint><idno type="ISSN">0266-0032</idno></series><idno type="istex">293CD78BBF76B3551D9B631970FF655B4B3DB475</idno><idno type="DOI">10.1111/j.1475-2743.2010.00305.x</idno><idno type="ArticleID">SUM305</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0266-0032</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Biosoil</term><term>Bulk density</term><term>boosted regression trees</term><term>forest soils</term><term>pedotransfer functions</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Soil bulk density (ρ) is an important physical property, but its measurement is frequently lacking in soil surveys due to the time‐consuming nature of making the measurement. As a result pedotransfer functions (PTFs) have been developed to predict ρ from other more easily available soil properties. These functions are generally derived from regression methods that aim to fit a single model. In this study, we use a technique called Generalized Boosted Regression Modelling (GBM; Ridgeway, 2006) which combines two algorithms: regression trees and boosting. We built two models and compared their predictive performance with published PTFs. All the functions were fitted based on the French forest soil dataset for the European demonstration Biosoil project. The two GBM models were Model G3 which involved the three most frequent quantitative predictors used to estimate soil bulk density (organic carbon, clay and silt), and Model G10, which included ten qualitative and quantitative input variables such as parent material or tree species. Based on the full dataset, Models G3 and G10 gave R2 values of 0.45 and 0.86, respectively. Model G3 did not significantly outperform the best published model. Even when fitted from an external dataset, it explained only 29% of the variation of ρ with a root mean square error of 0.244 g/cm3. In contrast, the more complex Model G10 outperformed the other models during external validation, with a R2 of 0.67 and a predictive deviation of ±0.168 g/cm3. The variation in forest soil bulk densities was mainly explained by five input variables: organic carbon content, tree species, the coarse fragment content, parent material and sampling depth.</div></front></TEI><affiliations><list><country><li>France</li><li>Italie</li></country><region><li>Aquitaine</li><li>Centre-Val de Loire</li><li>Grand Est</li><li>Lorraine (région)</li><li>Nouvelle-Aquitaine</li><li>Région Centre</li></region><settlement><li>Gradignan</li><li>Orléans</li><li>Velaine‐en‐Haye</li></settlement></list><tree><country name="France"><region name="Nouvelle-Aquitaine"><name sortKey="Jalabert, S S M" sort="Jalabert, S S M" uniqKey="Jalabert S" first="S. S. M." last="Jalabert">S. S. M. Jalabert</name></region><name sortKey="Arrouays, D" sort="Arrouays, D" uniqKey="Arrouays D" first="D." last="Arrouays">D. Arrouays</name><name sortKey="Boulonne, L" sort="Boulonne, L" uniqKey="Boulonne L" first="L." last="Boulonne">L. Boulonne</name><name sortKey="Jalabert, S S M" sort="Jalabert, S S M" uniqKey="Jalabert S" first="S. S. M." last="Jalabert">S. S. M. Jalabert</name><name sortKey="Jolivet, C" sort="Jolivet, C" uniqKey="Jolivet C" first="C." last="Jolivet">C. Jolivet</name><name sortKey="Martin, M P" sort="Martin, M P" uniqKey="Martin M" first="M. P." last="Martin">M. P. Martin</name><name sortKey="Renaud, J" sort="Renaud, J" uniqKey="Renaud J" first="J." last="Renaud">J. Renaud</name></country><country name="Italie"><noRegion><name sortKey="Montanarella, L" sort="Montanarella, L" uniqKey="Montanarella L" first="L." last="Montanarella">L. Montanarella</name></noRegion></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Bois/explor/CheneBelgiqueV2/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000573 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000573 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Bois
|area= CheneBelgiqueV2
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:293CD78BBF76B3551D9B631970FF655B4B3DB475
|texte= Estimating forest soil bulk density using boosted regression modelling
}}
| This area was generated with Dilib version V0.6.27. |  |