An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain
Identifieur interne : 000F47 ( Istex/Corpus ); précédent : 000F46; suivant : 000F48An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain
Auteurs : D. De La Rosa ; F. Mayol ; J. A Moreno ; T. Bons N ; S. LozanoSource :
- Agriculture, Ecosystems and Environment [ 0167-8809 ] ; 1999.
English descriptors
- KwdEn :
Abstract
Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (Vi) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (r=0.91) than the previous approach (r=0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.
Url:
DOI: 10.1016/S0167-8809(99)00050-X
Links to Exploration step
ISTEX:8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title><author><name sortKey="De La Rosa, D" sort="De La Rosa, D" uniqKey="De La Rosa D" first="D" last="De La Rosa">D. De La Rosa</name><affiliation><mods:affiliation>E-mail: diego@irnase.csic.es</mods:affiliation></affiliation><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Mayol, F" sort="Mayol, F" uniqKey="Mayol F" first="F" last="Mayol">F. Mayol</name><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Moreno, J A" sort="Moreno, J A" uniqKey="Moreno J" first="J. A" last="Moreno">J. A Moreno</name><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Bons N, T" sort="Bons N, T" uniqKey="Bons N T" first="T" last="Bons N">T. Bons N</name><affiliation><mods:affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Lozano, S" sort="Lozano, S" uniqKey="Lozano S" first="S" last="Lozano">S. Lozano</name><affiliation><mods:affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369</idno><date when="1999" year="1999">1999</date><idno type="doi">10.1016/S0167-8809(99)00050-X</idno><idno type="url">https://api.istex.fr/document/8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000F47</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000F47</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title><author><name sortKey="De La Rosa, D" sort="De La Rosa, D" uniqKey="De La Rosa D" first="D" last="De La Rosa">D. De La Rosa</name><affiliation><mods:affiliation>E-mail: diego@irnase.csic.es</mods:affiliation></affiliation><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Mayol, F" sort="Mayol, F" uniqKey="Mayol F" first="F" last="Mayol">F. Mayol</name><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Moreno, J A" sort="Moreno, J A" uniqKey="Moreno J" first="J. A" last="Moreno">J. A Moreno</name><affiliation><mods:affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Bons N, T" sort="Bons N, T" uniqKey="Bons N T" first="T" last="Bons N">T. Bons N</name><affiliation><mods:affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</mods:affiliation></affiliation></author><author><name sortKey="Lozano, S" sort="Lozano, S" uniqKey="Lozano S" first="S" last="Lozano">S. Lozano</name><affiliation><mods:affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Agriculture, Ecosystems and Environment</title><title level="j" type="abbrev">AGEE</title><idno type="ISSN">0167-8809</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1999">1999</date><biblScope unit="volume">73</biblScope><biblScope unit="issue">3</biblScope><biblScope unit="page" from="211">211</biblScope><biblScope unit="page" to="226">226</biblScope></imprint><idno type="ISSN">0167-8809</idno></series><idno type="istex">8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369</idno><idno type="DOI">10.1016/S0167-8809(99)00050-X</idno><idno type="PII">S0167-8809(99)00050-X</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0167-8809</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Agricultural management practices</term><term>Expert systems</term><term>Lowland areas</term><term>Neural networks</term><term>Water erosion</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (Vi) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (r=0.91) than the previous approach (r=0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.</div></front></TEI><istex><corpusName>elsevier</corpusName><author><json:item><name>D de la Rosa</name><affiliations><json:string>E-mail: diego@irnase.csic.es</json:string><json:string>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</json:string></affiliations></json:item><json:item><name>F Mayol</name><affiliations><json:string>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</json:string></affiliations></json:item><json:item><name>J.A Moreno</name><affiliations><json:string>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</json:string></affiliations></json:item><json:item><name>T Bonsón</name><affiliations><json:string>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</json:string></affiliations></json:item><json:item><name>S Lozano</name><affiliations><json:string>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Water erosion</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Lowland areas</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Expert systems</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Neural networks</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Agricultural management practices</value></json:item></subject><language><json:string>eng</json:string></language><originalGenre><json:string>Full-length article</json:string></originalGenre><abstract>Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (Vi) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (r=0.91) than the previous approach (r=0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.</abstract><qualityIndicators><score>8</score><pdfVersion>1.2</pdfVersion><pdfPageSize>612 x 792 pts (letter)</pdfPageSize><refBibsNative>true</refBibsNative><keywordCount>5</keywordCount><abstractCharCount>2184</abstractCharCount><pdfWordCount>6700</pdfWordCount><pdfCharCount>44204</pdfCharCount><pdfPageCount>16</pdfPageCount><abstractWordCount>323</abstractWordCount></qualityIndicators><title>An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title><pii><json:string>S0167-8809(99)00050-X</json:string></pii><genre><json:string>research-article</json:string></genre><host><volume>73</volume><pii><json:string>S0167-8809(00)X0052-7</json:string></pii><pages><last>226</last><first>211</first></pages><issn><json:string>0167-8809</json:string></issn><issue>3</issue><genre><json:string>journal</json:string></genre><language><json:string>unknown</json:string></language><title>Agriculture, Ecosystems and Environment</title><publicationDate>1999</publicationDate></host><categories><wos><json:string>science</json:string><json:string>environmental sciences</json:string><json:string>ecology</json:string><json:string>agriculture, multidisciplinary</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>1999</publicationDate><copyrightDate>1999</copyrightDate><doi><json:string>10.1016/S0167-8809(99)00050-X</json:string></doi><id>8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369</id><score>0.039806418</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>ELSEVIER</publisher><availability><p>©1999 Elsevier Science B.V.</p></availability><date>1999</date></publicationStmt><notesStmt><note type="content">Fig. 1: Flow chart of the agricultural soil erosion land evaluation ImpelERO model.</note><note type="content">Fig. 2: Pathway of the decision tree constructed for the management quality: crop protection. From the generalisation levels of the input characteristics, the path is followed until a severity level (1c=low to 4c=very high) is encountered.</note><note type="content">Fig. 3: Structure of the correlation-cascade neural network developed; showing the interrelationships between the land and management qualities: LQr=runoff erosivity, LQt=relief hazard, LQk=soil erodibility, MQc=crop protection, MQz=tillage translocation and MQy=productivity influence, to reproduce the vulnerability index (Vi).</note><note type="content">Fig. 4: Predicted versus observed soil erosion vulnerability index (Vi) for the 237 field units observed, one by five portions of columns plotted, in Andalusia region.</note><note type="content">Fig. 5: Sensitivity diagram showing effects of land and management qualities: LQr=runoff erosivity, LQt=relief hazard, LQk=soil erodibility, MQc=crop protection, MQz=tillage translocation and MQy=productivity influence, on soil erosion vulnerability index (Vi), through the neural network analysis.</note><note type="content">Fig. 6: Comparison of observed and predicted soil erosion values (Vc=vulnerability class) using the ACCESS model.</note><note type="content">Fig. 7: Comparison of observed and predicted soil erosion values (Vc=vulnerability class, and Vi=vulnerability index) using the ImpelERO model.</note><note type="content">Table 1: Relation of the selected 34 Major Land Resource Area (MLRAs) of lowland Andalucia region, which represent a total of 4323300ha</note><note type="content">Table 2: Monthly means of temperature and precipitation in Sevilla station, 1961–90 period, and oscilation range (minimum–maximum values) for all the 34 stations considered</note><note type="content">Table 3: Severity of erosion problem related in terms of status, risk, rate and vulnerability for the Andalucia region survey</note><note type="content">Table 4: Severity of erosion problem at the 237 field units observed in the Andalucia region</note><note type="content">Table 5: Relation of land/management qualities and subqualities, and associated land/management characteristics considered to develop the decision trees submodel</note><note type="content">Table 6: Neural network parameters of the ImpelERO model</note></notesStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title><author xml:id="author-1"><persName><forename type="first">D</forename><surname>de la Rosa</surname></persName><email>diego@irnase.csic.es</email><note type="correspondence"><p>Corresponding author. Tel.: +34-95-624711; fax: +34-954-624002</p></note><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation></author><author xml:id="author-2"><persName><forename type="first">F</forename><surname>Mayol</surname></persName><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation></author><author xml:id="author-3"><persName><forename type="first">J.A</forename><surname>Moreno</surname></persName><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation></author><author xml:id="author-4"><persName><forename type="first">T</forename><surname>Bonsón</surname></persName><affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</affiliation></author><author xml:id="author-5"><persName><forename type="first">S</forename><surname>Lozano</surname></persName><affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</affiliation></author></analytic><monogr><title level="j">Agriculture, Ecosystems and Environment</title><title level="j" type="abbrev">AGEE</title><idno type="pISSN">0167-8809</idno><idno type="PII">S0167-8809(00)X0052-7</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1999"></date><biblScope unit="volume">73</biblScope><biblScope unit="issue">3</biblScope><biblScope unit="page" from="211">211</biblScope><biblScope unit="page" to="226">226</biblScope></imprint></monogr><idno type="istex">8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369</idno><idno type="DOI">10.1016/S0167-8809(99)00050-X</idno><idno type="PII">S0167-8809(99)00050-X</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>1999</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (Vi) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (r=0.91) than the previous approach (r=0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.</p></abstract><textClass xml:lang="en"><keywords scheme="keyword"><list><head>Keywords</head><item><term>Water erosion</term></item><item><term>Lowland areas</term></item><item><term>Expert systems</term></item><item><term>Neural networks</term></item><item><term>Agricultural management practices</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="1998-12-02">Modified</change><change when="1999">Published</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="Elsevier, elements deleted: ce:floats; body; tail"><istex:xmlDeclaration>version="1.0" encoding="utf-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//ES//DTD journal article DTD version 4.5.2//EN//XML" URI="art452.dtd" name="istex:docType"><istex:entity SYSTEM="gr1" NDATA="IMAGE" name="gr1"></istex:entity><istex:entity SYSTEM="gr2" NDATA="IMAGE" name="gr2"></istex:entity><istex:entity SYSTEM="gr3" NDATA="IMAGE" name="gr3"></istex:entity><istex:entity SYSTEM="gr4" NDATA="IMAGE" name="gr4"></istex:entity><istex:entity SYSTEM="gr5" NDATA="IMAGE" name="gr5"></istex:entity><istex:entity SYSTEM="gr6" NDATA="IMAGE" name="gr6"></istex:entity><istex:entity SYSTEM="gr7" NDATA="IMAGE" name="gr7"></istex:entity></istex:docType><istex:document><converted-article version="4.5.2" docsubtype="fla" xml:lang="en"><item-info><jid>AGEE</jid><aid>1451</aid><ce:pii>S0167-8809(99)00050-X</ce:pii><ce:doi>10.1016/S0167-8809(99)00050-X</ce:doi><ce:copyright type="full-transfer" year="1999">Elsevier Science B.V.</ce:copyright></item-info><head><ce:title>An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</ce:title><ce:author-group><ce:author><ce:given-name>D</ce:given-name><ce:surname>de la Rosa</ce:surname><ce:cross-ref refid="AFF1"><ce:sup>a</ce:sup></ce:cross-ref><ce:cross-ref refid="CORR1">*</ce:cross-ref><ce:e-address>diego@irnase.csic.es</ce:e-address></ce:author><ce:author><ce:given-name>F</ce:given-name><ce:surname>Mayol</ce:surname><ce:cross-ref refid="AFF1"><ce:sup>a</ce:sup></ce:cross-ref></ce:author><ce:author><ce:given-name>J.A</ce:given-name><ce:surname>Moreno</ce:surname><ce:cross-ref refid="AFF1"><ce:sup>a</ce:sup></ce:cross-ref></ce:author><ce:author><ce:given-name>T</ce:given-name><ce:surname>Bonsón</ce:surname><ce:cross-ref refid="AFF2"><ce:sup>b</ce:sup></ce:cross-ref></ce:author><ce:author><ce:given-name>S</ce:given-name><ce:surname>Lozano</ce:surname><ce:cross-ref refid="AFF2"><ce:sup>b</ce:sup></ce:cross-ref></ce:author><ce:affiliation id="AFF1"><ce:label>a</ce:label><ce:textfn>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</ce:textfn></ce:affiliation><ce:affiliation id="AFF2"><ce:label>b</ce:label><ce:textfn>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</ce:textfn></ce:affiliation><ce:correspondence id="CORR1"><ce:label>*</ce:label><ce:text>Corresponding author. Tel.: +34-95-624711; fax: +34-954-624002</ce:text></ce:correspondence></ce:author-group><ce:date-received day="6" month="4" year="1998"></ce:date-received><ce:date-revised day="2" month="12" year="1998"></ce:date-revised><ce:date-accepted day="23" month="2" year="1999"></ce:date-accepted><ce:abstract><ce:section-title>Abstract</ce:section-title><ce:abstract-sec><ce:simple-para>Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (<ce:italic>V</ce:italic><ce:inf>i</ce:inf>) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (<ce:italic>r</ce:italic><ce:hsp sp="0.25"></ce:hsp>=<ce:hsp sp="0.25"></ce:hsp>0.91) than the previous approach (<ce:italic>r</ce:italic><ce:hsp sp="0.25"></ce:hsp>=<ce:hsp sp="0.25"></ce:hsp>0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:keywords class="keyword" xml:lang="en"><ce:section-title>Keywords</ce:section-title><ce:keyword><ce:text>Water erosion</ce:text></ce:keyword><ce:keyword><ce:text>Lowland areas</ce:text></ce:keyword><ce:keyword><ce:text>Expert systems</ce:text></ce:keyword><ce:keyword><ce:text>Neural networks</ce:text></ce:keyword><ce:keyword><ce:text>Agricultural management practices</ce:text></ce:keyword></ce:keywords></head></converted-article></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title></titleInfo><titleInfo type="alternative" lang="en" contentType="CDATA"><title>An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain</title></titleInfo><name type="personal"><namePart type="given">D</namePart><namePart type="family">de la Rosa</namePart><affiliation>E-mail: diego@irnase.csic.es</affiliation><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation><description>Corresponding author. Tel.: +34-95-624711; fax: +34-954-624002</description><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">F</namePart><namePart type="family">Mayol</namePart><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">J.A</namePart><namePart type="family">Moreno</namePart><affiliation>Instituto de Recursos Naturales y Agrobiologı́a de Sevilla (IRNAS), Consejo Superior de Investigaciones Cientificas (CSIC) P.O. Box 1052, 41080 Sevilla, Spain</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">T</namePart><namePart type="family">Bonsón</namePart><affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">S</namePart><namePart type="family">Lozano</namePart><affiliation>Escuela Superior de Ingenieros, Universidad de Sevilla, Isla de la Cartuja, 41092 Sevilla, Spain</affiliation><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="research-article" displayLabel="Full-length article"></genre><originInfo><publisher>ELSEVIER</publisher><dateIssued encoding="w3cdtf">1999</dateIssued><dateModified encoding="w3cdtf">1998-12-02</dateModified><copyrightDate encoding="w3cdtf">1999</copyrightDate></originInfo><language><languageTerm type="code" authority="iso639-2b">eng</languageTerm><languageTerm type="code" authority="rfc3066">en</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType></physicalDescription><abstract lang="en">Soil erosion by water is one of today’s most important environmental problems, in great part due to changes in agricultural land use and management. This paper illustrates the formulation, calibration, sensitivity and validation analysis of a hybrid model of expert decision trees and artificial neural networks (named ImpelERO) to evaluate the soil erosion process A total of 237 field units were selected, which represent 34 major land resource areas (MLRAs)for five traditional crops in the Andalusian region, southern Spain. The field units observed cover the whole range of erosion events, from what was considered very small to extreme erosion. Seventy-six per cent of the fields suffered small, moderate or large erosion problems. However, only 14% of the fields suffered very small, and 10% of the fields suffered very large or extreme erosion problems. Because of the complexity of the soil erosion process, and the interrelationships of the parameters, ImpelERO was developed as an Universal Soil Loss Equation (USLE) type model following traditional land evaluation analysis and advanced empirical modelling techniques. Using expert-decision trees, soil survey information and expert knowledge of the soil erosion process were combined through land and management qualities. An artificial neural network approach was then applied to capture the interactions between the land and management qualities and one output: vulnerability index (Vi) to soil erosion. The neural network was trained using the Correlation-cascade algorithm. The trained network estimated the output with a high degree of accuracy (maximum deviation 14%), and also had a good generalisation capacity. By means of correlation analysis, observed erosion vulnerability data were compared with predicted data using a previously developed model and using the ImpelERO model. The latter model gave more accurate results (r=0.91) than the previous approach (r=0.66). Along with the prediction of soil loss by water erosion, ImpelERO could be used as a optimisation tool for selecting the land use and management practices which satisfy the optimum environmental protection including reduction of soil erosion.</abstract><note type="content">Fig. 1: Flow chart of the agricultural soil erosion land evaluation ImpelERO model.</note><note type="content">Fig. 2: Pathway of the decision tree constructed for the management quality: crop protection. From the generalisation levels of the input characteristics, the path is followed until a severity level (1c=low to 4c=very high) is encountered.</note><note type="content">Fig. 3: Structure of the correlation-cascade neural network developed; showing the interrelationships between the land and management qualities: LQr=runoff erosivity, LQt=relief hazard, LQk=soil erodibility, MQc=crop protection, MQz=tillage translocation and MQy=productivity influence, to reproduce the vulnerability index (Vi).</note><note type="content">Fig. 4: Predicted versus observed soil erosion vulnerability index (Vi) for the 237 field units observed, one by five portions of columns plotted, in Andalusia region.</note><note type="content">Fig. 5: Sensitivity diagram showing effects of land and management qualities: LQr=runoff erosivity, LQt=relief hazard, LQk=soil erodibility, MQc=crop protection, MQz=tillage translocation and MQy=productivity influence, on soil erosion vulnerability index (Vi), through the neural network analysis.</note><note type="content">Fig. 6: Comparison of observed and predicted soil erosion values (Vc=vulnerability class) using the ACCESS model.</note><note type="content">Fig. 7: Comparison of observed and predicted soil erosion values (Vc=vulnerability class, and Vi=vulnerability index) using the ImpelERO model.</note><note type="content">Table 1: Relation of the selected 34 Major Land Resource Area (MLRAs) of lowland Andalucia region, which represent a total of 4323300ha</note><note type="content">Table 2: Monthly means of temperature and precipitation in Sevilla station, 1961–90 period, and oscilation range (minimum–maximum values) for all the 34 stations considered</note><note type="content">Table 3: Severity of erosion problem related in terms of status, risk, rate and vulnerability for the Andalucia region survey</note><note type="content">Table 4: Severity of erosion problem at the 237 field units observed in the Andalucia region</note><note type="content">Table 5: Relation of land/management qualities and subqualities, and associated land/management characteristics considered to develop the decision trees submodel</note><note type="content">Table 6: Neural network parameters of the ImpelERO model</note><subject lang="en"><genre>Keywords</genre><topic>Water erosion</topic><topic>Lowland areas</topic><topic>Expert systems</topic><topic>Neural networks</topic><topic>Agricultural management practices</topic></subject><relatedItem type="host"><titleInfo><title>Agriculture, Ecosystems and Environment</title></titleInfo><titleInfo type="abbreviated"><title>AGEE</title></titleInfo><genre type="journal">journal</genre><originInfo><dateIssued encoding="w3cdtf">199905</dateIssued></originInfo><identifier type="ISSN">0167-8809</identifier><identifier type="PII">S0167-8809(00)X0052-7</identifier><part><date>199905</date><detail type="volume"><number>73</number><caption>vol.</caption></detail><detail type="issue"><number>3</number><caption>no.</caption></detail><extent unit="issue pages"><start>177</start><end>286</end></extent><extent unit="pages"><start>211</start><end>226</end></extent></part></relatedItem><identifier type="istex">8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369</identifier><identifier type="DOI">10.1016/S0167-8809(99)00050-X</identifier><identifier type="PII">S0167-8809(99)00050-X</identifier><accessCondition type="use and reproduction" contentType="copyright">©1999 Elsevier Science B.V.</accessCondition><recordInfo><recordContentSource>ELSEVIER</recordContentSource><recordOrigin>Elsevier Science B.V., ©1999</recordOrigin></recordInfo></mods></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Agronomie/explor/SisAgriV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000F47 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 000F47 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Agronomie
|area= SisAgriV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:8D61DF7A6EFD7A3B8ADC01DB749742B3B638C369
|texte= An expert system/neural network model (ImpelERO) for evaluating agricultural soil erosion in Andalucia region, southern Spain
}}
| This area was generated with Dilib version V0.6.28. |  |