Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN)
In the work there were presented two pedotransfer models for determination of saturated hydraulic conductivity, generated by artificial neural networks (ANN). Models were learned based on empirical data obtained in laboratory, on 56 soil samples of differentiated texture. In the first model the inpu...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Publishing House of the University of Agriculture in Krakow
2017-12-01
|
| Series: | Acta Scientiarum Polonorum. Formatio Circumiectus |
| Subjects: | |
| Online Access: | http://www.formatiocircumiectus.actapol.net/pub/16_4_115.pdf |
| _version_ | 1819071145904701440 |
|---|---|
| author | Edyta Kruk Magdalena Malec Sławomir Klatka Andżelika Brodzińska-Cygan Jan Kołodziej |
| author_facet | Edyta Kruk Magdalena Malec Sławomir Klatka Andżelika Brodzińska-Cygan Jan Kołodziej |
| author_sort | Edyta Kruk |
| collection | DOAJ |
| description | In the work there were presented two pedotransfer models for determination of saturated hydraulic conductivity, generated by artificial neural networks (ANN). Models were learned based on empirical data obtained in laboratory, on 56 soil samples of differentiated texture. In the first model the input parameters were: characteristic diameters d10, d50, d60, d90, content of sand, silt and clay fractions, total porosity, bulk density and organic matter content. The MLP type of ANN was used. The best fitted model turned out MLP 10-10-1 with satisfactory quality parameters, for learning 0,996, for testing 0,754 and for validation 1,000. Global sensitivity analysis showed that the highest influence on explanation of relationship between saturated hydraulic conductivity in this model had: clay content (absolute influence 37.7%, d60 (17.1%), sand content (13.5%), d90 (6.0%), bulk density (5.9%) and total porosity (5.7%). The remaining parameters had absolute influence below 5.0%). The next generated ANN model was MLP 6-10-1, with six explaining parameters, of greatest influence. Correlation coefficient attained value 0.989 and 0.955 for the first and the second model. Mean percentage error pointed out underestimation in comparison to laboratory measurement. The values attained 35.9% and 54.8% respectively. Limitation of explaining parameters did not point high deterioration of the ANN model quality. |
| first_indexed | 2024-12-21T17:17:11Z |
| format | Article |
| id | doaj.art-8011e289c6e64c73b06b375512346031 |
| institution | Directory Open Access Journal |
| issn | 1644-0765 |
| language | English |
| last_indexed | 2024-12-21T17:17:11Z |
| publishDate | 2017-12-01 |
| publisher | Publishing House of the University of Agriculture in Krakow |
| record_format | Article |
| series | Acta Scientiarum Polonorum. Formatio Circumiectus |
| spelling | doaj.art-8011e289c6e64c73b06b3755123460312022-12-21T18:56:16ZengPublishing House of the University of Agriculture in KrakowActa Scientiarum Polonorum. Formatio Circumiectus1644-07652017-12-0116411512610.15576/ASP.FC/2017.16.4.115Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN)Edyta Kruk0Magdalena Malec1Sławomir Klatka2Andżelika Brodzińska-Cygan3Jan Kołodziej4University of Agriculture in KrakowUniversity of Agriculture in KrakowUniversity of Agriculture in KrakowUniversity of Agriculture in KrakowUniversity of Agriculture in KrakowIn the work there were presented two pedotransfer models for determination of saturated hydraulic conductivity, generated by artificial neural networks (ANN). Models were learned based on empirical data obtained in laboratory, on 56 soil samples of differentiated texture. In the first model the input parameters were: characteristic diameters d10, d50, d60, d90, content of sand, silt and clay fractions, total porosity, bulk density and organic matter content. The MLP type of ANN was used. The best fitted model turned out MLP 10-10-1 with satisfactory quality parameters, for learning 0,996, for testing 0,754 and for validation 1,000. Global sensitivity analysis showed that the highest influence on explanation of relationship between saturated hydraulic conductivity in this model had: clay content (absolute influence 37.7%, d60 (17.1%), sand content (13.5%), d90 (6.0%), bulk density (5.9%) and total porosity (5.7%). The remaining parameters had absolute influence below 5.0%). The next generated ANN model was MLP 6-10-1, with six explaining parameters, of greatest influence. Correlation coefficient attained value 0.989 and 0.955 for the first and the second model. Mean percentage error pointed out underestimation in comparison to laboratory measurement. The values attained 35.9% and 54.8% respectively. Limitation of explaining parameters did not point high deterioration of the ANN model quality.http://www.formatiocircumiectus.actapol.net/pub/16_4_115.pdfsaturated hydraulic conductivitypedotransfer functionsANN (artificial neural networks) |
| spellingShingle | Edyta Kruk Magdalena Malec Sławomir Klatka Andżelika Brodzińska-Cygan Jan Kołodziej Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) Acta Scientiarum Polonorum. Formatio Circumiectus saturated hydraulic conductivity pedotransfer functions ANN (artificial neural networks) |
| title | Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) |
| title_full | Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) |
| title_fullStr | Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) |
| title_full_unstemmed | Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) |
| title_short | Pedotransfer function for determining saturated hydraulic conductivity using Artificial Neural Network (ANN) |
| title_sort | pedotransfer function for determining saturated hydraulic conductivity using artificial neural network ann |
| topic | saturated hydraulic conductivity pedotransfer functions ANN (artificial neural networks) |
| url | http://www.formatiocircumiectus.actapol.net/pub/16_4_115.pdf |
| work_keys_str_mv | AT edytakruk pedotransferfunctionfordeterminingsaturatedhydraulicconductivityusingartificialneuralnetworkann AT magdalenamalec pedotransferfunctionfordeterminingsaturatedhydraulicconductivityusingartificialneuralnetworkann AT sławomirklatka pedotransferfunctionfordeterminingsaturatedhydraulicconductivityusingartificialneuralnetworkann AT andzelikabrodzinskacygan pedotransferfunctionfordeterminingsaturatedhydraulicconductivityusingartificialneuralnetworkann AT jankołodziej pedotransferfunctionfordeterminingsaturatedhydraulicconductivityusingartificialneuralnetworkann |