Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning
<p>Machine-learning algorithms are good at computing non-linear problems and fitting complex composite functions, which makes them an adequate tool for addressing multiple environmental research questions. One important application is the development of pedotransfer functions (PTFs). This stud...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2020-06-01
|
| Series: | SOIL |
| Online Access: | https://www.soil-journal.net/6/215/2020/soil-6-215-2020.pdf |
| _version_ | 1818188011028873216 |
|---|---|
| author | A. Gebauer M. Ellinger V. M. Brito Gomez M. Ließ |
| author_facet | A. Gebauer M. Ellinger V. M. Brito Gomez M. Ließ |
| author_sort | A. Gebauer |
| collection | DOAJ |
| description | <p>Machine-learning algorithms are good at computing
non-linear problems and fitting complex composite functions, which makes
them an adequate tool for addressing multiple environmental research questions.
One important application is the development of pedotransfer functions
(PTFs). This study aims to develop water retention PTFs for two remote
tropical mountain regions with rather different soil landscapes:
(1) those dominated by peat soils and soils under volcanic influence with high organic matter
contents and (2) those dominated by tropical mineral soils. Two tuning procedures were
compared to fit boosted regression tree models: (1) tuning with grid search,
which is the standard approach in pedometrics; and (2) tuning with
differential evolution optimization. A nested cross-validation approach was
applied to generate robust models. The area-specific PTFs developed outperform other more general PTFs. Furthermore, the first PTF for typical soils of
Páramo landscapes (Ecuador), i.e., organic soils under volcanic
influence, is presented. Overall, the results confirmed the differential
evolution algorithm's high potential for tuning machine-learning models.
While models based on tuning with grid search roughly predicted the response
variables' mean for both areas, models applying the differential evolution
algorithm for parameter tuning explained up to 25 times more of the response
variables' variance.</p> |
| first_indexed | 2024-12-11T23:20:08Z |
| format | Article |
| id | doaj.art-6f3a1d2c013245658248464a1b589a81 |
| institution | Directory Open Access Journal |
| issn | 2199-3971 2199-398X |
| language | English |
| last_indexed | 2024-12-11T23:20:08Z |
| publishDate | 2020-06-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | SOIL |
| spelling | doaj.art-6f3a1d2c013245658248464a1b589a812022-12-22T00:46:22ZengCopernicus PublicationsSOIL2199-39712199-398X2020-06-01621522910.5194/soil-6-215-2020Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learningA. Gebauer0M. Ellinger1V. M. Brito Gomez2M. Ließ3Department of Soil System Science, Helmholtz Centre for Environmental Research – UFZ, Halle (Saale), GermanyDepartment of Soil System Science, Helmholtz Centre for Environmental Research – UFZ, Halle (Saale), GermanyDepartamento de Recursos Hídricos y Ciencias Ambientales, Facultad de Ciencias Agropecuarias, Universidad de Cuenca, Cuenca, EcuadorDepartment of Soil System Science, Helmholtz Centre for Environmental Research – UFZ, Halle (Saale), Germany<p>Machine-learning algorithms are good at computing non-linear problems and fitting complex composite functions, which makes them an adequate tool for addressing multiple environmental research questions. One important application is the development of pedotransfer functions (PTFs). This study aims to develop water retention PTFs for two remote tropical mountain regions with rather different soil landscapes: (1) those dominated by peat soils and soils under volcanic influence with high organic matter contents and (2) those dominated by tropical mineral soils. Two tuning procedures were compared to fit boosted regression tree models: (1) tuning with grid search, which is the standard approach in pedometrics; and (2) tuning with differential evolution optimization. A nested cross-validation approach was applied to generate robust models. The area-specific PTFs developed outperform other more general PTFs. Furthermore, the first PTF for typical soils of Páramo landscapes (Ecuador), i.e., organic soils under volcanic influence, is presented. Overall, the results confirmed the differential evolution algorithm's high potential for tuning machine-learning models. While models based on tuning with grid search roughly predicted the response variables' mean for both areas, models applying the differential evolution algorithm for parameter tuning explained up to 25 times more of the response variables' variance.</p>https://www.soil-journal.net/6/215/2020/soil-6-215-2020.pdf |
| spellingShingle | A. Gebauer M. Ellinger V. M. Brito Gomez M. Ließ Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning SOIL |
| title | Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning |
| title_full | Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning |
| title_fullStr | Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning |
| title_full_unstemmed | Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning |
| title_short | Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning |
| title_sort | development of pedotransfer functions for water retention in tropical mountain soil landscapes spotlight on parameter tuning in machine learning |
| url | https://www.soil-journal.net/6/215/2020/soil-6-215-2020.pdf |
| work_keys_str_mv | AT agebauer developmentofpedotransferfunctionsforwaterretentionintropicalmountainsoillandscapesspotlightonparametertuninginmachinelearning AT mellinger developmentofpedotransferfunctionsforwaterretentionintropicalmountainsoillandscapesspotlightonparametertuninginmachinelearning AT vmbritogomez developmentofpedotransferfunctionsforwaterretentionintropicalmountainsoillandscapesspotlightonparametertuninginmachinelearning AT mließ developmentofpedotransferfunctionsforwaterretentionintropicalmountainsoillandscapesspotlightonparametertuninginmachinelearning |