Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index
Numerical modelling of water flow allows for the prediction of rainwater partitioning into evaporation, deep drainage, and transpiration for different seasonal crop and soil type scenarios. We proposed and tested a single indicator for drainage estimation, the soil drainability index (SDI) based on...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-04-01
|
| Series: | Soil Systems |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2571-8789/3/2/30 |
| _version_ | 1818108764750872576 |
|---|---|
| author | Ali Mehmandoost Kotlar Bo V. Iversen Quirijn de Jong van Lier |
| author_facet | Ali Mehmandoost Kotlar Bo V. Iversen Quirijn de Jong van Lier |
| author_sort | Ali Mehmandoost Kotlar |
| collection | DOAJ |
| description | Numerical modelling of water flow allows for the prediction of rainwater partitioning into evaporation, deep drainage, and transpiration for different seasonal crop and soil type scenarios. We proposed and tested a single indicator for drainage estimation, the soil drainability index (SDI) based on the near saturated hydraulic conductivity of each layer. We studied rainfall partitioning for eight soils from Brazil and seven different real and generated weather data under scenarios without crop and with a permanent grass cover with three rooting depths, using the HYDRUS-1D model. The SDI showed a good correlation to simulated drainage of the soils. Moreover, well-trained supervised machine-learning methods, including the linear and stepwise linear models (LM, SWLM), besides ensemble regression with boosting and bagging algorithm (ENS-LB, ENS-B), support vector machines (SVMs), and Gaussian process regression (GPR), predicted monthly drainage from bare soil (BS) and grass covered lands (G) using soil–plant–atmosphere parameters (i.e., SDI, monthly precipitation, and evapotranspiration or transpiration). The RMSE values for testing data in BS and G were low, around 1.2 and 1.5 cm month<sup>−1</sup> for all methods. |
| first_indexed | 2024-12-11T02:20:33Z |
| format | Article |
| id | doaj.art-fc5e391dd8bb4cf69e0add03320d3abc |
| institution | Directory Open Access Journal |
| issn | 2571-8789 |
| language | English |
| last_indexed | 2024-12-11T02:20:33Z |
| publishDate | 2019-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Soil Systems |
| spelling | doaj.art-fc5e391dd8bb4cf69e0add03320d3abc2022-12-22T01:24:05ZengMDPI AGSoil Systems2571-87892019-04-01323010.3390/soilsystems3020030soilsystems3020030Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability IndexAli Mehmandoost Kotlar0Bo V. Iversen1Quirijn de Jong van Lier2Tropical Ecosystems Division, University of São Paulo, Centre for Nuclear Energy in Agriculture (CENA/USP), Caixa Postal 96, Piracicaba (SP) 13416-903, BrazilDepartment of Agroecology, Aarhus Univ., Blichers Allé 20, 8830 Tjele, DenmarkTropical Ecosystems Division, University of São Paulo, Centre for Nuclear Energy in Agriculture (CENA/USP), Caixa Postal 96, Piracicaba (SP) 13416-903, BrazilNumerical modelling of water flow allows for the prediction of rainwater partitioning into evaporation, deep drainage, and transpiration for different seasonal crop and soil type scenarios. We proposed and tested a single indicator for drainage estimation, the soil drainability index (SDI) based on the near saturated hydraulic conductivity of each layer. We studied rainfall partitioning for eight soils from Brazil and seven different real and generated weather data under scenarios without crop and with a permanent grass cover with three rooting depths, using the HYDRUS-1D model. The SDI showed a good correlation to simulated drainage of the soils. Moreover, well-trained supervised machine-learning methods, including the linear and stepwise linear models (LM, SWLM), besides ensemble regression with boosting and bagging algorithm (ENS-LB, ENS-B), support vector machines (SVMs), and Gaussian process regression (GPR), predicted monthly drainage from bare soil (BS) and grass covered lands (G) using soil–plant–atmosphere parameters (i.e., SDI, monthly precipitation, and evapotranspiration or transpiration). The RMSE values for testing data in BS and G were low, around 1.2 and 1.5 cm month<sup>−1</sup> for all methods.https://www.mdpi.com/2571-8789/3/2/30evapotranspirationhydraulic conductivityHYDRUS-1Dsupervised learning modelssubsurface drainage |
| spellingShingle | Ali Mehmandoost Kotlar Bo V. Iversen Quirijn de Jong van Lier Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index Soil Systems evapotranspiration hydraulic conductivity HYDRUS-1D supervised learning models subsurface drainage |
| title | Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index |
| title_full | Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index |
| title_fullStr | Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index |
| title_full_unstemmed | Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index |
| title_short | Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index |
| title_sort | machine learning based prediction of drainage in layered soils using a soil drainability index |
| topic | evapotranspiration hydraulic conductivity HYDRUS-1D supervised learning models subsurface drainage |
| url | https://www.mdpi.com/2571-8789/3/2/30 |
| work_keys_str_mv | AT alimehmandoostkotlar machinelearningbasedpredictionofdrainageinlayeredsoilsusingasoildrainabilityindex AT boviversen machinelearningbasedpredictionofdrainageinlayeredsoilsusingasoildrainabilityindex AT quirijndejongvanlier machinelearningbasedpredictionofdrainageinlayeredsoilsusingasoildrainabilityindex |