Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index

Kotlar, Ali Mehmandoost; Iversen, Bo V.; Lier, Quirijn de Jong van

doi:10.3390/soilsystems3020030

Cited by 3 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In bare soil and grass over lands, Kotlar, Iversen, van Lier, et al. (2019) created a soil drainage index based on hydraulic conductivity in soil profiles, which outperformed typical modeling methods by using fewer parameters and data. This index was then integrated into SVM and Gaussian process regression to successfully predict drainage volumes with an RMSE ranging from 1.2 to 1.5 cm/month.…”

Section: Applicationsmentioning

confidence: 99%

Machine learning applications in vadose zone hydrology: A review

Li,

Nieber,

Kumar

2024

Vadose Zone Journal

View full text Add to dashboard Cite

Machine learning (ML) has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such developments. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic ML models with relatively limited applications of deep learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics‐based vadose zone principles into the ML approaches. To facilitate this interdisciplinary research of ML in vadose zone characterization and processes, a paradigm of knowledge‐guided machine learning is suggested along with other data‐driven and ML model‐based research suggestions to advance future research.

show abstract

Section: Applicationsmentioning

confidence: 99%

Machine learning applications in vadose zone hydrology: A review

Li,

Nieber,

Kumar

2024

Vadose Zone Journal

View full text Add to dashboard Cite

show abstract

Investigation of estimation of hydraulic parameters in heterogeneous soil

Ali

Zhang

Wang

et al. 2023

Proceedings of the Institution of Civil Engineers - Water Manag

View full text Add to dashboard Cite

Data availability for in situ spatial variability assessment of hydraulic parameters is always limited in the vadose zone. In this work, laboratory and in situ experimental methods of parameter estimation were compared to investigate the best estimation method for heterogeneous soil. The Marquardt–Levenberg and non-linear least-squares optimisation algorithms were used for parameter estimation. The simulation error was minimised by selecting sensitive parameters during the numerical solution. The shape factor n was found to be the most sensitive parameter, followed by water content θs, saturated hydraulic conductivity (SHC) and the inverse of the air entry α. Compared with the in situ cumulative infiltration and simultaneous methods, the outflow method resulted in the best fit by minimising the error. During the comparison of outflow and cumulative infiltration methods, only θs showed a significant difference (p = 0.00). On the other hand, SHC showed a non-significant difference (p = 0.439) when the outflow and simultaneous methods were compared. During model predictions, the SHC measured by the simultaneous method showed reasonable estimates for surface horizon and weak correlations (0.79 and 0.77) with deep soil water content, which could be improved by adding more hydraulic parameters. The cumulative infiltration numerical solution resulted in the most reliable estimates of hydraulic parameters for in situ heterogeneous soil.

show abstract

Soil Hydraulic Properties Determined by Inverse Modeling of Drip Infiltrometer Experiments Extended with Pedotransfer Functions

Kotlar

Varvaris²,

Lier

et al. 2019

Vadose Zone Journal

View full text Add to dashboard Cite

Core Ideas A robust PTF was developed to predict water contents at −1, −10, and −158 m tension. Drip infiltrometer experiments were inversely modeled to predict soil hydraulic properties. Both θ(h) and K(h) can be accurately estimated from experimental data together with PTFs. A transient flow experiment using automated drip infiltrometers (ADIs) was performed on soil columns (about 6 dm3) large enough to incorporate macropore flow effects. We investigated to what extent the estimated soil hydraulic parameters obtained from inverse modeling of these experiments are reliable. A machine learning based pedotransfer function (PTF) for prediction of water content at −1, −10, and −158 m pressure head was developed. Sensitivity analysis of the van Genuchten parameters (residual and saturated water content θr and θs, fitting parameters α, n, and λ, and saturated hydraulic conductivity Ks) in soils of sandy, silty, and clayey textures showed that the temporal variation of pressure heads in ADI scenarios was not sensitive to θr and θs. The other parameters were accurately estimated from numerically synthesized data. The uniqueness of the estimated parameters did not change when a bias, representing experimental error, was added to the data set. In actual columns, using the temporal and spatial pressure head data from the ADIs and the water contents in the drier range predicted by the developed PTF resulted in a precise estimation of the van Genuchten parameters. Not including the PTF water contents resulted in non‐uniquely estimated van Genuchten parameters.

show abstract

Machine Learning-Based Prediction of Drainage in Layered Soils Using a Soil Drainability Index

Cited by 3 publications

References 32 publications

Machine learning applications in vadose zone hydrology: A review

Machine learning applications in vadose zone hydrology: A review

Investigation of estimation of hydraulic parameters in heterogeneous soil

Soil Hydraulic Properties Determined by Inverse Modeling of Drip Infiltrometer Experiments Extended with Pedotransfer Functions

Contact Info

Product

Resources

About