04 Mar 2024
 | 04 Mar 2024
Status: this preprint is open for discussion.

Prediction of Hysteretic Matric Potential Dynamics Using Artificial Intelligence: Application of Autoencoder Neural Networks

Nedal Aqel, Lea Reusser, Stephan Margreth, Andrea Carminati, and Peter Lehmann

Abstract. Information on soil water potential is essential to assess soil moisture state, to prevent soil compaction in weak soils, and to optimize crop management. In lack of direct measurements, the soil water potential values must be deduced from soil water content dynamics that can be monitored at plot scale or obtained at larger scale from remote sensing information. Because the relationship between water content and soil water potential in natural field soils is highly ambiguous, the prediction of soil water potential from water content data is a big challenge. The hysteretic relationship observed in nine soil profiles in the region of Solothurn (Switzerland) is not a simple function of texture or wetting and drainage cycles but depends on seasonal patterns that may be related to soil structural dynamics. Because the physical mechanisms governing seasonal hysteresis are unclear, we developed a deep neural network model that predicts water potential changes using rainfall, potential evapotranspiration, and water content time series as inputs. To adapt the model for multiple locations, we incorporated a Deep Autoencoder Neural Network as a classifier. The autoencoder compresses the water content time series into a site-specific feature that is highly representative of the underlying water content dynamics of each site and quantifies the similarity of dynamic patterns. By adding the Autoencoder's output as an additional input and training the neural network model with three stations located in three major classes founded by the autoencoder, we predict matric potential for other sites. This method has the potential to deduce the dynamics of matric potential from water content data (including satellite data) despite strong seasonal effects that cannot be captured by standard methods.

Nedal Aqel, Lea Reusser, Stephan Margreth, Andrea Carminati, and Peter Lehmann

Status: open (until 04 May 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Nedal Aqel, Lea Reusser, Stephan Margreth, Andrea Carminati, and Peter Lehmann
Nedal Aqel, Lea Reusser, Stephan Margreth, Andrea Carminati, and Peter Lehmann


Total article views: 130 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
105 17 8 130 5 8
  • HTML: 105
  • PDF: 17
  • XML: 8
  • Total: 130
  • BibTeX: 5
  • EndNote: 8
Views and downloads (calculated since 04 Mar 2024)
Cumulative views and downloads (calculated since 04 Mar 2024)

Viewed (geographical distribution)

Total article views: 134 (including HTML, PDF, and XML) Thereof 134 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 12 Apr 2024
Short summary
The soil water potential (SWP) determines various soil water processes. Because it cannot be measured directly by remote sensing techniques, it is often deduced from volumetric water content (VWC) information. However, under dynamic field conditions, the relationship between SWP and VWC is highly ambiguous due to different factors that cannot be modeled with the classical approach. Applying a deep neural network with an autoencoder enables the prediction of SWP.