Preprints
https://doi.org/10.5194/nhess-2021-391
https://doi.org/10.5194/nhess-2021-391

  23 Dec 2021

23 Dec 2021

Review status: this preprint is currently under review for the journal NHESS.

Estimation of soil water holding capacity with Random Forests for drought monitoring

Yves Tramblay1 and Pere Quintana Seguí2 Yves Tramblay and Pere Quintana Seguí
  • 1HydroSciences Montpellier (University Montpellier, CNRS, IRD), France
  • 2Observatori de l’Ebre (OE), Ramon Llull University – CSIC, 43520 Roquetes, Spain

Abstract. Soil moisture is a key variable for drought monitoring but soil moisture measurements networks are very scarce. Land-surface models can provide a valuable alternative to simulate soil moisture dynamics, but only a few countries have such modelling schemes implemented for monitoring soil moisture at high spatial resolution. In this study, a soil moisture accounting model (SMA) was regionalized over the Iberian Peninsula, taking as a reference the soil moisture simulated by a high-resolution land surface model. To estimate soil water holding capacity, the parameter required to run the SMA model, two approaches were compared: the direct estimation from European soil maps using pedotransfer functions, or an indirect estimation by a Machine Learning approach, Random Forests, using as predictors altitude, temperature, precipitation, evapotranspiration and land use. Results showed that the Random Forest model estimates are more robust, especially for estimating low soil moisture levels. Consequently, the proposed approach can provide an efficient way to simulate daily soil moisture and therefore monitor soil moisture droughts, in contexts where high-resolution soil maps are not available, as it relies on a set of covariates that can be reliably estimated from global databases.

Yves Tramblay and Pere Quintana Seguí

Status: open (until 16 Feb 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Yves Tramblay and Pere Quintana Seguí

Yves Tramblay and Pere Quintana Seguí

Viewed

Total article views: 247 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
202 36 9 247 7 9
  • HTML: 202
  • PDF: 36
  • XML: 9
  • Total: 247
  • BibTeX: 7
  • EndNote: 9
Views and downloads (calculated since 23 Dec 2021)
Cumulative views and downloads (calculated since 23 Dec 2021)

Viewed (geographical distribution)

Total article views: 239 (including HTML, PDF, and XML) Thereof 239 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 18 Jan 2022
Download
Short summary
Monitoring soil moisture is important during droughts, but very few measurements are available. Consequently, land surface models are essential tools for reproducing soil moisture dynamics. In this study, a hybrid approach allowed to regionalize soil water content using a Machine Learning method. This approach proved to be efficient, compared to the use of soil properties maps, to run a simple soil moisture accounting model and therefore it can be applied in various regions.
Altmetrics