Preprints
https://doi.org/10.5194/hess-2020-271
https://doi.org/10.5194/hess-2020-271

  18 Aug 2020

18 Aug 2020

Review status: a revised version of this preprint was accepted for the journal HESS and is expected to appear here in due course.

Advances in Soil Moisture Retrieval from Multispectral Remote Sensing Using Unmanned Aircraft Systems and Machine Learning Techniques

Samuel N. Araya1, Anna Fryjoff-Hung2, Andreas Anderson2, Joshua H. Viers2,3, and Teamrat A. Ghezzehei2,4 Samuel N. Araya et al.
  • 1Earth System Science, Stanford University, Stanford, CA, USA
  • 2Center for Information Technology in the Interest of Society and the Banatao Institute, University of California, Merced, CA, USA
  • 3Department of Civil and Environmental Engineering, University of California, Merced, CA, USA
  • 4Life and Environmental Science, University of California, Merced, CA, USA

Abstract. We developed machine learning models to retrieve surface soil moisture (0–4 cm) from high resolution multispectral imagery using terrain attributes and local climate covariates. Using a small unmanned aircraft system (UAS) equipped with a multispectral sensor we captured high resolution imagery in part to create a high-resolution digital elevation model (DEM) as well as quantify relative vegetation photosynthetic status. We tested four different machine learning algorithms. The boosted regression tree algorithm gave the best prediction with mean absolute error of 3.8 % volumetric water content. The most important variables for the prediction of soil moisture were precipitation, reflectance in the red wavelengths, potential evapotranspiration, and topographic position indices (TPI). Our results demonstrate that the dynamics of soil water status across heterogeneous terrain may be adequately described and predicted by UAS remote sensing data and machine learning. Our modeling approach and the variable importance and relationships we have assessed in this study should be useful for management and environmental modeling tasks where spatially explicit soil moisture information is important.

Samuel N. Araya et al.

 
Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement
 
Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Samuel N. Araya et al.

Samuel N. Araya et al.

Viewed

Total article views: 426 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
293 127 6 426 36 10 10
  • HTML: 293
  • PDF: 127
  • XML: 6
  • Total: 426
  • Supplement: 36
  • BibTeX: 10
  • EndNote: 10
Views and downloads (calculated since 18 Aug 2020)
Cumulative views and downloads (calculated since 18 Aug 2020)

Viewed (geographical distribution)

Total article views: 379 (including HTML, PDF, and XML) Thereof 375 with geography defined and 4 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 19 Apr 2021
Download
Short summary
We developed machine learning models to estimate soil moisture levels from images taken by unmanned aircraft systems (drone). Our best model was able to estimate soil moisture with a mean absolute error of only 3.8 % volumetric soil water content. In addition to the images, we found rainfall, potential evapotranspiration, and topographic variables to be important in estimating soil moisture across the grassland.