Submitted as: methods for assessment of models 16 Jul 2021

Submitted as: methods for assessment of models | 16 Jul 2021

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

A new methodological framework by geophysical sensors combinations associated with machine learning algorithms to understand soil attributes

Danilo César de Mello1, Gustavo Vieira Veloso1, Marcos Guedes de Lana1, Fellipe Alcantara de Oliveira Mello2, Raul Roberto Poppiel2, Diego Ribeiro Oquendo Cabrero3, Luis Augusto Di Loreto Di Raimo4, Carlos Ernesto Gonçalves Reynaud Schaefer1, Elpídio Inácio Fernandes Filho1, Emilson Pereira Leite4, and José Alexandre Melo Demattê2 Danilo César de Mello et al.
  • 1Department of Soil Science, Federal University of Viçosa
  • 2Department of Soil Science, "Luiz de Queiroz" College of Agriculture, University of São Paulo, Av. Pádua Dias, 11, CP 9, Piracicaba, SP 13418-900, Brazil
  • 3Geography Department of Federal University of Mato Grosso do Sul, Av. Ranulpho Marques Leal, nº 3484 - Distrito Industrial CEP 79610-100 Três Lagoas/MS
  • 4Department of Geology and Natural Resources, Institute of Geosciences, University of Campinas, Rua Carlos Gomes, 250, Cidade Universitária, CEP 13083-855, Campinas/SP

Abstract. Geophysical sensors combined with machine learning algorithms have been used to understand the pedosphere system, landscape processes and to model soil attributes. In this research, we used parent material, terrain attributes and data from geophysical sensors in different combinations, to test and compare different and novel machine learning algorithms to model soil attributes. Also, we analyzed the importance of pedoenvironmental variables in predictive models. For that, we collected soil physico-chemical and geophysical data (gamma-ray emission from uranium, thorium and potassium, magnetic susceptibility and apparent electric conductivity) by three sensors, gamma-ray spectrometer – RS 230, susceptibilimeter KT10 – Terraplus and Conductivimeter – EM38 Geonics) at 75 points and, we performed soil analysis afterwards. The results showed varying models with the best performance (R2 > 0.2) for clay, sand, Fe2O3, TiO2, SiO2 and Cation Exchange Capacity prediction. Modeling with selection of covariates at three phases (variance close to zero, removal by correction and removal by importance), demonstrated to be adequate to increase the parsimony. The prediction of soil attributes by machine learning algorithms demonstrated adequate values for field collected data, without any sample preparation, for most of the tested predictors (R2 ranging from 0.20 to 0.50). Also, the use of four regression algorithms proved important, since at least one of the predictors used one of the tested algorithms. The performances of the best algorithms for each predictor were higher than the use of a mean value for the entire area comparing the values of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The best combination of sensors that reached the best model performance to predict soil attributes were gamma-ray spectrometer and susceptibilimeter. The most important variables were parent material, digital elevation model, standardized height and magnetic susceptibility for most predictions. We concluded that soil attributes can be efficiently modelled by geophysical data using machine learning techniques and geophysical sensors combinations. The technique can bring light for future soil mapping with gain of time and environment friendly.

Danilo César de Mello et al.

Status: open (until 15 Nov 2021)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on gmd-2021-153', Juan Antonio Añel, 10 Aug 2021 reply
    • AC1: 'Reply on CEC1', Danilo Mello, 16 Aug 2021 reply
  • RC1: 'Comment on gmd-2021-153', Anonymous Referee #1, 08 Oct 2021 reply

Danilo César de Mello et al.

Danilo César de Mello et al.


Total article views: 709 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
463 237 9 709 18 4 3
  • HTML: 463
  • PDF: 237
  • XML: 9
  • Total: 709
  • Supplement: 18
  • BibTeX: 4
  • EndNote: 3
Views and downloads (calculated since 16 Jul 2021)
Cumulative views and downloads (calculated since 16 Jul 2021)

Viewed (geographical distribution)

Total article views: 633 (including HTML, PDF, and XML) Thereof 633 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 25 Oct 2021
Short summary
We used soil parent material, terrain attributes and geophysics data from soil surface to test and compare different and unprecedented geophysical sensors combination, as well as different machine learning algorithms to model and predict several soil attributes. Also, we analyzing the importance of pedoenvironmental variables. The soil attributes were modelling throughout different machine learning algorithms and related to different geophysical sensors combinations.