Conversion between Soil Texture Classification Systems using the Random Forest Algorithm

Čistý, Milan; Celar, Lubomir; Minarič, Peter

doi:10.4137/aswr.s31924

Cited by 6 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The assignment of relative weights to textural classes was based on expert judgment. The map was reclassified on the basis of the criteria proposed by De Lannoy et al (2014) and Cisty et al (2015), who considered hydric models with high precision algorithms and techniques.…”

Section: Soil Texture (St)mentioning

confidence: 99%

Soil erosion risk zoning in the Ecuadorian coastal region using geo-technological tools

2019

View full text Add to dashboard Cite

The activation of erosive processes on the slopes of the drainage catchments in the Ecuadorian coastal region represents a serious environmental problem that results in a loss of soil resources with the consequent reduction in agricultural productivity, an increasing lack of areas with vegetation, and decreased water quality. These issues increase, among other things, the probability of flooding. The purpose of this research is to zone soil erosion risk in the Manabí province, which is located on the Ecuadorian coast. The methodology is supported by the employment of Geographic Information Technologies (GIT) in order to access and process information of interest, such as: precipitation data with which to calculate the R Factor of the Universal Soil Loss Equation (USLE); topography, which was employed to create a slope map from the Digital Elevation Model (DEM) ASTER; vegetation cover, by applying the Standard Difference Vegetation Index (NDVI); land use, through the interpretation of orthophotos and a field survey; and infiltration capacity, obtained by considering the texture of the soil. The factors were combined by means a weighted relationship (map algebra) and exploiting the potential of software for the design of Geographic Information Systems (GIS). The results indicate the presence of soils with little or without vegetal protection, in areas with steep slopes, moderate rainfall and extensive farming activities. These characteristics determine conditions of high soil erodability, and propitiate the triggering of erosive processes. This scenario shows the need for autonomous governments to implement policies, strategies and actions so as to promote the strengthening of institutional and community capacities that will promote environmental sustainability.

show abstract

Section: Soil Texture (St)mentioning

confidence: 99%

Soil erosion risk zoning in the Ecuadorian coastal region using geo-technological tools

2019

View full text Add to dashboard Cite

show abstract

“…Tree type ensemble algorithms were found to be successful in harmonizing different soil texture classification systems (Cisty et al, 2015) and prediction of soil bulk density (Chen et al, 2018;Dharumarajan et al, 2017;Ramcharan et al, 2017;Sequeira et al, 2014;Souza et al, 2016) but have not been intensively applied yet to derive input parameters for hydrological modelling (Koestel and Jorda, 2014;Tóth et al, 2014). Hengl et al (2018a) tested several machine learning algorithms (i.e.…”

Section: Introductionmentioning

confidence: 99%

Mapping soil hydraulic properties using random-forest-based pedotransfer functions and geostatistics

Szabó

Szatmári

Takács

et al. 2019

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Spatial 3-D information on soil hydraulic properties for areas larger than plot scale is usually derived using indirect methods such as pedotransfer functions (PTFs) due to the lack of measured information on them. PTFs describe the relationship between the desired soil hydraulic parameter and easily available soil properties based on a soil hydraulic reference dataset. Soil hydraulic properties of a catchment or region can be calculated by applying PTFs on available soil maps. Our aim was to analyse the performance of (i) indirect (using PTFs) and (ii) direct (geostatistical) mapping methods to derive 3-D soil hydraulic properties. The study was performed on the Balaton catchment area in Hungary, where density of measured soil hydraulic data fulfils the requirements of geostatistical methods. Maps of saturated water content (0 cm matric potential), field capacity (−330 cm matric potential) and wilting point (−15 000 cm matric potential) for 0–30, 30–60 and 60–90 cm soil depth were prepared. PTFs were derived using the random forest method on the whole Hungarian soil hydraulic dataset, which includes soil chemical, physical, taxonomical and hydraulic properties of some 12 000 samples complemented with information on topography, climate, parent material, vegetation and land use. As a direct and thus geostatistical method, random forest combined with kriging (RFK) was applied to 359 soil profiles located in the Balaton catchment area. There were no significant differences between the direct and indirect methods in six out of nine maps having root-mean-square-error values between 0.052 and 0.074 cm3 cm−3, which is in accordance with the internationally accepted performance of hydraulic PTFs. The PTF-based mapping method performed significantly better than the RFK for the saturated water content at 30–60 and 60–90 cm soil depth; in the case of wilting point the RFK outperformed the PTFs at 60–90 cm depth. Differences between the PTF-based and RFK mapped values are less than 0.025 cm3 cm−3 for 65 %–86 % of the catchment. In RFK, the uncertainty of input environmental covariate layers is less influential on the mapped values, which is preferable. In the PTF-based method the uncertainty of mapping soil hydraulic properties is less computationally intensive. Detailed comparisons of maps derived from the PTF-based method and the RFK are presented in this paper.

show abstract

“…Topographical parameters were calculated with SAGA GIS tools (Conrad et al, 2015) based on the digital elevation model. For the mapping of soil hydraulic properties all covariates were harmonized, projected to the Hungarian Uniform National Projection system, rasterized if necessary and resampled to 100 m resolution.…”

Section: Study Sitementioning

confidence: 99%

author comments

Szatmári¹

2019

Preprint

View full text Add to dashboard Cite

Spatial 3-D information on soil hydraulic properties for areas larger than plot scale is usually derived using indirect methods such as pedotransfer functions (PTFs) due to the lack of measured information on them. PTFs describe the relationship between the desired soil hydraulic parameter and easily available soil properties based on a soil hydraulic reference dataset. Soil hydraulic properties of a catchment or region can be calculated by applying PTFs on available soil maps. Our aim was to analyse the performance of (i) indirect (using PTFs) and (ii) direct (geostatistical) mapping methods to derive 3-D soil hydraulic properties. The study was performed on the Balaton catchment area in Hungary, where density of measured soil hydraulic data fulfils the requirements of geostatistical methods. Maps of saturated water content (0 cm matric potential), field capacity ( − 330 cm matric potential) and wilting point (−15 000 cm matric potential) for 0-30, 30-60 and 60-90 cm soil depth were prepared. PTFs were derived using the random forest method on the whole Hungarian soil hydraulic dataset, which includes soil chemical, physical, taxonomical and hydraulic properties of some 12 000 samples complemented with information on topography, climate, parent material, vegetation and land use. As a direct and thus geostatistical method, random forest combined with kriging (RFK) was applied to 359 soil profiles located in the Balaton catchment area. There were no significant differences between the direct and indirect methods in six out of nine maps having root-mean-square-error values between 0.052 and 0.074 cm 3 cm −3 , which is in accordance with the internationally accepted performance of hydraulic PTFs. The PTF-based mapping method performed significantly better than the RFK for the saturated water content at 30-60 and 60-90 cm soil depth; in the case of wilting point the RFK outperformed the PTFs at 60-90 cm depth. Differences between the PTF-based and RFK mapped values are less than 0.025 cm 3 cm −3 for 65 %-86 % of the catchment. In RFK, the uncertainty of input environmental covariate layers is less influential on the mapped values, which is preferable. In the PTF-based method the uncertainty of mapping soil hydraulic properties is less computationally intensive. Detailed comparisons of maps derived from the PTF-based method and the RFK are presented in this paper.

show abstract

Conversion between Soil Texture Classification Systems using the Random Forest Algorithm

Cited by 6 publications

References 24 publications

Soil erosion risk zoning in the Ecuadorian coastal region using geo-technological tools

Soil erosion risk zoning in the Ecuadorian coastal region using geo-technological tools

Mapping soil hydraulic properties using random-forest-based pedotransfer functions and geostatistics

author comments

Contact Info

Product

Resources

About