SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty

SoilGrids 2.0: получение глобальной информации о почвах с количественной оценкой пространственной неопределённости
G.B.M. Heuvelink, Laura Poggio, Luís Moreira de Sousa, N.H. Batjes, Bas Kempen, Eloi Ribeiro, David G. Rossiter
2021-06-14

SoilGrids 2.0digital soil mappingglobal soil property mapsmachine learningspatial uncertainty quantification
SoilGrids produces maps of soil properties for the entire globe at medium spatial resolution (250 m cell size) using state-of-the-art machine learning methods to generate the necessary models. It takes as inputs soil observations from about 240 000 locations worldwide and over 400 global environmental covariates describing vegetation, terrain morphology, climate, geology and hydrology. The aim of this work was the production of global maps of soil properties, with cross-validation, hyper-parameter selection and quantification of spatially explicit uncertainty, as implemented in the SoilGrids version 2.0 product incorporating state-of-the-art practices and adapting them for global digital soil mapping with legacy data. The paper presents the evaluation of the global predictions produced for soil organic carbon content, total nitrogen, coarse fragments, pH (water), cation exchange capacity, bulk density and texture fractions at six standard depths (up to 200 cm). The quantitative evaluation showed metrics in line with previous global, continental and large-region studies. The qualitative evaluation showed that coarse-scale patterns are well reproduced. The spatial uncertainty at global scale highlighted the need for more soil observations, especially in high-latitude regions.
1
Qualitative assessments showed that the predictions reproduce broad-scale spatial soil patterns, while uncertainty analysis identified high-latitude regions as priorities for additional observations.
2
Quantitative evaluations produced metrics consistent with previous global, continental, and large-region soil mapping studies.
3
SoilGrids 2.0 incorporates cross-validation, hyperparameter selection, and spatially explicit uncertainty quantification for global digital soil mapping.
4
SoilGrids 2.0 produces global 250 m maps of seven soil properties across six standard depths, extending to 200 cm.
5
The system integrates approximately 240,000 worldwide soil observations with more than 400 environmental covariates using state-of-the-art machine learning.

Global maps of soil properties produced by SoilGrids 2.0 (250 m resolution)

The accuracy, coarse-scale pattern reproduction, and spatially explicit uncertainty of global soil-property predictions

Publication Details
Publication Date
2021-06-14
Journal
Publisher
ISSN
Cited by
2790
Access Type
Author Information
Authors
G.B.M. Heuvelink
Laura Poggio
Luís Moreira de Sousa
N.H. Batjes
Bas Kempen
Eloi Ribeiro
David G. Rossiter
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%