Geoderma · 2025 · 13 citations · 42 references
AgricultureEarth SciencesSoil SciencesSoil ClassificationSoil PhysicsSoil ScienceSoil FormationSoil MappingSoil TaxonomySoil-landform RelationshipsSoil-landscape RelationshipsMulti-scale Soil MappingParent Material StratigraphySoil Mapping Unit DelineationSoil PropertiesSoil Property MappingSoil UnitsSpatial Soil ClassificationUnsupervised Classification
This study presents the development of a global pedogenon map, which classifies soil units based on similarities in their formation processes while excluding anthropogenic effects, at a high spatial resolution of 90 m. Designed for use by scientists, land managers, and policymakers, this map classifies distinct soil units —called pedogenons—based on key soil-forming factors, including parent material, climate, soil organisms, relief, and inherent soil properties. This study employs a combination of principal component analysis (PCA), non-metric-multidimensional-scaling (NMDS), and an unsupervised classification technique (K-means clustering) to capture the full range of soil formation processes across diverse landscapes of the world yielding an unbiased, globally applicable classification. The global pedogenon map enables soil capacity and condition monitoring on a global scale. This foundation is crucial for future identification of “least disturbed soils” to serve as a reference for any soil around the globe. By integrating this dataset with broader environmental data, the global pedogenon map supports global efforts to enhance soil security and monitor environmental changes.
42
Google Earth Engine: Planetary-scale geospatial analysis for everyone
Noel Gorelick, M. Hancher, Mike Dixon et al. · Remote Sensing of Environment · 2017 · 14K citations · Full text
The Shuttle Radar Topography Mission
T. G. Farr, P. A. Rosen, Robert E. Crippen et al. · Reviews of Geophysics · 2007 · 8.6K citations · Full text
Scikit-learn: Machine Learning in Python
PedregosaFabian, VaroquauxGaël, GramfortAlexandre et al. · Journal of Machine Learning Research · 2011 · 8.2K citations