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.

Snapshots derived from the World Soil Information Service (WoSIS) are served freely to the international community. These static datasets provide quality-assessed and standardised soil profile data that can be used to support digital soil mapping and environmental applications at broad scale levels. Since the release of the preceding snapshot in 2019, refactored ETL (extract, transform and load) procedures for screening, ingesting and standardising disparate source data have been developed. In conjunction with this, the WoSIS data model was overhauled, making it compatible with the ISO 28258 and Observations and Measurements (O&M) domain models. Additional procedures for querying, serving and downloading the publicly available standardised data have been implemented using open software (e.g. GraphQL API). Following up on a short discussion of these methodological developments we discuss the structure and content of the “WoSIS 2023 snapshot”. A range of new soil datasets was shared with us, registered in the ISRIC World Data Centre for Soils (WDC-Soils) data repository and subsequently processed in accordance with the licences specified by the data providers. An important effort has been the processing of forest soil data collated in the framework of the EU-HoliSoils project. We paid special attention to the standardisation of soil property definitions, description of the soil analytical procedures and standardisation of the units of measurement. The 2023 snapshot considers soil chemical properties (total carbon, organic carbon, inorganic carbon (total carbonate equivalent), total nitrogen, phosphorus (extractable P, total P and P retention), soil pH, cation exchange capacity and electrical conductivity) and physical properties (soil texture (sand, silt and clay), bulk density, coarse fragments and water retention), grouped according to analytical procedures that are operationally comparable. Method options are defined for each analytical procedure (e.g. pH measured in water, KCl or CaCl₂ solution, molarity of the solution, and soil / solution ratio). For each profile we also provide the original soil classification (i.e. FAO, WRB and USDA system with their version) and pedological horizon designations as far as these have been specified in the source databases. Three measures for “fitness for intended use” are provided to facilitate informed data use: (a) positional uncertainty of the profile's site location, (b) possible uncertainty associated with the operationally defined analytical procedures and (c) date of sampling. The most recent (i.e. dynamic) dataset, called wosis_latest, is freely accessible via various web services. To permit consistent referencing and citation, we also provide a static snapshot (in this case, December 2023). This snapshot comprises quality-assessed and standardised data for 228 000 geo-referenced profiles. The data come from 174 countries and represent more than 900 000 soil layers (or horizons) and over 6 million records. The number of measurements for each soil property varies (greatly) between profiles and with depth, this generally depending on the objectives of the initial soil sampling programmes. In the coming years, we aim to gradually fill gaps in the geographic distribution of the profiles, as well as in the soil observations themselves, this subject to the sharing of a wider selection of “public” soil data by prospective data contributors; possible solutions for this are discussed. The WoSIS 2023 snapshot is archived and freely available at https://doi.org/10.17027/isric-wdcsoils-20231130 (Calisto et al., 2023).

SoilGrids produces maps of soil properties for the entire globe at medium spatial resolution (250 metres 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 quality-assessed global maps of soil properties, with cross-validation, hyper-parameters 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 regions 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.

This study compares the performance of five popular equal-area projections supported by Free and Open Source Software for Geo-spatial (FOSS4G)—Sinusoidal, Mollweide, Hammer, Eckert IV and Homolosine. A set of 21,872 discrete distortion vindicatrices were positioned on the ellipsoid surface, centred on the cells of a Snyder icosahedral equal-area grid. These indicatrices were projected on the plane and the resulting angular and distance distortions computed, all using FOSS4G. The Homolosine is the only projection that manages to minimise angular and distance distortions simultaneously. It yields the lowest distortions among this set of projections and clearly outclasses when only land masses are considered. These results also indicate the Sinusoidal and Hammer projections to be largely outdated, imposing too large distortions to be useful. In contrast, the Mollweide and Eckert IV projections present trade-offs between visual expression and accuracy that are worth considering. However, for the purposes of storing and analysing big spatial data with FOSS4G the superior performance of the Homolosine projection makes its choice difficult to avoid.