This paper describes the technical development and accuracy assessment of the most recent and improved version of the SoilGrids system at 250m resolution (June 2016 update). SoilGrids provides global predictions for standard numeric soil properties (organic carbon, bulk density, Cation Exchange Capacity (CEC), pH, soil texture fractions and coarse fragments) at seven standard depths (0, 5, 15, 30, 60, 100 and 200 cm), in addition to predictions of depth to bedrock and distribution of soil classes based on the World Reference Base (WRB) and USDA classification systems (ca. 280 raster layers in total). Predictions were based on ca. 150,000 soil profiles used for training and a stack of 158 remote sensing-based soil covariates (primarily derived from MODIS land products, SRTM DEM derivatives, climatic images and global landform and lithology maps), which were used to fit an ensemble of machine learning methods—random forest and gradient boosting and/or multinomial logistic regression—as implemented in the R packages ranger, xgboost, nnet and caret. The results of 10–fold cross-validation show that the ensemble models explain between 56% (coarse fragments) and 83% (pH) of variation with an overall average of 61%. Improvements in the relative accuracy considering the amount of variation explained, in comparison to the previous version of SoilGrids at 1 km spatial resolution, range from 60 to 230%. Improvements can be attributed to: (1) the use of machine learning instead of linear regression, (2) to considerable investments in preparing finer resolution covariate layers and (3) to insertion of additional soil profiles. Further development of SoilGrids could include refinement of methods to incorporate input uncertainties and derivation of posterior probability distributions (per pixel), and further automation of spatial modeling so that soil maps can be generated for potentially hundreds of soil variables. Another area of future research is the development of methods for multiscale merging of SoilGrids predictions with local and/or national gridded soil products (e.g. up to 50 m spatial resolution) so that increasingly more accurate, complete and consistent global soil information can be produced. SoilGrids are available under the Open Data Base License.

We developed a comprehensive, gridded Global Soil Dataset for use in Earth System Models (GSDE) and other applications. The GSDE provides soil information, such as soil particle‐size distribution, organic carbon, and nutrients, and quality control information in terms of confidence level at 30″ × 30″ horizontal resolution and for eight vertical layers to a depth of 2.3 m. The GSDE is based on the Soil Map of the World and various regional and national soil databases, including soil attribute data and soil maps. We used a standardized data structure and data processing procedures to harmonize the data collected from various sources. We then used a soil type linkage method (i.e., taxotransfer rules) and a polygon linkage method to derive the spatial distribution of the soil properties. To aggregate the attributes of different compositions of a mapping unit, we used three mapping approaches: the area‐weighting method, the dominant soil type method, and the dominant binned soil attribute method. The data set can also be aggregated to a lower resolution. In this paper, we only show the vertical and horizontal variations of sand, silt and clay contents, bulk density, and soil organic carbon as examples of the GSDE. The GSDE estimates of global soil organic carbon stock to the depths of 2.3, 1, and 0.3 m are 1922.7, 1455.4, and 720.1 Gt, respectively. This newly developed data set provides more accurate soil information and represents a step forward to advance earth system modeling.

A comprehensive 30×30 arc‐second resolution gridded soil characteristics data set of China has been developed for use in the land surface modeling. It includes physical and chemical attributes of soils derived from 8979 soil profiles and the Soil Map of China (1:1,000,000). We used the polygon linkage method to derive the spatial distribution of soil properties. The profile attribute database and soil map are linked under the framework of the Genetic Soil Classification of China which avoids uncertainty in taxon referencing. Quality control information (i.e., sample size, soil classification level, linkage level, search radius and texture) is included to provide “confidence” information for the derived soil parameters. The data set includes 28 attributes for 8 vertical layers at the spatial resolution of 30×30 arc‐seconds. Based on this data set, the estimated storage of soil organic carbon in the upper 1 m of soil is 72.5 Pg, total N is 6.6 Pg, total P is 4.5 Pg, total K is 169.9 Pg, alkali‐hydrolysable N is 0.55 Pg, available P is 0.03 Pg, and available K is 0.61 Pg. These estimates are reasonable compared with previous studies. The distributions of soil properties are consistent with common knowledge of Chinese soil scientists and the spatial variations over large areas are well represented. The data set can be incorporated into land models to better represent the role of soils in hydrological and biogeochemical cycles in China.

Depth to bedrock serves as the lower boundary of land surface models, which controls hydrologic and biogeochemical processes. This paper presents a framework for global estimation of depth to bedrock (DTB). Observations were extracted from a global compilation of soil profile data (ca. 1,30,000 locations) and borehole data (ca. 1.6 million locations). Additional pseudo‐observations generated by expert knowledge were added to fill in large sampling gaps. The model training points were then overlaid on a stack of 155 covariates including DEM‐based hydrological and morphological derivatives, lithologic units, MODIS surface reflectance bands and vegetation indices derived from the MODIS land products. Global spatial prediction models were developed using random forest and Gradient Boosting Tree algorithms. The final predictions were generated at the spatial resolution of 250 m as an ensemble prediction of the two independently fitted models. The 10–fold cross‐validation shows that the models explain 59% for absolute DTB and 34% for censored DTB (depths deep than 200 cm are predicted as 200 cm). The model for occurrence of R horizon (bedrock) within 200 cm does a good job. Visual comparisons of predictions in the study areas where more detailed maps of depth to bedrock exist show that there is a general match with spatial patterns from similar local studies. Limitation of the data set and extrapolation in data spare areas should not be ignored in applications. To improve accuracy of spatial prediction, more borehole drilling logs will need to be added to supplement the existing training points in under‐represented areas.

The objective of this study is to develop a dataset of the soil hydraulic parameters associated with two empirical soil functions (i.e., a water retention curve and hydraulic conductivity) using multiple pedotransfer functions (PTFs). The dataset is designed specifically for regional land surface modeling for China. The authors selected 5 PTFs to derive the parameters in the Clapp and Hornberger functions and the van Genuchten and Mualem functions and 10 PTFs for soil water contents at capillary pressures of 33 and 1500 kPa. The inputs into the PTFs include soil particle size distribution, bulk density, and soil organic matter. The dataset provides 12 estimated parameters and their associated statistical values. The dataset is available at a 30 × 30 arc second geographical spatial resolution and with seven vertical layers to the depth of 1.38 m. The dataset has several distinct advantages even though the accuracy is unknown for lack of in situ and regional measurements. First, this dataset utilizes the best available soil characteristics dataset for China. The Chinese soil characteristics dataset was derived by using the 1:1 000 000 Soil Map of China and 8595 representative soil profiles. Second, this dataset represents the first attempt to estimate soil hydraulic parameters using PTFs directly for continental China at a high spatial resolution. Therefore, this dataset should capture spatial heterogeneity better than existing estimates based on lookup tables according to soil texture classes. Third, the authors derived soil hydraulic parameters using multiple PTFs to allow flexibility for data users to use the soil hydraulic parameters most preferable to or suitable for their applications.