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.

/pdf/soilgrids250m-global-gridded-soil-information-based-on-2xwvokp8f3.pdf

SoilGrids250m: Global gridded soil information based on machine learning

C4.5: Programs for Machine Learning (書評)

https://meetingorganizer.copernicus.org/EGU2014/EGU2014-2022.pdf

Digital soil mapping: A brief history and some lessons

. 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.

/pdf/soilgrids-2-0-producing-soil-information-for-the-globe-with-egkdd930cv.pdf

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

: A system is described which: (a) replaces existing sets of diverse terrain indices with a group of statistics for point-characteristics; (b) calculates all of these statistics in a single computer run from a single data set; and (c) utilizes available altitude matrix data The procedures are applicable to Altitude matrix data at any grid mesh From altitudes in each 3 x 3 submatrix, a quadratic surface is fitted and solved for its first and second horizontal and vertical derivatives at the central point This yields the slope gradient, slope aspect, profile convexity, and plan convexity at every point in the original matrix, except for the peripheral rows and columns These 'point' descriptors are presented as: (1) line-printer shaded maps; (2) histograms; (3) scatter plots of each pair; (4) matrix of pair-wise correlations, plus circular regressions on aspect, and several multiple regressions, and (5) summary (moment-based) statistics In general, the five basic descriptors have little relation to each other, except that gradient is usually a quadratic function of altitude A comparison is made with other approaches, such as spectral analysis and fractal modelling The long-distance persistence properties of terrain mean that considerable extra variance at long wavelengths is usually incorporated when the study area is extended Hence the auto correlation function varies with the length of series or size of area Variance of derivatives are also affected, but means, skews, and kurtoses are not

Statistical Characterization of Altitude Matrices by Computer. Report 6. An Integrated System of Terrain Analysis and Slope Mapping.

Soil mapping, classification, and pedologic modeling: History and future directions

Selecting appropriate machine learning methods for digital soil mapping

https://works.bepress.com/bradley_miller/9/download/

Impact of Multi-Scale Predictor Selection for Modeling Soil Properties

Soil Sci. Soc. Am. J. 76:1719–1727 doi:10.2136/sssaj2011.0303 Received 7 Sept. 2011. *Corresponding author (soils@msu.edu). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. Precision of Soil Particle Size Analysis using Laser Diffractometry Pedology

Precision of Soil Particle Size Analysis using Laser Diffractometry

We estimated the pre-settlement density and area of different classes of palustrine wetlands on the Des Moines Lobe based on soil characteristics. Six wetland classes, ranging from temporarily flooded to permanently flooded, were identified based on soil properties that persisted after artificial drainage. Prior to drainage, wetlands covered nearly half of the Des Moines Lobe and there were differences in both the types and relative abundance of wetlands among the four geologic subdivisions of the Lobe (Bemis, Altamont, and Algona till plains and Altamont Lake). In the flat Altamont Lake zone, the most common wetlands were equally split between temporarily flooded and saturated water regimes. Among the three till plain zones, saturated wetlands were the dominant wetland type. Differences in wetland distributions among the zones probably derive from differences in initial topography and post-glacial processes such as erosion-deposition processes and stream-network formation.

https://bioone.org/journals/wetlands/volume-29/issue-4/08-158.1/Spatial-Distribution-of-Historical-Wetland-Classes-on-the-Des-Moines/10.1672/08-158.1.pdf

Bradley A. Miller

Papers

Soil mapping, classification, and pedologic modeling: History and future directions

Selecting appropriate machine learning methods for digital soil mapping

Impact of Multi-Scale Predictor Selection for Modeling Soil Properties

Precision of Soil Particle Size Analysis using Laser Diffractometry

Spatial Distribution of Historical Wetland Classes on the Des Moines Lobe, Iowa