OpenLandMap-soildb: soil type probability - suborder: Xerolls

Sub-dataset: soil type probability - suborder: Xerolls

Description

Global annual maps of soil properties for 2000—2022 produced within the scope of the Land & Carbon Lab, integrating Digital surface/terrain model, vegetation/tillage indices, climatic/bioclimatic variables, and based on tree-based spatiotemporal Machine Learning. While the primary focus is on improving monitoring in global soil properties, the dataset provides wall-to-wall coverage across all terrestrial ecosystems and is organized into 300+ global mosaics in COG (Cloud Optimized GeoTIFF) format. Data are presented at 5-year intervals, across 3 standard depth intervals (0–30 cm, 30–60 cm, 60–100 cm), and cover 79 USDA soil taxonomy suborders. Original layers use the WGS84 Coordinate System (EPSG:4326) at a pixel resolution of 0.00025 degrees, and 0.00075 degrees with uncertainty (STAC and GEE). Layers archived on Zenodo are at 0.00075 degrees with uncertainty but include only the initial and final periods (2000–2005 & 2020–2022), including:

Soil Organic Carbon Content (g/kg)
As a key indicator of soil fertility, structure, and microbial activity, it represents the concentration of organic carbon in the fine earth fraction of the soil. Standard method of measurement is dry combustion using elemental analyzers (e.g., ISO 10694).

Soil Organic Carbon Density (kg/m³)
Represents the mass of organic carbon per unit volume of soil. It is derived as: SOC content × bulk density × (1 − coarse fragment volume fraction). This value is critical for estimating total carbon stocks and monitoring soil carbon changes over time.

Soil pH
Indicates the acidity or alkalinity of soil, affecting nutrient availability and microbial processes. Reported as pH measured in water solution (pH in H₂O).

Bulk Density (g/cm³)
Refers to the mass of dry fine earth (<2 mm) per unit volume, excluding coarse fragments. It reflects soil compaction and porosity, influencing water retention and root penetration. Commonly determined using the core method or calculated from pedotransfer functions.

Soil Texture Fraction
Defines the relative proportions of mineral particles by size. Texture influences water movement, nutrient holding capacity, and plant growth.

Clay content (%): Proportion of particles <0.002 mm in diameter.

Sand content (%): Proportion of particles between 0.05–2.0 mm (some definitions use 0.063 mm as lower threshold).

Silt content (%): Particles sized between 0.002–0.05 mm or up to 0.063 mm depending on classification system.

Textural fractions follow USDA or FAO particle size classifications.

Soil Type Probability
Probabilistic classification of soils based on USDA Soil Taxonomy at the subgroup level. Each pixel is assigned a probability distribution across potential soil types, based on legacy point data and environmental covariates.

30m layers can be accessed through STAC and Google Earth Engine GEE) through:

OpenLandMap STAC
https://stac.openlandmap.org

Google Earth Engine
https://code.earthengine.google.com/?asset=projects/global-pasture-watch/assets/gsm-30m

All modeling framework is publicly available at OpenLandMap GitHub - soildb

Data Detail

Time period: 2000-2022, in 5-year intervals (last period covers 2020–2022) for soil properties ; 2000-2022 static for soil type

Type of data: Spatiotemporal soil data base, with depth ranges and weighted percentage data for soil assessments and static soil type classification.

How the data was collected or derived: The data was derived using machine learning models.

Statistical methods used: Tree-based spatiotemporal machine learning

Depth reference: b30cm..60cm = below ground at 30-60cm interval

Limitations or exclusions in the data: no Antarctica; masking out permanent ice and deserts

Coordinate reference system: EPSG:4326

Bounding box (Xmin, Ymin, Xmax, Ymax): (-180, -56, 180, 76)

Spatial resolution: 0.00075 degree (~120m)

Image size: 360,000P, 132,000L

File format: Cloud Optimized Geotiff (COG) format

Dataset Contents

This dataset includes:

soil type probability - suborder: Xerolls

Related Identifiers

SOC density:
below ground 0cm-30cm 2000-2005 ,
below ground 0cm-30cm 2020-2022 ,
below ground 30cm-60cm 2000-2005 ,
below ground 30cm-60cm 2020-2022 ,
below ground 60cm-100cm 2000-2005 ,
below ground 60cm-100cm 2020-2022 ,

SOC content:
below ground 0cm-30cm 2000-2005 - part 1 ,
below ground 0cm-30cm 2000-2005 - part 2 ,
below ground 0cm-30cm 2020-2022 - part 1 ,
below ground 0cm-30cm 2020-2022 - part 2 ,
below ground 30cm-60cm 2000-2005 ,
below ground 30cm-60cm 2020-2022 ,
below ground 60cm-100cm 2000-2005 ,
below ground 60cm-100cm 2020-2022 ,

Bulk density:
below ground 0cm-30cm 2000-2005 ,
below ground 0cm-30cm 2020-2022 ,
below ground 30cm-60cm 2000-2005 ,
below ground 30cm-60cm 2020-2022 ,
below ground 60cm-100cm 2000-2005 ,
below ground 60cm-100cm 2020-2022 ,

Soil ph of water:
below ground 0cm-30cm 2000-2005 ,
below ground 0cm-30cm 2020-2022 ,
below ground 30cm-60cm 2000-2005 ,
below ground 30cm-60cm 2020-2022 ,
below ground 60cm-100cm 2000-2005 ,
below ground 60cm-100cm 2020-2022 ,

Soil textures fraction:
clay below ground 0cm-30cm 2000-2005 ,
clay below ground 0cm-30cm 2020-2022 ,
clay below ground 30cm-60cm 2000-2005 ,
clay below ground 30cm-60cm 2020-2022 ,
clay below ground 60cm-100cm 2000-2005 ,
clay below ground 60cm-100cm 2020-2022 ,
sand below ground 0cm-30cm 2000-2005 ,
sand below ground 0cm-30cm 2020-2022 ,
sand below ground 30cm-60cm 2000-2005 ,
sand below ground 30cm-60cm 2020-2022 ,
sand below ground 60cm-100cm 2000-2005 ,
sand below ground 60cm-100cm 2020-2022 ,
silt below ground 0cm-30cm 2000-2005 ,
silt below ground 0cm-30cm 2020-2022 ,
silt below ground 30cm-60cm 2000-2005 ,
silt below ground 30cm-60cm 2020-2022 ,
silt below ground 60cm-100cm 2000-2005 ,
silt below ground 60cm-100cm 2020-2022 ,

Soil type (Suborder):
Uderts ,
Calcids ,
Xerands ,
Orthents ,
Cryands ,
Ustalfs ,
Cryalfs ,
Aquepts ,
Udalfs ,
Cryolls ,
Durids ,
Usterts ,
Boralfs ,
Orthids ,
Udands ,
Torrerts ,
Histels ,
Rendolls ,
Aqualfs ,
Udepts ,
Xeralfs ,
Gelepts ,
Xerults ,
Fibrists ,
Ustepts ,
Xererts ,
Ustults ,
Aquands ,
Perox ,
Xerolls ,
Tropepts ,
Turbels ,
Udults ,
Aquents ,
Aquerts ,
Ustox ,
Aquods ,
Aquolls ,
Xerepts ,
Udox ,
Cryods ,
Ustolls ,
Aquults ,
Psamments ,
Arents ,
Fluvents ,
Humults ,
Vitrands ,
Udolls ,
Borolls ,
Orthels ,
Hemists ,
Wassents ,
Albolls ,
Salids ,
Cryepts ,
Saprists ,
Folists ,
Gypsids ,
Ochrepts ,
Cambids ,
Argids ,
Orthods ,

Data Details

Time period: 2000-2022

Type of data: soil type probability - suborder: Xerolls

How the data was collected or derived: Machine learning models.

Statistical Methods used: Random Forest.

Limitations or exclusions in the data: The dataset does not include Antarctica.

Coordinate reference system: EPSG:4326

Bounding box (Xmin, Ymin, Xmax, Ymax): (-180, -56, 180, 76)

Spatial resolution: 120m

Image size: 360,000P x 132,000L

File format: Cloud Optimized Geotiff (COG) format.

Layer information:

File Name
Unit
Scale
Data Type
No Data
Description

oc_iso.10694.1995.mg.cm3
kg/m³
10
UInt16
32767
Organic carbon density derived by multiply fine earth bulk density and organic carbon content

oc_iso.10694.1995.wpml
g/kg
10
UInt16
32767
Organic carbon content based on dry combustion weight percent

ph.h2o_iso.10390.2021.index
-
10
Byte
255
The pH, 1:1 soil-water suspension is the pH of a sample measured in distilled water at a 1:1 soil:solution ratio

bd.core_iso.11272.2017.g.cm3
g/cm³
100
UInt16
32767
Bulk density, <2mm fraction, dry is the weight per unit volume of the <2 mm fraction, with volume measured in laboratory

sand.tot_iso.11277.2020.wpct
%
1
Byte
255
Total laboratory-estimated sand 0.063 to 2.0 mm particle diameter

silt.tot_iso.11277.2020.wpct
%
1
Byte
255
Total laboratory-estimated silt 0.002 to 0.063 mm particle size

clay.tot_iso.11277.2020.wpct
%
1
Byte
255
Total clay is the soil separate with <0.002 mm particle diameter

soil.types_ensemble
%
1
Byte
255
Probability of soil type occurrence

Support

If you discover a bug, artifact, or inconsistency, or if you have a question please raise a GitHub issue here

Naming convention

To ensure consistency and ease of use across and within the projects, we follow the standard Ai4SoilHealth and Open-Earth-Monitor file-naming convention. The convention works with 10 fields that describe important properties of the data. In this way users can search files, prepare data analysis etc, without needing to open files.

For example, for oc_iso.10694.1995.wpml_m_30m_b30cm..60cm_20000101_20051231_g_epsg.4326_v20250204.tif, the fields are:

generic variable name: oc = organic carbon

variable procedure combination: iso.10694.1995.wpml = organic carbon content based on dry combustion weight per mille

Position in the probability distribution/variable type: m = mean

Spatial support: 30m

Depth reference: b30cm..60cm = below ground at 30-60cm interval

Time reference begin time: 20000101 = 2000-01-01

Time reference end time: 20051231 = 2005-12-31

Bounding box: g = global

EPSG code: EPSG:4326

Version code: v20250204 = version from 2025-02-04

Acknowledgement
This project is funded by European Commission through Open-Earth-Monitor project and by World Resources Institute through Land & Carbon Lab