iSDAsoil Total Carbon

  • The dataset provides predicted mean and standard deviation for total carbon at soil depths of 0-20 cm and 20-50 cm across Africa from 2001 to 2017.

  • Data was generated at a 30m pixel size using machine learning and over 100,000 soil samples by iSDA.

  • Pixel values require a specific back-transformation using the formula exp(x/10)-1.

  • Model accuracy can be low in dense jungle areas, potentially resulting in visible banding artifacts.

  • The dataset is available under the CC-BY-4.0 license and can be explored using Google Earth Engine.
ISDASOIL/Africa/v1/carbon_total
Dataset Availability
2001-01-01T00:00:00Z–2017-01-01T00:00:00Z
Dataset Provider
Earth Engine Snippet
ee.Image("ISDASOIL/Africa/v1/carbon_total")
Tags
africa aluminium isda soil

Description

Total carbon at soil depths of 0-20 cm and 20-50 cm, predicted mean and standard deviation.

Pixel values must be back-transformed with exp(x/10)-1.

In areas of dense jungle (generally over central Africa), model accuracy is low and therefore artifacts such as banding (striping) might be seen.

Soil property predictions were made by Innovative Solutions for Decision Agriculture Ltd. (iSDA) at 30 m pixel size using machine learning coupled with remote sensing data and a training set of over 100,000 analyzed soil samples.

Further information can be found in the FAQ and technical information documentation. To submit an issue or request support, please visit the iSDAsoil site.

Bands

Pixel Size
30 meters

Bands

Name Units Min Max Pixel Size Description
mean_0_20 g/kg 0 58 meters

Carbon, total, predicted mean at 0-20 cm depth
mean_20_50 g/kg 0 55 meters

Carbon, total, predicted mean at 20-50 cm depth
stdev_0_20 g/kg 0 151 meters

Carbon, total, standard deviation at 0-20 cm depth
stdev_20_50 g/kg 0 150 meters

Carbon, total, standard deviation at 20-50 cm depth

Terms of Use

Terms of Use

CC-BY-4.0

Citations

Citations:

  • Hengl, T., Miller, M.A.E., Križan, J., et al. African soil properties and nutrients mapped at 30 m spatial resolution using two-scale ensemble machine learning. Sci Rep 11, 6130 (2021). doi:10.1038/s41598-021-85639-y

Explore with Earth Engine

Code Editor (JavaScript)

var mean_0_20 =
'<RasterSymbolizer>' +
 '<ColorMap type="ramp">' +
  '<ColorMapEntry color="#000004" label="0-2" opacity="1" quantity="11"/>' +
  '<ColorMapEntry color="#0C0927" label="2-5.7" opacity="1" quantity="19"/>' +
  '<ColorMapEntry color="#231151" label="5.7-10" opacity="1" quantity="24"/>' +
  '<ColorMapEntry color="#410F75" label="10-12.5" opacity="1" quantity="26"/>' +
  '<ColorMapEntry color="#5F187F" label="12.5-13.9" opacity="1" quantity="27"/>' +
  '<ColorMapEntry color="#7B2382" label="13.9-15.4" opacity="1" quantity="28"/>' +
  '<ColorMapEntry color="#982D80" label="15.4-17.2" opacity="1" quantity="29"/>' +
  '<ColorMapEntry color="#B63679" label="17.2-19.1" opacity="1" quantity="30"/>' +
  '<ColorMapEntry color="#D3436E" label="19.1-21.2" opacity="1" quantity="31"/>' +
  '<ColorMapEntry color="#EB5760" label="21.2-23.5" opacity="1" quantity="32"/>' +
  '<ColorMapEntry color="#F8765C" label="23.5-26.1" opacity="1" quantity="33"/>' +
  '<ColorMapEntry color="#FD9969" label="26.1-29" opacity="1" quantity="34"/>' +
  '<ColorMapEntry color="#FEBA80" label="29-32.1" opacity="1" quantity="35"/>' +
  '<ColorMapEntry color="#FDDC9E" label="32.1-35.6" opacity="1" quantity="36"/>' +
  '<ColorMapEntry color="#FCFDBF" label="35.6-40" opacity="1" quantity="39"/>' +
 '</ColorMap>' +
 '<ContrastEnhancement/>' +
'</RasterSymbolizer>';

var mean_20_50 =
'<RasterSymbolizer>' +
 '<ColorMap type="ramp">' +
  '<ColorMapEntry color="#000004" label="0-2" opacity="1" quantity="11"/>' +
  '<ColorMapEntry color="#0C0927" label="2-5.7" opacity="1" quantity="19"/>' +
  '<ColorMapEntry color="#231151" label="5.7-10" opacity="1" quantity="24"/>' +
  '<ColorMapEntry color="#410F75" label="10-12.5" opacity="1" quantity="26"/>' +
  '<ColorMapEntry color="#5F187F" label="12.5-13.9" opacity="1" quantity="27"/>' +
  '<ColorMapEntry color="#7B2382" label="13.9-15.4" opacity="1" quantity="28"/>' +
  '<ColorMapEntry color="#982D80" label="15.4-17.2" opacity="1" quantity="29"/>' +
  '<ColorMapEntry color="#B63679" label="17.2-19.1" opacity="1" quantity="30"/>' +
  '<ColorMapEntry color="#D3436E" label="19.1-21.2" opacity="1" quantity="31"/>' +
  '<ColorMapEntry color="#EB5760" label="21.2-23.5" opacity="1" quantity="32"/>' +
  '<ColorMapEntry color="#F8765C" label="23.5-26.1" opacity="1" quantity="33"/>' +
  '<ColorMapEntry color="#FD9969" label="26.1-29" opacity="1" quantity="34"/>' +
  '<ColorMapEntry color="#FEBA80" label="29-32.1" opacity="1" quantity="35"/>' +
  '<ColorMapEntry color="#FDDC9E" label="32.1-35.6" opacity="1" quantity="36"/>' +
  '<ColorMapEntry color="#FCFDBF" label="35.6-40" opacity="1" quantity="39"/>' +
 '</ColorMap>' +
 '<ContrastEnhancement/>' +
'</RasterSymbolizer>';

var stdev_0_20 =
'<RasterSymbolizer>' +
 '<ColorMap type="ramp">' +
  '<ColorMapEntry color="#fde725" label="low" opacity="1" quantity="1"/>' +
  '<ColorMapEntry color="#5dc962" label=" " opacity="1" quantity="3"/>' +
  '<ColorMapEntry color="#20908d" label=" " opacity="1" quantity="4"/>' +
  '<ColorMapEntry color="#3a528b" label=" " opacity="1" quantity="5"/>' +
  '<ColorMapEntry color="#440154" label="high" opacity="1" quantity="6"/>' +
 '</ColorMap>' +
 '<ContrastEnhancement/>' +
'</RasterSymbolizer>';

var stdev_20_50 =
'<RasterSymbolizer>' +
 '<ColorMap type="ramp">' +
  '<ColorMapEntry color="#fde725" label="low" opacity="1" quantity="1"/>' +
  '<ColorMapEntry color="#5dc962" label=" " opacity="1" quantity="3"/>' +
  '<ColorMapEntry color="#20908d" label=" " opacity="1" quantity="4"/>' +
  '<ColorMapEntry color="#3a528b" label=" " opacity="1" quantity="5"/>' +
  '<ColorMapEntry color="#440154" label="high" opacity="1" quantity="6"/>' +
 '</ColorMap>' +
 '<ContrastEnhancement/>' +
'</RasterSymbolizer>';

var raw = ee.Image("ISDASOIL/Africa/v1/carbon_total");
Map.addLayer(
    raw.select(0).sldStyle(mean_0_20), {},
    "Carbon, total, mean visualization, 0-20 cm");
Map.addLayer(
    raw.select(1).sldStyle(mean_20_50), {},
    "Carbon, total, mean visualization, 20-50 cm");
Map.addLayer(
    raw.select(2).sldStyle(stdev_0_20), {},
    "Carbon, total, stdev visualization, 0-20 cm");
Map.addLayer(
    raw.select(3).sldStyle(stdev_20_50), {},
    "Carbon, total, stdev visualization, 20-50 cm");

var converted = raw.divide(10).exp().subtract(1);

var visualization = {min: 0, max: 60};

Map.setCenter(25, -3, 2);

Map.addLayer(converted.select(0), visualization, "Carbon, total, mean, 0-20 cm");
Open in Code Editor
Total carbon at soil depths of 0-20 cm and 20-50 cm, predicted mean and standard deviation. Pixel values must be back-transformed with exp(x/10)-1. In areas of dense jungle (generally over central Africa), model accuracy is low and therefore artifacts such as banding (striping) might be seen. Soil property predictions were …
ISDASOIL/Africa/v1/carbon_total, africa,aluminium,isda,soil
2001-01-01T00:00:00Z/2017-01-01T00:00:00Z
-35.22 -31.46 37.98 57.08
Google Earth Engine