Peatlands, waterlogged terrestrial wetland ecosystems, store large amounts of soil carbon and freshwater, constituting an important component of the global carbon and hydrologic cycles. Accurate information on the global distribution of peatlands is critically needed to simulate the effects of climate change on the global carbon and hydrologic balance. Peat-ML is a spatially continuous global map of peatland fractional coverage generated using machine learning techniques, training with climate, geomorphological and soil data, along with remotely-sensed vegetation indices. Available maps of peatland fractional coverage for 14 relatively extensive regions were used along with mapped ecoregions of non-peatland areas to train the statistical model. The dataset is published on 2021 (https://zenodo.org/records/5794336) in NetCDF format. That approach yielded an average r squared of 0.73 with a root mean squared error and mean bias error of 9.11% and -0.36%, respectively. the associated paper is published here.
Workflow for dataset generation (Joe R. Melton et al., 2022)
Example data visualization of peatland distribution in Indonesia
Earth Engine Snippet if dataset already in GEE
Loading Peatland Distribution for Indonesia
// Load administrative boundaries for Indonesia
var admin1 = ee.FeatureCollection("projects/sat-io/open-datasets/geoboundaries/HPSCGS-ADM1");
var geometry = admin1.filter(ee.Filter.eq('shapeGroup', 'IDN'));
Map.centerObject(geometry, 4);
Map.setOptions("Hybrid");
var peat = ee.Image("projects/ee-pinkychow1010/assets/global_peatland_2022") // just an example, not the identical dataset
.clip(geometry)
.unmask();
// Display the results
Map.addLayer(peat.clip(geometry),
{min: 0, max: 100, palette: ['#f7fcf5', '#c7e9c0', '#74c476', '#238b45', '#00441b']},
'Peatland Distribution', true
);
Enter license information
This work is distributed under the Creative Commons Attribution 4.0 License.
Keywords
peatland, soil carbon, wetland, ecosystem
Code of Conduct
[X] I agree to follow this project's Code of Conduct
Contact Details
pinkychow1010@gmail.com
Dataset description
Peatlands, waterlogged terrestrial wetland ecosystems, store large amounts of soil carbon and freshwater, constituting an important component of the global carbon and hydrologic cycles. Accurate information on the global distribution of peatlands is critically needed to simulate the effects of climate change on the global carbon and hydrologic balance. Peat-ML is a spatially continuous global map of peatland fractional coverage generated using machine learning techniques, training with climate, geomorphological and soil data, along with remotely-sensed vegetation indices. Available maps of peatland fractional coverage for 14 relatively extensive regions were used along with mapped ecoregions of non-peatland areas to train the statistical model. The dataset is published on 2021 (https://zenodo.org/records/5794336) in NetCDF format. That approach yielded an average r squared of 0.73 with a root mean squared error and mean bias error of 9.11% and -0.36%, respectively. the associated paper is published here.
Workflow for dataset generation (Joe R. Melton et al., 2022)
Example data visualization of peatland distribution in Indonesia
Earth Engine Snippet if dataset already in GEE
Loading Peatland Distribution for Indonesia
Enter license information
This work is distributed under the Creative Commons Attribution 4.0 License.
Keywords
peatland, soil carbon, wetland, ecosystem
Code of Conduct