[Dataset Title/Name]: Machine learning human footprint index (ml-HFI)

[ccmothes]

Contact Details

pkeys@bu.edu

Dataset description

This dataset introduces a novel machine learning-based Human Footprint Index (ml-HFI) with 300-meter spatial resolution, with values ranging from 0 to 100, where 0 represents intact natural areas and higher values indicate increasing human pressure. The ml-HFI is developed using a convolutional neural network (CNN) trained on an existing Human Footprint Index (HFI) dataset, with Landsat imagery as input features. This approach builds upon the approach by Keys et al. (2021) and removes dependencies on externally processed datasets, making it a fully self-sufficient index that only requires Landsat data for calculation. Landsat imagery serves as the input data, pre-processed using Google Earth Engine to remove cloud contamination and ensure consistent quality, including cloud, snow, and shadow masking, and annual median composites to reduce noise.

Dataset Details

Spatial Extent	Global
Spatial Resolution	300m
Temporal Resolution	Annual
Time Span	1999-2024

Citations and Links

Publication: https://doi.org/10.1088/3049-4753/ae2278
Dataset DOI: https://doi.org/10.5061/dryad.m63xsj4fk
Machine Learning Model Code: https://doi.org/10.5281/zenodo.17670299
ml-HFIv2 Web Application: https://mlhfi.com/

Earth Engine Snippet if dataset already in GEE

// Load the annual MLHFI collection
var annualCollection = ee.ImageCollection('projects/mlhfi-v2/assets/global_annual');

// Load the difference layer
var diffLayer = ee.Image('projects/mlhfi-v2/assets/diff_2022to2024_vs_1999to2001');

// Filter for specific years
var mlhfi1999 = annualCollection.filter(ee.Filter.stringContains('system:index', '1999')).first();
var mlhfi2024 = annualCollection.filter(ee.Filter.stringContains('system:index', '2024')).first();

// Define color palettes
// Viridis Inferno palette for annual layers
var infernoColors = [
  '000004', '1b0c41', '4a0c6b', '781c6d', 'a52c60',
  'cf4446', 'ed6925', 'fb9b06', 'f7d13d', 'fcffa4'
];

// Divergent palette: magenta (increase) to green (decrease)
var divergentColors = [
  '035403', '048204', '66ff66', 'f0f0f0', // greens to white
  'ffccff', 'ff66ff', 'ff00ff'             // white to magentas
];

// Visualization parameters
var annualVis = {
  min: 0.0,
  max: 100,
  palette: infernoColors
};

var changeVis = {
  min: -35,
  max: 35,
  palette: divergentColors
};

// Set map center (adjust as needed)
Map.setCenter(0, 0, 2); // Global view

// Add layers to map
Map.addLayer(diffLayer, changeVis, 'Change 1999-2001 → 2022-2024');
Map.addLayer(mlhfi1999, annualVis, 'MLHFI 1999');
Map.addLayer(mlhfi2024, annualVis, 'MLHFI 2024');


// Print collection info
print('Annual Collection:', annualCollection);
print('MLHFI 1999:', mlhfi1999);
print('MLHFI 2024:', mlhfi2024);
print('Difference Layer:', diffLayer);

Sample Code: https://code.earthengine.google.com/d50154ebb2309576dc09159ce1f15ef0

Enter license information

CC0 1.0 Universal

Keywords

human footprint, machine learning, Anthropocene, global change

Code of Conduct

• I agree to follow this project's Code of Conduct