Tutorials/Post - Remote Sensing, GIS, Earth System, Geo-AI/ML

Remapping National Land Use Land Cover (LULC) to Forest Resources Assessment (FRA) and Intergovernmental Panel on Climate Change (IPCC) classes

Land cover maps play an important role in decision making regarding land use, agriculture and food production, maintaining the ecosystem, and natural resource management, etc.

The National Land Use Land Cover (LULC) categories are usually more detailed and have more classes than the classes suggested by the Intergovernmental Panel on Climate Change (IPCC) and the Forest Resources Assessment (FRA) by Food and Agriculture Organization (FAO).

For instance, the LULC categories for Myanmar and the associated FRA and IPCC classes are shown in the figure below.


We use Google Earth Engine to convert our National LULC maps to the ones suggested by FRA and IPCC.

    1. Get the original and remapped classes.
var classStruct = {
  'unknown': {number: 0, color: '6f6f6f'},
  'surface water': {number: 1, color: '0000ff'},
  'snow and ice': {number: 2, color: '808080'},
  'mangroves': {number: 3, color: '556b2f'},
  'cropland': {number: 4, color: '7cfc00'},
  'urban and built up': {number: 5, color: '8b0000'},
  'grassland': {number: 6, color: '20b2aa'},
  'closed forest': {number: 7, color: '006400'},
  'open forest': {number: 8, color: '90ee90'},
  'wetlands': {number: 9, color: '42f4c2'},
  'woody': {number: 10, color: '8b4513'},
  'other': {number: 11, color: '6f6f6f'}

var remappedClassStruct = {
  'Forest': {number: 0, color: '006400'},
  'Grassland': {number: 1, color: '20b2aa'},
  'Cropland': {number: 2, color: '7cfc00'},
  'Settlements': {number: 3, color: '8b0000'},
  'Wetlands': {number: 4, color: '0000ff'},
  'Other land': {number: 5, color: '6f6f6f'},
    1. Get the original and remapped value for the raster
var originalValue = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11];
var mappedValue =   [5, 4, 5, 0, 2, 3, 1, 0, 0, 4,  0,  5];
    1. Load your landcover map and remap each of the images in the collection with your remapped value and with the appropriate color for each class
var collection  = ee.ImageCollection("projects/servir-mekong/LandCoverMyanmar");

for (var y = 2000; y < 2018; y++) {
  var startDate = ee.Date.fromYMD(y, 1, 1);
  var endDate = ee.Date.fromYMD(y, 12, 31);
  var img = ee.Image(collection.filterDate(startDate, endDate).first());
  img = img.select("classification").remap(originalValue, mappedValue);
  Map.addLayer(img, {palette: REMAPPED_PALETTE, min: 0, max: remappedClassNamesList.length - 1}, 'IPCC ' + y.toString(), false);

The link to the IPCC class is here and the link to the FRA class is here.

Happy Earth Observation 🙂


Calculating the Standing Age of the Forest in the Mekong Region using the Google Earth Engine


Change Detection in Google Earth Engine using Sentinel 1 images

1 Comment

  1. Suraj shah

    wow what an informatic blog
    love the management of category
    great work keep lighting us!!

Leave a Reply

Powered by WordPress & Theme by Anders Norén