Google Earth Engine Tutorial: Detect Los Angeles Wildfires

Google Earth Engine Tutorial: Detect Los Angeles Wildfires

Google Earth Engine (GEE) is a powerful platform for analyzing geospatial data, making it ideal for monitoring wildfires. This tutorial guides you through the process of detecting wildfires in Los Angeles using satellite imagery and GEE’s tools.

Step 1: Access Google Earth Engine

Create a Google account and access the GEE Code Editor at https://code.earthengine.google.com. This environment allows you to run geospatial analyses without installing software.

Step 2: Load Wildfire Data

GEE offers datasets like MODIS/006/MOD14A1 (active fire data) and USGS Landsat for burn scar detection. Use the following code to load and filter data:

var modisFire = ee.ImageCollection('MODIS/006/MOD14A1')
  .filterDate('2023-01-01', '2023-12-31')
  .filter(ee.Filter.geometry(ee.Geometry.Rectangle([-118.41, 33.82, -117.92, 34.30])));
  
var landsat = ee.ImageCollection('LANDSAT/LC08/C02/T1_L2')
  .filterDate('2023-01-01', '2023-12-31')
  .filter(ee.Filter.geometry(ee.Geometry.Rectangle([-118.41, 33.82, -117.92, 34.30])));

Step 3: Analyze Active Fires

Use the MODIS/006/MOD14A1 dataset to detect active fires. Highlight the fire pixels using a threshold on the fireMask band:

var activeFires = modisFire.select('fireMask');
Map.setCenter(-118.25, 34.05, 8);
Map.addLayer(activeFires, {palette: 'red'}, 'Active Fires');

Step 4: Visualize Burned Areas

Landsat imagery can identify burn scars by analyzing spectral indices like the Normalized Burn Ratio (NBR). Create a composite of pre- and post-fire images to compare changes:

var preFire = landsat.filterDate('2023-01-01', '2023-03-31').select(['SR_B.4', 'SR_B.5', 'SR_B.6']);
var postFire = landsat.filterDate('2023-06-01', '2023-09-30').select(['SR_B.4', 'SR_B.5', 'SR_B.6']);

var burnScars = preFire.map(function(img) {
  return img.select(['SR_B.4', 'SR_B.5', 'SR_B.6']).rename(['Red', 'Green', 'Blue'])
    .eeImage(right, 1, 1, 1);
});
Map.addLayer(burnScars, {bands: ['Red', 'Green', 'Blue'], min: 0, max: 3000}, 'Burned Areas');

Step 5: Export Results

Export the detected fire areas as a GeoTIFF or CSV using GEE’s export functions. For example:

Export.image.toDrive({
  image: activeFires,
  description: 'LA_Wildfires',
  fileNamePrefix: 'la_wildfire_detection',
  region: ee.Geometry.Rectangle([-118.41, 33.82, -117.92, 34.30]),
  crs: 'EPSG:4326',
  scale: 1000
});