Google Earth Engine Tutorial: Analyze Lake Shoreline Changes

Google Earth Engine (GEE) is a powerful platform for analyzing environmental changes. This tutorial will guide you through the process of analyzing lake shoreline changes using GEE, leveraging satellite imagery and time-series analysis. By the end, you’ll have a script to monitor and visualize shifts in lake boundaries over time.

Step 1: Set Up Your Environment

Before starting, ensure you have a Google account and access to the Google Earth Engine Code Editor. Log in to https://code.earthengine.google.com, then open the Code Editor.

To begin, load the GEE library in your script:

var ee = require('ee'); ee.Authenticate();

Run the script to authorize your account. Once authenticated, you’re ready to proceed.

Step 2: Define the Study Area and Time Range

Identify the lake and define its boundaries. For example:

var lakeRegion = ee.FeatureCollection('FAO/GAUL_SYM/2015');  
var selectedLake = lakeRegion.filter(ee.Filter.eq('ADM0_NAME', 'Lake Superior'));

Adjust the region selection based on your specific lake. Define the time range for analysis, such as the past 20 years:

var startDate = '2000-01-01';  
var endDate = '2020-12-31';

Step 3: Load and Filter Satellite Data

Select a satellite dataset, such as Landsat 8 or Sentinel-2, and filter imagery for the defined time and region:

var dataset = ee.ImageCollection("LANDSAT/LC08/C01/T1_SR")  
  .filterDate(startDate, endDate)  
  .filterBounds(selectedLake);

This example uses Landsat 8 Surface Reflectance data. Replace with other datasets if needed.

Step 4: Compute a Water Index

Use the Normalized Difference Water Index (NDWI) to identify water bodies:

var ndwi = dataset.map(function(image) {  
  return image.select(['SRB3', 'SRB4']).reduce(ee.Reducer.mean())  
    .multiply(10000)  
    .rename('NDWI');  
});

NDWI is calculated as (NIR – SWIR) / (NIR + SWIR). Adjust bands based on the satellite sensor.

Step 5: Create a Shoreline Mask

Generate a water mask by applying a threshold to the NDWI values:

var waterMask = ndwi.map(function(image) {  
  return image.select('NDWI').gt(0.3);  
});

Change the threshold value (e.g., 0.3) based on your lake’s characteristics. This helps isolate water pixels.

Step 6: Time-Series Analysis and Visualization

Aggregate the mask to compute the lake’s extent over time:

var shorelineAnalysis = waterMask.select('NDWI').median();  
Map.addLayer(shorelineAnalysis, {palette: ['blue'], max: 1}, 'Shoreline Change');

Export the analysis for further use with:

Export.image.toDrive({  
  image: shorelineAnalysis,  
  description: 'LakeShorelineAnalysis',  
  folder: 'GEE_Exports',  
  fileName: 'shoreline_mask',  
  region: selectedLake.geometry(),  
  scale: 30,  
  maxPixels: 1e10  
});

Step 7: Monitor Annual Changes

Create a time series by iterating over years and calculating the water area:

var years = ee.List.sequence(2000, 2020);  
var yearlyStats = years.map(function(y) {  
  var yearImage = dataset.filter(ee.Filter.calendarDate(y, 'year')).median();  
  var waterArea = yearImage.select('NDWI').gt(0.3).reduce(ee.Reducer.sum());  
  return ee.Feature(null, {year: y, waterArea: waterArea.get('NDWI')});  
});

Visualize the results using the Chart API or export to CSV for analysis.