Lab 2 - Hello Landsat!

Lab Instructions

1. Part 1 - Accessing an Image

   To begin, open the Code Editor and create a new script. We'll load a single Landsat 5 image using its unique ID:

   var first_image = ee.Image('LANDSAT/LT05/C02/T1_L2/LT05_118038_20000606');

   Understanding the code:

   • var first_image - Creates a variable to store our image
   • ee.Image() - Earth Engine function to load a single image
   • The long string is the image's unique identifier (like a file path)

   Click Run. Nothing visible happens yet—the image is loaded into memory, waiting to be used.

   Print the image to see its metadata (information about the image):

   print(first_image);

   Expand the output in the Console. You'll see this image has 19 different bands. The first six bands (SR_B1 through SR_B7) represent different portions of the electromagnetic spectrum:

   • SR_B1 - Blue light
   • SR_B2 - Green light
   • SR_B3 - Red light
   • SR_B4, SR_B5, SR_B7 - Infrared (invisible to humans)

   

2. Part 2 - Visualizing Single Bands

   Now let's add a band to the map. We use Map.addLayer():

   Map.addLayer(first_image);
   Map.setCenter(122, 32, 7);

   The image might look dark or washed out. That's because we haven't told Earth Engine how to display the values. Let's add visualization parameters:

   Map.addLayer(
       first_image,
       {
           bands: ['SR_B1'],
           min: 8000,
           max: 17000
       },
       'Band 1 (Blue)'
   );

   Understanding visualization parameters:

   • bands - Which band(s) to display
   • min - Pixel values below this appear black
   • max - Pixel values above this appear white
   • 'Band 1 (Blue)' - The layer name in the Layers panel

   Now add all three visible bands as separate layers:

   Map.addLayer(
       first_image,
       {bands: ['SR_B1'], min: 8000, max: 17000},
       'Band 1 (Blue)'
   );
   
   Map.addLayer(
       first_image,
       {bands: ['SR_B2'], min: 8000, max: 17000},
       'Band 2 (Green)'
   );
   
   Map.addLayer(
       first_image,
       {bands: ['SR_B3'], min: 8000, max: 17000},
       'Band 3 (Red)'
   );

   Try It

   Use the Inspector tool (right panel) to click on different areas of the map. Compare the band values between water, vegetation, and urban areas. Which band shows the most contrast between land and water?

3. Part 3 - True-Color Composites

   Instead of viewing bands separately, we can combine three bands into a single RGB composite. By assigning:

   • Red band → Red display channel
   • Green band → Green display channel
   • Blue band → Blue display channel

   We create an image that looks like a natural photograph:

   Map.addLayer(
       first_image,
       {
           bands: ['SR_B3', 'SR_B2', 'SR_B1'],
           min: 8000,
           max: 17000
       },
       'True Color (Natural)');

   

   Note: The order of bands matters! Earth Engine assigns bands to RGB channels in the order you list them: [Red, Green, Blue]. For Landsat 5, B3=Red, B2=Green, B1=Blue.

4. Part 4 - False-Color Composites

   Here's where remote sensing gets powerful. We can assign any band to any color channel—including bands our eyes can't see! The color-infrared composite puts the near-infrared band (B4) in the red channel:

   Map.addLayer(
       first_image,
       {
           bands: ['SR_B4', 'SR_B3', 'SR_B2'],
           min: 8000,
           max: 17000
       },
       'False Color (Color Infrared)');

   

   Why does vegetation appear red?

   Healthy plants strongly reflect near-infrared light (for temperature regulation). Since we've assigned NIR to the red display channel, areas with healthy vegetation glow bright red. This makes it much easier to identify vegetation than in true-color images!

   Try It

   Toggle between the True Color and False Color layers using the Layers panel. Can you identify agricultural fields that were hard to see in true color?