Exercise 1: The Electromagnetic Spectrum
All objects absorb and reflect light in predictable ways. By measuring how much light an object absorbs in some colors (or wavelengths) and reflects in others, we can tell a lot about that object. For instance, suppose there are three bananas on a table—one is dark brown with a wrinkled and leathery appearance, one is dull yellow in color with a smooth skin, and the third is pale green with a slick and shiny texture—which would you pick to eat and why? Noting its color and texture, most people would choose the yellow banana because these characteristics indicate it is ripe.
The banana example illustrates a concept basic to the art and science of remote sensing. The various colors (or wavelengths) of light that bounce off the banana and travel to your eye allow you to determine whether the banana is ripe, and you never have to touch the banana to make that judgment. This is similar to how “remote sensing” works. (The term “remote sensing” refers to the use of an artificial device to make measurements or observations of an object from a distance.) By observing and measuring things like color, shape, and texture, scientists can learn a lot about Earth’s environment.
Scientists use satellites to remotely sense the entire Earth’s surface and atmosphere. Space-based remote sensors collect measurements of radiant energy in red, green, and blue light as well as wavelengths of radiant energy (such as ultraviolet and infrared light) that human eyes cannot see. Again, knowing how much radiant energy in the visible and non-visible wavelengths an object absorbs, reflects and emits can teach us a lot about that object.
We classify radiant energy by wavelength and organize it into a chart known as the “electromagnetic spectrum.” (A portion of that chart is pictured below.) Wavelength, which we typically measure in micrometers, is the distance between the crests of the waves in a beam of light. The shorter the wavelength, the more energy the wave has.
Remote sensors measure and record how much light at specific wavelengths an object emits and reflects. Similar to the way your car radio tunes into specific frequencies of radio waves, or “bands,” scientists use space-based “spectroradiometers” to detect specific wavelengths of light, also called “bands.” Based upon their knowledge of how objects reflect and absorb certain wavelengths, scientists tailor spectroradiometers to be particularly sensitive to the bands that will tell them the most about the objects they are interested in. Certain bands reveal a lot of details about plants, while other bands are ideal for making images of clouds or the ocean surface.
Using the sensor’s obervations in bands of visible light (red, green, and blue), we can make a true-color image; that is an image that appears in natural color to our eyes. But how can we make pictures of objects using bands that our eyes cannot see? The only way to visually interpret bands of light that our eyes cannot see (such as ultraviolet or infrared) is to assign a color to those bands. In other words, it is possible to make meaningful images by taking three non-visible bands and coloring them red, green, and blue. We call such images “false color.” Let’s see how an area on Earth appears in true and false colors.
Using the pop-down menus below, select Landsat’s visible (red, green, and blue) bands and examine the resulting composite image. Refer to the graphic above to help you determine which channels are red, green, and blue.
Landsat: March 2000
Landsat: March 1991
QuickBird: March 2002
QuickBird (STRI zoom): March 2002
QuickBird (Tower zoom): March 2002
Next, using the same pop-down menus, select various combinations of infrared bands to be represented as red, green, and blue. You may assign whatever color you like to any of the six Landsat bands available there. Notice how the same scene looks very different using different combinations of bands.
Questions to consider:
- What features seem to be most prominent, or brightest, in each of the visible (red, green, and blue) bands? What features appear less prominent, or darker? Why do you think so?
- Now select various combinations of visible and infrared bands to see which combination offers to best contrast between water, vegetated land surface, and human-developed land surface. Which band combination seems to work best, and why? (Note: If you experience problems with ICE you may need to periodically clear your Web browser’s “cache.”)
Helpful hints when working with the ICE tool
- The individual red, green, and blue bands will initially appear dark on your screen. This is because those wavelengths of energy must travel through the atmosphere on their way up to the satellite sensor. During this trip, some of the light gets absorbed or scattered by gases and particles in the air.
- Use the sliders under each of the red, green, and blue thumbnails to adjust the brightness value for each of those bands. See if you can adjust the brightness values to make the composite image appear “natural color” to your eye.
- Each of these bands has a brightness value which ranges from 0 (black) to 255 (white), with 254 shades of gray in between. Use the Plot Transect feature to examine the brightness values of different surface types in this image. Click and drag on the scene to indicate where you want to draw your transect line. The resulting graph displays the brightness values in each band for every pixel along the line segment you draw.
- Click the Zoom & Roam button and then click on the scene to enlarge it. This can help you zoom in on features of interest in the image you may wish to examine.
- To learn more about what you can do with the Image Composite Editor, review the ICE User’s Guide.