Exercise 2: Green, Greener, Greenest

Down here on Earth, it’s easy for us to spot vegetation and figure out where it is thick and healthy and where it is sparse or struggling. The grass in our lawns, the woods at the park, or the corn field at the farm down the road are easy for us to identify. For a satellite orbiting 700 kilometers above the Earth, it’s not so easy. In the next paragraphs is an explanation of the most common way satellites identify vegetation.

Tropical forest canopy
With warm temperatures, plenty of sunlight, and an abundance of rainfall during most of the year, the rainforest of Panama has a thick, lush canopy with lots of green leafy vegetation. (Image courtesy of The Basic Science and Remote Sensing Initiative (BSRSI), a research program in the Department of Geography at Michigan State University.)

Vegetation reflects both visible light (the kind we see with our eyes) and other radiation from invisible parts of the electromagnetic spectrum in unique ways. These unique reflection patterns are vegetation’s “spectral signature.” Obviously, one of the spectral signatures of leafy vegetation is that it’s green. Vegetation is green because it absorbs almost all other wavelengths of visible light except green, which it reflects back to our eyes or to a satellite.

NDVI sample
Vegetation Indices are important tools for mapping and monitoring the world’s vegetation. By measuring how much visible and near-infrared light is reflected off the surface, scientists can gauge the “greenness” of the vegetation, to which they assign an index value ranging from - 0.1 to 0.9. (Image by Reto Stockli, using Landsat data courtesy Jerome Chave, Centre National pour la Recherche Scientifique, France)

The most commonly used technique for identifying vegetation in satellite data is to compare the amount of visible light reflected by a particular place on Earth to the amount of infrared light. Areas reflecting low levels of visible light and high values of infrared light are likely covered by leafy vegetation. Scientists use computer programs to subtract the amount of visible light observed by satellites from the amount of near-infrared light, and then divide that by the total amount of reflected light in those wavelengths.

Comparison of raw red and infrared data

Deciduous trees comprise as much as 40 percent of the forest in some parts of Panama. Many of these trees lose their leaves during the dry season, from January through February. The two images above show the same tropical forest tree, looking down from above. This tropical tree is leafless in the dry season (like forests in North America that lose their leaves in the winter), so only the branches are left. Below this leafless tree is a shorter tree that is full of leaves. In the red band, the fully-leaved tree below absorbs red light and the wood from the leafless tree above reflects red light. In the infrared, the fully-leaved tree below reflects a lot of light, more than the bare branches above. (Photos courtesy Stephanie Bohlman, U. of Washington)

This calculation produces an estimate of vegetation called the “Normalized Difference Vegetation Index,” or NDVI for short. NDVI values have no units; they can range from - 0.1 (low vegetation) to 0.9 (maximum detectable vegetation). NDVI images are usually colored on a scale of brown for low vegetation to dark green for dense vegetation. NDVI can reveal where natural and agricultural vegetation is under drought or temperature stress and where it is flourishing. Over many years, NDVI observations can show how vegetation and ecosystems may be appearing or disappearing in response to human influence or climate change.

NDVI example
NDVI is calculated from the visible and near-infrared light reflected by vegetation. Healthy vegetation (left) absorbs most of the visible light that hits it, and reflects a large portion of the near-infrared light. Unhealthy or sparse vegetation (right) reflects more visible light and less near-infrared light. The numbers on the figure above are representative of actual values, but real vegetation is much more varied. (Illustration by Robert Simmon, NASA GSFC).

In this next exercise, you can use the Image Compite Editor to produce your own NDVI measures. Using the pop-down menus, please select Landsat channel 4 as the first channel and channel 3 as the second channel. You can use the ICE tool’s math mode to perform the following operation: NDVI equals band 4 minus channel 3 divided by channel 4 plus channel 3 or written as a formula: NDVI = (near infrared - red light) divided by (near infrared + red light).

Landsat: March 2000

Landsat: March 1991

Questions to consider:

  1. Do you see any differences in the features of the tree in the red and near-infrared photos above? Explain any differences you see and what makes them look different.
  2. How is the way a satellite sees Earth’s vegetation similar to the way our eyes see it? How is it different?
  3. What kind of “spectral signature” would you expect a rainforest to reflect back to the satellite? Would it have a high NDVI value or a low NDVI value? Explain why you think so.
  4. Would the NDVI value of the Panama Rainforest be higher or lower than vegetation in a state park near where you live?

Helpful hints for using the ICE tool:

  • Select the bands you wish to use from the pop-down menus above and then click “build.”
  • ICE displays the bands you chose in the first two thumbnails, and the sum of those two bands added together in the third thumbnail.
  • Use the pop-down menus under the thumbnails to set up the desired math operation—the available options are add, subtract, multiply or divide.
  • Use the numbers in the fields under each thumbnail as multipliers. The default values are 1.0, which means “1.0 multiplied by the pixel values shown in the thumbnail.”
  • Once the multipliers and math operations are set as desired, click “compute.” The result of your operation will be displayed in the large window.
  • Using ICE to calculate NDVI is a two-step process. Once bands 4 and 3 are shown in the first two thumbnails, and their sum is shown in the third, go back and change the math function from “add” to “subtract” between first two thumbnails, and then change “result” to “divide” between the second two thumbnails. Clicking the “Compute” button performs the calculation and displays the resulting NDVI product in the large window.
  • To learn more about what you can do with ICE, read the User’s Guide.

back Exercise 1 | Exercise 3 next