How to Read Infrared Satellite Imagery

An image like this has plenty of information, but it's hard to know where to start working on it.

If you've ever looked at the Google satellite picture of your own neighborhood, you probably have an idea that interpreting a remote sensing image isn't the easiest task. Even knowing what should be in the picture, it can take a while to figure out what you're looking at. That's why scientists and engineers added infrared capabilities to satellite imaging systems: adding more spectral information makes it easier to identify features in the image, and also provides another layer of information not present in the visible pictures. A little knowledge about the behavior of infrared light will give some quick hints for interpreting infrared images.

Instructions

1. 

   Thermal infrared pictures can track clouds day or night by measuring their temperature and the temperature of their surroundings.

   Identify the type of image collected by the satellite. Most satellite images are pictures of reflected sunlight that makes its way through the atmosphere and is collected by the satellite. These are usually near-infrared images. Far-infared imagers collect very-long wave "thermal" radiation that is emitted by objects within the scene, whether or not the sun is shining.

2. Find the darkest regions of the near-infrared image. Because water strongly absorbs near-infrared light, clear water will appear black in infrared pictures. For the same reason, damp soil will appear distinctly darker than dry soil.

3. 

   Be careful when interpreting false color infrared images --- here the red is healthy forest and the green is a burned area.

   Compare the near-infrared intensity reflected from forested regions. Deciduous (leafy) trees reflect more infrared radiation than coniferous (needle-leaf) trees, so they will be lighter.

4. Combine the infrared image with data collected in visible wavelengths. For example, white sand and red sandstone both appear to be light gray in infrared wavelengths, so the visible light information is necessary to distinguish the two.

5. 

   Combining visible and infrared images and data from other sources provides the most value for image interpretation.

   Apply your background knowledge. For example, asphalt roads within a city can appear black in near-infrared images --- as if they were water. Using your background knowledge you can infer that the roads are most likely roads, and not canals.

6. Interpret the infrared imagery in concert with information from additional sources. For example, thermal infrared images show the temperature of the clouds. When connected to meteorological models, the temperature turns out to be dependent on the height of the clouds, so the temperature reading provides a measure of altitude.

Tips & Warnings

• Near-infrared images are often rendered in false color. For false color, healthy vegetation, for example, may be shown as red, while "stressed" vegetation shows as pink or yellow. Because false colors do not directly correspond to visible colors, use the guidelines supplied with the image to classify objects in the image.

References

• Texas A&M University---Kingsville: Aerial Photography and Photo Interpretation
• University of Colorado: Remote Sensing of the Environment

Resources

• Emporia State University: Landsat Image Interpretation
• University of Wisconsin: Interpreting Satellite Images