Satellite Eyes - Seeing Beyond the Rainbow

Satellites like Landsat can have sensors which "see" in visible light, but they may also be able to record radiation (such as infrared) that it is beyond our capability to see.

The Electromagnetic Spectrum

Visible light is only one kind of electromagnetic radiation that satellites can monitor. Infrared and radar are also part of the electromagnetic spectrum, and each represents radiation in a different wavelength. For example, yellow light has a longer wavelength than blue, and red is longer than yellow. Infrared and radar wavelengths are longer still. By collecting data in different regions of the spectrum, satellites can reveal information that would go undetected by our eyes alone.

Satellite Sensors

The same scene can appear very different when viewed by different satellite sensors. Imagery of California's Imperial Valley illustrates the differences in appearance, coverage, and detail acquired by various orbital missions.

Compare Satellite Images (40k)

Thermal infrared image from the Heat Capacity Mapping Mission (HCMM). This scene covers 15 times the area of a Landsat frame but can only resolve features larger than 600 meters (about 1900 feet). The warmest areas of the region, which show up in the lightest tones, are the deserts of Arizona and Mexico. Thermal imagery can reveal important information on the temperature and properties of surface materials. (26k jpg)
NASA Image

Radar image from the Seasat mission. The smooth surface of the dune field appears dark, while the rough texture and orientation of the mountains give them a brighter appearance. The light and dark patterns in the Sea are due to variations in the water surface caused by the blowing wind. (45k jpg)
NASA Image.

Processing An Image

A satellite image is often not a photograph at all. Complex satellite sensors do not record a scene on film, but instead, collect information that can be converted to computer images. The computer scenes are composed of a mosaic of tiny rectangles called picture elements or "pixels". By manipulating the image on the computer, different aspects of the terrain can be emphasized.

Digital Images

Digital images are collected as a series of numerical values, each value representing the amount of energy radiated or reflected from a unit of area on the Earth's surface. These images are collected in separate wavelength bands including, and beyond, the region of the spectrum detectable by film systems. Information from each band produces a discrete image from the swath of Earth below the spacecraft. Each image is composed of pixels which can be compared to the squares on a checkerboard. Individual pixel values (numbers representing the relative brightness of each point) are transmitted to a receiving station on Earth to produce the rows and columns of a numerical matrix comprising each scene. Pixels with high values will appear bright. Those with low values will be dark. A color composite image is produced by combining three images into one. Each of the three images of the scene, measured in different wavelengths, is assigned a different primary color: red, green, or blue. The brighter the pixel the more intense the color. When corresponding pixels (the same row and column) from each image are added together, the resulting color is a hue that represents the proportion of red, green, or blue from each of the three original digital images.

The large image shows a Landsat Thematic Mapper scene of Washington, D.C., including the Potomac and Anacostia rivers. The box indicates the area around the U.S. Capitol building that is enlarged below.
Digital images of the Capitol area of Washington, D.C., from three different spectral bands are combined into a "false-color" composite. The colors are called "false" because any primary color can be assigned to any band. Thus, vegetation can be featured in red by assigning this color to a near infrared band. Vegetation is very reflective in the near infrared and, therefore, has high brightness values in this band.
Images processed at the Center for Earth and Planetary Studies

Landsat

Radar