THE SCIENCE
What is satellite remote sensing? It is the name given to a host of activities designed to gather data and information about the Earth's surface without actually being in contact with it. This is done by using space-borne sensors to measure energy reflected or emitted by surface features and then to process this data into imagery from which information can be visually interpreted.
Remote sensing satellites orbit around the poles rather than along the equator. Unlike weather or communication satellites orbiting 25,000 miles above the Earth, most remote sensing satellites orbit at an altitude between 200-400 miles. This enables them to pass over every part of our planet in a shorter amount of time and produce multiple datasets for comparison.
A common misperception about these satellites is that they take photographs from space like a camera. Although there are some similarities between the sensors in these satellites and a digital camera, the actual process of interpreting satellite data is far more complicated. Whereas a digital camera records the brightness of a scene over a broad range of wavelengths corresponding to red, green, and blue light - satellites have one set of sensors for each of these very specific wavelengths. This high precision enables scientists to distinguish different vegetation and mineral types as well as other types of land use by the ways in which these materials reflect or emit light. This same precision makes it possible for artists to produce colorful images by applying the same image processing techniques.
Although our eyes only respond to a very narrow range of light from red through blue, the sun produces a much wider range of light than this visible spectrum. Except for a small amount of light that causes sunburn, shorter wavelengths of light (ultra-violet) are absorbed by the upper atmosphere. Longer wavelengths of light called "infra-red" readily pass through the atmosphere unnoticed by the human eye. Contrast is created between different surface features based on the manner in which these wavelengths of infra-red light are reflected.
Remote sensing satellites take advantage of this contrast by recording the brightness of an object at various wavelengths of light. These include visible, near-visible infra-red, short wavelength infra-red, and thermal infra-red. With the exception of longer wavelengths from the thermal band, all other infra-red wavelengths behave just like ordinary visible light - they are emitted by the sun and an object absorbs some of this energy and reflects the rest.
PIXELS, RESOLUTION and IMAGE PROCESSING
Just like a digital camera, a remote sensing satellite partitions its viewing region into a set of small rectangles corresponding to pixels. Satellites record the brightness of individual pixels at each of their target wavelengths. The area on the ground covered by a pixel is called the spatial resolution of the sensor and represents the smallest sized objects on the ground that can be resolved by the satellite.
Using 15 meter spatial resolution as an example, each pixel in the image corresponds to a 15m x 15m square on the ground. Each of these pixels has a series of numbers that represent the brightness at each of the recorded wavelengths of light. To produce an image that can be printed on paper or canvas or viewed on a monitor, this information must be mathematically manipulated to give just three values for each pixel. These values are determined by the primary colors that all other colors originate from - red, green and blue (RGB). This manipulation of RGB values is referred to as image processing.
Unfortunately, it is not possible to simply choose any three brightness values to represent the RGB colors. This would result in images that have very few visible colors and low contrast. Image processing tools are designed to produce RGB values that most effectively separate different types of surface objects. Attempting to assign different colors to objects of the same reflectivity would compromise the delicate balance of the image processing techniques developed by each artist.