Satellites, Non-Governmental High Resolution

High-resolution satellites, generally understood to be those with a spatial resolution of 2 meters (6.6 feet) or less, have the capability to provide forensic information from areas that are otherwise inaccessible to law enforcement officials.

Resolution is a measure of the ability of an image to depict detail. When used in reference to digital images such as those produced by remote sensing satellites, resolution generally refers to size of the pixels, or fundamental elements, comprising the image. A 2-meter resolution image consists of elements representing the average color or intensity of a 2x2 meter area of Earth's surface. Nothing smaller than 2x2 meters will be depicted as a distinct object. The smallest objects that can be clearly identified on an image, however, will be much larger than the resolution because many pixels are required to represent the characteristic shape or outline of an object. A 2-meter resolution satellite image might, therefore, show distinct images of buildings covering tens or hundreds of square meters, but not a small shed or automobile covering an area the size of a 2x2-meter pixel.

The best commercial satellites operating in 2005 had resolutions of 1 meter (3.3 feet) or less. However, intelligence satellites operated by the U.S. government were believed to have a resolution of about 2 centimeters (0.8 inches). Images with that resolution, however, have never been released for public use.

The first remote sensing satellites were built, launched, and operated by government agencies in the 1960s. In the interest of national security, images from these satellites were tightly controlled and generally inaccessible to civilian officials and forensic scientists. Imagery from the first Landsat satellites, launched by the United States in the 1970s, was publicly available but its low resolution (tens of meters) made it useful only for regional studies. After an attempt to privatize and eliminate government subsidies for the Landsat program in the 1980s, the United States passed the Land Remote Sensing Policy Act of 1992. This act emphasized the importance of satellite imagery, returned the Landsat program to government operation, mandated that its data be made available at cost, and included a provision for the licensing of commercially operated remote sensing satellites. At about the same time, the French government developed the SPOT (Satellite Pour l'Observation de la Terre) program and marketed its imagery through a subsidized corporation. Like Landsat imagery, however, SPOT imagery generally did not provide the resolution necessary for detailed forensic work.

The Landsat and SPOT satellites paved the way for a new generation of high-resolution commercial satellites that provide images detailed enough for forensic work. The commercial IKONOS satellite, launched by the multi-national Space Imaging consortium in 1999, orbits Earth at an altitude of 680 kilometers (422.5 miles) and provides panchromatic (black and white) images with 1-meter resolution. The EROS A1 satellite, built by the ImageSat International consortium in Israel and launched from Russia in 2000, provides 1.8-meter (6-foot) resolution. Its successor, the EROS B1, will have 0.70-meter (2.3-foot) resolution when operable in 2006. The highest resolution commercial satellite imagery available in 2005 came from the QuickBird satellite operated by the Colorado company DigitalGlobe. QuickBird produces 0.62-meter (2-foot) resolution panchromatic images and 2.4-meter (7.9-foot) resolution color images. The panchromatic images, in particular, are detailed enough to depict individual automobiles, pieces of machinery, or ground disturbance associated with illegal activities.

Panchromatic images have higher resolutions (smaller pixel size) than color, or multi-spectral, images. This is because digital imaging sensors have a fixed number of photosites that respond to light. When a panchromatic image is made, each photosite senses the total intensity of light. When a multi-spectral image is made, in contrast, the photosites must be divided among the spectral bands being depicted. Thus, a color image consisting of infrared, red, green, and blue bands would have one-fourth the resolution of a panchromatic image from a sensor with the same number of photosites.

One particularly high profile application of commercial high-resolution satellite imagery was the search for debris from the space shuttle Columbia, which exploded over Texas in 2003. The QuickBird satellite was immediately redirected to cover the accident area, and the resulting images showed areas of broken trees and highly reflected debris. The detailed satellite images allowed accident investigators to better document the extent of the debris field and recover pieces of the shuttle.

High-resolution commercial satellite imagery is also invaluable in the aftermath of natural disasters such as the 2004 Indian Ocean tsunami. There, it was used to help guide relief efforts and provided important information for researchers studying the effects of tsunamis.

Other applications of commercial satellite imagery in forensic science are less exotic. Government officials in Arizona, Georgia, and Minnesota have used satellite imagery to detect illegal cotton cultivation, logging, and pollution. Because satellites pass over any given location no more frequently than every few days, they are best suited for the characterization of slow processes such as growing crops or persistent problems such as air or water pollution. For the same reason, it is unlikely that satellite imagery will provide images that catch thieves, kidnappers, rapists, or murderers committing crimes.

Like photographs and videotapes, satellite images can be manipulated and must therefore be authenticated for use in court. Prosecutors or plaintiff's attorneys must establish that any processing or enhancement techniques used on the imagery were properly documented and followed accepted professional standards, whereas defendant's attorneys may question the authenticity of imagery used against their clients. Although some manipulation must be done in order to transform digital information into a visible image, it is critical to establish that the manipulation did not distort or otherwise misrepresent the area being depicted in the imagery.