Red stars show the locations where spectroradiometer readings were taken. The white dots are locations of archaeological sites that have been identified as being potentially eligible for listing on the National Register of Historic Places, a list of the most important archaeological sites. The background is the first version of the DDM. Red areas are those where the model predicts intact archaeological sites to be most likely, green areas are those least likely, yellow and organize designate areas where such sites are somewhat likely to be found. We can see here that this first version of the DDM performs well; the atmospheric correction that will be based on spectroradiometer readings should further improve the performance of the model.
The light that we see with our eyes is just a small fraction of the many bands of electromagnetic radiation emitted by the sun. The wavelengths of these other bands are longer or shorter than those of visible light. These invisible wavelengths are reflected (scattered) differently by the various materials that they encounter and certain instruments can record this. These data can be interpreted in ways that provide information about the world that we can’t see with our eyes, such as vegetation health or soil and rock types.
Interpreting such data from landscapes and land cover, and monitoring how it changes over time, can provide clues to the locations and importance of archaeological sites. The ways in which humans interact with landscapes alter land cover; these changes often persist for hundreds or thousands of years. Features constructed by humans, such as shelters for habitation or storage, irrigation systems, and fortifications, can leave traces in the landscape or land cover that remain long after the structures themselves are gone. Such structures can change the shape of the landscape itself, and materials and chemical compounds left behind from those structures or the activities that took place in and around them can sometimes be detected. All of these remnants can help archaeologists not only to find sites, but also to characterize and describe the sites they find.
Unfinished projectile point at an archaeological site where spectroradiometerdata were gathered.
Instruments that sense and record the radiation that our eyes cannot see can be carried by aircraft, satellites, or people. When aircraft or satellites collect this kind of data, the data must be corrected to remove the interference of the very air that the radiation must pass through, both when it comes from the sun to the ground and when it reflects from the ground to the sensor. One of the best ways to make this kind of adjustment, which is called atmospheric correction, is to use a handheld version of the same sensor and bring it to the place being studied. This way, one can compare the data collected with the device on the ground with the data collected by sensors on the aircraft or satellite, then adjust the aircraft or satellite data accordingly.
Atmospheric correction is especially necessary for detecting the small environmental differences that indicate where people lived, hunted, gathered food, and made tools. Most archaeological sites are not pyramids or temples, but places where people lived or worked in simple shelters. These sites, though perhaps not outwardly impressive, tell us how people lived, how they organized themselves in ways that allowed them to cooperate or caused them to fight, how they changed the environment and how environmental change influenced their lives, how they were like us and how they were different, and, overall, how the world of the past led to the world of today.
These landforms, called playas, are remnants of ancient lakes dating to many thousands of years ago. Sunlight reflected from these lakes provides spectral data that are especially useful in correcting for atmospheric interference that must be taken into consideration when analyzing spectral data collected from satellites and aircraft.
CSRM would like to thank the United States Department of Defense Legacy Resource Management Program for the support that made this project possible. This work is done in collaboration with the NASA Jet Propulsion Laboratory at Caltech (JPL/NASA) and the Johns Hopkins University Whiting School of Engineering Department of Applied Mathematics and Statistics.