CIESIN Thematic Guides The fundamental principles of remote sensing derive from the characteristics and interactions of electromagnetic radiation (EMR) as it propagates from source to sensor. The principles relate to the following: 1) the source of energy and the type and amount of energy it provides; 2) the absorption and scattering effects of the atmosphere on EMR; 3) the mechanisms of EMR interaction with Earth surface features; and 4) the nature of sensor response as determined by the type of sensor. Lillesand and Kiefer (1987) provide an overview of these principles in the chapter "Concepts and Foundations of Remote Sensing" in Remote Sensing and Image Interpretation. Other reviews of the fundamental principles of remote sensing are available in most remote sensing textbooks, including those by Avery and Berlin (1992), Szekielda (1988), Elachi (1987), Sabins (1987), and Swain and Davis (1978). A thorough treatment is available in the Manual of Remote Sensing (Simonett, D.S., ed., 1983).
Most satellite sensors detect EMR electronically as a continuous stream of digital data. The data are transmitted to ground reception stations, processed to create defined data products, and made available for sale to users on a variety of digital data media. Once purchased, the digital image data are readily amenable to quantitative analysis using computer-implemented digital image processing techniques. Sabins (1987) describes these techniques in the chapter "Digital Imaging Processing" in Remote Sensing: Principles and Interpretation. Some of these techniques (such as data error compensations, atmospheric corrections, calibration, and map registration) essentially involve preprocessing the data for subsequent interpretation and analysis. Another group of techniques is designed to selectively enhance the digital data and produce hard-copy image formats for interpreters to study. For these images, some of the principles and techniques of airphoto interpretation can be applied to manual analysis of the image information content. Lillesand and Kiefer (1987) discuss such analyses in Remote Sensing and Image Interpretation. A third major group of digital processing techniques involves information extraction through the implementation of a wide range of simple to complex mathematical and statistical operations on the numerical data values in the image. The results of these operations provide output such as derived information variables (that might relate to terrain brightness or vegetation condition), categorized land and water features, or images showing changes over time.
A discussion of remote sensing technology would not be complete without mention of geographic information systems (GIS). Satellite remote sensing represents a technology for synoptic acquisition of spatial data and the extraction of scene-specific information. GIS provides a computer-implemented spatially oriented database for evaluating the information in conjunction with other spatially formatted data and information that may be acquired from remote sensor data, maps, surveys, and other sources of spatially referenced information. Figure 1 illustrates the concept of spatial data integration in a GIS. The Environmental Systems Research Institute (1992) describes this technology further in "Lesson 1: Why GIS?"
GIS technology should aid human dimensions studies of global change by enabling the integration and joint analysis of human science data and natural science data. In "The Potential Methodological Impact of Geographic Information Systems on the Social Sciences," Marble (1990) notes that GIS will be particularly instrumental for heightening social science researchers' awareness of the spatial complexity that surrounds all societal structures and conditions much of human behavior. In "Landscape: A Unifying Concept in Regional Analysis," Crumley and Marquardt (1990) propose the concept of "landscape" for unifying the study of human interactions with their environment and explain that GIS is the tool that allows for practical study of landscape elements. Madry and Crumley (1990) provide an informative discussion of the development and use of a GIS for evaluating the historical interaction of environment and culture in the Arroux River valley in "An Application of Remote Sensing and GIS in a Regional Archaeological Settlement Pattern Analysis."
The emphasis on satellite-image data in this guide is not meant to diminish the roles of aerial photography and field work for providing "ground truth" data. Any comprehensive program of mapping landscape features (meaning the spatial manifestation of the relationship between humans and their environment, as proposed by Crumley and Marquardt, 1990), evaluating and quantifying their characteristics, and monitoring change generally requires supporting ground truth data to develop and verify the use of satellite-image data. Developing the intended use of satellite sensor data refers to establishing the qualitative associations or quantitative relationships one wants to implement; in other words, determining the capability to accomplish an objective using a particular type of data. Verification refers to assessment of performance and refinement of the capability. These activities enable establishing the link between satellite image data and the desired landscape information (both natural and human dimensions attributes).
The type of supporting ground truth data employed in specific studies varies relative to the scale of the primary data being used. When working with coarse-resolution continental- and global-scale satellite image data (such as Advanced Very High Resolution Radiometer data of 1-km spatial resolution), high-resolution satellite data (such as Landsat or Systeme Probatoire d'Observation de la Terra (SPOT)) may serve as supporting ground-truth data. On the other hand, scientists studying the human dimensions of global change with high-resolution satellite images may need to incorporate interpretations of airphotos and information acquired in the field to corroborate their intended use of the satellite data and validate their results.