1061 REMOTE SENSING Remote sensing is the act of acquiring information about an object from a distance. Environmental applications of remote sensing typically involve the collection of photographic or electronic images of the Earth’s surface or atmosphere from airborne or spaceborne platforms. Visual interpretation or com- puter processing can then be used to analyze these images. The history of remote sensing dates back to the 19th cen- tury, when the first aerial photographs were taken from bal- loons and kites. The invention of the airplane provided a better platform and aerial photography advanced rapidly after World War I. With the dawn of the Space Age in 1957, the field of photographic remote sensing expanded to include pictures taken from satellites and other space platforms, as demonstrated by pictures taken with a variety of camera types during the Apollo missions. At the same time, non-photographic remote sensing sys- tems such as electronic multispectral scanners (MSS) were developed for use on airborne platforms as well as on meteo- rological satellites, Earth resources satellites, and other spaceborne platforms. The 1990s have seen a further expan- sion of the field of remote sensing, with the appearance of new imaging radar satellites, imaging spectrometers, and high-resolution MSS systems, and with the improvement of methods for computer processing of remotely-sensed data. TYPES OF REMOTE SENSING SYSTEMS Several types of remote sensing systems can be differentiated based on the principles employed for measuring electromag- netic radiation. The most common types fall into four broad categories: photographic systems, videographic systems, multispectral scanners, and imaging radar systems. Within these categories, particular instruments are designed to oper- ate in specific portions of the electromagnetic spectrum. The Earth’s atmosphere scatters and absorbs many wavelengths of electromagnetic radiation, limiting the portion of the spec- trum that can be used for remote sensing. Photographic Systems Many types of cameras have been used to acquire photo- graphs of the Earth’s surface from airplanes and from space. Common formats include 35-mm, 70-mm, and 9 ϫ 9-inch film sizes, although specialized cameras that employ other film sizes are also used. Film types include black and white panchromatic, black and white infrared, color, and color infrared, covering the visible and near-infrared portions of the electromagnetic spectrum from approximately 0.4 to 0.9 µ m. (Photography in the ultraviolet range, from 0.3 to 0.4 µ m, is also possible but is rarely done due to atmospheric absorption and the need for quartz lenses.) Once the film has been processed, photographs can be electronically scanned at a variety of resolutions for use in a digital environment. Photographic systems provide relatively high-resolution images, with the nominal scale of a vertical aerial photo- graph being dependent on the focal length of the camera and the flying height of the sensor. 1 Videographic Systems Video cameras can be used to record images in analog form on videotape. Video systems have been designed to operate in the visible, near-infrared, and mid-infrared portions of the electromagnetic spectrum. The advantages of video systems include low cost, near-real-time image availability, and the ability to collect and store many image frames in sequence. The primary disadvantage of video is its low spatial resolu- tion, with approximately 240 lines per image for standard video cameras. 2,3 Multispectral Scanners MSS systems use electronic detectors to measure electro- magnetic radiation in selected bands of the spectrum from approximately 0.3 to 14 m m, including the visible and near-, mid-, and thermal-infrared regions. These individual bands may be fairly wide (greater than 0.2 m m in width) or quite narrow (less than 0.01 m m in width). The designs used for MSS systems fall into two categories. Across-track scanners employ a rotating or oscillating mirror to scan back and forth across the line of flight. Along-track (“push-broom”) scan- ners use a linear array of charge-coupled devices (CCDs) to scan in parallel along the direction of flight. Distinct subcat- egories of MSS systems include thermal scanners, which measure emitted radiation in the thermal infrared portion of the spectrum, and imaging spectrometers, or “hyperspectral scanners,” which generally collect data in over 100 continu- ous, narrow spectral bands, producing a complete reflectance spectrum for every pixel in the image. 1,4 C018_004_r03.indd 1061C018_004_r03.indd 1061 11/18/2005 11:05:35 AM11/18/2005 11:05:35 AM © 2006 by Taylor & Francis Group, LLC 1062 REMOTE SENSING Airborne and satellite MSS systems have become widely used in many environmental science and resource management applications. Examples of different types of MSS system include the following: • The U.S. Landsat satellite series includes two MSS systems. The original Landsat MSS (on board sys- tems launched between 1972 and 1978) includes four spectral bands in the visible and near-infrared portions of the spectrum, with a spatial resolution of 80 m. The Landsat Thematic Mapper (TM) instrument (since 1982) includes six bands in the visible, near-, and mid-infrared regions, with a spatial resolution of 30 m (and a thermal infrared band with a resolution of 120 m). Both instruments operate in an across-track configuration with a swath width of 185 km, and a current orbital repeat cycle of 16 days. • The HRV instrument on the French SPOT satel- lite series can collect data in either a single wide “panchromatic” band, with 10m resolution, or in three narrower bands in the visible (green and red) and near-infrared, with 20 m resolution. The orbital repeat cycle is 26 days, but the sensor’s ability to be rotated (via ground command) up to 27°ᎏ left or right allows more frequent imaging of a given location on the Earth’s surface. The HRV is an along-track scanner, with a swath width of 60 to 80 km depending on the viewing angle. Two identical HRVs are included on each SPOT satellite. Imaging Radar Systems Whereas the previous types of remote sensing systems oper- ate in the visible and infrared portions of the electromagnetic spectrum, imaging radar systems operate in the microwave portion of the spectrum, with wavelengths from approxi- mately 1 cm to 1 m. At these wavelengths, radar is unaf- fected by clouds or haze (shorter wavelength systems are used for meteorological remote sensing). In addition, radar systems are active sensors, transmitting their own radiation rather than passively measuring reflected solar or emitted radiation; thus, they can be operated at any time of day or night. Imaging radar systems are sensitive to the geometric structure and dielectric properties of objects, with the pri- mary determinant of an object’s dielectric properties being its liquid water content. Current satellite radar systems include the European ERS-series and the Canadian Radarsat, which each have a single 5-cm wavelength band, and the Japanese JERS-1 system with a 23-cm band. Several air- borne radar systems have been developed, such as the NASA/ JPL AIRSAR, which operates at multiple wavelengths. 5,6,7 Photographic cameras, video cameras, and multispectral scanners can be operated in a vertical configuration to mini- mize the geometric distortion of the image, or at an oblique angle to provide a side view of the landscape. Imaging radar systems are not operated vertically, but in a side-looking configuration with a broad range of possible look angles. ENVIRONMENTAL APPLICATIONS OF REMOTE SENSING Remote sensing has been used for a wide variety of applica- tions in the environmental sciences. Among the earliest uses of remote sensing was geologic mapping, including the dis- crimination of rock and mineral types, lineament mapping, and identifying landforms and geologic structures. Today, many types of remotely-sensed data are used for geologi- cal applications at a variety of spatial scales, ranging from high-resolution aerial photography, to thermal-scanner images, to lower-resolution Landsat images covering large areas. Agricultural applications of remote sensing are also common. Aerial photography and other remotely-sensed data are widely used as a base for soil mapping, while multispec- tral and thermal images are used for soil moisture mapping. Imaging radar systems, with their sensitivity to moisture- related dielectric surface properties, can also be used to mea- sure soil moisture. Multispectral visible and infrared data are used for crop classification and assessment, including moni- toring the health and productivity of crops, with the goal of predicting yields and identifying areas of crop damage. In forestry, aerial photographs are used to delineate timber stands and to estimate tree heights, stocking densities, crown diameters, and other variables relating to timber volume. Color infrared photography and multispectral imag- ery can be used to map forest types and to identify areas of stress due to pest infestations, air pollution, and other causes. Aerial and satellite imagery can be used to map the effects of wildfires, windthrow, and other phenomena in forested regions. Wildlife habitat can be assessed using remote sens- ing at a variety of scales. High-resolution aerial photography can also be used to assist in wildlife censuses in non-forested areas such as rangeland. Many aquatic and hydrological applications make use of remote sensing. Water pollution can be monitored using aerial photography or MSS systems, and imaging radar can be used to detect oil slicks. Thermal imagery is used to study currents and circulation patterns in lakes and oceans. Both optical and radar data are used to monitor flooding, including flooding beneath a forest canopy in the case of radar. Wetlands delin- eation and characterization can both be assisted by remote sensing. Radar systems are used to measure ocean waves, and both radar and optical images have been used to detect sea and lake ice. Remote sensing is often used to assist in site selection and infrastructure location, urban and regional planning, and civil engineering applications. Aerial photographs are often acquired with a significant overlap between adjacent photos, allowing heights to be measured using the stereoscopic effect. This process is extensively used for topographic map- ping and for creation of geometrically-correct orthorectified C018_004_r03.indd 1062C018_004_r03.indd 1062 11/18/2005 11:05:36 AM11/18/2005 11:05:36 AM © 2006 by Taylor & Francis Group, LLC REMOTE SENSING 1063 photographs to serve as base maps for other applications. Radar interferometry is also being used on an experimental basis for topographic mapping. IMAGE INTERPRETATION AND ANALYSIS Many environmental applications of remote sensing rely solely on visual image interpretation. In many cases, visual analysis is improved by stereo viewing of overlapping pairs of images. Increasingly, however, some degree of digital image processing is used to enhance and analyze remote sensing data. Simple image enhancement techniques include data stretches, arithmetic operations such as ratioing and differenc- ing, statistical transformations such as principal components analysis, and image convolution, filtering, and edge detec- tion. More complex image processing techniques include automated land use/land cover classification of images using spectral signatures representing different land cover types. 8 Most remote sensing applications require the collection of some form of reference data or “ground truth,” which is then related to features or patterns in the imagery. For exam- ple, pixels in a remotely-sensed hyperspectral image might be compared to a series of mineral spectra acquired from ground samples. Ground measurements of soil moisture, crop productivity, or forest leaf-area index (LAI) could be related to observed reflectance in a satellite image using linear regression. Often, ground truth locations are estab- lished using the Global Positioning System (GPS) to facili- tate the relation to a georeferenced image. One significant advantage of digital remotely-sensed imagery, whether collected electronically or as scanned pho- tographs, is the ability to use digital data in a geographic information system (GIS). Once a digital image has been georeferenced, it can be combined with a variety of other types of spatial data. This combination of image and non- image data can be used for a wide range of purposes from simple map updates to complex spatial analysis. 9,10,11,12 Remote sensing is a rapidly changing field, with more than twenty new satellite systems scheduled for launching in the next decade. Major sources of new data will be high- resolution (approximately 1 m) commercial systems and the various sensors comprising the Earth Observing System (EOS). REFERENCES 1. Lillesand, T.M. and R.W. Kiefer, 1994. Remote sensing and image interpretation, John Wiley and Sons, Inc., New York. 2. Mausel, P.W., J.H. Everitt, D.E. Escobar and D.J. King. 1992. Airborne videography: current status and future perspectives. Photogrammetric Engineering and Remote Sensing, vol. 58, no. 8, pp. 1189–1195. 3. Meisner, D.E. 1986. Fundamentals of airborne video remote sensing. Remote Sensing of Environment, vol. 19, no. 1, pp. 63–79. 4. Vane, G. and A. Goetz, 1993. Terrestrial imaging spectrometry: current status, future trends. Remote Sensing of Environment, vol. 44, no. 2/3, pp. 117–126. 5. Waring, R.H., J. Way, E.R. Hunt, Jr., L. Morrissey, K. Jon Ranson, J.F. Weishampel, R. Oren and S.E. Franklin, 1995. Imaging radar for ecosystem studies. Bio-Science, vol. 45, no. 10, pp, 715–723. 6. Way, J. and E.A. Smith, 1991. Synthetic aperture radar systems and their progression to the EOS SAR, IEEE Transactions on Geosciences and Remote Sensing, vol.29, no. 6, pp. 962–985. 7. Elachi, C., 1988. Spaceborne Radar Remote Sensing, IEEE Press, New York. 8. Jensen, J.R., 1986. Introductory Digital Image Processing: A Remote Sensing Perspective, Prentice-Hall, Englewood Cliffs, NJ. 9. Pequet, D.J. and D.F. Marble, Eds. 1990. Introductory Readings in Geographic Information Systems, Taylor and Francis, New York. 10. Star, J. and J. Estes. 1990. Geographic Information Systems, Prentice-Hall, Englewood Cliffs, NJ. 11. K. Hsu, X. Gao, S. Sooroshian and H.V. Gupta, 1997. Rainfall estima- tion from remotely sensed information using artificial neural networks. J. Appl. Meteorl., 36, 1176–1190. 12. Sooroshian, S., S.K. Shu, X. Gao, H. Gupta, B. Imam and D. Braithwaite, 2000. Evaluation of the PERSIANN system satellite-based estimates of tropical rainfall, Bull. Am. Hydrometeorol. Soc., 81, 2035–2046. JONATHAN CHIPMAN University of Wisconsin C018_004_r03.indd 1063C018_004_r03.indd 1063 11/18/2005 11:05:36 AM11/18/2005 11:05:36 AM © 2006 by Taylor & Francis Group, LLC . spectrometers, and high-resolution MSS systems, and with the improvement of methods for computer processing of remotely-sensed data. TYPES OF REMOTE SENSING SYSTEMS Several types of remote sensing. 1061 REMOTE SENSING Remote sensing is the act of acquiring information about an object from a distance. Environmental applications of remote sensing typically involve the collection of photographic. in a side-looking configuration with a broad range of possible look angles. ENVIRONMENTAL APPLICATIONS OF REMOTE SENSING Remote sensing has been used for a wide variety of applica- tions