|
| |
remote sensing |
|
| |
|
|
| |
The practice of collecting data by observing at a distance; the term is most often applied to observation of the Earth\'s surface from above, from either an aircraft or a satellite. The history of remote sensing extends back to the early days of photography, but the field received a major impetus during both World Wars, when very extensive use was made of aerial photography for reconnaissance. Photographs were taken from aircraft, often flying at high altitude to avoid enemy fire, then developed and interpreted on the ground. The science of air photo interpretation developed as an effort to systematize the detection of features from high-altitude aerial photography. Remote sensing received another boost during the Cold War, as instruments were developed to obtain high-resolution images from satellites flying above the Earth\'s atmosphere. Much of the imagery collected by the US using these systems in the 1960s has now become available to researchers.
Modern remote sensing uses digital instruments attached to satellites or aircraft. Passive systems measure the radiation received by the satellite in various parts of the electromagnetic spectrum, while active systems generate radiation of microwave or radar wavelengths and measure the proportion reflected from below. The radiation detected in a small area known as the instantaneous field of view (IFOV) is integrated, and recorded. A complete image is assembled as a two-dimensional array of pixels, and the spatial resolution of the image is determined by the linear dimensions of each pixel on the Earth\'s surface. Some early satellite systems recorded images on film, which was then ejected from the satellite and caught by an aircraft; but these systems were largely abandoned in favour of radio transmission of images in digital form to a ground receiving station. Early systems required several ground receiving stations, in order to ensure that at least one was in direct view of the satellite; modern systems either store images until the receiving station is in view, or relay the signal via a second satellite.
Applications of remote sensing can be divided into two broad categories, depending on the eventual use of the data. Some systems are designed to provide data that can be treated as measurements of some significant variable, and analysed directly, or used as input to simulation models; the so-called \'Ozone Hole\' over Antarctica was detected by such a system. Other systems are used primarily for mapping, in which case the image is used to identify and locate various types of features on the Earth\'s surface, such as growing crops, roads, urban development, or sea ice. In these latter cases the relationship between radiation and feature may be complex, and largely empirical. Remote sensing systems are now widely used to observe and forecast weather events; measure the precise elevation of the land and ocean surfaces; identify crops and forecast yields; evaluate timber resources; and a host of other practical applications. Low-altitude imagery from aircraft platforms is used to provide a source of higher accuracy to validate satellite-derived estimates; and there is increasing use of miniature remote-controlled aircraft for both civilian and military applications.
Of the hundreds of remote sensing systems now in Earth orbit, only a few are of significance for mapping applications. Much use has been made of the US Landsat series of satellites, which produce images in several visible and near-infrared areas of the spectrum at spatial resolutions down to 30 m. The French SPOT instruments have spatial resolutions down to 10 m. Several companies have been authorized to launch instruments with 1m resolution, beginning in 1998; and it seems likely that the products of high-resolution military satellites will also become increasingly available.
Over the past three decades much effort has gone into finding applications of remote sensing in social and economic domains, though the vast majority of applications address issues in the physical sciences and in natural resource management. A ground resolution of 30 m is sufficient to detect clearing and settlement in the Amazon basin, for example, or urban growth around Mexico City. But many other variables of interest to human geographers are simply invisible from above the surface of the Earth. Imagery from the new 1 m satellites may have potential for human geography, but has not yet been evaluated. It seems, however, that such imagery will be of great value for applications in urban infrastructure management, and for construction of geographic databases describing urban form.
Although it is not possible to count people from satellite imagery, good results have been obtained in studies that have used imagery to estimate small-area demographic statistics, and particularly housing statistics, by disaggregating large-area counts. For example, Langford, Maguire and Unwin (1991) demonstrated the use of these techniques for Leicestershire; Deichmann (1996) has used them to build a world population map at high spatial resolution; and Clarke, Hoppen and Gaydos (1997) have used them to calibrate spatial models of urban growth in the San Francisco Bay area.
The use of remote sensing for social and economic research raises a host of technical and social issues. The visible part of the spectrum, which is most useful for mapping applications, is obstructed by cloud, and certain areas of the Earth\'s surface, particularly in the tropics, are almost never totally cloud-free. Certain types of features and land-cover classes are easily differentiated, but others are not; remote sensing thus inevitably favours things that can be differentiated from space over things that cannot. Issues of surveillance and invasion of privacy arise when satellites fly over foreign countries, or when high spatial resolutions make it possible to detect and possibly to identify vehicles. While not normally associated with the term, modern technologies that make it possible to identify vehicle licence numbers from imaging devices at the roadside are also a form of remote sensing, and constitute collection of information about individuals without their informed consent. (MG)
References Clarke, K.C., Hoppen, S. and Gaydos, L. 1997: A self-modifying cellular automaton model of historical urbanization in the San Francisco Bay area. Environment and Planning B: Planning and Design 24: 247-61. Deichmann, U. 1996: A review of spatial population database design and modeling. Technical Report 96-3. Santa Barbara, CA: National Center for Geographic Information and Analysis. Langford, M., Maguire, D. and Unwin, D. 1991: The areal interpolation problem: estimating population using remote sensing in a GIS framework. In I. Masser and M. Blakemore, eds, Handling geographic information. Harlow: Longman, 55-77.
Suggested Reading Lillesand, T.M. and Kiefer, R.W. 1994: Remote sensing and image interpretation. New York: Wiley. Ryerson, R.A., ed., 1996: Manual of remote sensing, 3rd edn. Bethesda, Maryland: American Society for Photogrammetry and Remote Sensing. |
|
| |
|
|
| |
Bookmark this page:
|
|
| |
|
|
| |
<< former term |
|
next term >> |
|
|
|
|
|
| |
|
|
|
