Remote sensing is the process of collecting information about the environment without actually coming into direct contact with it. It typically involves the use of electronic sensors fitted to either a satellite or an aircraft. This enables the collection of data over large areas. It is often not possible to get this type of information at the continental scale using tradition ground-based techniques. Remote sensing also enables frequent repetition of data collection over the surface of the Earth. The remotely sensed data can be used and manipulated in geographical information systems. In this regard remote sensing has excellent potential to monitor environmental indicators on a regional levels and to identify trends over time. The production and interpretation of satellite images for the daily weather is a familiar example of the use of remote sensing.
Space-based remote sensing became operational with Landsat's launch in 1972. Since then, satellite remote sensing data have been increasingly available on a worldwide basis with increasing spatial and spectral resolution. The enormous data storage problems accompanying the technology's evolution have led to the development of the Geographic Information System (GIS) which accept large volumes of data, derived from a variety of sources, and efficiently store, manipulate, analyse and display these data according to user defined specifications, such as facilitating models.
Satellite remote sensing's main characteristics include the large area that can be covered by each image and the high frequency (continuous) with which data can be gathered over the same area. Features remote sensing can distinguish between include large forests and cities, to small streams and trees. Equally, remote sensing can monitor the environment at regional and continental scales, though the detail becomes necessarily less accurate with increased spatial resolution. With several series of satellites in orbit one can select or combine imagery and imagery techniques from a wide range of spatial (ground distance), spectral and temporal resolutions to maximize their potential uses. These combined functions have created a highly cost-effective technology whose information is particularly used in assessing, planning and managing natural resources.
Fundamental to the collection of remotely sensed data is that different physical objects ( e.g. vegetation, bare ground and water) reflect varying amounts of electromagnetic radiation when subjected to different wavelengths. Almost all wavebands can be used for remote sensing, ranging from radar to light to microwave. However, the most frequently used bands for environmental monitoring are the visible and infra-red wavebands. Sensors include photographic cameras, electronic multi-spectral scanners operating in the visible and infrared regions of the electromagnetic spectrum, and imaging radar systems. The data is enhanced in terms of accuracy and classification if digital image processing techniques are used.
The advantages of using remotely sensed information are complemented by a wide range of potential environmental indicators that can be measured and monitored. Some examples are: dryland salinity, soil erosion, presence of acid soils, sea-surface temperature, extent of seagrass beds, chlorophyll production in shallow coastal waters, sediment deposition along rivers and in estuaries, and bathymetric profiling. However, because remotely sensed data are not usually direct measurements, but rather correlations between the spectral signature and the specific parameter being observed, the information interpreted needs to be validated by ground-truthing.
The timely availability of accurate information now allows warning for the onset of drought, storms, locust plagues, forest diseases and the like that can be foreseen, thus allowing time to take necessary precautionary measures. It is argued that many natural disasters would have been less severe, therefore, if operational remote sensing and GIS had already been in place. As a result of the beneficial capabilities of remote sensing, the UN in 1993 recommended that the strengthening of indigenous remote sensing and GIS capacities in developing countries should be one of the prime objectives of the national authority responsible for providing the country with adequate tools for continuous monitoring and sustainable management of tropical forests.
This strategy features in the framework of Agenda 21 as formulated at UNCED (Rio de Janeiro, 1992), now coordinated by the United Nations Commission on Sustainable Development and implemented through national and local authorities.
In order to understand the Earth as a system, Agenda 21 recommends developing Earth observation systems from space which will provide integrated, continuous and long-term measurements of the interactions of the atmosphere, hydrosphere and lithosphere, and developing a distribution system for data which will facilitate the utilization of data obtained through such observation.
Programmes using the technology are already in place and include The Famine Early Warning System (FEWS) and the Africa Real Time Environmental Monitoring Using Imaging Satellite (ARTEMIS) of the UN Food and Agriculture Organization (FAO).
The delineation of biogeographic units of biodiversity through the use of remote-sensing imagery and analyses can better resolve the spatial distribution and seasonality of natural communities at continental scales. Capabilities exist to access and analyse data on the current status and distribution of remaining natural habitats throughout the world. For example, measuring the extent and level of degradation of native forests for different ecoregions is greatly facilitated through remote-sensing technologies. Mapping forest cover in perpetually cloudy regions is another area in which NASA for example may have unique capabilities. Another important issue is how effectively remote-sensing data can be used to assess the degradation of natural communities in deserts, grasslands, shrublands, and other non-forest terrestrial habitats. Moreover, biodiversity maps and analytical tools for freshwater and marine ecosystems are poorly developed relative to those for terrestrial systems. There is a need for updating and improving existing global or regional databases for remaining habitat cover and classification.