128
Chapter 14
use at the landscape scale
1 . LACOAST utilized classified remotely sensed
data (Landsat Multispectral Scanner, Landsat Thematic Mapper and SPOTHRV) to construct a land cover time series from 1976 to 1995 for a 10 km
buffer zone around the coastline of the European Union territories.
At continental and global scales, remote sensing has been used
successfully to assess changes in land cover, due, for example to natural and
anthropogenic disturbance (Hall et al. 1991; Belward et al. 1994; BourgeauChavez, Harrell et al. 1997). However, at scales appropriate to the needs of
most national, regional and local applications, the scientific literature is
almost devoid of practical examples of the application of remote sensing to
detect and monitor change. The reasons for this are not hard to discover.
Change detection, measurement and monitoring pose theoretical and
practical challenges that have yet to be fully met in the research context, far
less in any operational setting. Given the importance of changes in land use
as a driver of environmental quality, it is vital that these underlying research
issues are addressed, and that the enormous potential of remote sensing is
fully realized.
These research issues are numerous and highly inter-dependent;
moreover, solutions which are optimal for one application may be wholly
inappropriate for others. This paper addresses some of these issues, in the
context of the future use of remote sensing as an operational tool for
mapping land cover and for the detection and monitoring of change.
2.
CHANGE DETECTION AND ANALYSIS
Most remote sensing systems rely on polar-orbiting satellites or on
aircraft-mounted sensors. In consequence, data are typically captured as
scenes that are instantaneous in comparison to the rates of change typically
encountered in terrestrial landscapes. (Geo-stationary satellites have the
capacity for genuinely continuous monitoring, but only at very coarse spatial
resolutions). Methods for detecting or measuring change from remote
sensing therefore invariably depend on comparisons between data sets
acquired at intervals of time. This raises important issues regarding the
accuracy of the data sets on which these comparisons are based. Various
sources of potential error are identified below: they include spatial and
temporal effects and the extent to which a given land cover class may be
recognized unambiguously from its radiometric properties, perhaps under
differing conditions of solar irradiation or atmospheric turbidity. The key
point to bear in mind is that there is a limit to the capacity of any remote
1 http://www.ais.sai.jrc.it/environment/lacoast/index.html
Chapter 14
use at the landscape scale
1 . LACOAST utilized classified remotely sensed
data (Landsat Multispectral Scanner, Landsat Thematic Mapper and SPOTHRV) to construct a land cover time series from 1976 to 1995 for a 10 km
buffer zone around the coastline of the European Union territories.
At continental and global scales, remote sensing has been used
successfully to assess changes in land cover, due, for example to natural and
anthropogenic disturbance (Hall et al. 1991; Belward et al. 1994; BourgeauChavez, Harrell et al. 1997). However, at scales appropriate to the needs of
most national, regional and local applications, the scientific literature is
almost devoid of practical examples of the application of remote sensing to
detect and monitor change. The reasons for this are not hard to discover.
Change detection, measurement and monitoring pose theoretical and
practical challenges that have yet to be fully met in the research context, far
less in any operational setting. Given the importance of changes in land use
as a driver of environmental quality, it is vital that these underlying research
issues are addressed, and that the enormous potential of remote sensing is
fully realized.
These research issues are numerous and highly inter-dependent;
moreover, solutions which are optimal for one application may be wholly
inappropriate for others. This paper addresses some of these issues, in the
context of the future use of remote sensing as an operational tool for
mapping land cover and for the detection and monitoring of change.
2.
CHANGE DETECTION AND ANALYSIS
Most remote sensing systems rely on polar-orbiting satellites or on
aircraft-mounted sensors. In consequence, data are typically captured as
scenes that are instantaneous in comparison to the rates of change typically
encountered in terrestrial landscapes. (Geo-stationary satellites have the
capacity for genuinely continuous monitoring, but only at very coarse spatial
resolutions). Methods for detecting or measuring change from remote
sensing therefore invariably depend on comparisons between data sets
acquired at intervals of time. This raises important issues regarding the
accuracy of the data sets on which these comparisons are based. Various
sources of potential error are identified below: they include spatial and
temporal effects and the extent to which a given land cover class may be
recognized unambiguously from its radiometric properties, perhaps under
differing conditions of solar irradiation or atmospheric turbidity. The key
point to bear in mind is that there is a limit to the capacity of any remote
1 http://www.ais.sai.jrc.it/environment/lacoast/index.html