Geography Reference
In-Depth Information
of the wider countryside in Great Britain (see
below).
The production of land cover maps from
satellite images illustrates something of the
potential that remote sensing has for those
concerned with monitoring the environment. It
must be remembered, however, that there are
many more types of sensor than Landsat or SPOT.
The case study shown in Figure 40.2, used data
from the AVHRR sensor, which is different in
character to Landsat and SPOT.
The AVHRR sensor is carried by NOAA's
Polar Orbiting Environmental Satellites (POES)
and images are available for all areas on a daily
basis. By contrast, although Landsat and SPOT
are also in polar orbit, they pass over the same
place on the ground less frequently. For example,
Landsat 4 and 5 each have a repeat cycle of
sixteen days, but because their orbits are out of
phase, data for the same area can be acquired
once every eight days. It is likely that this
freqency of acquisition will be maintained with
the recent launch of Landsat 7. SPOT also has a
repeat cycle of twenty-six days, but because the
sensor is pointable, data for a given area may be
available more frequently. Thus in terms of the
frequency with which images of an area can be
acquired, the AVHRR sensor is said to have a
higher temporal resolution than either SPOT or
Landsat. However, AVHRR data have much
coarser spatial and spectral resolution than these
other systems; each pixel on the AVHRR image
used for Figure 40.2, for example, is 8×8 km.
Moreover, the AVHRR sensor records reflected
radiation in only two parts of the
electromagnetic spectrum, compared with seven
of Landsat, or the three or four of SPOT.
When looking at the way in which remotely
sensed images can be used to monitor and analyse
environmental change, it is important to
remember that different types of sensor will suit
different types of application. Moreover, as the case
study from Madagascar illustrates, it is important
that we find ways of integrating these images with
other sources of archive data so that long-term
change can be studied. As we look to the future, it
is evident that an increasingly wide range of
satellite and airborne sensors will be available, and
the challenge will be to use these new sources of
data to measure a range of environmental
parameters. A summary of some of the types of
data that will be produced by the next generation
of Earth observation satellites that are part of
NASA's EOS programme is given in Table 40.1.
Information about future European and Japanese
missions can be found at the end of the chapter.
Digital maps
Although remote-sensing systems are important
sources of information about the environment,
they are clearly not the only ones. The problems
of monitoring and analysing environmental
change often require information from a wide
range of sources if a full picture of environmental
change is to be achieved. We might need, for
example, information about soils, elevation,
administrative boundaries, population and so on.
The need for integration of many data sources
in a GIS is well illustrated by the case study for
Madagascar (Figure 40.1). Such studies illustrate
the importance of a second major data source for
our GIS, namely 'digital maps', that is maps that
can be read and manipulated in a computer. In
the Madagascar study, Green and Soussman (1990)
used digital maps of population and altitude to
examine some of the factors affecting the rate of
loss of rainforest.
Although maps can be scanned and converted
to raster images, like remotely sensed data, the
information held on digital maps can also be held
in
vector
format. Using this format, each object
on the map can be represented as a point, a line
or a polygon (or area feature). The information
about the spatial location of these features is
stored as a set of coordinates, and their
characteristics are recorded as a set of attributes.
Raster-scanned maps are little more than
pictures, used more often as a backdrop for the
display of other data rather than for analysis
per
se
. Vector data, by contrast, can be used in much
more flexible ways, because the lines, points and
areas can be made to correspond to real objects
on the ground.
