Environmental Engineering Reference
In-Depth Information
use the term ' habitat ' but sometimes also ' landscape ' ;
with a variable spatial scale depending on the organ-
isms under study. To avoid confusion, in this chapter,
we use
habitat
to explicitly address the 'value' of a
certain area for a (group of) species, and
landscape
to
describe a spatial matrix, which we consider as some-
thing more neutral and/or general. Moreover, in
section 5.4, we will differentiate between intact,
variegated, fragmented and relictual landscapes as
defi ned by the extent of habitat destruction (McIntyre
& Hobbs 1999), thus combining human- and organism-
oriented views on the condition or status of a given
landscape.
patches constitutes a
pattern
. Topography (e.g. eleva-
tion and slope), natural disturbances (e.g. due to fi re,
fl ooding or storms) and, increasingly, human activities
(e.g. farming, roads and buildings) are the principal
causes of landscape pattern.
There is often a remarkable similarity in patterns,
both over larger areas and over larger timescales. If
only one major environmental factor determines the
pattern, this may lead to the development of
zonation
,
perpendicular to direction or orientation of this factor.
Typical examples of factors that lead to vegetation
zonation include altitudinal gradients in mountainous
areas and hydrological gradients in river valleys (Figure
5.1), but also gradients in landuse intensity in the past.
Even in many cases where more than one factor affects
the distribution of patterns, relatively simple zonations
can occur, because many factors coincide to a high
degree. For instance, land use was (and in many places
still is) highly correlated to natural productivity and
ease of cultivation. In mountainous areas, most vil-
lages are situated in the intermontane valleys where
both climate and soil fertility are most favourable for
agriculture. In lowlands on the other hand, human
occupation started along the edges of river valleys in
close proximity to fertile fl oodplains, in places where
the risk of fl ooding was considered slight. In both cases,
whatever animal dung was available was used to
enhance the productivity of already productive sites
whereas infertile sites such as high mountain areas or
dry sandy sites were not fertilized at all. Human activi-
ties thus extended pre-existing productivity gradients,
thereby enabling increased niche differentiation and
biodiversity.
5.2.2
Scale
Scale is a central concept in landscape ecology, mainly
because temporal and spatial scales are highly corre-
lated. Small-scale processes tend to be more dynamic
and less predictable than large-scale processes and are
associated with fi ne-scale spatial structures. Different
organisms operate at different scales and may perceive
entirely different environments at the same spot. For
example, experimental studies by Wiens (1976)
showed that fi ne-scale spatial structures are highly rel-
evant for beetle movement whereas migrating geese
follow large-scale structures like coastlines, canals and
similar (Green
et al
. 2002 ). At fi rst sight this may seem
trivial, but in the context of restoration it implies that
the scales at which target organisms operate and at
which relevant processes occur should be identifi ed
explicitly. For example, restoring an ecosystem to
benefi t or rescue a marsh bird community, might be
achieved by fl ooding 5 ha of degraded wetland, whereas
it might be necessary to restructure a whole catchment
and reactivate a groundwater system to restore an eco-
system with specifi c groundwater - fed plant communi-
ties in the same site.
5.2.4
Borders and corridors
Adjacent patches are separated by borders
.
On a map
such borders are infi nitively thin and this suggests that
patches can be delineated sharply, but in practice there
are always transition zones ( ' ecotones ' or ' boundaries ' ;
e.g. Forman 1997). These may differ in width and often
contain specifi c species assemblages, highly deviating
from communities found in the interior of a patch.
Boundaries are not absolute barriers but may be rela-
tively sharp or fuzzy, and possess higher or lower resist-
ance to fl ows between landscape elements (section
5.3), thereby determining and affecting connectivity in
a landscape.
5.2.3
Patches and patterns
All landscapes are heterogeneous. Parts of a given
landscape that are considered to be uniform are called
patches
or
ecotopes
. As noted, the scale of a patch
depends entirely on the perspective of the observer. A
part of a landscape that is a patch for an elephant can
be a highly diverse region for a butterfl y. A group of
