each defined by application of different criteria.
But not all homogeneous areas are of equal significance. We are seeking meaningful types by
the identification of correspondence between
spatial patterns. That is, the manner in which
the variations within one ecosystem component
correspond to the variations within another component, particularly in ways that affect process.
For example, all steep slopes have shallow soils
and are susceptible to erosion.
1.3
The Process of Defining
The fundamental question facing all ecological
mappers is: How are the boundaries of systems to
be determined? The first, large-scale divisions of
the Earth’s surface are, quite obviously, the land
masses and the water areas, where ecological
processes take place in quite a different manner.
But how do we determine the distribution of
ecosystems within each?
Different methods have been used to identify
units where ecosystem components (i.e., climate,
vegetation, soil) are integrated in a similar way,
thereby classifying land as ecosystems. The
problem is boundaries on different component
maps rarely correspond to each other. Further,
overlay mapping techniques and cluster analysis
address neither the causes that generate different
ecosystem units nor why those units should be
distinguished.
Establishing a hierarchy of ecosystem
boundaries should be based on understanding of
the formative processes that operate to differentiate the landscape into ecosystems at various
scales (cf. Bailey 1985, 1987; Klijn and Udo de
Haes 1994; Godron 1994). The units derived
from such an approach are termed “genetic” in
that they follow the cycle of landscape evolution
(Fig. 1.5) and are predicated upon an understanding of the causal processes that control the pattern of ecosystems. Understanding spatial
relationships between causal mechanisms and
resultant patterns is the key to understanding
how ecosystems respond to management.
The genetic approach is based on the idea of
defining ecosystems using a deductive “top
down” approach. This is because subsystems
can be understood only within the context of
the whole; a mapping of ecosystems begins
with the largest units and successively subdivides
them.
The notion that process knowledge should
define landscapes has a long history in geography
and geology. William Morris Davis (1899)
argued for a genetic classification of landforms,
and both Herbertson’s (1905) natural regions of
the world and Fenneman’s (1928) physiographic
divisions of the United States are based on
Fig. 1.5 Schematic diagram illustrating the cycle of
landscape evolution beginning with (a) climate, (b) runoff
and landforms, and leading to (c) soil, vegetation, and
habitat formation. From Marsh (2005), p. 58. Used with
permission
4
1 Introduction
But not all homogeneous areas are of equal significance. We are seeking meaningful types by
the identification of correspondence between
spatial patterns. That is, the manner in which
the variations within one ecosystem component
correspond to the variations within another component, particularly in ways that affect process.
For example, all steep slopes have shallow soils
and are susceptible to erosion.
1.3
The Process of Defining
The fundamental question facing all ecological
mappers is: How are the boundaries of systems to
be determined? The first, large-scale divisions of
the Earth’s surface are, quite obviously, the land
masses and the water areas, where ecological
processes take place in quite a different manner.
But how do we determine the distribution of
ecosystems within each?
Different methods have been used to identify
units where ecosystem components (i.e., climate,
vegetation, soil) are integrated in a similar way,
thereby classifying land as ecosystems. The
problem is boundaries on different component
maps rarely correspond to each other. Further,
overlay mapping techniques and cluster analysis
address neither the causes that generate different
ecosystem units nor why those units should be
distinguished.
Establishing a hierarchy of ecosystem
boundaries should be based on understanding of
the formative processes that operate to differentiate the landscape into ecosystems at various
scales (cf. Bailey 1985, 1987; Klijn and Udo de
Haes 1994; Godron 1994). The units derived
from such an approach are termed “genetic” in
that they follow the cycle of landscape evolution
(Fig. 1.5) and are predicated upon an understanding of the causal processes that control the pattern of ecosystems. Understanding spatial
relationships between causal mechanisms and
resultant patterns is the key to understanding
how ecosystems respond to management.
The genetic approach is based on the idea of
defining ecosystems using a deductive “top
down” approach. This is because subsystems
can be understood only within the context of
the whole; a mapping of ecosystems begins
with the largest units and successively subdivides
them.
The notion that process knowledge should
define landscapes has a long history in geography
and geology. William Morris Davis (1899)
argued for a genetic classification of landforms,
and both Herbertson’s (1905) natural regions of
the world and Fenneman’s (1928) physiographic
divisions of the United States are based on
Fig. 1.5 Schematic diagram illustrating the cycle of
landscape evolution beginning with (a) climate, (b) runoff
and landforms, and leading to (c) soil, vegetation, and
habitat formation. From Marsh (2005), p. 58. Used with
permission
4
1 Introduction