28.6 Methodological Implications
meso-, and macroscale processes. The issue has been
of interest to ecologists, who are concerned with the
transfer of information and analyses across scales
(e.g., see Meentemeyer and Box, 1987; Turner et aI.,
1989; Lima and Zollner, 1996), in remote sensing
where the spatial resolution employed for analysis of
surface conditions affects interpretation (Malingreau
and Belward, 1992), and in linking information from
different sources gathered at different levels of spatial resolution (Brown et al., 1996).
Ecologists recognize that patterns of environmental variability are generated by complex dynamic interactions among and between species. Analysis of
environmental processes is made more complex
when the anthropogenic contributions to humanenvironment interactions are considered. Such interactions encompass a wide range of socioeconomic
and biophysical variables, each with its own spatial
and temporal characteristics. These have to be assessed in an integrated analytical structure (Izac and
Swift, 1994). Creativity is needed to comprehend the
processes acting both within and between levels, because the processes and states of a particular level
are the outcome of what is happening at that level
plus the impact of influences emanating from other
levels. At the microlevel, conditions are influenced
by what is happening at the microlevel, together with
impulses from meso- and macrolevels. Interactions
between scales, from the village to the nation and the
world, must therefore be recognized as determining
patterns of resource use. In this hierarchical structure,
each level is linked, and each level is independent to
some extent (Figure 28.2).
Furthermore, a change in the scale of analysis may
yield new insight as to causal processes: "Although
moving to a coarser scale may result in a loss of detail, it may also result in the appearance of emergent
properties due to synergisms at a higher level of system integration. These interactions were not seen at
fmer scales and were perhaps not acting if finer scale
meant a lower level of interaction" (Meentemeyer and
Box, 1987, pp. 20--21). Interpretation of such synergisms and interactions requires care to distinguish between causal interactions and mere statistical association or coincidence. Again, reference to a conceptual
model is critical to valid interpretation of phenomena.
This discussion also illustrates the fact that there
is no single appropriate scale at which to discuss
human-environmental systems (Levin, 1992). Ultimately, a reductionist approach would yield discussion of microsettings where individual people,
soil particles, and plants may be the focus. In the
context of human interactions with ecosystems, this
level of detail is superfluous and unmanageable. An
appropriate scale requires generalizations based on
429
these micro level patterns and processes, as well as
understanding of how they respond to and affect
broader-scale processes.
The issue remains, however, that the management
of natural resources implies interaction between societal and biophysical states and processes. This interaction takes place between and among different
scales. In both the ecological and the human sciences,
local to global interactions are recognized as contributing to patterns of resource use: "How do we
know which scale and spatial pattern we should use?
And how do we know when the results are meaningful and valid given the scale and resolution of the
data?" (Lam and Quattrochi, 1994, pp. 89-90).
The issue has been clearly stated in the context
of ecology by Levin (1992, p. 1943):
the problem of pattern and scale is the central problem
in ecology. Applied challenges, such as the prediction of
the ecological causes and consequences of global climate
change, require the interfacing of phenomena that occur
on very different scales of space, time, and ecological
organization. Furthermore, there is no single natural scale
at which ecological phenomena should be studied; systems generally show characteristic variability on a range
of spatial, temporal and organizational scales .... The
key to prediction and understanding lies in the elucidation of mechanisms underlying observed patterns.
An essential component of analysis is the explication of a theoretical--conceptual framework that
facilitates clarity of thought regarding the issue under consideration. Such a framework is needed to
ensure an appropriate definition of the analytical
model to be used. This model in tum will guide the
selection of variables, of the spatial scale and unit
of data collection. The conceptual framework will
also inform approaches to aggregation, analysis,
and interpretation and the generalization of results.
There are a number of approaches to this question. They range from case studies and strategies
for linking large-scale (small-area) studies to smallscale (large-area)3 information (McGwire et aI.,
3There is some confusion in the published literature over
the terminology to describe different scales. A number
of authors have sought to clarify the issue (Meentemeyer
and Box, 1987; Lam and Quattrochi, 1994). In a cartographic sense, small scale represents large areas; for example, a map of the world may be at a scale of
1 : 1,000,000, and large scale refers to small areas, for
example, a map of a city may be at the scale of 1 : 5000.
In this chapter we will follow this convention. Largescale spatial units might be used to examine local case
studies of village communities, field-level soil conditions, subcatchments of river basins, and, in vegetation
terms, the patch. Comparable small-scale units might include national social characteristics, generalized soil
maps, river basins, and continental vegetation zones.
meso-, and macroscale processes. The issue has been
of interest to ecologists, who are concerned with the
transfer of information and analyses across scales
(e.g., see Meentemeyer and Box, 1987; Turner et aI.,
1989; Lima and Zollner, 1996), in remote sensing
where the spatial resolution employed for analysis of
surface conditions affects interpretation (Malingreau
and Belward, 1992), and in linking information from
different sources gathered at different levels of spatial resolution (Brown et al., 1996).
Ecologists recognize that patterns of environmental variability are generated by complex dynamic interactions among and between species. Analysis of
environmental processes is made more complex
when the anthropogenic contributions to humanenvironment interactions are considered. Such interactions encompass a wide range of socioeconomic
and biophysical variables, each with its own spatial
and temporal characteristics. These have to be assessed in an integrated analytical structure (Izac and
Swift, 1994). Creativity is needed to comprehend the
processes acting both within and between levels, because the processes and states of a particular level
are the outcome of what is happening at that level
plus the impact of influences emanating from other
levels. At the microlevel, conditions are influenced
by what is happening at the microlevel, together with
impulses from meso- and macrolevels. Interactions
between scales, from the village to the nation and the
world, must therefore be recognized as determining
patterns of resource use. In this hierarchical structure,
each level is linked, and each level is independent to
some extent (Figure 28.2).
Furthermore, a change in the scale of analysis may
yield new insight as to causal processes: "Although
moving to a coarser scale may result in a loss of detail, it may also result in the appearance of emergent
properties due to synergisms at a higher level of system integration. These interactions were not seen at
fmer scales and were perhaps not acting if finer scale
meant a lower level of interaction" (Meentemeyer and
Box, 1987, pp. 20--21). Interpretation of such synergisms and interactions requires care to distinguish between causal interactions and mere statistical association or coincidence. Again, reference to a conceptual
model is critical to valid interpretation of phenomena.
This discussion also illustrates the fact that there
is no single appropriate scale at which to discuss
human-environmental systems (Levin, 1992). Ultimately, a reductionist approach would yield discussion of microsettings where individual people,
soil particles, and plants may be the focus. In the
context of human interactions with ecosystems, this
level of detail is superfluous and unmanageable. An
appropriate scale requires generalizations based on
429
these micro level patterns and processes, as well as
understanding of how they respond to and affect
broader-scale processes.
The issue remains, however, that the management
of natural resources implies interaction between societal and biophysical states and processes. This interaction takes place between and among different
scales. In both the ecological and the human sciences,
local to global interactions are recognized as contributing to patterns of resource use: "How do we
know which scale and spatial pattern we should use?
And how do we know when the results are meaningful and valid given the scale and resolution of the
data?" (Lam and Quattrochi, 1994, pp. 89-90).
The issue has been clearly stated in the context
of ecology by Levin (1992, p. 1943):
the problem of pattern and scale is the central problem
in ecology. Applied challenges, such as the prediction of
the ecological causes and consequences of global climate
change, require the interfacing of phenomena that occur
on very different scales of space, time, and ecological
organization. Furthermore, there is no single natural scale
at which ecological phenomena should be studied; systems generally show characteristic variability on a range
of spatial, temporal and organizational scales .... The
key to prediction and understanding lies in the elucidation of mechanisms underlying observed patterns.
An essential component of analysis is the explication of a theoretical--conceptual framework that
facilitates clarity of thought regarding the issue under consideration. Such a framework is needed to
ensure an appropriate definition of the analytical
model to be used. This model in tum will guide the
selection of variables, of the spatial scale and unit
of data collection. The conceptual framework will
also inform approaches to aggregation, analysis,
and interpretation and the generalization of results.
There are a number of approaches to this question. They range from case studies and strategies
for linking large-scale (small-area) studies to smallscale (large-area)3 information (McGwire et aI.,
3There is some confusion in the published literature over
the terminology to describe different scales. A number
of authors have sought to clarify the issue (Meentemeyer
and Box, 1987; Lam and Quattrochi, 1994). In a cartographic sense, small scale represents large areas; for example, a map of the world may be at a scale of
1 : 1,000,000, and large scale refers to small areas, for
example, a map of a city may be at the scale of 1 : 5000.
In this chapter we will follow this convention. Largescale spatial units might be used to examine local case
studies of village communities, field-level soil conditions, subcatchments of river basins, and, in vegetation
terms, the patch. Comparable small-scale units might include national social characteristics, generalized soil
maps, river basins, and continental vegetation zones.