|
LAND USE AS AN ECOLOGICAL VARIABLE
Steward T. A. Pickett;
Institute of Environmental Studies,
P.O. Box AB, Route 44A
Millbrook, NY 12545-0129;
STAPickett@aol.com
NOTE: The following document is posted to provide a look at
new ideas and developing components of the LUHNA project. No reproduction
or other use of the materials posted here should be undertaken without
consent of, and discussion with, the author. Please note e-mail
address, above.
Abstract
When are land use and land cover useful ecological or
environmental variables? In situations where the linkage between land cover
and an ecological response is direct, or when a sound model links land cover
or land use to ecological responses, they can serve as ecological variables.
However, without such a mechanistic connection they must be considered to be
promising exploratory factors.
Land Use and Land Cover
Land use is a description of how a parcel of land is employed by people, or
the degree to which that land reflects natural versus human-generated cover.
Land cover is a closely related concept, which more readily describes the
natural and human surfaces or structures that blanket the Earth. Land cover
is the more general variable, and land use is more specific. Land use
describes how a parcel of surface is employed -- or not -- by people, and
therefore requires more information than land cover. The two variables are
not redundant, because a given land cover type may be used in more than one
way.
Land use and land cover are often characterized using remote sensing with
subsequent confirmation by surveys on the ground or in private or public
records. Because of the reliance on remote sensing, the land use/land cover
categories that can be resolved are often rather crude. For instance,
residential land cover can sometimes be broken down in to high and low
densities, but many suburban developments may look, from high altitudes or
using a few satellite bands, like a closed canopy forest. Hence land cover,
and especially land use are not always easy parameters to measure. The
covers of the Earth's surface are, in fact, complex, three dimensional
bodies, rather than a simple layer that is draped over the geomorphology.
Such three dimensionality, is however, what makes land use and land cover so
important as ecological variables. Different cover types, due to their soil
and substrate, their vegetation depth, height and layering, their
hospitability, or hostility to the movement, shelter, and nourishment of
different animals, their permeability to flows of water, nutrients, wind, and
seeds, and their arrangement over space, all can influence biological
diversity and the provision of ecosystem services.
The three-dimensionality of land cover and land uses emphasizes that the
spatial arrangement of land covers and land uses is significant to
understanding their implications. Landscape comparisons may rest on knowing
simply the mixture of land use and land cover types. What types are present
in a landscape, and the relative proportions of the types may yield valuable
information about the structure and function of a landscape. Comparison of
contrasting landscapes using indices of land cover and land use richness, and
the evenness of land cover types may be especially revealing. However, still
more information can be gleaned from the actual map of land cover types and
quantification of neighborhood, distances between similar or contrasting
patch types, the nature and shape of boundaries, and the like.
Direct vs. Indirect Effects of Land Use
What does land use and land cover mean ecologically? In other words, can
land use or land cover be used as ecological variables? The answer depends
on the ecological arena and the research questions asked of that domain of
study.
If the ecological phenomena of interest respond directly to the land
use/land cover matrix, then the variables that describe land use should be
close determinants of the response. For example, birds may respond to the
neighboring vegetation, or mule deer may respond to the viscosity of the
boundary between forest and steppe.
Even in the cases of "direct" response, there may be one or more mechanisms
that an ecologist will have to interpose between the features of land use and
the ecological response. In the case of the birds, a more detailed mechanism
may be the avoidance of predators. In the case of the deer a similar kind of
mechanism may come into play, although the scale of sensitivity may differ
because the size and search range of the predators may differ. Both of these
cases interpose a relatively simple ecological model between the land use and
the ecological response.
These cases can be considered direct, but there still needs to be an
ecological model to translate between the abstractions of the map of land
cover types and the behavioral or functional response of the ecological
system. When traditional ecological variables are used, most ecologists who
are familiar with the subject may carry the general model in their heads, and
so land use/land cover can be readily perceived as a part of the roster of
environmental variables. However, even such direct uses of land use and land
cover have a relatively short history in ecology, dating in some cases to the
introduction of island biogeography, and in other more complex and subtle
cases, to the recognition of landscape ecology as a legitimate ecological
specialty.
There are other situations in which land use and land cover are not
environmental variables. Again, when such a disconnect occurs between land
use and ecology depends on the topic areas, the scales, and the question of
interest. One kind of situation where land use is a poor ecological variable
is one in which there is little direct connection between it and the
ecological response. Human population density is one aspect of land cover
that is often readily available. There may even be intriguing correlations
between human population and some ecological processes, such as number of
native species. However, the functional relationships in such a case are
likely to be indirect and multiple. Native biodiversity can be affected by
reduction in habitat area, which accounts for an estimated 80% of species
extinctions, for example. Alternatively, or in addition, loss of native
species can be due to pollution, deliberate human extirpation or harvest,
interaction with domestic predators, subsidy of wild predators, introduction
of new feral predators or exotic diseases, among other causes. While these
factors clearly relate to the nature and configuration of land use and land
cover types, they represent an array of more direct ecological interactions
that are affected by or allowed by the land use land cover setting. In order
for an ecologist to understand both the patterns of species richness relative
to land cover and use, and the mechanisms behind those patterns, an
explicitly ecological model of the factors involved, their interactions and
the outcomes of these interactions must be interposed between the land use
variables and the ecological responses.
Most attempts to use land use or land cover as ecological variables may fall
into the indirect category. This is because those sorts of variables have
not fallen within the domain of ecology for very long. Other variables such
as elevation, climate, geology, and the structure of habitats and
communities, are familiar to ecologists, and there are may formal and
informal models for linking those often coarse scale variables with processes
of ecological interest. For instance, climate variables are integrated
through models of energy balance, photosynthesis, and the like.
But there are few ecological models to relate land use and land cover to
important ecological functions. This is a serious lapse since human
influence is now so pervasive in the biosphere. Indeed it is spreading in
many places. Therefore modern ecology seems to demand a better understanding
of land use and land cover variables along with increased knowledge of
traditional and emerging areas of ecology.
An example of the link between ecology and land use, appears in the attempt
to understand the response of ecosystems to abstract gradients of the degree
of urbanization. Like climate or elevation, urbanization is a complex
ecological gradient that encompasses many factors. Unlike natural gradients,
for which ecologists have used direct measurement, multivariate statistics
and ordination, reciprocal and common garden transplants, simulation and
mechanistic models and the like, ecologists have not had practice in building
the models that translate from a complex and general variable to detailed
mechanistic ecosystem responses. However, such progress is possible and
desirable.
One exploratory attempt to assess the human role in forest function has been
made in a project on the New York City Metropolitan region. The metropolitan
area encompasses some of the densest and most highly developed land in the
U.S., as well as open, thinly populated farming and forested landscapes. In
other words it provides a wide range of conditions in which to discover the
relationships between human population, development and land use patterns.
Sample areas were established in several closed canopy forests in the
urbanized Bronx, suburbanized areas of Westchester County, New York, and
rural Litchfield County, Connecticut. All sites show the same bedrock and
soil types. The forests differed in key ecosystem parameters, such as litter
decay and nitrogen mineralization rates. What human factors might best
explain the ecosystem functions?
Several readily available variables were assembled for cells of 4x4 km
centered on the forest patches. Of the variables examined, traffic volume
explained the greatest proportion of the variation among forest processes.
Population density, and road density also explained large amounts of the
difference. However the fact that traffic volume can be more directly linked
to the input of nitrogen in the nearby atmosphere suggests that it is not
only the statistically best but also the ecologically most relevant variable.
Clearly other variables describing human impact and potential mechanisms
need to be examined and brought to bear via models so that the mechanisms can
be compared for their efficacy. The plausible relationships discovered by
the coarse scale statistical relationships, and the proposed mechanisms can
be tested by field transplants, common garden or laboratory experiments.
Land Use and a New Ecological Synthesis
The results from the New York urban to rural gradient show the way that land
use and land cover can enrich ecology. Because ecologists in the United
States have not focused on relationships and functions involving humans, there
are likely many patterns to be discovered. Not all such patterns will make
the transition from intriguing, to plausible, to mechanistically defensible.
But the search can profitably build from a broad base.
A second advantage of examining land use and land cover from an ecological
perspective is an expansion of the linkage between ecology and other
sciences. Humans, and the rich variety of behaviors, structures,
institutions, and processes they generate likely affect much of ecology.
However the interaction of ecologists with social scientists, historians,
economists, and geographers, among others who study humans and their
institutions, is in its infancy. Maturation of such interaction may generate
new questions about the human role in ecological systems.
With the integration of human and ecological studies, land use and land
cover can be used most intelligently to develop conceptual models and
empirical chains of causation that make them effective ecological variables.
New mechanisms and modes of interaction may emerge from the integrated study
of humans and ecological systems. Land use and land cover are the
appropriate entry points into the new arena.
Text-only
Privacy Policy and Disclaimers |
Accessibility |
FOIA
http://biology.usgs.gov/luhna/ecovar.html
Last Updated: Tuesday, 15-Aug-2000 08:26:50 MDT
|