18
Ecosystem Structure and
Function Modeling
Hope C. Humphries and Jill Baron
18.1 Introduction
An important component of ecological assessments
is the ability to predict and display changes in
ecosystem structure and function over a variety of
spatial and temporal scales. These changes can occur over short (less than 1 year) or long time frames
(over 100 years). Models may emphasize structural
responses (changes in species composition, growth
forms, canopy height, amount of old growth, etc.)
or functional responses (cycling of carbon, nutrients, and water). Both are needed to display
changes in ecosystem components for use in robust
ecological assessments. Structure and function
models vary in the ecosystem components included, algorithms employed, level of detail, and
spatial and temporal scales incorporated. They
range from models that track individual organisms
to models of broad-scale landscape changes. This
chapter describes models appropriate for ecological assessments. The models selected for inclusion
can be implemented in a spatial framework and for
the most part have been run in more than one system. Model assumptions, advantages, limitations,
and applications are discussed, and model features
are summarized in Table 18.1.
Models are needed in ecological assessments for
several reasons. The relatively short time periods
and limited resources often allotted for assessments
may preclude extensive empirical studies of ecosystem responses to changes. Experimental manipHope C. Humphries wishes to acknowledge partial funding provided by a Science to Achieve Results grant from
the U.S. Environmental Protection Agency ("MultiScaled Assessment Methods: Prototype Development
Within the Interior Columbia Basin"). Jill Baron wishes
to acknowledge the support of the U.S. Geological Survey Global Change Program.
ulation may not be feasible at the scale of a landscape or region due to prohibitive costs, and it may
be difficult to control for environmental changes
that have both direct and indirect effects on ecosystem components (Baker, 1989a). Adequate replication of large-scale experiments may not be possible, but many replications can be performed for
stochastic model simulations, providing an estimate of the range of potential effects (Turner et aI.,
1995). Models can generate output data that are fine
grained as well as spatially and temporally extensive (Smith and Urban, 1988).
Ecosystem processes such as succession may operate over very long periods of time due to the
longevity of many plant species. Modeling can be
used to extend the temporal and spatial scales considered in an assessment and can allow the user to
explore ecosystem dynamics following different
disturbance regimes with specified initial conditions (Dale et al., 1986). Environmental variables
can be changed while holding organism attributes
constant, and vice versa (Dunning et aI., 1995). Responses to various management scenarios can be
evaluated as a series of what-if questions. In addition, a simulation model represents a set of hypotheses about the functioning of a system and can
be used to summarize understanding of ecosystem
relationships in a quantitative and explicit manner
and suggest areas where further research is needed.
18.2 Models of
Ecosystem Structure
18.2.1 Transition Matrix Models
Transition matrix models, characterized by a matrix representing transitions in states of an ecological system over a specified period of time, include
257
Ecosystem Structure and
Function Modeling
Hope C. Humphries and Jill Baron
18.1 Introduction
An important component of ecological assessments
is the ability to predict and display changes in
ecosystem structure and function over a variety of
spatial and temporal scales. These changes can occur over short (less than 1 year) or long time frames
(over 100 years). Models may emphasize structural
responses (changes in species composition, growth
forms, canopy height, amount of old growth, etc.)
or functional responses (cycling of carbon, nutrients, and water). Both are needed to display
changes in ecosystem components for use in robust
ecological assessments. Structure and function
models vary in the ecosystem components included, algorithms employed, level of detail, and
spatial and temporal scales incorporated. They
range from models that track individual organisms
to models of broad-scale landscape changes. This
chapter describes models appropriate for ecological assessments. The models selected for inclusion
can be implemented in a spatial framework and for
the most part have been run in more than one system. Model assumptions, advantages, limitations,
and applications are discussed, and model features
are summarized in Table 18.1.
Models are needed in ecological assessments for
several reasons. The relatively short time periods
and limited resources often allotted for assessments
may preclude extensive empirical studies of ecosystem responses to changes. Experimental manipHope C. Humphries wishes to acknowledge partial funding provided by a Science to Achieve Results grant from
the U.S. Environmental Protection Agency ("MultiScaled Assessment Methods: Prototype Development
Within the Interior Columbia Basin"). Jill Baron wishes
to acknowledge the support of the U.S. Geological Survey Global Change Program.
ulation may not be feasible at the scale of a landscape or region due to prohibitive costs, and it may
be difficult to control for environmental changes
that have both direct and indirect effects on ecosystem components (Baker, 1989a). Adequate replication of large-scale experiments may not be possible, but many replications can be performed for
stochastic model simulations, providing an estimate of the range of potential effects (Turner et aI.,
1995). Models can generate output data that are fine
grained as well as spatially and temporally extensive (Smith and Urban, 1988).
Ecosystem processes such as succession may operate over very long periods of time due to the
longevity of many plant species. Modeling can be
used to extend the temporal and spatial scales considered in an assessment and can allow the user to
explore ecosystem dynamics following different
disturbance regimes with specified initial conditions (Dale et al., 1986). Environmental variables
can be changed while holding organism attributes
constant, and vice versa (Dunning et aI., 1995). Responses to various management scenarios can be
evaluated as a series of what-if questions. In addition, a simulation model represents a set of hypotheses about the functioning of a system and can
be used to summarize understanding of ecosystem
relationships in a quantitative and explicit manner
and suggest areas where further research is needed.
18.2 Models of
Ecosystem Structure
18.2.1 Transition Matrix Models
Transition matrix models, characterized by a matrix representing transitions in states of an ecological system over a specified period of time, include
257