190
Yiannis G. Matsinos, Wilfried F. Wolff, and Donald L. DeAngelis
Using spatially explicit individual-based simulation models, the above authors
have shown that such details of landscape configuration can have important
consequences for the dynamics of a population. This holds not only for mobile
animals, which have time and energy costs associated with their movement and
the exploitation of resources, but also for plants, for which seed dispersal and
survival may depend on details of the landscape.
From the above considerations, we conclude that traditional state variable
models, although very useful in providing theoretical insights, may sometimes be
too coarse to delineate the detailed mechanisms that can operate on complex
landscapes. This may not make a difference in some cases, but in other cases it
may.
Individual-Based Models
Spatially explicit individual-based modeling allows the effects of demographic
stochasticity, the internal complexity within populations, and the subtle complexities of population interactions on a landscape to be taken into account in a fairly
straightforward manner. In the individual-based approach, each individual in a
population is modeled, differing in its own set of characteristics (e.g., age, sex,
size, condition, social status) from all other individuals. This can reflect genetic
differences as well as the different experiences of each organism (e.g., movements
across a landscape, encounters with prey and predators, mating, and accidents),
which are subject to stochasticity. In these models, each individual organism is
also capable of making decisions (e.g., about their movements, foraging, predator
avoidance, mating) that may or may not depend on temporally and spatially
varying factors. In general, an individual and its interactions with the environment
and other members of the population are modeled by decision rules based on the
behavior and physiology of the organism.
In individual-based models, there are no state variables representing total population size or the sizes of various components of the population. Instead, these
quantities are found by summing over the set(s) of relevant individuals. These
models are almost by necessity computer simulation models, so there is seldom
any hope of using analytic approaches. The emphasis is on developing rigorous
and efficient computer codes.
The rapid increase in the execution speed and memory of computers during the
past decade has played a major role in the development of individual-based
models. Individual-based models are often complex, in the sense that they require
extensive coding and usually considerable computer power and time. However,
they are usually conceptually simple, at least much simpler than is generally
realized. The basic features of an individual-based model are the following:
• A set of interacting individuals
• A set of rules that describe how local interactions are implemented
• A physical landscape, including resources
Yiannis G. Matsinos, Wilfried F. Wolff, and Donald L. DeAngelis
Using spatially explicit individual-based simulation models, the above authors
have shown that such details of landscape configuration can have important
consequences for the dynamics of a population. This holds not only for mobile
animals, which have time and energy costs associated with their movement and
the exploitation of resources, but also for plants, for which seed dispersal and
survival may depend on details of the landscape.
From the above considerations, we conclude that traditional state variable
models, although very useful in providing theoretical insights, may sometimes be
too coarse to delineate the detailed mechanisms that can operate on complex
landscapes. This may not make a difference in some cases, but in other cases it
may.
Individual-Based Models
Spatially explicit individual-based modeling allows the effects of demographic
stochasticity, the internal complexity within populations, and the subtle complexities of population interactions on a landscape to be taken into account in a fairly
straightforward manner. In the individual-based approach, each individual in a
population is modeled, differing in its own set of characteristics (e.g., age, sex,
size, condition, social status) from all other individuals. This can reflect genetic
differences as well as the different experiences of each organism (e.g., movements
across a landscape, encounters with prey and predators, mating, and accidents),
which are subject to stochasticity. In these models, each individual organism is
also capable of making decisions (e.g., about their movements, foraging, predator
avoidance, mating) that may or may not depend on temporally and spatially
varying factors. In general, an individual and its interactions with the environment
and other members of the population are modeled by decision rules based on the
behavior and physiology of the organism.
In individual-based models, there are no state variables representing total population size or the sizes of various components of the population. Instead, these
quantities are found by summing over the set(s) of relevant individuals. These
models are almost by necessity computer simulation models, so there is seldom
any hope of using analytic approaches. The emphasis is on developing rigorous
and efficient computer codes.
The rapid increase in the execution speed and memory of computers during the
past decade has played a major role in the development of individual-based
models. Individual-based models are often complex, in the sense that they require
extensive coding and usually considerable computer power and time. However,
they are usually conceptually simple, at least much simpler than is generally
realized. The basic features of an individual-based model are the following:
• A set of interacting individuals
• A set of rules that describe how local interactions are implemented
• A physical landscape, including resources