12. Individual-Based Models and Assessment of Population Variability
191
Each individual in such models is characterized by a set of variables regarding
the different internal states (e.g., age, size). The spatial environment requires
another variable, the location of the individual. By keeping track of the location of
each individual, the model is thus able to identify which specific individuals might
be interacting at a given moment (e.g., competing for food while sharing a particular spatial locale, usually represented as a unit cell on a spatial grid).
Individual-based models are mechanistic in approach. They use information at
the level of the behavior and physiology of individual organisms. In principle,
there is no limit to the amount of detail that can be put into the description of the
interactions of individuals with their environment and with other individuals. This
makes the individual-based modeling approach especially suitable for species for
which individual organisms have been studied in some detail. Because complete
physiological and behavioral information is rarely available for any given species,
reasonable estimates must be made for those aspects that are not well known.
Sensitivity to a complex array of environmental conditions can be incorporated
into individual-based models. This includes stochasticity in environmental conditions on any time scale, because the time steps in the model are set at whatever
size is appropriate to include all important environmental effects on individuals.
The fact that individual-based models consider not only the numbers of organisms in populations but also the condition of the individuals within the population
confers another great advantage. There are conceivable cases in which population
size remains relatively constant through time but in which conditions of the
individuals are declining. Although the population may seem healthy from the
point of view of numbers alone, it may actually be on the brink of collapse. The
health of individuals could deteriorate long before this becomes manifest in a
numerical collapse of the population. Thus the models allow one to follow not
only the variability of numbers in a population, but the often equally important
variability in the internal state of the individuals within the population.
Before discussing in more detail the uses and limitations of spatially explicit
individual-based models, we describe a specific case study in some detail.
Case Study of an Individual-Based Model:
Wading Birds in the Everglades
The purpose of this model is to predict one aspect of population variability, the
variations in reproduction that can occur as a result of changing abiotic conditions
and the resultant changes in patterns of prey availability. Although it is not a
complete population model, it illustrates how this approach can be used to predict
population variations.
An individual-based model was developed by Wolff (1994) (see also Fleming
et al. 1994) to assist in answering questions about the decline of colonially nested
wading birds in the Florida Everglades. Over the past decades, wading bird
populations in the Everglades have experienced declines both in population numbers and in reproductive success concomitant with a series of anthropogenic
191
Each individual in such models is characterized by a set of variables regarding
the different internal states (e.g., age, size). The spatial environment requires
another variable, the location of the individual. By keeping track of the location of
each individual, the model is thus able to identify which specific individuals might
be interacting at a given moment (e.g., competing for food while sharing a particular spatial locale, usually represented as a unit cell on a spatial grid).
Individual-based models are mechanistic in approach. They use information at
the level of the behavior and physiology of individual organisms. In principle,
there is no limit to the amount of detail that can be put into the description of the
interactions of individuals with their environment and with other individuals. This
makes the individual-based modeling approach especially suitable for species for
which individual organisms have been studied in some detail. Because complete
physiological and behavioral information is rarely available for any given species,
reasonable estimates must be made for those aspects that are not well known.
Sensitivity to a complex array of environmental conditions can be incorporated
into individual-based models. This includes stochasticity in environmental conditions on any time scale, because the time steps in the model are set at whatever
size is appropriate to include all important environmental effects on individuals.
The fact that individual-based models consider not only the numbers of organisms in populations but also the condition of the individuals within the population
confers another great advantage. There are conceivable cases in which population
size remains relatively constant through time but in which conditions of the
individuals are declining. Although the population may seem healthy from the
point of view of numbers alone, it may actually be on the brink of collapse. The
health of individuals could deteriorate long before this becomes manifest in a
numerical collapse of the population. Thus the models allow one to follow not
only the variability of numbers in a population, but the often equally important
variability in the internal state of the individuals within the population.
Before discussing in more detail the uses and limitations of spatially explicit
individual-based models, we describe a specific case study in some detail.
Case Study of an Individual-Based Model:
Wading Birds in the Everglades
The purpose of this model is to predict one aspect of population variability, the
variations in reproduction that can occur as a result of changing abiotic conditions
and the resultant changes in patterns of prey availability. Although it is not a
complete population model, it illustrates how this approach can be used to predict
population variations.
An individual-based model was developed by Wolff (1994) (see also Fleming
et al. 1994) to assist in answering questions about the decline of colonially nested
wading birds in the Florida Everglades. Over the past decades, wading bird
populations in the Everglades have experienced declines both in population numbers and in reproductive success concomitant with a series of anthropogenic