Development of movement models to assess the spatial dynamics of marine fish populations.
Humston, Robert 2001
University of Miami, Rosenstiel School of Marine and Atmospheric Science, 254 pp.
greybar.jpg - 2645 Bytes

A simulation model is constructed which considers
the population dynamics of bonefish, Albula
vulpes, in a spatio-temporally articulated
seascape using biological and physical
environmental data describing Biscayne Bay. A
sub-model of behavioral movement of bonefish in
response to environmental cues is employed to
describe the dynamic distribution of individual
bonefish cohorts relative to features of their
habitat. This sub-model is developed from a
generalized framework based upon principles of
kinesis behavior described herein. The function
and performance of the movement model in
acquisition of spatial resources for growth is
analyzed with respect their subsequent influence
on bonefish stock dynamics predicted in
simulation. Two other movement behaviors are
inserted in simulation for comparison of results:
1) random walk; and 2) a model of restricted-area
search behavior. Simulations are designed to test
the null hypothesis that population dynamics are
not significantly influenced by the assumptions of
the behavioral movement sub-model. The results of
simulations are compared to elucidate the effect
behavioral assumptions of movement and their
practical application in simulation may have on
stock dynamic simulations incorporating the
spatial interactions of fish and habitat. It is
demonstrated that implicit differences in a priori
assumptions of behavioral movement can
significantly affect stock dynamics predicted in
simulation. The implications of simulation
results on existing hypotheses of bonefish biology
and stock dynamics are considered, which lead to
recommendations for further coordinated efforts in
both modeling and empirical studies. It is
concluded that assumptions dictating movement
behavior can significantly influence the
population dynamics predicted in simulation, yet
support of assumptions can be difficult to
demonstrate from empirical data. It is submitted
that a minimalist approach be adopted with respect
to cognitive mechanisms in the design of
behavioral movement models for incorporation into
simulations of fish population dynamics.