Matrix projection models and individual-based models (IBM) are commonly used for the
analysis and management of fish populations. Matrix models break down the population into
age or stage classes, while IBMs track individuals. I perform a series of quantitative
comparisons between the predictions of the two modeling approaches using the IBM as the
standard of comparison to demonstrate when individual variation, species interactions, and
spatial heterogeneity adversely affect matrix model performance. I first evaluate the matrix
approach for predicting yellow perch population responses when perch are involved in size-specific
predator-prey interactions with walleye. I created density-dependent and stochastic age-structured and stage-within-age matrix models from an Oneida Lake walleye-yellow perch IBM,
and then changed perch survival rates within the matrix models and IBM and compared their
predicted responses. The matrix models simulated yellow perch responses reasonably well when
density-dependent YOY survival was correctly defined. At least 20 years of data (IBM output)
were needed to correctly estimate the density-dependent relationships in the matrix models.
Second, I developed a 2-species matrix model by linking the elements between perch and
walleye matrix models. The 2-species model simulated yellow perch prey responses reasonably
well, but was unable to correctly predict walleye predator responses. Third, I developed a new
IBM that simulated a 6-species tidal marsh community on a fine-scale spatial grid of habitat
cells. The IBM was used to scale individual-level effects of lowered dissolved oxygen and
habitat degradation to population-level responses, and used to estimate relatively simple stage-based matrix models for grass shrimp and gulf killifish populations. Equilibrium analysis of the simple matrix models was insufficient for predicting population responses. This study showed
that stochastic, density-dependent matrix projection models were able to mimic density-
dependent survival processes and species interactions relatively well, while equilibrium analysis of simple matrix models was inadequate. The matrix approach consistently had trouble
estimating density-dependent and inter-specific growth relationships that were important for
accurate model predictions. I recommend the use of IBMs and relatively complicated matrix
models (stage-within-age, stochastic, density-dependent, multispecies) for simulation of fish
population and community dynamics.
| Identifer | oai:union.ndltd.org:LSU/oai:etd.lsu.edu:etd-07062007-155333 |
| Date | 09 July 2007 |
| Creators | Sable, Shaye Ellen |
| Contributors | Gregory Stone, James P. Geaghan, James H. Cowan, Jr., Kenneth A. Rose, Megan K. Lapeyre |
| Publisher | LSU |
| Source Sets | Louisiana State University |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | http://etd.lsu.edu/docs/available/etd-07062007-155333/ |
| Rights | unrestricted, I hereby certify that, if appropriate, I have obtained and attached herein a written permission statement from the owner(s) of each third party copyrighted matter to be included in my thesis, dissertation, or project report, allowing distribution as specified below. I certify that the version I submitted is the same as that approved by my advisory committee. I hereby grant to LSU or its agents the non-exclusive license to archive and make accessible, under the conditions specified below and in appropriate University policies, my thesis, dissertation, or project report in whole or in part in all forms of media, now or hereafter known. I retain all other ownership rights to the copyright of the thesis, dissertation or project report. I also retain the right to use in future works (such as articles or books) all or part of this thesis, dissertation, or project report. |
Page generated in 0.0015 seconds