Main content

Ecological modelling in a non-stationary world: factors affecting recruitment of yellow perch (Perca flavescens) in Lake Erie

Show full item record

Title: Ecological modelling in a non-stationary world: factors affecting recruitment of yellow perch (Perca flavescens) in Lake Erie
Author: Zhang, Fan
Department: Department of Integrative Biology
Program: Integrative Biology
Advisor: Nudds, ThomasMcLaughlin, RobertJiao, Yan
Abstract: Parametric variation in stock-recruitment models can cause non-stationary (hereafter, referred to as temporally variant) stock-recruitment relationships, with important implications for fisheries management. I examined potential mechanisms causing non-stationary stock-recruitment relationships of Lake Erie yellow perch (Perca flavescens). Non-stationary stock-recruitment relationships were detected across the lake between 1975 and 2013, but the frequency of detection decreased with decreasing time scale. Effects of spawning stock size and age structure on recruitment were weak across the lake between 1975 and 2013. Counterintuitively, negative age-related maternal effects were observed. Strong effects of yearling walleye biomass on recruitment were detected in the western and central basins between 1978 and 2013. By incorporating yearling walleye biomass into stock-recruitment models, non-stationary stock-recruitment relationship disappeared in the west-central basin, but remained in the western or east-central basins. In the western basin between 1999 and 2013, the critical period during which recruitment (year class) strength was determined was during the first four months following spawning in May. The best indicators of effects of reproduction, bottom-up, top-down and physical environment effects on recruitment during this critical period were spawning stock biomass, mean relative zooplankton biomass between May and August, absolute yearling walleye abundance in August and dissolved oxygen between May and August, respectively. Time-varying importance of these factors was detected, suggesting that variation in optimal model structure and complexity, as well as parameter variation, is important contributor to non-stationary recruitment dynamics. Overall, there is a trade-off between parameter estimation and parametric variation when determining appropriate length of time series for stock-recruitment analyses, but incorporating ecological mechanisms into stock-recruitment models can contribute to resolving this trade-off by reducing parametric variation while reducing uncertainty in parameter estimation. Nevertheless, such mechanisms may act at various scales of time, space and organizational complexity, posing challenges for future studies to model ecological effects on fish recruitment with data collected and analysed at appropriate scales.
Date: 2015-09
Rights: Attribution-NonCommercial-NoDerivs 2.5 Canada
Terms of Use: All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.

Files in this item

Files Size Format View Description
Zhang_Fan_201509_PhD.pdf 2.187Mb PDF View/Open Ph.D thesis

This item appears in the following Collection(s)

Show full item record

Attribution-NonCommercial-NoDerivs 2.5 Canada Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivs 2.5 Canada