Current invasive species research being conducted at MSU
Return to MSU Research

Project 3

Project Title: Estimating the relationship between sea lamprey-induced mortality on lake trout and observed marking rates.
Investigators: James R. Bence and Travis O. Brenden
Contact: James R. Bence (bence@msu.edu)

Abstract:  We will estimate the relationship between sea lamprey-induced mortality on lake trout and observed marking rates in the upper Great Lakes. We will do this by modifying existing lake trout stock assessment models so that the relationship is estimated at the same time other parameters of the assessments are determined. We will explore whether the data support current assumptions used in assessment models based on laboratory derived estimates of lethality and healing and an assumption that sea lamprey-induced mortality is directly proportional to attack rates inferred from marking statistics. Our suggested approach takes advantage of the existing assessment models to adjust for effects of fishing on the relationship, and of recent contrasting temporal patterns in marking rates among lakes. In more detail our methods are as follows. (1) Existing statistical catch-at-age models used to assess lake trout abundance and mortality in 1836 treaty waters of Lakes Huron, Superior and Michigan will be modified so that the parameters of all the models will be estimated simultaneously. These models currently assume a known direct proportionality between marking rates and the sea lamprey component of natural mortality. This approach will first be replaced by allowing the proportionality constant to be estimated at the same time as all other parameters in the assessment model. We will then explore model variants for which the proportionality constant is allowed to vary by lake or region within lake, or for which the proportionality assumption is replaced by a curvilinear relationship. In this process we will use a Bayesian approach and evaluate evidence in favor of different hypotheses based on posterior distributions and the Deviance Information Criterion, an information theoretic model selection approach.Although of critical importance, much uncertainty remains about the relationship between damage caused by sea lamprey on host populations and observed host marking statistics (see sea lamprey assessment theme and draft damage theme cited therein (Ebener and Christie July 2004 draft)). This proposal addresses the first research need expressed in Ebener and Christie in a draft damage theme paper for the GLFC to replicate the earlier analysis of Koonce and Pycha that relates sea lamprey marking to lake trout mortality.