SEA LAMPREY (PETROMYZON MARINUS) POPULATION DYNAMICS, ASSESSMENT, AND CONTROL STRATEGY EVALUATION IN THE ST. MARYS RIVER, MICHIGAN

Abstract

The St. Marys River is a major producer of invasive parasitic sea lampreys (Petromyzon marinus) to Lake Huron. My dissertation seeks to inform the management process for sea lamprey through a combination of statistical and simulation modeling. In Chapter 1, I developed a spatial age-structured model and applied it to the sea lamprey population in the St. Marys River. The model included a stock-recruitment function, spatial recruitment patterns, natural mortality, chemical treatment mortality, and larval metamorphosis. Recruitment was variable, and an upstream shift in recruitment location was observed over time. During 1993-2011, transformer escapement decreased by 86%. The model successfully identified areas of high larval abundance and showed that areas of low larval density contribute significantly to the population. In Chapter 2, I evaluated six methods of estimating sea lamprey density and abundance including the currently used sampling-based estimates, generalized linear and additive models, the population model from Chapter 1, and a hybrid approach. Methods were evaluated based on accuracy in matching independent validation data. The hybrid method was identified as the best method to inform sea lamprey control decisions in the St. Marys River due to its consistent performance. In Chapter 3, I used a resampling approach to estimate the effect of sampling intensity on the success of sea lamprey control and examined the economic tradeoff between assessment and control efforts. Sea lamprey control actions based on assessment outperformed those implemented with no assessment under all budget scenarios. The sampling intensity that maximized the number of larvae killed depended on the overall budget, with increased sampling intensities maximizing effectiveness under medium to large budgets. In Chapter 4, I conducted a management strategy evaluation using a stochastic simulation model to evaluate several fixed and survey-based Bayluscide-based treatment strategies for sea lamprey. The model incorporated population dynamics, sampling and assessment, and larval control actions. Treatment options with higher cost resulted in larger long-term reductions in transformer escapement, but increasing treatment effort did not result in a proportional decrease in transformer escapement. Survey-based treatment scenarios were the most desirable from both an economic and population control perspective.