**The title, authors, and abstract for this completion report are provided below.  For a copy of the completion report, please contact the GLFC via e-mail or via telephone at 734-662-3209**



Development and assessment of a predictive model of metamorphosis for Great Lakes' sea lamprey populations.


T.B. Steeves1, A.J. Treble1, and M.L. Jones2

1 Sea Lamprey Control Centre, 1 Canal Drive, Sault Ste Marie, ON, P6A 6W4

2Michigan State University, East Lansing, MI, USA







We examined the accuracy and utility of the length-dependent logistic regression models of lamprey metamorphosis currently used to forecast the parasitic lamprey production from Great Lakes tributaries. We found that lamprey weight, age, sex, and larval density, as well as stream temperature and location all to be significant variables in the model that used direct observation of metamorphosis as the dependent variable. Non-invasive measures of lipid content, such as total body electrical impedance and hydrostatic weight (the weight of the larvae while suspended in water) as a surrogate for lipids, were not significant in the model. The final model correctly predicted whether lamprey larvae captured in year t would remain as larvae or metamorphose in year t+1 in 95.4% and 87.5% of the observations, respectively. The model currently used to predict metamorphosis correctly forecasted 91.6% of larval and 33.3% of metamorphosed lampreys one year after capture. In a simulation study we evaluated the performance of various approaches to prioritizing streams for lampricide application, including the current model used by the lamprey control agents, and a simpler, length threshold model that uses larval lamprey length and abundance rather than forecasting the expected abundance of metamorphosed lamprey. We found that, when uncertainty in model predictions is low, the current “transformer cost/kill” method results in the largest suppression of lampreys through time. As model forecasts become less precise, we found that the length threshold model resulted in the highest level of lamprey suppression. Our results demonstrate that changes in the models and methods the control agents use to allocate control effort may result in greater suppression of lamprey abundance in the Great Lakes for the same cost.