Spatial and temporal variation in Great Lakes percid catch-per-effort data

Fishery-independent surveys are widely used to collect data on yellow perch and walleye across the Great Lakes basin.  Although used to fulfill many objectives, one of the most common uses of survey data is to infer changes in relative abundance over time, based on indices such as catch per effort (CPE).  However, our ability to detect changes in CPE is reduced by the large variability inherent in percid CPE time series.  Partitioning total variability into multiple spatial and temporal sources (i.e., variance components) is a powerful approach to accommodate variable time series data and for elucidating population trends and refining monitoring and assessment programs.  Furthermore, variability itself may be a useful indicator of large-scale ecological perturbations.  We are analyzing CPE time series from multiple percid populations using hierarchical statistical models to quantify the spatial and temporal variability across the Great Lakes basin.  We will discuss the results from these models in terms of (1) similarities and differences in variance components across Great Lakes, (2) the implications for monitoring programs, and (3) the usefulness of individual variance components as indicators of ecological change.