Research Document - 2004/087

Coastal exposure as a first-order predictor of the productive capacity of near shore habitat in the Great Lakes

By Randall, R.G., Minns, C.K., Brousseau, C.M.

Abstract

Regression tree analysis with coastal exposure (fetch distance) as a predictor of fish biomass was used to evaluate the productive capacity of near shore habitat in the Great Lakes. Regression tree models were developed using survey data collected at coastal wetlands, harbours and natural shorelines in 1994 (n=100) and validated using data from other areas surveyed in other years (1990 to 1999, n=273). Coastal habitat characteristics that influence fish distribution, including the occurrence and abundance of aquatic macrophytes, water temperature and substrate characteristics, were related to maximum fetch distance in a consistent manner in the model and validation data sets. Three classes of macrophyte density (absent, moderate and dense cover), were predicted from substrate size and fetch distance: plant cover was highest where the predominant particle size was fine (silt or smaller) and maximum fetch was < 12.6 km. Fetch was a significant predictor of the biomass of three species (Lepomis gibbosus, Perca flavescens, and Alosa pseudoharengus), each with different habitat preferences, and two fish community indices (Index of Biotic Integrity [IBI], and the Habitat Productivity Index [HPI]). IBI and HPI were used as measures of the diversity and production components of habitat productive capacity, respectively. For all fish response variables, classification was improved if fetch was used together with associated habitat attributes as predictors. The degree of resolution of habitat classification (number of classes that were distinct) was limited to 2 to 4 classes, depending on the fish response variable. Proportional reduction in error for the regression trees ranged between 0.30 and 0.76. Four classes of L. gibbosus habitat were determined and validated, but the number of habitat classes for P. flavescens and A. pseudoharengus was less. For the whole fish assemblage, four habitat classes were identified using IBI and HPI together in a two-axes approach for evaluating productive capacity, along with fetch and water temperature as predictors. Knowledge of site exposure and the associated habitat covariates can be used to determine and map first-order estimates of coastal habitat productive capacity in the Great Lakes.

Complete document (PDF, 537 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.