Macroinvertebrate Assemblage Responses to Anthropogenic Stressors: A Bioassessment Scoring Tool for Managing Stream Ecosystems In Eastern Ontario

Description
Title: Macroinvertebrate Assemblage Responses to Anthropogenic Stressors: A Bioassessment Scoring Tool for Managing Stream Ecosystems In Eastern Ontario
Authors: Théberge, Elysabeth
Date: 2016
Abstract: Using site-specific macroinvertebrate monitoring information and catchment-scale predictors, we developed multiple linear regression models to assess stream ecological status in eastern Ontario. With the developed equations, we hindcasted metric reference conditions for each site by setting anthropogenic activity variables to zero. Observed metric values were then scored based on their deviation from their expected reference value. These deviation scores were subsequently scaled by an estimate of the regional natural variability (i.e., distribution of residuals) observed for each metric. Scaled scores were then used to construct a summary index incorporating multiple, now compatible metrics. Although our simplified models were highly statistically significant, they often explained only a fraction of the observed variability in metric responses (R2 <0.34). Furthermore, there was evidence of lack of fit of the models, as indicated by the relatively high variability around predicted values when compared to the measurement error (variability between replicate samples). These results imply that diagnosis power of individual site condition assessments is low. However, the scaled deviation scores incorporate an evaluation of uncertainty in site condition diagnoses, given by the variability estimates (i.e., residuals) around each predicted metric reference value. This makes them suitable for use in a ‘probabilistic’ risk assessment framework – where a large deviation score indicates a higher probability (risk) that an observed metric value is outside of reference condition (i.e., impacted). By further manipulating anthropogenic activity levels in the developed model equations, it also becomes possible to test for alternative management scenarios and estimate associated risks. These features of the scaled metric deviation scores make them a useful tool in watershed management decisions.
URL: http://hdl.handle.net/10393/34263
http://dx.doi.org/10.20381/ruor-5178
CollectionThèses, 2011 - // Theses, 2011 -
Files