Establishing nutrient thresholds in the face of uncertainty and multiple stressors: A comparison of approaches using simulated datasets.

摘要: Abstract Various methods have been proposed to identify threshold concentrations of nutrients that would support good ecological status, but the performance these and influence other stressors on underlying models not fully evaluated. We used synthetic datasets compare ordinary least squares, logistic quantile regression, as well as, categorical based distribution nutrient categorised by biological status. The differed in their levels variation between explanatory response variables, were centered at different positions along stressor (nutrient) gradient. In order evaluate “multiple stressor” situations, another set with two was used. Ordinary squares regression most reliable when predicting concentration sole stressor; however, both had a tendency underestimate second present. contrast, produced strongly influenced level (nutrient enrichment, this case) relative they trying predict (good/moderate instance). Although all tested limitations presence stressor, upper quantiles seemed generally appropriate establish non-precautionary thresholds. For example, may be establishing targets for restoration, seeking minimise deterioration. Selection an should also attend regulatory regime (i.e. policy requirements environmental management context) within which it will used, ease communicating principles managers stakeholders.