Publications: Peer-reviewed journal articles (by staff)

Finding reference: a comparison of modelling approaches for predicting benchmarks for macroinvertebrate community metrics

20 December, 2016
CITATION

Clapcott JE, Goodwin EO, Snelder TH, Collier KJ and Neale WN. In press. Finding reference: a comparison of modelling approaches for predicting benchmarks for macroinvertebrate community metrics. New Zealand Journal of Marine and Freshwater Research 51(1): 44-59.

ABSTRACT

Reference benchmarks are needed to assess the contemporary status of rivers and to establish restoration targets. We developed predictive models to estimate site-specific reference values for a macroinvertebrate community index (MCI), which is used to indicate a range of human impacts on wadeable streams. We compared three statistical modelling approaches – general linear, boosted regression tree and random forest – and tested the effect of spatial scale on predictive accuracy by developing national and regional boosted regression tree models. Using fitted flexible models (boosted regression tree, random forest) and resetting predictors to reflect natural state provided the most accurate predictions of reference condition. Variation in reference MCI predictions from national and regional models was within the range observed from methodological and temporal variability. The proportion of native vegetation in upstream catchments was the primary predictor of MCI scores in all models, while secondary predictors varied regionally.