An invertebrate predictive model (NORTI) for streams and rivers: Sensitivity of the model in detecting stress gradients

Abstract

NORTI (NORThern Spain Indicators system) is a predictive model for assessing the ecological status of rivers of Northern Spain based on invertebrates. The system can be used to assign any test site to a type of river under minimal disturbed conditions. Macroinvertebrates were sampled with a multihabitat approach from 676 sites covering the variation in environmental conditions across Northern Spain, between 2000 and 2008 (n=1421 samples), including a spatial network of 108 reference sites selected by the absence of significant pressures. A multinomial logistic regression was conducted using the GAAC cluster-derived groups of reference sites as response variable. Obligatory typology factors, following WFD System A, were included as forced entry terms in the model, other potential predictors were selected using a forward stepwise procedure. Ecological quality ratios (EQRs) were estimated from the observed similarity between the faunal composition of the sample of interest (test sample) and the expected median similarity for the reference community of each river type. The model predictions as EQRs responded significantly to the most important pressures: sewages inputs, eutrophication, hydromorphological alterations, and intensive and low intensity agriculture, demonstrating its accuracy in detecting impact in Northern Spanish streams and rivers.