Analysis of macrobenthic communities in Flanders, Belgium, using a stepwise input variable selection procedure with artificial neural networks

Abstract

The effect of environmental conditions on river macrobenthic communities was studied using a dataset consisting of 343 sediment samples from unnavigable watercourses in Flanders, Belgium. Artificial neural network models were used to analyse the relation among river characteristics and macrobenthic communities. The dataset included presence or absence of macroinvertebrate taxa and 12 physicochemical and hydromorphological variables for each sampling site. The abiotic variables served as input for the artificial neural networks to predict the macrobenthic community. The effects of the input variables on model performance were assessed in order to identify the most diagnostic river characteristics for macrobenthic community composition. This was done by consecutively eliminating the least important variables and, when beneficial for model performance, adding previously removed ones again. This stepwise input variable selection procedure was tested not only on a model predicting the entire macrobenthic community, but also on three models, each predicting an individual taxon. Additionally, during each step of the stepwise leave-one-out procedure, a sensitivity analysis was performed to determine the response of the predicted macroinvertebrate taxa to the input variables applied. This research illustrated that a combination of input variable selection with sensitivity analyses can contribute to the development of reliable and ecologically relevant ANN models. The river characteristics predicting presence or absence of the benthic macroinvertebrates best were the Julian day, conductivity, and dissolved oxygen content. These conditions reflect the importance of discharges of untreated wastewater that occurred during the period of investigation in nearly all Flemish rivers.

Applications of artificial neural networks predicting macroinvertebrates in freshwaters

“…In addition, the variables have to be scaled in such a way as to be commensurate with the limits of the activation functions used in the output layer (Maier and Dandy 2000). Several authors (Chon et al , 2002Gabriels et al 2007;Obach et al 2001;Park et al 2003a, b;Schleiter et al 1999;Schleiter et al 2001;Wagner et al 2000) proportionally normalized the data between zero and one [0 1] in the range of the maximum and minimum values, while Dedecker et al (2004Dedecker et al ( , 2005a and Gabriels et al (2002) used the interval [À1 1]. Moreover the division of the dataset in folds for cross-validation is crucial for a good model training and evaluation.…”

“…On the other hand, one can start with all the available variables and remove one by one the least important ones (e.g. Beauchard et al 2003;Gabriels et al 2007). Disadvantages of these approaches are that they are computationally intensive and that they are unable to capture the importance of certain combinations of variables that might be insignificant on their own.…”

“…However, these kappa values also represent the information content of the dataset, and each dataset has a limit regarding extractable information. Consequently, differences in kappa threshold values (and evaluation classes) can be expected between disciplines in general and datasets in particular (Gabriels et al 2007). In an ecological context, Randin et al (2006) assess kappa values as follows: 0.00-0.40: poor; 0.40-0.75: good; 0.75-1.00: excellent.…”

Artificial neural network modelling of macrophyte indices based on physico-chemical characteristics of water

“…Advanced data analysis techniques, among which artificial neural networks have become particularly popular in recent years, and have been increasingly used in interpreting the results of environmental research (Gabriels et al, 2007;Iliadis & Maris, 2007;Samecka-Cymerman et al, 2009;Gevrey et al, 2010;Penczak et al, 2012). Statistical programs that are based on artificial neural networks are applicable where traditional methods of data analysis do not provide satisfactory results (Lencioni et al, 2007;Palialexis et al, 2011).…”

Development of artificial neural network models predicting macroinvertebrate taxa in the river Axios (Northern Greece)

“…Goethals et al (2002), Gabriels et al (2006) and Manel et al (2001) derived similar conclusions when predicting macroinvertebrate taxa using a limited set of environmental variables. On the other hand, their respective Cohen's kappa statistic had low values indicating that these predictions were based on chance.…”