Predictive models of collembolan diversity and abundance in a riparian habitat

The artificial neural network (ANN) was used in this work for modelling the abundance and diversity of hydrophilous Collembola on the microhabitat scale. The procedure was applied to a Collembolan assemblage of the northern Pyrenees. Six variables were retained to describe its structure: abundance of the three dominant species, species richness, overall abundance of Collembola, and Shannon index. Seven environmental variables were selected as explanatory variables: distance to water, soil temperature, water content, and proportion of mineral soil, moss, litter and rotten wood in the substrate. Correlations between observed values and values estimated by ANN models of the six dependent variables were all highly significant. The ANN models were developed from 83 samples chosen at random and were validated on the 21 remaining samples. The role of each variable was evaluated by inputting fictitious configurations of independent variables and by checking the response of the model. The resulting habitat profiles depict the complex influence of each environmental variable on the biological parameters of the assemblage, and the non-linear relationships between dependent and independent variables. The main results and the ANN potential to predict biodiversity and structural characteristics of species assemblages are discussed.