Ecological applications based on bacterial community abundance in reservoirs using an artificial neural network approach

Abstract

The objective of this study is to analyze the influence and predict abundance the heterotrophic bacteria (psychrophile; mesophile) and facultative oligotrophic bacteria as a reflection of ecological relationships in reservoirs and water quality. We used artificial neural networks (ANNs) to develop models based on input variables derived from two different reservoirs. The neural network models were developed using experimental data which is collected for ten years. Although reservoirs have a different position, different morphometric qualities, trophic state and dominant bacterial community there is a possibility of predicting these bacterial communities with the same input parameters. Comparing the modeled values by ANN with the experimental data indicates that neural network models provide accurate results. The important conclusion of this work is that ANNs can provide a flexible and applicable tool in monitoring water quality across bacterial communities in reservoirs.