| Popis: |
Human activities have severely deteriorated the Flemish river systems, and many functions such as drinking water supply, fishing, ... are threatened. Because their restoration entails drastic social (e.g. change in habits with regard to water use and discharge, urban planning) and economical (e.g. investment in nature restoration, wastewater treatment system installation) consequences, the decisions should be taken with enough forethought. Ecosystem models can act as interesting tools to support decision-making in river restoration management. In particular models that can predict the habitat requirements of organisms are of considerable importance to ensure that the planned actions have the desired effects on the aquatic ecosystems. In preliminary studies, Artificial Neural Network (ANN) models were tested and optimized to obtain the best model configuration for the prediction of the habitat suitability for Gammarus pulex based on the abiotic characteristics of their aquatic environment in the Zwalm river basin (Flanders, Belgium). Although, these ANN models are in general quite robust with a rather high predictive reliability, the model performance has to be increased with regard to simulations for river restoration management. In particular, spatial-temporal expert-rules have to be included. Migration kinetics (downstream drift and upstream migration) of the organism and migration barriers along the river (weirs, impounded river sections, ...) can deliver important additional information on the effectiveness of the restoration plans, and also on the timing of the expected effects. This paper presents an overview and quantification of the factors affecting the upstream and downstream movements of Gammarus pulex. During further research, ANN models will be used to predict the habitat suitability for Gammarus pulex after several restoration options. The migration models, implemented in a Geographical Information System (GIS), are applied to calculate the migration time to the restored parts of the river. In this way, decision makers have an idea whether and when a selected restoration option has the desired effect. |
