Fuzzy modelling of chlorophyll production in a Brazilian upwelling system

In the context of coastal management the aim of this paper is to present the development of a fuzzy model through the application of a Genetic algorithm in order to select the most appropriate set of variables and improve our understanding with a set of rules. The case studied is the chlorophyll response as bioindicator of ecological status in the Northeast coastal upwelling system of Rio de Janeiro state, Southeastern of Brazil. The prediction of the fuzzy model has shown an improved performance when compared to the traditional approaches as Multiple Regression modelling. The results show that the set of inferred rules can assess three different water masses. Despite the increased occurrence of upwelling is observed in spring-summer period and some instability of the model, it is able to forecast some chlorophyll peaks. We conclude that the sampling frequency is crucial to reach a better performance. Crown Copyright (C) 2009 Published by Elsevier B.V. All rights reserved.