fuzzy clustering analysis (fca)

Management decisions whose environmental impacts affect directly or indirectly surface waters must of necessity be based on adequate knowledge and information when water quality zoning and a clear picture of river water quality are sought. Water quality zoning is based on pollution criteria that are identified on the basis of different water quality parameters drawn from historical data and the water uses in the region. The aggregate of the data and parameters involved make river water quality modeling a complex process. In this paper, the Principal Component Analysis (PCA) is used to reduce the water quality parameters involved in the identification of river water pollution criteria. The method keeps those components with more variances. The results show that the first component transfers 93.59% of the variation in the data, while the first two and the first six components explain 96.67% and 99.99% of the variations, respectively. Based on the criteria thus identified, the fuzzy clustering analysis is used in a second stage of the study to classify the river intervals. For this purpose, the fuzzy water quality data are provided to generate the fuzzy similarity matrix based on the fuzzy relations. Then, the stabilized matrix and the clustering diagram are created. Finally, the river intervals are classified into similar categories using the proper thresholds. The efficiency of the proposed method is evaluated by employing water quality data collected from the Zayandehrood River monitoring stations.