Weighted fuzzy inference system for water quality management of Chirostoma estor estor culture

Abstract

The increase in overfishing and pollution in Pátzcuaro Lake, Michoacán, México, has generated a severe problem in the endemic species that live there; especially the Chirostoma estor estor that has been threatened. A computational model has been developed for the evaluation of water quality in freshwater intensive farming tanks to contribute to the conservation efforts of this species. The proposed model uses a fuzzy inference system weighted through a rule categorization process. Five parameters were selected for the analysis: dissolved oxygen, pH, temperature, total ammonia, and non-ionized ammonia, because they represent the primary set of parameters that affect the water quality and health of the species studied. The measurements were made on the Chirostoma farms, and different situations were studied to define weights of importance according to the situations of negative impact on water quality. The rules are maximized or minimized with those weights and then integrated into a final score (Chirostoma Water Quality Index (CWI). The CWI values have been compared with NSF and CCME. It is noted that the CWI values are more representative of the actual state of water quality for the Chirostoma estor estor. This is because the weighted fuzzy logic approach is sensitive to all parameters and can identify harmful situations by using the rules. CWI can be a useful tool for monitoring and managing the quality of tanks, as it provides helpful information to prioritize and maintain water quality, and this in turn favors the growth, reproduction, and preservation of the species.