Evaluation of Water Quality for Mangrove Ecosystem Using Artificial Neural Networks

Abstract

With the rapid urbanization and socioeconomic development, mangrove ecosystem, especially the water quality in the coastal environment is getting increasingly vulnerable. In our work, the model of back-propagation (BP) neural network was established based on the national standards of surface water (GB3838-2002). The resulting model was used to classify water quality of mangrove. Then the relationship between water quality and diseases and insect pests was analyzed. The results show that water quality of 1, 2 and 3 monitoring sites is the worst, and the water quality in August is significantly better than that in the other two months, which would be useful for recognizing the polluted areas and determining the priority preservation areas. Additionally, it is found that there is relevance between water quality and diseases and insect pests, which could provide basis for subsequent study on diseases and insect pests.