九州大学大学院生物資源環境科学府環境農学専攻生産環境科学教育コース水環境学研究分野
Laboratory of Water Environment Engineering, Course of Bioproduction Environmental Sciences, Department of Agro-environmental Sciences, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University
九州大学大学院生物資源環境科学府環境農学専攻生産環境科学教育コース水環境学研究分野
Laboratory of Water Environment Engineering, Course of Bioproduction Environmental Sciences, Department of Agro-environmental Sciences, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University
九州大学大学院生物資源環境科学府環境農学専攻生産環境科学教育コース水環境学研究分野
Laboratory of Water Environment Engineering, Course of Bioproduction Environmental Sciences, Department of Agro-environmental Sciences, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University
In this study, a method for analyzing water environment was proposed as a means of environmental assessment for contributing to the preservation and improvement of eutrophic reservoirs in rural areas from the viewpoint of data mining. Thus, water environmental characteristics were determined by periodically observing water quality factors related to eutrophication, such as chlorophyll-a concentration and measuring the levels of algal classes by clustering a database with a self-organizing map (SOM). As a result, datasets containing water temperature, chlorophyll-a from green algae and blue-green algae, total organic carbon, total nitrogen, and total phosphorus were classified into several clusters characterized by mean and maximum and minimum of these 6 factors. Moreover, water temperature and chlorophyll-a from green algae were the most important factors for the feature extraction of the clusters. In addition, the seasonal change in the water environment characteristics was quantitatively estimated by determining the temporal change patterns of the clusters. In conclusion, our results indicated that clustering of the database with SOM was an effective analysis tool for the assessment of the water environment based on the results of detailed observations.