Clustering evaluation of water quality for various classes of in-flow rivers in connected brackish lakes

Sangyeob Kim,Takuto Ando,Yasuhide Nakamura,Shohei Hayashi,Shun Kawaida

DOI: https://doi.org/10.1007/s10661-024-12670-7

2024-05-03

Abstract:Brackish waters and estuaries at the lower reaches of rivers accumulate organic matter and nutrients from various sources in the watershed. Sufficient light and shallow water depth stimulate phytoplankton growth, resulting in a more diversified ecosystem with higher trophic levels. For effective watershed management, it is crucial to characterize the water quality of all rivers, including small and medium-sized ones. Our field survey assessed water quality parameters in 26 inflow rivers surrounding Lakes Shinji and Nakaumi, two consolidated brackish lakes in Japan. The parameters included water temperature, salinity, chlorophyll-a, and nutrients. The study used hierarchical clustering. The Silhouette Index was used to assess clustering outcomes and identify any difficulties in dispersion across clusters. The 26 rivers surrounding Lakes Shinji and Nakaumi were classified into six groups based on their water quality characteristics. This classification distinguishes itself from earlier subjective methods that relied on geographical factors. The new approach identifies a need for improved management of river water quality. The results of the cluster analysis provide valuable insights for future management initiatives. It is important to consider these findings alongside established watershed criteria.