Water quality assessment of a river catchment by the composite water quality index and self-organizing maps

The river catchment management is a complex problem, which requires analysis of massive amounts of mandatory monitoring data on analytical, spatial, and temporal level. The present study deals with the application of a novel water quality assessment approach combining water quality index (WQI) developed by the Canadian Council of Ministers of the Environment (CCME) and self-organizing maps of Kohonen (SOM). The study is carried out by using long-term water quality monitoring data (2008-2018) collected from 10 sampling stations located in the Mesta River catchment, Bulgaria. In the first step, the assessment of annual water quality categories of each sampling station was estimated by the calculation of CCME WQI modified according to the Bulgarian legislation. In the next step, all annual water quality categories (n = 105) characterized by the factors used for CCME WQI calculation (scope, frequency, amplitude) underwent SOM analysis. The results of implementation of the WQI factors in SOM calculations reveal similar groups of annual water quality categories with specific CCME WQI patterns among the whole river catchment. The obtained patterns represent different water quality situations, which could be treated by different management activities. Additionally, the visualization advantages of the SOM algorithm allowed for detection of the water quality temporal changes and similarity between sampling stations among river catchment. Thus, the combination between integral CCME WQI assessment and multivariate technique SOM made it possible to achieve a precise river catchment water quality assessment, which could be used by water management authorities.

Automated summary

This study combines CCME WQI and SOM techniques to assess water quality in the Mesta River catchment, revealing similar water quality patterns among the 105 categories analyzed. The SOM algorithm aids in detecting temporal changes in water quality and similarities between sampling stations in the river catchment.