The evaluation of the spatial and temporal distribution of pollutants is a crucial issue to assess the anthropogenic burden on the environment. Numerous chemometric approaches are available for data exploration and they have been applied for environmental health assessment purposes. Among the unsupervised methods, Self-Organizing Map (SOM) is an artificial neural network able to handle non-linear problems that can be used for exploratory data analysis, pattern recognition, and variable relationship assessment. Much more interpretation ability is gained when the SOM-based model is merged with clustering algorithms. This review comprises: (i) a description of the algorithm operation principle with a focus on the key parameters used for the SOM initialization; (ii) a description of the SOM output features and how they can be used for data mining; (iii) a list of available software tools for performing calculations; (iv) an overview of the SOM application for obtaining spatial and temporal pollution patterns in the environmental compartments with focus on model training and result visualization; (v) advice on reporting SOM model details in a paper to attain comparability and reproducibility among published papers as well as advice for extracting valuable information from the model results is presented.

The water quality assessment of the surface water bodies (SWBs) is one of the major tasks of environmental authorities dealing with water management. The present study proposes a water quality assessment scheme for the investigation of the surface waters’ physicochemical status changes and the identification of significant anthropogenic pressures. It is designed to extract valuable knowledge from the Water Frame Directive (WFD) mandatory monitoring datasets. The water quality assessment scheme is based on the Canadian Council of Ministers of the Environment water quality index (CCME-WQI), trend analysis of estimated WQI values, and Principal Component Analysis (PCA) using calculated excursions during the determination of WQI values. The combination of the abovementioned techniques preserves their benefits and additionally provides important information for water management by revealing the latent factors controlling water quality, taking into account the type of the SWB. The results enable the identification of the anthropogenic impact on SWBs and the type of the corresponding anthropogenic pressure, prioritization and monitoring restoration measures, and optimization of conducted monitoring programs to reflect significant anthropogenic pressures. The proposed simple and reliable assessment scheme is flexible to introducing additional water quality indicators (hydrological, biological, specific pollutants, etc.), which could lead to a more comprehensive surface water quality assessment.

Deterioration of water quality is a major problem world widely according to many international non-governmental organizations (NGO). As one of the European Union (EU) countries, Bulgaria is also obliged by EU legislation to maintain best practices in assessing surface water quality and the efficiency of wastewater treatment processes. For these reasons studies were undertaken to utilize ecotoxicological (Microtox^®, Phytotoxkit F^TM, Daphtoxkit F^TM), instrumental (to determine pH, electrical conductivity (EC), chemical oxygen demand, total suspended solids (TSS), total nitrogen (N) and phosphorus (P), chlorides, sulphates, Cr, Co, Cu, Cd, Ba, V, Mn, Fe, Ni, Zn, Se, Pb), as well as advanced chemometric methods (partial least squares–discriminant analysis (PLS-DA)) in data evaluation to comprehensively assess wastewater treatment plants’ (WWTPs) effluents and surface waters quality around 21 major Bulgarian cities. The PLS-DA classification model for the physicochemical parameters gave excellent discrimination between WWTP effluents and surface waters with 93.65% correct predictions (with significant contribution of EC, TSS, P, N, Cl, Fe, Zn, and Se). The classification model based on ecotoxicological data identifies the plant test endpoints as having a greater impact on the classification model efficiency than bacterial, or crustaceans’ endpoints studied.