Abstract This paper describes the development of a model based on artificial neural networks (ANN) which aims to predict the concentration of nitrates in river water. Another 26 water quality parameters were also monitored and used as input parameters. The models were trained and tested with data from ten monitoring stations on the Danube River, located in its course through Serbia, for the period from 2011 to 2016. Multi-layer perceptron, standard three-layer network is used to develop models and two input variable selection techniques are used to reduce the number of input variables. The obtained results have shown the ability of ANN to predict the nitrate concentration in both developed models with a value of mean absolute error of 0.53 and 0.42 mg/L for the test data. Also, the application of IVS has contributed to reduce the number of input variables and to increase the performance of the model, especially in the case of variance inflation factor (VIF) analysis where the estimation of multicollinearity among variables and the elimination of excessive variables significantly influenced the prediction abilities of the ANN model, r – 0.91.

Paper presents results of researches carried out on various locations and immediate vicinity of mining and industrial activities of the northern and south-eastern part of Kosovo. Concentrations of As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Zn, Rn-222, as well as temperature and pH values of natural spring water were measured at 15 measuring sites (that belong to Zvečan, Leposavić and Novo Brdo municipalities), in April–May and September–October 2019. The quantification of heavy metals’ content was performed by applying ICP-OES method. In analysed samples a high content of As, Pb, Fe and Ni was found. Carcinogenic and non-carcinogenic risks due to the content of heavy metals in water were evaluated. Concentration of radon in water was measured by the alpha spectrometric method, and measured values range in the interval from 0.34 ± 0.12 to 341 ± 35 Bq/L. The yearly doses of inhalation and ingestion were determined for the measured concentrations of radon. Mutual correlation by the Pearson correlation coefficient, principal component analysis, cluster analysis and spatial distribution analysis of the researched parameters of sampled water were done. The most expressed mutual dependence of some heavy metals leads to the conclusion that they have the same anthropogenic origin.

No abstract

In this study, the results of research on radon activity concentrations in natural mineral waters, traditionally used for drinking but also for other needs, in rural and urban households in northern Kosovo are presented. Radon activity concentration in water was measured by the alpha spectrometric method with a RAD7 device. Radon activity concentrations in the 24 waters studied ranged from 1.6 ± 0.5 to 46.3 ± 6.3 Bq l−1, with an average activity concentration of 12.4 ± 2.0 Bq l−1, which was somewhat higher than the EPA recommended maximum activity concentration, but below the WHO recommended maximum. The contribution of radon activity concentrations in water was determined in relation to the total radon activity in air and enclosed space. The estimated annual effective doses of inhalation and ingestion radon from water were 109.4 ± 16.7 and 2.6 ± 0.4 μSv y−1, respectively.

No abstract