THE STRUCTURE OF THE VARIABILITY OF THE HYDROCHEMICAL COMPOSITION OF WATER IN LAKE-TYPE RESERVOIR

A qualitative and quantitative analysis of hydrochemical parameters of water is given in order to identify the factors that determine their spatial and temporal changes in a lake-type reservoir. Water samples were taken in 2019 and 2020 from the average level in spring (April), summer (July) and autumn (September) in the first week of the month in accordance with the requirements of GOST R 51592-2000 in three sections. The first target (1) is the shallow upper part (depth from 2 to 4 m); the second target (2) is the central part (depth from 5 to 7 m) and the third target (3) is the near – dam part (depth up to 12.2 m). Statistical analysis of the obtained data was performed using the unlimited Principal component analysis (PCA) technique and the limited redundancy analysis (RDA) technique. The effects were considered statistically significant at P<0.05, and useful for discussion-at P<0.10. It was found that, despite the flood increase in the level of chemical components in the water of the reservoir, most of them meet the requirements for fishing waters, with the exception of iron, copper, manganese, zinc, nickel and lead, which exceed the MPCVR from 1.1 to 45.0 times. The total variability of the hydrochemical composition of water in the reservoir, estimated by the PCA method, depends on the season of the year by 71.4 %. A similar result was obtained by the RDA method in a model with a single regressor. When all factors are taken into account in the RDA model, the variability of the water chemical composition is affected by the season of the year by 74.3 %, the year of research by 11.1 %, and the location of the target by 1.9 %. The primary indicators of water for the proportion of unexplained variability in both the PCA and RDA methods are manganese, bicarbonates, lead and aluminum, and pH.