Long-term spatiotemporal variability of ionic runoff of rivers in the Northern Dvina river basin

The article presents the results of the study dealing with the spatiotemporal variability of ionic runoff within the Northern Dvina River basin. Based on long-term hydrological and hydrochemical data collected from the state observation network of Roshydromet between 1990 and 2017, the values of ionic runoff and the modulus of ionic runoff have been calculated, both for the sum of ions and individually for specific components, namely sulfates, chlorides, bicarbonates, and calcium ion s. Bicarbonates dominated in the structure of ionic runoff, accounting for up to 50%. The ionic runoff of the Northern Dvina River reached its highest average long-term values at the outlet section near the village of Ust-Pinega; among the tributaries, the highest absolute values of ionic runoff were characteristic of the Vaga and Sukhona rivers. Based on the analysis of the Kendall rank correlation coefficient values, statistically significant trends in the changes of ionic runoff have been identified for the rivers of the Northern Dvina River basin. It is shown that the long-term dynamics of ionic runoff is characterized by a decreasing trend only for certain sections of the Northern Dvina and Sukhona rivers; for most other river sections, temporal changes in ionic runoff are statistically insignificant. Analysis of the identified trends in the runoff of individual macrocomponents revealed that most of them are decreasing and have a moderate strength of association. Increasing trends were observed only for the runoff of bicarbonates in the Luza and Sysola rivers and for sulfates in the Sukhona River. Based on a comparison of runoff modulus, the ionic runoff load on various sections of the Northern Dvina River basin has been evaluated. It was revealed that the highest load of the main ions runoff is characteristic of the Vym River, and the lowest for the Sysola River. For most of the studied rivers, the load in terms of the ionic runoff modulus is estimated as average.