Modeling of River Run-off in the Different Landscape Conditions on Plain Part of the European Part of Russia

Author(s):  E.A. Vedeneeva, Kazan Federal University, Kazan, Russia,

Issue:  Volume 42, № 4

Rubric:  Earth sciences

Annotation:  The paper reveals the mathematical modeling of river run-off modulus in the different landscape conditions on plain part of the European part of Russia. It describes the application of methods such as the generalized linear model (GLM) and the generalized additive model (GAM) to establishing of quantitative links between the river run-off modulus and the totality of its controlling factors. Regression models are successfully used for study river run-off on the local regional and global scale. The object of the research covers the river run-off in the small river basins. Geoinformation base «River basins of the European part of Russia» was used as input data. The construction of models was preceded by an analysis of the structure of explanatory variables, as well as the study of their mutual correlation. Calculations were carried out using programs written in the statistical environment R. In the end of the study, the author gives results of modeling river run-off. Nonlinear methods of modeling give better results compared to linear methods. The constructed models explain about 45-70 % of the variability of the data and well reflect conditions of river run-off formation. The best model was obtained for the steppe zone.

Keywords:  river run-off, geoinformation base, landscape zone, modeling, multiple regression.

Full text (PDF):  Download

Downloads count:  353