Long-term forecasting of the average monthly runoff of medium Arctic rivers on the example of the river Pur

Abstract

The article presents two options for long-term forecasting of the average monthly flow of Arctic rivers using the method of group accounting of arguments (MGAA). MGAA allows you to determine the optimal structure of the dependence and determine the parameters. The Pur River, which can be considered as a typical middle Arctic river, was chosen as the object of study. To test the quality of the prediction, the set of observations is split into two parts: a training sequence, which is used to estimate the coefficients by the least squares method, and a test sequence, which is used to evaluate the predictive model. This approach is especially valuable in cases where an integrated approach to planning the use and protection of water resources is needed, as well as when designing water management measures.