V.S. Kochergin, S.V. Kochergin
Marine Hydrophysical Institute of RAS, Sevastopol, Russia
Abstract
At present, variational assimilation methods and the method of conjugate equations are actively developed and used to solve oceanological problems. The assimilation algorithms of the measurement data are based, as a rule, on minimizing the quadratic functional of the forecast quality characterizing deviations of the model solution from the measurement data. The passive impurity transfer model acts as a constraint on the variation of input parameters. In this study, we consider a variational method for identifying the power of a point instant source of pollution based on solving conjugate problems and problems in variations. Numerical experiments were carried out using the hydrodynamic model of the Kerch Strait. The obtained flow fields were used to model the transport of passive impurities. Numerical experiments have shown that when using variational algorithms for identifying the power of an instantaneous point source, one iteration is sufficient. Additional iterations are necessary when identifying variables by space and time of the parameters of the pollution source. When a variational identification algorithm is implemented in the case of noisy measurement data, they are filtered. In general, numerical experiments showed the reliable operation of the considered algorithms for identifying the power of a pollution source, as applied to the model of passive admixture transport in the Kerch Strait.
Keywords
variational method, identification of input parameters, transport model, passive admixture, pollution propagation, assimilation of measurement data, the Kerch Strait
DOI
10.22449/2413-5577-2019-3-89-96