The paper presents the results of comparing the application of two different data assimilation methods. the method of generalized Kalman filtration (GKF) developed by the authors and standard objective ensemble interpolation method (EnOI) that is a partial case of extended Kalman filter scheme (EnKF). Those methods are applied in conjunction with the ocean dynamical model Hybrid Circulation Ocean Model (HYCOM). The methods are compared with respect to various criteria, in particular, with respect to minimum of the forecast error and with respect of a posterior error over a given time-interval. As observed data we used the Archiving Validating and Interpolating Satellite Observation (AVISO) i.e. altimetry data. It is shown that the method GKF has a number of advantages comparing with the method EnOI.

ocean modelling, data assimilation, generalized Kalman filter, ensemble interpolation method, satellite altimetry data

Mathematical modelling in actual problems of science and technics

Belyaev Konstantin Pavlovich,  kosbel55@gmail.com,  Институт океанологии им. П.П. Ширшова РАН

Kuleshov Andrey Aleksandrovich,  andrew_kuleshov@mail.ru,  orcid.org/0000-0002-4203-9953,  KIAM RAS

Smirnov Ilya Nikolaevich,  ismirnov@cs.msu.ru,  Московский государственный университет имени М.В. Ломоносова