[G04-2] Earth rotation and geodynamics II

[G04-2-03] EOP prediction based on the Copula method using multi-source data

^○Sadegh Modiri^{1, 2}, Robert Heinkelmann¹, Santiago Belda³, Jose M. Ferrandiz³, Harald Schuh^1,2 (1.GFZ German Research Centre for Geosciences, Potsdam, Germany, 2.Technische Universitat Berlin,Institute for Geodesy and Geoinformation Science, Berlin, Germany, 3.University of Alicante, EPS II, Applied Mathematics Alicante, Spain)

Prediction of the Earth Orientation Parameters (EOP) is provided by the International Earth Rotation and Reference Systems Service (IERS) Rapid Service Prediction Centre in USNO, Washington DC. Different methods have been developed and applied for the EOP prediction. However, the accuracy of EOP prediction still is not satisfactory even for a few days in the future. Therefore, new methods or a combination of the existing methods need to be investigated for improving the accuracy of the predicted EOP. Thus, the stochastic methods which analyze and exploit the dependency structure between multivariate data should be studied due to the stochastic nature of the EOP. There is a well-introduced method called Copula and we want to apply it for EOP prediction. The Copula method exploits linear and non-linear dependency between variables and it is a very powerful and efficient tool for dealing with multidimensional data and modeling the relation between parameters.
In this study, we analyze the impact of mass redistribution and movement within the Earth system e.g. solid Earth, atmosphere, ocean, hydrosphere, and cryosphere on EOP to get a more precise and reliable forecasting model. Our preliminary studies illustrate that Copula can be applied for capturing the dependency structure between EOP and angular momentum data. Finally, EOP will be predicted more accurately based on the derived dependency structure by evaluating the conditional distribution function given by the Copula model in the appropriate resolutions.