Adaptive algorithm for global carbon dioxide data assimilation

Abstract: The carbon dioxide concentration in the atmosphere plays an important role in climate change. Ensemble Kalman filter (EnKF) is a popular tool for atmospheric and oceanographic data assimilation. In this talk, I will first briefly introduce a theoretical framework of the global carbon flux and the ensemble Kalman filter for data assimilation. Then I will propose adaptive estimation procedures based on the maximum likelihood and least squares principles. The proposed methods are tested on the popular Lorenz-96 and Lorenz-3 models. The simulation results show that the forecast error covariance matrix is effectively corrected and the true system states are better retrieved.

Location:

MS 431 Math Science Building

Speaker:

Dr. Liqun Wang, Department of Statistics, University of Manitoba

More information at http://www.ucalgary.ca/events/calendar/adaptive-algorithm-global-carbon-dioxide-data-assimilation