Simultaneous ensemble-based state and parameter estimation for earth systems

Fuqing Zhang
Pennsylvania State University
Seminar Date: 
28. September 2018 - 11:00 - 12:00
Cinema, Ground Floor, NERSC

We seek to develop and apply a generalized data assimilation framework using Ensemble-based Simultaneous State and Parameter Estimation (ESSPE) that will facilitate data-model integration and uncertainty quantification for the broad weather, climate and earth science communities. Through augmenting uncertain model parameters as part of the state vector, the ESSPE framework allows for simultaneous state and parameter estimation using an ensemble Kalman filter (EnKF) through assimilating large-volume in-situ and remotely sensed heterogeneous observations such as those from radiosondes, radars and satellites. The ESSPE framework can be applied to identify key physical processes and their impacts, to better represent and parameterize these processes in dynamical models, to design better observation strategies, to understand predictability and nonlinearity of these processes, and to facilitate generalization of the knowledge from smaller-scale process understanding to larger- and system-scale impacts and parameterizations.