A Local Approach to Hybrid Data Assimilation
Abstract
A hybrid system with the local formulation is developed where the prior probability density function (pdf) consists of both statistical and dynamic information of the uncertainty. The dynamic information is provided by the monte caro approach. The local formulation is solved at every grid as in the local ensemble transform Kalman filter. The formulation is flexible in that it allows not only Gaussian but other stable pdfs.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2014
- Bibcode:
- 2014AGUFMNG23C..04I
- Keywords:
-
- 3315 Data assimilation;
- ATMOSPHERIC PROCESSES;
- 3325 Monte Carlo technique;
- ATMOSPHERIC PROCESSES;
- 1910 Data assimilation;
- integration and fusion;
- INFORMATICS;
- 4468 Probability distributions;
- heavy and fat-tailed;
- NONLINEAR GEOPHYSICS