Conditions for successful data assimilation
- Department of Mathematics, University of California, Berkeley California USA; Lawrence Berkeley National Laboratory, Berkeley California USA
We show, using idealized models, that numerical data assimilation can be successful only if an effective dimension of the problem is not excessive. This effective dimension depends on the noise in the model and the data, and in physically reasonable problems, it can be moderate even when the number of variables is huge. We then analyze several data assimilation algorithms, including particle filters and variational methods. We show that well-designed particle filters can solve most of those data assimilation problems that can be solved in principle and compare the conditions under which variational methods can succeed to the conditions required of particle filters. We also discuss the limitations of our analysis. Key Points Data assimilation is feasible only if an effective dimension is moderate. The effective dimension is moderate, if the noise satisfies a balance condition. Particle filters can work, even if the dimension of the problem is large.
- Research Organization:
- Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- AC02-05CH11231
- OSTI ID:
- 1581110
- Journal Information:
- Journal of Geophysical Research: Atmospheres, Vol. 118, Issue 20; ISSN 2169-897X
- Publisher:
- American Geophysical Union
- Country of Publication:
- United States
- Language:
- English
Similar Records
Realizable hyperuniform and nonhyperuniform particle configurations with targeted spectral functions via effective pair interactions
Early season prediction of within-field crop yield variability by assimilating CubeSat data into a crop model