Abstracts submitted by Farzad Kamalabadi

Data Assimilation Opportunities and Challenges for Investigating Transient Phenomena

Farzad Kamalabadi[1]; Mark Butala[1]; Richard Frazin[1]; Yuguo Chen[1]

[1] University of Illinois

Knowledge of the corona's three-dimensional (3D) structure is important for modeling the propagation of solar disturbances out to the near-Earth space environment. The current generation of Sun-observing spacecraft give the scientific community a historical opportunity to unravel unprecedented information about the 3D structure of the Sun's corona. However, full advantage of these sensing capabilities demands the concurrent development of next generation multi-sensor data processing strategies and algorithms.

Examples of such data include white-light and extreme ultraviolet (EUV) images of the solar corona, as measured routinely by a variety of dedicated space-based and ground instruments. Although Bayesian tomographic analysis of these data have produced reliable 3D reconstructions of persistent, large-scale structures, characterization of transient disturbances responsible for space weather phenomena requires developments in data assimilation, statistical estimation theory, and efficient algorithm design.

In this talk, I will describe a state-space framework through which one can estimate the state of the corona as a function of time. Since the 3D, time-dependent nature of the estimation scheme demands new algorithms that scale well with the problem size, I will describe new Monte Carlo filtering techniques which make use of recursive estimation and optimal filtering, in conjunction with new spatial localization concepts for line-integral measurement operators, that dramatically reduce computational complexity and enable global assimilative models of the solar corona.