SH14B-01 INVITED
Quantifying turbulence in solar magnetic fields: can this help predict solar eruptions?
Magnetic fields in solar active regions present us with a beautiful, although inextricable, complexity, and undergo dynamical evolution that is often far from predictable. This behavior is commonly attributed to the inherently turbulent, filamentary nature of magnetic fields in the solar atmosphere. We ask whether this turbulence and its manifestations can help us predict solar eruptions. To this purpose, we briefly outline the physical relationship between turbulence and (critical) self-organization, as well as their phenomenology, such as spatiotemporal intermittency, self-similar fragmentation, fractality, and multifractality of solar active-region magnetic fields. We also review the array of techniques that have been recently implemented to quantify these turbulent features and we apply them to numerous flaring and nonflaring active regions aiming toward quantitative flare prediction. Results and conclusions are presented in hopes to intrigue and stimulate further discussion on this fascinating, clearly outstanding, problem.
SH14B-02
Assimilative 3D Models of Density and Temperature in the Solar Corona
White-light and extreme ultraviolet images of the solar corona, as measured routinely by a variety of dedicated space- and ground-based instruments, offer an opportunity for empirical determination of the global, 3D distribution of density and temperature in the Sun's corona. In this work, we describe a 3D model for the estimation of coronal density from polarized brightness measurements and a coronal temperature model based on differential emission measure tomography. The computational solutions of the associated inverse problems, which utilize LASCO, MK-4, and EIT measurements at different solar rotation angles, yield reliable reconstructions of persistent, large-scale structures. The characterization of transient disturbances responsible for space weather phenomena, however, demands new developments in data assimilation and statistical estimation theory. We present a state-space framework capable of dynamically estimating the time-varying state of the corona. The 3D, time-dependent nature of the estimation scheme demands algorithms that scale well with the problem size. We describe recursive estimation techniques which dramatically reduce computational complexity and enable data-assimilative global models of the solar corona. Finally, implications for data assimilation with STEREO are discussed.
SH14B-03
Thermally peeling the Corona
The outer atmospheres of most low-mass stars, including the Sun, are composed of very hot plasma (1-50 MK) which is organized in spatially and thermally complex structures. A proper determination of these structures is necessary to decide the energetics of coronae, to establish their compositions, and to distinguish between different physical processes that may operate on them. The challenge of solar and stellar astrophysicists is thus to determine the temperature structure reliably, i.e., to establish how much of the observed intensity arises at what temperature. The data here are comprised of spectral lines from highly ionized species of elements such as Fe, Ne, O, etc. By measuring the intensities in specific lines, and calculating their emissivities from known atomic data, we can infer the shape of the underlying emission measure distribution. However, straightforward inversion solutions are subject to high-frequency instability, and we must carry out forward-fits to carry out the inference. We describe some of the challenges posed by this problem and discuss different methods of solutions, primarily based on a Markov-Chain Monte Carlo method.
SH14B-04
A New View of the Extreme Ultraviolet Corona from Wavelet-Processed EUV Images
Our knowledge of the structure and dynamics of the extreme ultraviolet solar corona has greatly increased over the last 11 years thanks to the observations from the Extreme-ultraviolet Imaging Telescope (EIT) aboard the SOHO spacecraft. The EIT images have revealed the early phases of coronal mass ejections (CMEs), discovered coronal waves associated with CMEs, recorded impressive post-CME loop systems and eruptive prominences, and detected reconnection at the base of coronal hole plumes, among other things. It would be natural to think that, by now, the EIT instrument has exhausted its discovery potential. We will demonstrate in this presentation that this is not the case. We have developed a wavelet-based image enhancement technique that exploits the multi-scale nature of the observed solar features, and treated the entire EIT database accordingly. The technique allowed us to remove the instrumental stray light background and enhance the fine coronal structures at the same time. The final images reveal such a wealth of structures and dynamics that they seem to have been obtained by a new instrument. The clarity of the enhanced images allows us to identify numerous and potentially interesting phenomena that were previously obscured by a background level that includes stray light and image noise. A few examples will be presented here, including application to the STEREO EUVI images. This presentation aims to bring the availability of this resource and its potential for significant discoveries to the attention of the solar physics community.
SH14B-05
Optimization Approach for the Computation of 3D Magnetohydrostatic Coronal Equilibria From Multi-Spacecraft Observations
We cannot measure the 3D coronal magnetic field and plasma pressure/density distribution directly. To derive these quantities we propose a modelling approach based on observational data from multiple instruments. Our aim is to use measurements of the photospheric magnetic field vector (e.g. from Hinode/SOT and in future from SDO/HMI) and plasma images from two viewpoints -as provided by STEREO- as input for a newly developed magnetohydrostatic optimization code. The resulting 3D magnetic field and plasma distribution is a self-consistent equilibrium within the magnetohydrostatic approach. Here we test our code with the help of an exact magnetohydrostatic equilibrium and extracted synthetic observational data, which allow us to evaluate the accuracy of our method. We find that the method reconstructs the equilibrium accurately, with residual forces of the order of the discretisation error of the exact solution. The correlation with the reference solution is better than 99.9 percent and the magnetic energy is computed accurately with an error of less than 0.1 percent. We are planning to use this method with real observational data as input as soon as possible.
SH14B-06
New Forecasting Factor for Solar Wind Velocity From EIT Observations
Various solar wind velocity forecasting methods at 1AU have been developed during the last decade, such as Wang-sheeley model and Hakamada-Akasofu-Fry Version 2 (HAFv2) model. Some authors have found that Coronal hole(CH) areas can be used to forecast the solar wind velocity with better results in low CME activity periods(e.g. Vršnak et.al.). The property of the solar surface is a good indication of the following interplanetary and geomagnetic activities. We analyzed all EIT284Å images in almost the whole solar cycle 23 and developed a new forecasting factor(Pch) from the brightness of the solar Extreme Ultraviolet Images. and a good relationship was found between the Pch and solar wind velocity V three days later probed by ACE spacecraft. A simple method of forecasting the solar wind speed near earth in low CME activity periods is presented. For Pch and solar wind velocity, the linear correlation coefficients is R = 0.89 from 21 September until 26 December. For comparison we also analysed the same period data as Vršnak(2007) who using the coronal hole areas AM as input parameters for predicting solar wind velocity. The linear least-squares fit of Pch with the 3-day lag solar wind velocity showed a correlation coefficient R = 0.70, which is better than the result using AM(R = 0.62). The solar wind speed could be expressed as V (km s-1) = 337 + 0.00868 × Pch. The average of relative difference between the calculated and the observed values amounts to |δ̄| ≈ 12.15%. Furthermore, for the ten peaks during the analysis period, AM and V just showed a correlation coefficient R = 0.32, much worse than using Pch factor which showed R = 0.75. Moreover, the Pch factor exterminated personal bias in the forecasting process, which existed in the method using AM as input parameters because the coronal hole boundary can not be easily determined since no quantitative criteria can be used to precisely locate coronal holes from observation. Finally, the expression of V by Pch is analysed, which showed the variation of background solar wind speed during the whole solar cycle 23.
SH14B-07
Mutual Information as a Non-linear Measure of Correlation in Multi-point Measurements in the Solar Wind
With increasing numbers of spacecraft in the solar wind new techniques for analyzing multi-spacecraft data are in demand. We discuss techniques for implementing mutual information and present results obtained for correlation of fluctuations in magnetic field, proton density and velocity between the WIND and ACE spacecraft at the minimum and maximum of the last solar cycle. The magnetic field, particle density and velocity measurements, as well as derived quantities such as Els\"{a}sser variables, are compared between ACE and WIND and correlation length scales are calculated. Linear cross correlation is also used as a comparison, indicating the degree to which non-linear and linear correlations differ, and the effect of the solar cycle is investigated. Mutual information shows a clear difference between the density and magnetic field measurements, which is harder to discern from the linear correlation. The effect of the solar driver is also clear in the mutual information measurements and the increase in scale length due to either solar coronal structure imposed on the solar wind, or non-universal turbulence is measured.