Flares I Posters
Presiding: J Zhang, George Mason University; S P Plunkett, Naval Research Laboratory
SP21A-01 0830h
Energetics of RHESSI X-Class Flares
The thermal and nonthermal energies of several X-class flares seen with the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) will be presented. The same techniques described by Emslie et al. (JGR, 109, A10104, 2004) are used to take the RHESSI imaging spectroscopic observations and compute the energies in the thermal plasma and in the nonthermal electrons as a function of time throughout the flares. Radiative and conductive cooling rates are estimated and total thermal and nonthermal energies are computed for each flare. Typically, the energy in nonthermal electrons integrated up to the time of peak soft X-ray emission is equal to or exceeds the energy in the thermal plasma at that time. This suggests that energy must have been converted into a form not visible with RHESSI and that the total energy released by the flares may be significantly greater than the sum of energies calculated from the RHESSI observations alone. This conclusion is supported by the high radiative energy seen with SORCE during the impulsive phase of the 28 October 2003 flare. The peak increase in total solar irradiance of 270 mW m-2 measured with SORCE was over two orders of magnitude higher than the peak soft X-ray flux seen with GOES or RHESSI. The implications of this new observation as compared to the energetics derived from the X-ray observations of that flare will be discussed along with the energetics analysis of most of the other X-class flares in October/November 2003.
SP21A-02 0830h
Transition Region Emission and the Energy Input to Thermal Plasma in Solar Flares
Understanding the energetics of solar flares depends on obtaining reliable determinations of the energy input to flare plasma. X-ray observations of the thermal bremsstrahlung from hot flare plasma provide temperatures and emission measures which, along with estimates of the plasma volume, allow the energy content of this hot plasma to be computed. However, if thermal energy losses are significant or if significant energy goes directly into cooler plasma, this is only a lower limit on the total energy injected into thermal plasma during the flare. We use SOHO UVCS observations of O VI flare emission scattered by coronal O VI ions to deduce the flare emission at transition region temperatures between 100,000 K and 1 MK for the 2002 July 23 and other flares. We find that the radiated energy at these temperatures significantly increases the deduced energy input to the thermal plasma, but by an amount that is less than the uncertainty in the computed energies. Comparisons of computed thermal and nonthermal electron energies deduced from RHESSI, GOES, and UVCS are shown.
SP21A-03 0830h
Temporal Evolution of Hard X-ray and UV Emission Sources in Solar Flares
In this work, we investigate, for a number of solar flares, the evolution of both UV hard x-ray emission as observed by TRACE and RHESSI, respectively. The spatial resolution of these two instruments provides an unprecedented capability to study the individual source regions producing the observed time profiles in these events. Confirming earlier results, it is found that individual bursts in the hard X-ray time profiles are correlated with the unresolved UV emission. However, the temporal evolution of individual sources provides a deeper insight into the flare energization process. We report a number of results: 1) the UV-to-UV temporal correlation analysis allows us to identify sources which exhibit similar time profiles and therefore may indicate conjugate footpoints of a large coronal loop, 2) the UV emission is distributed over a wider area than the hard X-ray emission, 3) the bulk of the hard X-ray emission would seem to be related to a different structure within the flaring volume but sharing a common footpoint with the UV emission, and 4) the temporal and spatial behavior observed would seem to indicate an interacting loop scenario for the flares considered.
SP21A-04 0830h
Center to Limb Variation of Hard X-ray Spectra from RHESSI
In this work we use the massive RHESSI flare database to measure the center to limb variation of the hard X-ray spectral index in the 50 to 100 keV energy range and the 100 to 300 keV energy range. If Coulomb collisions are the dominant scattering mechanism for accelerated electrons in solar flares, then there should be some variation in the steepness of the spectrum, the spectral index, from the center of the sun to the solar limb. The issue is complicated by the presence of backscatter of the radiation and the large spread in observed spectral index with respect to the expected center to limb variation. Previous work (Vestrand, et al, 1987, Apj 322, 1010) has shown that such a variation exists at high energies (300 to 1000 keV), but not necessarily at lower energies (Li, 1994, Apj 421, 381). It is also true that direct comparisons of energy spectra using multiple instruments have never found any noticeable anisotropy in solar flare X-rays (e.g., Kane, et al, 1988, Apj 326, 1017). The large number of flares and excellent spectral resolution of RHESSI should enable us to detect any center to limb variation of the spectra if it is present. This research is supported by NASA contract NAS5-98033.
SP21A-05 0830h
Comparison of Algorithms for Reconstructing Electron Spectra from Solar Flare Hard X-Ray Spectra
The Ramaty High Energy Solar Spectroscopic Imager (RHESSI) is yielding solar flare hard X-ray (HXR) spectra with unprecedented resolution and precision. Such spectra enable the reconstruction of the effective mean source electron spectrum F̅(E) by deconvolution of the photon spectrum I(ε) through the bremsstrahlung cross-section Q(ε,E). In this paper we report on an evaluation of three distinct "inverting" reconstruction techniques and one forward fitting procedure. We synthesized a variety of hypothetical F̅(E) forms, with a variety of empirical features designed to represent diagnostics of electron acceleration and transport processes, generated the corresponding I(ε) with realistic random noise added, and performed "blind" (i.e. without knowledge of F̅[E] in advance) recoveries of F̅(E) for comparison with the originally assumed forms. In most cases the inversion methods gave very good reconstructions of F̅(E). The forward fitting method did well in recovering large-scale features but, somewhat inevitably, failed to recover features outwith the parametric forms of F̅(E), such as dips, bumps and positive slopes. However, examination of the distribution of photon spectrum residuals over ε should in principle permit refinement of the parametric form used.
SP21A-06 0830h
Photon-Flux Saturation in Large Solar Flares
We use the spectral and spatial resolution of RHESSI to explore the behavior of electron fluxes and their associated currents in large solar flares, including the Halloween events of October-November 2003. The incident electron spectra at the flaring footpoints are derived from the RHESSI photon spectra. Spectral images are used to determine an upper limit as well as a spectrally-averaged estimation for the footpoint area. We find that, over a wide range of flare X-ray magnitudes, the integrated fluxes above 20 keV asymptotically approach a limiting value, suggesting some form of saturation in the particle production in flares. We discuss the implications of these results in terms of the assumed model of a stable non-thermal electron beam.
SP21A-07 0830h
Numerical Solution of the 2-D Momentum Diffusion Equation
The time-dependent momentum diffusion equation describes the evolution of a particle distribution function in response to resonant wave-particle interactions with plasma turbulence. As such, it is central to treatments of stochastic particle acceleration and transport in space and astrophysical plasmas. In either cylindrical (p∥,p⊥) or spherical (p, pitch-angle cosine Μ) momentum coordinates, this equation contains a mixed partial derivative, which is highly unstable to numerical finite difference schemes. This in turn precludes the use of many common numerical solution techniques, such as operator splitting or the ADI method. It is for this reason that the momentum diffusion equation is almost always averaged over one degree of freedom, in order to yield a more tractable 1-D equation (typically the pitch-angle averaged momentum diffusion equation, or equivalently the Fokker-Planck equation in energy). Instead, we present a solution that employs stochastic differential equations, which do not suffer from the above numerical instabilities, and which permit us to solve the full 2-D equation without approximation or averaging. The biggest obstacle with this method is taking, basically, the square root of a matrix; however, this can be dealt with effectively using Mathematica. We present results for the case of ions cyclotron resonating with Alfvén waves, and discuss how this numerical method can be easily generalized to include the effects of spatial transport or static electric fields. This work was supported by NASA grant NAG5-12824.
SP21A-08 0830h
Stochastic Ion Acceleration in 3He-Rich Solar Energetic Particle Events
The enhancements of heavy ions in impulsive solar energetic particle (SEP) events have long been seen as evidence for stochastic acceleration (SA) of particles by plasma wave turbulence. However, the lack of quantitative comparison of model predictions with observations prevents a solid conclusion about the underlying physical processes. By using the exact dispersion relation and taking into account the effects of background α-particles, we show that SA by parallel propagating waves can address the long standing challenge to SA models, namely the prominent enhancement and distinct spectrum of 3He ions in some impulsive solar flares. Encouraged by this and in light of recent detections of ultra-heavy ions in SEPs by ACE and WIND, we investigate the SA of all ions by parallel propagating waves. The observed ion spectra and enhancements are used to constrain the turbulence spectrum and energy partition among different wave branches. Based on this scenario, we explore the model parameter space to explain the recently discovered correlation between ion enhancements and the iron spectral index and the correlated enhancements of heavy ions. The limitations of the model are described and possible solutions to these difficulties are discussed. The work at Stanford is supported by NSF grant ATM-0312344, NASA grants NAG5-12111, NAG5 11918-1, and NASA grant PC 251429 through University of Maryland (PI: G. M. Mason).
http://www.stanford.edu/~liusm/post.html
SP21A-09 0830h
The Physics of Positron Annihilation in the Solar Atmosphere
We have developed an algorithm that calculates the 511-keV annihilation line spectrum and relative strength of the 3-gamma continuum for a wide range of physical environments from the flaring solar atmosphere to the interstellar medium. We present evaluated cross sections for each of the processes relevant to the positrons and their annihilation based on the most recent measurements and calculation. The calculation begins with a Monte Carlo code that follows positrons as they slow down to thermal energies or form positronium in flight. We then calculate the thermally-averaged annihilation and positronium production rates, and the positronium quenching rates. We apply the results to four specific environments (fully ionized, neutral, partially ionized and a non-LTE model of the quiet solar atmosphere), and calculate the relative strengths of each process and the combined total annihilation line spectrum. The results are compared with data obtained recently from the high spectral-resolution detectors of RHESSI. We find that positron annihilation in solar flares can occur in a wide variety of environments that can be quite different from currently-accepted solar atmospheres.
http://heseweb.nrl.navy.mil/gamma/solar/highe.html
SP21A-10 0830h
TRACE and SOHO/MDI Observations of Five X-Flares from AR10720, the "Fireworks Active Region", on January 15-20, 2005
Both TRACE and SOHO/MDI observed the five X-flares that were produced by AR10720 during January 15-20, 2005. This active region appears to be "special" because it is occurring so late in the solar cycle, about two-thirds of the way toward solar minimum. Also its structure is initially one large penumbral region with several umbral spots that are being stretched out east to west with rotating sunspots emerging in leading and trailing positions during the transit across the solar disk, similar to fireworks with pin-wheels attached. With a neutral line running almost east-west in between two of the stretched umbral regions and with opposite magnetic polarities on both sides of the neutral line adjacent to each other, the beta-delta region starts the "fireworks" with the first X-flare on January 15, continuing every one to one and one-half days over the next five days. Images and movies will be shown of these eruptive events in white light and 1600A from TRACE, and as magnetograms from MDI. Two of the three rotating sunspots also change their direction of rotation during this period. These X-flares add to those observed by TRACE since its launch in April 1998, all of which have been associated with rotating sunspots in the associated active regions. These X-flares consist of over half of the X-flares observed by GOES since April 1998 and many are associated with geo-effective CMEs in the form of solar energetic particle events. Rotating sunspots may be a good predictor of solar eruptions. This work has been supported by NASA, in part under the TRACE project at LMSAL (contract NAS5-38099) and in part under the MDI/SOHO project (NAG5-13261).
SP21A-11 0830h
Improving Flare Simulations by Combining Hydrodynamics Modeling with Stochastic Particle Transport
Currently flare models include heating due to non-thermal particle collisions as analytical, and often static, inputs to numerical hydrodynamics codes. While this has aided in our understanding of flare dynamics, it does not adequately represent the complexity of the interaction between thermal and non-thermal particles in a flaring region and cannot completely describe multi-wavelength observations of solar flares. In order to improve the current state-of-the-art flare simulations, we are combining thermal and stochastic non-thermal numerical modeling codes in order to simulate flare emission under a variety of theoretical and realistic solar conditions. The theoretical flare emission is then folded through the response functions of solar observatories in order to provide theoretical data that can be compared to observational results. This provides a means to test multiple flare models with observed flare behavior. This work was supported by NASA grant NAG5-12820
http://solar.physics.montana.edu
SP21A-12 0830h
Measurement and Interpretation of X-ray Visibilities with RHESSI
The Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) uses a set of 9 rotating modulation collimators to image solar flares at x-ray and gamma-ray energies. RHESSI's imaging information is encoded in the rapid time modulation (~2 to ~500 Hz) of the detected flux. This poster discusses two new approaches to converting this modulated time profile into images. Currently telemetered time-tagged photon data are presented to imaging algorithms as counts in a series of very short time bins. For long integrations, this approach has the disadvantage that the number of time bins can become very large ((>106)) and unwieldy. A new adaptation of folded epoch analysis is described that permits the "stacking" of data over multiple rotations. This can be done despite the distortion of the observed modulated time profile caused by variations in spacecraft pointing. For long integrations, stacking can dramatically compress the observational input to the imaging algorithms with little or no loss of imaging information. A second new development is the conversion of the stacked data into complex visibilities. Visibilities (which represent Fourier components of the source distribution) are determined from the amplitude and phase of short segments of the modulated x-ray waveform. They represent a compact, background-subtracted and fully-calibrated data product that is intermediate between the observed photon stream and images. Visibilities can be converted to images using reconstruction algorithms developed for radio interferometry. This poster also illustrates how RHESSI visibilities can be directly and quantitatively interpreted in terms of x-ray source size, shape and morphology.