The structure and evolution of the sources of solar activity directly affects the nature of space weather disturbances that reach the Earth. We have previously demonstrated that the loss of equilibrium and partial ejection of a coronal magnetic flux rope matches observations of coronal mass ejections (CMEs) and their precursors.In this paper we discuss the significance of such a partially-ejected rope for space weather. We will consider how the evolution and bifurcation of the rope modifies it from its initial, source configuration. In particular, we will consider how reconnections and writhing motions lead to an escaping rope which has an axis rotated counterclockwise from the original rope axis orientation, and which is rooted in transient coronal holes external to the original source region.

After summarizing the geometric relationship of the RHESSI flare X-ray sources with the TRACE flare loops, and the motion modes of RHESSI flare X-ray sources, we concentrate on the newly discovered downward motion of flare looptop X-ray source. In particular, we present a new argument for the flare loop shrinkage during the impulsive phase: TRACE 195 Å loop shrinkage. The temporal behavior of 195 Å loop shrinkage and its magnitude are consistent with that revealed from the RHESSI X-ray sources. Further study shows that the TRACE 195 Å loop shrinkage is in the form of oscillation, with a period of about 150 s and an amplitude of about 300 km. We discuss this oscillatory shrinkage with respect to a numerical simulation of reconnection model.

We discuss a large flare that was observed simultaneously by RHESSI in hard X–rays and by the Nobeyama Radio Heliograph (NoRH) in microwaves. The imaging observations made both by RHESSI and NoRH show many interesting features which may be relevant for producing realistic flare models.

We study CMEs observed by LASCO to have plane of the sky velocities exceeding 1500 km/sec. We find that these extremely fast CMEs are typically associated with flares accompanied by erupting prominences. Our results are consistent with a single CME initiation process that consists of three stages. The initial stage is brought about by the emergence of new magnetic flux, which interacts with the pre-existing magnetic configuration and results in a slow rise of the magnetic structure. The second stage is a fast reconnection phase with flaring, filament eruption and a sudden increase of the rise velocity of the magnetic structure (CME). The third stage consists of propagation in the corona. We discuss the sources of these CMEs and the need for improved understanding of the first and third stages.

The Solar Mass Ejection Imager (SMEI) was launched into a Sun-synchronous orbit in January 2003. Its mission objective is to detect and track coronal mass ejections (CMEs) from the Sun in order to improve space weather forecasts. In the three years since launch, over 200 CMEs, about 30 of which were Earth-directed, have been observed by SMEI. We have been able to track several of these CMEs from the SOHO LASCO coronagraphs ($<$8$^\circ$ from the Sun) through the SMEI field of view ($>$20$^\circ$) out to 0.5 AU and beyond, and to observe the morphology and evolution of distinctive features over this wide distance range. We report on comparisons of measurements of CME parameters made in the inner heliosphere with the more typical measurements made nearer the Sun with coronagraphs. We illustrate SMEI's capabilities and present key statistical results on basic CME parameters and the use of SMEI-type data in space weather forecasting models. For example, timely observations by SMEI of CMEs en route to Earth could be input to DoD's operational Hakamada-Akasofu-Fry solar wind model to correct or refine its real-time forecasts of approaching disturbances.

We present studies on a series of ICMEs detected by Ulysses and for which the solar sources on the Sun could be identified. EUV and white light data are used in order to correlate characteristics seen during eruption with those measured in-situ. Particularly, an attempt was made to find solar features that show a relationship with the type of ICME seen later (i.e. cloud or non-cloud ICME). For magnetic clouds (MC) the chirality of the magnetic field was then analyzed. Finally, the charge states of oxygen ions contained in ICMEs were used to obtain freezing-in temperatures and then compare these with the presence of flares occurring close (spatially and temporally) to the CME eruption.

We have found no solar feature that could be used to predict the presence of a MC in interplanetary space, they occur with the same frequency for cloud and non-cloud ICMEs. The chirality of the clouds seems to follow only weakly the hemisphere rule. The presence of solar flares do not seem to be correlated with the oxygen freezing-in temperatures seen in-situ.

Five flare/CME events were selected in this study. One is on May 12, 1997, for which there is only two active regions on the visible solar disc, and the magnetic configuration is rather simple. For other cases, many active regions were visible. They are the flare/CME events that occurred on Bastille Day of 2000, Oct. 28, 2003, Nov. 7, 2004 and Jan. 20, 2005. By tracing the spread of EUV dimming, which was obtained by SOHO/EIT 195 Å fixed-difference images, we studied the CME initiation and development on the solar disc. At the same time we reconstructed the 3D magnetic structure of coronal magnetic fields, extrapolated from the observed photospheric magnetograms by SOHO/MDI. In scrutinizing the EUV brightening and dimming propagation from CME initiation sites to large areas with different magnetic connectivities, we determine the overall coupling and interacting of multiple flux systems in the CME processes. Several typical patterns of magnetic connectivity are described and discussed in the view of CME initiation mechanism or mechanisms.

We analyzed brightness variations in widespread dimming regions occurred in several major eruptive events in 2003 and 2005 resulted in fast halo CMEs. In all cases brightness of remote dimming regions show some pre-eruptive growth, then gradual and fast decrease to minimum. The time interval of maximal decrease overlaps with X-flare peak and LASCO CME onset. Brightness variations in some dimmed regions are highly correlated before the CME onset and much less – after. It suggests a break of magnetic connectivity between previously linked coronal structures. Dimmings develop similar at 175 and 195 Å being shallower and sometimes delayed at 304 Å. In large-scale region (L$>$0.5 $R_\odot$) dimming propagates with a speed of $\sim$ 250 km/s which is typical for EIT-waves.

Time-distance helioseismology and data from SOHO/MDI are used for obtaining 3D images of subsurface sound-speed perturbations and maps of plasma flows, associated with emerging magnetic flux and flaring activity of large active regions in October 2003. The results reveal extremely complicated dynamical processes in the upper convection zone and indicate that subsurface shear flows may play an important role in magnetic energy release in solar flares. Strong X-class flares generated impulsive seismic waves (“sunquakes”), traveling through surrounding sunspots, thus providing new insight into the interaction of seismic waves with magnetic fields.

Employing the Poincaré index of isolated null-point in a vector field, we work out a mathematical method of searching 3D null-points in coronal magnetic fields. After introducing the theory of differential topology, we test the method by using the analytical model of Brown & Priest. The location of null-point identified by our method coincides precisely with the analytical solution. Finally we apply the method to search 3D null-points in coronal magnetic fields reconstructed based on observed magnetograms of two active regions (NOAA AR 10488 and AR 10720). We find that the 3D null-points seem to be a key element to the magnetic topology associated with flare occurrence.

Very few in situ measurements so far show the details of the interactions between CME and magnetosphere. Here we report a fortuitous observation that the Cluster spacecraft were located at the dusk-side magnetopause, meandering back and forth four times between the ICME and the magnetosphere, and observed part process of the interaction between the CME and the magnetosphere. The primarily analysis showed that: (1) a fast ion beam other than the ambient plasma was observed almost perpendicular to the magnetic field; (2) The left-hand-polarized kinetic Alfvén waves excitation by the ion beam; and (3) the resonances between the kinetic Alfvén waves and part of the beam particles lead to the pitch angle diffusion of resonant particles and formating of a monospheric distribution with nearly constant kinetic energy. The wave-particle resonance may pay an important role in the energy transfer and particle exchange across the magnetopause between the magnetosheath and magnetosphere during the CME impulsion.

I first review: a) the current state of knowledge of ion acceleration in solar flares; b) the physics of positron production and annihilation; and c) recent RHESSI data on solar flare annihilation radiation. I then show how the modeling of the positron production and annihilation in the chromosphere, coupled with the newly available high-resolution data on the 511 keV annihilation line, can have important physical implications w. r. t. the models: a) information on the temperature and density of the chromosphere; b) constraints on some of the physical characteristics of the flare and to some extent on the acceleration process.

Although I do mention past instruments (SMM and Yohkoh), this review focuses on the RHESSI satellite, considering the quantum leap it has constituted in the quality of the data it is providing and consequently the constraints it can place on models (of ion acceleration, annihilation environment, etc.).

We present preliminary results of the study of the photospheric physical conditions about 30 minutes before the 2N/M2 two-ribbon solar flare. The soft X-ray enhancement, $H_{\alpha}$ brightening and microwave bursts occured prior to the flare. We demonstrate the semiempirical models of the preflaring photosphere fot two kernels of the future flare. These models derived from the inversion reproduce the spectral observations in seven Fraunhofer lines. The models consist of two atmospheres: the magnetic flux tube and its nonmagnetic surroundings. The inferred models show strong changes of the photospheric physical conditions in the preflare phase. The preheating was found in the flux tube. Upward motions of the surrounding photospheric material were revealed. Significant discrepancies of the flux-tube temperature and density are found for two sites of the future flare. The inner kernel located near the filament was cooler in the photosphere than the outer kernel.

A series of drifting microwave bursts during the 30 March 2001 flare are analyzed using the Siberian Solar radiotelescope (SSRT) images at 5.7 GHz and dynamic spectra obtained simultaneously by the spectropolarimeters of National Astronomic Observatories in China (NAOC) in the range 5.2–7.6 GHz. While observing the event with the SSRT, the burst sources were simultaneously recorded at two frequencies, which allowed their relative spatial shifts to be measured and source velocity along the flare loop (observed in soft X-ray and ultraviolet emission) to be evaluated. Estimates were made of the plasma density gradient along the source movement direction, the plasma emission being assumed to be generated at the second harmonic. Drifting burst series occur during transient hard X-ray brightenings. Burst drift rates ranged from $-$10 to 20 GHz/s, with a mean value of about 6 GHz/s. The shape of the drift rate distribution around the mean value is nearly symmetric. It is suggested that the mean value distribution may be related to increased plasma density in the source of subsecond pulses. In particular, the corresponding density variations may be associated with magnetic reconnection processes.

Helioseismic holography is a technique used to image the sources of seismic disturbances observed at the solar surface. It has been used to detect acoustic emission, known as sun quakes, radiated from X-class solar flares. Since the seismic power emitted by the X-class flares has proved to be independent of the strength of the flare, we have undertaking a systematic search for seismic signatures from M-class solar flares, observed by SOHO-MDI.

We have detected significant acoustic emission from a few M-class solar flares. Preliminary results of the survey of M-type solar flares studied so far is available at: aira.astro.ro/~deanna/M.html.

We report the eruption of a small filament on 2001 September 28. Followed by a two-ribbon flare, the eruption was associated with a partial halo coronal mass ejection. During the eruption, coronal bipolar double dimmings were formed on the regions of opposite magnetic polarities. The optical counterparts of the coronal dimmings, i.e., the $H_{\alpha}$ dimmings, were clear and showed similar appearance and consistent development. A remarkable fact is that the two dimmings were preceded by impulsive EUV and $H_{\alpha}$ brightenings at the flare's rise phase, suggesting that the eruption can be explained by the tether-cutting model.

The problems of required conditions and possible consequences of the super - compression (up to $R_m\sim 3R_E$) of the Earth's magnetosphere by giant CME are investigated by the methods of laboratory and computer simulations. A useful relation between an expected magnetopause location $R_m^*$ and the kinetic plasma energy $E_0$ of spherical plasma cloud (exploded at distance $R_0$) was obtained $R_m^*/R_0\approx 0,75/{\ae}^{1/6}$ and tested by MHD – model of Nikitin & Ponomarenko (1994) with the using their main energetic criterion of the problem ${\ae}=3E_0 R_0^3/\mu^2$ (for magnetic moment $\mu$ of point obstacle in vacuum). This relation could describe rather well an observed compression ($R_m\sim 5-6 R_E$ for CME with energy $10^{32}\, ergs$ and effective value $E_0\sim 10^{33} ergs,$ into $4\pi$) and predicts $R_m^*\leq 3R_E$ in a probable case of Mega Flare with the total energy release $\sim 10^{34}\, ergs$ and possible $E_0\sim 5\cdot 10^{34}\, ergs$ according to Kane et al. (1995) and Tsurutani et al. (2003). Some most important features of the formation such Artificial Magnetosphere (AM) structure and its possible influence onto various geospheres media (or technosphere areas) could be successfully studied in the simulative experiments at KI-1 facility of ILP with Laser Plasmas (LP) of $E_0$ up to $kJ$ and dipole $\mu\sim 10^7\, G\cdot cm^3$ as was shown by Ponomarenko et al. (2001) and Zakharov (2003). But the main problem of such planned AMEX experiment (at ${\ae}\sim 50$ for $R_0=75\, cm$) is the influence of finite value of ion magnetization $\varepsilon_m=R_L/R_m^*$ based on the ion Larmor radius $R_L=mcV_0/ezB_d,$ where $V_0\sim 100\, km/s$ is the expansion velocity of LP and $B_d$ is the initial dipole field at the point $R_m^*.$ Of coarse, $\varepsilon_m\ll 1$ in a real space conditions (excluding cases of Mercury or asteroids, explored by Omidi et al. (2004)) while in the laboratory to fulfill both need constrains ${\ae}\gg 1$ and $\varepsilon_m\ll 1$ we should use a thermonuclear plasma and devices. To overcome this problem we did a 3D/PIC – calculations by hybrid model of Kyushu University, described by Muranaka et al. (2001), to find out a critical value of $\varepsilon_m^*$ $(\approx 0,2-\!0,3),$ which need for MHD – like interaction of exploding plasmas with magnetic dipole.

We report the temporal evolution of the long lasting solar flare observed on June 6, 2000 (15:00 – 17:00 UT) by the Brazilian Solar Spectroscope (BSS) at INPE. We emphasize the identification of the decimetric fine structures, such as “fiber” and “zebra” emissions, including the unique case of harmonic “zebra” emissions in the decimetric band reported, radio pulsations, type III bursts and variants, recorded during this event in the frequency range of (1.2 – 1.7 GHz). The main characteristics of fine structures recorded are presented.

An occulted flare occurred at about 06:07 UT on November 2, 2002. The RHESSI X-ray images show two separate parts. The lower part consists of a complete loop and the upper part a coronal source which well extends above the solar limb. The loop source shrank for about 4 minutes with a speed of $\sim$24 km s$^{-1}$ during the early impulsive phase and then expanded at $\sim$7 km s$^{-1}$, while the coronal source presented an upward motion at about 6 km s$^{-1}$. The temperature of the loop-top is always higher than that near the footpoints. The temperature of the loop increased with altitude, while that of the coronal source just showed opposite variation. These results indicate that the reconnection X-point of this flare is located in between the loop-top source and the coronal source.

We report on a complex study of a typical large W-limb CME event occurring on December 2, 2003 in the vicinity of AR 0508. It is associated with a prominence eruption which has been observed with the Pic du Midi Ha flux coronagraph as well as in EUV by both the SOHO/EIT and the Coronas/SPIRIT space telescopes. The eruption started with the emergence of a fast expanding loop between 9 and 10 UT, followed by a heating and an acceleration of the erupted material and resulted in three-part CME observed by LASCO after 10:50 UT. A temporal analysis of the prominence motion and the EUV dimming light curve have shown that both the frontal structure and the core of a CME were initiated simultaneously with the peak of the X-ray C7.2 flare. The total mass of the prominence including the H$\alpha$ filament and EUV filament channel is close to the mass of the core but significantly less than the total CME mass.