Current Projects
We use the unprecedented spectral and spatial resolution of RHESSI to explore the behavior of electrons and their associated currents in solar
flares. Spectral images are used to determine an estimate of the effective surface area for the different independent substructures
within each event. The incident electron spectra at those flaring  footpoints are derived from the RHESSI photon spectra. We find that,
over a wide range of flare X-ray magnitudes, the integrated photon flux above 20 keV asymptotically approaches a limiting value, suggesting a
saturation of the photon production in flares. The inferred particle fluxes in the beam, together with this saturation limit, are used to
determine the energy loss mechanism dominating the energetic particle transport in solar flares: one of the main results is that there seems
to be a relatively sharp transition from Coulomb dominated to return current dominated emission as the flares get more energetic. We also try
to compare our image analysis results to models of flare hard X-ray production, incorporating magnetic topology variation, for both cold and warm target situations. Furthermore, we use the flare substructure analysis to try to explain coronal heating.
We use Fokker-Planck modeling to investigate the electron temporal, spatial, energetic and directional distribution in flare loops. Multiple cases can be studied, such as converging magnetic fields in the loops, collisions of fast electrons with each other, reverse currents, collision and energy dissipation due to a warm target case at the loop footpoints, and asymmetric particle distribution injection in the loop top. Both thick target at the base of the loop, and thin target high in the corona can be used to infer the photon production in these cases.
We investigate the temporal and spatial relationship between filament eruptions and the production of hard X-ray emission using spatially resolved high cadence data from TRACE and RHESSI. In particular, we focus on comparing the characteristics of the hard X-ray production in 'successful' and 'failed' filament eruption cases. Our preliminary findings indicate even failed eruption events can generate significant energy release and hard X-ray emission with the hard X-ray production apparently correlated to the rate of expansion of the filament. The spatial distribution of the hard X-ray emission, while depending upon the overall strength of the event, also depends on the evolutionary behavior of the filament as it erupts, e.g. looplike versus "zipper"-like.
Coordinated observations of UV and hard X-ray emission in flares provide crucial diagnostic information regarding the relationship of active region topology to the flare-related energy release and potential CME initiation. We present analysis of the temporal relationship and spatial distribution of these disparate emissions for two X-class flares with high-cadence TRACE 1600Å observations and hard X-ray imaging from RHESSI. We verified the known temporal relationships from previous studies and compare the spatial distributions of the UV and hard X-ray emission to determine cases of spatial separation or extended development. The spatial distributions observed along with the temporal correlations require a complex 3-D topological picture involving the interaction of multiple flux systems fostering magnetic reconnection and thus flare energy release along a temporally evolving separator.

Eruptions of  soft X-ray sigmoids and the onset of fast coronal mass ejections (CMEs) are believed to be driven by the over-accumulation of coronal helicity. Thus, the helicity production and injection is an attractive issue recently. Both the emergence of twisted magnetic fields from below and the photospheric horizontal motions are the most likely means to produce and inject the helicity into the corona. We study the helicity injection for two strong solar events which occurring in NOAA 9684 and 9704 on Nov. 04 and 22 2001, which produced large flares (X1.0/3B and M9.9), associated with sigmoidal filament and soft X-ray eruptions,  very fast CMEs (1840 and 1437 km/s) and very strong proton events (31700 and 18900 pfu). Since the two events occurred on the solar west disk (W18 and W34), we are able to study in detail the helicity production and injection before the events, using more reliable MDI 96 m line-of-sight magnetograms and a local correlation tracking (LCT) method. NOAA 9684 is found to have a rotating sunspot, while NOAA 9704 shows significant horizontal motions. Thus, our results provid clues to the roles plays by twist or writhe helicity in the filament and sigmoid eruptions, which are associated with onset of the powerful CMEs.

Active regions with a delta magnetic configuration from 1996 to 2002 were selected to study how important a role the kink instability plays in such active regions.  We employ the systematic tilt angle of each active region as a proxy for the writhe of a fluxtube and the force-free parameter, alphabest, as a proxy for the magnetic field twist in the fluxtube. It is found that 65-67% of the 104 active regions have the same sign of twist and writhe, which violate the Hale-Nicholson and Joy's  Laws (HNJL) or the hemispheric helicity  rule (HHR). 68% (46/68) of these active regions produced more than five large flares. Active regions violating HNJL, but following HHR, have a much stronger tendency to produce X-class flares and/or strong proton events. Continuously clockwise rotation of magnetic configuration of a long-lived active region (AR 9604-9632-9672-9704-9738)which produced major flares, fast CMEs and much strong proton events shows that a kink instability would play very important role in such active regions. These results support the prediction for the presence of a kink instability, that the twist and writhe of the magnetic fields exhibit the same sign for delta active regions (Linton et al, 1998, 1999, and Fan et al., 1999). Finally, we analyze possible origins of the twist and writhe of the magnetic fields for the active regions with different relations between the twist and writhe.