Constraints on energy release in solar flares from RHESSI and GOES X-ray observations I. Physical parameters and scalings

Aims. We constrain energy release and particle acceleration processes in solar flares by means of comprehensively characterizing the physical parameters of both the thermal plasma and the accelerated nonthermal particles using X-ray data. Our aim is to bridge the gap between detailed case studies and large statistical studies. Methods. We obtained time series of spectral fits and images for 24 flares ranging from GOES class C3.4 to X17.2 using RHESSI hard X-ray observations. These data were used to derive basic physical parameters for the thermal plasma (using the isothermal approximation) and the injected nonthermal electrons (assuming the thick-target model). For the thermal component, this was supplemented by GOES soft X-ray data. We derived the ranges and distributions of the various parameters, the scaling with flare importance, and the relation between thermal parameters derived from RHESSI and GOES. Finally, we investigated the relation between thermal and nonthermal parameters. Results. Temperature and emission measure of the thermal plasma are strongly correlated with the peak GOES X-ray flux. Higher emission measures result both from a larger source volume and a higher density, with the latter effect being more important. RHESSI consistently gives higher temperatures and lower emission measures than GOES does, which is a signature of a multithermal plasma. The discrepancy between RHESSI and GOES is particularly pronounced in the early flare phase, when the thermal X-ray sources tend to be large and located higher in the corona. The energy input rate by nonthermal electrons is correlated with temperature and with the increase rate of emission measure and thermal energy. Conclusions. The derived relations between RHESSI-and GOES-derived thermal parameters and the relation between thermal parameters and energy input by nonthermal electrons are consistent with a two-component model of the thermal flare plasma. Both RHESSI and GOES observe a cooler plasma component (approximate to 10-25 MK) that is generated by chromospheric evaporation caused by a nonthermal electron beam. In addition, a hotter component (>= 25 MK) is only detected by RHESSI; this component is more consistent with direct in situ heating of coronal plasma. With the exception of the early impulsive phase, RHESSI observes a combination of the evaporated and the directly heated component.