Flare Parameters Inferred from a 3D Loop Model Data Base

We developed a data base of pre-calculated flare images and spectra exploring a set of parameters which describe the physical characteristics of coronal loops and accelerated electron distribution. Due to the large number of parameters involved in describing the geometry and the flaring atmosphere in the model used, we built a large data base of models (similar to 250 000) to facilitate the flare analysis. The geometry and characteristics of non-thermal electrons are defined on a discrete grid with spatial resolution greater than 4 arcsec. The data base was constructed based on general properties of known solar flares and convolved with instrumental resolution to replicate the observations from the Nobeyama radio polarimeter spectra and Nobeyama radioheliograph (NoRH) brightness maps. Observed spectra and brightness distribution maps are easily compared with the modelled spectra and images in the data base, indicating a possible range of solutions. The parameter search efficiency in this finite data base is discussed. 8 out of 10 parameters analysed for 1000 simulated flare searches were recovered with a relative error of less than 20 per cent on average. In addition, from the analysis of the observed correlation between NoRH flare sizes and intensities at 17 GHz, some statistical properties were derived. From these statistics, the energy spectral index was found to be delta similar to 3, with non-thermal electron densities showing a peak distribution less than or similar to 10(7) cm(-3), and B-photosphere greater than or similar to 2000 G. Some bias for larger loops with heights as great as similar to 2.6 x 10(9) cm, and looptop events were noted. An excellent match of the spectrum and the brightness distribution at 17 and 34 GHz of the 2002 May 31 flare is presented as well.