NONLINEAR PROPERTIES OF THE DYNAMICS OF BURSTS AND FLARES IN THE SOLAR AND STELLAR CORONAE

Solar and stellar flares are highly structured in space and in time, a s is indicated for example by their radio signatures: the narrowband s pikes, type III, type II and IV, and pulsation events. Structured in t ime are also the not flare related type I events (noise storms). The n ature of this fragmentation is still not clear. Either, it can be due to stochastic boundary or initial conditions of the respective process es, such as inhomogeneities in the coronal plasma. Or else, adetermini stic non-linear process is able to cause complicated patterns of these kinds. We investigate the nature of the fragmentation in time. The pr operties of processes we enquire are stationarity, periodicity, interm ittency, and, with dimension estimating methods, we try to discriminat e between stochasticism and low-dimensional determinism. Since the mea sured time series are rather short, the dimension estimate methods hav e to be used with care: we have developed an extended dimension estima te procedure consisting of five steps. Among others, it comprises agai n the questions of stationarity and intermittency, but also the more t echnical problems of temporal correlations,judging scaling and converg ence, and few data points (statistical limits). We investigate 3 event s of narrowband spikes, 13 type III groups, 10 type I storms, 3 type I I bursts and 1 type IV event of solar origin, and 3 pulsation-like eve nts of stellar origin. They have in common that all of them have stati onary phases, periodicities are rather seldom, and intermittency is qu ite abundant. However, the burst types turn out to have different char acteristics. None of the investigated time series reveals a low-dimens ional behaviour. This implies that they originate from complex process es having dimensions (degrees of freedom) greater than about 4 to 6, w hich includes infinity, i.e. stochasticity. The lower limit of the deg rees of freedom is inferred from numerical experiments with known chao tic systems, using time series of similar lengths, and it depends slig htly on the burst types.