Metric features of self-organized criticality states in sandpile models

A new characterization of self-organized criticality (SOC) states is develo ped by using metric features of the configuration's space. Quantities mainl y referring to the partition formalism, as mutual factorization, Shannon en tropy and Rohlin distances with their distributions and power spectra, are considered. Time series for these observables give account of geometrical a nd dynamical complexity through the interdependence of fractality and flick er noise. For Bak-Tang-Wiesenfeld and Manna automata, new indicators enforc e previous results given by standard parameters and allow a deeper insight into the structure of SOC configurations and their time behaviour. Moreover , we obtain indications regarding a possible split in the universality clas s of the two automata.