Sandpile-based model for capturing magnitude distributions and spatiotemporal clustering and separation in regional earthquakes

We propose a cellular automata model for earthquake occurrences patterned after the sandpile model of self-organized criticality (SOC). By incorporating a single parameter describing the probability to target the most susceptible site, the model successfully reproduces the statistical signatures of seismicity. The energy (magnitude) distributions closely follow power-law probability density functions (PDFs) with scaling exponent −5/3, consistent with the expectations of the Gutenberg–Richter (GR) law, for a wide range of the targeted-triggering probability values; this suggests that SOC mechanisms are still present in the model despite the introduction of the targeted triggering. Additionally, for targeted triggering probabilities within the range 0.004–0.007, we observe spatiotemporal distributions that show bimodal behavior, which is not observed previously for the original sandpile. For this critical range of values for the probability, model statistics show remarkable comparison with long-period empirical data from earthquakes from different seismogenic regions. The proposed model has key advantages, foremost of which is the fact that it simultaneously captures the energy, space, and time statistics of earthquakes by just introducing a single parameter, without disrupting the SOC properties of the sandpile grid. We believe that the critical targeting probability is a key requirement for SOC in seismicity, as it parametrizes the memory that is inherently present in earthquake-generating regions.