Scaling of rupture initiation from P-wave onset: insights from earthquakes and laboratory experiments

Despite recent advances from real observations, laboratory experiments and numerical modelling, the mechanisms governing earthquake generation and wave propagation are still not fully understood. Theoretical analyses and laboratory experiments increasingly show that seismic ruptures begin with a process of quasi-static slip accumulation over a limited region of the fault (referred to as the preparatory phase). Here, when a critical dimension (or area) is reached, the slip rapidly accelerates (referred to as the break-out phase) and creates a rupture front which finally propagates dynamically. While the breakout phase is observed at laboratory scale, no direct evidence is available at the scale of real-earthquake data. The unresolved question is whether the breakout phase has an influence on the final size of the forthcoming event. In other words, it remains unclear whether all earthquakes begin through a similar process—characterized by the exceedance of a yield stress and influenced by local frictional properties or geometric complexities of the fault surface—with the rupture extent determined during propagation, or whether fundamentally different initiation mechanisms govern the generation of small and large events.  Within the framework of the ERC FORESEEING project (https://www.foreseeing.eu/), we use the P-waves to shed light onto the mechanism of generation of seismic ruptures and to constrain the role of the parameters involved in the process. We investigate the onset of P-waves across multiple datasets, focusing on natural earthquakes with magnitudes Mw 1–4 from four well-instrumented regions: the Campi Flegrei region (Southern Italy), the Irpinia Near Fault Observatory (Southern Italy), the TABOO network (Central Italy), and The Geysers geothermal field (USA), for a total of thousands of events and available waveforms.

Following the approach of Longobardi et al. (2025), we analyse the behaviour of the low‐pass displacement vs. time curve (LPDT). We found that LPDT curves grows differently for micro (Mw 1-2) to small earthquakes (Mw 2-4), following a similar trend as observed for worldwide moderate-to-large events (Longobardi et al. 2025). For a limited number of events, we extended the analysis to laboratory scale - 300 mm fault length - experiments, where we use acoustic signals to investigate the relation between the LPDT and the seismic source allowing a direct comparison between elastic waves over a wide range of spatial and temporal scales. We will discuss the scaling of LPDT curves across diverse seismic environments and magnitude ranges, and its potential use for rapid source characterization, in the context of Earthquake Early Warning applications.