Spontaneous segmentation of slow slip events on a homogeneous planar fault

Abstract Slow slip events (SSEs) are transient aseismic fault slip events that often display segmentation along strike. While fault segmentation is commonly attributed to physical heterogeneities, numerical studies have shown that segmentation can also occur on homogeneous planar faults. In this study, we investigate the dynamics of SSEs on a homogeneous planar fault with velocity-weakening (VW) properties, focusing on the influence of the surrounding velocity-strengthening (VS) region. Based on numerical simulations, we reveal that the effective normal stress in the VS region ( σ VS ) alters the slip stability in the VW region in a non-monotonic manner. σ VS controls the propagation of SSEs into the VS region: under low σ VS , the rupture expands into the VS region, resulting in synchronous slip, whereas under high σ VS , the SSE is confined within the VW region, triggering afterslip in the VS region. This behavior is successfully captured by a simplified two-degree-of-freedom (2DOF) spring-slider model, and the non-monotonic change of the stability in the VW region with respect to σ VS is understood from linear stability analysis of the 2DOF model. σ VS also drives the change in SSE patterns. By systematically varying σ VS , we identify three distinct types of SSEs: periodic, quasi-periodic, and chaotic types, occurring in the VW region even with identical frictional properties. The low σ VS regime tends to produce periodic SSEs with spatiotemporally regular segmentation, while the high σ VS regime is characterized by chaotic SSEs exhibiting frequently varying segment sizes and boundaries. Quasi-periodic SSEs are regarded as a transition between the two. All types of SSEs spontaneously segment without imposing heterogeneities. We demonstrate, through linear stability analysis of the continuum fault model, that differences in slip patterns are attributed to the spatial scale of the dominant unstable linearized perturbation. While the diversity of SSE patterns has been reported in previous studies, this study identifies the condition and the mechanism that control the slip patterns.