Global evidence that fault complexity controls on-fault and off-fault deformation partitioning throughout the earthquake cycle

A major open question in earthquake science is how crustal deformation is partitioned between elastic strain accumulation on known faults and distributed deformation in the surrounding crust throughout the earthquake cycle. This distinction is critical for seismic hazard assessment but remains difficult to resolve because surface deformation reflects contributions from both sources. Here, we implement a framework that jointly estimates slip deficit rates on three dimensional faults and distributed moment rate sources in the crust, providing internally consistent estimates of their relative contributions and posterior uncertainties. Applying this approach across the western United States, eastern Mediterranean, Tibet, and New Zealand reveals a systematic dependence of deformation partitioning on fault system complexity. Mature, localized fault systems, including the Main Himalayan Thrust, San Andreas, North Anatolian, and Alpine faults, accommodate 70 to 90 percent of deformation between earthquakes on faults. In contrast, immature or diffuse systems, such as the Basin and Range, Tibetan Plateau, Intermountain Seismic Belt, western Anatolia, and northern New Zealand, accommodate only 30 to 60 percent on faults, with the remainder distributed off-fault. These results demonstrate that off-fault deformation is a fundamental component of geodetic strain rates, with its relative contribution governed by fault system complexity. Moreover, in light of recent evidence that cumulative fault-length distributions follow a power law with an exponent near -2 (Zou and Fialko, 2024), our results suggest that a significant fraction of off-fault deformation may be accommodated aseismically throughout the earthquake cycle.