AspSync included two work packages: WP1 analog modeling; WP2 statistics of natural subduction zones.
WP1A) 14 experiments, where the along-strike length of the barrier and of the asperities are varied systematically in the 0–10 cm range and in the 6–17 cm range respectively, have been realized. When scaled to nature, these values are well in the range of geodetically observed asperities. The experiments featured a frictionally segmented megathrust embedding two asperities with equal geometry and friction, separated by a barrier. Each experiment produced >80 analog earthquakes characterized by single asperities ruptures and/or asperities synchronization (i.e. both asperities fail during the same earthquake) depending on the boundary conditions.
WP1B) 12 experiments have been realized where the subduction velocity and seismogenic zone width have been varied systematically. With respect to the experiments of WP1A, here the models had no lateral frictional segmentation. Subduction velocity and width of the seismogenic zone have been proposed to exert a key role on interplate seismicity for their first-order control on deformation rate and coupling area between plates, respectively. AspSync investigated how the two selected parameters tune maximum magnitude, seismicity rate and moment release rate.
WP2) One of the most challenging questions for seismologist is which physical parameter, if any, tunes Mmax. AspSync updated an existing database including physical, geodynamical and seismological parameters of worldwide subduction zones and performed multivariate statistics to identify the key controlling factors tuning Mmax. The need of multivariate statistics represented a step forward in this field as the majority of previous studies were based on bivariate statistics. AspSync used the Pattern Recognition algorithm.
OVERVIEW OF THE RESULTS
Experiments focusing on asperities synchronization highlighted the importance of the barrier-to-asperity length ratio Db/Da in tuning megathrust seismicity. In particular, AspSync’s results highlighted that Db/Da displays a negative correlation with maximum magnitude and seismicity rate. Db/Da controls also the process of asperities synchronization. When Db/Da is < 0.5 asperities synchronization has been observed in the models and the percentage of events with synchronized asperities rupture is inversely proportional to Db/Da. AspSync also shed light on how the process of asperities synchronization takes place. In particular, AspSync imaged a sequence of cracks activating at different times, allowing the rupture to grow laterally.
AspSync’s experiments highlighted that Mmax is mainly controlled by the width of the seismogenic zone as the rupture potential increases with this parameter. No significant correlation between Mmax and subduction velocity has been observed in AspSync’s models as well as in natural subduction zones. Subduction velocity rather controls seismicity rate, in particular a high subduction velocity is associated to high seismicity rate. Moment release rate is controlled by both subduction velocity and width of the seismogenic zone. Consequently, the fastest subduction zones with medium to large seismogenic zone widths are expected to have the largest moment release rates.
The statistical analyses of AspSync highlighted the major role of length of the trench and the thickness of subducting sediments, the parameters concurring to enhance long ruptures in the trench-parallel direction. In particular, AspSync verified by means of Monte Carlo simulations that the occurrence of magnitude >8.5 earthquakes along the longest subduction zones characterized by a relatively high sediment supply does not appear to be random. A smooth or smoothened megathrust combined with a long subduction zone enhance the conditions for large trench-parallel extent of the rupture and, in turn, higher earthquake magnitudes.
EXPLOITATION AND DISSEMINATION
AspSyncs’ results have been summarized in 5 peer-review articles in international journals, available as author’s post-print or as DOIs on aspsync.wordpress.com. I have started publicizing it during the AGU meeting in Dec. 2018. From that moment the site received >200 views from various countries in Europe, Japan and USA.
AspSync’s results have been presented at the largest geo-sciences conferences: the European Geosciences Union EGU in ‘16 and ‘17 and the American Geosciences Union AGU in ‘16 and ‘17. A summary and future directions of AspSync will be presented at the EGU also this year. AspSync results have been presented also at smaller meetings and workshops (4 posters and 6 oral presentations in total).
