资源环境科技发展态势分析平台

We examine the effects of topography and material heterogeneity on geodetic inversions of shallow slow slip at the Hikurangi Subduction Margin, New Zealand. Offshore vertical data from absolute pressure gauges constrain our results. Using a finite element code to generate our Green's functions, we find that topography has a relatively small effect, reducing the seismic potency by only about 5% when compared to a flat model. When material heterogeneity from a New Zealand-wide seismic velocity model is taken into account, we find an extremely large effect, with the seismic potency increasing by 58% or more. In previous work, we found the opposite effect for deeper slip events, where models including elastic property variations result in similar to 20% more slip than when they are neglected. Our previous results for deeper events are consistent with a simplified model that considers the elastic property contrasts between a strong subducting slab and a weaker overriding plate. To understand the opposite behavior of our shallower models, we use synthetic forward models and view material property gradients in terms of additional body forces. We find that gradients in elastic properties can produce very different surface displacements near the source than simple contrasts, thus explaining the contrasting results for shallow slip events. Combined with our previous results for deeper events, this implies that neglecting material heterogeneity in slow slip inversions may result in overestimates in the amount of slip for deeper events and underestimates for shallow events.

Plain Language Summary Subduction zones, where one plate moves beneath another, are home to Earth's largest earthquakes as well as "slow slip events" (SSEs), which are slow earthquakes occurring over days to years. SSEs produce no ground shaking but produce movement of the Earth's surface that we can detect with methods such as GPS. By analyzing patterns of surface movement during an SSE we can determine the slip that occurred. Accurate estimates are important, because slip occurring during an SSE might otherwise be released in a damaging earthquake. Models to determine SSE slip generally assume a flat upper surface and properties that do not vary. When we consider these effects for a shallow SSE offshore of the North Island of New Zealand we find that topography has little effect on our results, but property variations result in much larger amounts of slip. Combined with previous results, if we neglect material property variations we will overestimate the amount of slip already released in deep SSEs and underestimate the amount of slip released in shallow SSEs. This impacts evaluation of earthquake hazards, particularly in terms of how much plate motion is accommodated by SSEs versus the amount expected during a subduction zone earthquake.