Current Research in EEPS: Dr. Christine Chesley, Lamont-Doherty Earth Observatory
Electromagnetic data at the Hikurangi Margin, New Zealand reveal fluid-rich subducting seamounts in a region of shallow slow slip
How does seafloor topography, that is seamounts, ridges, and other features that stick up above the sediment cover, affect subduction processes, particularly those related to megathrust fault slip? Before extensive geodetic networks and collocated seismic data were more widely available, it was hypothesized that subducting seamounts should act as rupture asperities whose size could generate large subduction earthquakes. Although some examples suggest that subducting topography may be associated with large megathrust events, more recent geophysical observations point to a link between such topography and aseismic creep, microseismicity, and slow earthquakes. Analogue models show that the geometric incompatibility of subducting topographic relief is accommodated by the generation of complex fracture networks in the overriding plate. These fault networks are unlikely to support earthquakes that rupture over large areas and are more likely to slip through a combination of small-earthquakes and creep. However, direct geophysical imaging of the complex fracture networks proposed and the hydrology of both the subducted topography and associated upper plate damage zones remains elusive. In this talk, I will present results from passive and controlled-source seafloor electromagnetic data collected at 3 trench-crossing profiles along the Hikurangi Margin, New Zealand, with an emphasis on the northern profile where active seamount subduction is present. I will show that the internal structure of a seamount on the incoming plate allows at least 3.2–4.7x more water than normal, unfaulted oceanic lithosphere to subduct. In the forearc, the data reveal a sediment-starved plate interface above a subducting seamount with similar electrical structure to the incoming plate seamount. A sharp resistive peak within the subducting seamount lies directly beneath a prominent upper plate conductive anomaly. The coincidence of this upper plate anomaly with the location of burst-type repeating earthquakes and seismicity associated with a recent slow slip event, directly links subducting topography to the creation of fluid-rich damage zones in the forearc that alter the effective normal stress at the plate interface by modulating fluid overpressure. In addition to severely modifying the structure and physical conditions of the upper plate, subducting seamounts represent an underappreciated mechanism for transporting a considerable flux of water to the forearc and deeper mantle.
Broad Audience Abstract:
As Earth scientists, we cannot always directly observe the processes we want to study, usually because of their inaccessibly deep locations or the amount of time they take to occur. For that reason, we have developed tools that allow us to estimate different properties of the subsurface so as to make educated guesses about what is occurring in areas of interest. Subduction zones, places where one lithospheric plate dives deep into the Earth beneath another plate, host the largest and most destructive earthquakes, which is why we are invested in understanding subduction systems more fully. At the northern end of the Hikurangi subduction zone, we collected electromagnetic (EM) data both on the incoming Pacific plate and as the plate subducts beneath the overriding New Zealand landmass. EM data are highly sensitive to conductors, like seawater, melt, and metallic minerals, and thus allow us to determine where these components are present in the subduction system. The incoming Pacific plate is not smooth near New Zealand but rather includes several features, called seamounts, that stick up above the sediment cover. Before our study, while it was recognized that these seamounts have some effect on subduction earthquakes, most researchers had not been considering how a seamount’s internal structure plays a role in the seamount’s contribution to earthquake fault slip. Using our EM data, we show that a seamount on the incoming plate contains a strong core and a large amount of water that, upon subduction, will allow it to damage the overriding plate while also transporting excess fluids to the plate interface or deeper. We also image a seamount that is already subducting. In the overriding plate above the seamount, we find a concentrated region of high conductivity that coincides with a cluster of seismicity from the 2014 slow earthquake. Our observations show that seamounts can be vessels for bringing water to the subduction system and may have some influence on the occurrence of slow earthquakes.