A document by Miriam Gauntlett. Click on the document to view its contents.

Microseismicity, induced by the sliding of a glacier over its bed, can be used to characterize frictional properties of the ice-bed interface, which are a key parameter controlling ice stream flow. We use naturally occurring seismicity to monitor spatiotemporally varying bed properties at Rutford Ice Stream, West Antarctica. We locate 230000 micro-earthquakes with local magnitudes from –2.0 to –0.3 using 90 days of recordings from a 35-station seismic network located ~40 km upstream of the grounding line. Events exclusively occur near the ice-bed interface and indicate predominantly flow-parallel stick-slip. They mostly lie within a region of interpreted stiff till and along the likely stiffer part of mega-scale glacial landforms. Within these regions, micro-earthquakes occur in spatially (<100 m radius) and temporally (mostly 1-5 days activity) restricted event-clusters (up to 4000 events), which exhibit an increase, followed by a decrease, in event magnitude with time. This may indicate event triggering once activity is initiated. Although ocean tides modulate the surface ice flow velocity, we observe little periodic variation in overall event frequency over time and conclude that water content, bed topography and stiffness are the major factors controlling microseismicity. Based on variable rupture mechanisms and spatiotemporal characteristics, we suggest the event-clusters relate to three end-member types of bed deformation: (1) continuous creation and seismogenic destruction of small-scale bed-roughness, (2) ploughed clasts and (3) flow-oblique deformation during landform-formation or along bedrock outcrops. This indicates that multiple processes, simultaneously active during glacial sliding, can accommodate stick-slip behaviour and that the bed continuously reorganizes.

The crystal orientation fabric of glacier ice severely impacts its strength and flow. Crystal fabric is therefore an important consideration when modelling ice flow. Here, we show that shear wave splitting (SWS) of glacial microseismicity can be used to invert for seismic anisotropy and ice fabric at Rutford Ice Stream (RIS). RIS is a fast-flowing Antarctic ice stream, a setting crucial for informing flow models. We present ~2000,000 SWS measurements from glacial microseismicity, registered at a 38-station seismic network located ~40 km upstream the grounding line. A representative subset of this data is inverted for ice fabric. Due to the character of SWS, which accumulates along the ray path, our method works best if additional information on the depth structure of the ice is available, which are radar measurements in our case. We find that the following three-layer model fits the data best: a broad vertical cone near the base of RIS (500 m thick), a thick vertical girdle, orientated perpendicular to flow, in the middle (1200 m thick) and a tilted cone fabric in the uppermost 400 m. Such a fabric causes a depth-dependent strength profile of the ice with the middle layer being ~3.5 times harder to deform along flow than across flow. At the same time, the middle layer is a factor ~16 softer to shear than to compression or extension along flow. If such a configuration is representative for fast-flowing ice streams, it would call for a more complex integration of viscosity in ice sheet models.

Mapping fluid accumulation in the crust is pertinent for numerous applications including volcanic hazard assessment, geothermal energy generation and mineral exploration. Here, we use seismic attenuation tomography to map the distribution of fluids in the crust below Uturuncu volcano, Bolivia. Seismic P-wave and S-wave attenuation, as well as their ratio (QP/QS), constrain where the crust is partially and fully fluid-saturated. Seismic anisotropy observations further constrain the mechanism by which the fluids accumulate, predominantly along aligned faults and fractures in this case. Furthermore, subsurface pressure-temperature profiles and conductivity data allow us to identify the most likely fluid composition. We identify shallow regions of both dry and H2O/brine-saturated crust, as well as a deeper supercritical H2O/brine column directly beneath Uturuncu. Our observations provide a greater understanding of Uturuncu’s transcrustal hydrothermal system, and act as an example of how such methods could be applied to map crustal fluid pathways and hydrothermal/geothermal systems elsewhere.

Oceanic transform faults are intriguing in that they do not produce earthquakes as large as might be expected given their dimensions. We use 1-year of local seismicity recorded on an array of ocean bottom seismometers (OBS) and geophysical data to study the seismotectonic properties of the Chain transform, located in the equatorial Mid-Atlantic. We extend our analysis back in time by considering stronger earthquakes (MW ≥ 5.0) from global catalogs. We divide Chain into three areas (eastern, central, and western) based on multi-dimensional OBS seismicity cluster analysis. Seismic activity recorded by the OBS is the highest at the eastern area of Chain where there is a lozenge shaped topographic high, a negative rMBA gravity anomaly, and only a few historical MW ≥ 5.5 events. OBS seismicity rates are lower in the western and central areas. However, these areas accommodate the majority of seismic moment release, as inferred from both OBS and historical data. We find no evidence of remote dynamic triggering and only weak evidence of tidal and static stress triggering. Higher b-values are significantly correlated with lower rMBA and also with shallower bathymetry, potentially related to thickened crust. Our results suggest high lateral heterogeneity along Chain: Patches with moderate to low OBS seismicity rates that occasionally host MW ≥ 6.0 earthquakes are interrupted by segments with abundant OBS activity but few historical events with 5.5 ≤ MW < 6.0. This segmentation is possibly due to variable fluid circulation and alteration, which may also be variable in time.

Seismicity along transform faults provides important constraints for our understanding of the factors that control earthquake ruptures. Oceanic transform faults are particularly useful due to their relatively simple structure in comparison to continental counterparts. The seismicity of several fast-moving transform faults has been investigated by local networks, but as of today there have not been many studies of slower spreading centres. Here we present the first local seismicity catalogue based on event data recorded by a temporary broadband network of 39 ocean bottom seismometers located around the slow-moving Chain Transform Fault (CTF) along the Mid-Atlantic Ridge (MAR) from March 2016 to March 2017. Locations are constrained by simultaneously inverting for a 1-D velocity model informed by the event P- and S-arrival times. Depths and focal mechanisms of the larger events are refined using deviatoric moment tensor inversion. We find a total of 972 events in the area. Most of the seismicity is located at the CTF (700) and Romanche transform fault (94) and the MAR (155); a smaller number (23) can be observed on the continuing fracture zones or in intraplate locations. The ridge events are characterised by normal faulting and most of the transform events are characterised by strike slip faulting, but with several reverse mechanisms that are likely related to transpressional stresses in the region. CTF events range in magnitude from 1.1 to 5.6 with a magnitude of completeness around 2.3. Along the CTF we calculate a b-value of 0.81 ± 0.09. The event depths are mostly shallower than 15 km below sea level (523), but a small number of high-quality earthquakes (16) are located deeper, with some (8) located deeper than the brittle-ductile transition as predicted by the 600˚C-isotherm from a simple thermal model. The deeper events could be explained by the control of seawater infiltration on the brittle failure limit.