Search for: All records

Advances in machine learning (ML) techniques and computational capacity have yielded state‐of‐the‐art methodologies for processing, sorting, and analyzing large seismic data sets. In this study, we consider an application of ML for automatically identifying dominant types of impulsive seismicity contained in observations from a 34‐station broadband seismic array deployed on the Ross Ice Shelf (RIS), Antarctica from 2014 to 2017. The RIS seismic data contain signals and noise generated by many glaciological processes that are useful for monitoring the integrity and dynamics of ice shelves. Deep clustering was employed to efficiently investigate these signals. Deep clustering automatically groups signals into hypothetical classes without the need for manual labeling, allowing for the comparison of their signal characteristics and spatial and temporal distribution with potential source mechanisms. The method uses spectrograms as input and encodes their salient features into a lower‐dimensional latent representation using an autoencoder, a type of deep neural network. For comparison, two clustering methods are applied to the latent data: a Gaussian mixture model (GMM) and deep embedded clustering (DEC). Eight classes of dominant seismic signals were identified and compared with environmental data such as temperature, wind speed, tides, and sea ice concentration. The greatest seismicity levels occurred at the RIS front during the 2016 El Niño summer, and near grounding zones near the front throughout the deployment. We demonstrate the spatial and temporal association of certain classes of seismicity with seasonal changes at the RIS front, and with tidally driven seismicity at Roosevelt Island.

 

Strong surface winds under extratropical cyclones exert intense surface stresses on the ocean that lead to upper-ocean mixing, intensified heat fluxes, and the generation of waves, that, over time, lead to swell waves (longer than 10-s period) that travel long distances. Because low-frequency swell propagates faster than high-frequency swell, the frequency dependence of swell arrival times at a measurement site can be used to infer the distance and time that the wave has traveled from its generation site. This study presents a methodology that employs spectrograms of ocean swell from point observations on the Ross Ice Shelf (RIS) to verify the position of high wind speed areas over the Southern Ocean, and therefore of extratropical cyclones. The focus here is on the implementation and robustness of the methodology in order to lay the groundwork for future broad application to verify Southern Ocean storm positions from atmospheric reanalysis data. The method developed here combines linear swell dispersion with a parametric wave model to construct a time- and frequency-dependent model of the dispersed swell arrivals in spectrograms of seismic observations on the RIS. A two-step optimization procedure (deep learning) of gradient descent and Monte Carlo sampling allows detailed estimates of the parameter distributions, with robust estimates of swell origins. Median uncertainties of swell source locations are 110 km in radial distance and 2 h in time. The uncertainties are derived from RIS observations and the model, rather than an assumed distribution. This method is an example of supervised machine learning informed by physical first principles in order to facilitate parameter interpretation in the physical domain.

 

Understanding deformation in ice shelves is necessary to evaluate the response of ice shelves to thinning. We study microseismicity associated with ice shelf deformation using nine broadband seismographs deployed near a rift on the Ross Ice Shelf. From December 2014 to November 2016, we detect 5,948 icequakes generated by rift deformation. Locations were determined for 2,515 events using a least squares grid‐search and double‐difference algorithms. Ocean swell, infragravity waves, and a significant tsunami arrival do not affect seismicity. Instead, seismicity correlates with tidal phase on diurnal time scales and inversely correlates with air temperature on multiday and seasonal time scales. Spatial variability in tidal elevation tilts the ice shelf, and seismicity is concentrated while the shelf slopes downward toward the ice front. During especially cold periods, thermal stress and embrittlement enhance fracture along the rift. We propose that thermal stress and tidally driven gravitational stress produce rift seismicity with peak activity in the winter.

 

Gravity waves impacting ice shelves illicit a suite of responses that can affect ice shelf integrity. Broadband seismometers deployed on the Ross Ice Shelf, complemented by a near‐icefront seafloor hydrophone, establish the association of strong icequake activity with ocean gravity wave amplitudes (A_G) below 0.04 Hz. The Ross Ice Shelf‐front seismic vertical displacement amplitudes (A_SV) are well correlated withA_G, allowing estimating the frequency‐dependent transfer function from gravity wave amplitude to icefront vertical displacement amplitude (T_GSV(f)).T_GSV(f) is 0.6–0.7 at 0.001–0.01 Hz but decreases rapidly at higher frequencies. Seismicity of strong icequakes exhibits spatial and seasonal associations with different gravity wave frequency bands, with the strongest icequakes observed at the icefront primarily during the austral summer when sea ice is minimal and swell impacts are strongest.

 

Flexure and extension of ice shelves in response to incident ocean surface gravity waves have been linked to iceberg calving, rift growth, and even disintegration of ice shelves. Most modeling studies utilize a plate bending model for the ice, focusing exclusively on flexural gravity waves. Ross Ice shelf seismic data shows not only flexural gravity waves, with dominantly vertical displacements, but also extensional Lamb waves, which propagate much faster with dominantly horizontal displacements. Our objective is to model the full-wave response of ice shelves, including ocean compressibility, ice elasticity, and gravity. Our model is a 2D vertical cross-section of the ice shelf and sub-shelf ocean cavity. We quantify the frequency-dependent excitation of flexural gravity and extensional Lamb waves and provide a quantitative theory for extensional Lamb wave generation by the horizontal force imparted by pressure changes on the vertical ice shelf edge exerted by gravity waves. Our model predicts a horizontal to vertical displacement ratio that increases with decreasing frequency, with ratio equal to unity at ~0.001 Hz. Furthermore, in the very long period band (<0.003 Hz), tilt from flexural gravity waves provides an order of magnitude larger contribution to seismometer horizontal components than horizontal displacements from extensional Lamb waves. 

Abstract Ambient seismic recordings taken at broad locations across Ross Ice Shelf and a dense array near West Antarctic Ice Sheet (WAIS) Divide, Antarctica, show pervasive temporally variable resonance peaks associated with trapped seismic waves in near-surface firn layers. These resonance peaks feature splitting on the horizontal components, here interpreted as frequency-dependent anisotropy in the firn and underlying ice due to several overlapping mechanisms driven by ice flow. Frequency peak splitting magnitudes and fast/slow axes were systematically estimated at single stations using a novel algorithm and compared with good agreement with active source anisotropy measurements at WAIS Divide determined via active sources recorded on a 1 km circular array. The approach was further applied to the broad Ross Ice Shelf (RIS) array, where anisotropy axes were directly compared with visible surface features and ice shelf flow lines. The near-surface firn, depicted by anisotropy above 30 Hz, was shown to exhibit a novel plastic stretching mechanism of anisotropy, whereby the fast direction in snow aligns with accelerating ice shelf flow. 

Abstract Ocean swell interacting with Antarctic ice shelves produces sustained (approximately, 2×106 cycles per year) gravity-elastic perturbations with deformation amplitudes near the ice front as large as tens to hundreds of nanostrain. This process is the most energetically excited during the austral summer, when sea ice-induced swell attenuation is at a minimum. A 2014–2017 deployment of broadband seismographs on the Ross Ice shelf, which included three stations sited, approximately, 2 km from the ice front, reveals prolific swell-associated triggering of discrete near-ice-front (magnitude≲0) seismic subevents, for which we identify three generic types. During some strong swell episodes, subevent timing becomes sufficiently phase-locked with swell excitation, to create prominent harmonic features in spectra calculated across sufficiently lengthy time windows via a Dirac comb effect, for which we articulate a theoretical development for randomized interevent times. These events are observable at near-front stations, have dominant frequency content between 0.5 and 20 Hz, and, in many cases, show highly repetitive waveforms. Matched filtering detection and analysis shows that events occur at a low-background rate during all swell states, but become particularly strongly excited during large amplitude swell at rates of up to many thousands per day. The superimposed elastic energy from swell-triggered sources illuminates the shelf interior as extensional (elastic plate) Lamb waves that are observable more than 100 km from the ice edge. Seismic swarms show threshold excitation and hysteresis with respect to rising and falling swell excitation. This behavior is consistent with repeated seismogenic fracture excitation and growth within a near-ice-front damage zone, encompassing fracture features seen in satellite imagery. A much smaller population of distinctly larger near-front seismic events, previously noted to be weakly associated with extended periods of swell perturbation, likely indicate calving or other larger-scale ice failures near the shelf front. 

Abstract Ice shelves play a critical role in modulating dynamic loss of ice from the grounded portion of the Antarctic Ice Sheet and its contribution to sea-level rise. Measurements of ice-shelf motion provide insights into processes modifying buttressing. Here we investigate the effect of seasonal variability of basal melting on ice flow of Ross Ice Shelf. Velocities were measured from November 2015 to December 2016 at 12 GPS stations deployed from the ice front to 430 km upstream. The flow-parallel velocity anomaly at each station, relative to the annual mean, was small during early austral summer (November–January), negative during February–April, and positive during austral winter (May–September). The maximum velocity anomaly reached several metres per year at most stations. We used a 2-D ice-sheet model of the RIS and its grounded tributaries to explore the seasonal response of the ice sheet to time-varying basal melt rates. We find that melt-rate response to changes in summer upper-ocean heating near the ice front will affect the future flow of RIS and its tributary glaciers. However, modelled seasonal flow variations from increased summer basal melting near the ice front are much smaller than observed, suggesting that other as-yet-unidentified seasonal processes are currently dominant. 

Abstract The Ross Ice Shelf (RIS) is host to a broadband, multimode seismic wavefield that is excited in response to atmospheric, oceanic and solid Earth source processes. A 34-station broadband seismographic network installed on the RIS from late 2014 through early 2017 produced continuous vibrational observations of Earth's largest ice shelf at both floating and grounded locations. We characterize temporal and spatial variations in broadband ambient wavefield power, with a focus on period bands associated with primary (10–20 s) and secondary (5–10 s) microseism signals, and an oceanic source process near the ice front (0.4–4.0 s). Horizontal component signals on floating stations overwhelmingly reflect oceanic excitations year-round due to near-complete isolation from solid Earth shear waves. The spectrum at all periods is shown to be strongly modulated by the concentration of sea ice near the ice shelf front. Contiguous and extensive sea ice damps ocean wave coupling sufficiently so that wintertime background levels can approach or surpass those of land-sited stations in Antarctica.