Abstract: Ice shelf collapse can herald subsequent grounded ice instability. However, robust understanding of external mechanisms capable of triggering rapid changes remains elusive. Improved understanding therefore requires improved remote and in-situ measurements of ice shelf properties and the environmental processes that can affect them. Using nearly three years of continuous data from a recently deployed 34-station broadband seismic array on the Ross Ice Shelf, we analyze persistent temporally varying, anisotropic near-surface resonant wave modes at frequencies above ~1 Hz that are highly sensitive to small changes in elastic shelf properties to depths of several m. These modes exhibit both progressive (on the scale of months) and rapid (on the scale of hours) amplitude and frequency changes. The largest and most rapid excursions are associated with local storms and with a large regional ice shelf melt event in January 2016, but changes are apparent throughout the records, which we attribute related to abrupt and progressive structural evolution of the firn layer. These modal variations can be forward modeled and inverted for structural changes using synthetic spectra calculated via numerical wave propagation, and thus present an opportunity for 4-D structural monitoring of local shallow glacial elasticity and structure using continuous high-frequency seismic recordings.
Slidecast:
https://vimeo.com/277185719
Abstract: The Texas Seismological Network, established by the 84th session of the Texas State Legislature, currently consists of 25 stations that comprise a backbone network across the state, supplemented by several portable arrays in four areas of localized seismicity. All portable installations have both a seismometer and an accelerometer. One array, located in the Snyder-Cogdell area, consists of 7 portable stations with interspatial distances varying between 2 and 25 km. Another, located in the Delaware basin, consists of 7 portable stations in addition to 2 permanent stations located in that area. Interstation distances in that region are 50-80 km. To better understand the site amplification for stations in west Texas, and get qualitative information of the site classification, we perform site characterization using the horizontal-to-vertical spectral ratio (HVSR) for earthquake and noise data. For these areas, we focus our investigation on earthquakes with local magnitudes that range between 1.5 and 3.1. For earthquakes, we analyzed data starting at the S arrival and extending to the point where the signal dropped to normal background noise levels. For noise analysis, we used 24 hour records for each site, with earthquakes removed. Our analysis is concentrated on the frequency range between 1 and 50 Hz. We find consistency in HVSR values across multiple earthquakes for the same stations with peaks that indicate soil resonant frequencies. Additionally, we observe a correlation between the resonant frequencies identified for stations within similar geological formations, indicating a transition from higher to lower frequencies moving from stiffer to looser materials. This trend is supported by results from the noise data.
Slidecast:
https://vimeo.com/277188088
Abstract: The Tsunami Ready Pilot Project is a community performance based recognition project for coastal communities interested in strengthening their tsunami preparedness, adapting and adopting good practices to minimize loss of life and property. Ten important elements are considered in the pilot project guidelines, divided in three major components: mitigation, preparedness and response. Grenada is a tri-island state – Grenada, Carriacou and Petite Martinique – located in the southern Caribbean, with a total estimated population of 108,339 people for 2018. History shows that tsunamis have affected Grenada during the past centuries. In 1867, a magnitude 7.5 earthquake originated in the Anegada Trough – USVI, produced a tsunami that reached the island with a registered wave height of 3.1 m (10.2 ft) at Gouyave. Records also show that in 1902 pyroclastic flows from the eruption of the Soufriere Volcano of St. Vincent, could have generated the waves of up to 1 m (3.3 ft) that were observed in harbors of Grenada and other neighnoring islands. To fulfill the mitigation component of the Tsunami Ready Guidelines, tsunami hazards for Grenada were identified using ComMIT (tsunami modeling software). Seven different earthquake sources along the Lesser Antilles, Venezuela/Trinidad, and the South Caribbean Deformed Belt were modeled as well as a simulated wave produced by the Kick em’ Jenny (KeJ) submarine volcano. The St. Patrick parish located north to Grenada, was selected for piloting the Tsunami Ready project due to the proximity of this community to the KeJ volcano and their interest in preparedness for Tsunamis. During the CaribeWave 2017 exercise 21,000 people from the community were registered as participant. Continuing the preparedness steps for tsunamis, an evacuation map with tsunami evacuation routes and assembly points were determined and approved thru community engagement. Tsunami signage artwork and locations were also discussed with locals in a meeting organized by the National Disaster Management Agency. The Tsunami Plan with Standard Operating Procedures needs to be completed to receive the verification visit of the National Tsunami Board. Hopefully during the current semester, St. Patrick will be designated as the first Tsunami Ready community in Grenada and the first one considering a volcano as tsunami source.
Poster:
Implementation of the Tsunami Ready Pilot Project_ SSA18
Abstract: Auto-correlation of teleseismic P-wave coda is a recently developed technique capable of imaging crustal-scale features based on reflection signals from seismic discontinuities. Traditional P-wave receiver function analysis maps velocity contrasts based on P-to-S conversions and has been well utilized in a great many locations, but encounters difficulties in seismically complex environments: for example, where high impedance contrasts result in large magnitude reverberations that overwhelm the primary phases, or where high velocity layers or sloping interfaces invalidate assumptions of near-normal incidence. Deconvolution and multiple constraint alogrithms have been developed to address these issues but often introduce additional complexities and computational overhead. Auto-correlation has the advantages of being fast (when computed via the Cross Correlation Theorem), is not reliant on tuning parameters (e.g., water levels), directly exploits reverbatory phases for locating reflection horizons, and can be combined with a transformation to PSH wavevector coordinates to fully partition P- and S-wave energy into separate channels for joint inversion matching. We present a benchmark comparison of receiver function and auto-correlation inversions utilizing the first year of data from a 39 station broadband array spanning the Mackenzie Mountains, Yukon, Canada. Initial surveys of this region indicate a variety of seismically complicated crustal-scale structures, including large-scale strike-slip faulting, sloping interfaces, high velocity intrusions, and low velocity cratonic sediments. We also present application of the autocorrelation methodology to floating and grounded ice stations in Antarctica, where receiver function methods encounter particular difficulties due to the icecap and ocean.
Poster:
MGB_SSA_2018.compressed
Abstract: The presence of Tectonic Earthquake Swarms (TES) and sequences in the north and northeast of the island of Puerto Rico in the northeastern Caribbean have been recorded by the Puerto Rico Seismic Network (PRSN) since it started operations in 1974. A great wealth of seismic data has been archived during this period such that at least 10 years of catalog data can be used to characterize the seismic activity in the Puerto Rico-Virgin Islands (PRVI) region. This research uses two declustering methods to identify cluster events in the PRVI block. The first method, known as Model Independent Stochastic Declustering (MISD), filters the catalog sub-set into cluster and background seismic events. The second method uses a spatio-temporal algorithm applied to the catalog in order to link the separate seismic events into clusters. After using these two methods, identified clusters were classified into either earthquake swarms or seismic sequences. Results have allowed to identify and classify 128 clusters categorized in 11 distinctive regions based on their centers, and their spatio-temporal distribution have been used to determine interplate dynamics. This analysis yields implication on the behavior of the plate interface between the North American and Caribbean plates along the Puerto Rico trench.
Poster:
Characterizing_the_Temporal_and_Spatial_Distribution_of_Earthquake_Swarms_in_the_Puerto_Rico___Virgin_Island_Block (1)
Abstract: The epicentral region of the April 16, 2016 Mw 7.8 Pedernales earthquake lies north of the intersection of the Carnegie ridge and the Ecuador subduction zone. This segment has ruptured on decadal time scales in a series of megathrust events, Mw 8.8 (1906), Mw 7.8 (1942), Mw 7.7 (1958), and Mw 8.2 (1979), suggesting that accumulated strain is released through large earthquakes clustered in space and time. The 2016 rupture coincides with the rupture area of the Mw 7.8 1942 earthquake. Seismic data recorded by a dense array of 82 stations from Ecuador’s national network and an international rapid response effort were processed and analyzed. Multiple combinations of detection and association algorithms were assessed to address the challenge of producing robust earthquake catalogs from aftershock sequences. Calibrated relocations using the Hypocentroidal Decomposition approach were determined for a subset of events for which we combine phase readings from local and temporary stations with regional and teleseismic phase readings from the NEIC. Preliminary aftershock locations through August 2017 (11502 events, ≥ 6 phases, errors <10km, Mc1.8) show a sharp downdip limit of aftershocks and are characterized by a series of event clusters. Two distinct streaks of aftershocks outline the north and south ends of the rupture, and a third streak occurs between the two patches of slip defined by finite fault models. Aftershocks north of the main rupture are sparse until an earthquake swarm ~2 months after the mainshock in the vicinity of Esmeraldas, the portion of the subduction zone that ruptured in 1958. Eight months after the mainshock, a series of aftershocks in the Atacames area, north of the main rupture, caused additional structural damage. The spatial distribution of background seismicity and aftershocks is quite similar suggesting that background seismicity likely includes long-lived aftershocks from the ’06, ‘42, and ’58 large earthquakes as well as swarms associated with slow-slip events.
Poster:
LSC_SSA_mtg_v4
