We study Sn-to-Lg conversion at regional distances due to significant crustal thickening, particularly in the context of using Sn and Lg amplitude ratios (Sn/Lg) to identify continental mantle earthquakes. We further enhance recent developments in computational seismology to perform 2.5D simulations up to 5 Hz and 2,000 km. Our simulations compare propagation in a reference, constant-thickness crust from a source at three depths straddling the Moho, to 48 models of the same three sources propagating through Moho ramps of four different widths (dips) at four different distances from the source. We compare our synthetics to data from 12 earthquakes recorded on the HiCLIMB array across Tibet, of which six events from northwestern Tibet traverse no major crustal-thickness variation, and six located south of the Himalaya cross a major Moho ramp. Our observations on real data show that amplitude perturbations on individual Sn and Lg waves are smooth and mostly limited to near the ramp end. Even the more-pronounced amplitude variations seen in our simulations show that Sn/Lg for mid-crustal earthquakes is consistently lower than those for mantle earthquakes. Hence we can directly compare Sn/Lg for ramp-crossing and non-ramp-crossing earthquakes and identify new mantle earthquakes in northern India. Sn-to-Lg converted waves may be readily detected near the Moho ramp end through an enhancement in high-frequency content. In addition, we observe higher frequency content in Lg from crustal than from mantle earthquakes, which offers a new discriminant for continental mantle earthquakes based on frequency content of Lg waves alone.

We study Sn-to-Lg conversion at regional distances due to significant crustal thickening, particularly in the context of using Sn and Lg amplitude ratios (Sn/Lg) to identify continental mantle earthquakes. We further enhance recent developments in computational seismology to perform 2.5D simulations up to 5 Hz and 2,000 km. Our simulations compare propagation in a reference, constant-thickness crust from a source at three depths straddling the Moho, to 48 models of the same three sources propagating through Moho ramps of four different widths (dips) at four different distances from the source. We compare our synthetics to data from 12 earthquakes recorded on the HiCLIMB array across Tibet, of which six events from northwestern Tibet traverse no major crustal-thickness variation, and six located south of the Himalaya cross a major Moho ramp. Our observations on real data show that amplitude perturbations on individual Sn and Lg waves are smooth and mostly limited to near the ramp end. Even the more-pronounced amplitude variations seen in our simulations show that Sn/Lg for mid-crustal earthquakes is consistently lower than those for mantle earthquakes. Hence we can directly compare Sn/Lg for ramp-crossing and non-ramp-crossing earthquakes and identify new mantle earthquakes in northern India. Sn-to-Lg converted waves may be readily detected near the Moho ramp end through an enhancement in high-frequency content. In addition, we observe higher frequency content in Lg from crustal than from mantle earthquakes, which offers a new discriminant for continental mantle earthquakes based on frequency content of Lg waves alone.

We evaluate three identifiers of continental mantle earthquakes motivated by surface-wave normal-mode theory, the amplitude ratio of Sn to Lg, and the frequency content of Sn and of Lg, after wave propagation through continental crustal thinning. These methods are flexible and easy to apply, allowing for potential new discoveries of unusual continental mantle earthquakes, but they rely on guided waves whose propagation is dependent on the uniformity of their waveguides. For a range of Moho models, we perform 2.5D axisymmetric simulations that reach the conventional distance and frequency ranges of observational studies, and we compare results from four different source depths straddling the Moho. Using our synthetics and also six south-Tibet earthquakes recorded by an array in Bangladesh, we show that our Sn/Lg identifier is robust in the presence of crustal thinning on the continents, but the identifying frequency contents of Sn and Lg are easily obscured.

Earthquakes are frequently accompanied by electromagnetic (EM) anomalies. These anomalies are thought to be caused by earthquakes but the generation mechanism is still unclear. The piezoelectric effect has been proposed as a possible mechanism but the EM responses to earthquakes due to such an effect has not been well understood. In this article, we study the EM signals generated by an earthquake source due to the piezoelectric effect. We develop a semi-analytical method to solve the seismic and EM fields in a 3D layered model and conduct numerical simulations to investigate the characteristics of the EM fields. The results show that the earthquake can generate two kinds of EM signals. One is the early-EM signal which arrives earlier than the seismic wave. The other is the co-seismic EM signal accompanying the seismic wave. For an earthquake the co-seismic electric field can reach ~10 μV/m and the magnetic field can reach ~10-4 nT. We also study the sensitivity of the co-seismic EM fields to the rock conductivity. The results show that the co-seismic EM fields are mainly affected by the conductivity of the shallow layer, and they are also affected by the conductivity of the deep layer when the top layer is thin.

Whether deep earthquakes beneath the southern East African Rift, down to 44 ±4 km, occur above or below the Moho has remained controversial. We explore a new technique, relying on the amplitude ratios of Sn and Lg seismic waves, to test earthquake depths relative to the Moho. Sn and Lg waves propagate through reflections in the mantle lid and the crust, respectively, so Sn is only strongly excited by events with hypocenters below the Moho and Lg by events above the Moho (see presentations in S024: Wang and Klemperer, and Chen et al., this meeting). We first validated our Sn/Lg method with an earthquake accepted to be below the Moho in the offshore extension of the East African Rift System, a magnitude 4.6 event at 14.7 3.4 km in the Mesozoic oceanic crust of the Mozambique Channel. Even given the Miocene-to-Recent volcanism (associated with the offshore extension of the East African Rift System), this event - if its depth is correct - is almost certainly in the mantle. As predicted from our Sn/Lg models, regional seismograms for this event have strong Sn but essentially no Lg. We then applied our method to the 1998 m=4.7 earthquake beneath southern Lake Malawi to determine its position relative to the Moho. Yang & Chen (JGR, 2010) report it to be 44 km deep in a region where crustal thickness varies rapidly from ~35 km within the rift to ~45km beneath the adjacent craton. The published location beneath Lake Malawi suggests the earthquake may have occurred in the mantle but Craig et al. (GJI, 2011) note that it lies within error of the Moho. Our analysis shows rather small Sn but strong Lg energy, so we believe it is likely in the lower crust. Given our results, either the current depth is incorrect or the event is not beneath the axis of the rift. Neither solution easily explains the phases Yang & Chen described as underside PmP reflections, which were used to suggest this earthquake occurred below the Moho. A possible answer may lie in strong 3D structure around the hypocenter, which may either be affecting our Sn and Lg observations, or may offer a multi-pathing solution to the delayed “underside PmP” phases

The occurrence of earthquakes in the continental upper mantle is highly debated, and bears directly on lithospheric rheology (e.g. “jelly-sandwich” vs. “crème-brulée” models). Because Sn waves travel only below the Moho, a detection of high Sn amplitudes indicates that the source earthquake occurred below the Moho. In contrast, because Lg waves propagate in the crust, strong Lg amplitudes signify a crustal earthquake above the Moho. In this project, we use Sn/Lg amplitude ratios as evidence to support prior identifications of mantle earthquakes. We develop a novel workflow to analyze data from the IRIS database. Using the expected velocities of Sn and Lg waves, we calculate the RMS amplitude of the Sn and Lg windows, to determine the Sn/Lg amplitude for each recording. To validate our approach, we apply our methods to the well-recorded 2013 Mw 4.8 Wyoming event, reported to be 76 km deep, and the 2016 Mw 4.8 Wyoming event, just 12 km deep. Contrary to simple expectation, the deep Wyoming earthquake does not show a strong Sn/Lg ratio. However, the Sn/Lg amplitude ratio for the deep (upper-mantle) event is significantly larger than the equivalent ratio at the equivalent station for the shallow (upper-crustal) event (Figure 1). We apply the same algorithm to Californian events reported to be in the crust, near the Moho, and in the mantle, to test our methodology and theory. We present our results in the form of raypath maps and record sections for each earthquake that we studied. Our results show that the 1D assumptions of the Sn/Lg theory are successful in Wyoming and corroborate the existence of rare deep earthquakes, which indicates that some parts of the mantle possesses brittle properties like the upper crust. However, in other areas our 1D assumptions are insufficient, as shown by contrasting results from the East African Rift (see adjacent poster by Espinal et al.).

We studied seven earthquakes in the southern East African Rift System (EARS) with catalog depths of 10 to 33km, in locations where the Moho is thought to be at ~32 km depth (CRUST 1.0). Our earthquakes include three relocated by Yang and Chen (JGR, 2010) to be significantly deeper and to be below the Moho. We independently assessed whether the events occurred above or below the Moho using the Sn/Lg method (Wang et al., AGU Fall Meeting 2019; see also adjacent poster by Chen et al.). In a 1D earth, sub-Moho earthquakes produce strong Sn and weak Lg signals, and intra-crustal earthquakes produce weak Sn and strong Lg arrivals. All seven events we studied were characterized by low Sn/Lg, including the three earthquakes interpreted as upper-mantle events by Yang and Chen (2010) (their events M3 and M5 in Malawi and T12 in Zambia). Although low Sn/Lg is elsewhere associated with crustal events we suspect that, in the East African Rift, events in the shallow upper mantle that produce strong Sn at the source may be recorded at regional distances with low Sn/Lg due to Sn-to-Lg conversion at the deepening Moho at the rift margins. CRUST 1.0 suggests crustal thicknesses reach 45 km beneath the cratons adjacent to the East African Rift, with average Moho dips of 5-10°. Hence even the deepest earthquake reported by Yang and Chen (JGR, 2010), at 44±4 km, could undergo significant Sn-to-Lg conversion. Our findings highlight the importance of careful interpretation of Sn/Lg ratios and motivates our ongoing work to model 2D propagation effects.

Deep earthquakes in the lower continental lithosphere - the lower crust and uppermost mantle - are frequently too poorly located in depth to be definitively labelled as having occurred above or below the Mohorovičić discontinuity (Moho; base of the crust). Our Sn/Lg methodology utilizes two regional seismic waves to determine the depth of an earthquake relative to the Moho: Sn and Lg waves, which are phases that propagate through reflections in the mantle lid and the crust, respectively. Therefore, an analysis of Sn and Lg waves can provide a robust understanding of an earthquake’s depth relative to the Moho. We present our Sn/Lg analysis through reduced-velocity record sections, which show Sn and Lg energy in the waveform, allowing measurements of RMS amplitudes, and maps of ray-paths and Sn/Lg amplitude ratios which allow us to visualize the propagation of Sn and Lg in all directions. We demonstrate the efficacy of our approach by applying it to a well-known upper-mantle earthquake in Wyoming and a shallow earthquake with a similar epicenter. We then use our method to study other deep-crustal/upper-mantle earthquakes in North America. A cluster of earthquakes with reported depths from 10–50 km ± ~10 km, spans the border between Alberta, Canada and Montana, U.S.A. where the crustal thickness increases from ~30 km in the SW to ~45km in the NE. For an earthquake occurring in the crust-mantle transition zone, the dipping Moho should tend to block Sn (decrease the Sn/Lg ratio) in the direction of crustal thickening, and tend to block Lg (increase the Sn/Lg ratio) in the direction of crustal thinning. We studied seven earthquakes of magnitude>2.5. A m=2.7 earthquake, previously reported (USGS PDE) to be at 50±10 km where CRUST 1.0 shows a 49 km Moho depth, and a m=3.5 reported at 38 km depth above a nominal 43 km Moho, both show much stronger Sn/Lg ratios than earthquakes with nearby epicenters at nominal dots of 15 and 21 km. Hence the “50-km” and the “38-km” earthquakes must occur in the upper mantle, or so close to the upper mantle as to preferentially excite Sn. The better-recorded of these earthquakes also shows clear evidence of Sn enhancement along azimuths into regions of thinner crust.