Introduction
Metastable Olivine Wedge Beneath the Japan Sea Imaged by
Abstract
Here, we use cross-source interferometry, which converts deep earthquakes into virtual seismometers, to detect the MOW seismic signature without the influence of shallow heterogeneities. Using data from Hi-net, we confirm the existence of a MOW beneath the Sea of Japan and constrain its geometry to be ~30 km thick at a depth of 410 km and gradually thin to a depth of at least 580 km.
Introduction
Laboratory experiments have shown that the incorporation of a small amount of H2O leads to a marked increase in the rate of transformation of olivine into ringwoodite through a hydrolytic attenuation process (Du Frane et al., 2013). A thermal plate without MOW could satisfactorily predict high-resolution seismic arrival times, but the inclusion of MOW offers only a subtle improvement over the data fit (Koper et al., 1998).
Inter-source Interferometry
The right panel shows the optimal focal depth of the individual station searching for the highest cross-correlation coefficient. Slab velocity and density profiles and the MOW are constructed based on a thermal model tuned for the Japan subduction zone (see section 2.6 for thermal modeling details).
Results
Next, with the real data from Hi-net, we obtain the 0.2-5 Hz strain responses for all six earthquake pairs from D1/D2 to S1/S2/S3 following the same interferometry procedures (Figure 2.13). Furthermore, our MOW model also provides a good fit to the interferometric observations at three virtual receivers from the other deep earthquake D2 (Figure 2.14).
Discussion and Conclusions
The depth range of our proposed MOW shows an extremely dry Pacific slab core (<75 wt ppm) in the MTZ beneath the Sea of Japan (Figure 2.20; Kawakatsu and Yoshioka, 2011; Du Frane et al., 2013). Therefore, water associated with these faults must be almost completely expelled into the mantle at intermediate depths (Kawakatsu and Watada, 2007), carrying negligible amounts of water into the MTZ (Green II et al., 2010). Or rather than relating the water reservoir in the MTZ to actual subduction, the wet MTZ may be associated with other tectonic processes (Green II et al., 2010; Hirschmann, 2006).
Appendix
In the calculation, the velocity model was constructed using a combination of Crust1.0 (Laske et al., 2013) and IASP91 (Kennett and Engdahl, 1991). For the horizontal locations D1 and S1, we took the results from the ISC-EHB catalog, on the basis of which the Slab2.0 model was created (Hayes et al., 2018). The maximum intraplate P and S wave velocity perturbations are 4.5% and 6.0%, respectively, which are in good agreement with the inferred velocity anomalies from seismic wave studies ( Zhan et al., 2014a ).
Acknowledgement
Recent progress of seismic observation networks in Japan – Hinet, F-net, K-NET and KiK-net. In: Earth, planets and space56.8. Due to the plate-parallel ray paths of the six earthquake pairs (Figure 3.1), the 5 Hz waveforms are more sensitive to Melt-type models than to Fault-type models. Therefore, we propose that the elongated within-plate heterogeneity decreases from 2.5% to <1.0% in the MTZ under the Sea of Japan (Figure 3.11).
Small-Scale Intra-slab Heterogeneity Weakens Into the Mantle
Abstract
Small-scale intraplate heterogeneity is well documented seismically in multiple subduction zones, but its nature remains elusive. Here, we illustrate that the intersource interferometry method, which turns deep earthquakes into virtual receivers, can resolve small-scale intraslab heterogeneity in the mantle transition zone. Combined with previous studies that suggest high heterogeneity (∼2.5%) at intermediate depths, we conclude that small-scale within-slab heterogeneity weakens as slabs descend.
Introduction
Nevertheless, given the limited seismological observations in the mid-depth range, there is no widely accepted petrological understanding of small-scale intraplate heterogeneity. In this paper, we first validate the cross-source interferometry technique for small-scale intraplate heterogeneity through numerical simulations. Next, we invoke different models of intraplate heterogeneity in the MTZ beneath the Sea of Japan to explain the interferometric observations.
Deep Slab Model for Inter-source Interferometry
The black solid and dashed lines show the subducting Pacific plate (Hayes et al., 2018) and a MOW model under the Sea of Japan (Shen and Zhan, 2020). To describe the small-scale heterogeneity, we adopt the Von K´𝑎rm ´𝑎n-type autocorrelation function (ACF) given as (Sato et al., 2012). We also increase linearly from the plate surface (0.5%) to the bottom (2.5%), called the Melt_basal model.
Synthetic Test
Following the same procedure, we also compare the stacked cross-correlations with theoretical strain waveforms for other intra-slab heterogeneity models. Interestingly, regardless of the type of intra-slab heterogeneity, the long period (0.2–2 Hz) interferometric waveforms do not differ significantly in waveform, confirming that MOW (the controlling factor in this frequency band) is required for to match the interferometry. -source interferometric observations in Japan (Figure 3.1b; Shen and Zhan, 2020). Therefore, the 5 Hz interferometric waveforms are essential to constrain the small-scale intra-slab heterogeneity in the MTZ.
Results
Compared to the Melt models, the Fault-type models show similar but less complex waveform variations for heterogeneity level 𝜀 greater than 1.5% (Figure 3.3b). Similar to the standard tests above, the complexity of the high-frequency strain waveform of the D1-S1 pair is related to the scattering strength (𝜀) for the Melt-type models, but shows little dependence on the heterogeneity level of the Fault-type models ( Figure 3.4 a). Moreover, the probability of the Melt_constant0.5 and Fault_constant1.0 models does not exceed 30% (Figure 3.8), so the upper limit of the level of intraplate heterogeneity below 410 km is probably less than 0.5%.
Discussion
The level of heterogeneity within the plate below 410 km is indicated on the left side of each seismogram. The inter-plate seismic scatterers at intermediate depths along this plate (Furumura and Kennett, 2005) can therefore be explained by the short scale tortuous faults (Garth and Rietbrock, 2014b). Given the existence of a MOW (Shen and Zhan, 2020) requiring an extremely dry sheet core (<75 wt-ppm) (Du Frane et al., 2013; Kawakatsu and Yoshioka, 2011) and the lack of a P interplate wave coda in the MTZ (this study), the dehydration process of the plate core is nearly complete by 410 km.
Conclusion
As the slab subducts, elevated pressure and temperature cause a series of hydrous mineral dissolution reactions that secrete aqueous fluid to fertilize the mantle wedge (Van Keken et al., 2011). Water released from the breakdown of serpentinite can gradually seep upward along intra-plate strata characterized by a tectonic stress gradient (Faccenda et al., 2012). On the contrary, our observations shed light on the dehydration process of slab cores, which could lead to (re)hydration of the slab crust and the formation of dense oxyhydroxide phases (e.g. Buchen et al., 2021; Karato, 2006; Ohira et al., 2019; Ohtani, 2020) .
Acknowledgement
For example, reconstructing the plate velocity based on the plate temperature profile (Lu et al., 2019; Shen and Zhan, 2020). Due to the lack of plate expansion effect in scenarios without high-velocity plates (e.g., plate models with 0.0% velocity perturbation), the values for 𝑡∗. But we should also point out the challenge of global applications of the plate operator method.
Estimating the Slab Seismic Velocity Perturbation Using a Slab
Abstract
Given that high velocity plates act similarly to a weakening operator that broadens seismograms, we develop a plate operator method that uses teleseismic waveforms from subduction zone earthquakes. With synthetic 2D tests, we demonstrate that the plate operator method is able to measure plate velocity perturbation via an apparent attenuation factor (𝑡∗ . 𝑠) that is insensitive to complicated earthquake rupture processes. Applying this technique to the Kuril subduction zone, we determine a velocity amplitude of 4% for the slab core, which is close to previous values derived from waveform modeling.
Introduction
Accurate seismic structure, at least in the upper mantle, is a prerequisite for inferring the thermal state of the subducting slab (Cammarano et al., 2003; Cammarano et al., 2009; Stixrude and Lithgow-Bertelloni, 2012). To reduce the effect of earthquake mislocation, Lu et al. 2019) simultaneously inverted seismic velocity and source locations. Using such waveform sensitivity, Zhan et al. 2014a) suggested a 4% faster slab core than ambient mantle in the Kurile subduction zone.
Methodology: Slab Operator
Applying this technique to the Kuril subduction zone, we estimate a velocity amplitude of ∼4% for the stick core, which is close to previous values derived from waveform modeling. So, the Green's function𝐺𝐴𝑖 can be estimated as:. where𝜂𝑠𝑙 𝑎 𝑏 represents the sheet effect on the Green's function and is referred to hereafter as the sheet operator. Therefore, to describe the slab operator 𝜂𝑠𝑙 𝑎 𝑏 analytically, we adopt a modified Futterman damping function that focuses exclusively on the waveforms (Shearer, 2019).
Synthetic Test
This is probably an artifact due to the broadened waveform of 𝑈𝐴 𝑗 in the plate operator method that contradicts our second assumption (Equation 4.12). Since both 𝑊𝑟 𝑒 𝑓 and 𝑊𝑚 𝑎 𝑠𝑡 𝑒𝑟 contain the event source terms 𝐴 and 𝐵 (equations 4.11 and 4.14), our plate operator method is theoretically capable of solving the plate velocity anomaly by using teleseismic waveforms with complex source-time functions. Therefore, our synthetic tests confirm that our plate operator method can resolve the plate speed discrepancy using 𝑡∗.
Application
With abundant deep earthquakes across the globe (Figure 4.11), the slab operator method provides an efficient way to accurately evaluate the relationship between slab thermal state and corresponding seismic velocity anomaly. For example, the teleseismic distances of deep earthquakes in South America are in the Atlantic Ocean. While for the cold Fiji-Tonga subduction, most teleseismic areas are covered by the Indian Ocean (Figure 4.11).
Conclusions
In: Physics of Earth and the Planetary Interiors 306, p. 2012). Seismic wave propagation and scattering in the heterogeneous Earth. The bottom panel shows the two-year stacked cross-correlation record section for a virtual source in the center of the array. In addition, the frequency analysis of dv/v at the western end of the DAS cable yields similar results (Figure 5.8), amplifying a seasonal manifestation from the top 10 meters.
High-resolution Vadose Zone Water Saturation Monitoring Using
Abstract
Water in the critical zone is vital, but current monitoring techniques mainly focus on surface water content, with few surveying the vadose zone. The emerging distributed acoustic sensing (DAS) technology offers an affordable and scalable solution for deploying ultra-dense, large-aperture seismic arrays, thus making it feasible for vadose zone monitoring. With two years of ambient sound recorded on the Ridgecrest DAS array in California, the resulting seismic changes (dv/v) reveal an unprecedented high-resolution spatiotemporal evolution of water saturation in the vadose zone.
Introduction
In recent decades, enormous advances in technology from various disciplines have enabled critical area hydrological monitoring (Parsekian et al., 2015), which has improved our vision of water cycle dynamics (Brooks et al., 2015; Sprenger et al., 2019). Alternatively, given the diffuse noise field, seismic interferometry can extract the Green's function of seismic waves propagating from one seismometer to another (Campillo and Paul, 2003; Shapiro et al., 2005). As an array of tens of kilometers of aperture with channel spacing of several meters, DAS can record high-frequency wavefields offline (Atterholt et al., 2022), which have great potential to increase the spatiotemporal resolution of long-term subsurface monitoring (Tribaldos and Ajo-Franklin, 2021).
Data and Methods
Here we use only an 8 km linear segment along an east-west oriented main road to provide Rayleigh-type dominated surface waves (Figure 5.1b). These collected surface waves have been successfully exploited to image subsurface seismic structure and identify fault zones in the Ridgecrest (Yang et al., 2022). The cyan waveform represents the crosstalk function for a particular receiver that is 60 channels away from the virtual source.
Results
In response to rainfall, dv/v can temporarily drop and leave horizontal deviations across the DAS cable (Figure 5.3b vs. 5.3c). To quantify its depth extent, we calculate sensitivity kernels for a set of 10-m layers at depths ranging from the surface to 70 m at 10-m intervals (Figure 5.7b). To intuitively compare the depth sensitivity kernels with observations, we focus on slope changes of dv/v as a function of frequency by scaling dv/v to the same values at 2.93 Hz (Figure 5.7b).
Discussion
Similarly, rapid temperature changes can affect water saturation in the root zone through evaporation (Figure 5.11a and 5.11c). 99 variation with a contribution to the dv/v amplitude of up to 0.3% (Figure 5.11b), favoring a seasonal source in the vadose zone. Meanwhile, during dry seasons, evaporation will drain water saturation from the vadose zone, leading to a positive dv/v (Figure 5.11b vs. 5.11c).
Conclusion
In this way, we suggest that more water can be stored in the vadose zone during wet seasons, resulting in negative dv/v (Figure 5.11b vs. 5.11c). Nevertheless, surface soil moisture and seasonal dv/v measurements appear to have a phase shift (Figure 5.11b), which may indicate an important role for evapotranspiration diffusing through some thickness. Still, more detailed studies are required to quantitatively link our dv/v observations to the water cycle in the vadose zone.
Acknowledgement
Assessment of the resolution and accuracy of the Moving Window Cross Spectral technique for monitoring temporal variations in the Earth's crust using environmental seismic noise. An overview of the available methods for estimating soil moisture and its implications for water management. For example, the seismic structure and dynamics of the plate core at intermediate depths (70–350 km) are still poorly constrained, leaving the origins of lower-plane seismicity mysterious (for example, desiccation versus thermal breakout).
Conclusion and Future Direction
