No abstract

In this study, we undertake a renewed investigation of up-bent reflections seen in seismic time sections from the Baltic Sea, Bay of Kiel. These warped reflections stretch over the entire vertical extent of the sections, from Permian to Quaternary strata, and underlie tunnel valleys. Previous studies interpreted these structures as anticlines, explaining them together with adjacent faults and disrupted strata as the consequence of ice-load-induced salt tectonics. This conclusion would have influenced theories on how tunnel valleys formed. However, well data from tunnel valleys in other regions supported the interpretation of the up-bent reflections as imaging artefacts (pull-ups). A newly acquired long-offset, multichannel seismic data set images all strata from Base Zechstein up to the seafloor. Owing to the length of the streamer and a shallow water depth, the data display significant moveout and refracted waves, allowing the application of different quantitative methods to investigate velocities. By generating partialoffset sections, we reveal an offset dependence in the imaging of the up-bent structures caused by a local, near-surface high-velocity zone. This also explains a smoothing of the up-bending with depth in the seismic image. A velocity model gained by a traveltime tomography shows positive velocity anomalies in the upper strata correlating with tunnel valleys resolved in the reflection images. A pre-stack depth migration performed with a velocity model containing a high-velocity zone results in a seismic image almost free of the observed up-bent reflections. High-frequency reflection seismic data confirm this result as it shows a detailed image of a tunnel valley with a phase-reversed bottom reflection caused by the velocity inversion at the base of the high-velocity valley fill deposits. Hence, there is consistent evidence that all up-bent reflections in the Bay of Kiel are imaging artefacts (pull-ups) that formed beneath tunnel valleys. A salt tectonic control on tunnel valley evolution is, consequently, not likely. This study is the first purely seismic data-driven study that proves high-velocity valley fill deposits. Our findings imply that extra care must be taken when interpreting reflection undulations as tectonic features where glacial deposits are present.

Seismic waveforms are governed by the physical properties of the subsurface in which they propagate. Most modern conventional seismic imaging methods utilize only a small proportion of the seismic waveform, neglecting a wealth of additional information hidden within the full coda (Yilmaz, 2001). Full-waveform inversion (FWI) utilizes the entire seismic coda to produce accurate quantitative models of these governing properties (P-and S-wave velocity, attenuation, density, and anisotropy) (Tromp, 2020). First theorized in the 1980s, the FWI method looks to match the observed wavefield by iteratively updating a starting model, using linearized local inversion, to solve the full nonlinear inversion problem (Lailly & Bednar, 1983;Tarantola, 1984;Virieux & Operto, 2009). However, it took three decades for FWI to become widely utilized, particularly to higher frequencies because of its high computational requirements. Each doubling of the maximum FWI frequency requires the node spacing and sample interval to be halved, leading to an eight and 16 times increase in compute time in 2D and 3D, respectively, making high-frequency FWI prohibitively expensive. In the meantime, seismic reflection imaging has been the prime method for active-source subsurface imaging and exploration (Yilmaz, 2001).

Recurring slow slip along near-trench megathrust faults occurs at many subduction zones, but for unknown reasons, this process is not universal. Fluid overpressures are implicated in encouraging slow slip; however, links between slow slip, fluid content, and hydrogeology remain poorly known in natural systems. Three-dimensional seismic imaging and ocean drilling at the Hikurangi margin reveal a widespread and previously unknown fluid reservoir within the extensively hydrated (up to 47 vol % H 2 O) volcanic upper crust of the subducting Hikurangi Plateau large igneous province. This ~1.5 km thick volcaniclastic upper crust readily dewaters with subduction but retains half of its fluid content upon reaching regions with well-characterized slow slip. We suggest that volcaniclastic-rich upper crust at volcanic plateaus and seamounts is a major source of water that contributes to the fluid budget in subduction zones and may drive fluid overpressures along the megathrust that give rise to frequent shallow slow slip.