Abstract Subduction zones host some of Earth's most damaging natural hazards, including megathrust earthquakes and earthquake‐induced tsunamis. A major control on the initiation and rupture characteristics of subduction megathrust earthquakes is how the coupled zone along the subduction interface accumulates elastic strain between events. We present results from observations of slow slip events (SSEs) in Cascadia occurring during the interseismic period downdip of the fully coupled zone, which imply that the orientation of strain accumulation within the coupled zone can vary with depth. Interseismic GPS motions suggest that forces derived from relative plate motions across a shallow, offshore locked plate interface dominate over decadal timescales. Deeper on the plate interface, below the locked (seismogenic) patch, slip during SSEs dominantly occurs in the updip direction, reflecting a dip‐parallel force acting on the slab, such as slab pull. This implies that in subduction zones with obliquely convergent plate motions, the seismogenic zone of the megathrust is loaded by forces acting in two discrete directions, leading to a depth‐varying orientation of strain accumulation on the plate interface.
more »
« less
The signature of accumulated permanent uplift, northern Cascadia subduction zone
Uplift of the overriding plate at a subduction zone denotes interseismic strain accumulation, which is subsequently released during a megathrust earthquake. Although most interseismic strain is thought to be released elastically, observations of uplifted coastal regions at subduction zones worldwide indicate that some strain may result in permanent uplift. The Grays Harbor and Willapa Bay (Washington, USA) coastal region of the Cascadia subduction zone hosts flights of marine terraces testifying to late Pleistocene rock uplift. Our new detailed mapping of the marine terraces recognizes nine new units, including estuarine and fluvial sediments. Luminescence dating, relative age based on soil maturity and terrace elevation, and an evaluation of previous ages from fossil shells collectively constrain the probable ages of three estuarine units to sea-level high stands during Marine Isotope Stages 5a, 5c, and 5e. We estimate an average uplift rate of 0.4 ± 0.1 mm/yr for the terraced estuarine units, consistent with other Pleistocene uplift and incision rates in Cascadia. When compared with observed interseismic vertical deformation, these rates suggest that about one-tenth of interseismic strain may become permanent. The values are permissible within the uncertainties of uplift based on regional estimates of interseismic vertical strain rates and of coseismic subsidence.
more »
« less
- Award ID(s):
- 2103713
- PAR ID:
- 10579329
- Publisher / Repository:
- Cambridge University Press
- Date Published:
- Journal Name:
- Quaternary Research
- Volume:
- 117
- ISSN:
- 0033-5894
- Page Range / eLocation ID:
- 98 to 118
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Landslides are a significant hazard and dominant feature throughout the landscape of the Pacific Northwest. However, the hazard and risk posed by coseismic landslides triggered by great Cascadia Subduction Zone (CSZ) earthquakes is highly uncertain due to a lack of local and global data. Despite a wealth of other geologic evidence for past earthquakes on the Cascadia Subduction Zone, no landslides have been definitively linked to such earthquakes, even in areas otherwise highly susceptible to failure. While shallow landslides may not leave a lasting topographical signature in the landscape, there are thousands of deep-seated landslides in Cascadia, and these deposits often persist for hundreds of years and multiple earthquake cycles. Synthesizing newly developed inventories of dated large deep-seated landslides in the Oregon Coast Range, we use statistical methods to estimate the proportion of these types of landslides that could have been triggered during past great Cascadia Subduction Zone earthquakes. Statistical analysis of high-precision dendrochronology ages of landslide-dammed lakes and surface roughness-dated bedrock landslides reveal Cascadia Subduction Zone earthquakes may have triggered 0–15 % of large deep-seated landslides in the Oregon Coast Range over multiple earthquake cycles. Our results refine estimates from previous studies and further suggest that coseismic triggering accounts for a small fraction of the total deep-seated bedrock landslides mapped in coastal Cascadia. However, if the real rate of coseismic landslide triggering during CSZ earthquakes is near our estimated upper bound for the 1700 CSZ earthquake, we estimate up to 2400 coseismic large deep-seated landslides could occur in the Oregon Coast Range in a single earthquake. These findings suggest Cascadia is consistent with global observations from other subduction zones and that coseismic landslides may still represent a serious geohazard in the region.more » « less
-
Abstract The details of subduction zone locking place constraints on the characteristics of megathrust events. Due to the lack of significant present‐day seismicity along the Cascadia subduction interface, geodetic data are used to assess subduction locking along the margin. We isolate the subduction signal from other tectonic signals within the Cascadia GPS field, to assess the details of plate‐interface locking. Apparent coupling determined by a simple homogenous elastic half‐space inversion cannot everywhere reproduce the subduction component of the GPS field. Consequently, we explore the relationships among upper‐plate strength, locking depth and the resulting surface velocity signal using 2D finite element models. When the upper plate is composed of a weak material, trenchward of a strong backstop, we find that the down‐dip limit of locking relative to the location of the weak‐to‐strong transition controls how upper‐plate deformation is spatially distributed. If locking extends into the stronger material, as observed in central Cascadia, the surface velocity field propagates farther inland than expected from a simple homogeneous elastic model. In contrast, in southern Cascadia, because locking terminates within the weak accretionary margin, upper‐plate shortening is localized within the weaker material, particularly in the region between the end of locking and the strong Klamath terrane. This behavior provides a possible mechanism for producing the high (geodetic and permanent) uplift rates, plate‐motion‐parallel shortening, and crustal exhumation observed in many active and fossil subduction zones.more » « less
-
Abstract At subduction zones, the down‐dip limit of slip represents how deep an earthquake can rupture. For hazards it is important ‐ it controls the intensity of shaking and the pattern of coseismic uplift and subsidence. In the Cascadia Subduction Zone, because no large magnitude events have been observed in instrumental times, the limit is inferred from geological estimates of coastal subsidence during previous earthquakes; it is typically assumed to coincide approximately with the coastline. This is at odds with geodetic coupling models as it leaves residual slip deficits unaccommodated on a large swath of the megathrust. Here we will show that ruptures can penetrate deeper into the megathrust and still produce coastal subsidence provided slip decreases with depth. We will discuss the impacts of this on expected shaking intensities.more » « less
-
Abstract River terraces are commonly used to infer climate and tectonic histories. Yet, it is increasingly recognised that other processes, such as river capture, can affect river terrace genesis and incision rates and patterns. In this study, we conduct a field‐based investigation of river terrace sequences along the Kolokithas and Varitis Rivers in central Crete, Greece, that share a confluence and preserve geomorphic evidence for the recent capture of the Kolokithas headwaters by the Varitis. We use digital topographic analysis, mapping, and optically stimulated luminescence (OSL) geochronology to quantify the river terrace and bedrock incision response to river capture. Topographic analysis indicates the Varitis captured ~30 km2of drainage area from the Kolokithas. We find differences in terrace characteristics, number of terraces, and incision rates and patterns on the adjacent valleys. The Kolokithas has four terrace levels, and the Varitis has five. All terraces are strath terraces, except for the oldest on the Kolokithas, a ~8 m thick fill terrace that starkly contrasts the time‐equivalent ~1–2 m thick strath terrace on the Varitis. Relative and absolute age control suggests three Pleistocene terraces were emplaced during cooler climate intervals, and two Holocene terraces are perhaps because of anthropogenic disturbances. The incision patterns differ on each valley, with generally more incision upstream on the Varitis relative to the Kolokithas. Incision rates on the Varitis are roughly twice as high as on the Kolokithas, but the average incision rate of both valleys combined is comparable to coastal rock uplift rates derived from marine terraces. Collectively, our results suggest that fluvial systems are sensitive to climate and tectonic processes even when affected by geomorphic disturbances, like river capture and beheading. However, care must be taken when interpreting river terraces as direct records of climate and tectonic processes, particularly when working on a single river valley.more » « less