Speaker Presentation Summaries

Speaker Presentation Summaries:
Proceeding of the 3rd International Biennial
Workshop on Subduction Processes, University of Alaska, Fairbanks, June 10-14, 2002

BASAL SUBDUCTION EROSION AND THE FORMATION OF THE ALEUTIAN TERRACE AND UNDERLYING FOREARC BASIN.

David W. Scholl; Department of Geophysics, Stanford University, Stanford, CA, 94305, and U.S. Geological Survey, Menlo Park, CA, 94025 (dscholl@usgs.gov or dscholl@pangea.stanford.eduundefinedundefined)

Roland von Huene; Department of Geology, University of California, Davis, CA, USA
95616, and GEOMAR, Kiel, Germany (rhuene@mindspring.com)

Holly F. Ryan; U.S. Geological Survey, Menlo Park, CA 94025 (hryan@usgs.gov)

INTRODUCTION: It is generally supposed that the structural depression forming large forearc basins, like that underlying the Aleutian Terrace, is caused by the upward and outward growth of
an accretionary prism. However, an improved base of offshore data reveals that many forearc basins are underlain by continental or arc basement rock that has been thinned in place. Thinning is best explained by basal subduction erosion. We explore here the notion that the Aleutian forearc basin (AFB) records a late Cenozoic episode of enhanced crustal thinning caused by the underthrusting Pacific plate.

BACKGROUND INFORMATION: The Aleutian Terrace is a broad, 40-50-km wide mid-slope bench that runs along virtually the length (~2000 km) of the Aleutian forearc. This prominent geomorphic feature extends seaward from a depth of 3000-4000 m at the based of the upper landward trench slope to a depth of about 4500 m at the southern edge of the terrace. Bathymetric highs are common along the southern side of the terrace, beyond and below which the lower landward trench slope descends to the trench floor (~7000 m). In cross-slope profile, the surface of the terrace is either gently basinal in contour or, more typically, sloping irregularly downward toward the top of the lower slope.

Seismic reflection and gravity data document that the terrace is the surface expression of a thickly (2-3 km) sedimented forearc basin (Grow, 1973; Harbert et al., 1986; Scholl et al., 1987; Ryan and Scholl, 1989; Ryan and Scholl, 1993).
Structurally, the AFB is a broad depression or swale in the surface of the ridge's basement of igneous rock. The axis of the basin strikes parallel to the ridge. Major faults do not border the ridge side of the AFB, but faulting and folding are typical of its outer, forearc-high margin. Drilling and dredging
establish that the basin began to form rapidly about 5-6 Myr ago in the latest Miocene (Scholl et al., 1987). The basin's fill of mostly Pliocene and Quaternary beds is a richly diatomaceous sequence of turbidite, hemipelagic, and ash debris shed from the Aleutian Ridge. The basinal sequence overlies
older coarse-grained volcaniclastic beds (sandstone and siltstone) of Miocene and Oligocene age. This section, ~1 km thick, drapes the upper trench slope but does not thicken where it passes beneath the AFB. The clastic sequence is thus a pre-basinal forearc accumulation. Beneath the crust of the AFB, at a sub-sea level depth near 15 km, the underunning slab of Pacific lithosphere is
virtually horizontal.

BASAL SUBDUCTION EROSION: At erosional convergent margins, lower plate underthrusting thins forearc crust by detaching rock from the upper plate and transporting this material to the mantle. Evidence for basal subduction erosion of a forearc is (1) rapid (0.3-0.5 km/Myr) and
substantial (3-5 km) subsidence, (2) offshore truncation of cratonic rock, (3) retrograde (landward) migration of the arc-magmatic front, and (4) the coastal and offshore occurrence of arc magmatic rocks. Globally, at convergent margins bordered by no observable or small- and medium-width accretionary prism (~5-40 km), the long-term (~10's of Myr), the average rate of subduction erosion is at least ~40 km3/Myr/km of trench.

EVIDENCE FOR SUBDUCTION EROSION ALONG THE ALEUTIAN RIDGE: Presently, only circumstantial evidence exists that subduction erosion has thinned the forearc crust of the Aleutian Ridge. This docket of observations includes: (1) the landward (northward) migration of the
arc magmatic front by ~30 km since the early Oligocene and 20 km since the middle Miocene; (2) the regional occurrence of a deeply (~1 km) subside shelf edge bordering the southern side of the ridge's wave-planed summit platform, and (3) the recovery along the inner side of the terrace of coarse clastic sediment from the pre-basinal deposits of Oligocene and Miocene age (Scholl et
al., 1987). It is inferred that the older of these units has subsided 3-4 km similar to the shallow-water deposits recovered by drilling or dredging the deeply submerged (2-5 km) outer forearcs of northern Japan, Tonga, northern Chile?, Peru, Costa Rica, Guatemala, and Mexico.

THE HYPOTHESIS: It is recognized that immediately seaward of the AFB a sizable accretionary prism forms the lower landward trench slope. But we propose that the tectonic mechanism that formed the AFB basin is not only the addition of an accretionary prism but more dominantly the consequence of subcrustal erosion (Ryan and Scholl, 1995). If we are correct in supposing that the older pre-basin volcaniclastic deposits are shallow water deposits, then it seems likely their subsidence to a depth of at least 6 km began well before the late Cenozoic formation of the AFB. Pre-basin subsidence can be explained by a background or long-term rate of subduction erosion of 40 km3 / Myr / km of ridge. However, construction of the AFB calls for an enhanced rate of basal erosion during the past 5-6 Myr that most prominently thinned the mid-slope area.

We speculate that the underthrusting of a nearly horizontal slab covered by a thick sequence of subducted sediment (McCarthy and Scholl, 1985) caused the enhancement that built the structural depression of the AFB. The heightened transport of fluids beneath the base of the forearc is presumed to be involved. Increased sediment subduction can be linked to rapid trench-axis sedimentation initiated by uplift and glaciation of eastern Gulf of Alaska drainages. Slab flattening suggest the Aleutian Ridge has been driven seaward over a mantle-anchored Pacific slab. We note that the axis of the AFB is situated above the inner edge of the flat slab, where it bends sharply downward to plunge below the crestal mass of the ridge. Great subduction zone earthquakes occur just beneath the inner edge of the AFB.

REFERCENCES:

Grow, J., 1973, Crustal and upper mantle structure of the central Aleutian arc: Geol Soc. Amer. Bull. v.84, p. 2169-2192;

Harbert, W. P., D.W. Scholl, T.L. Vallier, A.J. Stevenson, and D.M. Mann, 1986, Major
evolutionary phases of a forearc basin of the Aleutian ter­race--relation to north Pacific tectonic events and the formation of the Aleutian subduction complex: Geology, v. 14, p. 757-761;

McCarthy, J., and D.W., Scholl 1985,
Mechanisms of subuction accretion along the central Aleutian Trench: Geol. Soc. Amer. Bull. v., 96, p. 691-701;

Ryan, H.F., and D.W Scholl,. 1989, The evolution of forearc structures along an oblique convergent
margin, central Aleutian Arc: Tectonics, v. 8, p. 497-516;

Ryan, H.F., and D.W. Scholl, 1993, Geologic implications of great interplate earthquakes along
the Aleutian Arc: Jour. Geophys. Res., v. 98, p. 22,135-22,146;

Ryan, H. F., and D.W Scholl,1995, Deep reflectors beneath the central Aleutian forearc: implications
for the geometry of the subducting slab [abs]: Eos, AGU, v. 76;

Scholl, D. W., T.L.Vallier, and A.J.Stevenson,, 1987, Geologic evolution and petroleum
geology of the Aleutian Ridge, in Scholl, D. W., Grantz, A., and Vedder, J. G., (eds), Geology and resource potential of the continental margin of western North America and adjacent ocean
basins--Beaufort Sea to Baja California: Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, v. 6, p. 124-155, Houston, Texas.

The hydrogeological systems of incoming oceanic plates and overriding convergent margins of subduction zones: Insights from studies of the Middle America Trench.

César R. Ranero, ICREA at Instituto de Ciencias del Mar, CSIC, Pg. Marítim de la Barceloneta 37-49, 08003 Barcelona, Spain. e-mail: cranero@icm.csic.es

At the WHOI geodynamics seminar I will present recent observations and conceptual models of two important hydrogeological systems of subduction zones: The system of the incoming oceanic plate and the forearc system. I will also present some ideas on the implications of the models for shallow and deep tectonic processes and recycling of water
in the mantle.

The physical and chemical structure of subducting oceanic slabs is presently poorly known. After the realization of the meaning of oceanic trenches in the late 60’s, most research in the 70’s and early 80’s was dedicated to explain their large-scale geometry. Subsequent research on subduction zones has been greatly focused on overriding plates, and the oceanic plates have been comparatively overlooked. However, recent highresolution data indicate that oceanic lithosphere physical and chemical structure changes dramatically just prior to subduction. Those changes which -with the limited available data- we are just starting to appreciate, possibly have major implications for a number of first order processes, like the storage and transport of water into the subarc mantle (and
perhaps deeper) and the generation of intermediate depth (~70-350 km depth) seismicity. When oceanic plates bend to form the trench (prior to subduction) faulting -related to bending- cuts into the lithosphere across the igneous crust and into the upper mantle. The data indicate that faults possibly provide open paths for water percolation deep into the plates, where fluids may alter the oceanic crust and transform mantle peridotites into serpentine. The hydration of oceanic lithosphere at trenches may provide most of the fluids transported in slabs and released under arcs, and perhaps some fluids that are transported deeper into the mantle. Plate hydration may also play an important role on the occurrence of intra-slab intermediate-depth earthquakes.

The distribution and flow of fluid is commonly related to tectonics at all settings. At convergent margins this relationship has been widely studied at accretionary prisms, but at convergent margins where tectonic erosion affects overriding plates fluid distribution and tectonics are far less understood.

An integrated study of geophysical, geochemical and geological observations along the erosional type Middle America Trench indicates a hydrogeological system distinctly different from those described at accretionary prisms. The observations show how the hydrogeological system influences long-term tectonic erosion and the transition with depth from aseismic to seismogenic behavior along the plate boundary, where large earthquakes nucleate. In this hydrogeological system most fluid appears to come from pore water or chemically-bond water in subducting sediment. Where fluid is more abundant along the plate boundary, the overriding plate is actively being thinned, and fractures and subsides to form the continental slope. Most fluid originally contained at the plate boundary migrates by focused flow across a fractured overriding plate, contrasting with conceptual models of
accretionary margins where the decollement has been inferred to be the main fluid flow conduit. Seismogenic behavior at the plate boundary begins where fluid appears to be less abundant indicating a first order control on subduction zone thrust earthquakes.

Proceeding of the 3rd International Biennial
Workshop on Subduction Processes,
University of Alaska, Fairbanks, June 10-14, 2002

BASAL SUBDUCTION EROSION AND THE FORMATION OF THE ALEUTIAN TERRACE AND UNDERLYING FOREARC BASIN.

David W. Scholl; Department of Geophysics, Stanford University, Stanford, CA, 94305, and U.S. Geological Survey, Menlo Park, CA, 94025 (dscholl@usgs.gov or
dscholl@pangea.stanford.eduundefinedundefined)

Roland von Huene; Department of Geology, University of California, Davis, CA, USA 95616, and GEOMAR, Kiel, Germany (rhuene@mindspring.com)

Holly F. Ryan; U.S. Geological Survey, Menlo Park, CA 94025 (hryan@usgs.gov)

INTRODUCTION: It is generally supposed that the structural depression forming large forearc basins, like that underlying the Aleutian Terrace, is caused by the upward and outward growth of
an accretionary prism. However, an improved base of offshore data reveals that many forearc basins are underlain by continental or arc basement rock that has been thinned in place. Thinning is best explained by basal subduction erosion.
We explore here the notion that the Aleutian forearc basin (AFB) records a late Cenozoic episode of
enhanced crustal thinning caused by the underthrusting Pacific plate.

BACKGROUND INFORMATION: The Aleutian Terrace is a broad, 40-50-km wide mid-slope bench that runs along virtually the length (~2000 km) of the Aleutian forearc. This prominent geomorphic feature extends seaward from a depth of 3000-4000 m at the based of the upper landward trench slope to a depth of about 4500 m at the southern edge of the terrace. Bathymetric highs are common along the southern side of the terrace, beyond and below which the lower landward trench slope descends to the trench floor (~7000 m). In cross-slope profile, the surface of the terrace is either gently basinal in contour or, more typically, sloping irregularly downward toward the top of the lower slope.

Seismic
reflection and gravity data document that the terrace is the surface expression of a thickly (2-3 km) sedimented forearc basin (Grow, 1973; Harbert et al., 1986; Scholl et al., 1987; Ryan and Scholl, 1989; Ryan and Scholl, 1993). Structurally, the AFB is a broad depression or swale in the surface of the ridge's basement of igneous rock. The axis of the basin strikes parallel to the ridge. Major faults do not border the ridge side of the AFB, but faulting and folding are typical of its outer, forearc-high margin. Drilling and dredging establish that the basin began to form rapidly about 5-6 Myr ago in the latest Miocene (Scholl et al., 1987). The basin's fill of mostly Pliocene and Quaternary beds is a richly diatomaceous sequence of turbidite, hemipelagic, and ash debris shed from the Aleutian Ridge. The basinal sequence overlies older coarse-grained volcaniclastic beds (sandstone and siltstone) of Miocene and Oligocene age. This section, ~1 km thick, drapes the upper trench slope but
does not thicken where it passes beneath the AFB. The clastic sequence is thus a pre-basinal forearc accumulation. Beneath the crust of the AFB, at a sub-sea level depth near 15 km, the underunning slab of Pacific lithosphere is virtually horizontal.

BASAL SUBDUCTION EROSION: At erosional convergent margins, lower plate underthrusting thins forearc crust by detaching rock from the upper plate and transporting this material to the mantle. Evidence for basal subduction erosion of a forearc is (1) rapid (0.3-0.5 km/Myr) and
substantial (3-5 km) subsidence, (2) offshore truncation of cratonic rock, (3) retrograde (landward) migration of the arc-magmatic front, and (4) the coastal and offshore occurrence of arc magmatic rocks. Globally, at convergent margins bordered by no observable or small- and medium-width accretionary prism (~5-40 km), the long-term (~10's of Myr), the average rate of subduction erosion is at least ~40 km3/Myr/km of trench.

EVIDENCE FOR SUBDUCTION EROSION ALONG THE ALEUTIAN RIDGE: Presently, only circumstantial evidence exists that subduction erosion has thinned the forearc crust of the Aleutian Ridge. This docket of observations includes: (1) the landward (northward) migration of the
arc magmatic front by ~30 km since the early Oligocene and 20 km since the middle Miocene; (2) the regional occurrence of a deeply (~1 km) subside shelf edge bordering the southern side of the ridge's wave-planed summit platform, and (3) the recovery along the inner side of the terrace of coarse clastic sediment from the pre-basinal deposits of Oligocene and Miocene age (Scholl et
al., 1987). It is inferred that the older of these units has subsided 3-4 km similar to the shallow-water deposits recovered by drilling or dredging the deeply submerged (2-5 km) outer forearcs of northern Japan, Tonga, northern Chile?, Peru, Costa Rica, Guatemala, and Mexico.

THE HYPOTHESIS: It is recognized that immediately seaward of the AFB a sizable accretionary prism forms the lower landward trench slope. But we propose that the tectonic mechanism that formed the AFB basin is not only the addition of an accretionary prism but more dominantly the consequence of subcrustal erosion (Ryan and Scholl, 1995). If we are correct in supposing that the older pre-basin volcaniclastic deposits are shallow water deposits, then it seems likely their subsidence to a depth of at least 6 km began well before the late Cenozoic formation of the AFB. Pre-basin subsidence can be explained by a background or long-term rate of subduction erosion of 40 km3 / Myr /km of ridge. However, construction of the AFB calls for an enhanced rate of basal erosion during the past 5-6 Myr that most prominently thinned the mid-slope area.

We speculate that the underthrusting of a nearly horizontal slab covered by a thick sequence of subducted sediment (McCarthy and Scholl, 1985) caused the enhancement that built the structural depression of the AFB. The heightened transport of fluids beneath the base of the forearc is presumed to be involved. Increased sediment subduction can be linked to rapid trench-axis sedimentation initiated by uplift and glaciation of eastern Gulf of Alaska drainages. Slab flattening suggest the Aleutian Ridge has been driven seaward over a mantle-anchored Pacific slab. We note that the axis of the AFB is situated above the inner edge of the flat slab, where it bends sharply downward
to plunge below the crestal mass of the ridge. Great subduction zone earthquakes occur just beneath the inner edge of the AFB.

REFERCENCES: Grow,
J., 1973, Crustal and upper mantle structure of the central Aleutian arc: Geol Soc. Amer. Bull. v.84, p. 2169-2192;

Harbert, W. P., D.W. Scholl, T.L. Vallier, A.J. Stevenson, and D.M. Mann, 1986, Major evolutionary phases of a forearc basin of the Aleutian ter­race--relation to north Pacific tectonic events and the formation of the Aleutian subduction complex: Geology, v. 14, p. 757-761;

McCarthy, J., and D.W., Scholl 1985, Mechanisms of subuction accretion along the central Aleutian Trench: Geol. Soc. Amer. Bull.

Ikuko Wada, School of Earth and Ocean Sciences, University of Victoria, Canada Pacific Geoscience Centre, Geological Survey of Canada At subduction zones, viscous coupling between the subducting slab and the overriding mantle drives “corner flow” beneath the forearc-backarc region. The flow brings in hot mantle material from greater depths to keep the mantle wedge warm. However, low forearc surface heat flows indicate a cold and thus stagnant forearc mantle. It is
hypothesized that elevated pore fluid pressure and the presence of weak hydrous minerals along the plate interface cause slab-mantle wedge decoupling, resulting in the stagnant and cold forearc mantle. We model this system using a two-dimensional steady state finite element model. The mantle wedge is assumed to have a dislocation-creep rheology, and the effects of interface
weakening is approximated by imposing a thin low-viscosity layer along the plate interface. Decoupling occurs when the interface layer is weaker than the mantle wedge, causing the wedge flow above the decoupled interface to stop. The maximum depth of decoupling controls the thermal conditions in the mantle wedge and the slab beneath it and therefore is the key to most primary thermal and petrological processes in subduction zones. We apply the model to a number of subduction zones to investigate their thermal structure and its implications to subduction zone
processes. In the models, the maximum decoupling depth is constrained by surface heat flow data and the presumed subarc mantle temperature of > 1250°C. We find that the optimal maximum decoupling depth for most subduction zones is in the range of 70 to 80 km. For all subduction zones, the stagnant part of the forearc mantle wedge above the maximum decoupling depth is sufficiently cold to allow serpentine to be stable, but the actual degree of serpentinization depends on the availability of fluids. For subduction zones with a young and warm slab such as Cascadia and Nankai, dehydration of the subducting crust peaks beneath the stagnant part of the mantle wedge, providing ample fluid for serpentinization. For subduction zones with an old and cold slab such as NE Japan and Hikurangi, crustal dehydration peaks at greater depths, and significantly less metamorphic fluid
is released beneath the stagnant part of the mantle wedge and lower degree of serpentinization is expected except at ocean-ocean subduction zones such as Mariana and Kermadec. Beneath the arc, cold slabs provide large fluid flux into the hot and flowing part of the mantle wedge to promote melt production and arc volcanism. In contrast, young slabs are significantly drier at subarc depths, and a lower level of arc volcanism is expected.