Effects of triple junction migration and glacioeustatic cyclicity on evolution of upper slope morphologies, offshore Eel River Basin, northern California

Effects of triple junction migration and glacioeustatic cyclicity on evolution of upper slope morphologies, offshore Eel River Basin, northern California

Available online at www.sciencedirect.com R Marine Geology 199 (2003) 307^336 www.elsevier.com/locate/margeo E¡ects of triple junction migration and...

4MB Sizes 1 Downloads 12 Views

Available online at www.sciencedirect.com R

Marine Geology 199 (2003) 307^336 www.elsevier.com/locate/margeo

E¡ects of triple junction migration and glacioeustatic cyclicity on evolution of upper slope morphologies, o¡shore Eel River Basin, northern California Robert L. Burger a;b; , Craig S. Fulthorpe b , James A. Austin Jr. b b

a Department of Geological Sciences C1100, The University of Texas at Austin, Austin, TX 78712-1101, USA University of Texas Institute for Geophysics, 4412 Spicewood Springs Road, Building 600, Austin, TX 78759-8500, USA

Received 24 July 2002; received in revised form 22 May 2003; accepted 16 June 2003

Abstract The upper continental slope of the Eel River Basin is affected both by uplift and seismicity associated with the northward-migrating Mendocino Triple Junction (MTJ) and by glacioeustatic fluctuations. As a result, seismically imaged slope sequences vary dramatically along strike. The southern slope is dominated by the Humboldt Slide, a stack of nine deformed sequences. Older, undeformed sequences thicken landward, while slide sequences maintain a constant travel-time thickness. Onset of deformation was abrupt. Within the slide, reflector amplitudes alternate: high-amplitude reflectors are wavy to shingled, and suggest increasing deformation with depth. Low-amplitude reflectors are sub-parallel to wavy throughout. High- and low-amplitude sequence ‘couplets’ suggest repetitive deposition of contrasting lithologies, a response we ascribe to late Pleistocene glacioeustatic cyclicity. If each ‘couplet’ represents a V100 kyr late Pleistocene sea-level cycle, then Humboldt Slide deformation began V450 ka. This is approximately coeval with onset of deformation on the adjacent shelf V500 ka, previously attributed to uplift and seismicity associated with northward encroachment of the MTJ. We believe migration of the MTJ triggered the slide, and is the cause for continuing deformation today, by sediment creep along internal glide planes. North of the Humboldt Slide, we interpret an upper slope anticline as a seaward extension of the Little Salmon Fault Zone (LSFZ) previously identified landward. Both faulting and folding extend to the seafloor, indicating that deformation continues. An adjacent buried anticline suggests that deformation has also jumped northward within the LSFZ on some parts of the upper slope. North of the LSFZ, the slope is dominated by downslope-trending channels and gullies. These features are v-shaped, vertically stacked, and increase in depth downslope. They are mostly infilled; their physiographic expression is generally subdued by sediment draping. Channel fills generally consist of basal highamplitude reflectors, overlain by reflectors of lower amplitudes. Several large channels on older surfaces are concentrated near the northern end of the seismic coverage; smaller channels are more evenly distributed along the margin on younger surfaces. Channels also extend increasingly landward on progressively younger surfaces, indicating migration in that direction with time. These incisions are clearly inactive during highstands, as at Present; slope deposition today is dominated by draping of hemipelagic sediments. Channels form and migrate landward through headward erosion during base-level changes, when both shoreface and fluvial sediment sources are more proximal to the upper slope. The increasing lateral distribution of channels through time supports an increase in shoreface erosion

* Corresponding author. Present address: Joint Oceanographic Institutions, Inc., 1755 Massachusetts Avenue NW, Suite 700, Washington, DC 20036-2102, USA. Fax: +1-202-462-8754. E-mail address: [email protected] (R.L. Burger).

0025-3227 / 03 / $ ^ see front matter A 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0025-3227(03)00194-4

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

308

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

north of the LSFZ. A possible explanation is a known decrease in fluvial sediment input from the north, combined with MTJ-related uplift in the south. Along-strike contrasts in upper-slope sequence morphology in the southern Eel River Basin are clearly a function of both regional tectonism and proximity to sediment sources, influenced by baselevel changes. Humboldt Slide sequences, while deformed, are continuously influenced by systematic lithological variations occurring as a result of glacioeustatic cycles. Effects of base-level changes similarly control the distribution and fill of incisions on the northern slope. All of these observations reinforce previous seismic investigations of the offshore Eel River Basin shelf, and confirm that the competing effects of tectonism and glacioeustacy on the preservation of continental margin stratigraphy can be differentiated. A 2003 Elsevier B.V. All rights reserved. Keywords: Eel River Basin; continental slope; multichannel seismic; sequence stratigraphy; sediment transport; tectonics

1. Introduction Upper continental slopes are generally characterized by gradients of gravitational potential, resulting in slumps, slides, and frequent debris £ows. Erosional features, such as chutes, gullies, channels and canyons are formed commonly in slope environments (Galloway, 1998). Unfortunately, these downslope processes occur at a wide variety of scales, producing complex subsurface features that, when buried, may be di⁄cult to interpret seismically (Barnes and Lewis, 1991; Baraza et al., 1999). Nonetheless, the tectonically active Eel River Basin is an excellent setting for evaluating the accumulation, preservation, and post-depositional deformation of sedimentary strata on an upper continental slope. The combination of high rates of sediment supply from the Eel River (Brown and Ritter, 1971), late Pleistocene glacioeustatic sea-level £uctuations of su⁄cient magnitude (e.g. Ruddiman et al., 1989) to expose the adjacent shelf, and active deformation o¡shore (Clarke, 1992; Gulick et al., 2002; Burger et al., 2001, 2002; Gulick and Meltzer, 2002), including frequent earthquakes concentrated near the southern end of the basin (Couch et al., 1974; Field and Barber, 1993), all have resulted in an along-strike variation in upper-slope morphologies (Fig. 1). The southern Eel River Basin shelf and upper slope have already been extensively studied to evaluate sediment transport processes (Walsh and Nittrouer, 1999; Ogston and Sternberg, 1999; Alexander and Simoneau, 1999), sediment preservation rates at decadal scales (Drake, 1999; Leithold and Hope, 1999; Sommer¢eld and Nit-

trouer, 1999), and process-based in£uences on upper-slope mass wasting (Field and Edwards, 1981; Field and Barber, 1993; Lee et al., 1981, 1999, 2002). Ultra-high resolution (V3.5 kHz), Huntec deep-towed single-channel seismic pro¢les have been used to investigate previously recognized slope morphologic features, such as the Humboldt Slide in the south (Gardner et al., 1999; Lee et al., 2002) and downslope-trending gullies in the north (Field et al., 1999; Spinelli and Field, 2001). However, longer-term evolution and preservation of the upper slope in the Eel River Basin, speci¢cally over time scales of 105 ^ 106 yr, have not yet been investigated in detail. In this study, multichannel seismic (MCS) data collected in the southern o¡shore Eel River Basin in 1996, as part of the O⁄ce of Naval Research (ONR) STRATAFORM initiative (Nittrouer and Kravitz, 1996; Fig. 1), have been used to image slope sequences and evaluate their formation and preservation over V105 -yr time scales. We infer major controls on sediment preservation, including post-depositional deformation, from the midPleistocene to the Present by interpreting sequence geometries and associated seismic facies in the context of known downslope processes. We then integrate these observations with observed modern slope conditions, and develop a predictive sequence-stratigraphic model for this upper slope, extending a previously proposed shelf model for the Eel River Basin (Burger et al., 2002).

2. Geologic setting The Eel River Basin is part of a late Cenozoic

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

309

Fig. 1. The Eel River Basin, illustrating the 1996 STRATAFORM MCS geophysical survey tracks (light lines) with respect to onshore topography and bathymetry. Numbered MCS pro¢les (bold line segments) and map areas (solid outlines) correspond to locations of Figs. 3^10 and 13. Bathymetry (Go¡ et al., 1999) in meters: contour interval 10 m from 0 to 100 m, 50 m from 100 to 200 m, and 100 m for all depths s 200 m. Bold contour represents the 150 m isobath, the approximate location of the shelfbreak. Dashed contours indicate the approximate extent of the upper slope marginal plateau. Onshore topography is from Mayer et al. (1999); inset is adapted from Clarke (1987). MTJ = Mendocino Triple Junction; MFZ = Mendocino Fracture Zone; SAF = San Andreas Fault; CSZ = Cascadia subduction zone LSFZ = Little Salmon Fault Zone; CM = Cape Mendocino; CS = Cape Sebastian.

forearc located along the northwestern U.S. continental margin ; it extends V210 km from near Cape Mendocino, California to Cape Sebastian, Oregon (Bachman and Crouch, 1987; Clarke, 1992 ; Fig. 1). Approximately 90% of the basin lies o¡shore, bordered by the Cascadia subduc-

tion zone to the west and the Mendocino Fracture Zone to the south. As much as 2500 m of Neogene^Holocene marine sediments have accumulated in the basin, unconformably overlying a Middle Jurassic^early Tertiary deformed accretionary prism known as the Franciscan Complex

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

310

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

(Ogle, 1953; Blake and Jones, 1974; Clarke, 1992; Gulick et al., 2002). Basin-¢ll exposed onshore indicates an overall coarsening-upward sequence during the Miocene, from V14 to 4 Ma, from deep-water lithofacies to paralic deposits (McCrory, 2000), followed by rapid uplift with resultant shoaling starting V3.5 Ma (McCrory, 1989). Onshore sediments suggest a quiescent tectonic environment until the late Pliocene; folding of basin strata commenced V2.0^1.5 Ma (McCrory,

1989), followed by thrust faulting V1.0 Ma (Carver, 1987; McCrory, 1996). Eel River Basin sediments are today in£uenced both by E^NE compression related to Gorda^ North America plate convergence and N^S compression and marginal uplift associated with northward migration of the Mendocino Triple Junction (MTJ; Nilsen and Clarke, 1987; Clarke, 1992; Gulick and Meltzer, 2002; Fig. 1). The MTJ is a di¡use region of uplift and faulting

Fig. 2. Eel River Basin MCS grid with respect to regional seismic activity. Earthquake epicenters (historical observations and measurements) are from 1860 to 1980, compiled mainly by Couch (1980). Note the E^W oriented concentration near 40‡30PN, associated with locations of the MTJ and Mendocino Fracture Zone (see also Fig. 1, inset). Bathymetry in meters. Figure adapted from Field and Barber (1993).

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

formed by the intersection of the Paci¢c, North America and Gorda plates (Fig. 1). Strike-slip deformation associated with the San Andreas Fault system occurs south of the MTJ, and subductionrelated folding and faulting associated with Gorda^North America plate convergence occurs to the north (Fig. 1, inset). The MTJ was formed near Baja, California V29 Ma (McKenzie and Morgan, 1969; Atwater, 1970) and has been migrating northward since V24 Ma (Atwater, 1970). N^S compression associated with its migration has warped the southern Eel River Basin; folds and faults trend NNW^SSE in the northern basin, but become NW^SE and almost E^W nearer the MTJ (Woodward-Clyde Consultants, 1980; Carver, 1987; Clarke, 1992; Orange, 1999; Gulick and Meltzer, 2002). Folds and faults onshore and in the southern o¡shore basin probably re£ect compression and uplift associated with this northward migration (Carver, 1987; McLaughlin et al., 1994; Furlong and Govers, 1999; Gulick et al., 2002). Although the entire Eel River Basin is subject to earthquakes associated with the adjacent Cascadia subduction zone (Fig. 1, inset), seismicity is concentrated at the basin’s southern end, nearest the MTJ (Couch et al., 1974; Field and Barber, 1993; Fig. 2). The Eel River provides most modern sediment to the o¡shore Eel River Basin, with a mean annual suspended sediment load s 107 t/yr (Brown and Ritter, 1971). However, annual sediment discharge is highly episodic, occurring primarily during winter storm events that vary in severity annually (Brown and Ritter, 1971; Leithold and Hope, 1999). The Mad River also contributes sediment to this margin, but its mean annual discharge is an order of magnitude lower than that of the Eel River (Nittrouer, 1999). Modern £ood deposits indicate a prominent N^NW component of sediment transport during the current sea-level highstand, as a result of winds and currents prevailing during periods of maximum river discharge associated with winter storms (Borgeld, 1985; Bray and Greengrove, 1993; Wheatcroft et al., 1996). However, sediment transport to the south is also observed (Ogston and Sternberg, 1999), suggesting that highstand sediment transport is complex in this energetic setting. In con-

311

trast, sediment transport from the Eel River during lowstands is presumed to have been primarily W^SW, discharging directly into or near the head of Eel Canyon (Burger et al., 2001 ; Fig. 1). The southern o¡shore Eel River Basin is characterized by a smooth, nearly featureless shelf (Go¡ et al., 1999; Fig. 1), suggesting that high modern sedimentation rates ¢ll topography caused by local tectonic deformation. The shelf dips V0.2^0.3‡ seaward, gradually steepening towards the modern shelf-break. Although best described as a transitional zone, the shelf-break has been generally de¢ned as the 150-m isobath (Nittrouer, 1999 ; Fig. 1). The slope steepens to V3^ 5‡ at V150^450 m, before leveling out to V1‡ at V450^800 m. This V10^20 km wide marginal plateau is most pronounced in the southern basin (Fig. 1). A lower slope, dipping at V5‡ but locally up to V15^20‡ (Gulick, 1999), extends from V800 m to the foot of the slope at V2700 m (Alexander and Simoneau, 1999). The upper slope of the southern basin is divided by a NW^SE trending anticline that breaches the sea£oor below a water depth of V350 m (Fig. 1). This feature is correlative along strike with a shelfal zone of folding and faulting, interpreted by Burger et al. (2002) as the seaward extension of the onshore Little Salmon Fault Zone (LSFZ; Ogle, 1953; Carver, 1987; Clarke and Carver, 1992; Clarke, 1992; McCrory, 1995). Physiography changes on either side of the breached anticline: to the south, the upper slope and adjacent marginal plateau are dominated by an arcuate-shaped, mass-wasting feature previously identi¢ed as the Humboldt Slide (Fig. 1; Field and Edwards, 1980, 1981; Lee et al., 1981; Brooks et al., 1991, Field and Barber, 1993; Gardner et al., 1999). In contrast, the upper slope to the north has a more constant gradient and is less a¡ected by such large-scale mass-wasting processes. The upper slope there is dominated by a series of downslope-trending incisions below water depths of V250^300 m; these features range in size from small gullies V1^3 m in depth (Field et al., 1999; Spinelli and Field, 2001) to larger channels up to V50 m deep. In this paper, we investigate the upper slope of the southern basin seismically, from the shelf-break to the mar-

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

312

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

ginal plateau. Our study encompasses the Humboldt Slide, the breached anticline, and a portion of the channeled/gullied slope to the north (Fig. 1), and builds upon a seismic analysis of the adjacent continental shelf using these MCS pro¢les (Burger et al., 2001; Burger et al., 2002).

3. Methodology Approximately 2200 km of high-resolution MCS pro¢les were collected in the southern o¡-

shore Eel River Basin in 1996 by the University of Texas Institute for Geophysics (UTIG) and Lamont-Doherty Earth Observatory (LDEO) (Fulthorpe et al., 1996). The survey used a 48channel (12.5-m group interval) hydrophone streamer (a back-up 48-channel, 15-m group interval streamer was used for some pro¢les) and a 45/45 in3 G.I. airgun ¢ring at V12.5 m intervals. Data were sampled every 0.5 ms and displayed after bandpass ¢ltering at 10^475 Hz. Predictive deconvolution was applied to remove sea£oor multiples, and the data were f-k migrated.

Fig. 3. Dip seismic pro¢le 41, showing sequences mapped through the axis of the Humboldt Slide (‘HS’; for location of pro¢le, see Fig. 1). Numbered solid lines represent sequence-bounding ‘HS’ surfaces; bold, steeply dipping lines represent faults. True dips indicated at top left. Inset ¢gures from the middle of the ‘lower slide’ (right) and from near the toe of the slide (left) indicate sequences characterized by high, low and mixed amplitudes, as well as representative internal sequence geometries (see also Table 1). Numbered lines at top indicate the locations of crossing pro¢les.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Forty-one (V50%) of the seismic pro¢les were processed at UTIG and the rest at LDEO. This investigation is based on both sets of pro¢les; spacing is V800 m, with the exceptions of larger 1.5^5 km pro¢le spacings near gas-charged areas near the shelf-break and at the western end of the grid (Yun et al., 1999; Fig. 1). We estimate the vertical seismic resolution to be V5^10 m and the horizontal resolution to be V15^50 m (Burger, 2002). Depth conversions assume a mean sonic velocity of 1.8 km/s, a reasonable estimate for the ¢ne-grained, unconsolidated terrigenous ¢ll of this basin. Interpretations were made using GeoQuest0 seismic interpretation software. Seis-

313

mic surfaces were mapped by identifying re£ection truncations, including lapping relationships, and/or distinct changes in amplitudes or seismic facies (Burger, 2002).

4. Results 4.1. Humboldt Slide The upper shelf and marginal plateau within the Humboldt Slide area are bounded by the interpreted LSFZ anticline to the north and the shelf-break to the east (Figs. 1, 3 and 4). Vertical

Fig. 4. Seismic pro¢le 56, showing along-strike variations in upper slope sequence morphology (for location of pro¢le, see Fig. 1). Numbered solid lines represent mapped sequence-bounding surfaces (see also Table 1); bold, steeply dipping lines represent faults. None of the surfaces could be correlated across the LSFZ, so Humboldt Slide (‘HS’) surfaces are not presumed correlative with similarly numbered north slope (‘NS’) surfaces. Insets highlight portions of sequences characterized by high, low and mixed amplitudes within the Humboldt Slide, strike-oriented internal geometries, as well as stacked channels imaged near the southwest end of the MCS grid (left inset). Numbered lines at top indicate the locations of crossing pro¢les. LSFZ = Little Salmon Fault Zone.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

314

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

to steeply seaward-dipping normal faults occur just seaward of the shelf-break (Fig. 3). Dip-slip o¡sets along these faults range up to V18 m. Re£ectors between faults are in some places chaotic, but in most areas are coherent within individual fault blocks (Fig. 3). These re£ectors commonly display a convex-upward morphology, suggesting that they have undergone compression as well as dip-slip deformation (Fig. 3). Faulting and associated sediment deformation occur along the steepest portion of the upper slope, V3^6‡ in the line of section. We term this area the ‘upper slide’, as it appears distinct from deformation farther seaward (Fig. 3). Seaward of the faulted zone, the slope inclination decreases to V1‡ towards the marginal plateau. Wavy, shingled and chaotic re£ectors are prevalent, being di¡erent in appearance from the ‘upper slide’ region (Fig. 3). Faulting is minor to absent, also in contrast to the ‘upper slide’. We term this deeper-water deformed area the ‘lower slide’ (Fig. 3). The upper and lower slides are separated occasionally by a narrow (V1^3 km) zone of undeformed re£ectors, generally coincident with an in£ection point of the modern sea£oor (Fig. 3). Just seaward of the shelf-break, at the head of the slide, gullies and small downslopetrending channels occur (Fig. 1 ; Gardner et al., 1999). These channels are poorly imaged, presumably as a result of the presence of interstitial gas along the shelf-break and uppermost slope (Yun

et al., 1999), and they disappear downslope. They are not imaged at all in the lower slide. Twelve surfaces, labeled HS1^HS12, have been mapped seismically in this area (Fig. 3 and Table 1). All can be mapped within the lower slide, but they cannot be correlated beyond the upper slide to the shelf-break. These surfaces locally truncate underlying re£ectors within the lower slide (Fig. 5), but generally appear conformable further seaward (Fig. 3). Sequences bounded by these mapped surfaces fall into two categories, based on observed seismic facies (see Table 2): Type 1 represents landward-thickening, generally undeformed strata, composed of parallel to landward-divergent re£ectors, while Type 2 displays generally constant thickness, with varying degrees of internal deformation (Fig. 3), including subparallel, wavy, shingled, and chaotic re£ectors. Type 1 sequences underlie Type 2 sequences. All Type 1 sequences underlie surface HS4, while all Type 2 sequences overlie that surface. The transition between the two sequence types is sudden. HS4 is the prominent marker of the change in seismic facies (Fig. 3). A cyclic alternation of re£ector amplitudes is apparent throughout the Humboldt Slide sequences. Most of the mapped sequences are characterized by either relatively high- or low-amplitude re£ectors, alternating throughout the section. This pattern is best imaged in the southern portion of the lower slide (Fig. 3, inset). Sequences

Table 2 Seismic facies of Humboldt Slide sequences Sequence (de¢ned by bounding surfaces)

Sequence type

Amplitude

Seismic facies (in lower slide)

HS12^Sea£oor HS11^HS12 HS10^HS11 HS9^HS10 HS8^HS9 HS7^HS8 HS6^HS7 HS5cHS6 HS4^HS5 HS3^HS4 HS2^HS3 HS1^HS2 Below HS1

Type Type Type Type Type Type Type Type Type Type Type Type Type

High Low High Low High Moderate/mixed Moderate/mixed Low High Low High Low High

wavy wavy wavy subparallel^wavy shingled wavy^shingled shingled^chaotic subparallel^wavy shingled subparallel^parallel subparallel^parallel subparallel^parallel subparallel

2 2 2 2 2 2 2 2 2 1 1 1 1

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

Table 1 Identifying seismic characteristics of mapped slope surfaces Humboldt Slide surfaces Truncation of underlying re£ectors

Amplitude change across surface Facies change across surface

Other distinguishing characteristics

HS1 HS2 HS3 HS4

moderate^high high high moderate

prominent, angular low-angle, locally low-angle not apparent

high to low low to high high to low low to high

none none none subparallel/parallel to shingled

HS5

high

local

high to low

shingled to subparallel/wavy

HS6

not distinctive

not apparent

low to moderate in deformed areas

subparallel/wavy to shingled/ chaotic

HS7

moderate^high

local

minor

HS8

high

wavy/shingled to shingled

HS9 HS10

moderate^high high

prominent; angular in deformed areas prominent not apparent

high to low, only apparent in limited areas low/moderate to high

continuous continuous; small channel incision prominent channel incision downlap surface; irregular/faulted in deformed areas continuous; mostly undeformed; conformable with overlying re£ectors identi¢ed primarily by changes in amplitude and seismic facies across surface very irregular in deformed areas

high to low low to high

shingled to subparallel/wavy subparallel/wavy to wavy

HS11

low^moderate

local

high to low

minor

HS12

low^moderate

not apparent

low to high

minor

Amplitude change across surface Other distinguishing characteristics

downlap surface; irregular/faulted(?) in places continuous wavy; conformable with overlying seismic facies identi¢ed primarily by changes in amplitude across surface identi¢ed primarily by changes in amplitude across surface

Channeled/gullied slope surfaces Surface Surface amplitude

Channel incisions

Truncation (non-incised areas)

NS1 NS2 NS3 NS4 NS5 NS6 NS7 NS8 NS9 NS10 NS11

prominent prominent prominent prominent apparent apparent apparent prominent very minor prominent prominent

not apparent not apparent local, near shelf-break low angle, local low angle, local minor, near shelf-break prominent, near shelf-break prominent, near shelf-break prominent, near shelf-break minor, near shelf-break low angle, local

a

high moderate^high moderate high very high moderate^high low^moderate moderate moderate moderate moderate^high

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

Surface Surface amplitude

Correlative with shelf surface 11a low^high high^low Mappable onto shelf

Correlative with shelf surface 13a

From Burger et al. (2002). 315

316

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Fig. 5. Variable-intensity image of a portion of seismic pro¢le 41, showing detailed features of seismic facies imaged within the Humboldt Slide (for location of pro¢le, see Figs. 1 and 3). White arrows indicate locations where mapped sequence boundaries truncate underlying re£ectors. Truncation is localized; all mapped surfaces do not display truncation on every pro¢le. Note the transition within the high-amplitude sequences from a dominant wavy seismic facies in shallower sequences to a dominant shingled seismic facies at greater depths (see also Table 2). Numbered lines at top indicate the locations of crossing pro¢les.

characterized by lower amplitudes generally display less internal deformation and are dominated by subparallel to wavy re£ectors. Higher-amplitude sequences are more discontinuous and display more internal deformation, as suggested by wavy to shingled re£ectors. Overall, the degree of deformation within the high-amplitude sequences increases with depth, as shallow sequences dominated by wavy re£ectors (HS10^HS11, HS12^Sea£oor; Figs. 3 and 5) are replaced downward by shingled re£ectors (HS4^HS5, HS8^HS9; Figs. 3 and 5).

The lateral extents of deformed re£ectors within all sequences above HS4 in the lower slide have been mapped on dip-oriented pro¢les (Fig. 6). These deformed areas vary in size between sequences, ranging from V64 km2 to V107 km2 . This variation is not gradational vertically; areas of deformation within adjoining sequences di¡er by as much as 43 km2 (e.g. compare sequences HS5^HS6 and HS6^HS7 ; Fig. 6). All sequences mapped within the Humboldt Slide area pinch out to the north, against the breached LSFZ anticline (Fig. 4). They also thin

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

317

Fig. 6. Lateral extents of deformed re£ectors within each mapped sequence in the lower Humboldt Slide (see also Figs. 3^5). Deformation was identi¢ed on all dip-oriented seismic lines, where deformation is most apparent. Sequence names, corresponding to bounding surfaces speci¢ed in Figs. 3 and 4, are indicated at the top left of each map. Dominant amplitudes that characterize each sequence are indicated at the bottom left of each map. Note that the high-amplitude sequences have slightly larger deformed areas than low-amplitude sequences, and that the two mixed-amplitude sequences have the largest deformed areas. Locations of maps with respect to the MCS grid are indicated in Fig. 1.

to the west, toward an apparent uplift or anticline that extends seaward of the MCS coverage (Fig. 3). Towards the shelf, sequences extend into the upper slide, but they cannot be correlated farther landward due to a combination of pervasive deformation, interference from a strong sea£oor multiple, and wipeout of re£ectors as a result of presumed disseminated gas concentrated along the shelf-break (Yun et al., 1999; Fig. 3). Humboldt Slide sequences are also truncated to the southwest by an unnamed, E^W trending submarine canyon (Fig. 7) that intersects the seismic coverage at the southern ends of strike pro¢les 62 and 64 (Fig. 7). This canyon is clearly expressed in the bathymetry near 40‡50PN, 124‡40PW (Fig. 1). Mapped channels may represent tributaries to an earlier manifestation of this modern canyon. Two mapped surfaces, HS2 and HS3, display prominent, v-shaped incisions near the southern end of the seismic grid (Fig. 4). Smaller channels overlie these features, de¢ned by HS4 and HS5 (Fig. 4). The HS3 channel ap-

pears to trend NE^SW. However, no buried incisions occur above HS5. 4.2. Little Salmon Fault Zone The LSFZ, a NW^SE trending zone of faultcored folding, de¢nes the northern boundary of the Humboldt Slide (Figs. 1 and 4). Landward, folding associated with Little Salmon Fault motion is more deeply buried; we interpret that folding ended beneath the shelf V700 ka (Burger et al., 2002). In contrast, an anticline breaches the sea£oor on the slope and recent sequences onlap against it, suggesting that folding there continues (Figs. 1 and 4). In places on the upper slope, particularly near Line 60, the LSFZ is composed of two adjacent anticlines (Fig. 8). The southern anticline is buried, indicating folding there is no longer active, whereas the northern anticline extends further seaward and breaches the sea£oor (Fig. 8), suggesting a northward jump of the LSFZ. Faulting extends to the sea£oor within

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

318

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Fig. 7. Uninterpreted and interpreted portion of strike pro¢le 64, showing an unnamed modern submarine canyon that truncates Humboldt Slide sequences to the southwest. Numbered solid lines represent mapped sequence-bounding surfaces; bold, steeply dipping lines represent faults. Numbered lines at top indicate locations of crossing pro¢les. See Fig. 1 for location of this pro¢le and bathymetric expression of the unnamed canyon.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

319

Fig. 8. Uninterpreted and interpreted portion of strike pro¢le 60, showing adjacent anticlines that characterize the LSFZ in places on the upper slope, especially near this pro¢le. Numbered solid lines represent mapped sequence-bounding surfaces; bold, steeply dipping lines represent faults. Note that younger sequences thin towards the northern, more sur¢cial anticline; bounding surfaces onlap against it, suggesting recent, and possibly continuing, uplift. In contrast, sequences maintain generally constant thicknesses across the southern, more deeply buried anticline, suggesting that folding there is less active. However, faulting in both anticlines extends to the sea£oor, indicating that both structures are still deforming. Numbered lines at top indicate the locations of crossing pro¢les. Location of pro¢le is shown in Fig. 1.

both anticlines, indicating that faulting remains active in both structures. 4.3. Gullied slope north of LSFZ North of the LSFZ, slope physiography di¡ers from that observed to the south. Faulting seaward of the shelf-break and evidence of large-scale slope failure are largely absent. Instead, only small areas are characterized by growth faulting and minor rotational failures (Fig. 9). Inclination of the uppermost slope is similar to that in the Humboldt Slide area, V3‡^4.5‡ (in the line of

section). The marginal plateau is less prominent just north of the LSFZ (Fig. 1); slope inclinations there do decrease seaward of the steep uppermost slope, as in the Humboldt Slide area, but only slightly, ranging from V1.2‡ to 3.5‡ (in the line of section). Even further to the north, the marginal plateau is largely absent (Figs. 1 and 9). The most distinctive feature on the upper slope north of the LSFZ is an extensive, stacked succession of downslope-trending channels (Figs. 4 and 10). Most of these channels are developed on seismic surfaces mappable both between channels and upslope to the shelf-break. Eleven surfaces have

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

320

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Fig. 9. Uninterpreted and interpreted dip-oriented seismic pro¢le 97, showing mapped upper slope sequences in the gullied slope north of the LSFZ (see Fig. 1 for location). Numbered solid lines represent mapped sequence boundaries; bold, steeply dipping lines represent faults. True dips indicated at top left. Note the continuous slope gradient and prevalence of evenly bedded, sea£oor-parallel re£ectors. Arrows indicate re£ector truncations, concentrated near the shelf-break. Areas of mapped surfaces that deviate from the sea£oor-parallel trend may represent oblique crossings of downslope-trending channels. Numbered lines at top indicate locations of crossing pro¢les.

been mapped, based mainly on observed incisions ; other identi¢ers include truncation of underlying re£ectors and/or changes in seismic facies across these surfaces (Fig. 9 and Table 1). Several also truncate underlying re£ectors just seaward of the shelf-break (Fig. 9). Between channels, surfaces are generally parallel to the sea£oor, forming part of a series of continuous, evenly spaced, concordant re£ectors (Figs. 9 and 10). Most channels are v-shaped and vertically

stacked (Fig. 10). Channels deepen downslope; depths range from the lower limits of vertical seismic resolution, V5 m, to a maximum of V110 m (at 1.8 km/s). Channel widths range from V50 m to V1100 m; most channels are V100^500 m wide. Channel widths generally increase downslope. Width/depth ratios range from V1.5 to V42; most are between 5 and 25, and generally decrease downslope. The number of channels on each surface initially increases downslope, as

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

321

Table 3 North slope (‘NS’) channel characteristics

Mean channel width (m)

Mean channel depth (m)

Mean width/depth ratio

Number of mapped channels

Line #

NS11

NS10

56 60 62 64 66 56 60 62 64 66 56 60 62 64 66 56 60 62 64 66

119 136.9 278.7 265.2 261.7 8.3 13.4 25.5 34.6 50.5 15 11.9 11.2 8.8 5 23 43 23 21 15

235.4 158.7 289 283.9 232.2 14.5 18 29.3 38.9 44.7 20.6 9.8 10.5 8.4 4.9 6 50 24 20 15

NS9

130.7 147.2

10.6 17.7

13.3 8.3

3 1

channels develop, then decreases further downslope as channels deepen and coalesce (Fig. 10). Channel characteristics for each surface are summarized in Table 3. Many of the channels are characterized by a distinctive ¢ll pattern. High-amplitude basal ¢lls are overlain by progressively lower-amplitude re£ectors that appear to drape and ultimately bury the channels (Fig. 11). Buried channels are more deeply incised than channels expressed at the modern sea£oor; draped sediments that appear to cap channel ¢lls preserve subdued surface expressions of most previously active channels (Figs. 10 and 11). In the same upper slope area a¡ected by channeling, gullies V1^3 m deep have been described on Huntec boomer single-channel seismic pro¢les to depths of V65 m below the sea£oor (Field et al., 1999; Spinelli and Field, 2001). These features are below the vertical resolution of the MCS data (Figs. 9 and 10), and are generally more closely spaced, V100^1000 m apart (Spinelli and Field, 2001), than the channels, which are generally 6 1^2 km apart (Fig. 10). We therefore de¢ne ‘channels’ as features observable with the V5 m

NS8

NS7

147.2 145 260.6 226.6 134.9 9.9 14.5 27.2 32.1 38.1 20.5 11.7 10.2 7.2 4.4 5 22 23 19 14

163.2 127.8 176.7 176.1 314 7.6 9.6 11.8 18.1 14.2 21.9 14 16.1 11.5 16.5 11 19 19 14 5

NS6

NS5

162.9 187.9 229.5 55.3

171.7 140.9 294.4 449.7

10.6 12.4 21 33.6

10.6 13.3 33 27.7

16.2 17 12.3 13.2

18.2 10.5 8 13.8

7 6 7 2

2 2 3 3

NS4

NS3

515 223.8 302.3 269.8 607.1 62 29.6 29.1 30.1 41.2 12.2 9.6 11.2 9.8 12.4 2 4 12 12 4

185 125.4 222.6 197 402.3 8.3 12 17.7 19.7 55.9 21.6 11.4 13.2 11.2 6.2 7 9 6 16 5

NS2

NS1

417 241.2

686.3 601 564.2

30.7 21.2

77.9 60.2 56.6

12.9 12.1

8.8 10.6 10

3 6

1 2 1

vertical resolution of the MCS data, and ‘gullies’ as the morphologically similar, though smaller, features below our seismic resolution. According to Spinelli and Field (2001), the gullies merge into larger channels downslope, suggesting that the gullies may simply represent tributaries and upslope extensions of the channels described in this study, and are therefore likely formed by the same processes. The lateral distribution of channels changes through time. Younger surfaces NS8, NS10, and NS11 display a large number of similarly sized channels, distributed fairly evenly across the upper slope (Figs. 4 and 10). In contrast, older surfaces NS1^NS7 display a smaller number of larger channels, concentrated at the northern end of the seismic coverage (Fig. 10). In contrast to surfaces mapped south of the LSFZ, surfaces NS5, NS8 and NS11 can be correlated landward across presumed gas-charged sediments at the shelf-break. Therefore, their ages can be estimated, using a previously proposed shelf chronostratigraphy based on ties to industry wells and correlations to dated unconformities onshore (Burger et al., 2002). We tentatively correlate NS11

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

322

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Fig. 10. Uninterpreted (A) and interpreted (B) portions of strike seismic pro¢les 60, 64 and 66, showing the downslope evolution of slope channel geometries (locations of pro¢les are shown in Fig. 1 and in inset). Numbered solid lines represent mapped sequence-bounding surfaces; bold, steeply dipping line segments represent faults. Numbered lines at top indicate locations of crossing pro¢les. Note the expression of larger channels on older surfaces basinward, as well as the increase in lateral distribution of channels on younger surfaces. Channel highlighted on Line 64 (panel B) is shown in greater detail in Fig. 11.

with shelf surface 14 (Burger et mated to have formed V43 ka. correlates with shelf surface 11 2002), with an estimated age of

al., 2002), estiNS5 tentatively (Burger et al., V300 ka. NS8

appears to overlie shelf surface 12 (Burger et al., 2002), and truncates it near the shelf-break. Therefore, NS8 must be only slightly younger than surface 12, estimated at V203 ka.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

323

Fig. 10 (Continued).

5. Discussion 5.1. Tectonism vs. glacioeustacy and the evolution of Eel River Basin upper slope morphology Previous investigations of seismic sequences on the Eel River Basin continental shelf have indi-

cated that even in this tectonically complex setting, the e¡ects of glacioeustatic cyclicity appear to be of primary importance in controlling late Pleistocene^Recent sediment distribution and preservation across this margin (Burger et al., 2002). However, tectonism locally overrides the presumed glacioeustatically controlled shelf stack-

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

324

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Fig. 11. Variable-intensity seismic image, showing detail of a slope channel from pro¢le 64 (see Fig. 10B for channel location). Note the high-amplitude basal ¢ll, buried by re£ectors of decreasing amplitude, suggesting a ¢ning-upward progression of channel-¢ll sediment.

ing pattern, demonstrating that relative roles of tectonism and glacioeustacy can vary over short distances (Burger et al., 2001, 2002). A similar contrast in modern slope morphologies (Fig. 1) indicates that these in£uences also vary along strike on the adjacent upper slope. 5.1.1. Contrasting preserved lithologies in the Humboldt Slide area The Humboldt Slide is located only V20 km from the mouth of the Eel River, the major sediment source for the o¡shore basin. However, the slide is also located just north of Eel Canyon, a prominent avenue for sediment bypassing during

lowstands (Burger et al., 2001). We expect resultant variations in sediment supply to a¡ect the morphological evolution of the Humboldt Slide area. Lowstand sediment input was likely reduced, as a result of bypassing to Eel Canyon (Figs. 1 and 12C; Burger et al., 2001). Sediment input from smaller sources to the north, such as the Mad River, may also have been minimal, if north-northwest directed sediment transport on the shelf, as indicated by a major 1995 £ood deposit (Borgeld, 1985; Bray and Greengrove, 1993; Wheatcroft et al., 1996), also prevailed during lowstand periods. Nonetheless, the sequence stratigraphic model proposed for the adjacent continental shelf (Burger et al., 2002) suggests that signi¢cant shelf sediment deposited during highstands is removed by shoreface erosion during both falling stages and transgressions and then transported basinward. Therefore, highstand shelf sediments must have been winnowed and redeposited on the upper slope south of the LSFZ (Figs. 12A, 12B, and 12D), leaving the coarsest fraction on the shelf as thin lag deposits (Burger et al., 2002). In contrast, a large proportion of sediment discharged o¡shore during highstands, as at Present, occurs primarily during large winter storm events (Brown and Ritter, 1971). A signi¢cant portion of this modern sediment bypasses the shelf, based on measured sediment budgets (Wheatcroft et al., 1997; Alexander and Simoneau, 1999; Sommer¢eld and Nittrouer, 1999). Some of this missing shelf sediment is probably transported to deep water by way of Eel Canyon (Mullenbach and Nittrouer, 1998).

Fig. 12. Interpreted areas of sediment transport, erosion and deposition on the upper Eel River Basin slope during di¡erent sealevel stages. (A) Highstand: some £ood-plume-borne sediment from the Eel River bypasses the shelf to the upper slope, depositing ¢ne-grained sediment in the Humboldt Slide area and hemipelagic sediment on the more distal slope north of the LSFZ. Slope channels and gullies north of the LSFZ are inactive. Some sediment may also bypass the shelf and upper slope to the deep ocean through Eel Canyon. (B) Falling stage: increasing amounts of Eel River sediment bypass the shelf and upper slope by way of Eel Canyon. Shelf sediment is eroded at the shoreface, supplying increasing amounts of ¢ne-grained sediment to the upper slope. The Mad River discharges sediment progressively closer to the upper slope, increasing slope sediment input north of the LSFZ. Channels north of the LSFZ reactivate along pre-existing trends, and also become bypass conduits. (C) Lowstand: most Eel River sediment bypasses the shelf and upper slope through Eel Canyon. Minimal sediment deposition is expected in the Humboldt Slide area. The Mad River extends to the shelf edge and sediment from it causes erosion of slope channels directly seaward of its mouth; more distal channels may be inactive. (D) Transgression: bypass of sediment through Eel Canyon decreases, and deposition of shoreface-eroded shelf sediment resumes in the Humboldt Slide area. However, as shelf sediment sources become more distal, slope channels north of the LSFZ begin to in¢ll, ¢rst by relatively coarse basal lags, then by ¢nergrained hemipelagic sediment (see Fig. 11).

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

However, highstand, ¢ne-grained sediment from the modern Eel River must also reach the upper slope in the Humboldt Slide area (Figs. 1 and 12A), transported in surface plumes during £oods, and in nepheloid layers during shelf resuspension events (Walsh and Nittrouer, 1999). Such an extrapolation of the Burger et al.

325

(2002) shelf model to the upper slope suggests that alternating deposition of falling stage/transgressive and highstand sediments can explain the ‘couplets’ of high- and low-amplitude sequences observed south of the LSFZ (Fig. 3). We interpret the high-amplitude slope sequences as highstand sediments, because we expect these directly depos-

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

326

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

ited sediments to be coarser-grained and therefore more re£ective than sediments winnowed during falling stages and transgressions. Low-amplitude sequences must therefore represent these winnowed sediments. The low-amplitude sequences appear to drape high-amplitude sequences (e.g. Fig. 3, HS11^HS12 sequence), as we would expect if these sequences were ¢ner-grained. Our hypothesis is also corroborated by the fact that the shallowest lower slide sequence is high-amplitude, as expected during the modern highstand (Fig. 3, HS12^Sea£oor). Shallow (1^3 m) cores and box cores from within the Humboldt Slide have shown that this sur¢cial sediment is primarily silt, with grain sizes ranging from V7 to 8 P (Field and Edwards, 1981; Gardner et al., 1999). Deeper sampling is needed to test whether buried lowamplitude sequences consist of ¢ner-grained clays and silts.

If each observed seismic couplet of a high- and low-amplitude sequence represents falling stage/ transgressive and highstand sediment packages, respectively, then they may also represent a single sea-level cycle. If so, we estimate that each couplet could represent V100 kyr, as late Pleistocene glacioeustatic £uctuations have been dominated by V100 kyr orbital eccentricity cycles (Ruddiman et al., 1989). The Humboldt Slide would then have been initiated V450 ka, as nine sequences/ V4.5 couplets occur between surface HS4, when deformation began, and the sea£oor (Figs. 3 and 4). A second estimate of the age of the slide can be made by considering the morphological change in sequences above surface HS4 (Fig. 3). Similar shifts in morphology observed on the adjacent shelf have been attributed to approach of the MTJ at V500 ka (Burger et al., 2002). We suspect that northward encroachment of the MTJ

Fig. 13. Uninterpreted and interpreted dip pro¢le 57r, showing possible correlation of surface HS4 in the Humboldt Slide area and a prominent shelf unconformity with an estimated age of V500 kyr (Burger et al., 2002). Seismic wipeout due to presumed gas-charged sediments near the shelf-break and strong sea£oor multiples make this correlation uncertain, but the overall inclinations of overlying re£ectors imply that these surfaces must be roughly correlative. This implies that HS4 was formed V500 ka. True dips shown at top left; bold, steeply dipping line segments represent faults. Numbered lines at top indicate locations of crossing pro¢les. See Fig. 1 for location of the seismic pro¢le.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

triggered the Humboldt Slide, sometime between V450 and 500 ka. Surface HS4, representing the slide’s initiation, must then be coeval with a surface of that age beneath the adjacent shelf. Although such a correlation cannot be made unequivocally across the shelf-break, HS4 appears correlative with a prominent V500 ka shelf unconformity (Fig. 13; Burger et al., 2002). Climate changes must also play a role in the variability of preserved sediments on the upper slope. During lowstand (glacial) periods, climate in this region was cooler and wetter than at present, and Eel River sediment load has been estimated at V25% greater during the last lowstand than at present (Syvitski and Morehead, 1999). However, increased discharge during lowstand periods would not signi¢cantly change the model as outlined above, because we would still expect much of this sediment to bypass through Eel Canyon. Although the relative in£uence of climate changes cannot discerned from the seismic data alone, it remains an important in£uence on sediment distribution that must also be considered when ground truth is obtained. 5.1.2. Sediment supply controls slope incision north of the LSFZ In contrast to the Humboldt Slide area, upper slope sequences north of the LSFZ do not display regular alternations of seismic amplitudes (Fig. 4). This slope lies further from the Eel River mouth (Fig. 1); we expect that modern Eel River £ood plumes reaching this part of the slope must be uniformly ¢ner-grained, which is con¢rmed by studies of modern sediment transport and depositional processes on the slope north of the LSFZ breached anticline (Alexander and Simoneau, 1999). Concordant, parallel re£ectors, observed in most areas of the northern upper slope (Fig. 9), also suggest a hemipelagic settling dominated by ¢ne-grained sediment (e.g. Knebel, 1984). However, the prominent network of buried channels (Fig. 10) indicates that changes in downslope transport occur. Topographic expression of buried channels persists at the modern sea£oor, but the shallowest channels are mostly in¢lled (Figs. 10 and 11), indicating that these channels are inactive today (Figs. 11, 12A and 14A). A

327

model for Eel River Basin gully formation on this slope proposed by Field et al. (1999) and Spinelli and Field (2001) con¢rms that smaller gullies are also inactive during highstands. During falling sea level, sediment input to the upper slope increases, as a result of both shoreface erosion as the shoreline regresses and direct £uvial input by way of an increasingly proximal mouth of the ancestral Mad River (Figs. 12B and 14B). Resultant increasing input of coarser-grained sediment initiates downslope sediment £ows and bypassing, which reactivates upper slope channels along preexisting trends (Fig. 10). At maximum lowstand, the ancestral Mad River extends to the shelf-edge, discharging sediment directly to the upper slope (Figs. 12C and 14C). As a result, channels directly seaward of the river mouth are preferentially eroded, which may explain observed larger channels at the northern end of the MCS grid (Fig. 10). During early transgression, slope channels remain active only as long as shelf sediment sources remain proximal. As £uvial and shoreface sources become more distal, upper slope channels begin to in¢ll (Figs. 12D and 14D). Coarsest sediments are deposited ¢rst; numerous channels display high-amplitude basal ¢ll (Fig. 11), suggesting the presence of coarse-grained lag deposits. Amplitudes of in¢lls progressively decrease upward (Fig. 11), presumably as shelf sediment sources recede and remaining channel topography is buried by ¢ner-grained sediment. All channels today are draped by low-amplitude re£ectors, representing highstand hemipelagic deposition (Figs. 9, 11 and 14E). 5.2. Proposed mechanisms of formation 5.2.1. Humboldt Slide Previous investigations of the Humboldt Slide have considered its formation. Geotechnical analyses of shallow cores within the slide zone by Lee et al. (1981) conclude that the sediment is somewhat overconsolidated ; they suggest that sediment failure is initiated by earthquake accelerations. More recently, Gardner et al. (1999) have used ultra-high-resolution, shallow-penetration (V65 m) Huntec boomer single-channel seismic pro¢les to conclude that deformation within the

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

328

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

slide has occurred by extension-related shearing, followed by rotation and folding of shearbounded blocks. Deformation began in the middle of the slide and then propagated both upslope and downslope, based on seismic interpretations that the largest displacements occur in the middle of the slide, with progressively less translation upslope and downslope. Gardner et al. have suggested that a combination of factors contributed

to deformation: frequent earthquakes, bubblephase gas in the sediment reducing shear strength, high rates of sediment accumulation, and tectonic oversteepening due to local uplift. In contrast to these earlier investigations, which considered only near-surface structures, our MCS images address the Humboldt Slide’s deeper internal structure and therefore its longevity. In the upper slide, faulting extends to the sea£oor (Figs. 3 and 4); mass wasting and brittle sediment deformation within the upper slide remain active. However, faulting is generally absent further seaward. The transition between undeformed and deformed sequences in the lower slide is sudden; sequences appear deformed only above surface HS4 (Figs. 3, 5 and 13). In addition, landwardthickening, Type 1 sequences below surface HS4 become parallel, Type 2 sequences of generally uniform thickness above (Fig. 3), suggesting a change in sediment supply. We suggest that after HS4, sediment supply and/or preservation rates decreased, perhaps associated with initiation of Humboldt Slide-related deformation. A similar change in sequence morphology is observed on the adjacent shelf (Burger et al., 2002). Sediment preservation rates decrease dra-

Fig. 14. Proposed sequence of upper slope channel development north of the LSFZ in response to sea-level £uctuations (see also Fig. 12). Arrows indicate the direction of proposed sediment transport; arrow sizes schematically represent the estimated magnitude of sediment supply and transport. Dashed line indicates upslope limit of channels during initial highstand. (A) Highstand: channels inactive, upper slope dominated by hemipelagic sedimentation. (B) Falling stage: channels reactivate along pre-existing trends, as shoreline regresses and sediment supply from £uvial sources and shoreface erosion increases. Channels also migrate upslope as a result of headward erosion. (C) Lowstand: Channels erode preferentially seaward of ancestral Mad River mouth. Other slope channels further from the river mouth may be inactive at maximum lowstand, due to a potential decrease in shoreface erosion during that stage. (D) Transgression: channels are active as shoreface transgresses and erodes the shelf. As sediment supply becomes more distal, channels in¢ll, initially with coarse-grained material followed by progressively ¢nergrained sediment (see Fig. 11). (E) Highstand: channels inactive again, and hemipelagic draping resumes. However, channels extend further upslope than during previous highstand (see Fig. 10).

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

matically at V500 ka, a trend that has continued to the Present. Burger et al. (2002) have interpreted the observed change in shelf sequence development as a response to uplift associated with the approach of the northward-migrating MTJ, causing a resultant decrease in shelf accommodation space. We interpret that the changes in sequence morphology on the upper slope to be a similar response to the MTJ. If so, then the Humboldt Slide began V500 ka, the same age assigned to the morphological change beneath the shelf (Burger et al., 2002). The reasonable correlation between surface HS4 and the V500 ka surface beneath the shelf (Fig. 13; Burger et al., 2002) supports this hypothesis. We have already hypothesized that uplift and increased seismicity associated with the northward-migrating MTJ initiated formation of Eel Canyon (Burger et al., 2002). Eel Canyon is a major sediment conduit to the deep sea, both during lowstands (Burger et al., 2001) and to a lesser extent during highstands (Mullenbach and Nittrouer, 1998). A concomitant decrease in sediment supply to the Humboldt Slide since formation of the canyon could explain the observed landward reduction in preserved slope sequence thicknesses above HS4 (Fig. 3). Furthermore, prominent channel incisions below surface HS4 die out above surface HS5 (Fig. 4), supporting reduced sediment supply to the Humboldt Slide after HS4 time. Periodic strong earthquakes have been cited as one of the most likely mechanisms that contribute

329

to sediment failures in the Eel River Basin (Lee et al., 1981; Field and Barber, 1993; Gardner et al., 1999). An earthquake of magnitude 6 occurs on average every 10 yr (Couch et al., 1974). Earthquake-induced sediment failures on continental slopes have been widely described, at many scales. For example, small-scale failures, involving V103 m3 of sediment, have been observed in the Gulf of Corinth (Papatheodorou and Ferentinos, 1997). In contrast, extremely large failures involving volumes s 1012 m3 also occur, such as the Storegga, Vigra, MWre, and Tampen slides o¡ western Norway (Bugge, 1983; Bugge et al., 1987, 1988; Evans et al., 1996). A lack of observed sediment failures on the upper slope north of the LSFZ (Fig. 9) supports our hypothesis that MTJ-related uplift and seismicity, concentrated near the southern end of the Eel River Basin (Fig. 2), initiated the Humboldt Slide. The Humboldt Slide is moderate-sized ; each of the deformed sequences involves volumes of V109 ^1010 m3 , based on measured areas of seismic deformation (Fig. 6) and estimated mean sequence thicknesses within these areas (Table 4). The total volume of the Humboldt Slide based on these estimates is V3.22U1010 m3 (Table 4). Imaged stratigraphic successions of many tectonically active slopes, including the eastern North Island of New Zealand (Barnes and Lewis, 1991), the Dead Sea (Niemi and Ben-Avraham, 1994), the northeast Faeroe margin (Van Weering et al., 1998) and the eastern Gulf of Cadiz (Baraza et al., 1999) display similar stacked sequences of deformed sediments, sug-

Table 4 Computations of estimated Humboldt Slide volumes Sequence (de¢ned by bounding surfaces)

Estimated average thickness (m)a

HS4^HS5 49 HS5^HS6 56 HS6^HS7 43 HS7^HS8 38 HS8^HS9 50 HS9^HS10 35 HS10^HS11 50 HS11^HS12 31 HS12^Sea£oor 33 Total estimated Humboldt Slide volume a

Measured deformed area (km2 ; see Fig. 6)

Calculated volume (m3 )

75 64 107 107 97 86 85 66 67

3.7U109 3.6U109 4.6U109 4.1U109 4.8U109 3.0U109 4.2U109 2.0U109 2.2U109 3.22U1010

Thickness calculated using mean estimated sonic velocity of 1800 m/s (twtt).

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

330

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

gesting that recurring seismicity creates long-term instabilities. The sudden initiation of Humboldt Slide deformation, probably coincident with onset of MTJ-related seismicity, supports the view that earthquake-induced sediment loading both triggered the slide initially and continues to deform its component sediments. Repetitive, cyclic seismic facies observed between sequences (Figs. 3 and 4 ; Table 2), and widely varying areas of deformation from one sequence to the next (Fig. 6), suggest multiple episodes of deformation after initiation of the Humboldt Slide. Seaward thickening of sequences in the slide zone would be expected if large amounts of downslope sediment translation have occurred. However, such downslope thickening in the Humboldt Slide does not occur, indicating more limited downslope sediment translation (Figs. 3 and 5). These observations support the conclusions of Gardner et al. (1999) that only a small amount of overall displacement occurs within the slide. However, observed truncations by mapped surfaces within deformed areas also suggest that they represent multiple glide planes within the slide (Fig. 5). Some downslope movement along these planes must also occur. Radiographs of split cores in the shallow slide zone show contorted bedding (Field and Edwards, 1981; Lee et al., 1981), con¢rming that plastic sediment deformation facilitates downslope motion. Such sediment creep and associated near-surface wavy seismic morphology have long been identi¢ed along the U.S. Atlantic continental slope (Aaron et al., 1980; Knebel, 1984; Silva and Booth, 1984) and in the Beaufort Sea (Hill et al., 1982). Repeated episodes of creep-dominated deformation attributed to long-term earthquake activity have also been documented, most notably along the continental slope of Israel (Almagor and Wiseman, 1982; Hawkins, 1983). Based on the imaged wavy re£ector morphologies in the shallow Humboldt Slide sequences, and the lack of imaged faults in the lower slide (Fig. 3), a process of earthquake-induced slow, gravitationally driven downslope motion of ¢ne-grained sediment may be today dominant in the Humboldt Slide. The observed wavy seismic facies in some Humboldt Slide sequences could also represent bed-

forms resulting from the action of slope-parallel currents (Gardner et al., 1996) or turbidity currents (Lee et al., 2002), rather than evidence for deformation. However, we consider this possibility unlikely because of observed truncation by interpreted internal glide planes (Fig. 5), and the large di¡erences in deformation across some surfaces in adjacent sequences (Fig. 6). Low-amplitude sequences generally display a lower degree of internal deformation then high-amplitude sequences, so their wavy facies could result from passive draping over deformed sediments. If so, these sequences would be expected to have smaller lateral extents of irregular re£ectors. That is con¢rmed by our maps (Fig. 6), which show the average mapped area for deformation of high-amplitude sequences to be V81 km2 , whereas the average mapped area for low-amplitude sequences is only 72 km2 (Fig. 6). Interestingly, the two mixed-amplitude sequences, HS6^HS7 and HS7^HS8, each exhibit the largest deformed area, 107 km2 (Fig. 6). This widespread deformation, is coupled with intense internal deformation (Figs. 3 and 4) suggests either two periods of an increased degree of sediment reworking, or that these sequences are composed of more easily deformable sediment. We suggest that an unusually high-amplitude glacioeustatic fall, perhaps inducing intense shelf erosion of sediment exhibiting a variety of grain sizes, contributed to their formation. Increased precipitation onshore inducing greater sediment loads might also result in greater mixed-sediment input to the shelf and slope, so climate shifts could also explain the observed seismic facies. Unfortunately, the lithologic makeup of the mixed-amplitude sequences cannot be understood until samples are obtained. Their formation represents a divergence from the cyclical process that appears to dominate the remainder of the Humboldt Slide succession. Nonetheless, for the most part the overall succession south of the LSFZ suggests a cyclical process of sediment deposition and preservation that has operated since before the Humboldt Slide was initiated and that has continued to the Present. Observed di¡erences in seismic facies suggest contrasting lithology and/or depositional energy.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Changes in alongshore currents, such as the California Current system (Hickey, 1979), in response to sea level and climate cycles, likely contribute to the cyclical variation in preserved facies. We interpret high amplitudes at many sequence boundaries as high impedance contrasts between contrasting lithologies. These bounding surfaces become conformable seaward of the slide, suggesting that they were not formed solely by downslope processes, but by primary depositional variations. Di¡erences in seismic facies between the two sequence types could also be enhanced by di¡erential compaction. Deformation within the lower slide appears to become more pronounced with depth, especially within high-amplitude sequences (Figs. 3 and 5). In contrast, low-amplitude sequences appear similar at all depths. All these observations suggest that lithological di¡erences contribute to low- vs. high-amplitude seismic response. Based on observed lateral and vertical seismic characteristics of Humboldt Slide sequences, along with previous seismic observations of the shallow slide (Gardner et al., 1999), we interpret a three-stage process for development of the upper slope in the Humboldt Slide region: 1. Sea-level variations, coupled with location of the Humboldt Slide proximal to both the Eel River and Eel Canyon, result in a cyclical pattern of sediment deposition south of the LSFZ. Lithologically distinct highstand vs. falling stage/transgressive sediment inputs form highand low-amplitude sequence couplets in the preserved record, respectively. Each couplet represents a single, V100 kyr sea-level cycle dominant since the late Pleistocene. Changes in sediment load and shifts in regional current systems in response to the sea-level variations likely contribute to the observed variability of slope facies. 2. Sur¢cial sequences undergo gravitationally driven creep deformation and/or fault-bounded rotation with limited downslope motion, facilitated in part by reductions in shear strength from periodic, earthquake-induced loading. Deformation occurs preferentially near the sea£oor, as the highest creep potential is at pre-

331

consolidation stress (Silva and Booth, 1984). However, contrasting shear strengths between sequences of di¡ering lithologies result in differential amounts of displacement between sequences ; lithological boundaries may act as zones of weakness and become glide planes within the slide zone (Fig. 5). 3. As sequences become more deeply buried, consolidation stress increases and creep potential decreases (Booth et al., 1983). Wavy morphology observed in shallow sequences is replaced by a stacked, shingled seismic facies at greater depths (Figs. 3 and 5), possibly as small-scale faults (near or below seismic resolution) form and sediments within sequence packages undergo continuing small-scale rotation as more coherent units. Di¡erential compaction may also accentuate contrasting seismic facies resulting initially from di¡ering grain sizes. The absence of shingled facies within low-amplitude sequences (Figs. 3 and 5) suggests that those ¢ner-grained sequences have higher shear strengths at depth, limiting continued creep deformation. 5.2.2. Gullied slope north of LSFZ The distribution of downslope-trending channels along the northern Eel River Basin upper slope, and the evolution of their morphologies through time, suggest that both sea-level changes and regional tectonism in£uence their development. Repeated channel formation suggests a cyclical process of formation, and their symmetrical, v-shaped cross-sections (Figs. 4 and 10) indicate the dominance of downslope processes; they appear una¡ected by slope-parallel currents. The channels are vertically stacked in most areas, suggesting reactivation along pre-existing trends. Width/depth ratios, mainly 5^25 (Table 3), indicate that the slope is mud-dominated and that these channels are stable (Galloway, 1998). We believe that the channels ¢rst become active during sea-level regressions, when seaward-directed sediment input to the upper slope increases (Fig. 12B^D). Because younger channels are expressed progressively farther upslope (Figs. 4 and 10), channels must migrate upslope through time as reactivation occurs. Similar headward erosion is

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

332

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

observed in canyons along the U.S. Atlantic coast (Pratson and Coakley, 1996). Channel distribution across the margin changes through time; older channels are concentrated to the north, while younger ones are more widely distributed (Fig. 10). These shifts re£ect the sensitivity of channel development to sediment supply. On the adjacent shelf, isochrons of preserved sequences suggest that, during the mid-Pleistocene, a major £uvial source discharged sediment to the margin in the general vicinity of the modern Mad River mouth (Burger et al., 2002; Fig. 1). Since then, the Eel River has progressively become the dominant sediment source (Burger et al., 2002). The observed transition, from fewer large channels concentrated at the north end of the grid to smaller, more numerous and uniformly distributed channels (Fig. 10), probably re£ects this diminution of the northern £uvial source and transition to Eel River-dominated £uvial input. More recent channels re£ect a line-source of sediment, perhaps a response to increased shoreface erosion across this high-energy margin (Fig. 12). MTJ-related uplift may have accentuated such shoreface erosion during both falling stages and transgressions since V500 ka, ultimately causing the observed shift to a broader distribution of slope channels north of the LSFZ (Fig. 10). 5.3. Implications of age estimates The correlations that have been attempted north of the LSFZ with dated shelf surfaces (Burger et al., 2002) provide possible age constraints for some mapped upper slope features. If NS5 correlates with shelf surface 11, estimated at V300 ka (Burger et al., 2002), more than one generation of slope channels form during a V100 kyr sea-level cycle, as at least ¢ve prominently incised slope surfaces (NS6, NS7, NS8, NS10, and NS11; Fig. 9) overlie NS5. Channel formation at this greater frequency may occur because they are incised during both falling stage and transgression (Fig. 14B,D), but become largely inactive during lowstands as well as highstands, when shelf gradients stabilize and shoreface erosion decreases (Fig. 14A,C). Resultant decreased sediment input to the slope could

accentuate hemipelagic sediment input to channels during those times. However, better ages are needed by sampling these slope surfaces to constrain links between incision episodes and known global cyclicity.

6. Conclusions Variation of seismically observed slope features imaged along strike in the southern Eel River Basin exempli¢es how diverse in£uences, including seismicity, known Pleistocene base-level changes, and related spatial variations in sediment supply, can a¡ect preserved slope morphologies. Northward migration of the MTJ caused a fundamental shift in the accumulation, preservation and postdepositional deformation of sedimentary sequences V500 ka. This shift is best expressed by sudden initiation of sediment deformation in the Humboldt Slide area, and a general decrease in sediment preservation rates seaward of the shelfbreak due to increased sediment-bypassing through the presumed newly formed Eel Canyon. These e¡ects are most pronounced at the southern end of the Eel River Basin. Multiple stacked, deformed sequences suggest a sudden initiation for the Humboldt Slide and con¢rm that it continues to deform. Alternating sequence couplets of di¡ering amplitudes and seismic facies result from deposition of lithologically distinct, highstand vs. falling stage/transgressive sediment packages, deposited during late Pleistocene^Recent V100 kyr glacioeustatic cycles. Lowstand sediments preferentially bypass this area through Eel Canyon. Based on the number of observed sequences, their interpreted representation of known sea-level cycles, and circumstantial evidence from the shelf that the Humboldt Slide was initiated approximately coincident with initial encroachment of the MTJ into the southern basin, the Humboldt Slide may be V450^500 kyr old. The gullied slope north of the LSFZ suggests a di¡erent response to base-level £uctuations, sediment supply and regional tectonism. Large-scale mass wasting features are not observed, suggesting that MTJ-related seismicity and tectonism

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

does not yet deform this portion of the margin. Instead, the slope is dominated by downslopetrending channels and gullies that appear to be sensitive to changes in sediment supply modulated by base-level changes. Long-term changes in channel distribution suggest decreasing importance of £uvial point-sources north of the LSFZ and increasing e⁄ciency of shoreface erosion along the margin through time. Morphology of these channels and the seismic character of their ¢lls suggest that they are incised by downslopeeroding £ows between falling stage and early transgression, then in¢lled during late transgression as £uvial and shoreface sediment sources become more distal. Channels are inactive and ¢lled by hemipelagic sediment during highstands, as at Present, and possibly during lowstands as well. Channels may incise during maximum lowstand, but possibly only directly seaward of an ancestral river mouth, as was true of channels seaward of the Mad River. Along-strike contrasts in preserved upper slope sequence morphologies in the southern Eel River Basin appear to be a function of both regional tectonics and changing proximity to local shelf sediment sources, in£uenced by base-level £uctuations. The proximity of the Humboldt Slide region to the northward-migrating MTJ is likely the causative factor in initiating slope failure there, while the more distal northern slope is relatively una¡ected by earthquake-induced mass wasting. Seismic observations of upper slope sequences in the southern Eel River Basin reinforce previous seismic interpretations of the adjacent shelf (Burger et al., 2002), which conclude that geophysical studies of tectonically complex margins can di¡erentiate forcing e¡ects of tectonism and glacioeustacy on the preservation of stratigraphy.

Acknowledgements This work was supported by O⁄ce of Naval Research (ONR) Contracts N00014-96-1-0359 and N00014-02-1-0032 to C.S.F. and J.A.A., part of the STRATAFORM initiative (program manager, J. Kravitz), and ONR’s AASERT pro-

333

gram, Contract N00014-97-1-0555, to C.S.F. and J. Go¡, UTIG. We thank G. Mountain and J. Diebold, Lamont-Doherty Earth Observatory (L-DEO), for their contributions to data acquisition and processing. Thoughtful reviews by James Gardner and Lionel Carter improved the manuscript, and we thank them for their e¡orts. S. Saustrup, UTIG, and P. Buhl, L-DEO, supervised seismic data processing. We also thank the captain and crew of R/V Wecoma, cruise #9605B. University of Texas Institute for Geophysics Contribution No. 1642.

References Aaron, J.M., Butman, B., Bothner, M.H., Sylwester, R.E., 1980. Maps showing environmental conditions relating to potential geologic hazards of the United States northeastern Atlantic continental margin. U.S. Geological Survey Miscellaneous Field Studies Map, MF-1193, 3 sheets. Alexander, C.R., Simoneau, A.M., 1999. Spatial variability in sedimentary processes on the Eel continental slope. Mar. Geol. 154, 243^254. Almagor, G., Wiseman, G., 1982. Submarine slumping and mass movements on the continental slope of Israel. In: Saxov, S., Nieuwenhuis, J.K. (Eds.), Marine Slides and Other Mass Movements. Nato Conference Series IV - Marine Sciences, vol. 6, pp. 95^128. Atwater, T., 1970. Implications of plate tectonics for the Cenozoic tectonic evolution of western North America. Geol. Soc. Am. Bull. 81, 3513^3536. Bachman, S.B., Crouch, J.K., 1987. Geology and Cenozoic history of the northern California margin; Point Arena to Eel River. In: Ingersoll, R.V., Ernst, W.G. (Eds.), Cenozoic Basin Development of Coastal California, Rubey Vol., pp. 124^145. Baraza, J., Ercilla, G., Nelson, C.H., 1999. Potential geologic hazards on the eastern Gulf of Cadiz slope (SW Spain). Mar. Geol. 155, 191^215. Barnes, P.M., Lewis, K.B., 1991. Sheet slides and rotational failures on a convergent margin: the Kidnappers Slide, New Zealand. Sedimentology 38, 205^221. Blake Jr., M.C., Jones, D.L., 1974. Origin of Franciscan me¤langes in northern California. In: Dott Jr., R.H., Shaver, R.H. (Eds.), Modern and Ancient Geosynclinal Sedimentation; Problems of Palinspastic Restoration. SEPM Spec. Publ. 19, 345^357. Booth, J.S., Silva, A.J., Jordan, S.A., 1983. Slope-stability and creep susceptibility of Quaternary sediments on the northeastern United States continental slope. In: Denness, B. (Ed.), Seabed Mechanics. Graham and Trotman, London, pp. 65^75. Borgeld, J.C., 1985. Holocene stratigraphy and sedimentation

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

334

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

on the northern California continental shelf. Ph.D. Dissertation Thesis, University of Washington, Seattle, WA, 177 pp. Bray, N.H., Greengrove, C.L., 1993. Circulation over the shelf and slope o¡ northern California. J. Geophys. Res. 98, 18119^18146. Brooks, J.M., Field, M.E., Kenicutt, M.C.II., 1991. Observations of gas hydrates o¡shore northern California. Mar. Geol. 96, 103^109. Brown III, W.M., Ritter, J.R., 1971. Sediment transport and turbidity in the Eel River Basin, California. U.S. Geological Survey Water-Supply Paper W 1986, 70 pp. Bugge, T., 1983. Submarine slides on the Norwegian continental margin, with special emphasis on the Storegga area. Continental Shelf and Petroleum Research Institute (IKU) Publ. 110, Trondheim, 152 pp. Bugge, T., Befring, S., Belderson, R.H., Eidvin, T., Jansen, E., Kenyon, N.H., Holtedahl, H., Sejrup, H.P., 1987. A giant three-stage submarine slide o¡ Norway. Geo-Mar. Lett. 7, 191^198. Bugge, T., Belderson, R.H., Kenyon, N.H., 1988. The Storegga Slide. Philos. Trans. R. Soc. London A 325, 357^388. Burger, R.L., 2002. Mid-Pleistocene to Present stratigraphic responses in a tectonically-driven depositional setting: o¡shore Eel River Basin, northern California. Ph.D. Dissertation, The University of Texas at Austin, TX. Burger, R.L., Fulthorpe, C.S., Austin, J.A., Jr., 2001. Late Pleistocene channel incisions in the southern Eel River Basin, northern California: implications for tectonic vs. eustatic in£uences on shelf sedimentation patterns. Mar. Geol. 177, 317^330. Burger, R.L., Fulthorpe, C.S., Austin, J.A., Jr., Gulick, S.P.S., 2002. Lower Pleistocene to Present structural deformation and sequence stratigraphy of the continental shelf, o¡shore Eel River Basin, northern California. Mar. Geol. 185, 51^ 183. Carver, G.A., 1987. Late Cenozoic tectonics of the Eel River Basin region, coastal northern California. In: Schymiczek, H., Suchland, R. (Eds.), Tectonics, Sedimentation, and Evolution of the Eel River and Other Coastal Basins of Northern California. San Joaquin Geol. Soc., pp. 61^72. Clarke Jr., S.H., 1987. Late Cenozoic geology and structure of the onshore-o¡shore Eel River Basin, northern California. In: Schymiczek, H., Suchland, R. (Eds.), Tectonics, Sedimentation, and Evolution of the Eel River and Other Coastal Basins of Northern California. San Joaquin Geol. Soc., pp. 31^40. Clarke, S.H., Jr., 1992. Geology of the Eel River Basin and adjacent region: implications for late Cenozoic tectonics of the southern Cascadia Subduction Zone and Mendocino Triple Junction. AAPG Bull. 76, 199^224. Clarke, S.H., Jr., Carver, G.A., 1992. Late Holocene tectonics and paleoseismicity, southern Cascadia subduction zone. Science 255, 188^192. Couch, R.W., 1980. Seismicity and crustal structure near the north end of the San Andreas fault system. In: Streitz, R., Sherburne, R. (Eds.), Studies of the San Andreas Fault Zone

in Northern California. California Div. Mines and Geology Spec. Report 140, 187 pp. Couch, R.W., Victor, L.P., Keeling, K.M., 1974. Coastal and o¡shore earthquakes of the paci¢c northwest between 39‡ and 49‡ latitude and 123‡ and 131‡ longitude. School of Oceanography, Oregon State University Press, Corvallis, OR, 67 pp. Drake, D.E., 1999. Temporal and spatial variability of the sediment grain-size distribution on the Eel shelf: the £ood layer of 1995. Mar. Geol. 154, 169^182. Evans, D., King, E.L., Kenyon, N.H., Brett, C., Wallis, D., 1996. Evidence for long-term instability in the Storegga Slide region o¡ western Norway. Mar. Geol. 130, 281^292. Field, M.E., Barber Jr., J.H., 1993. A submarine landslide associated with shallow sea£oor gas and gas hydrates o¡ northern California. In: Schwab, W.C., Lee, H.J., Twichell, D.C. (Eds.), Submarine Landslides: Selected Studies in the U.S. Exclusive Economic Zone. U.S. Geol. Surv. Bull. 2002, pp. 151^157. Field, M.E., Edwards, B.D., 1980. Slopes of the southern California borderland: a regime of mass transport, in processes of the Quaternary depositional environments of the Paci¢c Coast. In: Field, M.E., Bouma, A.H., Colbourn, I.P., Douglas, R.G., Ingle, V.C. (Eds.), Paci¢c Coast Paleogeography Symposium 4, SEPM Paci¢c Section, Bakers¢eld, CA, pp. 169^184. Field, M.E., Edwards, B.D., 1981. Large submarine slump o¡ Eureka, California. AAPG Bull. 65, 924^925. Field, M.E., Gardner, J.V., Prior, D.B., 1999. Geometry and signi¢cance of stacked gullies on the northern California slope. Mar. Geol. 154, 271^286. Fulthorpe, C.S., Mountain, G.S., Austin, J.A., Jr., Buhl, P., Diebold, J., Go¡, J.A., Schuur, C.L., Yun, J., 1996. STRATAFORM high-resolution MCS survey, Eel River basin, Northern California margin; shelf/slope stratigraphy and processes. AGU 1996 Fall Meeting, EOS Trans. AGU 77, 330. Furlong, K.P., Govers, R., 1999. Ephemeral crustal thickening at a triple junction: the Mendocino crustal conveyor. Geology 27, 127^130. Galloway, W.E., 1998. Siliciclastic slope and base-of-slope depositional systems: component facies, stratigraphic architecture, and classi¢cation. AAPG Bull. 82, 569^595. Gardner, J.V., Field, M.E., Lee, H.J., Prior, D.B., 1996. Are the Humboldt Slide ‘blocks’ slide blocks or antidunes? EOS Trans. AGU 77, F330. Gardner, J.V., Prior, D.B., Field, M.E., 1999. Humboldt Slide - a large shear-dominated retrogressive slope failure. Mar. Geol. 154, 323^338. Go¡, J.A., Orange, D.A., Mayer, L.A., Clarke, J.E.H., 1999. Detailed investigation of continental shelf morphology using a high-resolution swath sonar survey: the Eel margin, northern California. Mar. Geol. 154, 255^269. Gulick, S.P., 1999. Seismic studies of the southern Cascadia Subduction Zone near the Mendocino Triple Junction. Ph.D. Dissertation, Lehigh University, 257 pp. Gulick, S.P., Meltzer, A.M., Clarke, S.H,Jr., 2002. E¡ect of

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

R.L. Burger et al. / Marine Geology 199 (2003) 307^336 the northward migrating Mendocino Triple Junction on the Eel River forearc basin, California: stratigraphic development. Geol. Soc. Am. Bull. 114, 178^191. Gulick, S.P., Meltzer, A.M., 2002. E¡ect of the northward migrating Mendocino Triple Junction on the Eel River forearc basin, California: structural evolution. Geol. Soc. Am. Bull. 114, 1505^1519. Hawkins, A. B., 1983. Submarine slope failures. In: Denness, B. (Ed.), Seabed Mechanics. Graham and Trotman, London, pp. 51^64. Hickey, B.M., 1979. The California Current System - hypotheses and facts. Prog. Oceanogr. 8, 191^279. Hill, P.R., Moran, K.M., Blasco, S.M., 1982. Creep deformation of slope sediments in the Canadian Beaufort Sea. GeoMar. Lett. 2, 163^170. Knebel, H.J., 1984. Sedimentary processes on the Atlantic continental slope of the United States. Mar. Geol. 61, 43^74. Lee, H.J., Edwards, B.D., Field, M.E., 1981. Geotechnical analysis of a submarine slump, Eureka, California. Proc. O¡shore Technology Conference, Houston, TX. OTC 4121, 53^59. Lee, H.J., Locat, J., Dartnell, P., Israel, K., Wong, F., 1999. Regional variability of slope stability: application to the Eel margin, California. Mar. Geol. 154, 305^322. Lee, H.J., Syvitski, J.P.M., Parker, G., Orange, D., Locat, J., Hutton, E.W., Imran, J., 2002. Distinguishing sediment waves from slope failure deposits: ¢eld examples, including the ‘Humboldt slide’, and modeling results. Mar. Geol. 192, 79^104. Leithold, E.L., Hope, R.S., 1999. Deposition and modi¢cation of a £ood layer on the northern California shelf: lessons from and about the fate of terrestrial particulate and organic carbon. Mar. Geol. 154, 183^196. Mayer, L., Fonesca, L., Pacheco, M., Galway, S., Martinez, J., Hou., T., 1999. STRATAFORM. CD-ROM: http://www. strataform.unh.edu/. McCrory, P.A., 1989. Late Neogene geohistory analysis of the Humboldt Basin and its relationship to convergence of the Juan de Fuca Plate. J. Geophys. Res. B 94, 3126^ 3138. McCrory, P.A., 1995. Evolution of a trench-slope basin within the Cascadia subduction margin: the Neogene Humboldt Basin, California. Sedimentology 43, 223^247. McCrory, P.A., 1996. Evaluation of fault hazards, northern coastal California. U.S. Department of the Interior, U.S. Geological Survey Open File Report 96-656. McCrory, P.A., 2000. Upper plate contraction in the vicinity of the Mendocino Triple Junction: implications for partitioning of strain. Tectonics 19, 1144^1160. McKenzie, D.P., Morgan, W.J., 1969. Evolution of triple junctions. Nature 224, 125^133. McLaughlin, R.J., Sliter, W.V., Frederiksen, N.O., Harbert, W.P., McCulloch, D.S., 1994. Plate motions recorded in tectonostratigraphic terranes of the Franciscan Complex and evolution of the Mendocino Triple Junction, northwestern California. U.S. Geol. Surv. Bull. B 1997, 60 pp. Mullenbach, B.L., Nittrouer, C.A., 1998. Supply of £uvial

335

sediment to the head of the Eel Canyon. EOS Trans. AGU (suppl.) 79, F466. Niemi, T.M., Ben-Avraham, Z., 1994. Evidence for Jericho earthquakes from slumped sediments of the Jordan River delta in the Dead Sea. Geology 22, 395^398. Nilsen, T.H., Clarke Jr., S.H., 1987. Geological evolution of the late Cenozoic basins of northern California. In: Schymiczek, H., Suchland, R. (Eds.), Tectonics, Sedimentation, and Evolution of the Eel River and Other Coastal Basins of Northern California. San Joaquin Geol. Soc., pp. 15^29. Nittrouer, C.A., 1999. STRATAFORM: overview of its design and synthesis of its results. Mar. Geol. 154, 3^12. Nittrouer, C.A., Kravitz, J.H., 1996. STRATAFORM: A program to study the creation and interpretation of sedimentary strata on continental margins. Oceanography 99, 146^152. Ogle, B.A., 1953. Geology of the Eel River Valley area, Humboldt County, California. California Div. Mines Bull. 164, 128 pp. Ogston, A.S., Sternberg, R.W., 1999. Sediment-transport events on the northern California continental shelf. Mar. Geol. 154, 69^82. Orange, D.L., 1999. Tectonics sedimentation. and erosion in northern California: submarine geomorphology and sediment preservation potential as a result of three competing processes. Mar. Geol. 154, 369^382. Papatheodorou, G., Ferentinos, G., 1997. Submarine and coastal sediment failure triggered by the 1995, Ms = 6.1 R Aegion earthquake, Gulf of Corinth, Greece. Mar. Geol. 137, 287^304. Pratson, L.F., Coakley, B.J., 1996. A model for the headward erosion of submarine canyons induced by downslope-eroding sediment £ows. Geol. Soc. Am. Bull. 108, 225^234. Ruddiman, W.F., Raymo, M.E., Martinson, D.G., Clement, B.M., Backman, J., 1989. Pleistocene evolution: northern hemisphere ice sheets and north Atlantic Ocean. Paleoceanography 4, 353^412. Silva, A.J., Booth, J.S., 1984. Creep behavior of submarine sediments. Geo-Mar. Lett. 4, 215^219. Sommer¢eld, C.K., Nittrouer, C.A., 1999. Modern accumulation rates and a sediment budget for the Eel Shelf: a £ooddominated depositional environment. Mar. Geol. 154, 227^ 241. Spinelli, G.A., Field, M.E., 2001. Evolution of continental slope gullies on the northern California margin. J. Sediment. Res. 71, 237^245. Syvitski, J.P., Morehead, M.D., 1999. Estimating river-sediment discharge to the ocean: application to the Eel margin, northern California. Mar. Geol. 154, 13^28. Van Weering, T.C.E, Nielsen, T., Kenyon, N.H., Akentieva, K., Kuijpers, A.H., 1998. Large submarine slides on the NE Faeroe continental margin. In: Stoker, M.S., Evans, D., Cramp, A. (Eds.), Geological Processes on Continental Margins: Sedimentation, Mass-Wasting and Stability. Geol. Soc. London Spec. Publ. 129, 5^17. Walsh, J.P., Nittrouer, C.A., 1999. Observations of sediment £ux to the Eel continental slope, northern California. Mar. Geol. 154, 55^68.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart

336

R.L. Burger et al. / Marine Geology 199 (2003) 307^336

Wheatcroft, R.A., Borgeld, J.C., Born, R.S., Drake, D.E., Leithold, E.L., Nittrouer, C.A., 1996. The anatomy of an oceanic £ood deposit. Oceanography 9, 158^162. Wheatcroft, R.A., Sommer¢eld, C.K., Drake, D.E., Borgeld, J.C., Nittrouer, C.A., 1997. Formation of a £ood-deposit on the northern California continental shelf. Geology 25, 163^ 166.

Woodward-Clyde Consultants, 1980. Evaluation of the potential for resolving the geologic and seismic issues at the Humboldt Bay Power Plant Unit No. 3. Unpublished ¢nal report prepared for Paci¢c Gas and Electric Company, 309 pp. Yun, J.W., Orange, D.L., Field, M.E., 1999. Subsurface gas o¡shore of northern California and its link to submarine geomorphology. Mar. Geol. 154, 357^368.

MARGO 3362 14-8-03 Cyaan Magenta Geel Zwart