Mass-transport-deposits-and-gas-hydrate-occurrences-in-the-U

Mass transport deposits and gas hydrate occurrences in the Ulleung Basin,East Sea e Part 1:Mapping sedimentation patterns using seismic coherency

N.A.Scholz a ,M.Riedel a ,b ,*,J.-J.Bahk c ,D.-G.Yoo c ,B.-J.Ryu c

a

School of Earth and Ocean Science,University of Victoria,Canada

b

Natural Resources Canada,Geological Survey of Canada e Paci ?c,Sidney Subpision,9860West Saanich Road,Sidney,BC V8L4B2,Canada c

Korea Institute of Geoscience and Mineral Resources,KIGAM,Daejeon,Republic of Korea

a r t i c l e i n f o

Article history:

Received 6September 2011Received in revised form 9March 2012

Accepted 14March 2012

Available online 3April 2012Keywords:

Mass transport deposits Seismic coherency attribute Outrunner blocks Gliding tracks Pressure ridges Gas hydrate

Depositional environments

a b s t r a c t

Seismic coherency measures,such as similarity and dip of maximum similarity,were used to characterize mass transport deposits (MTDs)in the Ulleung Basin,East Sea,offshore 00da2ecbcc175527072208f6ing 2-D and 3-D seismic data several slope failure masses have been identi ?ed near drill site UBGH1-4.The MTDs have a distinct seismic character and exhibit physical properties similar to gas hydrate bearing sediment:elevated electrical resistivity and P-wave velocity.Sediments recovered from within the MTDs show a reworked nature with chaotic assemblage of mud-clasts.Additionally,the re ?ection at the base of MTDs is polarity reversed relative to the sea ?oor,similarly to the bottom-simulating re ?ector commonly used to infer the presence of gas hydrates.The MTDs further show regional seismic blanking (absence of internal re ?ectivity),which is yet another signature often attributed to gas hydrate bearing sediments.At the drill site UBGH1-4,no gas hydrate was recovered in sediment-cores from inside a prominent MTD unit.Instead,pore-?lling gas hydrate was recovered only within thin turbidite sand layers near the base of the gas hydrate stability zone.With the analysis of seismic attributes,the seismic character of the prominent MTD (Unit 3)was investigated.The base of the MTD unit exhibits deep grooves interpreted as gliding tracks from either outrunner blocks or large clasts that were dragged along the paleo-sea ?oor.Similar seismic features were identi ?ed on the sea ?oor although the length of the gliding tracks on the sea ?oor is much shorter (a few hundred meters to w 1km),compared to over 10km long tracks at the base of the MTD.The seismic coherency attributes allowed to estimate the volume of the failed sediment as well as the direction of the ?ow of sediment.Tracking the MTD and extrapolating its spatial extent from the 3-D seismic volume to adjacent 2-D seismic pro ?les,a possible source region of this mass failure was de ?ned w 50km upslope of Site UBGH1-4.

Crown Copyright ó2012Published by Elsevier Ltd.All rights reserved.

1.Introduction

As part of the South-Korean gas hydrate program regional 2-D and 3-D seismic data were acquired in the Ulleung Basin,East Sea,off Korea (Fig.1),to characterize the sediment-depositional environments and map possible gas hydrate accumulations (Ryu et al.,2009;Park,2008;Park et al.,2008;Lee et al.,2005).The Ulleung Basin Gas Hydrate Drilling Expedition 1(UBGH1)in 2007(see Fig.1for location of drill sites)consisted of a logging-while-drilling (LWD)operation at ?ve sites (UBGH1-1,UBGH1-4,UBGH1-9,UBGH1-10,and UBGH1-14),spot-coring operation at

three of those sites (UBGH1-4,UBGH1-9,and UBGH1-10),as well as wire-line logging and vertical seismic pro ?ling at Site UBGH1-9.The core and logging data together with the pre-drilling seismic data were used to assess the gas hydrate potential and to estimate the possible amount of in place methane resource in the basin (Park et al.,2008;Chun et al.,2008).

Regional seismic blanking has initially been considered as an indicator for the presence of gas hydrates within the Ulleung Basin,following similar observations elsewhere (e.g.at the Blake Ridge (Lee and Dillon,2001)and the Cascadia margin (Riedel et al.,2002)).However,several authors suggested that sedimentation in the slope regions of the Ulleung Basin is dominated by MTDs,(e.g.Lee et al.,2001;Lee and Kim,2002;Bahk et al.,2004).Compared to the sea ?oor the base of the MTDs is polarity reversed indicating a reduction in acoustic impedance.The widespread occurrence of MTDs across the Ulleung Basin together with their polarity-reversed characteristics and an accompanying lack in seismic

*Corresponding author.Natural Resources Canada,Geological Survey of Cana-da e Paci ?c,Sidney Subpision,9860West Saanich Road,Sidney,BC V8L4B2,Canada.Tel.:t12503636422.

E-mail address:mriedel@nrcan.gc.ca (M.

Riedel).

Contents lists available at SciVerse ScienceDirect

Marine and Petroleum Geology

journal h omepage:ww w.elsevi 00da2ecbcc175527072208f6/locate/marp

etgeo

0264-8172/$e see front matter Crown Copyright ó2012Published by Elsevier Ltd.All rights reserved.doi:10.1016/j.marpetgeo.2012.03.004

Marine and Petroleum Geology 35(2012)91e 104

internal re ?ectivity can complicate the de ?nition of the base of the gas hydrate stability zone (GHSZ).Seismically the base of the GHSZ is seen as a bottom-simulating re ?ector (BSR),also having an opposite-to-sea ?oor re ?ection polarity.In this study we focus on the characteristics of MTDs in seismic,core,and log data and show that seismic blanking,as well as log-derived increased P-wave velocity and electric resistivity can also be caused by the occurrence of MTDs and not just by pore-?lling gas hydrate.The goal is to understand and de ?ne the three-dimensional geometry of the MTD especially gliding tracks and outrunner blocks.The understanding of the depositional history of a MTD is an important component in the basin-wide assessment of gas hydrate as potential resource for Korea,and to predict coarse-grained sandy sections that are known to represent preferred host sediments for gas hydrates (e.g.Clennell et al.,1999;Torres et al.,2008).The role of MTDs and possible linkages with gas hydrate occurrences have previously been studied especially at the Storegga Slide offshore Norway (e.g.Bouriak et al.,2000;Bünz et al.,2005;Mienert et al.,2005).Gas

hydrate dissociation may have played a role in the slide-initiation at the Norwegian margin (e.g.Mienert et al.,1998;Vogt and Jung,2002).Slide mechanics,possible trigger mechanisms,and sliding behavior such as hydroplaning of the slide have been extensively studied at the Storegga Slide as well (e.g.De Blasio et al.,2005;Gauer et al.,2005;Kvalstad et al.,2005).

Understanding the ?ow behavior of subaqueous slides is of major importance to the protection of any existing or future offshore infrastructure.The Ulleung Basin has experienced a large number of slope failures in the past (Lee et al.,1996)and results show that it most likely is still experiencing underwater sliding events.

This study presents the ?rst elements of a risk analysis in the Ulleung Basin that will become especially necessary if gas hydrates are found to be a viable energy resource for South Korea.With the completion of UBGH1and recently also UBGH2,an assessment of local gas hydrate occurrence has already been conducted (Kim et al.,2011;Bahk et al.,2011)and gas hydrate production tests will soon be under way.Should those test results demonstrate the feasibility of a decades-long hydrocarbon-exploitation based on gas hydrates then mass ?ow dynamics will become a major issue for production safety.

With this in mind we want to highlight the ef ?cacy of seismic attributes such as the measurement of coherency (or similarity)between neighboring traces as a means to thoroughly study MTD ?ow behavior as well as to discern MTDs from gas hydrate hosting sediment and to characterize important sediment features.Our interpretation of seismic coherency is guided by comparisons with studies by Nissen et al.(1999)from the Nigerian continental slope,Prior et al.(1982)from the Kitimat-slide,off British Columbia,as well as by Posamentier and Walker (2006).In the companion paper by Riedel et al.the LWD logging data are further used to estimate gas hydrate concentrations and linking fracture patterns seen on electrical resistivity data to the seismically de ?ned ?ow-pattern in the MTD.

2.Geologic setting of the Ulleung Basin

The Ulleung Basin,a continental back-arc basin,lies on the eastern margin of the Eurasian Plate and is separated from the Paci ?c by the Japanese Islands.The East Sea,to the west of Japan,has its origin in the Oligocene when crustal thinning and sea ?oor spreading led to the development of the Ulleung,Japan,and Yamato basins (e.g.Chough and Barg,1987;Lee et al.,2001;Ryu et al.,2009).

These basins in the East Sea are separated from each other by the Korea Gap and continental remnants such as the Oki Bank and the Korea Plateau (Park,2007).The Ulleung Basin is bound by the steep slopes of the Korean Peninsula to the west and by the Korea Plateau to the north (Fig.1a).To the east and southeast lie the more gentle slopes of Oki Bank and the Japanese Islands.The ocean ?oor of the Ulleung Basin is fairly smooth in the center,and deepens north-eastwards from 1000m water depth on the basin margin to about 2300m near the Korea Gap (Park,2007).The crustal thick-ness decreases from 10km in the south to 5km in the central part,thus having a wedge-like shape (Lee and Suk,1998).This was interpreted by Lee and Suk (1998)as an indication that the main sediment source and the location of maximum subsidence lies in the south.As the depositional energy of the mass transport processes dropped with increasing distance to the source,inter-bedded turbidites and hemipelagic sediments were deposited further to the northeast of the basin.The mass transport processes were reduced in scale but reached regions further to the north.Lee et al.(2001)mention two distinct phases consisting of widespread distribution of MTDs in the late Neogene and extensive turbidite and hemipelagic sedimentation in the Pleistocene

and

Figure 1.Ulleung Basin,East Sea,South Korea;(a)location of the Ulleung Basin in the East Sea;(b)Area (see inset-box in (a))of the 2-D seismic lines and the 3-D survey,as well as the location of the drilling sites from UBGH1expedition of 2007.

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104

92

Holocene.Uplift and deformation in the south and southeast, having started in the middle Miocene due to back-arc closure,are seen as causes for the enormous volumes of sediments found in the basin(Lee et al.,2001).The southern basin subsided due to sedi-ment loading and lithospheric cooling.The Ulleung Basin margin experienced a subsidence rate of700m/m.y.until the middle Miocene,with an especially rapid initial phase,characteristic for the Ulleung Basin(Chough and Barg,1987).This was followed by a phase of uplifting in the late Miocene and slight subsidence in the Pliocene e Pleistocene.As the number of mass transport?ows decreased in the late Miocene,a prominent change in sedimentary facies in the central basin occurred,with a transition from coarse-grained,high energy deposits to more?ne-grained-low-energy turbidites and hemipelagic sediments(Lee and Suk,1998).With its large number of MTDs the sedimentation pattern in the Ulleung Basin is different from the sedimentation in the Japan and Yamato basins in the East Sea,which both are dominated by slowly deposited,hemipelagic sediments,and turbidites.

3.Data and methods

The data set to investigate the area around drill site UBGH1-4 consists of2-D and3-D multichannel seismic lines(Fig.1b).The data were acquired in2005and2006,using the research vessel TAMHAE-II of the Korea Institute of Geoscience and Mineral Resources(KIGAM).The seismic pro?les were acquired using an airgun source volume of4.9l(1035in3),and a streamer of240 channels,with shot and group intervals of12.5m and6.25m, respectively.The3-D data set as a bin-spacing of25m by25m, with a total of320inlines and1000cross-lines(Fig.1b).The3-D data set covers an area of400km2surrounding the drilling site UBGH1-4.The frequency spectrum of the2-D and3-D seismic surveys ranges from10to80Hz,but dominantly at50Hz,which corresponds to an approximate vertical resolution of8m(using average seismic velocities of1600m/s).

To map the MTD and to depict its features,geometric attributes such as seismic coherency(semblance)and dip of maximum similarity were used.Seismic coherency is a measure of similarity between adjacent traces with low values indicating the presence of fractures,faults,and other incoherent features.Seismic coherency is able to show these features more clearly than common seismic amplitude data(e.g.Bahorich and Farmer,1995;Marfurt et al., 1998;Chopra,2002;Chopra and Marfurt,2006).The algorithm used in this paper is based on calculating the semblance,which is a form of seismic coherency.This is done by computing the cross-correlation between adjacent traces as a function of variable time lag(e.g.Marfurt et al.,1998).The method of calculating the semblance dates back to the work of Taner and Koehler(1969)who used it as a tool for conventional seismic velocity analysis.The advantage of this method next to obtaining clearer structural images is the absence of an interpretative bias as the stratigraphic analysis is done before picking faults.Additionally,the seismic attribute dip of maximum similarity,which gives the rate at which similarity changes,is very effective in depicting features such as channels and fractures where the similarity between traces decreases quickly.

The interpretation of the features found was guided by the results of Nissen et al.(1999),who used seismic coherency to interpret the depositional settings of a MTD offshore Nigeria,by Prior et al.(1982)on the Kitimat-slide offshore British Columbia, Canada,as well as by the research of Posamentier and Walker (2006).

In addition to the2-D and3-D seismic data,the interpretation was supported by LWD-derived resistivity,P-wave velocity, gamma-ray,and porosity at drill site UBGH1-4,as well as split-core images,and physical properties(density,shear strength)of the recovered core to de?ne the vertical extent of the MTD and related seismic signatures.

4.Seismic attributes

Several examples of seismic lines cutting through the3-D seismic data volume are analyzed for the illustration of the observed re?ection characteristics and sedimentary units(Fig.2

).

Figure2.Arbitrary lines taken from3-D seismic data volume(for location see Fig.3)(a)Line1is taken along the major MTD(Unit3a)and inpidual units are highlighted;(b)Line 2crossing the MTD Unit3parallel to Line1through the3-D volume,but containing only Sub-Unit3b,green dashed line indicating the bottom horizon of Units3a and3b;(c)Line3 cutting through Sub-Unit3b with indication of location of the assumed BSR.(For interpretation of the references to color in this?gure legend,the reader is referred to the web version of this article.)

N.A.Scholz et al./Marine and Petroleum Geology35(2012)91e10493

The seismic lines in Figure 2(see Fig.3for the location of the arbitrary lines)are chosen arbitrarily through the seismic volume but are overall in the direction of and perpendicular to the assumed ?ow-path of the main MTD.Based on the absence of distinct sediment layers as well as on the internal blanking and prominent top and bottom re ?ections in seismic amplitude data we identify ?ve separate units,henceforth called Units 1e 5:

Unit 1is comprised of the most recent MTD,identi ?ed as an seismically almost transparent layer right beneath the sea ?oor re ?ection.Based on a P-wave velocity of 1500m/s this unit has a thickness of 25m at Site UBGH1-4.Underneath,Unit 2is made up of about 30e 40m thick relatively uniform,sea ?oor-parallel re ?ections,comprises hemipelagic and pelagic sediments mixed with thin turbidites (Bahk et al.,2011).

The main focus of our study,Unit 3,is underlying the strati ?ed hemipelagic and pelagic section.Based mainly on differences in the internal re ?ection characteristics and in the amplitude of the top-re ?ection (Fig.3),Unit 3was pided into two Sub-units 3a and

3b.The arbitrary lines in Figure 2provide a good image of Sub-Unit 3a cutting through older deposits.Sub-Unit 3a shows almost no internal re ?ections (Fig.2a),while Sub-Unit 3b reveals internal,chaotic re ?ections of only a few tens of meters (Fig.2b,c).

Below Unit 3lies an approximately 70m thick stratigraphic sequence (labeled Unit 4),which is composed of a mix of hemi-pelagic sediments and a series of thinner MTDs.The base of Unit 4is dif ?cult to assess as it is near (or at)the depth of the BSR.In part,Unit 4resembles a similar seismic character as Unit 2.The sedi-ments below Unit 4show intervals of chaotic re ?ections caused by the presence of free gas below the GHSZ.We have not de ?ned deeper units as they are not essential for investigating the sedi-mentation history of the MTD (Unit 3a and b)and the presence of gas hydrates and/or free gas below the GHSZ.

Two prominent re ?ections framing the top and bottom of Unit 3as well as the present-day sea ?oor were manually picked to delineate this MTD and aid in subsequent volume-calculations.The picks of the different horizons were then used to image different aspects of the data such as time (depth),amplitude,similarity (coherence),and dip of maximum similarity.They are useful tools to better describe the sedimentation characteristics and the behavior of the ?ow.Figure 4shows the two-way travel-time for both the top and bottom re ?ection of the MTD (including the picks of the adjacent 2-D lines).The travel-time of the upper re ?ection increases steadily from the southwest to the northeast with a markedly larger increase in TWT in the north-eastern corner of the mapped horizon.The bottom horizon of the MTD has a similar depth increase which trends in the same direction (especially when including the 2-D seismic lines).It additionally reveals a broad-ening distinct graben-like incision,forming a fan-like topography which is only weakly observable in the upper re ?ection.Equally noteworthy is the presence of long,straight,and deep grooves.Figure 5shows the re ?ection amplitude,similarity,and dip of maximum similarity attributes for the MTD-base.Overall,the amplitude of the bottom re ?ection (Fig.5a)is quite homogeneous compared to the top-re ?ection (Fig.3).It additionally reveals stri-ations that are low in amplitude.However,the seismic similarity image (Fig.5b)shows a clearer image of these striations (darker colors represent low seismic similarity).The dip of maximum similarity (Fig.5c),an attribute representing the rate of change in similarity,increases the sharpness and thus reveals a larger number of these groove-like features compared to amplitude or similarity.The grooves have a distinctive pattern,mainly consisting of elon-gated closely spaced pairs of bands lower in re ?ection amplitude and similarity with high dip of maximum similarity values (Fig.6).The grooves cross-cut through the study area from the southwest to the northeast,then bend slightly,and ?nally develop into a fan-like run-out near drill site UBGH1-4.

In contrast to the area covered by Sub-unit 3a,regions in the south and south-east of the bottom re ?ection belonging to Sub-unit 3b have a rather chaotic distribution of features with overall low coherence values (area labeled Zone 1in Fig.6d).By means of the seismic coherence attribute,those strongly discontinuous features which can ’t be seen in Figure 3now become apparent,thereby revealing a de ?ection towards the east.This de ?ection is even stronger than the one seen in the striations of the bottom re ?ection within Sub-unit 3a.The change in the degree of coherence between the north-western and south-eastern parts is consistent with the change in the internal re ?ection characteristics as seen in Figure 2a e c.This shift in seismic character as well as coherence suggests a different deposition mechanism for the Sub-units 3a and 3b.Furthermore,grooves belonging to Sub-unit 3a strike at angles between 20 and 40 when measured clockwise from the north.The striations belonging to Unit 3b all show an angle of about 50 ,being much shorter in

length.

Figure 3.Map of the amplitudes contained in part of the 3-D data set belonging to the top of Unit 3(MTD)re ?ection event.This unit was subpided into two units:3a and 3b,using the distinct re ?ection amplitude characteristics.The extent of Sub-unit 3a is highlighted by the orange dashed line.Location of the three arbitrary lines in Figure 2a e c shown in yellow;locations of the in-line and cross-line containing Site UBGH1-4shown in light pink.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104

94

00da2ecbcc175527072208f6parison to the seismic characteristics of the sea ?oor To be able to reconstruct the depositional history of the MTD Unit 3and its features,we calculated the same seismic attributes for the sea ?oor (Fig.7),which itself is the top of the most recent MTD in the study area.The sea ?oor re ?ection shows lower absolute values in the rate of change in similarity compared to the bottom horizon of Unit 3and a stronger varying topography.A prominent saddle-like structure striking southwest-northeast is seen.The shallower southern portion of the 3-D data coverage has a much rougher character in sea ?oor topography than the struc-tural low to the north.Figure 7c and d are detailed views of sea ?oor-features that are located in the north-western part of the 3-D seismic data coverage.Three small sea ?oor elevations (labeled A,B,C,and D in Fig.7c)are associated with short grooves,similar to those seen at the base of the MTD (Sub-unit 3a)initi-ating in the northwest corner of the present sea ?oor.Figure 7d shows several larger blocks that seem to have been passed by other material,as suggested by grooves bending around them.The consistently high values in the dip of maximum similarity in Figure 7b e d running straight from north to south are due to acquisition foot-print and insuf ?cient statics correction.The sea ?oor blocks (labeled A,B,C,and D in Fig.7d)seem to have traveled past one another on their way down the slope stopping

approximately in line with each other.Also,these three blocks are linked to short,a few hundred meter long grooves.The features in Figure 7c (labeled 1,2,and 3),however,appear to have been blocking subsequent material that ?owed downslope,as seen by the grooves bending around them.This shows that the material of the MTD (Unit 1)did not ?ow as a single mass,but likely consists of several events and larger blocks of material that slid past each other.It is also worth noting that the outrunner blocks seen on top of the sea ?oor are in ?uenced by earlier topographic features (Fig.8).Elevated portions of the top of the MTD (Unit 3)have apparently in ?uenced subsequent deposition up to the current sea ?oor.Mini-basins have developed in the shadow of the older blocks but must have not been completely ?lled by the time of the most-recent MTD event,to still in ?uence modern sea ?oor topography.

5.Sedimentology and physical properties

The identi ?cation of the ?ve units within the upper-most 200m below sea ?oor was based on the seismic re ?ection character (as depicted in the arbitrary lines shown in Fig.2).At Site UBGH1-4,LWD-data as well as spot-coring with physical property measure-ments further allow to describe these units in terms of their sedi-mentology and physical

properties.

Figure 4.Two-way travel-time for the (a)top and (b)bottom re ?ection of the MTD including picks of the 3-D data volume as well as adjacent 2-D seismic lines.

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 10495

The top-Unit 1shows no internal seismic re ?ectivity.Photo-graphs of the top core acquired at this site reveal sediments of chaotic (blocky)nature as well as strong internal deformation typical for MTDs (Fig.9a e c).The sediments show clear signs of stratigraphic disturbance and reworked material.Two cores taken between 60and 70mbsf (UBGH1-4B-4H and UBGH1-4B-5H)show a clear transition from the hemipelagic sedimentation of Unit 2(Fig.9d)to the underlying MTD e Unit 3(Fig.9e).Cores taken through the base of Unit 3(UBGH1-4C-3H)show a transition from strongly reworked sediments (Fig.9f)to an underlying sequence of predominant low-energy deposition and hemipelagic sediments (Fig.9g)in core UBGH1-4C-4H.The deeper portion of Unit 4consists of a mix of thin,only few centimeter thick sandy turbidites (Fig.9h)and mud-dominated,un-deformed hemipelagic sediment as seen in core UBGH1-4C-6H.The thin sandy turbidites were gas hydrate bearing as identi ?ed from pore-water chlorinity freshening and infrared core imaging (Kim et al.,2011).

These observations can be tied to the log LWD-data (Fig.10).A time-depth conversion was obtained from tying the sea ?oor,and the top,and bottom of the MTD (Unit 3)to the seismic data,fol-lowed by integration of the velocity log to increase resolution of the time-depth conversion.Log properties obtained by LWD

(Gamma-

Figure 5.Bottom horizon of MTD,showing (a)amplitude,(b)similarity,and (c)dip of maximum similarity;green lines showing the locations of the arbitrary lines (Fig.2e c),orange dot shows the location of drill site UBGH1-4;(d)interpretation of the features found in 5a e 5c:dotted orange line outlines the base of Sub-unit 3a,blue-dotted line outlines two zones of Sub-unit 3b with different character in striations similar to the grooves of Sub-unit 3a,but with different orientation and length.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this

article.)

Figure 6.Zoom on two prominent grooves near the drill site UBGH1-4showing (a)amplitude,(b)similarity,and (c)dip of maximum similarity.Same color-bars and scales as in Figure 5are used.

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104

96

Figure 7.Seismic attributes of the sea ?oor:(a)depth as two-way travel-time (TWT),(b)dip of maximum similarity,(c)zoom of portion of sea ?oor with outrunner blocks,and (d)zoom of sea ?oor with distinct ?ow-pattern (grooves)around a series of outrunner blocks.The red line is inline 521shown in Figure 8.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this

article.)

Figure 8.(a)Seismic amplitude data of inline 521showing the outrunner blocks (ORBs)on the sea ?oor with the underlying compressional features as indicated by the green circles;(b)interpretation of the geometry in inline 521(drawing not to scale).(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 10497

ray,P-wave velocity,electrical resistivity,density,and neutron porosity)are compared in Figure 10:the P-wave velocity,electrical resistivity,and density show a marked increase over the interval of the MTD (Unit 3)whereas porosity and gamma-ray values decreased.The synthetic seismogram (Fig.11)calculated using a wavelet derived from the 3-D seismic data and utilizing the LWD logs of velocity and density show an overall good correlation to the acquired 3-D data,especially for the prominent re ?ections attrib-uted to the MTD (Unit 3)and base of gas hydrate stability.Changes in physical properties are gradational within the MTD (Unit 3),thus the synthetic seismogram reveals a lack of signi ?cant re ?ections.A tie of the LWD data to the seismic volume is shown in Figure 12for the inline and cross-line at Site UBGH1-4.

It is worth to point out two additional although much thinner zones similar to the main MTD (Unit 3)in their log-property trends.The ?rst interval is about 10m thick and occurs at a depth of 38e 48mbsf (2.544e 2.555s TWT).This interval is sit-uated within seismic Unit 2,but can ’t be fully resolved by the frequency range of the seismic data (dominantly 50Hz,equivalent to a spatial resolution of about 8m,using a log-derived velocity of 1525m/s and a resolution limit of 1/4of the wavelength).The second interval is only about 4m thick (148e 152mbsf,2.680e 2.685s TWT).It is located within seismic Unit 5and coincides with a faint seismic re ?ection (polarity reversed relative to the sea ?oor).The fact that this lower zone stands out seismi-cally may be related to the overall lack of other impedance contrasts in the immediate vicinity of this re ?ection,whereas the shallower MTD (though twice as thick)is within a package of more internal acoustic variability.

The sediment-cores collected at Site UBGH1-4allowed sedi-mentological descriptions and detailed physical property measurements (Fig.13).Porosity and bulk density measured on the top three cores in hole UBGH1-4B as well as the bottom core in Hole UBGH1-4C (UBGH1-4C-6H)match overall fairly well with the LWD 00da2ecbcc175527072208f6rger discrepancies exist around the top and base of the MTD

(Unit 3),which may be the result of the complex reworked nature of the sediments,and limited useful zones on the core for physical property measurements due to post-recovery drying and disrup-tion of the sediments.However,marked differences in shear strength,sand-content,and grain density occur around the top and base of Unit 3.The sediments from core UBGH1-4C-3H situated within the MTD (Unit 3)have higher sand and lower silt content in contrast to sediments from core UBGH1-4C-4H situated below the MTD (Unit 3).These two cores also show a drop in grain density and a marked increase in shear strength across the lower boundary of the MTD (Unit 3).

6.Physical dimensions of Unit 3

Top and bottom of Unit 3are used to estimate the extent,thickness,and volume of the MTD.Where possible we followed the MTD from inside the 3-D coverage to the overlapping and adjacent 2-D seismic lines to assess its physical 00da2ecbcc175527072208f6ing the picked horizons on both,the 3-D volume and 2-D seismic lines,the approximate surface area of the MTD was de ?ned (Fig.14a).The estimated spatial extent of the mass transport deposit amounts to about 210km 2.The thickness of Unit 3was calculated using velocity from the LWD data and small regions where picking of the horizons was impossible were interpolated using a gridding-algorithm (Fig.14b).

With an estimated MTD-volume of 1.28?1010m 3and using the empirical relationship by De Blasio et al.(2005)between volume V and ?ow-speed c of the form c ?V 1/6,the MTD may have moved downslope with a speed of about 50m/s.Where the slope angle of the paleo-sea ?oor decreased in the main basin ?oor the MTD ?ow started to spread out laterally.The total area covered by the MTD (and thus total volume)cannot be completely calculated as the zone where its thickness is below seismic detection limits cannot be determined.Thus the inferred velocity is probably an approxi-mate lower

bound.

Figure 9.Examples for core images from Site UBGH1-4:(a)Core UBGH1-4B-1H,sections 5,(b)Core UBGH1-4B-2H section 2,(c)and Core UBGH1-4B-2H section 4showing reworked nature of the top-MTD of Unit-1;(d)Core UBGH1-4B-4H section 1of hemipelagic sediment in Unit 2;(e)core UBGH1-4B-5H,section 6from the top of MTD Unit 3;(f)core UBGH1-4C-3H section 1;(g)core UBGH1-4C-4H section 2with hemipelagic layered mud;(h)Core UBGH1-4C-6H,section 1with interbedded thin sandy turbidites and hemipelagic mud;red ellipses show parts with reworked,disturbed sediment used to identify this section as a MTD.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104

98

Inferring the topography of the paleo-surface and ?ow-direction of the MTD from the seismic attributes shown in Figure 5,we projected the in ?ll path to a possible source region,which seems to lie somewhere along the southern basin rim about 50km to the south of the 3D data set at current water depths of around 1000m (Fig.15).The MTD was likely sliding downslope into the northeast direction towards drill-site UBGH1-4,which is slightly rotated further east from inferred modern tracking-paths based on current sea ?oor topography.The MTD top-re ?ection displays a rather rugged surface topography (e.g.Figs.2c,and 7b),which could be due to buttressing mechanisms causing sediment accumulation on top of the ?ow-body as described by Garziglia et al.(2008).The compressional features observed in our data set show similarities to the pressure ridges in Nissen et al.(1999),but are different from those described by Prior et al.(1982),who showed a feature much larger in scale having a terrace-like 00da2ecbcc175527072208f6pared to the descriptions of Prior et al.(1982),the elevations seen in our data are

con ?ned to a smaller area and the undulating character of the pressure ridges is missing.

Arbitrary line 1(Fig.2a)is also used to determine the dip-angle of the current sea ?oor and the base of the MTD.Assuming a water velocity of 1485m/s,the sea ?oor dips gently with an angle of w 0.38 .The paleo-sea ?oor representing the gliding plane of the MTD is dipping at a slope as low as 0.42 e 0.45 and steepens only slightly to a value 2.2 towards the south-western corner of the 3-D data coverage area.The range in dip-values is obtained by assuming sediment interval-velocities varying between 1500m/s and 1600m/s.7.Sliding processes

The several broad scars (or grooves),which the MTD (Unit 3a)left along the paleo-surface reach up to w 10m in depth.Judging from the pattern of scars as seen in Figures 5and 6,the MTD (Unit 3a)seems to have taken a path leading from southwest

to

Figure 10.Logging-While-Drilling (LWD)results for drill site UBGH1-4;from left to right:Depth (in meter below rig ?oor (mbrf),meter below sea ?oor (mbsf)and two-way time (TWT)in seconds,conversion control points after velocity integration are shown as red dots),Gamma-ray,P-wave velocity,electrical resistivity (ring),bulk density,and neutron porosity;gray box highlights the location of Unit 3a.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)

N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 10499

northeast,thereby widening slightly towards the north-east to

form a fan-like deposition.This widening probably occurred due to

the particular topography of the paleo-surface (gliding plane)and

may also be associated with a reduction in speed of the ?owing

sediment 00da2ecbcc175527072208f6ing the dip of maximum similarity attribute it

becomes evident that the grooves leading from southwest to

northeast have lengths of several kilometers amounting up to more than 10km.Those tracks have edges with a very large in dip of maximum similarity suggesting steeply inclined sidewalls.7.1.Bottom scars There are two possible explanations for the development of the bottom 00da2ecbcc175527072208f6paring Figure 6b and c with the

features Figure 11.Synthetic seismogram using the Logging-While-Drilling (LWD)P-wave velocity and bulk density logs at Site UBGH1-4and a seismic wavelet extracted from the 3-D seismic data volume.Top,and base re ?ection of the MTD units correspond well to the real data (orange arrows).Also,three smaller-MTDs identi ?ed in the log do show some correlation to the real data (green arrows).(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this

article.)Figure 12.(a)Seismic line 717with resistivity-log from borehole UBGH1-4shown in blue (b)cross-line number 955with resistivity-log;note the grooves at the bottom of Unit 3a.(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104

100

described by Nissen et al.(1999),the incisions seen in the bottom

re ?ection of the MTD can either be interpreted as gliding tracks of

outrunner blocks that separated from the main body of the MTD.

Alternatively,they can also be seen as scars caused by sediment

clasts that were dragged along with the main MTD body.

a)Hypothesis 1:Outrunner blocks

In the case that the grooves were caused by material that outran

the main body,the scars were caused by blocks that would have cut

into substrate and left deep,narrow tracks on the paleo-surface.

We

Figure 13.Physical properties of the recovered core (orange squares),compared to LWD-results (blue lines).From left to right:Core recovery in Hole UBGH1-4B and UBGH1-4C (blue boxes indicate intended length of core,black adjacent boxes show actually achieved recovery),porosity,bulk density,gamma-ray,shear strength,grain density,sand-content.

[Core-naming convention:H:hydraulic piston core,R:rotary pressure core,P:percussion pressure core,T:temperature tool,D:drilled interval;cores annotated by asterisk are shown in Fig.9].(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this

article.)

Figure 14.(a)Spatial extent and (b)thickness of Unit 3;the spatial extent was esti-

mated by including picks of the 3-D volume and adjacent 2-D

lines.Figure 15.Interpretation of possible ?ow-direction of recent MTDs (black arrows,perpendicular to bathymetric lines)and mapped paleo-MTD (red arrows)with possible source region (red dotted zone).(For interpretation of the references to color in this ?gure legend,the reader is referred to the web version of this article.)N.A.Scholz et al./Marine and Petroleum Geology 35(2012)91e 104101

suggest that the tracks are either part of an older?ow-event and have then been?lled in by the MTD that makes up Sub-Unit3a.Or alternatively,they could have been formed by parts of this MTD itself,outrunning the main body which subsequently?lled the fresh gliding tracks.The decrease in shear strength and bulk density,which was observed beneath the bottom of the MTD Unit3 (Fig.13)supports the latter possibility and could be an explanation for the observation that although the gliding tracks just had formed,they weren’t eroded by the sliding material that followed. The length of the gliding tracks as seen in Figures5and6,which constitutes the major contribution to the inferred extent of the MTD,leads to the assumption that the outrunner blocks traveled at least10km.The grooves are exceeding the area of the3-D seismic data set and there are no inpidual blocks seen in the re?ection of the bottom horizon.According to Nissen et al.(1999)and Prior et al. (1982),outrunner blocks can be formed by an early spill-over of the MTD material,which separates from the main body due to tensional forces.The reason for this complete separation between blocks and main body can be attributed to hydroplaning.Hydro-planing is thought to occur when the front of a MTD pushes through ambient water giving rise to hydrodynamic pressures, which then deform the frontal part of the?ow,thus enabling the penetration of a wedge-shaped layer of water underneath the debris.This layer reduces the basal friction and induces tensile stresses farther behind the front,causing a possible detachment and decoupling with respect to the main slide body.Matching with our observations,distances between the outrunner blocks and the main MTD have been reported to amount to several kilometers in many cases(e.g.Mohrig et al.,1998;Ilstad et al.,2004;De Blasio et al.,2005).

De Blasio et al.(2005),who studied the dynamics of subaqueous gravity?ows of the Storegga Slide,considered several possible reasons for such long run-out distances.In addition to hydro-planing as a hypothesis,they also see the progressive reduction in yield strength under a high shear rate as a possible contributing factor to long run-out distances.This is a consequence of a collapsing soil structure and the accompanying increase in pore pressure.The decrease in yield strength may be caused by the mixing with seawater and/or water-rich hemipelagic sediments.De Blasio et al.(2005)also consider the increased mobility of the outrunner blocks in respect to the main body as a result of a combination of the two mechanisms,which they called‘shear wetting’.According to their results,the?ow-velocity of MTDs is largely the consequence of an equilibrium between the component of the gravity force parallel to the ground-surface and the drag force caused by water.Another contributing factor is the internal friction at the base of the MTD body.De Blasio et al.(2005)state that the maximum?ow-velocity for hydroplaning scales with V1/6,where V is the volume of the MTD and that the?ow-velocity of the front is the highest velocity throughout the?ow-unit.

b)Hypothesis2:Embedded sediment clasts

Alternatively,the scars or striations could be interpreted as erosional features formed by embedded clasts at the base of the MTD(e.g.Posamentier and Walker,2006).Such embedded clasts within a MTD would be seismically visible,likely in form of chaotic re?ections,or short re?ector-elements,such as seen within Sub-Unit3b.Iverson(1997)states that surge heads carry the greatest concentration of large sediment clasts within a MTD.He describes the head as relatively dry and as restraining the subsequent more ?uid,water-saturated debris.There are two mechanisms by which large clasts accumulate at the head:Either they are acquired in transit of the?ow and subsequently retained,or they migrate through the?ow-body to the head.The mechanisms can be attributed to kinetic sieving,selective entrainment or because gravity and boundary drag do not suf?ce to force the clasts through the small voids,which repeatedly open and close during defor-mation of the MTD,leaving them as a kind of residue near the front. The grain size segregation mechanisms within MTDs are compli-cated and probably include more than one process.

8.Discussion

Seismic coherency such as similarity and dip of maximum similarity were used as effective tools to map MTDs in the southern part of the Ulleung Basin,East Sea,offshore Korea.Generally, features of a MTD can be discerned from the seismic data by means of their re?ection and attribute characteristics.Guided by results of the study of Nissen et al.(1999),we also calculated seismic attri-butes coherency(similarity)and dip of maximum similarity and retrieved a clearer image of the components of the main MTD(Unit 3)and sea?oor(top of MTD Unit1).The interior body of the MTD Unit3has a fairly homogenous seismic character(i.e.acoustic blanking),previously proposed as indicator for the presence of gas hydrate especially when occurring in conjunction with elevated P-wave velocities.The log-and core-derived physical properties show an elevated electrical resistivity and P-wave velocity,but no gas hydrate was recovered from this unit.The increase in resistivity and P-wave velocity is the result of coinciding reduced porosity and increased bulk-density.The MTD is framed by a strong upper and lower seismic re?ection,suggesting large contrasts in physical properties between the MTD and the surrounding sediments.The base of the MTD shows a polarity reversal relative to the sea?oor re?ection(reduction in impedance)identical to the BSR,commonly interpreted as the base of the GHSZ.

The Sub-Unit3a is characterized by the absence of internal re?ectivity and the base of the unit shows long grooves exceeding the dimension of the3-D seismic data coverage,without visible out-runner blocks.In contrast,the base of Sub-Unit3b shows short-lived striations and the body of this Sub-Unit is characterized by short-lived,chaotic re?ection elements.Therefore,we propose that the striations of sub-Unit3a are left by outrunner blocks,which traveled ahead of the main MTD body(likely as a result of hydroplaning)and that striations of Sub-unit3b are the result of embedded clasts.

Using the seismic character of sub-Units3a and3b one can propose two alternate scenarios for deposition of the MTD Unit3: One option is that sediments of Sub-Unit3b were less?uidized and more consolidated than those of Sub-Unit3a,which resulted in separate depositional characters:deeper and longer striations associated with Sub-Unit3a,and shorter striations with Sub-Unit 3b.In this scenario,both sub-Units were deposited at the same time.Alternatively,Sub-Unit3a could have been deposited later than Sub-Unit3b,and thus may have cut through the deposits of the older MTD Sub-Unit3b,eroding and incorporating substrate. Sub-Unit3a was more?uidized allowing longer grooves to develop than what is seen at the base of Unit3b(that was not eroded).

With a previously established empirical relation between volume and?ow-velocity,we inferred a velocity of at least50m/s for the main MTD body despite the shallow basin slope angle. Velocities of this magnitude have been reported elsewhere such as in the case of the Storegga(De Blasio et al.,2005)and the Currituck slides(Locat et al.,2009)but our result rather poses an upper limit of the possible?ow-speed.Additionally,run-out distances are found to amount to at 00da2ecbcc175527072208f6rge traveling distances in low slope angle environments have been studied before(e.g.Hampton and Locat,1996;Mohrig et al.,1999;Locat and Lee,2002).As an example,the Amazon and the Mississippi Fan both reveal equally long run-out distances despite a smoothly dipping sea?oor topography(e.g.Hampton and Locat,1996;Maslin et al.,2005).

N.A.Scholz et al./Marine and Petroleum Geology35(2012)91e104 102

9.Conclusions

Seismic coherency has proven to be an effective tool for the study of?ow behavior of MTDs.The dip of maximum similarity is especially capable of enhancing the contrast of the MTD and sea?oor traits.Gliding tracks are depicted as two narrow,parallel lines thus making it possible to discern them from other features such as faults which would be visible as only one line of disconti-nuity in the seismic coherency attribute.

With the help of seismic coherency we were able to infer the physical dimensions of the MTD(Unit3),its maximum velocity,and the extent of its run-out.The entire MTD(Unit3)is assumed to have followed a slightly slanted path to the northeast cutting through an older MTD complex.The estimated spatial extent and approximate volume of the MTD derived from the3-D and adjacent 2-D seismic data amounts to210km2and1.28?1010m3,respec-tively.The possible source-region lies w50km upslope from Site UBGH1-4as inferred by following the MTD top and bottom re?ections from the3-D seismic data coverage to adjacent2-D seismic lines further south of the3D data volume.The main body could have traveled with a velocity as high as50m/s and its out-runner blocks have covered a distance of at least10km.

Since the3D data does not capture the total extent which the outrunner blocks traveled and since mass failure and the occurrence of outrunner blocks still are ongoing phenomena in the Ulleung Basin,further studies of slope failure and outrunner blocks are necessary.Long distance run-out,high?ow velocities,and the large, fast traveling outrunner blocks all present risk factors for offshore infrastructure and the impact of a collision with a production facility could lead to signi?cant damage that would pose a threat to human life and could cause possible widespread pollution. Acknowledgments

We would like to thank all scientists involved in the?rst scienti?c drilling expedition in the Ulleung Basin in2007as well as all crew and technical staff onboard the vessel Rem-Etive.We also want to thank KIGAM for allowing data distribution,especially the 3-D seismic data volume at drill site UBGH1-4.This is Earth-Science-Sector(ESS)contribution number:20100042. References

Bahk,J.J.,Han,S.-J.,Khim,B.-K.,2004.Variations of terrigenous sediment supply to the southern slope of the Ulleung Basin,East/Japan Sea since the Last Glacial Maximum.Geosciences Journal8(4),381e390.

Bahk,J.-J.,Kim, D.-H.,Chun,J.-H.,Son, B.K.,Kim,J.-H.,Ryu, B.-J.,Torres,M., Schultheiss,P.,Collett,T.,Riedel,M.,2011.Gas hydrate occurrences and their relation to hosting sediment properties:results from UBGH2,East Sea.In: Proceedings of the7th International Conference on Gas Hydrates(ICGH2011). Bahk,J.-J.,Um,I.-K.,Holland,M.,,2011.Core lithologies and their constraints on gas-hydrate occurrence in the East Sea,offshore Korea:results from the site UBGH1-9.Journal of Marine and Petroleum Geology28(10),1943e1952. Bahorich,M.,Farmer,S.,1995.The coherence cube.The Leading Edge14(10), 1053e1058.

Bouriak,S.,Vanneste,M.,Saoutkine,A.,2000.Inferred gas hydrates and clay diapirs near the Storegga Slide on the southern edge of the V?ring Plateau,offshore Norway.Marine Geology163,125e148.

Bünz,S.,Mienert,J.,Vanneste,M.,Andreassen,K.,2005.Gas hydrates at the Storegga Slide:constraints from an analysis of multicomponent,wide-angle seismic data.Geophysics70(5),B19e B34.doi:10.1190/1.2073887.

Chopra,S.,2002.Coherence Cube and beyond.First Break20,27e33.

Chopra,S.,Marfurt,K.,2006.Seismic Attributes e A Promising Aid for Geologic Interpretation.CSEG Recorder,Special Edition.

Chough,S.K.,Barg,E.,1987.Tectonic history of the Ulleung Basin Margin,East Sea (Japan Sea).Geology15,45e48.

Chun,J.H.,Kwon,Y.I.,Lee,H.Y.,Kim,H.J.,Kim,J.H.,Park,Y.S.,Jeon,C.W.,Lee,J.W., An,G.O.,Son,B.K.,Ko,J.H.,Ryu,B.J.,Sunwoo,D.,Oh,J.H.,Lee,Y.J.,Cheong,T.J., Lee,J.Y.,Kang,N.K.,2008.Shallow gas hydrate in deep sea sediments in the Ulleung back-arc basin,offshore Korea.In:6th International Conference on Gas Hydrates(ICGH2008),Vancouver,Canada,pp.86.Clennell,M.B.,Hovland,M.,Booth,J.S.,Henry,P.,Winters,W.J.,1999.Formation of natural gas hydrates in marine sediments1.Conceptual model of gas hydrate growth conditioned by host sediment properties.Journal of Geophysical Research104,22,985e23,003.

De Blasio,F.V.,Elverh?i,A.,Issler,D.,Harbitz,C.B.,Bryn,P.,Lien,R.,2005.On the dynamics of subaqueous clay rich gravity mass?ows e the giant Storegga slide.

Norway,Marine and Petroleum Geology22,179e186.

Garziglia,S.,Migeon,S.,Ducassou,E.,Loncke,L.,Mascle,J.,2008.Mass-transport deposits on the Rosetta province(NW Nile deep-sea turbidite system,Egyptian margin):characteristics,distribution,and potential causal processes.Marine Geology250,180e198.

Gauer,P.,Kvalstad,T.J.,Forsberg,C.F.,Bryn,P.,Berg,K.,2005.The last phase of the Storegga Slide:simulation of retrogressive slide dynamics and comparison with slide-scar morphology.Journal of Marine and Petroleum Geology22, 171e178.

Hampton,M.A.,Locat,J.,1996.Submarine landslides.Review of Geophysics34(1), 33e59.

Ilstad,T.,De Blasio,F.V.,Elverhoi,A.,Harbitz,C.B.,Engvik,L.,Longva,O.,Marr,J.G., 2004.On the frontal dynamics and morphology of submarine debris?ows.

Marine Geology213,481e497.

Iverson,R.M.,1997.The physics of debris?ows.Reviews of Geophysics35,245e296. Kim,G.Y.,Yi,B.Y.,Yoo,D.G.,Ryu,B.J.,Riedel,M.,2011.Evidence of gas hydrate from downhole logging data in the Ulleung Basin,East Sea.Journal of Marine and Petroleum Geology28(10),1979e1985.

Kvalstad,T.J.,Andresen,L.,Forsberg,C.F.,Berg,K.,Bryn,P.,Wangen,M.,2005.The Storegga Slide:evaluation of triggering sources and slide mechanics.Marine and Petroleum Geology122(7),245e256.

Lee,G.H.,Suk,B.-C.,00da2ecbcc175527072208f6test Neogene-Quaternary seismic stratigraphy of the Ulleung Basin,East Sea(Sea of Japan).Marine Geology146,205e224.

Lee,G.H.,Kim,J.H.,Han,S.J.,Kim,D.C.,2001.Seismic stratigraphy of the deep Ulleung Basin in the East Sea(Japan Sea)back-arc basin.Marine and Petroleum Geology18,615e634.

Lee,G.H.,Kim,B.,2002.In?ll history of the Ulleung Basin,East Sea(Japan Sea)and implications on source rocks and hydrocarbons.Marine and Petroleum Geology 19,829e845.

Lee,H.J.,Chough,S.K.,Yoon,S.H.,1996.Slope-stability change from late Pleistocene to Holocene in the Ulleung Basin,East Sea(Japan Sea).Sedimentary Geology 104,39e51.

Lee,J.H.,Baek,Y.S.,Ryu,B.J.,Riedel,M.,Hyndman,R.D.,2005.A seismic survey to detect natural gas hydrate in the East Sea of Korea.Marine Geophysical Research26,51e59.

Lee,M.W.,Dillon,W.P.,2001.Amplitude blanking related to gas hydrate concen-tration.Marine Geophysical Researches22(2),101e109.

Locat,J.,Lee,H.J.,2002.Submarine landslides:advances and challenges.Canadian Geotechnical Journal39,193e212.

Locat,J.,Lee,H.,ten Brink,U.S.,Twitchell, D.,Geist, E.,Sansoucy,M.,2009.

Geomorphology stability and mobility of the Currituck slide.Marine Geology 264,28e40.

Marfurt,K.J.,Kirlin,R.L.,Farmer,S.L.,Bahorich,M.S.,1998.3-D seismic attributes using a semblance based coherency algorithm.Geophysics63(4),1150e1165. Maslin,M.,Vilela,C.,Mikkelsen,N.,Grootes,P.,2005.Causes of catastrophic sedi-ment failures of the Amazon Fan.Quaternary Science Reviews24,2180e2193. Mienert,J.,Posewang,J.,Baumann,M.,1998.Gas hydrates along the northeastern Atlantic Margin:possible hydrate-bound margin instabilities and possible release of methane.In:Henriet,J.P.,Mienert,J.(Eds.),Gas Hydrates:Relevance to World Margin Stability and Climate Change.Geological Society of London Special Publication,vol.137,pp.275e291.

Mienert,J.,Vanneste,M.,Buenz,S.,Andreassen,K.,Ha?idason,H.,Sejrup,H.P.,2005.

Ocean warming and gas hydrate stability on the mid-Norwegian margin at the Storegga Slide.Marine and Petroleum Geology22(1e2),233e244.

Mohrig,D.,Ellis,C.,Parker,G.,Whipple,K.X.,Hondzo,M.,1998.Hydroplaning of subaqueous debris?ows.Geological Society of America Bulletin110,387e394. Mohrig,D.,Elverh?i,A.,Parker,G.,1999.Experiments on the relative mobility of muddy subaqueous and subaerial debris?ows and their capacity to remobilize antecedent deposits.Marine Geology154,117e129.

Nissen,S.E.,Haskell,N.L.,Steiner,C.T.,Coterill,K.L.,1999.Debris?ow outrunner blocks,glide tracks,pressure ridges identi?ed on the Nigerian continental slope using3-D seismic coherency.The Leading Edge,595e599.

Park,C.H.,2007.Hydrography of the East Sea.In:Lee,K.-S.,Kim,W.-S.(Eds.),Ocean Atlas of Korea e East Sea.National Oceanographic Research Institute,Incheon, pp.66e75.

Park,K.P.,2008.Gas hydrate exploration activities in Korea.In:Proceedings of the6th International Conference on Gas Hydrates,ICGH2008,Vancouver2008,pp.7. Park,K.-P.,Bahk,J.-J.,Kwon,Y.,Kim,G.-Y.,Riedel,M.,Holland,M.,Schultheiss,P., Rose,K.,the UBGH-1Scienti?c Party,2008.Korean National Program Expedition Con?rm Rich Gas Hydrate Deposits in the Ulleung Basin,East Sea.DOE-National Energy Technology Laboratory(NETL)Fire in the Ice Methane Hydrate News-letter,6e9pp.00da2ecbcc175527072208f6/technologies/oil-gas/publications/ Hydrates/Newsletter/HMNewsSpring08.pdf(last accessed02.09.11.). Posamentier,H.W.,Walker,R.G.,2006.Deep-water turbidites and submarine fans.

In:Facies Models Revisited.SEPM Spec.Publ.,vol.84,pp.397e520.

Prior, D.B.,Bornhold, B.C.,Coleman,J.M.,Bryant,W.R.,1982.Morphology of

a submarine slide,Kitimat arm,British Columbia.Geology10,588e592. Riedel,M.,Hyndman,R.D.,Spence,G.D.,Chapman,N.R.,2002.Seismic investiga-tions of a vent?eld associated with gas hydrates,offshore Vancouver Island.

N.A.Scholz et al./Marine and Petroleum Geology35(2012)91e104103

Journal of Geophysical Research,JGR Solid Earth107(B9),2200.doi:10.1029/ 2001.

Ryu,B.-J.,Riedel,M.,Kim,J.-H.,Hyndman,R.D.,Lee,Y.-J.,Chung,B.-H.,Kim,I.-S., 2009.Gas hydrates in the western deep-water Ulleung Basin,East Sea of Korea.

Journal of Marine and Petroleum Geology26,1483e1498.

Taner,M.T.,Koehler,F.,1969.Velocity spectra e digital computer derivation and applications of velocity functions.Geophysics34,859e881.Torres,M.E.,Tréhu,A.M.,Cespedes,N.,Kastner,M.,Wortmann,U.G.,Kim,J.-H., Long,P.,Malinverno,A.,Pohlman,J.W.,Riedel,M.,Collett,T.,2008.Methane hydrate formation in turbidite sediments of northern Cascadia,IODP Expedition 311.Earth and Planetary Science Letters271,170e180.

Vogt,P.R.,Jung,W.Y.,2002.Holocene mass wasting on upper non-polar continental slopes d due to post-glacial ocean warming and hydrate dissociation?

Geophysical Research Letters29.doi:10.1029/2001GL013488.

N.A.Scholz et al./Marine and Petroleum Geology35(2012)91e104 104

本文来源：https://www.bwwdw.com/article/teol.html