bsrs and associated reflections as an indicator of gas hydrate and free gas accumulation: an example...

1. Introduction

Bottom simulating reflections (BSRs) are widely rec-ognized in seismic profiles around the Japanese island-arcsystem and their total distribution area is 44,000 km2.This BSR distribution is highly heterogeneous and 75 %of them are recognized at well-developed present accre-tionary prism (the Nankai accretionary prism) and forearcbasins (the Enshu Trough, the Kumano Basin, the MurotoBasin, the Tosa Basin and the Hyuga Basin) along theNankai Trough (Baba and Yamada, 2000). The Nankaiaccretionary prism has developed as a large accretionarywedge since the Miocene (e.g. Shipboard Scientific Party,1991) due to the NNW-directed subduction of thePhilippine Sea Plate (Seno and Maruyama, 1984). Mostaccreted sediments are coarse terrestrial clastics whichare derived from the Japan island-arc system. The seg-ment of the Philippine Sea Plate subducting at the Nankaiis called ‘Shikoku Basin’, a former back-arc basin of theIzu-Bonin arc, thus the heat flow at Nankai is exception-ally high (Yamano et al., 2003). The forearc basins havebeen formed between the older accreted belt of theJapanese island-arc and the Nankai accretionary prism,

and subsequently filled with sediments of submarine fansystems (Fig. 1).

It is commonly believed that BSRs are resulted froma change in P-wave velocity; from high-velocity sedi-ments containing gas hydrates to lower velocity sedi-ments containing a small amount of free gas (e.g. Yuanet al., 1999; Ecker et al., 1998). Therefore, BSRs areregarded as an indicator of hydrate-bearing sedimentsthat can be a seal to the free gas underneath.

Two models for the gas hydrate accumulation havebeen proposed. In the first, gas hydrates are generated in-situ from organic carbon contained in the sediments of agas hydrate zone (Claypool and Kaplan, 1974) whereas inthe second, gas hydrates are derived from upward migrat-ing fluids containing light hydrocarbons (Hyndman andDavis, 1992). In addition to these models, the gas hydratesare re-concentrated by “hydrate recycling” (e.g. Paull etal., 1994; von Heune and Pecher, 1999) in which gashydrates form again from upflow of dissociation gasderived from pre-existing gas hydrates when the base ofthe gas hydrate stability field (BGHS) moves upward bytectonic uplift or with burial.

In this study, we try to elucidate an accumulationprocess of gas hydrates in the Nankai accretionary prism

RESOURCE GEOLOGY, vol. 54, no. 1, 11–24, 2004

11

BSRs and Associated Reflections as an Indicator of Gas Hydrate and Free Gas Accumulation: An Example of Accretionary Prism and

Forearc Basin System along the Nankai Trough, off Central Japan

Kei BABA and Yasuhiro YAMADA*

JAPEX Research Center, Japan Petroleum Exploration, Co. Ltd., 1-2-1 Hamada, Mihama-ku, Chiba 261-0025, Japan[e-mail: [email protected]]

* Department of Civil and Earth Resources Engineering, Graduate School of Engineering, Kyoto University,Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501, JapanReceived on September 19, 2003; accepted on February 25, 2004

Abstract: Multi-channel seismic data obtained from the Nankai accretionary prism and forearc basin system has been studied toelucidate the migration and accumulation process of gas to the BGHS and examine the distribution pattern of BSRs and charac-teristic reflections associated with them.

BSRs are distributed widely in the Nankai accretionary prism and associated forearc basins (33,000 km2) and 90 % ofthem have migration and recycling origins. The widest distribution of the BSRs can be seen at the prism. A correlation betweenthe BSR distributions and prism size shows that the BSRs tend to be more well-developed in a prism of large size. This suggeststhat a large prism may produce much amount of gas-bearing fluids that migrate to the BGHS and form the BSRs (tectonic con-trol). In the forearc basins, the BSRs are identified at topographic highs, anticlines and basin margins (structural control).

The upward migration of gas-bearing fluids is carried out through permeable sand layers and as a result, the distribution ofBSRs is confined to alternating beds of sand and mud facies (sedimentary control). However, if there is enough time for upwardmigration and accumulation of gas to the BGHS, the BSRs can be generated widely in low-permeable mud facies (time control).

Those results imply that structural, tectonic, sedimentary and time controls are primary factors to decide the distribution ofBSRs in the Nankai Trough area.

Keywords: BSR, gas hydrate, methane hydrate, fluid migration, seismic, accretionary prism, forearc basin, Nankai Trough

and forearc basin system by examining the distributionpattern of BSRs and characteristic reflections associatedwith them. The 2D multi-channel seismic data we usedare all of the seismic records obtained by Japan PetroleumExploration, Co., Ltd. and Japan National Oil Co., includ-ing the newest 2D high-resolution survey. The sedimenta-ry facies we used was obtained from seismic facies analy-sis in which each seismic facies was calibrated by geolog-ic data of the MITI (Ministry of International Trade andIndustry; currently METI; Ministry of Economy, Tradeand Industry) Exploratory test well “Nankai Trough” (theMITI Nankai Trough wells).

2. Distribution Pattern of BSRs

2.1. Classification of BSR

The BSRs distributed around the Japanese island-arcsystems are classified into the following four types basedon the distribution configuration of BSRs and accumula-tion process of gas hydrates (Baba and Uchida, 1999).

(1) Ridge type BSR is generated at a topographic highwhich has been uplifted with formation of anticlinal struc-ture, and is characterized by a convex-shaped reflection(Figs. 2 and 3a). It is supposed that this type BSR isformed by: i) upward migration and accumulation of gastoward the anticlinal structure, and ii) hydrate recyclingresulting from upward movement of the BGHS caused by

tectonic uplift with formation of the structure.(2) Buried anticline type BSR is generated at an anticli-

nal structure which is completely buried by younger sedi-ments after the end of the tectonic movement, and is char-acterized by planar-shaped reflections parallel to theseafloor (Figs. 2 and 3b). It is supposed that this type ofBSR is formed by: i) upward migration and accumulationof gas toward the anticlinal structure, and ii) hydrate recy-cling resulting from upward movement of the BGHS withburial. During the formation of the anticlinal structure,hydrates might be accumulated by the same process as theridge type BSR.

(3) Basin margin type BSR is generated at a marginalpart of a sedimentary basin, and is characterized by aplanar-shaped BSR which cuts stratigraphic reflectionsdipping gently toward the basin center (Figs. 2 and 3c).It is supposed that this type BSR is formed by: i)upward migration and accumulation of gas toward thebasin margin passing through permeable layers (e.g.sand layer), and ii) hydrate recycling resulting fromupward movement of the BGHS with burial.

(4) Accretionary prism type BSR generates at accre-tionary prism, and is characterized by a distribution pat-tern which is rarely controlled by either geologic structureor topography (Figs. 2 and 3d). Such a distribution patterncan be considered to be common in an accretionary prism,because strong and diffusive upward fluid flow occurs as

K. BABA and Y. YAMADA12 RESOURCE GEOLOGY :

0

11000

mm

0

34°N

33°N

32°N

31°N

30°N

35°N

131°E 132°E 133°E 134°E

135°E 139°E137°E 138°E136°E

Shikoku

Tosa Basin

Muroto Basin

Kyushu

Hyuga Basin

Shikoku Basin

Forearc basin

Accretionary prism

Deformation front

Enshu Trough

Kumano Basin

N a n k a i

T r o u g h

Nankai Accretionary Prism

Honshu

4000m

2000m

3000m

1000

m

3000m

4000

m

0 100km

2000

m

1000

m

Fig. 1 Location map of the study area.

vol. 54, no. 1, 2004 BSRs in the Nankai Accretionary Prism and Forearc Basins 13

Fig. 3 Seismic examples of BSR types. a: Ridge type BSR from southern flank of the Shima Spur; b: Buried anticline type BSRfrom southwestern margin of the Kumano basin; c: Basin margin type BSR from southwestern margin of the Kumano basin.Dashed line indicates base of the basin; d: Accretionary prism type BSR from off Tosa basin. Dashed line indicates thrust fault.

Ridge Type Buried Anticline Type

Basin Margin Type Accretionary Prism Type

Fig. 2 Schematic illustration for BSR of ridge type, buried anticline type, basin margin type and accretionary prism type.Black arrows indicate the migration of gas-bearing fluids.

a result of tectonic thickening and loading in the accre-tionary prism (Davis et al., 1990; Hyndman and Davis,1992). In addition, hydrate recycling may be acceleratedby continuous uplift of the accretionary prism caused bylong-lived accretion of underplate sediments.

2.2. Distribution and origin of BSR

On the basis of the classification, the distribution ofBSRs is investigated in the Nankai accretionary prismand forearc basins (Figs. 4-1 and 4-2). In the figure, the


Fig. 4-1 Distribution map of four BSR types.

ridge type BSR is mainly recognized at fold highs andseamounts in the forearc basins, the fold highs formed byfold fault systems in the prism and trench-slope break.The buried anticline type BSR is mainly found at anticli-nal structures formed in forearc basins. The basin margintype BSR develops typically at marginal part of theKumano Basin, the Muroto Basin, and the Tosa Basin.The accretionary prism type BSR which shows the widestdistribution of all types is recognized at the slope betweentrench-slope break and deformation front, and its mainoccurrence is located off Shikoku. The distribution area ofthese four types occupies about 90 % of all BSRs in thisstudy area. Such high percentage suggests that the most ofthe BSRs are formed by upward migration and accumula-tion of gas and hydrate recycling.

This explanation is supported by geochemical investi-gation because TOC values of less than 1 % were reportedin the sediments recovered from ODP Site 808 (Bernerand Koch, 1993) and the MITI Nankai Trough wells. Inthe case of such low TOC values, it is difficult to form gashydrates from an in-situ carbon source of gas in the sedi-ments of the gas hydrate zone (Minshull et al., 1994;Waseda, 1998). Further support for the requirement of gasmigration and/or hydrate recycling may be given by theexistence of a highly concentrated hydrate zone (70-80 mthick) observed just above the free gas zone in the succes-sion of the MITI Nankai Trough wells. The hydrate con-tent was 60-70 % of pore volume (porosity 40-50 %).These values were obtained from chloride anomalies inpore water and physical log analysis (TOC and hydrate


Fig. 4-2 Distribution map of four BSR types.

content are unpublished data from the well report of theMITI Nankai Trough wells). Occurrence of such highconcentration of gas hydrate can be interpreted from themodel in which the gas hydrate amount increases dramati-cally just above the BGHS when the gas is advected frombelow (Paull et al., 1994). A similar interpretation is givenby mathematical simulation using 1-dimensional staticmodel to simulate accumulation process of marine gashydrate. In this simulation, strongly hydrated sedimentscan be calculated just above the free gas zone by settingupward migration of gas-bearing fluid below the BGHS(Baba and Katoh, 2004).

2.3. Relation between BSR distribution and prism size

As shown in Figures 4-1 and 4-2, the accretionaryprism type BSR has the widest distribution ofall BSR types and its main distribution is rec-ognized off Shikoku where the Nankai accre-tionary prism is the largest. Such concentra-tion of BSRs in the prisms has been consid-ered to be common (e.g. Shipley et al., 1979;Kvenvolden and Barnard, 1983), because theaccretionary prism is dominated by active andlarge volumes of fluid flow in comparisonwith normal basins (e.g. Bray and Karig,1985; Moore and Vrolijk, 1992). The distrib-ution of the BSRs is reduced toward north-east, offshore Tokai, where the size of theprism is also smaller (Figs. 4-1 and 4-2). Suchchange can be examined semi-quantitativelyby using a ratio of width of the accretionaryprism to that of the total summed extent ofobserved BSRs on the same section lines(Fig. 5). This plot presents a relationship inwhich the ratio is directly proportional to thewidth of the prism. The result indicates clear-ly that the distribution of BSRs iscontrolled by the size of the prism.This suggests that an increase ofprism size accelerates hydrocarbongeneration and expulsion of gas-bear-ing fluids that migrate upward to theBGHS. As tectonic thickening andloading progresses with growth of theprism, the sediment porosity shouldbe decreased with the increase of vol-ume of fluid expulsion, which canbring hydrocarbons to shallow depth.

3. Direct Indicators of FluidMigration on Seismic Records

In the previous paragraph, theimportance of upward migration and

accumulation of gas to the BGHS began to emerge fromthe examination of origin of BSRs by using the distribu-tion pattern of BSRs. To support such interpretation, wetry to find more decisive evidence on seismic records.

3.1. High amplitude reflections

Seismic amplitude can help to discriminate gas-bearingsediments from water-bearing ones (e.g. Laberg andAndreassen, 1996; Grauls, 2001). The gas-bearing sedi-ments can generate anomalously high amplitude reflec-tions because even small amounts of gas (1-2 %) maycause a marked P-wave velocity drop (Domenico, 1976).

Such high amplitude reflections are recognized on sev-eral seismic lines in this study area. In an example, Figure6 shows a thick zone of high amplitude reflections


Fig. 6 Thick zone of high amplitude reflections underneath BSR. BSRs aremarked by black arrows. The location of this figure is Line A on Figure 13.

1

0.8

0.6

0.4

0.2

00 20 40 60 80 100

Rat

io o

f tot

al s

umm

ed e

xten

t of B

SR

s to

wid

th o

f acc

retio

nary

pris

m o

n a

sam

e lin

e

Width of Accretionary Prism (km)

Fig. 5 Cross plot of width of accretionary prism versus ratio of totalsummed extent of observed BSRs to width of accretionary prism. Thetotal extent of BSRs and width of accretionary prism are measured onthe same section line.

beneath the BSR (about 0.3 second thick) developing inalternating beds of sand and mud. In this case, high ampli-tude reflections are considered to result from the gas-bear-ing sands in which gas is trapped by the gas hydrate sealdeveloped above the BSR. The base of the high amplitudereflections may correspond to the base of the gas column.In addition, Figure 6 shows several high amplitude reflec-tions with reverse polarity which may be generated by aP-wave velocity change from high-velocity water-bearingmud (impermeable facies) to low-velocity gas-bearingsand (permeable gas reservoir).

Figure 7 also shows the large-scale distribution of high

amplitude reflections developing in alternating beds ofsand and mud. The high amplitude reflections are termi-nated to the left by the unconformity and display anuppermost horizon at the BSR. This configuration mayexpress the upward migration of gas-bearing fluidsalong the unconformity from the greater depths andtheir accumulation in a stratigraphic trap sealed byhydrate-bearing sediments above the BSR. In this case,the unconformity acts as a seal because of the low-per-meability mud-dominant facies above it.

Figure 8 is a remarkable example of control of the BSRby upward migration of gas-bearing fluids. At the right


Fig. 8 Example of high amplitude reflection indicating upward migration of gas-bearing fluid. BSRs are marked by blackarrows. The location of this figure is Lind C on Figure 13.

Fig. 7 Example of high amplitude reflections indicating upward migration and accumulation of gas to BGHS. BSRs aremarked by black arrows. The location of this figure is Line B on Figure 13.

side of this figure the clear BSR can be recognized. ThisBSR fades to the left and finally terminates at the centralpart of this figure, associated with the development ofanomalously high amplitude reflections underneath anunconformity. Such disappearance of the BSR may beexplained by the inhibition of upward migration of gas-bearing fluids by the unconformity trap. In this case, thehigh amplitude reflection zone underneath the unconfor-mity may indicate free gas trapped by the unconformity.This interpretation emphasizes the existence of upflow ofgas-bearing fluids.

3.2. Upwarping of BSRs

Figure 9 shows the upwarp of a BSR recognized in theanticlinal fold. This anticlinal fold is pierced by a muddiapir associated with a mud volcano. Themud diapir is the most long-lived fluidexpulsion structure. The mud diapir risesvertically from greater depths and is com-posed of plastic mud having higherfluid/mud temperature than the surround-ing strata. Gas is a common component ofmud diapirs in an accretionary prism(Carson and Screaton, 1998). In general,the upwarping of the BSR is considered toresult from local shallowing of the BGHScaused by local heating from ascendingflow of deep fluids passing through themud diapir piercing the anticlinal fold.However, the fluid expulsion from greaterdepths may contaminate the gas hydrateswith thermogenic gases.

The distribution area of the upwarpingin the 2D high-resolution seismic surveyarea is summarized in Figure 13. Theseareas are consistent with the locations ofanticlinal folds pierced by mud diapirs.

4. Sedimentary Controls on BSR Distribution

The sedimentary control on BSR distribution is recog-nized at the northern flank of the Daini Atsumi Knoll(Fig. 10). In the right side of this figure, the clear BSR isrecognized in alternating beds of sand and mud facies.This BSR is terminated to the left by the unconformitycovered by the mud-dominant facies. Above the uncon-formity, the low-amplitude BSR develops again in alter-nating beds of sand and mud facies intercalated by themud-dominant facies. Therefore, it is clear that the BSRsare generated selectively in alternating beds of sand andmud facies. Such selective generation of the BSRs direct-ly indicates the important role of sand layers as a fluidconduit for upward migration of gas-bearing fluids. Such


Fig. 10 Example ofsedimentary controlof BSR. BSRs aremarked by blackarrows. The locationof this figure is LineE on Figure 13.

Fig. 9 Example of upwarping of BSR and associated mud volcano. BSRs aremarked by black arrows. The location of this figure is Line D on Figure 13.

fluid conduit model of sand layer has been assumed bymany workers (e.g. Moore, 1989; Pecher et al., 2001).

On the contrary, in the Shima Spur, the BSRs are dis-tributed widely even in mud-dominant facies (Figs. 3a and14). In this case, there may be enough time for upwardmigration and accumulation of gas to the BGHS passingthrough low-permeability mud facies, because the ShimaSpur is thought to be an old structure formed during theMiddle to Late Miocene based on the tectonic historyanalysis using seismic records. On the other hand, theAtsumi Daini Knoll is thought to be a very young struc-ture formed during the Quaternary to the Recent. Such ashort period cannot provide enough time for gas-bearingfluids to infiltrate into low-permeability mud facies and,as a result, the migration and accumulation of gas may beconfined to permeable sand layers.

From the above considerations, it is revealed that thesedimentary controls on BSR distribution are influencedby the time available for migration of gas-bearing flu-ids. This result implies that time is one of the importantfactors in determining the distribution of BSRs, in addi-tion to the sedimentary controls.

5. BSR Distribution Model for Accretionary Prismand Forearc Basin System

As shown in Figure 1, several forearc basins liebetween the older accreted belt of the Japanese island-arcand the Nankai accretionary prism. In the forearc basins,BSRs are recognized mainly at anticlinal structures(buried anticline type BSR), topographic highs (ridge typeBSR) and basin margins (basin margin type BSR)(Figs. 4-1 and 4-2). These BSRs are formed by upward migrationof gas-bearing fluids generated from a thick pile of sedi-ments deposited from submarine fan systems. In this case,

the turbidite sand layers intercalated by submarine fandeposits provide the permeable fluid conduit for theupward migration of the fluids. Such a migration andaccumulation style has been widely observed in petrole-um geology.

In contrast to the forearc basin situation, in the Nankaiaccretionary prism, the BSRs are characterized by a distri-bution pattern which is rarely controlled by either geolog-ic structure or topography (Figs. 4-1 and 4-2). Such a dis-tribution pattern is thought to be caused by strongly diffu-sive flow of gas-bearing fluids occurring as a result of tec-tonic thickening and loading of accreted sediments. Suchdiffusive fluid flow may occupy a significant proportionof the total fluid loss from the prism (Carson andScreaton, 1998). The gas source is the accreted trench andocean floor deposits consisting of hemipelagic and pelagicmud with low TOC values (Berner and Koch, 1993). Themigration of gas-bearing fluids by the diffusive flow maybe mainly carried out through permeable sand layers,because the Nankai accretionary prism is classified as acoarse clastic accretionary wedge. The micro-fracture sys-tem formed by the tectonic loading may also contribute tothe diffusive fluid flow. However, with the decrease of theprism size, the flux of diffusive fluid flow reduces and, asa result, the migration and accumulation style of hydro-carbons comes to approach to that of a normal basin. Inthis case, the BSR distribution is limited to the anticlines(ridges and knolls) which are commonly thought to be themost effective structure to accumulate hydrocarbons (seeoff Kumano basin and off Enshu Trough in Fig. 4-1). Inaddition, the focused fluid flow may occur along fault sys-tems (e.g. Moore and Vrolijk, 1992). In this case, the flu-ids may be derived from a deep source. Figure 11 showsthe direct evidence of focused fluid flow from the sub-ducting sediments. Another fluid flow may occur below


Fig. 11 Mud volcano developing at the toe of the prism, off Shikoku. BSRs are marked by black arrows. The upwarping of BSRis distinct. Perhaps, the warm deep-fluid comes up directly passing through mud diapir or thrust fault system from overpres-sured subducting sediments. In the Nankai accretionary prism, the subducting sediments have been supposed to be overpres-sured (e.g. Moore et al., 1991; Byrne et al., 1993).

the decollement zone (e.g. Moore, 1989) and/or along thedecollement zone (e.g. Minshull and White, 1989; Mooreand Vrolijk, 1992). In the Nankai accretionary prism, suchfluid flow is suggested by the chemical anomaly of porefluid developing at the decollement zone (You et al.,1993; Saffer and Bekins, 1998). The local distribution ofBSRs may be modified by slumping or local uplift bythrusting (Ashi and Tokuyama, 1994).

The other major distribution of the BSRs is recognizedat the trench-slope break where large topographic highsare developed between the forearc basin and the accre-tionary prism, and many buried anticline-type and ridge-type BSRs are distributed at mound-shaped or ridge-shaped highs associating with anticlinal structures. In thetrench-slope break, the highly fractured strata may bedominant because the deformation in the older portion ofthe prism, where sediments have undergone considerablecompaction, is characterized by brittle fracture and dila-tion (Moore and Vrolijk, 1992). This implies that frac-tures constitute an important system to migrate gasupward to the BGHS in the trench-slope break.

The origin of various types of BSRs in the Nankaiaccretionary prism and forearc basins are summarizedin the schematic illustration of BSR distribution modelshown in Figure 12.

6. Other Characteristic Reflections Associated withBSRs

In this chapter, we will describe the characteristicreflections associated with BSRs in the 2D high-resolu-tion seismic survey area. These reflections are observedon the 2D high-resolution seismic records only.

6.1. High amplitude reflection zone just below BSR

Figure 14 shows an example of a high amplitude

reflection zone developed just below a BSR. The thick-ness of the high amplitude zone is several tens of mil-liseconds. This phenomenon is widely recognized in the2D high resolution seismic survey area (Fig. 13). Thevelocity analysis reveals that the high amplitude zonecorresponds to low velocity zone (Inamori and Hato,2004). Therefore, the high amplitude reflections arethought to express the gas-charged sediments forming afree gas zone. The sedimentary facies of this figure issupposed to be mud-dominant. On the other hand, in thecase of alternating beds of sand and mud, the highamplitude zone is characterized by a jagged pattern asshown in Figure 15. This jagged pattern may be derivedfrom gas-bearing fluids reserved in permeable sand lay-ers of alternating beds. The similar high amplitudereflections just below a BSR have been reported fromthe Blake Ridge (Holbrook et al., 1996).

The origin of the free gas zone causing the highamplitude reflection zone is considered to be; i) upwardmigration and accumulation of gas, ii) gas hydrate recy-cling, or iii) partial melting of the bottom of the gashydrate zone caused by regional uplift or sea-level fall.According to our consideration in the previous para-graph, cases i) and ii) may be the most likely.

6.2. Amplitude attenuation below BSR

Figure 16 shows the remarkable amplitude attenua-tion of seismic reflections occurring below a BSR. Suchamplitude attenuation frequently masks the stratigraphicreflections below BSRs. In general, it is known that gas-charged sediments attenuate P-wave, which often pre-cludes imaging beneath the gas zone (Pecher andHolbrook, 2000). A recent study in the Blake Ridgealso shows strong P-wave attenuation in the free gaszone (Wood et al., 2000). Furthermore, in this studyarea, the distribution of the amplitude attenuation is


Older Accreted Belt

Forearc Basin

Accretionary Prism

Trench

Oceanic BasementDecollement

Mud volcanoFocused fluid flow

Mud diapirDiffusive fluid flow

Fig. 12 Schematic illustration of BSR distribution model in accretionary prism and forearc basin system. Black and whitearrows indicate the migration of gas-bearing fluids.

generally consistent with that of the high amplitudereflection zone just below the BSR (Fig. 13). The highamplitude reflection zone is considered to result fromgas-charged sediments in the free-gas zone, and thus isconsistent with the argument that amplitude loss results

from attenuation of P-waves above the zone of weak-ened reflections.

6.3. High amplitude reflections just above BSR

Figure 17 shows an example of high-amplitude


Fig. 14 Example of high amplitude reflection zone justbelow BSR developing in mud-dominant facies. BSRsare marked by black arrows. The location of this figure isLine F on Figure 13.

Fig. 15 Example of jagged pattern of high amplitude reflec-tion zone just below BSR developing in alternating bedsof sand and mud facies. BSRs are marked by blackarrows. The location of this figure is Line G on Figure 13.

34°30N

34°00N

136°30E 137°00E 138°00E137°30E

33°30N

Fig. 13 Location map of characteristic reflections associating with BSRs in 2D high resolution seismic survey area. Lines Ato J indicate the location of seismic examples shown in figures.

reflections just above BSR developing in alternatingbeds of sand and mud facies. In this figure, we can findthe strong reflections with positive polarity developingjust above the BSR. Such reflections may be derivedfrom the sand layers of alternating beds cemented byhighly concentrated gas hydrates, because the replace-ment of pore fluids by rigid gas hydrates causes anincrease in P-wave velocity (Dvorkin et al., 2000). Infact, Inamori and Hato (2004) have reported high P-wave velocity from similar reflections developing justabove a BSR. Therefore, the top of a zone of high-amplitude reflections is thought to express the top ofstrongly hydrated sediments. This explanation is sup-ported by the discussion in section 2.2. in which thestrongly hydrated sediments have been assumed justabove the BSR. The distribution of this type of reflec-tion is limited to small areas as shown in Figure 13.

6.4. Double BSR

Figure 18 shows an example of a double BSR con-sisting of an upper BSR and a lower BSR with negativepolarity. The origin of the double BSR has not beencompletely elucidated yet. However, the lower BSRmay be a relic BSR representing the former location ofBGHS. In this case, the double BSR represents tempo-ral, transient conditions moving toward new PT-regimecaused by recent tectonic uplift, sea-level fall orincrease of bottom water temperature (Matsumoto,2000). The double BSR may also represent the upperand lower boundaries of the transitional zone betweenthe hydrate zone and the free gas zone assumed byDavie and Buffet (2003). The distribution of double

BSRs is limited to a small area as shown in Figure 13.

7. Conclusion

Ninety percent of the distribution area of BSRs in thestudy area is occupied by ridge type, buried anticlinetype, basin margin type and accretionary prism typeBSRs. The first three types are thought to be formed byupward migration of gas-bearing fluids and hydraterecycling, and the last type is thought to be formed bystrong upward diffusive flow and hydrate recyclingcaused by the tectonic process of sediment accretion.Therefore, most of the BSRs in the Nankai accretionaryprism and forearc basin system are supposed to be ofmigration and recycling origin. This conclusion is com-patible with the geochemical evidence of low TOC val-


Fig. 17 Example of high amplitude reflectionsjust above BSR. BSRs are marked by blackarrows. The location of this figure is Line Ion Figure 13.

Fig. 18 Example of double BSR. The location of this fig-ure is Line J of Figure 13.

Fig. 16 Example of amplitude attenuation below BSR. BSRsare marked by black arrows. The location of this figure isLine H on Figure 13.

ues and the gas hydrate occurrence characterized bystrongly hydrated sediments developing just above thefree-gas zone. In addition, the characteristic configura-tion of high amplitude reflections and the upwarping ofthe BSR associated with mud diapirs is direct evidenceindicating the existence of upward migration of gas-bearing fluids to the BGHS.

In the forearc basins, the BSRs are recognized at topo-graphic highs, anticlines and basin margins. Such a distri-bution pattern indicates the existence of structural controlon the migration and accumulation of gas to the BGHS.On the other hand, in the Nankai accretionary prism, theBSRs are mainly caused by an active and large volume ofdiffusive fluid flow derived from tectonic thickening andloading with sediment accretion and, furthermore, the dis-tribution of the BSRs is strongly controlled by the prismsize. This suggests that the tectonic control is one of theimportant factors in determining BSR distribution in theprism. In addition, the BSR distribution confined to alter-nating beds of sand and mud designates the existence ofsedimentary control. In contrast, in mud-dominant facies,the wide distribution of BSRs formed by long-livedmigration of gas-bearing fluids indicates an importance oftime control. Those four controls are thought to be themain constituents to determine the distribution pattern ofBSRs in the Nankai accretionary prism and forearc basinsystem. In addition, TOC in the sediments is a fundamen-tal control in determining the distribution of BSRs (geo-chemical control).Acknowledgments: We are grateful to the Japan NationalOil Corporation (JNOC) providing METI fundamentalgeophysical exploration data. We also appreciate Mr.Hato, Mr. Oikawa and Mr. Inamori for their valuablesuggestions. This work was performed as a part of theproject by MH 21 Research Consortium in Japan.

References

Ashi, J. and Tokuyama, H. (1994) Development of the Nankaiaccretionary prism and the formation of gas hydrate.Gekkan Chikyu, 539–544 (in Japanese).

Baba, K. and Uchida, T. (1999) Newly discovered BSRs inmarginal seas of eastern Japan. Jour. Japan Assoc.Petroleum Technol., 64, 346 (in Japanese).

Baba, K. and Yamada, Y. (2000) A tectonic control on methanehydrate distribution at island arc systems. Western PacificGeophysics Meeting, AGU, Eos, Transactions, AGU, 81,WP59.

Baba, K. and Katoh, A. (2004) Application of a simulationmodel for the formation of methane hydrate to the NankaiTrough and the Blake Ridge: natural examples of twoend-member cases. Resource Geol., 54, 125–135.

Berner, U. and Koch, J. (1993) Organic matter in sediments ofSite 808, Nankai accretionary prism, Japan. in Hill, I. A.,Taira, A., Firth, J. V. and others (eds.) Proc. ODP Sci.Results, 131, 379–385, Ocean Drilling Program, College

Station, Texas.Bray, C. J. and Karig, D. E. (1985) Porosity of sediments in

accretionary prisms and some implications for dewateringprocesses. Jour. Geophys. Research, 90, 768–778.

Byrne, T., Maltman, A., Stephenson, E., Soh, W. and Knipe,R. (1993) Deformation structures and fluid flow in the toeregion of the Nankai accretionary prism. in Hill, I. A.,Taira, A., Firth, J. V. and others (eds.) Proc. ODP Sci.Results, 131, 83–101, Ocean Drilling Program, CollegeStation, Texas.

Carson, B. and Screaton, E. J. (1998) Fluid flow in accre-tionary prisms: evidence for focused, time-variable dis-charge. Rev. Geophys., 36, 329–351.

Claypool, G. E. and Kaplan, L. R. (1974) The origin and dis-tribution of methane in marine sediments. in Kaplan, I. R.(ed.) Natural Gases in Marine Sediments, 99–139, PlenumPress, New York.

Davie, M. K. and Buffet, B. A. (2003) Sources of methane formarine gas hydrate: inferences from a comparison ofobservations and numerical models. Earth Planet. Sci.Lett., 206, 51–63.

Davis, A. A., Hyndman, R. D. and Villinger, H. (1990) Ratesof fluid expulsion across the northern Cascadia accre-tionary prism: constraints from new heat flow and multi-channel seismic reflection data. Jour. Geophys. Research,95, 8869–8889.

Domenico, S. N. (1976) Effect of brine-gas mixture on veloci-ty in an unconsolidated sand reservoir. Geophysics, 42,882–895.

Dovorkin, J., Helgerud, M. B., Waite, W. F., Kirby, S. H. andNur, A. (2000) Introduction to physical properties and elas-ticity models. in Max, M. D. (ed.) Natural Gas Hydrate inOceanic and Permafrost Environments, 245–260, KluwerAcademic Publishers, Dordrecht.

Ecker, C., Dvorkin, J. and Nur, A. (1998) Sediments with gashydrates: internal structure from seismic AVO. Geo-physics, 63, 1659–1669.

Grauls, D. (2001) Gas hydrates: importance and applicationsin petroleum exploration. Marine Petroleum Geol., 18,519–523.

Holbrook, W. S., Hoskins, H., Wood, W. T., Stephen, R. A.,Lizarralde, D. and Leg 164 Science Party (1996) Methanehydrate and free gas on the Blake Ridge from verticalseismic profiling. Science, 273, 1840–1843.

Hyndman, R. D. and Davis, E. E. (1992) A mechanism for theformation of methane hydrate and seafloor bottom-simulat-ing reflectors by vertical fluid expulsion. Jour. Geophys.Research, 97, 7025–7041.

Inamori, T. and Hato, M. (2004) Detection of methanehydrate-bearing zone from seismic data. Resource Geol.,54, 99–104.

Kvenvolden, K. A. and Barnard, L. A. (1983) Hydrates of natur-al gas in continental margins. in Watkin, J. S. and Darke, C.L. (eds.) Studies in Continental Margin Geology. Mem.Amer. Assoc. Petroleum Geol., 34, 631–640.

Laberg, J. S. and Andreassen, K. (1996) Gas hydrate and freegas indications within the Cenozoic succession of theBjornoya Basin, western Barents Sea. Marine PetroleumGeol., 13, 921–940.



Matsumoto, R. (2000) Double BSR in the eastern NankaiTrough: fact or artifact. Western Pacific GeophysicsMeeting, AGU, Eos, Transactions, AGU, 81, WP63.

Minshull, T. and White, R. (1989) Sediment compaction andfluid migration in the Makran accretionary prism. Jour.Geophys. Research, 94, 7387–7402.

Minshull, T. A., Singh, S. C. and Westbrook, G. K. (1994)Seismic velocity structure at a gas hydrate reflector, off-shore western Colombia from full waveform inversion.Jour. Geophys. Research, 99, 4715–4734.

Moore, J. C. (1989) Tectonics and hydrogeology of accre-tionary prisms: role of the decollement zone. Jour. Struct.Geol., 11, 95–106.

Moore, J. C. and Vrolijk, P. (1992) Fluids in accretionaryprisms. Rev. Geophys., 30, 113–135.

Moore, G. F., Karig, D. E., Shipley, T. H., Taira, A., Stoffa, P.L. and Wood, W. T. (1991) Structural framework of theODP Leg 131 area, Nankai Trough. in Taira, A., Hill, I. A.,Firth, J. V. and others (eds.) Proc. ODP Initial Rept., 131,15–20, Ocean Drilling Program, College Station, Texas.

Paul, C. K., Ussler, W. and Borowski, W. S. (1994) Sources ofbiogenic methane to form marine gas hydrates. in Sloan, J.H. and Hnatow, M. A. (eds.) International Conference onNatural Gas Hydrates, 392–409, Plenum Press, New York.

Pecher, I. A. and Holbrook, W. S. (2000) Seismic methods fordetecting and quantifying marine methane hydrate/freegas reservoirs. in Max, M. D. (ed.) Natural Gas Hydrate inOceanic and Permafrost Environments, 275–294, KluwerAcademic Publishers, Dordrecht.

Pecher, I. A., Kukowski, N., Huebscher, C., Greinert, J.,Bialas, J. and the GEOPECO Working Group (2001) Thelink between bottom-simulating reflections and methaneflux into the gas hydrate stability zone – new evidence fromLima Basin, Peru Margin. Earth Planet. Sci. Lett., 185,343–354.

Saffer, D. M. and Bekins, B. A. (1998) Episodic fluid flow inthe Nankai accretionary complex: timescale, geochemistry,flow rates, and fluid budget. Jour. Geophys. Research, 103,30351–30370.

Seno, T. and Maruyama, S. (1984) Paleogeographic recon-struction and origin of the Philippine Sea. Tectonophysics,102, 53–84.

Shipboard Scientific Party (1991) Geological background andobjectives. in Taira, A., Hill, I. A., Firth, J. V. and others(eds.) Proc. ODP Initial Rept., 131, 5–14, Ocean DrillingProgram, College Station, Texas.

Shipley, T. H., Houston, M. H., Buffler, R. T., Shaub, F. J.,McMillen, K. J., Ladd, J. W. and Worzel, J. L. (1979)Seismic evidence for widespread possible gas hydratehorizons on continental slopes and rises. Amer. Assoc.Petroleum Geol. Bull., 63, 2204–2213.

von Heune, R. and Pecher, A. I. (1999) Vertical tectonics andthe origins of BSRs along the Peru margin. Earth Planet.Sci. Lett., 166, 47–55.

Waseda, A. (1998) Organic carbon content, bacterial methano-genesis, and accumulation processes of gas hydrates inmarine sediments. Geochem. Jour., 32, 143–157.

Wood, W. T., Holbrook, W. S. and Hoskins, H. (2000) In-situmeasurements of P-wave attenuation in the methanehydrate- and gas-bearing sediments of the Blake Ridge. inPaull, C. K., Matsumoto, R., Wallace, P. J. and Dillon, W.P. (eds.) Proc. ODP Sci. Results, 164, 265–272, OceanDrilling Program, College Station, Texas.

Yamano, M., Kinoshita, M., Goto, S. and Matsubayashi, O.(2003) Extremely high heat flow anomaly in the middle partof the Nankai Trough. Phys. Chem. Earth, 28, 487–497.

You, C. F., Gieskes, J. M., Chen, R. F., Spivack, A. andGamo, T. (1993) Iodide, bromide, manganese, boron, anddissolved organic carbon in interstitial waters of organiccarbon rich marine sediments: Observation in the Nankaiaccretionary prism. in Hill, I. A., Taira, A., Firth, J. V. andothers (eds.) Proc. ODP Sci. Results, 131, 165–174,Ocean Drilling Program, College Station, Texas.

Yuan, T., Spence, G. D. and Hyndman, R. D. (1999) Seismicvelocity studies of a gas hydrate bottom-simulating reflectoron the northern Cascadia continental margin: amplitudemodeling and full waveform inversion. Jour. Geophys.Research, 104, 1179–1191.

bsrs and associated reflections as an indicator of gas hydrate and free gas accumulation: an example...

Documents