- Idand Correlation

Correlation of Seismic Reflectors with the CRP-3 Drillhole, Victoria Idand Basin, Antarctica

Abstract - Seismic rel'lcction ilatti collccteil i i ~ the region ol'f.shoi-c Ciipc Roberts i'eveal a seaward dippins scismic sequence that thickens into the Victoria Land Biisin. Drillhole CRP-3 cored this sctliicnce to ti depth of 039 mbsf (equiviilcnt to 1025 111s two-way-time below the sctil'loor). This e x t e ~ ~ ~ l e d the totiil thickness of Ccnozoic strata cored b y CRP to 1500 m. The CRP-3 tilso cored 116 111 into Devonian sandstone basement At least 10 reflection events can he identificcl in the scismic data. These events are related to litliolosic changes within CRP-3 core and down-hole electrical logs by (Jeriving a time-depth relationship from whole-core velocity measurements. Data from a vertical scismic profile experiment in CRP-3.

together with synthetic seismograms and the depth converted scismic section. enable tlie correltition of 9 key seismic reflection events (p-X) and 4 ma,jor seismic scq~icnccs (U-X) with sedimentary sequence boundaries and lithostratigraphic sub-units clocumentecl in the core. Seismic sequence U is correlated to sedimentary cycles 19 to 25 i n tlie lower part ol' CRP-2/2A and to cycles I and 2 in upper part 01' CRP-3. Seismic sequence V corresponds to sedimentary cycles 3 to 6. Seismic sequences U and V arc both dated tit around 3 I lo 32 Ma. Sequence W includes the sctlimcntary cycles 7 to 23 beneath ( c . 460 inhsf). and sequence X corresponds to the lowest part of the sedimentary section, including 40 m of dolerite brcccia a n d conglomerate resting on the Devonian Beacon sandstone beneath. Sequences W and X arc estimated to be around 33 Ma. and possibly as old as 34 Ma. The sequences. U-X, are all within V5 of Cooper et al. (1987) and RSS-1 of Brancolini et al. (1995) and dip >1X0 to the northeast.

INTRODUCTION

Seismic data provides essential spatial context for the high-resolution history of ice margin and water depth changes derived from drill hole data. For this reason an important component of the stratigraphic drilling project a t C a p e Roberts ( C R P Project) is successfully correlating the Oligocene and Miocene stratigraphy f r o m c o r e to s e i s m i c reflection d a t a (Cape Roberts Science Team, 1998, 1999, 2000). The seismic stratigraphic framework developed at CRP-1 and CRP-212A and integrated with seismic mapping in the Victoria L a n d Basin have been reported in Henrys et al., 2000 and Davey et al., 2000. A critical element of this analysis was the correlation between seismic sequences and equivalent lithostratigraphic and stratigraphic boundaries.

In this paper, we extend our analysis to drill site CRP-3 using synthetic seisinograms derived from logs of CRP-3 drill core and the corresponding processed VSP, multi- and single- channel seismic reflection d a t a that c r o s s c l o s e to t h e d r i l l s i t e . We a l s o establish the geometry of major seismic sequences by

tracingthem hundreds to thousands of meters in both coast-parallel and coast-orthogonal directions, and attempt to relate erosional and depositional features in the seismic records to core features. Throughout this paper we follow the nomenclature of Sheriff ( 1 980) where seismic sequences are mappable seismic facies b o u n d e d by u n c o n f o r m i t i e s ( o n l a p , t o p l a p , a n d d o w n l a p ) . W h e r e t h e r e is no e v i d e n c e f o r an unconformity then we refer to these bounding packets a s s e i s m i c units. S e i s m i c s e q u e n c e s r e s u l t f r o m sediments within a time-stratigraphic depositional unit that implies a certain age interval.

Seismic stratigraphy of the Victoria Land Basin w a s first established by Cooper et al. (1987), who identified three major basin wide seismic units (V3- V5). In addition, Brancolini et al., (1995) extended t h e Victoria Land Basin stratigraphy developed by C o o p e r et al. (1987) by correlating t h e R o s s S e a seismic sequences (RSS 1-RSS8), from the central and e a s t e r n R o s s S e a . I n M c M u r d o S o u n d a n d C a p e Roberts region this seismic stratigraphy was further refined into 10 seismic sequences (K ^- T ) and the V 3 , V 4 , a n d V5 units c o u l d be i d e n t i f i e d a s an Corresponding author (s.henrys@gns.cri.nz)

Fig. I - Location and bathymetry map of Cape Roberts showing the distribution of the sedimentary sequences (V3. V4. and V5). The licavy grey lines are magnetic anomaly lineaments interpreted as faults by Bozzo et al. ( 1 9 9 7 ) . West of Roberts Ridge. NW-SE U-ending lineaments coincide with mapped normal faults (Hamilton et al.. 2001) and are proposed as the main boundary faults that separates ~ l i c sedimentary sequence from the basement granitic rocks (V7). which crop out at the coast. The extent of shallow igneous bodies. inferrcd from magnetic data (Bozzo et. al.. 1997) is also shown in the shaded pattern. Ship tracks of multi- and single-channel seismic data are arcy lines ^- tracks named in the text are bold and annotated with shot points.

amalgamation of mappable sequences. For example, sequence R, S , and T were identified as part of the Cooper et al. ( 1 9 8 7 ) "super sequence" V 4 and inferred to correspond with early Oligocene and late Eocene rocks at depths greater than 366 m in CIROS-1 (Barrett et al., 1995; Bartek et al., 1996) - see Tab. 1 of Henrys et al. (1998) for correlation between the different nomenclatures. A number of these seismic sequences, including the three sub-seismic sequences of V4, were previously mapped across Roberts Ridge (Henrys et al., 1994, 1998) and identified in core

recovered at CRP-2. Nine additional separate seismic events have been identified on SCS lines crossing the CRP-3 drill site and complete the mapping of V5 t o basement in this region.

SEISMIC REFLECTION DATA

The primary data set for drillhole correlation is the single-channel seismic (SCS) data from the NBP9601 c r u i s e (Hamilton e t al., 1 9 9 8 , 2 0 0 1 , and

g l ). T w o profiles cross close to the drill site:

NBPOOO 1-89 passes H-W and 3 0 0 m south of the tlrill liok- and NBP9601-92 passes N-S, about 800 in east (l:ig. 1). These data sets include a minimum ainoinil of processing to preserve both true amplitu~le and reliitive t r a c e - t o - t r a c e a m p l i t u d e s , i . e . , n o deconvolntion or S-k filtering, or trace averaging were applied. However, the core at CRP-3 penetrated strata below (lie seafloor multiple. The use of multi-channel seismic ihita can significantly reduce seafloor multiple energy iind improves the correlation of events below the niulliplc (Hamilton et. al., 1998). Whereas multi- channel seismic helped in the interpretation of CRP- 2 / 2 A , these data were f o u n d not to b e useful for correhiting data beneath the multiple near CRP-3.

A l s o . r e p r o c e s s i n g of N B P 9 6 0 1 - 8 9 failed to sigiiil'icantly improve the quality of data below the multiple in the vicinity of CRP-3 (i.e. below 92 sec two-way travel time). The reason for this is the high impedance at the seafloor and rapid decay of reflected energy after 2 sec.

The near-offset (normal incidence) seismic traces f o r N B P 9 6 0 1 - 8 9 w e r e m i g r a t e d u s i n g velocity i n f o r m a t i o n derived f r o m o u r r e p r o c e s s i n g and velocity measurements from cores. Near-offset data are displayed in all the figures 2 to 5 and both near- offset iind migrated data are shown in figure 3. We noted that the effect of migrating data is to steepen dipping structures, but migration of water-bottom m u l t i p l e s c a u s e s artifacts - "smiles" - a n d . in particular, impairs interpretation of reflectors through the multiple (see Fig. 3b). To overcome this problem, and correctly position reflection events in depth, we depth-migrated (Fig. 4c) the line drawing of travel- t i m e arrivals in f i g u r e 4b. H o w e v e r , w e used all available seismic information, both near-trace and migrated images, in helping to refine our interpreted depth sections.

VERTICAL SEISMIC PROFILE AND MEASUREMENTS ON CORES

VERTICAL SEISMIC PROFILE (VSP)

VSP travel-time data can b e used to determine velocities to serve as a basis f o r comparison with down-hole sonic and core measurements and can also b e used to tie directly into marine seismic reflection data, resulting in reliable depth-time conversion for seismic reflection data.

Three separate three-component vertical seismic profiles (VSPs) were completed at CRP-3 (see Cape R o b e r t s S c i e n c e T e a m , 2 0 0 0 ) b u t o n l y t h e Z- component data for the near -offset VSP are reported here. Processing of the data includes trace display and editing. Down-going waves are marked by clear first arrivals and a low frequency ( < l 0 Hz) complex coda t h a t i n c l u d e s t h e s o u r c e s i g n a t u r e a n d s h a l l o w

revcrbwilions i n the sea ice ancl reflectio~~s from the seiil'looi'. This wavetrain masks np-going rcl'lection r r i v a l s . Separation of down-going a ~ u l up-going waves was carried out by median filtering a n d polygon mute i n (lie fi.eqiieiicy-waveiiiiniher (S-k) dom;iin. '1'0 compare the VSP to two-way traveltime SCS data, the VSP traces have a static shift applied;

equal to the first-arrival times. A n 80 insec corridor al'tcr the first arrival is used to correlate well data t o SCS reflcctioii data (see Fig. 2). A detailed analysis o f the three-component VSP data is to follow in a separate publication.

VELOCITY 1j.s DEPTH

A n important application of V S P data is t o provide a c c u r a t e velocity-depth data for t h e f o r m a t i o n s penetrated by the well. Velocity measurements of the core provide substantially better ilepth resolution, but the VSP results arc much less sub.ject to borehole conditions. Indeed, they average over a much larger volume of the formation so that they provide velocities that are more representative of seismic reflection velocities than those obtained from core or core plugs. Interval velocities are calculated from travel times picked off first-arrivals of near- offset VSP data (± msec accuracy) and plotted as a function of depth together with core velocities (Cape Roberts Science Team, 2000; and Niessen, personal communication) in figure 2. The core velocities are median filtered over a sliding 10 m window of core.

Our filter will also reject outliers if they are greater than 2.5 times the median value. A comparison of the velocity and time-depth curves (Fig. 2) show they are remarkably similar, with VSP velocities consistently about 0.2 km/s (or -7%) higher than core velocities at depths greater than 450 meters below sea floor (mbsf). In contrast, VSP velocities are systematically slower than whole-core velocities in the top half of the hole. We d o not know why there is a systematic mismatch in velocity values, or why it changes sign down the hole but w e suspect temperature is o n e possibility. The whole-core velocities are normalized to 20°C whereas in situ temperatures increase from about -1.8' at the sea floor to 23.0' at 870 mbsf. The sensitivity of CRP-3 velocities to temperature is not known. A 20-25'C temperature change produces a m i n o r ( < l % ) d e c r e a s e in velocity f o r very l o w - porosity rocks but a 5.0-5.5% velocity change for the very high-porosity sediments studied by Shumway (1958). The difference is attributable to mineral vs.

water compressibilities. It follows that a correction for temperature s h o u l d b e applied to t h e w h o l e c o r e measurements but we have not done so here.

V S P velocities a r e dominated by anomalies of wavelengths greater than 5 0 m. High velocity peaks associated with thin (<20 m) conglomerate beds are n o t r e s o l v e d i n t h e V S P t r a v e l t i m e d a t a . T h e dominance of high-velocity clasts within the core

Migrated Seismic Section

/Â¥7,i; 3 - Cut away seclion NBP9601-89 with synthetic scismogi-a111 I'roin CRP-3: ( a ) near-offset seismic data: ( h ) . tiinc-inigi'alvtl seisinic

data. At least 9 of he most prominent seismic rel'lectors ( p to X ) can he i(le11liI'icil (Io\\I-I 10 7S3.1 I mbsf. (equivalent to 1030 mscc l\\'l hsl) in the SCS and VSP clala.

results in the two-way traveltime to depth conversions d i f f e r i n g by about 2 0 to 3 0 m b e t w e e n 3 0 0 a n d 500 mbsf (see Fig. 2). In general. velocities in CRP-3 are about 3.2 43.6 km/s, apart from the first 80 in of core, where velocity is close to 2.0 km/s. and a 50 in dolerite shear and conglomerate zone from 790 to 823 inbsf, where velocity is greater than 4.5 km/s. Wet bulk density values range from 1.63 to 3.16 Mg 111'~

with the average value b e i n g 2.4 M g m--. T h e intervals of the core corresponding to high velocities are also where measured densities are highest.

Core-plug measurements of velocity vs. pressure s h o w that in situ v e l o c i t i e s a r e h i g h e r t h a n atmospheric-pressure velocities by a percentage that depends on depth. For example at 900 mbsf depth core velocities need to increase by 9% (Jasrard et al., this volume). We found that by making this rebound c o r r e c t i o n to c o r e m e a s u r e m e n t s p r o d u c e d c o r e velocities that were closer to V S P velocities and a time-depth curve that matched the V S P first-arrival traveltime picks and therefore w e used corrected velocities in subsequent analysis.

In summary, data from core measurements, down- h o l e tools, a n d V S P a g r e e v e r y well a n d w e a r e t h e r e f o r e confident in u s i n g t h e c o r e v e l o c i t y measurements that have been corrected for rebound for determining synthetic seismograms where there is an absence of sonic log data.

GENERATION OF SYNTHETIC SEISMOdRAMS Reliable down-hole density and velocity data. as well as seismic impedances z = p Vp, and rel'lection coefficients R ⁼ (z, - zl)/(z7

+

Synthetic seismograms were generated using ;I

reflectivity algorithm (Kennett. 198 1 ) for n o r m a l incidence data. T h e method used here is similar to that reported for CRP-2/2A (Henrys et al., 2000) and CIROS--1 (Bucker et al.. 1998). Input is in terms o f P-wave velocity. density. attenuation, and depth o r thickness of horizontal layers. Attenuation is assumed to be near z e r o in a l l c a s e s p r e s e n t e d h e r e . N o attenuation measurements were made on core samples a n d , u n l e s s Q ( s e i s m i c wave attenuation quality factor) is low. velocity dispersion effects are assumed to b e s m a l l . Velocity a n d d e n s i t y values f o r t h e different layers, apart from the sediments immediately below the seafloor. were derived from measurements o n c o r e s a m p l e s . T h e p r o p e r t i e s of the s e a f l o o r (density, velocity) were determined by trial-and-error matching of the observed and calculated water bottom reflections. To accurately match the sea floor and other reflection events. w e convolved the synthetic impedance function with a source wavelet derived f r o m t h e s e i s m i c d a t a . An estimate of the source wavelet for SCS data can be derived by summing

Fig. 4 - Seismic section for dip line NBP9601-89: (a) uninterpreted: (b). with interpretation of seismic stratigraphic sequence V5 and subsequences U, V. W. X together with major seismic reflector events identified in CRP-2/2A and CRP-3: (c), depth migrated section

traces a l o n g t h e s e a f l o o r h o r i z o n . The convolved synthetic traces were sub,jected to the same processing sequence as the observed single-channel seismic clatii n u 1 s h o t g a t h e r s ( i . e . s a m e f'ilter and g a i n s ) a n d displayed with identical plotting parameters.

RESULTS

CORRELATION O F SEISMIC RI-;FI,ITCTORS WITH CRP-3

A n integrated seismic reflection, VSP, a n d c o r e litliology plot is shown i n figure 2. We have used the whole-core velocity a n d VSI' data to derive time- depth conversion curves to map the seismic reflection section (top) t o depth. These curves are overlaid o n

the t i m e a l i g n e d VSP. O n the bottom. the st;u-ki-il V S P d a t a a r c coinp;n't~(I to t i m e - c o n v e r t c i l c o r e velocity meastirements. I n i ~ l d i t i o ~ i . we h a v e nseil core velocity ini(l density diitii to derive a downholc reflection coefficient profile to associate the seisinir datii to (lie l i t l i o l o g i ~ ~ l logs (l-'ig. 2).

CRP-3 reached a depth of 939.42 mbsf. eqtiivah*n~

to 1030 msec two--way travel time below s e ; ~ level (twt bsl) haset1 o n core velocities and the VSIJ check shot s u r v e y . At least 10 of t h e most p r o m i n ~ i i l seismic reflectors (seismic events) can be identif'it~d down to this depth o r above in tlie S C S a n d VSF data. (o to W ) . Table 1 s i i m m a r i ~ e s the co1-reliition between seismic reflectors from line NB1960 1 ! S O :nid lithostratipapliical units i n Cl<]'-3. Seismic events on S C S d a t a ( 0 . p, S. L I , a n d v ) w e r e d e t e r m i n e d b y correlating the highest positive amplitude wavelet. tlwt

Tab. I

-

Seismic Sequence

Base of seismic sequence U

Base of seismic sequence V

Base of seismic sequence W

Base of seismic sequence X at 823 mbsf

Comments and iiiferences

Sandy Diamictitc; corresponds to a ma.jor rcl'lcclor triiccd to a 100-m wide bench on the seafloor Ma,jor impedance changc between sandy miidstone and conglomerate.

Base of core-physical property unit

I .

A s c r i e s o f t h i n - b e d d e d conglomerates gives a sharp change in velocity c 225 mbsf.

Change in lithology from n i e d i ~ i n - grained sandstone to conglomerate.

Corresponds to a significant velocity change.

Minor velocity increase associated wit11 thin (up to 1.5 in thick) conglomerate beds.

Corresponds to a strong reflector in VSP data and an increase in velocity. Base of core-physical

12.3 mark an increase in core and VSP velocity. Near the base of core-physical property unit 3.

Most likely conglomerates with LSU 13.1 and an increase in core and V S P velocity. A marked change in magnetic susceptibility occurs at 630 mbsf ( b a s e of core-physical property unit 3). A prominent unconformity on strike seismic reflection profiles.

Too of dolerite breccia. Marked bv

>4.5 km/s velocity. B a s e of core-physical property unit 4 . Dolerite breccia and conglomerate 40 ¹¹¹ thick

Notes; LSU - Lithostratigraphical Sub-Unit; * strongest and most persistent reflectors: + not prominent on large scale near trace plot but observed in VSP data.

are lak'rally continuous away from the drill hole. ancl can Ix- related to the cored section. However, detailed linkii;!.cs are uncertain b e c a u s e of 1 ) the low resolution of the seismic signal (wavelength

-

2) iiiicn~tiiinty in the traveltime depth curve (cstim;itcd to be * 10- 15 m). and 3) the surface seismic data are convolved with a complex source wavelet.

WC have used V S P s e i s m o g r a m s . c o r e measurement impedance data (Fig. 2), and changes ill

physical properties that extend over about 20 111 to refin(; o u r correlations. For example, the highest reflection coefficients are encountered in the highest velocity dia~~iictiteslconglomerates. Continuous layers o f these lithologies will yield bright and laterally continuous reflectors.

In t h e s y n t h e t i c s e i s m o g r a m s ( F i g . 3 ) , t h e computed reflections represent only the information sampled in the borehole. Differences between [lie synthetic seismograms and the seismic reflection data (Fig. 3a,h) can arise because the Fresnel zone of the reflection data takes in a larger area than just the borehole (about 300 m radius at 1 S and 20 Hz) and therefore includes reflections generated by rocks and s t r u c t u r e s s u r r o u n d i n g t h e b o r e h o l e . Given the limitations of the data we have been able to establish that all of our 9 seismic events are close to or are associated with a lithostratigraphic sequence boundary (Fig. 2 and Tab. 1). Mapping of seismic data in the region by Cooper et al. (1987), has identified at least 3 major seismic unit boundaries (V3, V4, and V5) identified on the basis of regional unconformities and acoustic velocities - two of these unit boundaries (V3lV4 and V4lV5) were encountered in CRP-212A.

In addition, these units have been further refined. into a t least 1 0 s e i s m i c s e q u e n c e s , based on detailed mapping of SCS data (Henrys et al., 2000, Henrys et al, 1998; Bartek et al, 1996). We have chosen in this p a p e r t o c o n t i n u e t h e c o n v e n t i o n of g r o u p i n g individual seismic units and sequences bounded by seismic events (p-X) into larger sequences (e.g., U-X), b e c a u s e they offer t h e c h a n c e of h i g h l i g h t i n g s i g n i f i c a n t basin-wide u n c o n f o r m i t i e s that c o u l d correspond to climate and tectonic episodes. However, n o t all lithostratigraphic sequences are detected as seismic sequences. Reflection data, for example, are b l i n d to c h a n g e s in m a g n e t i c a n d l o r r a d i o m e t r i c p r o p e r t i e s and w h e r e t h e r e i s n o s i g n i f i c a n t accompanying variation in impedance (Bucker et al., this volume). The correspondence between the seismic s e q u e n c e s , and s e i s m i c r e f l e c t o r e v e n t s , a n d lithostratigraphic units in the CRP3-3 drill site are summarized in table 1. T h e sequences, U-X, are all w i t h i n V 5 of C o o p e r e t a l . ( 1 9 8 7 ) a n d R S S - 1 of Brancolini et al. (1995).

Seismic Sequence V5 (sequence U)

The seismic sequence boundary V4lV5 (base of s e i s m i c s e q u e n c e T) w a s r e c o g n i z e d a t a b o u t 4 4 0 mbsf in CRP-212A a n d i s believed t o b e a

sii:nil'iciint clii~o~~osti~iitigrophic brc;ik i n CRP-212A at W M;> (Wilson et i l l . 2 0 0 0 ) . DiITercntiation into seismic sill>-sctliienccs and chiiriicte~~i~,atio~~ of seismic i i c i c s within V5 was not m;ulc i n Henrys et a l . ( 2 0 0 0 ) . However. S C S seismic data cilong s t r i k e ( N B P O f t O 1-03) and beneath the V4lV5 unconformity s h o w laterally (Iiscontinnous a n d low a m p l i t u d e reflectors. Sequence 0 is correlated to sedimentary seiliiences 19 a n d 25 i n CRP-212A. I n C R P - 3 s e q u e n c e U is correlated with the base of t h i s s e d i m e n t a r y s e q n c n c e 2 anil is identified as a prominent reflector o n VSP data ("p" at 95.48 mbsf).

A marked velocity and density gradient is identified iicross this interval.

Seismic Sequence V5 (sequence V)

Seismic facics that comprise V are characterized o n both strike (north-south) and dip (east-cast) lines by l a y e r s of high a m p l i t u d e discontinuous a n d hummocky sub-horizontal reflectors separated by layers of low amplitude. To the north, the Mackay Sea Valley cuts through Roberts Ridge and exposes these older strata. On NBP9601-92 rcflectors within sequence V appear as foresets and clownlap onto a regional unconformity (reflector "S" a n d to the base of lithostratigraphic Sequences 6 at CRP-3). Also channels cut these layers, the widest being about 2 km and about 70 111 deep (see F i g 5). The character of this sequence is similar to younger sequences S and T (Fig. 4 of Henrys et al.. 2000). In CRP-2/2A,

"S" corresponds to a single lithostratigraphic sequence 1 1 a n d "T" to a n a m a l g a m a t i o n of 5 t h i n lithostratigraphic sequences. These sequences a r e interpreted as glaciomarine deposits each interpreted to record a cycle of glacial advance and retreat with attendant changes in palaeobatl~ymetry (Fielding et al., 2000). Similarly, seismic sequence V is correlated to 4 thick lithostratigrapliic sequences (sequences 3, 4, 5 , a n d 6 ) in the CRP-3 drill site and may also b e i n t e r p r e t e d as c y c l i c a l g l a c i o m a r i n e s t r a t a t h a t correspond to changes in glacio-eustatic sea-level (Fielding et al. this volume; Naish et al. this volume).

T h e vertical resolution of the seismic data are not sufficient to resolve individual sequences recognized rams, in the core but we infer, from synthetic seismo*

that the reflection signal in both CRP-212A and CRP- 3 i s dominated by high velocity and high density diamictiteslconglomerates that occur at the base of depositional cycles (Fielding et a l . , this volume).

Sedimentary sequences 1-7 have been dated by SrISr a n a l y s i s of m o l l u s c s (Lavelle, this volume), and correlation of magnetic polarity stratigraphy (Florindo et al., this volume; Hannah et al., this volume) to be c. 31 Ma.

Seismic Sequence V5 (sequence W)

Seismic sequence W is correlated to 16 thin (less than 28 m thick) lithostratigraphic sequences. T h e b a s e of W ("U") i s m a r k e d a s a h i g h a m p l i t u d e

I-';,?. .1 - Seismic section for strike line NBP9601-92: ( a ) iininterprclcd: (1)). with interpretation of seismic straligrapllic unit V5 and subsequences U. V. W. X together with major seismic reflector events identified i n CRP-3.

reflection corresponding to an erosional unconformity on NBP9601-92 (Fig. 5 ) and corresponds to the base of l i t h o s t r a t i g r a p h i c s e q u e n c e 2 2 at C R P - 3 ( c . 460 mbsf; near the deepest of the interpreted sea- level driven cycles recognizable in the core (Naish et al., this volume). This boundary is also marked by a sharp change in magnetic susceptibility (Cape Roberts Science Team, 2000). Dip profiles showing sequence W also reveal layers with high amplitude reflections, simi1a.r to those observed in the overlying sequence, V. However, on strike profiles, amplitudes are low and t h e internal g e o m e t r y of this s e q u e n c e i s poorly imaged. We are able to trace the base of sequence W approximately 7 k m north of CRP-3, to shot point 1675 011 line NBP9601-92, but further north these deeper sequences could not be confidently mapped.

No clear age model has been established for the core spanning sedimentary sequences deeper that 7 but, seismic sequence W spans a normallreverse polarity change (C12rlC13n) at c. 33 M a (Florindo et al., this volume).

Seismic Sequence V5 (sequence X)

Low amplitude, a l m o s t opaque. seismic facies characterize seismic s e q u e n c e X on both d i p a n d strike reflection profiles. At CRP-3, the sequence X corresponds to the interval c. 460 to 783 mbsf and

c o m p r i s e s a thick succession of seismic;illy monotonous sandstone intervals punctuated by I'i~~ing- upward conglomerate-sandstone units. However, i n this interval synthetic seismogran~s show that weak amplitude reflections dominate. A 4 0 111 thick high- velocity dolerite breccia and conglomerate occurring at the base of the sandstones is calculated t o give a high-amplitude reflection, which is absent o n strike and d i p s e i s m i c profiles. We infer from this that either the dolerite layer is local to the vicinity of' CRP-3 or is extensive and masks a coherent basement wide reflector If the dolerite is local then brccciated rocks may be associated with a shear zone in f a ~ ~ l t e d basement rocks and implies deposition near the fault escarpment. On the other hand. strike line reflection data (Fig. 5) reveals an opaque mound structure near shot point 1900 extending up to 2 km north of CRP-3 that points to a more extensive feature. The base of the s e q u e n c e X is dated at 3 4 M a ( C a p e Roberts Science T e a m , 2000) and is the contact with mid Devonian sandstones of the Beacon Supergroup. The observation that the base of this sequence is a weak reflector, on seismic data away from the drill hole, can a l s o b e e x p l a i n e d by t h e small d i f f e r e n c e i n velocity between Beacon rocks and those sandstones of t h e overlying C e n o z o i c ( s e e Biicker e t a l . this volume and Barrett and Froggatt, 1978).

ISaseiiiciit

Seisniic sequence V7 is acoustic hase~iiei~t for the x g i o n . a n d c o m p r i s e s rocks that pro-dated the Victoriii l.;incl B a s i n . This unit w;is i i ~ ~ e r p r e t e c l by Cooper ct a l . ( 1 9 8 7 ) to include i g n e o u s and metamorpliic rocks o f Ordovician and older age. and probably tlie sedimentary Beacon Supergoup. The top of V7 was infessed to lie at approximately 1500 mbsS

( l .0 scr twt-bsf: see Figure 22 of Cooper et al., 1987) iciieatli tlie western margin of Roberts Ridge. but CRY-3 cored the Beacon Siipei'groiip at 823. l 1 mbsf.

Beacon sti'ata, and the ii~ico~il'ormity corresponding to the Ccno/.oic and older rocks, therefore. lie within s e q u e n c e VS identif'ied b y C o o p e r et a l . ( 1 9 8 7 ) . Indeed. our correlation lias iclentif'icd a reflector about 3 0 0 mscc ( 4 5 0 m ) above the VS/V7 boundary (identified in as the reflector intersecting the sea floor near sliot point 300 in plate 2 of Cooper at al. 1987) as the l)cvo~iian/Ce~iozoic unconf'ormity (see Fig. 5).

Further, we suggest that the reflector identified by Cooper et al. (1987) as representing the top of V7.

m a r k s a n i g n e o u s and m e t a m o r p h i c b a s e m e n t underlying the Beacon strata and close to the top of sheet-like magnetic bodies modeled by Bozzo et al.

( 1 997).

S e i s m i c d a t a collected to d a t e , i n c l u d i n g h i g h - resolution seismic data from NBP9601, are not able t o ackl s i g n i f i c a n t l y new i n f o r m a t i o n a b o u t t h e b a s e m e n t s t r u c t u r e and the C e n o z o i c I B e a c o n

~inconl'onnity since the regioii beneath 7 7 0 mbsf is o b s c u r e d by a s t r o n g multiple. However. depth conversion of the seismic data (Fig. 4 c ) shows the steeping dip with depth trend. observed in CRP-212A.

continues in CRP-3 where dips steepen from c. 11' up to 18' above the conformity truncating the Beacon s t r a t a . North-south striking r e f l e c t i o n p r o f i l e s (NBP9601-92 and Figure 5) appear to show that the Beacon/Cenozoic basement unconformity dips to the e a s t - n o r t h e a s t . i n a g r e e m e n t w i t h d i p m e t e r a n d borehole televiewer data (Jarrard et al.. this volume).

CONCLUSION

Seismic data provides essential spatial context for t h e high-resolution history of ice margin and water d e p t h changes from drill hole data (Henrys et al., 2000). Although the resolution of reflecting horizons in the NBP9601 data is limited to around 2 0 m, in t h e best case. individual horizons could b e traced several kilometers along strike (parallel to the ancient coastline). and many hundreds of meters along dip (normal to the coast). The internal geometry of the strata1 packages has been investigated to establish the lateral continuity of seismic facies a n d s e q u e n c e architecture.

Data from a vertical seismic profile experiment in CRP-3, together with synthetic seismograms and the depth converted seismic section, enable the correlation

of 0 key seismic rel'lection events (p-X) and 4 ma,jor norlli (.~iisicrly (lipping (1117 ( o ISo) seismic .sequences ( U - X j witli s e d i i n c ~ ~ t a r y sequence boundaries and litliostr;itipi';ipliic siih-units tlocumcnted in the core.

Sequence I1 is correlated to sedimentary sequences 19 ind 25 i n ('RP-212A and 1 and 2 i n CRP-3; the base of \vhicli is ; i t 0 0 mbsf. S e i s m i c s e q u e n c e V corresponds to sedimentary sequences 3 to A. U and V arc hoth (liited ;it c. 3 I Ma. Secluence W represents a n ;imal~aiiiation of (lie remaining sedimentary cycles present i n C R P - 3 ( 7 to 2 3 ) to 444 mbsf. T h e age model is not well de~crmined in this interval but W is most likely e;irly Oligocene in age (c. 32-33 Ma). The r c m a i n i n g s e q u e n c e , X . is interpreted as early Oligoccne to liite I-iocene in cigc. The base of X at 9 2 3 inhsf i n ( ' R P - 3 is the base of a 4 0 m thick d o l e r i t e breccia aiul c o n g l o m e r a t e that separates iindcrlyinflDevonian age Beacon sedimcnts froin Ceno/,oic strata and dated at 34 Ma. The sequences, U - X , are all within VS of Cooper et al. (1987) and RSS-l of Brancolini et al. (1995).

The oldest C e n o ~ o i c strata beneath Roberts Ridge, which comprise the upper part of V5 and were cored

i i i CRP-212A and CRP-3, have now been dated as

early Oligoccne (or possibly latest Eocene) to early M i o c e n e . T h e y are t h u s much y o ~ i n g e i - that Palaeocene or Cretaceous, as interpreted by Cooper &

Davey ( 1985) and Cooper et al. ( 1 987).. Furthermore, the lower part of VS. deeper than 823.1 1 mbsf in CRP-3 a n d extending to a depth of approximately 1280 mbsf. are most likely Devonian age Beacon rocks. The V5lV7 boundary of Cooper et al. (1987) lies approximately 340 m below the base of CRP-3 (from the intersection of NBP96-89 and USGS-401) and contrary to what is reported by Hamilton et al.

(2001), probably marks an igneous and metamorphic basement underlying the Devonian Beacon strata and c l o s e to the top of s h e e t - l i k e m a g n e t i c bodies modeled by Bozzo et al. (1997).

Finally, t h e a b s e n c e of any c o r e strata of Paleocene or Cretaceous age, requires a reassessment of seismic data in the western Ross Sea along with a reevaluation of the evolution of the Victoria Land Basin.

A C K N O W L E D G E M E N T S - C R P is a n i n t e r n a t i o n a l p r o j e c t , Alfred-Wegener-Institute f o r P o l a r and M a r i n e Research, DFG (German Science Foundation) and the BGR (Federal Institute for Geosciences and Natural Resources.

Hannover) funded the German part. Seismic reflection data acquisition. in the vicinity of Cape Roberts, were funded by the USA National Science Foundation (grants NSF-OPP- 9220848 and NSF OPP-93 167 10) to Bartek. New Zealand participation is f u n d e d by t h e Foundation f o r Research S c i e n c e a n d T e c h n o l o g y G r a n t C 0 5 8 15. We a l s o a c k n o w l e d g e l o g i s t i c a l s u p p o r t p r o v i d e d by t h e N e w Zealand and USA Antarctic Programs. Scientist and support staff contributing to the C a p e Roberts Project freely gave their advice and opinion and greatly enriched the science.

l .36 S . A . Henrys rl al.

W e a r e g r a t e f u l t o 1'1'ccI 1);ivcy. Aliin C o o p e r , 1,;nir;i O c ~\ni;iirtii.';i. In: /\,K. Cooper it I-'..!, Diivcy (Hcls.). 77ir A i i ~ ( i i ~ ' l i i '

Stintis. C h r i s S o r l i e n , I'ctcr Hiirrctl, ;incl C h r i s l-'ielding I'ol. ( ~ o n t i n e n t a l l\-ldi.y;;~i: (ico/o,q,v & ( / ' ( ~ / n . . w r . \ o f tlic W ( , \ ~ < - I I I t h e i r h e l p in i i - n p i w i n g t h e n ~ a n n s c r i p ~ ;HKI coiitrihu\itig t o / \ \ I \ , \ Sea, /~,'(ii'//i ,S<-i, ,S('r. ~ ' i i ~ ~ ~ ~ i ~ ~ ~ - l ' c ~ ~ j / i ~ ~ (.'cuiid! f i j i ' / ' . ' I I I ' J . I ~ ^i'i

~lliiici'(~l R < ~ , \ o t i / ~ ~ ~ c . ~ . I':ort/~ Sciences .Sci'/c.s, 5 B . I Iousioi~, 'I'(*A;I'., discussions o n Victoria Lain! Biisin seismic stratigraphy.

27 7 6 .

B;~rrett P.J.. Henrys S.A.. 13;irtek 1,,ll.. Urancolini G.. Busetti M , . Davey F.J., H a n n a h M.J.. (V Pync A.[<., 1005. Gcology of the

~ V a r g i i i of the Vicloria I.and Basin off' C a p e R o b c r t s . Southwest R o s s S e a . 111: C o o p e r A . K . . B a r k e r !'.l<'.. &

Brancolini G . (etls.1. Aiitiii'clir R e s e c ~ ~ . ( ~ / i .S(,i.ic~.s. 6 8 . AGU Washington D C. 181-207,

l3;irrett P.J. & 1~rogs;;itt. , ('.l'. 1978. I l e ~ ~ s i t i e s . porositics. ;tiul s e i s m i c v e l o c i t i e s ol' s o m e r o c k s from Victoria L a n d , Antarctica. Alcir Zctiliinii .lina.iiiil of Ci('ologj, a n d Geop/i,~sic,s.

21. 175-187.

Biirtck L . R . . Hcnrys S.A., Aiulerson J.B.. & Barrett P.J.. 1996.

S e i s m i c S t r a t i g r a p h y of M c M u r d o S o u n d . A n t a r c t i c a : Implications tor Glacially Influenced Early Ccno/.oic Eustatic Change. Marine U e o l ~ ~ y ~ . . 130, 79-98,

H o u o E.. Datiiaskc 11.. Caneva G.. Chiappini M,. Ferraccioli I:,.

Gainbetta M . & Mcloni A . . 1997. A high r e s o l u t i o n aeromagnetic survey over proposed drill sites off sl~ore of Cape Roberts in the southwestern Ross Sea (Antarctica). In: Ricci C . A . (ed). The Antiirclic Region: Geolo,qical Evolution and Processes. Siena. Term Antartiea Publication. 1 129- 1 134.

Brancolini G.. Cooper A.K.. & Corcn F., 1995. Seismic Faeies and Glacial History in the Western Ross S e a ( A n t a r c t i c a ) . I n : Coopcr A.K., Barker P.P.. & Brancolini G. (eels.). Antarctic Research Series. 68. AGU Washington D C. 209-233.

Biicker C.J.. H e n r y s S . A . . & Wonik T.. 1998. Revision of the Cenozoic seismic velocity structure of the CIROS- 1 drillhole.

Antarctica. and implications for further d r i l l i n g off C a p e Roberts. Terra Aii/c~i.ticu. 5. 28 1-289.

Biicker C.J.. Jarrard R.D.. Nicssen F. & Wonik T.. 2001. Downhole temperature. radiogenic lic;it production, and heat flow from the C R P - 3 d r i l l h o l e . Victoria Land B a s i n . A n t a r c t i c a . Term Amcirticci. This volume.

Cape Roberts Science Team. 1998. Initial Report on CRP-I. Cape Roberts Project. Antarctica. Terra An/(irtica. 5. 1-187

Cape Roberts Science Team. 1999. Studies from the Cape Roberts Project. Ross Sea. Antarctica. Initial Report 011 CRP-2. Term Aiifni-tica. 6 . 1-173.

Cape Roberts Science Team. 2000. Studies from the Cape Roberts Project. Ross Sea. Antarctica. Initial Report on CRP-3. Terra Antarricc~. 7. 1-209.

Cooper A.K. & Davey F.J., 1985. Episodic rifting of Phanerozoic rocks in the Victoria L,and basin. Western Ross Sea. Antarctica.

Science. 229. 1085- 1087.

C o o p e r A.K.. Davey F.J. & B c h r e n d t J.C.. 1 9 8 7 . S e i s m i c stratigraphy and structure of the Victoria Land Basin. Ross Sea.

. . .

Â¥\iit(irlirii 7. 2 15-220.

I-'rililiiip ( ' . l < . , Naisli 'I'.R.. Wooll'c K . J . & 1,avcllc M , / \ , .'OOll 1 ' _{' ; I (} .' _W analysis and scqiicncc stratis;ral~liy of ('RI' . ' / ? A , McMin'tIo Soiiml. Ant;in:lic;i. '1'i'rra Aii~ir~i-ticti, 7 . 323 3.3s.

i t - l d i i ~ g ('.R,. Naisli T.R. & Woolfc K,J.. 2001. Fiicies : ~ r r l i i l i \ ' l i ~ ~ r of ~ l i r ('RI'-.< drillhole. Victoria 1,and Basin. Antamii'it. 'I'ln'i volillllc.

1l;iniilton R..!.. Sorlicn ('.('.. l.uycndyk B.. 13;irick I,.A.. ^i<i l l c i n v ~ ,?.A., \WK. Tectonic regimes and structiiral li-enil.s i ~ l ' l ' ( ' ; I ] ) < ,

Robn'ts, Antarticii. Terra Aiit(irticii. 5. 26 1-272.

Ilainilton K.S.. l,nycndyk B., Sorlicn C.C.. & Bartek 1 . . A . , .'tMil (\'io/oic tectonics of Hie C a p e R o b e r t s Rift liiisin ;l110 I'r;ins;int;ircfic M o u n t a i n F r o n t . soiithwestcrn k o s s Si1;i<

Aiitiirkicii. '/'f~/oiii('\. 20. 125-342.

1en1-ys S.A.. Biickcr, C.J.. Bartck. L,.R.. Bannister. S.. Nirssrn, I , , , awl Wonik. 'S.. 2000. Correlation of Seismic R c f l c r ~ o r s willl CRP 212A. Victoria Land Basin. Antarctica. fii~rc~ Aiitiirtirn. 7, 22 1-210.

Henrys S.A.. Bartek L.A.. Brancolini G., L ~ ~ y c n d y k B.. Il;miilloi~

R.J.. Sorlien C.C.. & Davey. F.,).. 1998. Seismic .stratigrr:ililiy ol the prc-Qiiartcrnary s t r a t a o f f C a p e R o b e r t s anti tln'ir correl;itio~i with str;it;i cored in the CIROS-1 clrillholi1. ' I ~ I ' I I I Ani(ii~tic(~. 5. 273-279.

Henrys S.A.. Bartck L.R.. Anclei-son J.B.. & Barrett .. l'..! IO0.I.

Seismic Stratigrapliy in McMurdo Sound: Correlation of' l ligh Resolution Data S e t s Trrni An/(~rticci. 1. 373-374.

J. . .. R.D.. Biicker C.J.. Wilson T.J. & Paulsen T.. 2001. Hcdiling (lips from the CRP-3 drillhole, Victoria Land Basin. Ant;lix'tii.".t.

This volume.

Kcnnett B.L.N.. 1981. Elastic wave propagation in strati1'in.l nicilia.

In: Advances ^{~ I I} applied mechanics. Academic Press. Inc. 2 1 . 79- 167.

Naish T . R . B a r r e n P.J. D u n b a r G.P. W o o l f e K.S.. D n n n A.(;..

Henrys S . A . Claps M , . Powell R.D. & Fielding Cl-'. 2001.

Sedimentary cyclicity in C R P drillcore,Victoria Li~ncl Basin.

Antarctica. This volume.

Sheriff R.E.. 1980. Seismic stratigrpahy. 1nternation;il Ilnmaii Resources Development Corporation, Boston, 227p.

Sliumway G.. 1958. Sound velocity versus temperature in water- saturated sediments. Geoph\sics. 23. 494-505

Wilson G.S.. Bohaty. S . M,. Fieldins. C.R.. Floi-inclo. F.. Ilaniiali.

M.J.. Harwoocl. D.M.. Mclntosh. W.C.. Naish. T.R., Roberts.

A.P., Sagnotti. L . . Scherei-. R.. Strong. C.P.. Verosiih. K . [ < . , Villa. G . . Watkins. D . K . . Webb. P.N.. Woolfe. K..!.. 2000.

C h r o n o s t r a t i ~ r a p h y of t h e CRP-212A Drillhole. Ross Sea.

Antarctica. Terra Aniarfica, 7. 647-654.