Montana Tech LibraryDigital Commons @ Montana Tech

Graduate Theses & Non-Theses Student Scholarship

Spring 2016

Processing and Interpretation of Three-Component Vertical Seismic Profile Data, Ross Sea,AntarcticaMeltem AkanMontana Tech of the University of Montana

Follow this and additional works at: http://digitalcommons.mtech.edu/grad_rsch

Part of the Geology Commons, and the Other Engineering Commons

This Thesis is brought to you for free and open access by the Student Scholarship at Digital Commons @ Montana Tech. It has been accepted forinclusion in Graduate Theses & Non-Theses by an authorized administrator of Digital Commons @ Montana Tech. For more information, pleasecontact sjuskiewicz@mtech.edu.

Recommended CitationAkan, Meltem, "Processing and Interpretation of Three-Component Vertical Seismic Profile Data, Ross Sea, Antarctica" (2016).Graduate Theses & Non-Theses. 64.http://digitalcommons.mtech.edu/grad_rsch/64

Processing and Interpretation of Three-Component Vertical

Seismic Profile Data, Ross Sea, Antarctica

by

Meltem Akan

A thesis submitted in partial fulfillment of the

requirements for the degree of

MS Geosciences/Geophysical Engineering

Montana Tech of the University of Montana

2016

ii

ABSTRACT

The Antarctic Geological Drilling Program (ANDRILL) AND-2A drill hole was drilled,

cored and logged in southern McMurdo Sound (SMS) in the western Ross Sea, Antarctica during

the austral summer of 2007. A single near-offset, over-sea ice vertical seismic profile (VSP) was

collected in the AND-2A drill hole as a part of the logging program. The source for VSP data

collection was a Generator Injector (GI) air-gun which was suspended by a cable through a hole

made in the sea ice. The reason for selecting a GI air gun was the minimization of the bubble

pulse effects which are prevalent in all explosive sources placed in the water column.

Both time-depth curves obtained from the measurement of the whole core velocity and

the VSP P-wave are similar. This is an indication that sediments are well consolidated at shallow

depths. The P-wave corridor stack, the surface seismic, and the whole core data synthetic seismic

traces are in a good agreement at the SMS site. Four major seismic reflectors are present in the

corridor stack and they match well with the important regionally extensive unconformities,

which represent major tectonic events and change in climate.

Mode converted S-waves appear to have been generated at the base of the water column.

S-wave splitting analysis of the two horizontal components was used to determine the dominant

fracture orientation from anisotropy. S-wave velocity is the least in a direction perpendicular to

major faults in the region and the slower velocity could be caused by the fractures at depth. The

interpreted subsurface fracture orientations suggest that the minimum horizontal compressive

stress is approximately in the east-west direction and that the maximum horizontal stress is

oriented north-south.

Keywords: VSP, Seismic, Antarctica, ANDRILL, Sea ice, Ross Sea.

iii

Dedication

This thesis is dedicated to my family for their endless encouragement and support in all the

stages of my life.

iv

Acknowledgements

During the preparation of my thesis, I have had support and guidance from remarkable

individuals that I owe them a debt of gratitude.

First and foremost, I would like to express my sincere gratitude to my supervisor, Prof.

Marvin Speece, for allowing me into the geophysics program at Montana Tech. This thesis

project would never have been completed without his continuous academic and research support

and invaluable guidance over the course of my studies.

I thank my thesis committee members Dr. Khalid Miah and Mr. David Reichhardt for

taking time to review this work. Their comments and suggestions enabled me to improve this

thesis.

I would like to express my appreciation to my sponsor, Turkish Petroleum Corporation

(TPAO) for providing scholarship support to have a master’s degree.

I owe many thanks to my friend Brad Rutherford for his friendship and patience. His help

during the editing process was invaluable and unforgettable for me. I would like to acknowledge

my fellow graduate students Mason Porter, Taylor Stipe, and Leif Knatterud for their input and

assistance in the completion of this project.

Finally, I would like to extend my special thanks to my wonderful family for their

continuous support and encouragement. Mom, dad and my siblings, you are wonderful parents

and wonderful friends.

v

Table of Contents

ABSTRACT ............................................................................................................................................ II

KEYWORDS: VSP, SEISMIC, ANTARCTICA, ANDRILL, SEA ICE, ROSS SEA. .......................... II

DEDICATION ............................................................................................................................................ III

ACKNOWLEDGEMENTS ........................................................................................................................... IV

TABLE OF CONTENTS ................................................................................................................................ V

LIST OF TABLES ....................................................................................................................................... VII

LIST OF FIGURES .................................................................................................................................... VIII

1. CHAPTER 1 ....................................................................................................................................... 1

1.1. Introduction ............................................................................................................................. 1

1.2. Objectives ................................................................................................................................ 3

1.3. Previous Related Studies .......................................................................................................... 4

1.4. Southern McMurdo Sound Project......................................................................................... 11

2. CHAPTER 2 ..................................................................................................................................... 16

2.1. Geologic Background ............................................................................................................. 16

2.2. Antarctic Ice ........................................................................................................................... 16

2.3. East Antarctic Ice Sheet ......................................................................................................... 17

2.4. West Antarctic Ice Sheet ........................................................................................................ 17

2.5. Crustal Block Component of East and West Antarctica ......................................................... 18

2.6. Orogenic History of Antarctica .............................................................................................. 20

vi

2.7. Geological Setting of McMurdo Sound .................................................................................. 21

2.8. McMurdo Sound Stratigraphy ............................................................................................... 24

3. CHAPTER 3 ....................................................................................................................................... 27

3.1. VSP Data Acquisition ............................................................................................................. 27

3.1. Vertical Seismic Profile Equipment ........................................................................................ 30

3.2. Multi Component VSP ............................................................................................................ 32

3.3. Converted Wave Theory ........................................................................................................ 33

3.4. Methods ................................................................................................................................. 35

3.5. Processing .............................................................................................................................. 37

3.6. Velocity Profile ....................................................................................................................... 40

3.7. Hodogram Analysis ................................................................................................................ 41

3.8. Wavefield Separation ............................................................................................................ 49

3.9. Predictive Deconvolution ....................................................................................................... 54

3.10. Corridor Stack ........................................................................................................................ 57

3.11. Seismic Well Tie ..................................................................................................................... 60

3.12. Regional seismic correlation and stratigraphic context ........................................................ 63

3.13. Discussion and Results ........................................................................................................... 67

4. CHAPTER 4 ....................................................................................................................................... 68

4.1. Brief Theory of Seismic Anisotropy ........................................................................................ 68

4.2. S- wave Splitting .................................................................................................................... 69

4.3. Methodology ......................................................................................................................... 70

4.4. S-wave Processing ................................................................................................................. 71

4.5. Discussion and Results ........................................................................................................... 78

5. CONCLUSIONS ................................................................................................................................. 82

6. BIBLIOGRAPHY ................................................................................................................................ 84

vii

List of Tables

TABLE 1. SUMMARY OF THE TECHNICAL SPECIFICATIONS OF THE MCMURDO DRILL-SITES (HUSTON ET AL., 2006). 12

TABLE 2. AND-2A VERTICAL SEISMIC PROFILE ACQUISITION PARAMETERS 29

TABLE 3. STANDARD "Q" WIRE LINE BIT SIZES. 29

viii

List of Figures

FIGURE 1. LOCATION MAP OF DRILL-SITES IN SOUTHERN MCMURDO SOUND. 4

FIGURE 2. SUMMARY OF CORE DATA, PLOTTED AGAINST TIME, FOR DRILL-SITES IN THE WESTERN ROSS SEA. 8

FIGURE 3. WATER DEPTH AND CORE RECOVERY AT DRILL-SITES IN SOUTHWESTERN ROSS SEA. 9

FIGURE 4. ANDRILL DRILLING SYSTEM. 10

FIGURE 5. ANDRILL DRILL-SITE LOCATIONS IN MCMURDO SOUND. 13

FIGURE 6. MAP OF THE SEISMIC SURVEY LINES USED TO HELP CHOOSE THE SMS PROJECT AND-2A DRILL-SITE. 15

FIGURE 7. EAST ANTARCTICA, AND WEST ANTARCTICA GLACIER OUTLETS AND MAJOR GEOGRAPHICAL AND TECTONIC SETTINGS . 18

FIGURE 8. LOCATION OF THE MAIN CRUSTAL BLOCKS IN WEST ANTARCTICA. 19

FIGURE 9. LOCATION OF THE MCMURDO SOUND REGION AND VICTORIA LAND BASIN . 23

FIGURE 10. STRATIGRAPHIC UNITS AND DIATOM BASED AGES FOR THEAND-2A BOREHOLE. 26

FIGURE 11. BOREHOLE LININGS FOR THE FIVE PHASES OF DOWNHOLE EXPERIMENTS IN THE SOUTHERN MCMURDO SOUND PROJECT. 28

FIGURE 12. SCHEMATIC SURVEY OF DOWNHOLE LOGGING EXPERIMENTS IN THE SOUTHERN MCMURDO SOUND PROJECT. 28

FIGURE 13. SCHEMATIC VIEW OF THE BUBBLE PULSE EFFECT FROM AN AIR GUN. 31

FIGURE 14. SOURCE SLED, SNOW STREAMER AND DRILLING RIG. 32

FIGURE 15. SCHEMATIC DIAGRAM OF THE 3-C GEOPHONE. 33

FIGURE 16. COMPARISON OF ACQUISITION GEOMETRY BETWEEN CONVERTED AND P-WAVES. 35

FIGURE 17.TYPES OF VSP SURVEYS. 36

FIGURE 18. PROCESSING FLOW FOR THE 2D VSP DATA. 37

FIGURE 19. AND2-A VSP DATA COMPONENTS. 39

FIGURE 20. FIRST BREAK PICKS FOR A PORTION THE Z COMPONENT DATA. 40

FIGURE 21. TRAVEL TIME VERSUS TRUE VERTICAL DEPTH CURVE AND VELOCITY VERSUS TRUE VERTICAL DEPTH. 41

FIGURE 22. SCHEMATIC OF VSP GEOPHONE ORIENTATION PROBLEM. 42

FIGURE 23. P-WAVES AND S-WAVE PARTICLE MOTIONS. 43

FIGURE 24. VISUAL EXPLANATION OF HODOGRAM ANALYSIS AND THE REAL DATA ANALYSIS. 44

FIGURE 25. COMPARISON OF HMAX AND HMIN COMPONENTS. 45

FIGURE 26. VISUAL EXPLANATION OF SECOND HODOGRAM ANALYSIS. 47

FIGURE 27. COMPARISON OF HMAX* AND Z* COMPONENTS. 48

FIGURE 28. WAVEFIELD SEPARATION USING MEDIAN FILTERING. 50

ix

FIGURE 29. WAVEFIELD SEPARATION USING THE F-K FILTERING TO REMOVE DOWNGOING ENERGY. 52

FIGURE 30. COMPARISON OF RESULTS AFTER APPLICATION OF THE MEDIAN AND F-K FILTERS. 53

FIGURE 31. AUTOCORRELATION WINDOW USED FOR DERIVING DECONVOLUTION PARAMETERS. 55

FIGURE 32. UPGOING WAVEFIELD OF VERTICAL COMPONENT BEFORE AND AFTER THE DECONVOLUTION. 56

FIGURE 33. CORRIDOR STACK PROCEDURE OF THE MEDIAN FILTERED, DECONVOLVED UPGOING DATA SET. 58

FIGURE 34. CORRIDOR STACK PROCEDURE OF THE F-K FILTERED, DECONVOLVED UPGOING WAVES. 59

FIGURE 35. SYNTHETIC SEISMOGRAM CREATED FROM SONIC LOG, DENSITY LOG, AND RICKER WAVELET. 61

FIGURE 36. CORRELATION OF THE CORRIDOR STACK AND SYNTHETIC SEISMOGRAM WITH THE SURFACE SEISMIC DATA. 62

FIGURE 37. SEISMIC STRATIGRAPHIC FRAMEWORK FOR THIS STUDY. 65

FIGURE 38. FINAL INTERPRETED GEOLOGIC SECTION OVERLAYING THE DEPTH CONVERTED SEISMIC SECTION. 66

FIGURE 39. VTI AND HTI ANISOTROPY WITH SYMMETRY AXIS. 68

FIGURE 40. S-WAVE SPLITTING PROCESS. 70

FIGURE 41. WAVE MODE CONTRIBUTION FOR THE CASE OF A DEVIATED BOREHOLE. 71

FIGURE 42. P, SV, AND SH WAVES CAPTURED BY A TRIAXIAL GEOPHONE 72

FIGURE 43. S-WAVE SEPARATION USING THE F-K FILTER. 74

FIGURE 44. S-WAVES AFTER WAVEFIELD SEPARATION. 75

FIGURE 45. S-WAVE SEPARATION USING THE F-K FILTER. 76

FIGURE 46. S-WAVES AFTER WAVEFIELD SEPARATION. 77

FIGURE 47. PRINCIPLES OF S-WAVE SPLITTING IN A FRACTURED ROCK 79

FIGURE 48. GEOGRAPHICAL FEATURES AND HORIZONTAL S-WAVE COMPONENT DIRECTIONS. 80

1

1. Chapter 1

1.1. Introduction

Antarctica has a vast intraplate rift framework with broad, deep sedimentary

basins. These give us the best modern site for improving our understanding of the

stratigraphic architecture for sedimentary rift basins which are influenced by glaciers.

The Antarctic Geological Drilling Program (ANDRILL) is an international

scientific program designed to improve our understanding of Antarctica’s role in the

global system. ANDRILL drilling was carried out in the spring of 2006 for the McMurdo

Ice Shelf (MIS) project and in 2007 for the Southern McMurdo Sound (SMS) project.

The MIS and SMS drilling locations were chosen based on geophysical information from

surface seismic surveys collected on ice shelf and sea-ice respectively.

Multichannel seismic reflection data collected from a sea ice platform in Ross Sea

generally exhibits poor quality and has led to poor overall results (Proubasta, 1985; Davy

and Alder, 1989; Horgan and Bannister, 2004). Essentially, the poor-quality data

compromise the overall results even though the common standards for multichannel

marine air-gun seismic surveys were met. Generally, data collected on sea ice in

McMurdo Sound is very poor because of the issues associated with bubble pulse

generation from the seismic source and the impact of reflections from the base of the sea

ice (Horgan and Bannister 2004).

To avoid the source coupling issues and the bubble pulse effects, Horgan and

Bannister (2004) recommended the use of an air gun source instead of explosive sources.

They also identified the potential mechanical and logistical issues related to applying an

air-gun in cold sea ice environment.

2

During the past two decades vertical seismic profiles (VSPs) helped to improve

seismic reflection data processing and interpretation. An interpreted VSP dataset is a

high-resolution subsurface image that creates a link between geology and geophysics.

The VSP procedure involves the recording of a seismic signal generated at the surface of

the earth with geophones secured at various depths in a well (Hardage, 1983, 2001;

Toksöz and Stewart, 1984; Stewart, 2001). With a VSP geometry, both downgoing and

upgoing seismic events can be recorded over time and depth. A VSP is often recorded in

a lower-noise environment than surface seismic profile because the geophones are

located down in the borehole. VSP data improves the structural, stratigraphic, and

lithological interpretation of surface seismic recordings (Hardage, 1983, 2001; Stewart,

2001).

Previous VSP studies attempted to use shear-wave anisotropy to determine the

dominant fracture orientation (Majer et al., 1988; Daley et al., 1988). Our knowledge of

the subsurface properties can be improved through seismic anisotropy analysis in the

form of natural fracture characterization and subsurface stress regime identification.

Limitations of current techniques for characterization of subsurface anisotropy properties

is one of the challenges encountered in incorporating seismic anisotropy studies with

VSPs (Tsvankin, 2001; Helbig and Thomsen, 2005; Tsvankin and Grechka, 2011).

Generally, S-waves are more sensitive than P- waves to anisotropy.

This thesis will investigate the processing and interpretation of three-component

VSP data collected on sea ice from a single near-offset survey in the Ross Sea,

Antarctica. I will combine P-wave and S-wave processing to identify seismic anisotropy

that might be related to fractures and determine stratigraphic boundaries. I will develop a

3

methodology for focusing S-wave particle motion to determine fracture orientation,

which has implications on past and current stress states. The results will be compared

with surface seismic reflection data to identify seismic reflectors, correlate regional

reflectors and gain an understanding of the VSP data in the context of regional geology.

1.2. Objectives

This study has the following four objectives:

1) Document the regional geology in Southern McMurdo Sound region. Even though,

Antarctica plays a key role in the global system, basic geological and geophysical data of

the continent are missing because most of the geologic record is covered by the ice sheet

that covers the continent.

2) Develop processing algorithms for P-waves in near-offset VSP data to image the

surface structure and lithology. The main goal of this thesis is to apply a geophone

rotation process to orient the geophone components correctly, and to separate the

downgoing P-waves mode from S-waves.

3) Interpret S-wave birefringence and characterize fractures by determining the S-wave

window and data rotations.

4) Compare the results to the surface seismic data and interpret them in a regional

context.

4

1.3. Previous Related Studies

The stratigraphic record of the Cenozoic era is insufficiently studied because of

lack of geophysical and geologic data in that region. This problem is exacerbated because

of a succession of ice-grounding events that caused considerable erosion which removed

much of the sediment record.

Strata drilling on the continent of Antarctic started with the Dry Valley Drilling

Project (DVDP) from 1970 to 1974. The four phases of drilling that took place in the

region: the US-led Deep Sea Drilling Program (DSDP) of 1973; USA/New

Zealand/Japan DVDP of 1974; the New Zealand led drilling in McMurdo Sound (1979-

1986) and the multinational Cape Roberts Program (CRP) in northwest McMurdo Sound

in 1997-1999 (Figure 1).

Figure 1. Location map of drill-sites in southern McMurdo Sound. The drill-site locations are black

dots and black triangles locate Neogene volcanic edifices (Fielding et al., 2012).

5

1.3.1. Deep Sea Drilling Project of 1973

Four DSDP holes were bored during Leg 28 in 1973 on the continental shelf of

central Ross Sea (Figure 1). The DSDP site lies on a transect running from the southwest,

close to the Ross Ice Shelf (RIS), toward northeast. The most comprehensive record

originated from the DSDP Site 270, which comprises mudstone, ice-rafted debris and

diamictite. An important finding of the DSDP Leg 28 was that the Antarctic glaciations

traversed an interval of time that was greater than that of the Northern Hemisphere ice

sheets (Hayes and Frakes, 1975).

1.3.2. Dry Valley Drilling Project of 1974

Onshore drilling near McMurdo Sound was done by the DVDP in 1974 (Figure

1). The resulting cores show a fundamentally fjordal sequence. The longest core, DVDP

11 from Taylor Valley, encompasses some of the intervals from late Miocene to

Holocene (McKelvey, 1981; Ishman and Rieck, 1992; Winter and Harwood, 1997).

The facies are dominated by diamictite, conglomerate, and sandstone. The

diamictites have formed near the grounding line of an ice shelf. The conglomerate and

sandstone show the existence of melt water.

After the onshore drilling, DVDP drilled an offshore 65-m hole under the sea ice

offshore at DVDP-15.

The experience obtained from this project led to new drilling techniques with

regard to drilling from land, sea-ice and ice shelf platforms.

1.3.3. New Zealand Drilling of 1979–1986

New Zealand drilling started in 1979 with the McMurdo Sound Sediment and

Tectonic Studies (MSSTS)—1 in 1979 and then continued with the Cenozoic

6

Investigations in the western Ross Sea (CIROS) project in 1986 (Barrett, 1989), offshore

from the stream centerline of Ferrar Glacier. The 702-m-long CIROS-1 was the deepest

borehole until the Cape Roberts Program drilled CRP-3 in 1999.

With 98% recovery, the CIROS-1 core became the standard for subsequent

research drilling and made paleoenvironmental, biostratigraphic and paleomagnetic

examinations possible. Two years prior to this drilling, CIROS-2 had been drilled near

the end of Ferrar Glacier, yielding a Pliocene-Pleistocene record of vacillations of ice

from both the east and the west (Barrett and Hambrey, 1992) (Figure 1). This core has

facies comparable to those in DVDP-11 and apparently shows a fjordal setting. Volcanic

ash layers have permitted radiometric dating.

1.3.4. Cape Roberts Project of 1997–1999

Despite the drilling achievements of the 1970s and 1980s, the basic question of

when Antarctic glaciations started had not been answered. In the mid-1990s a

multinational collaboration attempted to answer this question.

CRP-1 cored a mid-quaternary sequence that included glacigenic sediments like

today’s, and also a fossiliferous carbonate sequence showing much warmer surface

temperatures than today’s. The lower part of the core consists of lower Miocene strata

that included diamictite, sandstone and siltstone, reflecting cyclic variation in ice mass

and sea level.

CRP-2/2A cored lower Oligocene to lower Miocene strata, retrieving the same

chronostratigraphic interval as CIROS-1 70 km the south, with a better preserved

microfossil collection. The mix of biostratigraphic, isotopic and paleomagnetic

7

information identifies a stratigraphic record practically identical to that for middle and

low latitudes today (Harwood et al., 2002).

The CRP-3 cored glaciomarine strata between 3 and 823 meters below the

seafloor (mbsf) have been dated as before Oligocene in age. However, a recent Eocene

age is probable for the lower part of the core, taking into account magnetic polarity data

(Harwood et al., 2002). Sediment aggregation rates were likely high through this section.

CRP delivered a record of paleological and tectonic changes from which we can derive

an understanding of Antarctica in the worldwide framework (Harwood et al., 2002).

During 1999, VSPs were gathered in the CRP drill hole as part of the Cape

Roberts Project (CRP) in order to calibrate the seismic reflection data. An explosive

source was utilized to generate an energy pulse in a water column, and a clamped three-

component geophone was used to collect data. Henrys et al. (2001) asserted that a VSP in

the CRP-3 gave information on velocity and depths as well as P-wave corridor stack in

good agreement with the borehole sediment contacts and surface seismic data. However,

the observed upgoing reflections were complicated by the presence of multifaceted wave

-train.

The VSP profile from the CRP drill hole provides a detailed correlation to the

lithostratigraphy used by Fielding et al. (1998, 2000 and 2001) to create a tectono-

stratigraphic model.

The drill-site information plotted against a time (Figure 2) indicates the existence

of gaps in the Cenozoic record. The most distinctive gap encompasses the middle and

late Miocene Epoch. Some, shorter gaps in the record at individual sites are identified as

8

ice-grounding events. Given the fast pace of sedimentation in glacial environments, the

sediment recovered during drilling represents a small portion of Cenozoic time.

Figure 2. Summary of core data, plotted against time, for drill-sites in the western Ross Sea,

illustrating the relative sediment recovery and disconformities (Harwood et al., 2002).

During the period, spanning from 1975 to 1999, sea ice has been utilized as a

drilling platform in the McMurdo Sound region where seven drill holes were made, as

illustrated in Figure 3. (Previous drilling was undertaken by the Dry Valley Drilling

Project (DVDP) onshore).

9

Figure 3. Water depth and core recovery at drill-sites in southwestern Ross Sea (Harwood et al.,

2002).

Lessons learned during CRP allowed ANDRILL to address specific logistic and

drilling requirements for accomplishing its objectives. A new drilling tool (drill fix and

coring tool, in addition to a riser scheme) was assembled as an improvement over the

drilling equipment used throughout the Cape Roberts Project (Figure 4) (ANDRILL

International Science Proposal, 2003).

The new system had the capacity to perform on both shore-fast sea ice and on an

ice-shelf platform and to retrieve stratigraphic records of up to about 1200 m at water

depths of approximately 1000 m.

10

Figure 4. ANDRILL drilling system. Panel (A) Sea riser styles used from 1975 to 1999 and proposed

for ANDRILL sea ice and ice shelf drilling. From 1975 to 1986 a riser tensioned with an anchor mass

and integral steel pressure vessel buoyancy was used. From 1997-1999 the CRP project also used

buoyancy modules to reduce the loads on the sea ice. Both risers are self-supporting and float free of

the sea ice. The third style uses the ice shelf for support after anchoring into the sea floor (ANDRILL

International Science Proposal, 2003). Panel (B) is the new drill rig that was required to achieve the

proposed ANDRILL objectives. This system allowed quicker coring rates and deep drilling, which

was not possible with previous equipment (Huston et al., 2006).

11

1.4. Southern McMurdo Sound (SMS) Project

ANDRILL drilling occurred in two phases: the McMurdo Ice Shelf (MIS) phase

and the Southern McMurdo Sound (SMS) phase (Naish et al., 2007). Lithologies and

depositional environments for the MIS and SMS phases are summarized in Table 1. The

results from the project offer valuable insight into two major issues:

1) The development and behavior of the cryospheric system of Antarctica.

2) The evolution and timing of the key tectonic episodes in the Antarctic region

and sedimentary basin stratigraphic development.

In the MIS phase, a 1285-meter-long hole was drilled, AND-1B, in the northern

corner of the Ross Ice Shelf, in the austral summer 2006–2007 (Figure 5). In the SMS

phase, AND-2A, an 1138-meter-long hole, was drilled during the summer of 2007–2008

(Florindo et al., 2008). The AND-1B and AND-2A boreholes are situated at the junction

of the Transantarctic Mountains (TAM), the Victoria Land Basin (VLB) and the Erebus

Volcanic Province (Figure 5). The AND-1B borehole is located in the Erebus Volcanic

Province. The AND-2A borehole is situated between the TAM to the west and the Ross

Island to the east.

12

Table 1. Summary of the lithologies and depositional environments of the McMurdo drill-sites

(Huston et al., 2006).

Facies Description Interpretation

1 Diatomite , opal- bearing or calcareous, fossil-rich

mudstone, typically intensively and diversely

bioturbated, little if any dispersed gravel

Open marine shelfal conditions (60-100 m

water depth), minimal ice influence.

2 Siltstone to very fine gravel and calcareous fossils,

soft- sediment deformation structures

Open marine shelfal conditions (60-100m

water depth), deposition from suspension

settling, distal ice influence

3 Interstratified siltstone and very fine-grained

sandstone, bioturbated and/or stratified with ripple-

scale structures, typically restricted and modest-

intensity bioturbation, dispersed calcareous fossils,

soft-sediment deformation.

Open marine to restricted coastal settings

(10-80 m water depth), deposition from

suspension settling, gentle currents and

waves, and minor ice rafting.

4 Stratified, fine to coarse-grained sandstone, dispersed

to locally concentrated small gravel, typically well

sorted, with flat lamination, ripple cross-stratification.

Open marine shore face to upper offshore

(10-40 m water depth), minor ice rafting of

gravel.

5 Muddy fine-grained sandstone to sandy siltstone with

dispersed gravel, typically soft-sediment-deformed or

unstratified, locally fossiliferous, rarely bioturbated.

Open marine conditions (60-100 m water

depth) affected by hemipelagic fallout and

supply of sediment from ice rafting

6 Delicately stratified, interlaminated fine- and coarse-

grained siltstone, fine-grained sandstone and

diamictite, dispersed gravel including diamictitie

clasts, some lamination rhythmic at sub-sediment and

brittle deformation.

Distal glaciomarine conditions affected by

hemipelagic fallout, bottom currents and

ice-rafting

7 Stratified diamictite, generally clast-poor, muddy to

sandy matrix, commonly fossiliferous.

Open, proximal glaciomarine conditions,

in distal portions of grounding-line fans.

8 Massive diamictite, clast-poor to clast-rich and

typically sandy matrix, some alignment of clast long-

axes, shear structures including boxwork fracturing

and brecciation.

Ice proximal glaciomarine (grounding-line

fan) to subglacial environments.

9 Interbedded conglomerate and sandstone, thin units

typically interbedded with other coarse-grained facies,

locally fossiliferous.

Ice-proximal environments in the presence

of meltwater, various possible depositional

settings.

10 Vesicular, porphyritic, basaltic lava Primary sheet flows erupted from a

proximal vent or fissure.

11 Volcanistic breccias, monomict, composed of basaltic

clasts in basaltic matrix.

Autobrecciation of subaerially erupted

lavas

12 Pyroclastic lapilli tuff, composed of glassy juvenile

clasts of coarse ash to lapilli size

Settling of pyroclasts through the water

column

13 Volcanic sedimentary breccias and sandstones,

comprising texturally unmodified to modified juvenile

volcanic clasts and grains, local ripple cross-

laminaton.

Reworking of volcanistic deposits in

shallow marine waters

13

Figure 5. ANDRILL drill-site locations in McMurdo Sound. As a part of the ANDRILL SMS Project,

successful downhole experiments were performed at the 1138-meter-deep AND-2A drill location

situated in McMurdo Sound. Both AND-2A and the previous AND-1B are positioned in the western

section of the Antarctic Rift base, known as the VLB, adjoining the TAM (Fielding et al., 2011).

To select a drill-site on a stable sea-ice platform, additional seismic data were

obtained in 2005. These new data extended the targeted Miocene section that was first

identified in open-water marine seismic reflection data, southwards into a land-fast sea-

ice region where the annual breakout risk was minimal (Betterley et al., 2007; Speece et

al., 2009). The new seismic lines were used to map regional seismic reflectors to enable

drill-site selection.

The SMS Project was intended to recover a stratigraphic section that included a

substantial portion of lower to upper Miocene strata. The target strata were observed in

the PD-90-46 seismic lines (Figure 6) and mapped to the SMS drilling sites using on-sea

14

ice seismic reflection data that were recorded in 2005. ANDRILL Thunder Sled 2005

Line One (ATS-05-01)-on-sea-ice, multichannel seismic data—was tied into PD-90-46.

PD-90-46 is part of an existing marine seismic data grid gathered in McMurdo Sound

(Anderson et al., 1992; Betterly et al., 2007) (Figure 6). Subsequently; a sub-network of

seismic surveys of southern McMurdo Sound was created using ATS-05-01, ATS-05-02,

ONH-08-01, ONH-08-02, and PD-90-46 (Pekar et al., 2013).

15

Figure 6. Map of the seismic survey lines used to help choose the SMS Project AND-2A drill-site. (A) SMS project seismic surveys ATS-05-01 and ATS-

05-02 (red lines) were tied to the PD-90-46 (black line) (Betterly et al., 2007). (B) Fence diagram of PD-90-46 ATS-05-01, and ATS-05-02. The spacing

found between CMP locations on ATS-05-01 and ATS-05-02 has a value of 12.5 m and shot locations on PD-90-46 have a spacing of 45 m. Each profile

displays 1.5 s of strata and 200 ms automatic gain control has been applied to these data. The interpreted seismic reflectors in colors represent

disconformities that record glaciers’ advance and retreat (Harwood et al., 2004).

16

2. Chapter 2

2.1. Geologic Background

Antarctica’s geological and tectonic evolution is less known compared to the

other continents. Our knowledge about the global system will continue to remain

incomplete primarily because of this unless sufficient information is obtained about

tectonic, paleoclimatic and paleoenvironmental events in the high southern latitudes.

Targets under land and ice have become accessible through the ANDRILL program and

its ability to provide a regular platform as well as great penetration depth and core

percentage recovery.

Three geologic processes have dominated the formation of the continent:

tectonics, the accumulation and dispersion of ice, and volcanism. Antarctica began its

development with the formation of the East Antarctic Craton in the early Precambrian

and continued until the end of the Nimrod Orogeny (Goodge et al., 2010). East and West

Antarctica are the two major continental blocks of the continent of Antarctic. The

Transantarctic Mountains (TAM) extend approximately 4500 kilometers from north to

south (Figure 7), creating a structural division between the eastern and western portions

of Antarctica (Anderson, 1999).

2.2. Antarctic Ice

As the Antarctic ice streams move toward the sea, they become part of floating

ice shelves and break terrestrial contact. The point at which the ice loses contact is known

as the grounding line. The grounding line not only changes positions between glacial

maximums and minimums, but can also change seasonally as well. The largest changes in

17

ice shelves occur at the ice front and are caused by the constant barrage of waves. The

calving of icebergs caused by a mass balance deficiency occurs predominantly during the

austral summer, when seasonal sea ice is no longer present and the thermal limit for the

ice shelf is exceeded by the local climate (van der Veen, 2002).

2.3. East Antarctic Ice Sheet

The East Antarctic Ice Sheet (EAIS) is the larger of two ice sheets. It has an

average thickness of 3 km and covers about 2/3 of the continent. Although considered a

terrestrial ice sheet because it is predominantly grounded above sea level, some sub-sea-

level basins are filled in its interior. Drainage from the sheet is directed downward and

toward the coast, making it primary divergent. Exceptions exist near the Amery Ice Shelf

and the TAM. The divergent pattern associated with the ice drainage tends to decrease

the velocity of the ice as it reaches the coast.

2.4. West Antarctic Ice Sheet

Grounded approximately 1 km below sea level, the West Antarctic Ice Sheet

(WAIS) is a marine ice sheet. The drainage from this sheet is predominantly convergent.

Therefore, ice velocities increase as the flows make their way toward the sea. High ice

velocities suggest multiple processes are contributing to ice flow, including internal

deformation, basal sliding and ground deformation. Internal deformation not only allows

ice to flow in a more fluid manner, but also increases surface area contact within the flow

itself, creating more friction and therefore more heat. Anderson (1999) contends that

friction is greatest near the center of the flow, causing the bed beneath to thaw while the

bed adjacent to the stream divide remains frozen. Thawing of the ice stream bed means

18

that sediments are looser and can readily be removed, creating advantageous conditions

for ground deformation. Subsequent ice streams can reach up to five kilometers in width,

and are responsible for close to 90% of the ice leaving WAIS (Anderson, 1999).

Figure 7. East Antarctica, and West Antarctica glacier outlets and major geographical and tectonic

settings (Geological Survey (U.S.), 2007).

2.5. Crustal Blocks of East and West Antarctica

East Antarctica is formed mostly by the East Antarctic craton, which is one of the

oldest cratonic elements on earth. East Antarctica formed in the early Archean and

reached maturity by the middle Proterozoic (Anderson, 1999). The rocks that have been

found in East Antarctica are moderately older than those of West Antarctica (Siegert,

2008). The East Antarctic shield is comparable in lithology and structure to the shields of

South Africa, India, Madagascar, Sri Lanka, and Australia: all were parts of Gondwana

before their separation around 200 million years ago (Yatsenko, 2008) .

19

West Antarctica is made up of four main crustal blocks: the Ellsworth-Whitmore

Mountains, Antarctic Peninsula, Marie Byrd Land, and Thurston Island blocks (Figure 8).

Figure 8. Location of the main crustal blocks in West Antarctica. Crustal block names: AP=

Antarctic Peninsula; TI= Thurston Island; EWM= Ellsworth-Whitmore Mountains; WARS= West

Antarctic Rift System; MBL =Marie Byrd Land (Dalziel and Lawver, 2011).

The Ellsworth-Whitmore Mountains are usually considered to be one crustal

structure. The Ellsworth-Whitmore block is rotated nearly orthogonal to the contiguous

stratigraphic sequence of the TAM. Strike-slip faulting is believed to account for the

rotated block (Anderson, 1999). Separated from the Ellsworth-Whitmore Mountains by

the Byrd Subglacial Basin, Marie Byrd Land is located east of the Ross Sea. The Marie

Byrd Land block was fragmented during its migration from the Gondwana configuration.

20

ANDRILL has focused on the McMurdo Sound located in the western Ross Sea.

This morphological depressed region’s major structural features, the Ross Embayment, is

along the boundary zone between east and west Anatrctica .

2.6. Orogenic History of Antarctica

In the Precambrian, a series of orogenic events gave rise to the modern East

Antarctic craton. Beginning around 4000 Ma, the Napier Orogeny gave rise to the

cratonic nucleus. Subsequent orogenies added to the nucleus throughout the Archean: the

Rayner Orogeny (~3500 Ma) and the Humboldt Orogeny (~3000 Ma).

Events of deposition and submarine volcanism continued along the proto-Pacific

margin of East Antarctica until 1000 Ma. The later Ruker and Nimrod Orogenies were

characterized by metamorphism and magnetism in the deep crust that are attributed to

crustal collisions that led to craton thickening (Goodage et al., 2001). Continuing

metamorphism ended the Nimrod Orogeny, leaving a reworked proto-Pacific margin in

East Antarctica.

The central TAM has a convergent tectonic history. The ancestral TAM formed

as older sequences were overturned during deformation, resulting in isoclinal and

overturned folds.

The East Antarctic craton caused orogenic development at the end of the

Precambrian. The Paleozoic marked the beginning of orogenic events in West Antarctica

as the West Antarctic crustal blocks moved closer to their current positions. During the

early Cambrian the ancient TAM underwent significant erosion while an epicratonic sea

spread across the majority of West Antarctica (Anderson, 1999). Locked within the

present TAM, are shallow marine clastics, platform carbonates, deep-water turbidities

21

and volcanoclastic strata. With the onset of the Ross Sea Orogeny during the early

Paleozoic, metamorphism and plutonism dominated the continent and aided in the

formation of the present TAM. The early Ross Sea Orogeny is marked by the extension

along the TAM of marginal phases of bimodal magmatism. As the orogeny progressed,

tectonic events intensified, leaving sedimentary sequences metamorphosed and intruded

by granitic batholiths in areas of deformation. A westward dipping subduction zone on

the paleo-Pacific border of East Antarctica explains similarities in the geochemical

analysis of Pacific batholiths in areas of deformation (Anderson, 1999).

Extension in the Ross Embayment is the most likely cause for thinning crust and

resulting subsidence. Uplift in TAM is poorly constrained but is believed to have its

beginnings in the early Cenozoic (~55-50 Ma). Erratic block faulting along the eastern

margin of TAM has led to a complex step-fault system. Most major Cenozoic mountain

ranges on Earth were created by the volcanism of thrusting and folding, or andesitic

volcanism, but the TAM are the result of vertical crust movement and fault block tilt

(Fitzgerald et al., 1986). Orogenic events of the late Cretaceous and Cenozoic were, for

the most part, confined to the Ross Embayment and the TAM region of Victoria Land.

2.7. Geological Setting of McMurdo Sound

The sediments of the Ross Embayment are extremely important to our

understanding of the relationships among the climatic, tectonic, depositional, and eustatic

changes that took place during Cenozoic times. The Ross Embayment is characterized as

a failed Cenozoic rift system and rifting that occurred in the Ross Embayment caused

faulting to occur and therefore created accommodation space for sediment deposition

(Cooper et al., 1987; Davey and Brancolini, 1995; Stagpoole, 2004).

22

McMurdo Sound is a small basin in the southwestern portion of the Ross Sea,

located at approximately 77°45′S and 165°16′E. The McMurdo Sound region is at the

junction of the West Antarctic Rift system (Figure 9A).The ANDRILL McMurdo Sound

Project was inspired by the westward location of the VLB, which is filled with a large

amount of Neocene–Recent volcanism from the Erebus Volcanic Province (Kyle, 1990),

which includes Neogene Mounts Morning, Bird, Discovery, and Terror as well as the

active volcanos of Erebus (Figure 9B).

The VLB, situated northwest of the AND-2A drill location, contains a ~14-

kilometer-thick sequence of the Mesozoic and Cenozoic sediment supplied from the

TAM after the volcanic detritus from the Erebus Volcanic Province (Kyle 1990; Kyle et

al., 1992; Cooper et al., 2007; Martin et al., 2010; Nyland et al., 2013).

The sedimentation that occurred in the VLB was controlled by the eastward tilting

fault block that was created during past rifting events; consequently, sediments in the

VLB dip and thicken towards the east (Figure 9). Sediments located on the western edge

of the VLB were targeted for previous drilling (AND-2A; CIROS-1; CRP-1, 2, 3;

MSSTS-1; and DVDP-15) to recover older sediments that are shallowest along this

western margin.

Figure 9 shows a schematic view of the tectonic setting of the VLB. VLB is a

semi-graben which is titled downwards to the east and restricted by a huge down towards

the west border fault. The sediment fills dips and thickness towards the east. The front

fault zone of the TAM, which is located at the western side of the basin, forms the border

between VLB and the uplifted TAM. As recorded during the Cape Roberts drilling, the

TAM margin of VLB was a ramp margin with a gentle slope during the Oligocene–

23

Miocene period, when the VLB was still in a developmental phase (ANDRILL

International Science Proposal, 2003).

At the Discovery Accommodation Zone (DAZ) (Figure 9), where the rift-flank

steps are around 100 km to the west of the TAM, the VLB either ends or is equalized, and

the volcanism of the Late Cenozoic is focused (Wilson, 1999). The depositional

accommodation space provided by additional weight of the volcanos flexes the crust and

allows for sediment accumulation in flexural-moat basins (Figure 9).

Figure 9. Location of The McMurdo Sound region and Victoria Land Basin (VLB). (A) The

McMurdo Sound region is indicated with red box and Victoria Land Basin is indicated with blue box

(ANDRILL International Science Proposal, 2003). (B)The VLB with its major geologic features

(Harwood et al., 2006).

The mixture of rubble derived from the high elevations of TAM spread out in the

open space of the VLB (Zattin et al., 2012). This space also shielded the silt coming from

the erosive impacts of the ice masses that uprooted the Antarctic ice record (Zattin et al.,

2012). Thus, Upper Eocene silts are the most recognized post-Paleozoic residue

recovered to date from the stratigraphic west edge using a CIROS-1 drill hub (Wilson et

al., 1998; Davey et al., 2001). Additionally, the Eocene strata partially cover the

24

Devonian silt from the Taylor Group (Davey et al. 2001; Paulsen et al. 2011). Thus, the

AND-2A drill-site is in position to receive sediments from a TAM base connecting the

ice sheet of EAIS, the Ferrar Glacier and the Mulock Glacier and from the volcanic focal

point of the Late Neogene Erebus Volcanic Area around Southern McMurdo Sound

(Panter et al., 2008 ; Sandroni and Talarico, 2011). To the south, the Byrd Group could

provide carbonate-based rocks from the Pleistocene (Stump et al., 2006), which are

plentiful within the lowered Byrd Glaciers (Licht et al., 2005).

2.8. McMurdo Sound Stratigraphy

The stratigraphic and tectonic arrangement of the McMurdo Sound is well

understood because of the integrated core and seismic data obtained over the last thirty

years (Naish et al., 2007). The character of the stratigraphic succession in the southern

VLB varies considerably from the basal, uppermost Eocene strata upward.

The (AND-2A) drill core recovered numerous unique intervals truncated by

disconformities, including:

1) The lower Miocene interval, ranging between 1138.54 and 800 mbsf.

2) The middle to early Miocene interval with a thickness of 600 m and ranging

between 800 and 223 mbsf (meters below sea floor) , inclusive of an extended section

that includes two Miocene climate optimums, that are truncated by a disconformity .

3) A recent Miocene interval between 223 and 0 mbsf, which is related to the

AND-1B core.

Mt Erebus’ rapid volcanic loading caused the VLB to deepen at c. 1.5 Ma while

shallow sea sediment accumulated in the lower section for AND-2A. The middle to lower

25

Miocene strata indicates repetitive lithological changes related to climatic fluctuations,

glacial proximity, and fluctuating sea level along the shallow marine coastline bordering

the Trans-Antarctic ranges. Fossils indicate non-polar climatic conditions resembling

similar ones that experienced in the south of New Zealand and high turbidity along the

coast and significant sedimentary discharge emanating from river run-offs. Figure 10

shows the AND-2A borehole lithology (Fielding et al., 2008) and ages (Taviani et al.,

2008).

26

Figure 10. Stratigraphic units and diatom based ages for theAND-2A borehole (Fielding et al., 2008; Taviani et

al., 2008).

27

3. CHAPTER 3

3.1. VSP Data Acquisition

On November 2007, an on-sea ice VSP survey at the SMS AND-2A drill hole was

conducted during phases II and V of drilling (Figures 11 and 12). The acquisition

parameters used for the near-offset over-sea ice VSP are listed in Table 2. Equipment

used in the VSP data acquisition is explained in the next section.

Core samples were collected using diamond coring rigs and wireline systems.

During this process, the drilling rods remained in the borehole while the core barrel was

brought back up to the surface with the line wire. The five commonly used tube sizes for

the wireline are listed in Table 3. Tube size selection is based on requested rock core

diameter and the depth.

PQ, HQ, and NQ drill string were used for this experiment. The PQ base was at

229.24 mbsf but the VSP data were collected using dual casings (PQ and HQ) during the

phase II. The VSP was repeated for the PQ part of the borehole in phase V after

completion of downhole and hydraulic fracture logging at the bottom of the hole and after

pulling out the HQ casing. In Figure 11 five stages of drilling with casing size and depth

are shown. Because of a sand layer at ~340 mbsf, the initial plan for an intermediate

logging run from a PQ base at ~200 to ~700 mbsf (assumed to be the HQ base) was

abandoned because of the complex conditions of the drill hole. After the HQ drilling

ended at 1008 mbsf because of the adverse condition of the hole, an approximate 130 m

extra NQ diameter hole was drilled to 1138 mbsf. After completion of the NQ hole, the

28

HQ drill string was replaced and logging was restarted within the PQ casing (Wonik et

al., 2008). Figure 12 shows different logging tools in these five phases of logging.

Figure 11. Borehole linings for the five phases of downhole experiments in the Southern McMurdo

Sound Project, AND-2A borehole (Wonik et al., 2007).

Figure 12. Schematic survey of downhole logging experiments in the Southern McMurdo Sound

Project, AND-2A borehole (Wonik et al., 2007).

29

Table 2. AND-2A Vertical Seismic Profile Acquisition Parameters

Table 3. Standard "Q" wire line bit sizes.

Size Hole (outside) diameter, mm Core (inside) diameter, mm

AQ 48 27

BQ 60 36.5

NQ 75.7 47.6

HQ 96 63.5

PQ 122.6 85

Survey type Near-offset VSP

Offset 70 m in the structural up-dip direction from the

drill-site

Source Type Seismic Systems 210- in3 generator-injector air

gun

Source Mode Harmonic, generator and injector volumes of 105

in3 each

Air gun Position

18 m below the ice level

Air Gun Pressure

2000 psi

Top Depth 386m

Bottom Depth 1388m

Receiver Spacing 6 m

Air Gun Pressure 3.4 x 106 pa (2000 psi )

Sampling Rate 2 ms

Record Length 3 s

Borehole Inclination ~ 2.25 °.

30

3.1. Vertical Seismic Profile Equipment

3.1.1. The Bubble Pulse Effect: Background

The main role of any seismic source is to infuse acoustic energy through the earth.

This energy is often provided by an explosive source. Unfortunately, inside a water

column the explosion also forms secondary shock waves because air bubbles that form as

gases are released by the source that oscillate, growing and contracting, until the energy

in the bubble is disseminated (Figure 13).

The bubble effect can be observed in the final seismic cross section because of the

repeating reflections that they cause. Understanding the stratigraphy underneath the

surface seismic lines is not possible unless the impact of the bubble pulse is minimized,

either during the acquisition or the processing of the data. Despite the use of a

deconvolution approach, reducing or eliminating the bubble effect with data processing is

difficult.

The bubble effect created by an air-gun may lead to significant problems, if a

single gun is used as the seismic source. The bubble can be damped by using a generator–

injector (GI) air gun that contains two air chambers with compressed air in both chambers

(Figure13). A single GI air gun was used as the seismic source for the AND-2A VSP

experiment, whereby a secondary bubble was injected into a bubble first released by the

generator chamber of the GI gun to cancel the oscillation of the first bubble.

31

Figure 13. Schematic view of the bubble pulse effect from an air gun. The air pressure inside this

large air bubble released into the water is greater than the surrounding water pressure initially, and

therefore the bubble starts to expand. When the bubble reaches its maximum volume as a result of

this expansion, the air pressure inside the bubble is lower than the pressure of the surrounding

water, and because of the surrounding high pressure the bubble suddenly shrinks. This contraction

emits a signal back into the water as if a second acoustic source exists. However, since this second

signal occurs during the contraction of the bubble, it has reverse polarity compared to the first

signal. Following this, the bubble begins to expand again as a result of the increasing internal

pressure caused by the shrinking volume, and emits another signal into the water (Dondur, 2009).

Two separate recording systems were utilized to obtain the necessary information.

The main system located at the well head recorded the VSP. The geophones in the

borehole measured the particle motion in three orthogonal directions (a vertical in

addition to two horizontal components) to record both P-wave and S-wave modes. A

second recording system was housed in the source sled and recorded surface geophones.

The surface geophones were used for quality control of timing errors (Figure 14). The

surface system used six 30 Hz gimbaled vertically-oriented geophones.

A separate room for the sled source was modified to have air tanks fed by gas-

powered compressors. The tanks supplied the air to the GI air gun source which was

suspended from a winch with a rotating arm.

32

Figure 14. Source sled, snow streamer and drilling rig. (A) Generator-Injector (GI) air gun was

located in the recording room of the source sled. (B) Gimbaled geophone and the snow streamer

cable. (C) Outside view of the drilling rig in McMurdo Station.

3.2. Multi Component VSP

Multicomponent data analysis is undertaken using a rectangular polarization

coordinate system. Thus, three traces of seismic components—Sx, Sy, and Sz—are

recorded at all receiver stations.

During data acquisition, the receiver in the borehole picks up signals with two

horizontal components and a single vertical component. The horizontal components are

termed the X and Y components, and the vertical one is the Z component. A right-handed

coordinate system is used to describe data acquisition characteristics in multicomponent

2-D seismic data acquisition (Figure 15).

33

Figure 15. Schematic diagram of the 3-C geophone. The X, Y and Z directions indicate the

orientation of the VSP tool’s actual geophones (Stewart et al., 2001).

3.3. Converted Wave Theory

Seismic reflection surveys utilizing compressional P-waves have been performed

in hydrocarbon investigations for a long time. P-wave data are generally accessible and

high-quality, making the utilization of P-wave technology predominant in seismic

investigation (Sheriff and Geldart, 1982).

In the past, P-waves were believed to be adequate for imaging fractures and 3D

channels, creating seismic images of reservoirs, and so forth. However, S-waves travel

through rocks too. S-wave, have the capacity to improve subsurface characterization and

improve detection of anisotropy. S-waves are more sensitive to anisotropy than P-waves

and, thus can help in the estimation of anisotropic parameters. If P-wave and S-wave are

studied together, then features related to anisotropy and stratigraphic boundaries can be

identified more precisely than with just pure P-wave data.

Marine environments are normally associated with petroleum reservoirs, because

petroleum source rocks are derived from ancient marine life. Modern marine seismic

survey techniques enable efficient data collection, because air guns allow for inexpensive

rapid shooting while the survey vessel remains in constant motion.

34

In the marine environment, since seawater does not support S-waves, shear

sources are basically nonexistent. Therefore, S-wave data are typically gathered utilizing

a P-wave source. Energy goes down from the source as a P-wave, experiences mode

conversion at an interface, and comes back to the receiver in the borehole as an S-wave.

The ocean bottom functions as a conversion interface where compressional wave energy

is converted to S-wave energy. S-waves that have been produced through the

transformation of P-waves in the subsurface are called “converted waves” (C-waves).

Offset VSP data provide converted wave data with which to calibrate P-wave

techniques to detect vertical fracture orientation and map hydrocarbon saturation and the

oil-water contact. Converted waves allow for a richer subsurface characterization by

providing S-wave properties along with P-wave properties.

Figure 16 shows the illustration of mode conversion. P-wave energy, produced by

a P-wave source, travels via the ground and at all acoustic impedance discontinuities

partial energy is transformed into shear wave (S-wave) energy. The focus of converted

wave processing is on analyzing this upcoming S-wave energy.

35

Figure 16. Comparison of acquisition geometry between converted and P-waves. When working with

converted waves, the reflection point is considered the conversion point, where the P energy is

converted into S- wave energy. This reflection point is different from midpoint between the source

and receiver (x/2) and is depth dependent (Converted Wave Seismic Surveys, 2016).

Multicomponent seismic data with converted S-waves provides rock, fluid and

lithological information and provides for better characterization of subsurface anisotropy

than that provide by P-wave data alone (Hardage et al., 2011). Azimuthal S-wave

anisotropy (birefringence/vertical S-wave splitting) analysis has generated significant

interest (Sun and Jones, 1993).

3.4. VSP Methods

A VSP is ideally suited to determine the geological properties of a targeted

horizon and acquire seismic images of the subsurface. VSPs help the interpretation of

surface seismic reflection data.

The important features of vertical seismic profiling are the following:

- The capacity to create a seismic trace at a borehole that is more reliable than a

seismogram synthesized from density and sonic curves.

36

- The ability to aid correlation of surface seismic sections with borehole logs.

- The shorter transit distances in a VSP causes higher resolution when compared

with surface seismic data.

- The capability to record a reflector signal that is not dependent on surface

reception.

- The possibility of corridor stacking: a procedure that stacks all upgoing events to

find the major reflectors and correlate with the surface seismic data.

In general, a VSP survey uses receivers that are down in a borehole and sources

are on the surface. A zero-offset VSP survey is achieved when the sources are positioned

at a short distance from the well relative to the depth of investigation. If a VSP survey has

several sources at multiple offsets from the borehole, it is a walkaway VSP survey.

Figure 17 shows a simple illustration of a VSP survey.

Figure 17.Types of VSP surveys : (A) Zero-offset VSP (B) Offset VSP and (C) Walkaway VSP. Blue

circles indicate sources, and red circles are geophone positions. Yellow lines define downgoing ray

paths, and green lines indicate upgoing ray paths (Campbell et al., 2005).

37

3.5. Processing

The processing techniques and an explanation of the major processing steps that

were used are discussed in the subsection. Modifications made to the raw data during the

processing procedure include geometry definition, band-pass filtering, trace editing,

component rotations, amplitude recovery, deconvolution and corridor stacking. The

processing flow used in this thesis is shown in Figure 18.

Figure 18. Processing flow for the 2D VSP data. Processing of the data included trace display and

editing. The flow begins in the top left corner of the figure and ends in the bottom right corner.

38

The geometry of the data is entered into the program manually from the field

observer report. For instance, the depth of the receiver is typically gotten from field

observer reports, and then entered into the trace headers manually.

Prior to the complete geometry setup, the data are separated into three data files

corresponding to the X, Y and Z components. Figure 19 shows the components after the

separation. In a near-offset VSP, the Z component data contain an upgoing and a

downgoing P-wave. Thus, the energy shown on a Z component data provides information

about the P-wavefield.

I picked first-breaks on the vertical raw Z component for each trace to mark the

first arrival of source-generated energy. This first break picks were used create to a P-

wave velocity profile (Figure 20). Also the picks were used to separate upgoing waves

from the downgoing waves (Coulombe, 1993).

39

Figure 19. AND2-A VSP data components (A) X component. (B) The other horizontal component (Y). (C) Vertical component (Z).

40

Figure 20. First break picks for a portion the Z component data. The selected time in each trace

marks the first arrival of source-generated energy. Notice the consistent trend with the exception of a

few outliers, which would indicate a problem with the first break picks and/or geometry. The outliers

1046-1030 m might be caused by a casing change while the outlier at 1334 m is believed to be caused

by a timing error. These outliers are disregarded in this study.

3.6. Velocity Profile

The VSP travel time data can be used to determine velocities, which can then be

correlated with sonic log data. These velocities are used for time to depth conversion for

surface seismic reflection data.

Figure 21 shows the travel time versus true vertical depth curve (left plane) that

was obtained from the first-break picks in Figure 20. The right plane shows the velocity

and interval velocity versus true vertical depth profile.

41

Figure 21. Travel time versus true vertical depth curve (left) and velocity versus true vertical depth

(right). The left figure shows the first break times of the raw vertical component plotted against

depth. The right figure shows the corresponding interval velocity (red) with mean velocity (blue).

3.7. Hodogram Analysis

Multicomponent analysis is used to estimate the orientation of geophones In

addition to its role in data processing. Inconsistency in geophone orientations (Figure 22)

can cause unwanted phase shifts and amplitude variation in data.

42

Figure 22. Schematic of VSP geophone orientation problem (Di Siena et al., 1984). The problem of

orientation is repeated at every station. Hp refers the reference ray path.

To address inconsistency, the horizontal components need to undergo hodogram

analysis, so that the energy of the component is maximized towards the source in a

particular time window around the first-breaks (DiSiena et al., 1984).

Two hodogram rotations were performed on the data. The first is applied to the X

and Y components to orient the data to obtain new HMAX component (pointing toward

to the source), where the energy that contains the primary P- and SV-wavefields

(vertically polarized S wave) is maximized, and new HMIN component, where the

energy that would have a SH-wavefield (horizontally polarized S wave) is minimized.

Figure 23 shows the, P- and S-wave particle motion.

43

Figure 23. P-waves and S-wave particle motions. P-waves and S-waves can be reflected or

transmitted by an incident P-wave once the source is positioned within certain distance of the rig. P-

wave particle motion is orthogonal to the direction of propagation. The polarization of the SV-waves

are polarized in the vertical plane, while the polarization of the SH-waves takes place in the

horizontal plane (Balaguera et al., 2014)

Figure 24A illustrates the rotation procedure. For this study, the particle motion of

the P-wave energy from the two horizontal components was plotted by choosing the

angle to maximize the energy on one component and minimize it on the other component

(Figure 24B).

Figure 25 shows the comparison result of the analysis of HMAX and HMIN

components after the first hodogram orientation.

44

Figure 24. Visual explanation of hodogram analysis and the real data analysis. (A) Projection of X and Y data components by data rotation onto the

vertical-plane vector (modified from Hinds et al., 1989). (B) A hodogram of the two inputs (X and Y). The angle (θ) defined with this analysis is used to

find the proper orientation of the horizontal components. P-wave arrivals are compared on two horizontal components in order to maximize energy on

one component at the expense of the other component to create HMAX and HMIN.

45

Figure 25. Comparison of HMAX and HMIN components. Rotation analysis is conducted for every receiver and source. (A) Raw X component.

(B) Raw Y component. (C) HMAX component. (D) HMIN component. Black square indicates the maximization of the energy on HMAX component

and minimizes it on the HMIN component. SV-waves will be picked from the rotated HMAX component, and SH-waves will be picked from the rotated

HMIN component in further processing steps.

46

The second hodogram rotation orients the data in the plane framed by the source

and well. Inputting the HMAX and Z components for the second rotation results in

HMAX* and Z* wavefields that are rotated in the source and well plane. In principle, the

HMAX* component should contain the downgoing and upgoing SV-waves while Z*

should contain the downgoing and upgoing P-wavefields. A sketch of this second rotation

is displayed in Figure 26.

A horizontal hodogram of the direct P-wave arrivals is shown in Figure 26. The

HMAX and HMIN components resulting from this analysis were compared with the raw

horizontal components in Figure 27.

47

Figure 26. Visual explanation of second hodogram analysis. (A) Orthogonal coordinate system showing HMAX* and Z* components (modified from

Hinds et al., 1989). (B) Hodogram of the two inputs (HMAX and Z). Hodogram resembles an ellipse pointing in the direction of the azimuth of the

downgoing wave. The angle decided in hodogram analysis is used to calculate HMAX* and Z*. P-wave arrivals are compared on the Z and HMAX

components.

48

Figure 27. Comparison of HMAX* and Z* components. (A) Raw vertical component. (B) Rotated Z* component. (C) HMAX component. (D) Rotated

HMAX* component. After the polarization, the HMAX* data will have downgoing and upgoing SV-wave energy. The Z* axis data contains

predominately downgoing and upgoing P-wave energy.

49

3.8. Wavefield Separation

3.8.1. Wavefield Separation Using a Median Filter

Upgoing and downgoing waves need to be separated for further processing.

Stewart (1985) notes that a popular method of wavefield separation is the median filter,

which utilizes the linear nature of the downgoing P-wavefield to keep the events aligned

in the same direction. A median filter functions by selecting the middle value from a

sequence of numbers arranged in ascending order. The moving window of the data

waiting to be filtered is the source of the numbers. The importance of the median filter in

VSP data processing results from its rejection of noise spikes and its passage over step

functions without altering them (Hardage, 1983).

Different median filtering lengths of 9, 11, 13, 14, 15, 19, and 21 points were used

to test the median filter for this thesis. The best result was obtained from 15-median filter

(Figure 28).

50

Figure 28. Wavefield separation using median filtering. (A) Raw Z component. (B) Upgoing waves of VSP displayed following application of the 15-

median filter known as the alpha trim median filter.

51

3.8.2. Wavefield Separation Using an f-k Filter

Another method of wavefield separation entails the use of an f-k filter. This

technique requires that adequate spatial sampling of data be conducted to prevent aliasing

(Poletto et al., 2004).

f-k filters are applied by separating waves using velocity passes (upgoing P-

waves) and reject filters (downgoing P-waves) (Sengbush and Al-Fares, 1987).

Upgoing P-waves reside in the negative wavenumber quadrant of the f-k spectrum

and downgoing P-wave arrivals reside in positive wavenumber quadrant of the f-k

spectrum. Figure 29 shows the f-k wavefield separation with a defined rejection zone to

obtain upgoing P-waves.

The best results were obtained with the median filter, however results were

obtained with the f-k filter will also be used to correlate the VSP corridor stack to

correlate with seismic data (Figure 30).

52

Figure 29. Wavefield separation using the f-k filtering to remove downgoing energy. Raw Z component (A) was transformed into the f-k domain (B).

Downgoing events reside in the negative wave number of analysis window and the upgoing events reside in the positive wave number of analysis

window. A pie filter was used to attenuate downgoing energy. (C) Downgoing energy was removed in the f-k domain. (D)Upgoing energy after f-k

filtering (inset images from Solano, 2004).

Figure 30. Comparison of results after application of the median and f-k filters. (A) The raw vertical component. (B) Upgoing waves after the median

filter was applied. (C) Upgoing wavefield after f-k filter was applied.

54

3.9. Predictive Deconvolution

A predictive deconvolution algorithm was applied to the data to enhance the

reflections and attenuate multiples. Deconvolution predicts the time and magnitude of

multiples and removes them from the data.

Two passes of predictive deconvolution were applied. First, short period

deconvolution used a prediction lag equivalent to the second zero crossing of the auto

correlation (Figure 31) to suppress short period ghost, bubble pulse effects, and multiples.

Then, long period deconvolution, using the second large peak of the auto correlation was

used to reduce water bottom multiples.

The results of the deconvolution can be seen in Figure 32 and compared to the

data prior to the deconvolution. The reflections in Figure 32 have been improved because

the wavelets are narrower.

55

Figure 31. Autocorrelation window used for deriving deconvolution parameters. First, short period

deconvolution used a prediction lag equivalent to the second zero crossing of the auto correlation to

suppress ocean-bottom multiples. Then, long-period deconvolution used the second large peak of the

auto correlation to reduce water bottom multiples. Multiples can be repeated several times with

reversing polarity. This happens when an upgoing wave hits the water/air interface, the reflection

coefficient (fraction of incident energy reflected from interface called reflection coefficient) becomes

negative which means the reflected wave has reverse polarity.

56

Figure 32. Upgoing wavefield of vertical component before and after the deconvolution. (A) Vertical component upgoing wavefield prior to

deconvolution. (B) The same upgoing waves after predictive deconvolution. The black ellipse present in both panels is centered at the same point and

shows a clear difference following the application of deconvolution. After the predictive deconvolution algorithm was implemented, long and short

period multiples have been minimized.

57

3.10. Corridor Stack

The next step is the creation of a corridor stack of the deconvolved upgoing data

set. In this procedure, the upgoing primary reflection events are summed and their two-

way arrival times are shifted to yield a single trace. This allows correlation with the

surface data (Hardage, 2000; Sheriff, 2002).

The data were flattened to first break times by adding a time shift given by the

equation.

𝑇𝑖𝑚𝑒 𝑠ℎ𝑖𝑓𝑡 = 𝑇𝑖𝑚𝑒 𝐷𝑎𝑡𝑢𝑚 − 𝐹𝑖𝑟𝑠𝑡 𝐵𝑟𝑒𝑎𝑘 𝑇𝑖𝑚𝑒𝑠 (1)

An exponential gain was applied to balance the amplitudes. The VSP Normal Move Out

(NMO) algorithm was applied and then the time was multiplied by two to place the VSP

data at two-way time (TWT). A top mute was applied to eliminate artifacts of downgoing

wavefield removal. The results of flattening to TWT time are shown in Figures 33.

Application of median filtering improved the continuity of upgoing events before

stacking. Then a very narrow corridor mute was applied and the data were stacked into

one trace, and the final stacked trace was repeated 10 times (Figure 33).

The use of a top mute, to eliminate artifacts that were present in the median filtered data

to remove downgoing events, removed much of the shallow reflectors on the corridor

stack. For this reason data from f-k filtering to remove the downgoing events was also

used to create a second corridor stack. The f-k filtered data did not require as severe of a

top much and so shallow reflectors were presented (Figure 34).

58

Figure 33. Corridor stack procedure of the median filtered, deconvolved upgoing data set. (A) The deconvolved upgoing P-waves of the vertical

component. (B) The same data after being transferred into TWT time and after being NMO corrected. A top mute was applied to eliminate artifacts of

downgoing wavefield removal. (C) VSP outside the corridor mute. A very narrow corridor mute was used to eliminate seismic noise. (D) Corridor stack

of the deconvolved upgoing data.

59

Figure 34. Corridor stack procedure of the f-k filtered, deconvolved upgoing waves. (A) The deconvolved upgoing P-waves of the vertical component.

(B) The same data after being transferred into TWT time and after being NMO corrected. (C) VSP outside the corridor mute. A very narrow corridor

mute was used to eliminate seismic noise. (D) Corridor stack of the deconvolved upgoing data.

60

3.11. Seismic Well Tie

Sonic and density logs are used to create a synthetic seismogram. The

measurements for density and velocity are utilized for determining the acoustic

impedance, which is needed to calculate the reflection coefficient and create a synthetic

seismogram.

The sonic log continuously records the time in microseconds required for a sound

wave to travel through the sonic measurement used for this project were collected on core

that was removed from the borehole. The sonic log travel time depends on formation

lithology, porosity and fluid saturation.

The formation density log measures the bulk density of the formation by

bombarding the core with gamma rays and detecting those that are scattered back to the

sensor. Density logs give valuable information on total porosity and gas-bearing

formations.

To create a synthetic seismogram, an acoustic impedance log was calculated from

velocity and density values that have been measured at the same depths along the core.

These acoustic impedance data were utilized to calculate a reflection coefficient (RC)

series. The RC series was then convolved with a wavelet and displayed as a wiggle trace

(synthetic seismogram). Figure 35 shows the synthetic seismogram created with the

Ricker wavelet. Figures 36 and 37 show the correlation of surface seismic data with

synthetic seismogram, and corridor stack. After the generation of synthetic seismogram,

the surface seismic data were prepared for a depth interpretation. VSP velocities were

used to convert the combined surface section to depth, which was used for interpretation.

61

Figure 35. Synthetic seismogram created from sonic log, density log, and Ricker wavelet.

62

Figure 36. Correlation of the (A) corridor stack and (B) synthetic seismogram with the (C) ATS-05-02 surface seismic data. The spacing between CDP

locations is 12.5 m. (I) Corridor stack which was generated from data where downgoing waves were removed by using a median filter. (II) Corridor

stack which was generated from data where downgoing waves were removed by using an f-k filter.

63

3.12. Regional seismic correlation and stratigraphic context

To understand the interpreted section, I analyzed the stratigraphy and lithology

of formations in the region. Correlations of the regional seismic stratigraphy with the

AND-2A drill core are listed as follows.

Two seismic units were identified. The first unit is the Late Oligocene – Early

Miocene unit which is made up of sequence boundaries that are identified based on

truncated reflectors and shows increase in the glacier ice conditions. The second unit is

the late Early Miocene to Present unit which includes truncated underlying reflectors and

shows progress of glacier ice over the Ross Sea. Age dates were found by correlation of

seismic reflectors with CIROS–1 for the Oligocene strata and by tying the ATS – 05

seismic lines into AND–2A to determine the ages of Miocene reflectors (Pekar et al.,

2013). Existing seismic reflection horizons identified in seismic reflection data also

matched with the corridor stack. The AND–2A core shows the presence of important

reflectors like Ri, Rh, Rg, Rf, Re2, Rf2, Ri and Re (Figures 37 and 38). Three seismic

units have been identified by other researches for this region. These are discussed below.

Stratigraphic section is changing between the upper Oligocene middle and

Miocene and younger reflectors.

Unit 1: Eocene to Lower Oligocene reflectors

Unit 1 is bounded by Ra and Rc reflectors, which have been described by Pekar et

al., 2013. These reflectors were not identified in this study.

64

Unit 2: Lower Miocene to Upper Oligocene reflectors (Main Rift Interval)

(Lower to Upper Oligocene: 29–23 Ma)

Unit 2 is bounded by middle Oligocene Rc and middle Miocene Rg reflectors. Early

Miocene reflectors were traced from AND-2A to CIROS-1 by Pekar et al. (2013), who

identified two new reflectors, Re2 and Rf2.The Re reflector from Fielding et al. (2006) is

tied to the AND-2A core at 1020 mbsf. Rf2 reflector was tied to the AND-2A core at 450

mbsf. The Re2 reflector truncates the other reflectors below. This reflector tied to the

AND-2A core at about 720 mbsf. The Rf reflector from Fielding et al., 2006 was tied to

the AND-2A core at 800 mbsf.

Unit 3: Recent to Middle Miocene reflectors (Passive Thermal Subsidence)

(covers Lower Middle Miocene 23- ~13 Ma)

Middle Miocene and younger reflectors from Ross Sea have been the best studied

seismic units in southern McMurdo Sound (Davy & Alder 1989; Anderson & Bartek

1992; Brancolini et al. 1995; Bartek et al. 1996; Fielding et al., 2006). Rg, Rh, and Ri

reflectors belong to this unit. The Rg reflector has moderate amplitude and is tied to the

AND-2A at about 290 mbsf. The Rh reflector has moderate amplitude and tied to AND-

2A core at 250 mbsf. The Ri reflector is not observed in the corridor stack created with

median filter but seen in the corridor stack which was created with an f-k filter and is tied

to AND-2A at about110 mbsf.

65

Fielding

et al.

(2008a)

Pekar et

al., 2013

This

study

Age

(Ma)

1 Ri Ri

Ri 4 Ma

2 Rh Rh

Rh 14.3 Ma

3 Rg Rg Rg

15.5–14.9

Ma

4 Rf2

Rf2 16.1 Ma

5 Re2

Re2 19.4 Ma

6 Rf Rf

Rf 17.8–18.7

Ma

7 Re Re

Re 23 Ma

8 Rd

9 Rc Rc

25–33 Ma

10 Rb

11 Ra Ra

33.9 Ma

12 Ra.5

35 Ma

Figure 37. Seismic stratigraphic framework for this study, showing correlation to other seismic

stratigraphic schemes for the region. Colored lines represent seismic reflectors that bound

interpreted seismic units and are tied to AND-2A core.

66

Figure 38. Final interpreted geologic section overlaying the depth converted seismic section. Regional seismic reflections are identified in a surface

seismic section and correlated to whole core data. (a)Surface seismic data in time. (b) A whole core time depth curve (blue) is overlain with the time

depth curve obtained from VSP first arrivals (red). The whole core sonic (P-wave) velocities were measured on the core taken from the well. (c) Whole

core velocity. (d) Density data shown with a 50-point smoothing applied. (e) Regional seismic reflectors are overlain with their respective names. Ri, Rh,

Rg Rf2, Re2, Rf and Re from Pekar et al. (2013). (f) Seismic stratigraphy of AND-2A well (Figure 10) (Taviani et al., 2008). (g) Synthetic seismogram

created from density and velocity core data. (h) P-wave corridor stack obtained from median filtered upgoing waves. (i) P-wave corridor stack obtained

from f-k filtered upgoing waves converted to depth. (j) Surface seismic data in depth.

67

3.13. Discussion and Results

The VSP P-wave time-depth curve obtained from the first-break picks showed

good agreement with the time-depth curve obtained from the measurement of the whole

core velocity. Traditionally, these time depth curves do not display such similarity at

depth because seismic waves are sampled in the vertical plane while core velocities are

sampled in the horizontal plane. As sediments are compacted with depth, anisotropy

occurs between the vertical and horizontal directions because, compaction leads to

preferential mineral alignment and layering. The fact that these curves agree may suggest

over compaction from the top and down due to the ice loading and/or sediment

consolidation through the entire section.

Near-offset VSP deconvolution is used to eliminate multiples. This technique is

based on optimizing the deconvolution parameters from downgoing waves and then

applying these parameters to the upgoing waves. However, in this research two passes of

predictive deconvolution were used instead and this made a significant improvement in

wavelet shaping and multiple suppressions.

Major seismic reflectors in the corridor stack correspond to the most important

stratigraphic boundaries that are observed in the core. However, shallow reflectors (Ri,

Rh, and Rg) did not show up in the corridor stack created with median filtered,

deconvolved upgoing waves because of the top mute used to eliminate the artifacts of the

downgoing wavefield removal. On the other hand, a P-wave corridor stack obtained

from f-k filtered upgoing waves shows the shallow reflections.

68

4. CHAPTER 4

4.1. Brief Theory of Seismic Anisotropy

Anisotropy is defined as a physical property variation which is dependent upon

the direction that the measurement is taken (Sheriff, 2002). Fractures, small cracks,

mineral alignment, and sedimentary layering can be some reasons for anisotropy.

Many seismic anisotropy observations to date reveal that anisotropy can be

divided into two main types. Both are classified as transverse anisotropy, but they have

different symmetries.

An anisotropy system having symmetry along the vertical axis (VTI) is produced

by the natural horizontal alignment of the materials of sedimentary rocks (Wild, 2011).

Usually vertical or dipping fractures and cracks within a medium create the horizontal

transverse isotropy (HTI). Figure 39 shows horizontal layering characteristics defined by

the VTI anisotropy and vertical fracturing characteristics defined by the HTI anisotropy.

Figure 39. VTI and HTI anisotropy with symmetry axis. (A) VTI anisotropy characterizing

horizontal layering. (B) Vertical fracturing that occurs in fractured reservoirs is characterized by

HTI anisotropy (Wild, 2011).

69

Techniques to study seismic anisotropy include:

S-wave splitting: the two polarized S-waves arrive at different times.

Azimuthal anisotropy: the arrival times or apparent velocities of seismic waves at

a given distance from an event depend on azimuth.

4.2. S- wave Splitting

S-wave anisotropy is regularly evaluated by measuring S-wave splitting (Alford,

1986; Crampin, 1985) in VSP data. S-wave splitting, also called seismic birefringence,

happens when a polarized S-wave enters an anisotropic medium.

Figure 40 shows the splitting process. S-wave particle motion splits into two

unique polarizations (for a fast S-wave and a slow S-wave) when an S-wave is

propagated through an anisotropic medium.

The time lag (time distinction between the fast and the slow S-wave) is the most

direct confirmation for anisotropy. S-wave splitting is routinely seen in three-component

VSPs, and time lags between fast and slow S-waves give us a way to find fracture

orientations and anisotropy rates.

70

Figure 40. S-wave splitting process. An S-wave in an isotropic medium moves into an anisotropic

medium accompanied by a fast and a slow direction (red and blue planes), on which the first pulse

splits, to accumulate a “time lag” (Crampin, 1981).

4.3. Converted-Wave Methodology

To obtain the two independent seismic sections, one with the P-wave while the

other one with the PS-waves, is the main goal of the converted-wave method. Generally,

horizontal components record PS-wave data and the vertical component records P-wave.

However, some energy leakage might occur between the two parts. This causes both the

vertical and horizontal component recordings to include both P-wave and SV-energy

(Figure 41). This makes the converted mode processing becoming more complex. As a

result of using three-component geophones and vertical propagation of P and S-waves in

a near-offset VSP, the particle motion of these waves could be separated easily.

71

Figure 41. Wave mode contribution for the case of a deviated borehole. The triaxial geophone

package receives contributions from P waves, SV-waves and SH-waves on all geophones. Even for

the near-offset source location at S2, the Z channel data will contain wave modes other than just P-

waves. To separate the various wavefield contributions, the data on three geophones is rotated using

the hodogram analysis. This is the geometry used in the analysis of upgoing rays for the deviated

borehole.

4.4. S-wave Processing

I investigated the horizontal components of VSP data for possible S-wave

arrivals, which should be slower than P-wave arrivals. Two sub-modes for S-waves exist:

In SV the particle motion is parallel to the vertical plane whereas in SH the particle

motion is parallel to the horizontal plane.

S-waves propagate in a direction that is perpendicular to the direction of motion,

P-waves travel in a direction parallel to the direction of motion (Figure 42). Particle

motions for P- and mode converted SV-waves are contained within the same plane. Mode

converted SH-waves are not mechanically coupled to either P- or SV waves (Zoeppritz

Equations) (Tatham and Campbell, 2013).

72

Figure 42. P, SV, and SH waves captured by a triaxial geophone (Baziw & Verbeek, 2016).

The orientation of the two horizontal components of the three-component

geophones needs to be corrected for the rotation of the tool in the borehole. When the PS-

wavefield data are used, the PP-wave must be eliminated from the horizontal component.

Hodogram analysis is used to rotate the S-wave data to in-line (radial component) and

cross-line (transverse component) coordinate system, by choosing the angle to maximize

the energy in one component while minimize it in the other component. SV-waves were

picked from the rotated HMAX* component and SH-waves were picked from the rotated

HMIN component.

Once the HMAX* and HMIN components were generated, I applied an f-k filter

with a polygon mute to remove the P-wave energy from the S-wave energy prior to

downgoing S wave velocity analysis.

73

Figures 43 and 44 show the HMAX* component and Figures 45 and 46 show the

HMIN component after application of an f-k filter.

Possible S-waves are highlighted in Figures 44 and 46 and show low velocities (1600 ±

130.5 m/s and 1900 ±112.7m/s) relative to P- wave velocities. This velocity range is

consistent with what I might be expected for shallow unconsolidated sediments.

However, tube waves travel at the speed closer to the speed of water (1500 m/s)

and these events that I have interpreted as S-wave might be tube waves instead.

74

Figure 43. S-wave separation using the f-k filter. (A) The rotated HMAX* component transformed

into the f-k domain. (B) The S-wave energy has a velocity of 1614 ± 130.5 m/s. (C) HMAX*

component in the f-k domain with the rejection filter to keep S- waves as much as possible. (D) The

S-wave after the f-k filter was applied.

75

Figure 44. S-waves after wavefield separation. (A) HMAX* component of the data (B) S-waves after wavefield separation using the f-k filter. S-Waves

travel more slowly and arrive later than P-waves.

76

Figure 45. S-wave separation using the f-k filter. (A) The rotated HMIN component transformed

into the f-k domain. (B) The S-wave energy has a velocity of 1976 ±112.7m/s. (C) HMIN component

in the f-k domain with the rejection filter to keep S-waves as much as possible. (D) The S-wave after

the f-k filter was applied.

77

Figure 46. S-waves after wavefield separation. (A) HMIN component of the data (B) S-waves after wavefield separation using the f-k filter.

78

4.5. Discussion and Results

The separation of the PS-wave energy from the total wave field is the main purpose for

converted wave processing. Unfortunately, the S-wave data gathered from these data are

discontinuous, and I could not follow the downgoing S-waves to produce a reasonable velocity

from first-break picks. Upgoing S-waves are not identified. Many of the S-wave arrivals were

difficult to pick. I was not able to rotate these data yet a third time to obtain components to

maximize the velocity separation between the fast and slow directions because of the poor

quality of these data.

Only one initial shear-wave polarization can be produced depending on the dip of the

boundary where the mode-conversion occurs. If the polarization of the converted S-wave is in a

symmetry plane, no S-wave splitting will be observed. Near-offset VSPs are not expected to

produce significant mode conversion of P- to SV, and little to no SH.

S-wave arrivals may be contaminated by P-wave energy which can distort information

for the S-wave splitting. This problem can be eliminated with the use of effective P- and S-

wave separation techniques, without distorting S-wave polarizations. (Dankbaar 1985; Dankbaar

1987; Dillon, Ahmed & Roberts 1988; Esmersoy 1990).

Although the near-offset geometry and lack of high quality S-wave data, S-wave splitting

was observed after hodogram rotation. The HMAX*(radial) component travels as a slow S-wave

and the HMIN (transverse) component travels as a fast S-wave (Figure 47).

79

Figure 47. Principles of S-wave splitting in a fractured rock (Hardage, 2011). Fast S-wave is polarized in the

same direction as the fracture orientation. Slow S-wave which is polarized in a direction perpendicular to the

fractures.

The velocity difference on rotated HMAX* and HMIN components indicates the

existence of anisotropy in the near sea floor strata. Velocities on HMAX* (perpendicular to the

fractures) and HMIN (parallel to the fractures) components are in agreement with the strike of

major faults that are parallel to the coastline adjacent to the TAM Front (Figure 48). The

orientation of the fault planes and fractures in the survey region would likely tend to be oriented

perpendicular to the minimum horizontal compressive stress regime which is interpreted from

the radial component (HMAX*) (Lorenz et al. 1990). The maximum horizontal compressive

stress direction which is interpreted to be most closely aligned with the transverse component

(HMIN) is oriented approximately in a north-south direction roughly parallel to the TAM Front

(Wilson and Paulsen, 2000).

80

Figure 48. Geographical features and horizontal S-wave component directions. HMAX* is the direction of the

slower S-wave velocity and HMIN is the direction of the faster S-wave velocity. These have been interpreted

as approximately the minimum horizontal compressive stress direction and maximum horizontal stress

direction respectively (modified from Dunbar et al., 2008).

81

The low S-wave velocities suggest the existence of shallow marine sediments in this

study. Typically, sea bottom sediments have downgoing P-wave velocities greater than 2500

m/s, and downgoing S-wave velocity greater than 1000 m/s for hard sea bottom (Tatham and

Stoffa, 1976). The velocity ratio of P-wave to S-wave (Poisson's ratio) in marine sediments is

important as an indicator of the degree of the sediment consolidation. If the velocity ratio of

Vp/Vs is < 2.0, the sediment is consolidated and if the ratio of Vp/Vs is > 2.0, the sediment is

unconsolidated (Graden and Harris, 1986). In this study Vp/Vs is > 2.0 suggesting the sediments

are well consolidated.

VSP data may also be contaminated by the tube waves. The low S-wave velocities seen

in Figures 42 and 43 could be caused by tube waves which often appear strong on horizontal

components due to the relatively low signal to noise ratio. Tube waves, have velocity similar to

what is expected for the S-waves in these sediments for low velocity layers (White, 2000). If this

is indeed the case, the observed S-wave splitting might be caused by error in picking velocities

from tube waves on the horizontal components and, as a consequence, not represent S-wave

splitting at all.

82

5. CONCLUSIONS

This study shows that a single near-offset, over-sea ice multi-component VSP survey

using a GI air-gun was a useful project. A GI air-gun successfully suppressed the bubble pulse

effects caused by using explosives in the water column. The multicomponent geophone used in

the survey produced high quality seismic P-wave data.

Whole core velocities from a sonic velocity tool were used to derive a time-depth curve

which was compared with the VSP P-wave velocity curve. The VSP velocity curve represents

vertical velocities while sonic velocity tool records horizontal velocities. The P-wave velocities

determined from the VSP data show a good agreement with the velocity trends in the smoothed

sonic velocity tool velocities. Overloading by ice during the successive advances and retreats of

glaciers removes differential compaction with depth on the sediments and this along with partial

consolidation of sediments might be a reason for this agreement because these effects would tend

to minimize any anisotropy that we might have otherwise been expecting to observe.

Four major seismic reflectors are present in the corridor stack and they match very well

with the important regionally extensive unconformities. These unconformities represent major

tectonic events and change in climate. Major seismic reflectors in the corridor stack correspond

to the most important stratigraphic boundaries that are observed in the core.

Mode converted S-waves appear to have been generated at the base of the water column.

S-wave splitting analysis of the two horizontal components was used to determine the dominant

fracture orientation from anisotropy. S-wave velocity is the least in a direction perpendicular to

major faults in the region and the slower velocity could be caused by the fracture orientation at

depth. The interpreted subsurface fracture orientation suggests that the minimum horizontal

83

compressive stress is in roughly east-west direction and that the maximum horizontal stress is

oriented north-south.

84

6. BIBLIOGRAPHY

Anderson, J. B. (1999). Antarctic marine geology. London, England: Cambridge University Press.

Anderson, J. B., & Bartek, L. R. (1992). Cenozoic glacial history of the Ross Sea revealed by

intermediate resolution seismic reflection data combined with drill-site information. The

Antarctic Paleoenvironment: A Perspective on Global Change: Part One, 231-264.

ANDRILL International Science Proposal. (2003). ANDRILL: Investigating Antarctica’s role in Cenozoic

global environmental change (ANDRILL Contribution No. 2). Lincoln, NE: ANDRILL Science

Management Office, University of Nebraska–Lincoln.

Balaguera, A., Useche, M., Romero, J., Meza R., Rivas J., Busnego H., & Zambrano

A.(2014).Applications of the shear wave acquired from a Zero Offset VSP, in the seismic

characterization of the Morichal Member, Oficina Formation, in the Faja Petrolífera del Orinoco,

Venezuela.

Barrett, P. J. (1989). Antarctic Cenozoic history from the CIROS-1 drillhole, McMurdo Sound. New

Zealand Department of Scientific and Industrial Research Bulletin, 245- 254.

Barrett, P. J. (2007). Cenozoic climate and sea level history from glacimarine strata off the Victoria Land

coast, Cape Roberts Project, Antarctica. Glacial Sedimentary Processes and Products, 259-287.

Barrett, P. J., & Hambrey, M. J. (1992). Plio-Pleistocene sedimentation in Ferrar Fjord, Antarctica.

Sedimentology, 39, 109–123.

Bartek, L. R., Henrys, S. A., Anderson, J. B., & Barrett, P. J. (1996). Seismic stratigraphy of McMurdo

Sound, Antarctica: Implications for glacially influenced early Cenozoic eustatic change? Marine

Geology, 130, 79–98.

Baziw, E. & Verbeek G. (2016). Source Wave Design for Downhole Seismic Testing,

http://www.bcengineers.com/images/BCE_Technical_Note_4.pdf, accessed date January 12,

2016.

Betterly, S. J., Speece, M. A., Levy, R. H., Harwood, D. M., & Henrys, S. A. (2007). A novel over-sea

ice seismic reflection survey in McMurdo Sound, Antarctica. Terra Antarctica, 14, 97–106.

Campbell, A., Fryer, A., & Wakeman, S. (2005). Vertical seismic profiles—More than just a corridor

stack. The Leading Edge, 24, 694–697, doi: 10 .1190/1.1993259

Converted Wave Seismic Surveys (2016). CREWES Research Links

http://www.crewes.org/ResearchLinks/Converted_Waves/Page2.html, accessed date January 12,

2016.

Cooper, A., Adam, L., Coulter, R., Eby, G., & McIntosh, W. (2007). Geology, geochronology, and

geochemistry of a basanitic volcano, White Island, Ross Sea, Antarctica. Journal of Volcanology

and Geothermal Research, 189–216.

Cooper, A. K., Davey, F. J., & Behrendt, J. C. (1987). Seismic stratigraphy and structure of the Victoria

Land basin, western Ross Sea, Antarctica. In A. K. Cooper & F. J. Davey (Eds.), The Antarctic

continental margin: Geology and geophysics of the western Ross Sea (Earth Science Series 5B,

pp. 27–66). Houston, TX: Circum-Pacific Council for Energy and Mineral Resources.

Coulombe, C. A. (1993). Amplitude-versus-offset analysis using vertical seismic profiling and well log

data (Master’s thesis). Department of Geology and Geophysics, University of Calgary.

Crampin, S. A. (1981). Review of wave motion in anisotropic and cracked elastic-media. Wave Motion, 3,

343–391.

85

Crampin, S. A. (1985). Evaluation of anisotropy by shear-wave splitting. Geophysics, 50, 142–152.

Daley, T. M., McEvilly, T. V., & Majer, E. L. (1988). Analysis of P and S wave vertical seismic profile

data from the Salton Sea scientific drilling project. Journal of Geophysical Research, 93,

doi:10.1029/88JB03073.

Dalziel, I. W. D., & Lawver, L. A. (2001). The lithospheric setting of the West Antarctic ice sheet. The

West Antarctic Ice Sheet: Behavior and Environment, Antarct. Res. Ser, 77, 29-44.

Dankbaar, J. W. M. (1985). Separation of P-and S-waves. Geophysical Prospecting, 33(7), 970-986.

Dankbaar, J. W. M. (1987). Vertical Seismic Profiling—separation of P-and S-waves. Geophysical

Prospecting, 35(7), 803-814.

Davey, F. J., Barrett, P. J., Cita, M. B., van der Meer, J. J. M., Tessensohn, F., Thomson, M. R. A., . . .

Woolf, K. J. (2001). Drilling for Antarctic Cenozoic climate and tectonic history at Cape Roberts,

Southwestern Ross Sea. Transactions American Geophysical Union, 589–590.

doi:10.1029/01EO00339.

Dillon, P. B., Ahmed, H., & Roberts, T. (1988). Migration of mixed-mode VSP wavefields. Geophysical

Prospecting, 36(8), 825-846.

Disiena, J. P., Gaiser, J. E., & Corrigan, D. (1981). Three-component vertical seismic profile: orientation

of horizontal components for shear wave analysis: Presented at 51st SEG meeting. Los Angeles.

Dondur, D. (2009). Deniz sismiginde veri islem. JFMO egitim yayinlari, 11,294.

Dunbar, G., Atkins, C., Magens, D., & Niessen, F. Studies from the ANDRILL Southern McMurdo

Sound Project, Antarctica—Initial Science Report on AND-2A: Terra Antartica.

Esmersoy, C., Chang, C., Kane, M., Coates, R., Tichelaar, B., & Quint, E. (1998). Acoustic imaging of

reservoir structure from a horizontal well. The Leading Edge, 17(7), 940-946.

Falconer, T., Pyne, A., Wilson, D., Levy, R. H., Nielsen, S., & Petrushak, S. (2008). Operations overview

for the ANDRILL Southern McMurdo Sound project, Antarctica, Studies from the ANDRILL

Southern McMurdo Sound project, Antarctica. Terra Antarctica, 15, 41–48.

Fielding, C. R., Woolfe, K. J., Howe, J. A., & Lavelle, M. (1998). Sequence stratigraphic analysis of

Cape Roberts Project, McMurdo Sound, Antarctica. Terra Antartica, 5, 353-361.

Fielding, C. R., Naish, T. R., Woolfe, K. J., & Lavelle, M. J. (2000). Facies analysis and sequence

stratigraphy of CRP-2/2A, Victoria Land Basin, Antarctica. Terra Antarctica, 7(3), 323–338.

Fielding, C. R., Naish, T. R., & Woolfe, K. J. (2001). Facies architecture of the CRP-3 drillhole, Victoria

Land Basin, Antarctica. Terra Antartica, 8, 217-224.

Fielding, C. R., Henrys, S. A., & Wilson, T. J. (2006). Rift history of the western Victoria Land Basin: A

new perspective based on integration of cores with seismic reflection data. In D. Fütterer, D.

Damaske, G. Kleinschmidt, H. Miller, & F. Tessensohn (Eds.), Antarctica contributions to global

earth sciences (pp. 309–318). Berlin, Germany: Springer.

Fielding, C. R., Atkins, C. B., Basset, K, N, Dunbar, G.B.(2008a). Sedimentology and stratigraphy of the

AND-2A core, ANDRILL Southern McMurdo Sound Project, Antarctica. In D. M. Harwood, F.

Florindo, F. Talarico, & R. H. Levy, (Eds.), Studies from the ANDRILL, Southern McMurdo

Sound project, Antarctica. Terra Antarctica, 15, 77–112.

Fielding, C. R., Whittaker, J., Henrys, S. A., Wilson, T. J., & Naish, T. R. (2008b). Seismic facies and

stratigraphy of the Cenozoic succession in McMurdo Sound, Antarctica: Implications for

tectonic, climatic, and glacial history. Palaeogeography, Palaeoclimatology, Palaeoecology, 260,

Fielding, C. R., Browne, G. H., Field, B., Florindo, F., Harwood, D. M., Krissek, L. A., ... &

86

Pekar, S. F. (2011). Sequence stratigraphy of the ANDRILL AND-2A drillcore, Antarctica: A

long-term, ice-proximal record of Early to Mid-Miocene climate, sea-level and glacial

dynamism.Palaeogeography, Palaeoclimatology, Palaeoecology, 305(1), 337-351.8–29.

Fielding, C. R., Blackstone, B. A., Frank, T. D., & Gui, Z. (2012). Reservoir potential of sands formed in

glaciomarine environments: An analogue study based on Cenozoic examples from McMurdo

Sound, Antarctica Geological Society, London, Special Publications, 211-228.

Fitzgerald, P. G., Sandiford, M., Barrett, P. J., & Gleadow, A. J. W. (1986). Asymmetric extension

associated with uplift and subsidence of the Transantarctic Mountains and Ross Embayment.

Earth & Planetary Science Letters, 81, 67–78. doi:10.1016/0012-821X(86)90101-9.

Fitzgerald, P. A. U. L. (2002). Tectonics and landscape evolution of the Antarctic plate since the breakup

of Gondwana, with an emphasis on the West Antarctic Rift System and the Transantarctic

Mountains. Royal Society of New Zealand Bulletin, 35, 453-469.

Florindo, F., Harwood, D., Levy, R., & SMS Project Science Team. (2008). ANDRILL’s success during

the 4th international polar year. Scientific Drilling, 6(6), 29–31. doi:10.2204/iodp.sd.6.03.2008

Gaiser, J., Ward, R., & DiSiena, J. (1982). Three component vertical seismic profiles: Polarization

measurements of P-wave particle motion for velocity analysis. SEG Technical Program

Expanded Abstracts, 162–165.

Gardner, G. H. F., & Harris, M. H. (1968, January). Velocity and attenuation of elastic waves in sands.

In SPWLA 9th Annual Logging Symposium. Society of Petrophysicists and Well-Log Analysts.

Garotta, R., Granger, P. Y., & Dariu, H. (2002). Combined interpretation of PP and PS data provides

direct access to elastic rock properties. The Leading Edge, 21(6), 532-535.

Goodge, J. W. (2007). Metamorphism in the Ross orogen and its bearing on Gondwana margin tectonics.

Geological Society of America Special Papers, 419, 185–203.

Goodge, J. W., & Finn, C. A. (2010). Glimpses of East Antarctica: Aeromagnetic and satellite magnetic

view from the central Transantarctic Mountains of East Antarctica. Journal of Geophysical

Research: Solid Earth (1978–2012), 115(B9).

Geological Survey (U.S.), British Antarctic Survey., United States., & National Science Foundation

(U.S.). (2007). Landsat Image Mosaic of Antarctica (LIMA). Reston, Va.: U.S. Depaertment of

the Interior, U.S. Geological Survey.

Hamilton, R. J., Luyendyk, B. P., Sorlien, C. C., & Bartek, L. R. (2001). Cenozoic tectonics of the Cape

Roberts Rift Basin and Transantarctic Mountains Front, Southwestern Ross Sea, Antarctica.

Tectonics, 20(3), 325–342.

Hardage, B. A. (1983). Vertical seismic profiling, Part A: Principles ,v. 14A. London Geophysical Press.

Hardage, B. A. (2000). Vertical seismic profiling: Principles (3rd updated and revised ed.) . New York.

Pergamon Seismic Exploration, Vol. 14.

Hardage, B. A. (2001). Developments, trends, and future directions in vertical seismic profiling and

crosswell seismic profiling. Society of Exploration Geophysicists Recorder, v. 35, no. 7, 72–78.

Hardage, B. A., DeAngelo, M. V., Murray, P. E., & Sava, D. (2011). Multicomponent seismic

technology: Society of Exploration Geophysicists, Geophysical References Series No. 18,318 ,

Hardage, B. (2011). GC Fracture Identification and Evaluation Using S Waves.

Harwood, D. M., Lacy, L., & Levy, R. (2002). Future Antarctic Margin Drilling: Developing a Science

Program Plan for McMurdo Sound--Report of a Workshop Oxford, UK April 5–7, 2001.

87

Harwood, D. M., Florindo, F., Levy, R. H., Fielding, C. R., Pekar, S. F., & Speece, M. A. (2004).

ANDRILL Southern McMurdo Sound Project Scientific Prospectus: ANDRILL Contribution

5. University of Nebraska-Lincoln, Lincoln, Nebraska, 29.

Harwood, D., Cowie, J., Florindo, F., Naish, T., Powell, R., & Pyne, A. (2006). Deep drilling with the

ANDRILL program in Antarctica. Scientific Drilling.

Hayes, D. E., & Frakes, L. A. (Eds.). (1975). Initial reports of the deep sea drilling project (Report No.

28). U. S. Government Printing Office.

Helbig, K., & Thomsen, L. (2005). 75-plus years of anisotropy in exploration and reservoir seismics: A

historical review of concepts and methods. Geophysics, 70, 9ND–23ND.

Henrys, S. A., Bucker, C. J., Bartek, L., Bannister, S., Niessan, F., & Wonik, T. (2000). Correlation of

seismic reflectors with CRP-2/2A. Terra Antarctica, 7, 221–230.

Henrys, S. A., Bucker, C. J., Niessen, F., & Bartek, L. R. (2001). Correlation of seismic reflections with

drillhole CRP-3, Victoria Land Basin, Antarctica. Terra Antarctica, 8, 127–136.

Hinds, R. C., Anderson, N. L., & Kuzmiski, R. D. (1999).VSP interpretive processing: Theory and

practice. Society of Exploration Geophysicists open File Publications, 56-67.

Hinds, R. C., & Kuzmiski, R. D. (2001). VSP for the interpreter/processor for 2001 and beyond: Part1.

Canadian Society of Exploration Geophysicists Recorder, 26(9), 84–95.

Horgan, H., & Bannister, S. C. (2004). Explosive source seismic experiments from a sea ice platform,

McMurdo Sound 2003. Institute of Geological & Nuclear Sciences Science Report, 2004(15), 23.

Huston, M., Gilbert, N., & Newman, J. (2006). Final Comprehensive Environmental Evaluation (CEE)

for ANDRILL McMurdo Sound Portfolio.Christchurch, Antarctica New Zealand.

The Intergovernmental Panel on Climate Change (2007). Climate change 2007: The physical basis. S.

Solomon, D. Qin, M. Manning, M. M. K. Averyt, M. M. B. Tignor, H. L. Miller, Jr., & Z. Chen

(Eds.). Retrieved from www.ipcc.ch.

Ishman, S. E., & Rieck, H. J. (1992). A late Neogene Antarctic glacio-eustatic record, Victoria Land

Basin margin, Antarctica. Antarctic Research Series, 56, 327–347.

Kyle, P., Moore, J., & Thirlwall, M. (1992). Petrologic evolution of anorthoclase phonolite lavas at

Mount Erebus, Ross Island, Antarctica. Journal of Petrology, 33, 849–875.

LeMasurier, W. E. (1990). Late Cenozoic volcanism on the Antarctic plate: an overview. Volcanoes of the

Antarctic Plate and southern oceans, 1-17.

Li, Y., Zhao, X., Zhou, R., Dushman, D., & Janak, P. (2005). 3C VSP tomography inversion for

subsurface P- and S-wave velocity distribution. SEG Technical Program Expanded Abstracts,

2625–2628.

Licht, K. J., Lederer, J. R., & Swope, J. (2005). Provenance of LGM glacial till (sand fraction) across the

Ross embayment, Antarctica. Quaternary Science Reviews, v.24, Issues 12–13, 1499–1520.

Liu, E., Queen, J. H., Li, X. Y., Chapman, M., Maultzsch, S., Lynn, H. B., & Chesnokov E. M. (2003).

Observation and analysis of frequency-dependent anisotropy from a multicomponent VSP at

Bluebell-Altamont field, Utah. Journal of Applied Geophysics, 54, 319–333.

Lorenz, J. C., Finley, S. J., & Warpinski, N. R. (1990). Significance of Coring-Induced Fractures in

Mesaverde Core, Northwestern Colorado (1).AAPG Bulletin, 74(7), 1017-1029.

Majer, E. L., McEvilly, T. V., Eastwood, F. S., & Myer, L. R. (1988). Fracture detection using P-wave

and S-wave vertical seismic profiling at The Geysers. Geophysics, 53, 76–84.

88

Martin, A. P., Cooper, A. F., & Dunlap, W. J. (2010). Geochronology of Mount Morning, Antarctica:

Two-phase evolution of a long-lived trachyte-basanite-phonolite eruptive center. Bulletin of

Volcanology, 72, 357–371.

Mckay, R., Browne, G., Carter, L., Cowan, E., Dunbar, G., Krissek, L., . . . Wilch, T. (2009). The

stratigraphic signature of the late Cenozoic Antarctic ice sheets in the Ross embayment.

Geological Society of America Bulletin, 121, 1537–1561.

McKelvey, B. C. (1981). The lithologic logs of DVDP cores 10 and 11, eastern Taylor Valley. Dry valley

drilling project, 63-94.

Mukasa, S. B., & Dalziel, I. W. D. (2000). Marie Byrd Land, West Antarctica: Evolution of Gondwana’s

Pacific margin constrained by zircon U‐Pb geochronology and feldspar common‐Pb isotopic

compositions. Geological Society of America, 112, 611–627.

Müller, T., Gurevich, B., & Lebedev, M. (2001). Seismic wave attenuation and dispersion resulting from

wave-induced flow in porous rocks: A review. Geophysics, 75, 75A147–175A164.

Naish, T., Powell, R., & Levy, R. H. (Eds.). (2007). Studies from the ANDRILL, McMurdo ice shelf

project, Antarctica—Initial Science Report on AND-1B. Terra Antarctica, 14, 328.

Nyland, R. E., Panter, K. S., Rocchi, S., Di Vincenzo, G., Del Carlo, P., Tiepolo, M., . . . Gorsevski, P.

(2013). Volcanic activity and its link to glaciation cycles: Single-grain age and geochemistry of

Early Miocene volcanic glass from ANDRILL AND-2A core, Antarctica. Journal of Volcanology

and Geothermal Research, 250, 106–128.

Panter, K. S., Talarico, F., Bassett, K., Del Carlo, P., Field, B., Frank, T., . . . the ANDRILL-SMS Science

Team. (2008). Petrologic and geochemical composition of the AND-2A Core, ANDRILL

Southern McMurdo Sound project, Antarctica. Terra Antarctica, 15, 147–192.

Passchier, S., Browne, G., Field, B., Fielding, C. R., Krissek, L. A., Panter, K., . . . the ANDRILL-SMS

Science Team. (2012). Early and Middle Miocene Antarctic glacial history from the sedimentary

facies distribution in AND-2A, Ross Sea, Antarctica. Geological Society of America Bulletin,

123, 2352–2365.

Paulsen, T., Encarnación, J., Valencia, V. A., Roti, J., & Rasoazanamparany, C. (2011). Detrital U-Pb

zircon analysis of an Eocene McMurdo Erratic sandstone, McMurdo Sound, Antarctica. New

Zealand Journal of Geology and Geophysics, 54, 354–360.

Pekar, S. F., Speece, M. A., Wilson, G. S., Sunwall, D. S., & Tinto, K. J. (2013). The Offshore New

Harbour Project: deciphering the Middle Miocene through Late Eocene seismic stratigraphy of

Offshore New Harbour, western Ross Sea, Antarctica. Geological Society, London, Special

Publications, 381(1), 199-213.

PetroIndependencia, B. P., & PetroIndependencia, A. Z. P. (2014). Applications of the shear wave

acquired from a Zero Offset VSP, in the seismic characterization of the Morichal Member,

Oficina Formation, in the Faja Petrolífera del Orinoco, Venezuela.

Poletto, F., & Miranda, F. (2004). Seismic while drilling fundamentals of drill-bit seismic for

exploration,Vol. 35, Italy: ELSEVIER.

Solono,G.E.(2004). Zero offset VSP data processing and attenuation estimates in the oil sands,

Department of Physics, University of Alberta, Edmonton, Alberta, p.106.

Sandroni, S., & Talarico, F. (2011). The record of Miocene climatic events in AND-2A drill core

(Antarctica): Insights from provenance analyses of basement clasts. Global Planet Change, 75,

31–46.

89

Sengbush, R. L., & Al-Fares, M. H. (1987). Optimal processing of VSP data. Monograph presented at the

Offshore Technology Conference, Houston, TX.

Sheriff, R. E. (2002). Encyclopedic dictionary of applied geophysics (4th ed., Vol. 13).Society of

Exploration Geophysicists.

Sheriff, R. E., & Geldart, L. P. (1982). History, theory, and data acquisition; Exploration seismology

(Vol. 1). Cambridge, England: Cambridge University Press.

Siegert, M. J. (2008). Antarctic subglacial topography and ice-sheet evolution. Earth Surface Processes

and Landforms, 33, 646–660.

Speece, M. A., Levy, R. H., Harwood D. M., Pekar S. F., & Powell R. D. (2009). New seismic methods to

support sea ice platform drilling. Scientific Drilling, 7, 40–43.

Stewart, R. R. (2001). VSP: An in-depth seismic understanding. Recorder, 7, 79–83.

Stewart, R. R., Ferguson, R., Miller, S. L. M., Gallant, E., & Margrave, G. (1996). The Blackfoot seismic

experiments: Broad-band, 3C-3D, and 3-D VSP surveys: Canadian Society of Exploration

Geophysicists Recorder, 6, 7-10

Stewart, R., Gaiser, J., Brown, R., & Lawton, D. (2002). Converted-wave seismic exploration: Methods.

Geophysics, 67, 1348–1363.

Stewart, R. R., Gaiser, J. E., Brown, R. J., & Lawton, D. C. (2003). Converted-wave seismic exploration:

Applications. Geophysics, 68, 40–57. doi:10.1190/1.1543193

Stewart, R. R., & Lawton, D. C. (1994, September). 3C-3D seismic polarity definitions. Notes from

CREWES Workshop on 3C-3D Seismic Exploration, Location.

Stewart, R. R., Pye, G., Cary, P. W., & Miller, S. L. M. (1995). Interpretation of P-SV seismic data:

Willesden Green, Alberta: CREWES Research Report, 5.

Stump, E., Gootee, B. F., & Talarico, F. (2006). Tectonic model for development of the Byrd Glacier

discontinuity and surrounding regions of the Transantarctic Mountains during the

Neoproterozoic–Early Paleozoic. In D. K. Fütterer, D. Damaske, G. Kleinschmidt, H. Miller, &

F. Tessensohn (Eds.), Antarctica: Contributions to global earth sciences (pp. 181–190). New

York, NY: Springer, Berlin, Heidelberg. doi:10.1007/3-540-32934-X_22

Sun, Z., & Jones, M. (1993). VSP multi-algorithm shear-wave anisotropy study. CREWES Research

Report 5, 601-622.

Tatham, R. H., & Stoffa, P. L. (1976). V p/V s-A potential hydrocarbon indicator. Geophysics, 41(5), 837-

849.

Tatham, R. H., & Goolsbee, D. V. (1984). Separation of S-wave and P-wave reflections offshore western

Florida. Geophysics, 49, 493–508.

Tatham, R. H., & Campbell, T. A. (2013, June). Rotation of Shear-Wave Components at Non-Normal

Angles of Incidence. In 75th EAGE Conference & Exhibition incorporating SPE EUROPEC

2013.

Thomsen, L. (1999). Converted-wave reflection seismology over inhomogeneous anisotropic media.

Geophysics, 64, 678–690.

Thomsen, L. (2002). Understanding seismic anisotropy in exploration and exploitation (Vol. 5, pp. 1-28).

Society of Exploration Geophysicist.

Toksöz, M. N., & Stewart, R. R. (1984).Vertical seismic profiling: Advanced concept. Geophysical Press,

419.

90

Tsvankin, I., & Lynn, H. B. (1999). Special section on azimuthal dependence of P-wave signatures—

Introduction. Geophysics, 64, 1139–1142.

Tsvankin, I., & Grechka, V. (2011). Seismology of azimuthally anisotropic media and seismic fracture

characterization. In A. B. Editor (Series Ed.), Geophysical Series: No. 17 : Society of Exploration

Geophysicists.

Taviani, M., Hannah, M., Harwood, D. M., Ishman, S. E., Johnson, K., Olney, M., ... & Cantarelli, V.

(2008). Palaeontological characterisation and analysis of the AND-2A Core, ANDRILL Southern

McMurdo Sound Project, Antarctica.

van der Veen, C. J. (2002). Polar ice sheets and global sea level rise: How well can we predict the future?

Global and Planetary Change, 32, 165–194.

Wild, P. (2011). Practical applications of seismic anisotropy. first break,29(5), 117-124.

Wilson, G. S., Roberts, A. P., Verosub, K. L., Florindo, F., & Sagnotti, L. (1998).

Magnetobiostratigraphic chronology of the Eocene–Oligocene transition in the CIROS-1 core,

Victoria Land margin, Antarctica: Implications for Antarctic glacial history. Geological Society

of America Bulletin, 110, 35–47.

Wilson, T. J. (1999). Cenozoic structural segmentation of the Transantarctic rift flank in southern Victoria

Land. Global Planet Change, 23, 105–127.

Wilson, T., & Paulsen, T. S. (2000). Brittle deformation patterns of CRP-2/2A, Victoria Land Basin,

Antarctica. Terra Antartica, 7(3), 287-298.

Winter, D. M., & Harwood, D. M. (1997). Integrated diatom biostratigraphy of late Neogene drillholes in

southern Victoria Land and correlation to Southern Ocean records. The Antarctic Region:

Geological Evolution and Processes, 985-992.

Winterstein, D. F., & Meadows, M. A. (1991). Shear-wave polarizations and subsurface stress directions

at Lost Hills field. Geophysics, 56, 1331–1348.

White, J. E. (2000). Seismic Wave Propagation: Collected Works of JE White (No. 21). SEG Books.

Wyatt, D. K., & Wyatt, S. B. (1984). Determining subsurface structure using the vertical seismic profile.

In M. N. Toksoz & R. R. Stewart (Eds.), Vertical seismic profiling part B: Advanced concepts:

Geophysical Press.

Wonik, T., Grelle, T., Handwerger, D. A., Jarrard, R. D., McKee, A., Patterson, T., ... & Speece, M.

(2008). Downhole measurements in the AND-2A borehole, ANDRILL southern McMurdo Sound

Project, Antarctica. Terra Antarctica, 15(1), 57-68.

U.S Geological Survey, 2007, Landsat Image Mosaic of Antarctica (LIMA): U.S Geological Survey Fact

Sheet 2007–3116, 4 p.

Yatsenko, A. (2008). TECTONIC IMPLICATIONS OF AND-IB AND AND-2A BOREHOLE DATA,

MCMURDO SOUND REGION, ANTARCTICA (Doctoral dissertation, The University of

Utah).

Zattin, M., Andreucci, B., Thomson, S. N., Reiners, P. T., Talarico, F. M. (2012). New constraints on the

provenance of the ANDRILL AND-2A succession (western Ross Sea, Antarctica) from apatite

triple dating. Geochemistry, Geophysics, 13, Q10016. doi:10.1029/201 2GC004357.