17
Australian Journal of Earth Sciences (2001) 48, 439–455 INTRODUCTION This study aims to delineate the spatial distribution and 3-D architecture of the unconsolidated sediment facies outlined by the seabed sampling programs of the National Museum of Victoria (Beasley 1966), the Ministry of Conservation (Seedsman & Marsden 1980), and the Marine and Freshwater Resources Institute (Greilach et al. 1996). An earlier paper (Holdgate et al. 1981) was produced on the Late Pleistocene channels that were identified seismically in the central part of Port Phillip, but the complete study of Port Phillip has only now been finalised. This paper therefore completes the results of these surveys and further aug- ments the studies of nearshore sand deposits undertaken by the Ports and Harbours Division of the Board of Works up until the late 1980s (Buckley & Clark 1987). To our knowledge, little subsequent stratigraphic work has been carried out on the Port Phillip subsurface marine geology, except for the shallow (400 mm) coring of Greilach et al. (1996). Environmental impacts of future developments along the Port Phillip coastline may impinge on the waters and bottom sediments of Port Phillip Bay, and this study should provide a basis for evaluating unconsolidated sedi- ment thicknesses as potential pollutant repositories. GEOGRAPHICAL AND GEOLOGICAL SETTING Port Phillip Bay is located centrally on the southern coast- line of Victoria. It includes an area of 1167 km 2 of almost land-locked water with a maximum length in the north to south direction of 50 km, and a maximum width in the east to west direction of 32 km (Figure 1). On the western side it includes an embayment 25 km wide known as the Outer Geelong Harbour and Corio Bay, and in the north is the embayment of Hobsons Bay (Figure 2). Water depths in Port Phillip generally range from 9.1 to 23.7 m (referred to lowest astronomical tide), except in the south where a large <5.0 m deep area of sand shoals known as the Nepean Bay Bar occurs (Keble 1946) (Figures 2, 3). The sandbars form a veneer of variable thickness resting on indurated Pleistocene aeolianites. They almost close across the entrance to Port Phillip except at The Rip where a narrow 3.2 km-wide opening maintains oceanic exchange between Port Phillip and Bass Strait. In The Rip water depths exceed 60 m locally. Marine geology of Port Phillip, Victoria G. R. HOLDGATE, 1 * B. GEURIN, 2 M. W. WALLACE 1 AND S. J. GALLAGHER 1 1 School of Earth Sciences, University of Melbourne, Vic. 3010, Australia. 2 Department of Natural Resources and Environment, 250 Victoria Parade, East Melbourne, Vic. 3002, Australia. The marine geology of Port Phillip is described in detail, based on data from seismic profiling, vibro- coring and grab sampling. Three major unconsolidated facies can be distinguished: sands and muddy sands peripheral to the present coastline, muds covering the major central region, and channel fills of muds and sands. The first two facies units result from an increase in wave sorting towards the coast, reworking of Tertiary and Quaternary sandstone outcrops around the coast, and a dominant mud supply from river sources into the central area. The distribution and thicknesses of the unconsolidated facies have been augmented by a shallow-seismic program that reveals the thicknesses of the modern sediments overlying an older surface comprised of consolidated clays and sandy clays of Pleistocene or older age. In central Port Phillip, muds and sands up to 27 m-thick have infilled Pleistocene channels cut into underlying consolidated units. Sediments immediately above the channel bases show characteristic seismic patterns of fluvial deposition. The presence of peat deposits together with gas phenomena in the water column suggest organic breakdown of channel-fill deposits is releasing methane into the bay waters. Outside the channel areas, carbon-14 dating indicates that the unconsolidated sediments largely post-date the last glaciation sea-level rise (<6500 a BP), with an early Holocene period of rapid deposition, similar to other Australian estuaries. Stratigraphic and depositional considerations suggest that the undated channel-fill sequences cor- relate with the formation of cemented quartz–carbonate aeolianite and barrier sands on the Nepean Peninsula at the southern end of Port Phillip. Previous thermoluminescence dating of the aeolianites suggests that channel-fill sequences B, C and D may have been deposited as fluvial and estuarine infills over the period between 57 and 8 ka. The eroded surface on the underlying consolidated sediments is probably the same 118 ka age as a disconformity within the Nepean aeolianites. Further estuarine and aeolianite facies extend below the disconformity to 60 m below sea-level, and may extend the Quaternary depositional record to ca 810 ka. Pliocene and older Tertiary units progres- sively subcrop below the Quaternary northwards up the bay. KEY WORDS: marine geology, Port Phillip, Quaternary, sediments, seismic profiles. *Corresponding author: [email protected]

Marine geology of Port Phillip, Victoria

Embed Size (px)

Citation preview

Page 1: Marine geology of Port Phillip, Victoria

Australian Journal of Earth Sciences (2001) 48, 439–455

INTRODUCTION

This study aims to delineate the spatial distribution and 3-Darchitecture of the unconsolidated sediment facies outlinedby the seabed sampling programs of the National Museumof Victoria (Beasley 1966), the Ministry of Conservation(Seedsman & Marsden 1980), and the Marine andFreshwater Resources Institute (Greilach et al. 1996). Anearlier paper (Holdgate et al. 1981) was produced on the LatePleistocene channels that were identified seismically in thecentral part of Port Phillip, but the complete study of PortPhillip has only now been finalised. This paper thereforecompletes the results of these surveys and further aug-ments the studies of nearshore sand deposits undertakenby the Ports and Harbours Division of the Board of Worksup until the late 1980s (Buckley & Clark 1987). To ourknowledge, little subsequent stratigraphic work has beencarried out on the Port Phillip subsurface marine geology,except for the shallow (400 mm) coring of Greilach et al.(1996). Environmental impacts of future developmentsalong the Port Phillip coastline may impinge on the watersand bottom sediments of Port Phillip Bay, and this studyshould provide a basis for evaluating unconsolidated sedi-ment thicknesses as potential pollutant repositories.

GEOGRAPHICAL AND GEOLOGICAL SETTING

Port Phillip Bay is located centrally on the southern coast-line of Victoria. It includes an area of 1167 km2 of almostland-locked water with a maximum length in the north tosouth direction of 50 km, and a maximum width in the eastto west direction of 32 km (Figure 1). On the western sideit includes an embayment 25 km wide known as the OuterGeelong Harbour and Corio Bay, and in the north is theembayment of Hobsons Bay (Figure 2). Water depths in PortPhillip generally range from 9.1 to 23.7 m (referred to lowestastronomical tide), except in the south where a large<5.0 m deep area of sand shoals known as the Nepean BayBar occurs (Keble 1946) (Figures 2, 3). The sandbars form aveneer of variable thickness resting on induratedPleistocene aeolianites. They almost close across theentrance to Port Phillip except at The Rip where a narrow3.2 km-wide opening maintains oceanic exchange betweenPort Phillip and Bass Strait. In The Rip water depths exceed60 m locally.

Marine geology of Port Phillip, VictoriaG. R. HOLDGATE,1* B. GEURIN,2 M. W. WALLACE1 AND S. J. GALLAGHER1

1School of Earth Sciences, University of Melbourne, Vic. 3010, Australia.2Department of Natural Resources and Environment, 250 Victoria Parade, East Melbourne, Vic. 3002,Australia.

The marine geology of Port Phillip is described in detail, based on data from seismic profiling, vibro-coring and grab sampling. Three major unconsolidated facies can be distinguished: sands and muddysands peripheral to the present coastline, muds covering the major central region, and channel fillsof muds and sands. The first two facies units result from an increase in wave sorting towards the coast,reworking of Tertiary and Quaternary sandstone outcrops around the coast, and a dominant mudsupply from river sources into the central area. The distribution and thicknesses of the unconsolidatedfacies have been augmented by a shallow-seismic program that reveals the thicknesses of themodern sediments overlying an older surface comprised of consolidated clays and sandy clays ofPleistocene or older age. In central Port Phillip, muds and sands up to 27 m-thick have infilledPleistocene channels cut into underlying consolidated units. Sediments immediately above thechannel bases show characteristic seismic patterns of fluvial deposition. The presence of peat depositstogether with gas phenomena in the water column suggest organic breakdown of channel-filldeposits is releasing methane into the bay waters. Outside the channel areas, carbon-14 dating indicates that the unconsolidated sediments largely post-date the last glaciation sea-level rise(<6500 a BP), with an early Holocene period of rapid deposition, similar to other Australian estuaries.Stratigraphic and depositional considerations suggest that the undated channel-fill sequences cor-relate with the formation of cemented quartz–carbonate aeolianite and barrier sands on the NepeanPeninsula at the southern end of Port Phillip. Previous thermoluminescence dating of the aeolianitessuggests that channel-fill sequences B, C and D may have been deposited as fluvial and estuarineinfills over the period between 57 and 8 ka. The eroded surface on the underlying consolidatedsediments is probably the same 118 ka age as a disconformity within the Nepean aeolianites. Furtherestuarine and aeolianite facies extend below the disconformity to 60 m below sea-level, and mayextend the Quaternary depositional record to ca 810 ka. Pliocene and older Tertiary units progres-sively subcrop below the Quaternary northwards up the bay.

KEY WORDS: marine geology, Port Phillip, Quaternary, sediments, seismic profiles.

*Corresponding author: [email protected]

Page 2: Marine geology of Port Phillip, Victoria

440 G. R. Holdgate et al.

Shipping channels across the Nepean Bay Bar follow thenaturally scoured tidal depressions of West Channel andSouth Channel (Figure 3). Dredging in the Hobsons Bayarea at the shallower northern end of Port Phillip main-tains deeper shipping channels through to the Yarra Riverand the Melbourne wharf areas between Williamstown andPort Melbourne. Similar dredged channels allow shallowerdraught vessels through the Outer Geelong Harbour andCorio Bay to Geelong.

Offshore reefs and sandbanks occur around the periph-ery of Port Phillip particularly near headlands andpromontories. The central deeper areas form a uniformbasin-shaped depression except in the south where deepwaters abut steeply against the shallow sand shoals of theNepean Bay Bar.

Hydrodynamic models have been studied in detail forthe Port Phillip Bay Environmental Study (Walker 1997;Walker & Sherwood 1997). Measured vector mean currentsin Port Phillip vary with season and location, e.g. south-directed currents along the northeastern side have near-surface and near-bottom currents up to 0.25 m/s and

0.2 m/s, respectively (Walker 1997). Mud plumes from YarraRiver flood discharges have been observed to extend as farsouth as Brighton. In contrast, the central bay area recordsvaridirectional currents with speeds of less than 0.1 m/s.Rates of water exchange with Bass Strait have also beencalculated and vary between 10 days in the south to >280days in the north and in Corio Bay (Walker 1997; Walker &Sherwood 1997). Restrictions to tidal flow at The Rip producecurrents of up to 3.9 m/s. Tide ranges in Port Phillip aresmall (less than a metre at spring tides). The ebb and flood-tide peaks between the entrance and the north end inHobsons Bay are delayed by up to 3 hours (Jones 1980; Bird1993). Ocean swells entering Port Phillip Heads are rapidlydiffracted and weakened. Therefore, other than at theentrance, Port Phillip wave heights of up to 1–1.5 m aregenerated largely by winds blowing across the bay, limitedby available fetch.

The only large freshwater input to Port Phillip is theYarra River. This river has a mean flow average of 21 m3/s(Walker 1997), but has been measured at flood times up to500 m3/s (SECV 1970). Smaller freshwater inputs include theKororoit Creek at Williamstown and the Werribee River atPoint Cook. In consequence, Port Phillip salinity decreasesnorthwards from 35‰ at the Rip to 33‰ near the YarraRiver mouth. Particulates (total non-filterable residue)transported into Port Phillip primarily from the YarraRiver are estimated by Harris et al. (1996) to be of the orderof 51 000 t/y, although most input is believed to occur asshort flood events. Total input is believed to be approxi-mately 85 000 t/y (including from the atmosphere, butexcluding coastal erosion: Walker & Sherwood 1997).

The Port Phillip Sunklands are part of the Port PhillipBasin and Ballan Graben (Abele 1988), the initial formationof which can be traced to the development of an intracra-tonic rift across the southern margins of the continent.This is a precursor to the rifting system along which LateCretaceous separation of the Antarctic and Australianlandmasses finally occurred. The subparallel Rowsley andSelwyn Fault Systems that border Port Phillip (Figure 1)and the east–west Curlewis Fault control the limits to aninfilled part of the ancient rift. The oldest infill sediments—the Lower Cretaceous Otway Group—form the basal sedi-mentary sequence in the southern part of the bay.Subsequent crustal collapse between the two main faultsresulted first in the deposition of Eocene terrestrial sedi-ments (the Werribee Formation), and Older Volcanics; thena marine incursion resulted in the deposition of Oligo-Miocene marl and limestone (the Torquay Group). Finallya return in the Pliocene to marginal marine and terrestrialregimes including volcanic lava flows resulted in a cappingof younger sediments including the Moorarbool ViaductFormation, Brighton Group, Newer Volcanics, WannaeueFormation and Bridgewater Formation (Figure 1). In all,over 1 km of post-Early Cretaceous sediments and volcanicswere deposited in the southern regions of the Port PhillipBasin.

Tectonic uplift in the Mid to Late Tertiary raised theBellarine Peninsula and maintained the topographic ele-vation of the Mornington Peninsula. The subsequentdevelopment of the Yarra and Werribee deltas during theLate Tertiary and Quaternary produced the present-daytopography of Port Phillip. Recent sea-level and climatic

Figure 1 Geological map of Port Phillip area, Victoria (adaptedfrom Geological Survey of Victoria 1:1 000 000-scale geologicalmap 1993).

Page 3: Marine geology of Port Phillip, Victoria

Marine geology of Port Phillip, Vic. 441

changes have modified, but not greatly changed thismorphology.

Early structural features together with more recent geo-logical events have determined the present shape of PortPhillip. The western side between Geelong and Melbourneis a basalt plain of low relief modified in the coastal regionby beach deposits and sediments of the Werribee and YarraRiver deltas. The Bellarine Peninsula to the south is a horststructure consisting of a thin cover of Lower Tertiary sed-iments and volcanics capping arkoses of the Otway Group.In contrast, the eastern side of Port Phillip has a greaterelevation with cliffs and bluffs formed of Tertiary and oldersediments. Sediments of the Brighton Group are found incliffs near Melbourne; the more resistant Black RockSandstone forms steep bluffs, while the occurrence of theeasily eroded Red Bluff Sandstone has allowed embaymentsto develop. The coastline between Beaumaris andFrankston is a long continuous crescent-shaped beach withassociated low dunes of Pleistocene to Holocene age.Between Frankston and Dromana, movement on the

Selwyn Fault has resulted in the development of high cliffsin the uplifted Tertiary sediments. Three granite bodiesform the main topographic features on this side of the bayat Olivers Hill, Mt Martha and Arthurs Seat (Figure 3). TheNepean Peninsula practically closes Port Phillip to thesouth, and consists of Bridgewater Formation dune lime-stone overlying Tertiary carbonate sediments.

PREVIOUS WORK AND OBJECTIVES OF THISSTUDY

Investigations into the bottom sediments of Port Phillipwere completed by Beasley (1966, 1969, 1971), Link (1967),Buckley and Clark (1987) and Greilach et al. (1996). Adetailed map of the distribution of sea-bottom sedimentsis reproduced on Figure 3 (adapted from Buckley & Clark1987). Investigations into the distribution and reserves ofsand and its coastal movements have been undertaken by the Ports and Harbours Division (Public Works

Figure 2 Bathymetric map of Port Phillip, Victoria (adapted from Commonwealth of Australia Port Phillip 100 000-scale map).

Page 4: Marine geology of Port Phillip, Victoria

442 G. R. Holdgate et al.

Department/Ministry of Transport) and the GeologicalSurvey (Geurin 1974a, b; McCutcheon 1979; Smith 1981;Buckley & Clark 1987). Other reports include seismic andcoring results along the Esso ethane pipeline (Esso 1971),seismic profiling near Mordialloc (Dolan 1973), seismic pro-filing in the northern part of Port Phillip (Walsh 1978), andsediment distribution and movement in Corio Bay(Seedsman & Marsden 1980). Additional sediment charac-teristics over the top 5 cm of seabed cores (including per-meability, consolidation, viscosity, settling velocity, grainsize and bulk density) were determined by Greilach et al.(1996) for environmental studies in Port Phillip.

The objective of the 1973–1982 Geological Survey ofVictoria study was to correlate the known distribution ofthe surface sediments with the aid of a sub-bottom seismicprofiler (Figure 4), and to provide additional informationon the immediate subsurface geology of Port Phillip andits relationships to the outcrop geology onshore. Followingthe sub-bottom seismic survey, targets were selected fromthe seismic records and were drilled using a vibrocorer toobtain samples to help identify the main seismic events(Figure 4; Tables 1, 2). Additional vibrocore sites were

drilled for the Ports and Harbours investigation of theperipheral sandy areas for beach-restoration projects. Theresults are referred to in general in this report. Moredetailed core descriptions and results from this work arelisted in the references (Geurin 1974a, b; McCutcheon 1979;Smith 1981; Buckley & Clark 1987) and Table 2.

FIELD PROGRAM TECHNIQUES

Seismic surveys

Workboats were loaned and crewed by the former Ports and Harbours Division of the Public Works Department.Position fixing was by use of a Decca TrisponderNavigation System that gave coordination accurate towithin 1 m or less. These were later converted intoAustralian Map Grid (AMG) coordinates. Sextants and com-passes were used in the northern end of Port Phillip. Thesegave less accurate results and were only used between fixedpoints. Boat speeds during the seismic surveying variedbetween 5 and 9 km/h. Generally, seismic surveys were not

Figure 3 Seabed sediment distribution map of Port Phillip showing the principal textural classes (adapted from Buckley & Clark 1987).

Page 5: Marine geology of Port Phillip, Victoria

Marine geology of Port Phillip, Vic. 443

carried out in seas where wave heights exceeded 1 m as thisaffected the clarity of the records.

The sub-bottom profiler comprises three parts of asystem manufactured by Ocean Research Equipment Inc.—SO Model 132A over-the-side transducer; Model AOTransceiver; and Model 4600 Graphic recorder. The4-transducer array was fixed over the side of the boat to hang 2.4 m below the surface. Its swivel mountings enable it to maintain a vertical position irrespective ofthe motion of the boat. The transducer has a maximum output of 10 kW, and adjustments of the pulse length,frequency and filters were required to maintain the best image possible. Increasing frequency decreasedpenetration and increased resolution, and vice versa.In general, satisfactory results were obtained using a sweep time of 125 ms, a pulse length of 1.0 ms, and a band plus filter width of 1.0, and frequency of 4.5 kHz.The resulting seismic profiles appeared to give better def-inition to the more unconsolidated Holocene and UpperPleistocene strata, compared to the deeper penetrationboomer seismic results from University of Melbourne data

(Walsh 1978). The latter appeared to show better definitionin the more consolidated units below.

Thirty-six seismic runs totalling 656 km were madeduring 18 field days in 1977 and 1978 (Figure 4), and approxi-mately 85% of this distance had trisponder position fixing.Seismic penetration to 50 ms two-way-time (TWT) (approxi-mately 37.0 m) below the seabed was obtained in favourablesub-bottom conditions, such as soft muds, with resolutionof single beds down to 1 ms (0.7 m). In sandy conditions pen-etration was reduced, and often the sea floor was totallyreflective to the seismic frequency, producing a series ofsea-floor multiples. Consolidated clays were similarly diffi-cult to penetrate with a few exceptions, and hard rock wastotally reflective. However, most of the study area consistsof soft muds overlying consolidated sediments. The latterhave been deeply eroded in the centre of Port Phillip and have later been infilled by soft muds with some sandlayers.

Conversion of seismic TWT to depth was assumed to be approximately similar to a signal velocity in water of1.5 km/s. The velocity of unconsolidated sediments is

N

WerribeeWerribeeW

Portarlington

BELLARINEPENINSULA

BASS STRAIT

NEA

PEINSUN

LA0 5 10 15 km

MORNIN

GTON PENIN

SULA

Dromana

Mornington

Mordialloc

Brighton

Altona

MELBOURNEarra

erRiver

Werribee

Werribee

W

rR

iver

***

****

***

**

*

*******

*

***

*

****

*

*

***

**

11E

11F

7D7C7B

7A

10C10B

C145,240

C148,110

4

10A

8B

8D

9B9A

6A

6Bi*

5B*5B*5C

E3

C14

E3

4

C145,990*

ESSOPIPELINE

R2222

R2

R30

R30R25

R24

R24

R14

R17

R15

R32

R19

R32

R20

21A

R20

R19

R20

R18

R16R16R18

R20

R15

R12

R11

R11

R14

R11R21

R19

R17

R34

R10

R31

Portarlington depression

The

Rip

145 00

38 00

145 00

38 00

MudIsland

***AD1R = n

)

Figure 4 Location of the GeologicalSurvey Port Phillip sub-bottom pro-filing seismic runs and core locali-ties mentioned in text. Outline ofthe subsurface Portarlingtondepression indicated together withthe location of Figure 5. 14C dates (a BP) from cores indicated. E3,38, Esso pipeline cores; AD1,Australian Dredging core. Figureadapted from Holdgate et al. 1981.

Page 6: Marine geology of Port Phillip, Victoria

444 G. R. Holdgate et al.

generally taken as being between 1.5 and 1.8 km/s. At most coring sites the assumed depth using a velocity of1.5 km/s was found to agree with the depths found in thecores.

Drilling and sampling surveys

The surface sediments around the peripheral sandy areasof Port Phillip have been extensively sampled by grabsampling, vibrocoring, and dredging and diver sampling aspart of the Ports and Harbours Port Phillip Coastal Study.The same coring equipment was used in this survey. It con-sisted of a 32 mm-diameter stainless-steel tube below whichan air-driven vibration head was mounted. The drill waslowered to the sea floor, and was operated via flexible hosesfrom the boat-based vacuum pump and compressor. Thevacuum assists penetration of the substrate and retentionof the core. The core was retrieved by vibration out of thecore tube into a core liner on board the vessel. The maxi-mum core length obtained in Port Phillip was 5.0 m,although most cores finished in stiff clay prior to thisdepth. The corer is capable of cutting through thick shellsand soft rock.

Twenty-eight sites were chosen from the sub-bottomprofiling for coring for this survey, with a total recovery of49.31 m (Table 1). The average core length was 1.76 m witha range between 0.53 and 4.20 m. The locations of core sitesare shown on Figure 4. Fifteen samples were submitted for

nannofossil studies to Russell Wilks (at the time with theDepartment of Geology, University of Melbourne), and fivecarbon-14 dates were carried out by the Australian NationalUniversity (ANU) and University of Sydney (SUA).

Reports by Esso Standard Oil (Aust) Ltd, described theresults of a seismic and drilling program involving 80 boresto a maximum depth of 6.0 m along the line of the Essoethane pipeline. The descriptions of the strata obtainedhave been useful for this study (Esso 1971). Esso reports onresults from seismic profiling are summarised by Jackman(1971). Additional samples were obtained from theAustralian Dredging Company following a drilling programalong the Port Melbourne Channel and are described indetail by Holdgate (1977).

RESULTS

Peripheral sands

Sand occurs in a band around the periphery of Port Phillipand on the Nepean Bay Bar (Figure 3). Around most of PortPhillip the sand occurs in two main zones, nearshore sand-bars and offshore sand, that are separated by major breaksin slope. Beaches occur in embayments and at the foot ofmany of the cliffs and bluffs around the coastline. Thebeach faces (as formed by wave uprush) usually have a slopesteeper than 1 in 10.

Table 1 Geological Survey vibrocore site locations, nannofossil samples and stratigraphy, Port Phillip.

Vibrocore AMG Coordinates Water Core Nannofossil Nannofossil Coode Island Fishermanssite no. Easting Northing depth (m) recovery (m) sample depth (m) age Silt (m) Bend Silt (m)

1A 317356 807589 13.7 1.31 1.31 tr, nd 0–1.31 nr1B ? ? 13.0 0.53 – – 0–0.53 nr1C 317589 808879 21.2 1.00 0.96 nd 0–1.00 nr2A 317106 806229 16.4 0.94 – – 0–0.50 0.50–0.942B 317415 805673 17.1 2.40 – – 0–2.40 nr4A 312645 803125 18.5 0.62 0.60 barren 0–0.37 0.37–0.624B 312757 802572 22.6 0.76 0.75 tr, nd 0–0.76 nr4C 312901 802523 24.6 3.07 3.06 barren 0–3.07 nr5B 312372 798221 26.0 0.77 0.74 tr, nd 0–0.47 0.47–0.775C 312599 798324 32.9 3.75 – – 0–3.75 nr6A 308957 791889 30.1 0.66 – – 0–0.40 0.40–0.666B 311500 791917 32.9 0.75 0.74 tr, nd 0–0.56 0.56–0.756Bii 311511 791936 35.2 2.80 2.79 Holocene 0–2.80 nr7A 309862 782497 37.7 2.00 2.00 barren 0–0.89 0.89–2.007B 311107 782821 39.4 0.75 0.74 tr, nd 0–0.20 0.20–0.757C 312660 783535 39.7 1.42 1.41 tr, nd 0–1.42 nr7D 317743 784786 36.3 3.80 – – 0–3.80 nr8B Corio Bay ? 2.06 – – 0–1.55 1.55–2.068D Corio Bay 17.4 1.14 1.13 barren 0–0.93 0.93–1.149A 316474 790963 31.0 3.17 3.16 barren 0–3.17 nr9B 320342 792322 27.1 3.08 – – 0–2.93 2.93–3.0810B 308936 778224 41.1 1.33 1.32 barren 0–1.19 1.19–1.3310C 309485 778431 41.0 4.04 4.00 tr, nd 0–4.04 nr11E 325991 767651 30.1 0.61 – – 0–0.41 0.41–0.6111F 326146 767999 30.4 0.71 – – 0–0.55 0.55–0.7111G 326693 5768509 8.8 1.32 – – 0–1.04 1.04–1.3221A 310564 786177 37.7 0.57 0.56 tr, nd 0–0.48 0.48–0.57CB1 Corio Bay Buoy No.11 10.3 4.20 – – 0–4.20 nrCB2 Corio Bay 200 m N Buoy 3 ? 1.11 – – 0–1.11 nr?

tr, trace; nd, no age; nr, not reached.

Page 7: Marine geology of Port Phillip, Victoria

Marine geology of Port Phillip, Vic. 445

NEARSHORE SANDBAR ZONE

The nearshore sandbar zone extends from sea-level toapproximately 2 m below the Australian Height Datum(AHD). It generally has a slope of between 1 in 80 and 1 in300, and may be narrow with only one or two parallel bars,or very wide as in the Rosebud–Rye area where it extendsoffshore approximately 500 m and contains numerous barsin a complex pattern. The pattern and positions of the sand-bars are basically stationary, being formed in response tothe wave regime, but they may change orientation and pat-tern slightly in response to seasonal variations. The near-shore sandbar zone usually consists of fine sand, whereasthe beach and offshore sands are considerably coarser.The sandbar sand is generally thicker than 3 m (deepestcore in the sandbar zone) unless there is a shallowsubsurface occurrence of ‘basement’ material (Tertiary orQuaternary). The sandbars are considered to be the areasof greatest sand mobility as they experience the most waveactivity and contain the finest sand (Figure 3).

OFFSHORE SANDBAR ZONE

The outer edge of the nearshore sandbar zone dropssteeply to the much flatter offshore sand zone. This usuallyhas a slope of between 1 in 800 to 1 in 1000. No further majorchange in slope occurs at the outer limit of this zone wherethe sediments change to muds. Wave action on the sea floorin this zone is slight, being confined mainly to small oscil-latory currents generated by the passage of storm wavesand probably causing some winnowing of fines. Withincreasing water depth the sand becomes muddy, indicatingnegligible current winnowing.

The offshore sand zone differs on the east and westcoasts in Port Phillip. On the west coast it contains homo-geneous medium to fine sand, whereas down the east coastit ranges from very coarse to fine sand, which is taken toindicate more variety in grainsize of the source rocks. Inaddition, on the east coast, the zone contains pockets ofcoarse to very coarse sand that lie parallel to the coast fromBrighton to Dromana, between 500 m and 2 km offshore, andgenerally occur between 7 and 11 m water depths (Figures2, 3). These deposits are considered to comprise lag materialderived from wave erosion of Tertiary rocks during therising Holocene sea-level.

Most cores in the offshore sand zone penetrate throughthe sand into ‘basement’. The sand is generally less than1 m thick with a maximum thickness of 3 m off MarthaPoint. The sands generally show little stratificationalthough in many of the cores there visually appeared tobe an increase in clay content with depth.

The physical regime on the Nepean Bay Bar differs fromthe other offshore sandbar zones. Unconsolidated sedi-ments on the bar display wave and ripple morphologies andare constantly sorted by tidal currents. Wave action is ofsecondary importance. The bar comprises a sand-coveredshelf overlying Quaternary calcarenite, with four deeptidal channels radiating from the 60 m-deep Rip. SouthChannel is the deepest channel and is 12–31 m deep. It is themain shipping channel for larger vessels. On top of theNepean Bay Bar the water is generally less than 5 m deep

(Figure 2). The sand cover of the Nepean Bay Bar is up to6 m thick and displays tidal-delta morphology. The sand isfine grained (<2.0 �), with a 50% calcareous content, andappears to be largely derived from the dune limestoneunderneath. Sand within the channels is similar, but con-tains a coarser biogenic component.

AGE DATING OF THE PERIPHERAL SANDS

The peripheral sand facies are considered equivalent to thePort Melbourne sands in the Yarra Delta (Neilson 1967). Inaddition, a number of carbon-14 dates were carried out atthe time of the surveys, but these remain uncorrectedwithin the context of recent procedures detailed by Stuiveret al. (1998). However, we feel the dates are important, andare representative of the spread of ages likely from thesestrata. In view of the absence of more recent (corrected)dates, they lie within the constraints of accuracy requiredfor this stratigraphic study (e.g. tree-ring corrections forHolocene dates would only alter by a few hundreds ofyears). Two carbon-14 dates of shells at the base of theperipheral sands overlying consolidated basement havebeen obtained. Bowler (1966) quoted a 14C date (N-155) of5990 � 160 a BP on a bivalve Anadara shell at the base of 73 cm of marine sands (14.3 m water depth) off Sandring-ham. A Geological Survey date (SUE 995) gave 5240 �

135 a BP for shells at the base of 1.2 m of marine sands overlying stiff basement clay near Martha Point (17 mwater depth) (Figure 4).

Black organic sediments containing plant fragmentswere found below marine sands in three cores, and werecarbon-dated. Near Rosebud (8.5 m water depth) compactedquartz sands below 1.0 m of organic-rich sands gave a 14Cdate of 8110 � 170 a BP (SUE 993). Near Sorrento (4 m waterdepth) organic silts below 3.5 m of sand, returned a 14C ageof 7010 � 190 a BP (SUE 992) (Figure 4). Another core 5 maway intersected dune limestone at only 1.7 m water depth,suggesting the carbonaceous layers occur in a depressionin the dune limestone. Offshore from Altona in Core 4B(22.6 m water depth), plant material from peaty sandsbelow 0.64 m of fine to coarse shelly sands gave a 14C age of8290 � 280 a BP (SUE 1126) (Figure 4).

The dated carbonaceous layers are considered to bedeposits from lagoonal, intertidal and other peat environ-ments formed close to the contemporary sea-levels duringthe early part of the Holocene marine transgression. Theoverlying marine sandy sediments were therefore depositedduring the last 6500 (14C years) BP highstand of southeastAustralia (Thom & Roy 1985).

Older peripheral sand units may be represented in theoffshore St Kilda area where distinctly stratified sandsoccur in 5–6 m water depth. Closely spaced coring estab-lished the existence of three sand layers comprising anupper 1.5 m-thick grey medium shelly sand overlying 20 cmof discontinuous mottled yellow to dark-grey soft to firmshelly clay, overlying orange to brown clayey sand con-taining iron-stained bivalve shells. This rested upon stiffclays of the basement surface. Carbon dating of the iron-stained shells returned an age of 34 900 � 2000 a BP (SUA994). As this date is near the limit of practical 14C detection and the possibility of contamination by modern carbon

Page 8: Marine geology of Port Phillip, Victoria

446 G. R. Holdgate et al.

Table 2 Geological Survey Port Phillip core descriptions. Table 2 (cont.)

Vibrocore 1A0–0.59 m: dark blue-grey to black sticky silty mud.0.59–0.99 m: grey brown shelly silty and sandy mud.0.99–1.31 m: grey silty mud, slightly shelly, sticky.

Vibrocore 1B0–0.12 m: grey brown sandy mud, slightly shelly.0.12–0.26 m: grey muddy sand with abundant shells.0.26–0.34 m: light brown very shelly muddy sand.0.34–0.53 m: grey brown very shelly sticky clay.

Vibrocore 1C0–0.95 m: dark grey silty and sandy mud with scattered shells.0.95–1.00 m: light grey-brown medium clean sand.

Vibrocore 2A0–0.25 m: grey brown sandy mud, slightly shelly.0.25–0.50 m: dark grey very shelly quartz sand with Anadara

shells at base.0.50–0.95 m: mottled brown and grey sticky clay, traces of

shelly fossils (contaminants?).Vibrocore 2B

0–1.56 m: grey slightly silty mud, sticky, traces of shells.1.56–2.06 m: grey sandy mud with some shells.2.06–2.40 m: grey mud, sticky, traces of shells.

Vibrocore 4A0–0.37 m: brown muddy sand with shells, grading down to

shelly quartz clean sand.0.37–0.41 m: hard cemented band of iron-stained carbonate

nodules.0.41–0.62 m: mottled grey and light brown stiff clay.

Vibrocore 4B0–0.27 m: grey brown medium to fine quartz sand with shells.0.27–0.65 m: coarse quartz and very shelly sand.0.65–0.74 m: dark brown peaty coal, some coarse quartz sand

on top.0.74–0.77 m: coarse brown sand, slightly muddy, no shells.

Vibrocore 4C0–1.57 m: grey muddy sand and sandy mud with some shells,

grading down to:1.57–3.07 m: grey sticky silty mud with minor shells.

Vibrocore 5B0–0.65 m: brown sandy mud with disseminated shells,

particularly near the base.0.65–0.77 m: blue-grey stiff clay, brown mottling, traces of

shells (contaminants?).Vibrocore 5C

0–0.32 m: brown-grey sandy mud with shells.0.32–3.75 m: medium dark grey silty mud and mud with minor

shells.Vibrocore 6A

0–0.40 m: brown muddy shelly sand, quartz sand, very shelly at the base, oysters.

0.40–0.66 m: blue-grey and brown mottled stiff clay containing brown stained quartz pebbles, traces of shells (contaminants?).

Vibrocore 6B0–0.56 m: brown muddy shelly quartz sand.0.56–0.64 m: blue-grey mottled sandy clay.0.64–0.75 m: blue-grey and brown mottled stiff clay.

Vibrocore 6Bii0–2.80 m: sticky brown-grey (top) grading to dark-grey silty

mud, disseminated shells.Vibrocore 7A

0–0.89 m: grey-brown sandy mud, abundant shells including oysters.

0.89–2.00 m: grey-brown mottled stiff silty clay and clay,bands of iron-stained stiff clays and sands every 0.7 m.

Vibrocore 7B0–0.20 m: grey-brown silty mud.0.20–0.40 m: dark-blue grey and brown stiff clay with an

oxidised clay band at the top.Vibrocore 7C

0–0.72 m: blue-grey-brown silty mud with rare shells.0.72–1.16 m: brown sandy mud with shells at the base.1.16–1.42 m: stiff dark grey clayey mud.

Vibrocore 7D0–3.07 m: medium greeny-grey-brown mud with traces of

shells.3.07–3.58 m: grey sandy mud grading down to muddy quartz

sand with shells at the base.3.58–3.80 m: sticky firm silty mud, dark grey, with traces of

shells, becoming brown mottled at the base.Vibrocore 8B

0–0.47 m: brown shelly sand.0.47–0.96 m: dark grey shelly mud with weathered zone at base.0.96–1.55 m: blue-grey sticky mud.1.55–2.06 m: mottled blue-grey and brown stiff clay, nodules of

carbonate cemented sands.Vibrocore 8D

0–0.38 m: brown shelly mud.0.38–0.93 m: blue-grey mud, rare shells more abundant

towards the base.0.93–1.14 m: mottled blue-grey and brown stiff clay, weathered

zone at top.Vibrocore 9A

0–3.17 m: brown-grey slightly shelly mud, blue-grey at depth.Vibrocore 9B

0–2.93 m: brown and blue-grey mud, scattered and layered shelly horizons.

2.93–3.08 m: mottled grey and brown stiff clay, weathered at the top.

Vibrocore 10B0–1.19 m: grey and blue-grey shelly mud, very shelly at the

base with oysters.1.19–1.33 m: red-brown mottled stiff clay and sandy clay,

weathered zone at top.Vibrocore 10C

0–0.30 m: brown shelly mud.0.30–3.26 m: blue-grey mud, some shells, firmer towards base

with less shells.Vibrocore 11E

0–0.41 m: brown shelly muddy sand, abundant shells at the base.

0.41–0.61 m: greeny-blue-grey and purple-grey mottled stiffclay and clayey silt, weathered zone at the top.

Vibrocore 11F0–0.55 m: brown shelly muddy sand, abundant shells at the base.0.55–0.71 m: blue-grey and brown mottled stiff clay.

Vibrocore 11G0–1.04 m: brown shelly fine sand and muddy sand, shelly and

greyer at the base.1.04–1.32 m: mottled dark grey and brown stiff sandy clay,

clayey sand over the top 7 cm, ligneous coal band at 1.10 m.Vibrocore 21A

0–0.48 m: brown shelly silty mud.0.48–0.57: grey-brown and blue-grey mottled stiff clay.

Vibrocore Corio Bay Core 1 buoy 11 (CB1) (Hopetown Channel)0–3.60 m: brown and blue-grey sandy mud, shelly and sandy

towards the base.3.60–3.94 m: light grey-white sticky mud, mottled bands of

brown mud.3.94–4.20 m: dark blue-grey sticky mud.

Page 9: Marine geology of Port Phillip, Victoria

possible, the date is taken as a minimum only. The iron-staining of the shells suggests that they were exposed tosubaerial weathering at an earlier date and they may rep-resent a deposit from an earlier transgression.

BASEMENT UNDER THE PERIPHERAL SANDS

The peripheral sands are mainly reflective to the seismicfrequencies used by the sub-bottom profiler, and thick-nesses were identified from cores. Basement sediments inthe cores usually consisted of firm to stiff sandy clays orclayey sands around the edges and calcarenite under theNepean Bay Bar. Generally only 10–20 cm of the basementsediments were retrieved. This material generally showedevidence of subaerial weathering with mottling and smallroot fragments. The shallow penetration and often weath-ered nature of the basement material made geologicalidentification difficult, but the basement material in mostof the cores in the peripheral sandy areas appears to beweathered Tertiary sediments similar to those occurringin the nearest adjacent coastal cliffs.

The occurrence of Tertiary basement could be verifiedin the vicinity of submarine outcrops of Tertiary rocks.For example, offshore from St Kilda (5.5 m water depth) and Sandringham (9 m water depth) basement comprisedcompact quartz sand with root fragments. BetweenBrighton and Black Rock (offshore from the TertiaryBrighton Group outcrops: Figure 1), cores recovered stiffmottled ochre and grey sandy to silty clays similar to the Black Rock Sandstone (Abele 1988). Other coresrecovered green to greyish firm clayey fine to coarse sand similar to the Red Bluff Sandstone (Abele 1988).Offshore from Mentone, basement material of a dark green-brown glauconitic clayey fine sand appears similarto Black Rock Sandstone outcropping at beach level at Red Bluff. On exposure this material rapidly oxidised to a ferruginous sand. Offshore from Seaford (11.6 m waterdepth), basement material was recovered comprising stiff mottled orange to grey clayey medium sand with rootfragments and hard grey concretionary limestone similarto the Miocene Balcombe Clay (Abele 1988). Similarmaterial offshore from Dromana and Rosebud basementrock contained small shell fragments and here it may bePleistocene in age similar to the Bridgewater Formationsand ridges that underlie the Nepean Peninsula (Mallett &Holdgate 1985).

Data using the University of Melbourne boomer seismicprofiles tend to confirm the presence of Tertiary sediments,considered to be probable Brighton Group, immediatelyunderlying the peripheral sands in the northeastern parts

of Port Phillip (Walsh 1978). In the northwestern parts ofPort Phillip, a totally reflective hard surface beneath theperipheral sands was interpreted as Newer Volcanicsbasalt (Walsh 1978).

Central muds

A layer of soft-grey to dark-grey mud containing some silt,sand and marine shells occupies the central 500 km2 of PortPhillip (Figure 3). This thickens to 9.1 m on seismic at thecentre of the bay. At the margins, the mud layer includessilty beds and is a facies equivalent of the peripheral sands.Good acoustic penetration was obtained through the mudlayer using the sub-bottom seismic profiler. The seismicresults show the central muds disconformably overlie aneroded surface consisting of acoustically reflective clay, orin some cases an eroded surface of harder rocks, such asbasalt or sandstone.

The mud layer occurs as an elongated pear-shaped area(Figure 3) covering central Port Phillip with a maximumeast–west width between Frankston and Portarlington of23 km, and a maximum north–south length betweenWilliamstown and Rosebud of 48 km. In the northern partof Port Phillip north of seismic line Run 24 (Figure 4) thewestern margin of the central muds is limited by the rapidshallowing of acoustic basement that forms a small escarp-ment on the sea floor, and the western edge to an infilledchannel. North of Run 22 the mud layer appears to beconfined to the channel, and becomes silty in the infilledchannels of Hobsons Bay.

Coring shows the central muds vary in thickness froma few centimetres to over 6 m. Sub-bottom profiling indi-cates that the mud thickness is quite variable as it fillschannels and other topographic features eroded in theunderlying harder formations. An average thickness of4.4 m is maintained over most of the study area, but incentral Port Phillip where it forms the upper layer of aninfilled system of channels, its thickness from profiling canbe up to 9.1 m.

Colour and textural variations to the central muds occurboth laterally and vertically. Generally they become siltytowards the peripheral sandy areas. Detailed analysis of14 core samples (at 5–10 cm core depth) in this area suggeststhe mean grain size is 3.0 � (fine to very fine sand), with asilt component ranging between 10 and 45% (Greilach et al.1996). In the infilled channel areas it becomes darker withdepth, and more consolidated with fewer marine shells, andmay include sandy beds (e.g. core 7D: Figure 4). The basalcontact with the stiff clays of the underlying basement sur-face can be extremely shelly in places. A separate area ofsimilar soft mud covers an area of 100 km2 in the GeelongOuter Harbour that is divided into two basins by a rockyand sandy ridge between Point Wilson and Curlewis. Theeastern basin is a subsurface depression filled by up to36 ms (26 m) of soft, acoustically similar muds to the cen-tral Port Phillip. Its extent defined from seismic profilingis shown on Figure 4, and a typical line across it betweenPoint Richards and Point Wilson (part of Run 33) is shownon Figure 5. The reflective basement beneath this depres-sion can be traced westwards up to basalt outcrops on thePoint Wilson – Curlewis Ridge, suggesting that basaltunderlies the whole of the depression that we refer to as

Marine geology of Port Phillip, Vic. 447

Table 2 (cont.)

See Table 1 for location and other details.

Vibrocore Corio Bay Core 2, 200 m north of buoys 3 and 4(CB2)

0–0.74 m: brown-grey silty mud, abundant shells at the base.0.74–1.00 m: blue-grey sticky clay with shells, some brown

mottling.1.00–1.11 m: brown sandy clay, disseminated coarse quartz

sands.

Page 10: Marine geology of Port Phillip, Victoria

the Portarlington depression. It appears unconnected toother Upper Quaternary channel features (see below) andits origins are uncertain. We consider it could possiblyrepresent a volcanic maar infilled by Upper Pleistocene toHolocene sediments.

AGE OF THE CENTRAL MUDS

R. Wilks (pers. comm. 1981) examined cores of the centralmud for nannofossils. Of the eight samples submitted, Core6Bii contained age diagnostic species 2.8 m below theseabed (35.2 m water depth) including the key zonal fossilsGephyrocapsa oceanica and Emiliania huxleyi, which indi-cate that the sample comes from the youngest Quaternarynannofossil zone with an age range of between 0.27 and 0 Ma. A number of the other samples contained rarenannofossils from which no age was obtained. The marinemacrofossils, which are ubiquitous in the mud, are also gen-erally indicative of a fairly recent age and include thebivalve Anadara trapezia, thought to have made its firstappearance in Victorian coastal waters during the lastinterglacial stage at around 120 ka (Gill 1972).

The central muds are considered to be equivalent in ageto the Coode Island Silt that occurs in the Yarra Delta(Neilson 1967). Here it is described as soft dark organic finesandy and silty clays that, in bores, occur down to 30 mbelow sea-level. Various uncorrected carbon-14 dates havebeen obtained from plant material at its base. They includea tree stump at Spencer Street (at –19 m) dated at 8300 �

210 a BP (Y-151), and wood at –28 m dated at 12 810 � 210 a BP(GaK-1101) (both dates by Gill 1968). The Coode Island Siltand, by inference, the central muds, have been depositedduring or since the Holocene transgression.

BASEMENT UNDER THE CENTRAL MUDS

Basement under the central muds was defined both by cor-ing and using the sub-bottom seismic profiler. In the lattercase, it appears to be acoustically reflective to the seismicfrequencies used. In the central mud areas, basement liesat a deeper level than under the peripheral sands and its

surface is far more irregular. Acoustically, it was detectedbelow a variable thickness of soft penetrable muds and channel-filling sediments. On the seismic profiles thebasement surface appears to be quite irregular and this istaken to indicate that it has been formed by periods ofsubaerial erosion. In rare instances, some reflectivemarkers may appear within the basement, but these arelocalised and discontinuous. In the central Port Phillip this

448 G. R. Holdgate et al.

Figure 5 Sub-bottom profile part of Run 33 between Point Wilsonand Point Richards, Corio Bay,showing the southern half of thePortarlington depression. Verticalscale in milliseconds (two-way-time)where 10 ms � 7.2 m. For location ofseismic section see Figure 4.

Figure 6 Map of the Late Pleistocene channel system in PortPhillip, also showing the relevant parts of the sub-bottom profileseismic lines shown in Figures 5, 8–11 (adapted from Holdgateet al. 1981).

Page 11: Marine geology of Port Phillip, Victoria

erosion surface includes former river channels that havebeen infilled by soft sediments to a thickness of 38 ms(27.7 m).

Coring could only penetrate to basement outside themain infilled channel areas. Here it was found to consist ofcompacted and mottled brown, yellow and grey stiff claysor sandy and silty clays, sometimes with thin beds of clayeysands and calcareous nodules. The top surface layer is oftenoxidised dark-brown or black, but similar iron-stainedlayers can also occur below this surface. The basementclays have been penetrated to 3.2 m and are thought to beequivalent to the stiff clays that are found in the YarraDelta under the soft Holocene muds. In the Yarra Deltathese sediments are referred to as the Fishermans Bend Silt (Neilson 1967), and are dated as being slightly youngerthan the mainly underlying Burnley Basalt that has aradiometric age of 0.81 Ma (Page 1968). Constrained by theCoode Island Silt above, and the date on the basalt below,the Fishermans Bend Silt is considered to be MiddlePleistocene in age (Neilson 1967). Calcareous material,recovered in eight core samples in Port Phillip, was exam-ined for nannofossils, but no age-diagnostic taxa were found(R. Wilks pers. comm. 1981). Rare foraminifers recoveredfrom one of the Australian Dredging cores also containednon-diagnostic species (Holdgate 1977). On some seismicruns, faint discontinuous reflectors can be seen in base-ment clays suggestive of cut-and-fill structures. They indi-cate the basement clays to be over 25 ms (18.2 m) thick. Stiffbasement clays occur as far south as Mornington, andextend beneath the Geelong Outer Harbour. Their relation-

ship with the Tertiary formations outcropping around the shores of Port Phillip is poorly understood because theperipheral sands cover their boundary. It is presumed fromtheir geomorphic relationships that the stiff clays thin outagainst the older underlying Tertiary sediments. In the off-shore Williamstown area they abut against sea-floor out-crops of Quaternary basalts.

Infilled channels

A soft-sediment infilled channel system down the centre ofPort Phillip marks the ancestral course of the Yarra Riverand Werribee River during past glacial periods of low sea-level when the sea floor of Port Phillip was exposed(Figures 6, 7). These infilled channels contain up to threeseismically identifiable units that are not recognisedelsewhere. The channelling appears to have cut down into acoustically non-penetrable basement (clays?)probably similar to the stiff basement clays found along the channel margins. A number of vibrocores were sitedon and adjacent to the channel edges to confirm theseismic results. A previous paper (Holdgate et al. 1981) fully described the morphology, stratigraphy and age ofthese infilled channel systems and only a brief summaryis presented here.

CHANNEL MORPHOLOGY

An infilled channel system can be traced seismically inHobsons Bay beginning at Princes Pier where it is up to

Marine geology of Port Phillip, Vic. 449

Figure 7 North–south stratigraphic cross-section A–A’ between the Yarra Delta and Bass Strait, showing the interpreted Quaternarygeology (and sequence ages) from sub-bottom profiling in Port Phillip, correlated to the Yarra Delta (part of cross-section 3 of Neilson1988) and the Nepean Peninsula (cross-section figure 4 of Mallett & Holdgate 1985). Inset shows location of cross-section and the centralchannel system of Port Phillip.

Page 12: Marine geology of Port Phillip, Victoria

1.5 km wide and 8 ms (5.8 m) deep (Figures 6–8). A singlechannel (the Yarra Channel) swings south-southwest acrossHobsons Bay passing close to the Williamstown area. Itwidens and deepens to the south becoming 3 km wide and17 ms (12.4 m) deep at around the latitude of Mordialloc(Run 25: Figures 6, 9). Shortly south of here it splits into two branches of approximately equal width, or a secondchannel commences at this point. The eastern palaeo-Yarrachannel is approximately twice the thickness of the west-ern (Yarra) channel, with a soft sediment infill over 35 ms(25.2 m) deep (Run 24: Figures 6, 10). This dual channelsystem continues south to the latitude of Portarlington. AtRun 19 both Yarra and palaeo-Yarra channels bifurcate intoa number of deep, but narrow sub-branches with up to38 ms (27.7 m) of sediment infill. At this point the western

Yarra Channel is joined by a third channel emanating fromthe mouth of the Werribee River (Figures 6, 11). South ofRun 19 the channels are progressively lost on the seismicrecord beneath sands of the Nepean Bay Bar or appear tojoin up again in the Capel Sound area at the baysideentrance to the South Channel (Figures 6, 7). It is mostlikely that they would have run along the South Channelpassing out into Bass Strait via The Rip. Shortly before thechannels are lost in the Capel Sound area, their baseappears to have been cut to a maximum depth of 66 ms(48 m) below present sea-level.

Channel floor irregularities indicate the valley floorswere cut by a number of shifting and meandering rivers that left isolated ridges in the subchannel clays.Channel sides on the seismic profiles appear almost

450 G. R. Holdgate et al.

Figure 9 Sub-bottom profile part ofRun 25 west of the Esso PipelineBuoy 3, northern Port Phillip, show-ing the buried Yarra Channel fea-tures. Note the incomplete infill onthe western side. Vertical scale inmilliseconds (two-way-time) where10 ms � 7.2 m. For location of seis-mic section see Figure 6.

Figure 8 Sub-bottom profile part ofRun 27 between the Webb DockSwinging Basin and the PortMelbourne dredged channel inHobsons Bay, showing the buriedYarra Channel features. Verticalscale in milliseconds (two-way-time)where 10 ms � 7.2 m. For location ofseismic section see Figure 6.

Page 13: Marine geology of Port Phillip, Victoria

vertical in places, but calculations correcting for verticalexaggeration suggest the side angles were between 5 and10°.

CHANNEL STRATIGRAPHY AND AGES

The stratigraphy of the channel-filling sequences isdetermined from the seismic profiles in which four main sequences can be recognised by their seismic faciescharacteristics (Figures 7–11). These were fully detailed byHoldgate et al. (1981) and only a summary is presentedbelow. Most of the lower three sequences are too deep to bereached by conventional vibrocoring techniques. Theirages, as interpreted below, are shown on Figure 12 togetherwith the sea-level curve for the last 140 000 years (derivedfrom Aharon & Chappell 1986 and Labeyrie et al. 1987). Thefour sequences are referred to from top to bottom asSequences A–D. Their disposition is shown as a strati-graphic north–south cross-section drawn along the channel

systems between the Yarra Delta and the Nepean Peninsula(Figure 7).

Sequence A is the same sediments as the central mudsoutside the channels except it is usually more than doublethe thickness. Maximum cored interval is 6.05 m (Esso coreNo. 38), but on seismic it may exceed 10 ms (7.2 m). From14C dating it was deposited as marine infilling muds dur-ing the Holocene transgression.

Sequence B in the western Yarra Channel is denselylayered, strongly reflective and shows lateral accretion fromthe channel sides towards the channel centres. In places it shows drape over channel-floor irregularities. Goodexamples of this sequence are shown on Figures 8–11. Twocoreholes reached the upper levels of this sequence (EssoCore 3 and AD Core 1). In these two coreholes over 2 m offine to coarse clayey sands underlie Sequence A muds.Lithologies of the deeper layers of Sequence B areunknown, but may include coarse sands and gravels. In the Yarra Channel, Sequence B directly overlies acoustic

Marine geology of Port Phillip, Vic. 451

Figure 11 Sub-bottom profile partof Run 17, 8 km southeast of theWerribee River mouth, central PortPhillip, showing the buriedWerribee Channel features. Notethe incomplete infill. Vertical scalein milliseconds (two-way-time)where 10 ms � 7.2 m. For location ofseismic section see Figure 6.

Figure 10 Sub-bottom profile partof Run 24 east of the Spoil GroundBuoy, central Port Phillip, showingthe buried Yarra Channel (A, B)overlying the palaeo-Yarra Channelfeatures (C, D). Vertical scale inmilliseconds (two-way-time) where10 ms � 7.2 m. For location of seis-mic section see Figure 6.

Page 14: Marine geology of Port Phillip, Victoria

basement, and can be up to 8 ms (5.8 m) thick. The reflec-tivity of this unit appears to decrease to the south sug-gesting sediment fining, becoming more clayey in thisdirection. This sequence can be traced as a reflectivelyweaker middle infill unit in the eastern palaeo-YarraChannel (Figure 10). It is interpreted to represent fluvialderived aggradation sediments deposited as point bars dur-ing the last glacial sea-level fall and lowstand.

Sequence C first appears on Run 24 only in the easternpalaeo-Yarra channel (Figure 7). It is acoustically similarto the central muds of Sequence A, and can be up to 12 ms(8.7 m) thick. Sequence C appears to have been cut into bythe overlying Sequence B, so that the top of Sequence C isno shallower than 22 m below sea-level. It is presumed torepresent an earlier marine and estuarine mud sequence,deposited during relatively higher sea-level times whenmarine flooding of Port Phillip occurred (at least up to Run24). Using the sea-level curve (Figure 12) a date for SequenceC has been inferred at approximately 47–22 ka.

Sequence D is a basal, strongly layered reflective unit confined to the eastern palaeo-Yarra Channel andshows a similar lateral aggradational morphology toSequence B. On some runs it appears to form a veneer onthe western channel walls. By analogy, it represents afluvial aggradational sedimentary deposit preceding the47–22 ka flooding event, and could occur during the sea-levelfall at approximately 57–47 ka. As this sequence appears tooverlie acoustically similar clays to outside the channelareas, the base of the Yarra and palaeo-Yarra Channels areassumed to disconformably overlie the Fishermans BendSilt.

DISCUSSION

Sea-floor irregularities

In many instances where sub-bottom profiles cross theYarra Channel, irregularities in the form of mounds and

hollows were observed on the sea floor with up to 2 m ofrelief. They always overlie the deeper infilled channels. Inother areas coincident with the channels the sea floor maybe depressed to some extent below the surrounding seabed( as shown on Figures 8, 9). The channel from the WerribeeRiver consistently shows this feature along most of itslength (Figure 12). However, the palaeo-Yarra Channellacks any sea-floor bathymetric expression. Possibleexplanations for the Yarra Channel irregularities include:(i) they are a result of scouring bottom currents; (ii) theyresult from differential settling and compaction of theinfilled channel sequences; and (iii) the Yarra Channel isthe youngest channel feature and has not completelyinfilled to reach equilibrium with the rest of the PortPhillip bay floor. The first suggestion is the least likely asbottom currents are unlikely to be constrained to sub-surface features. Therefore some combination of (ii) and(iii) is the most likely explanation.

Holocene sedimentation rates

Evidence for sedimentation rates for Sequence A (centralmuds) in the Holocene transgression comes from the 14C dates at core site 18 (Bowler 1966) where the top 73 cm of sediments represent total accumulation within the last 5990 years, giving a compacted sedimentation rate of 12 cm/1000 y. There is a further 900 cm of SequenceA sediments below this stratigraphic level in the middle if the infilled channels (that may not be older than 10 000 years). Therefore a faster compacted sedimentationrate in the early part of the Holocene transgression ofup to 224 cm/1000 y is required. This accords to the findingsof Thom and Roy (1985) where rapid sedimentation rates and progradation immediately followed terminationof the post-glacial transgression (6500–6000 14C a BP),but slowed towards the present. A slower average sedi-mentation rate for the whole 25 m of channel infill of44 cm/1000 y is obtained if the channels are not older than 57 Ka.

452 G. R. Holdgate et al.

Figure 12 Interpreted stratigraphiccorrelation of Port PhillipSequences (A–D) and NepeanPeninsula aeolianites (dated by ther-moluminescence: Zhou et al. 1994)with the Late Pleistocene toHolocene sea-level record from HuonPeninsula (after Aharon & Chappell1986) and the pre-120 ka record fromBarbados (after Labeyrie et al. 1987).Temperature/humidity estimationsfrom Schleiger (1981).

Page 15: Marine geology of Port Phillip, Victoria

These rates appear to be slower than rates quoted forother estuarine, barrier and marine shelf areas in south-eastern Australia: e.g. 1–2 m/1000 y for barrier sands offCape Byron (Thom & Roy 1985); 0.5 m/1000 y for mid-shelfmuds (Thom & Roy 1985). However, translated into anannual rate of 0.44 mm/y, they are of a similar magnitudeto the 0.3 mm/y modern sedimentation rate estimated byHarris et al. (1996) for Port Phillip, based on current sedi-ment budgets and sediment inputs.

Gas phenomena

An interesting phenomenon, possibly related to sedimentcompaction, can be observed as the large number of‘rooftop-shaped’ smudges in the water column, particularlyabove the infilled channels or thicker areas of muds(Figures 5, 8–10). The water column above the more sandysea bed areas appears to lack this effect (or where sea-floorspoil ground dumps occur: Figure 10), suggesting the causeis not due to occurrence of fish in the water column. TheSIE sub-bottom profiler handbook (SIE 1975) refers to simi-lar phenomena in the Gulf Coast area of the USA and sug-gests that they result from subsea gas seepages. Similareffects were observed by Watkins and Worzel (1978) inclu-ding gas bubbles in the interstitial waters of the sedimentsproducing a ‘white-out’ effect or smearing of sub-bottomreflectors, and associated small mud volcanoes on the seafloor as a result of gas emanations. Although the latter twophenomena are not exactly correlatable to the Port Phillipseabed events, it is likely that methane production fromorganic breakdown of peaty deposits in the thicker areasof muddy sediments may cause some of the surface dis-ruptions, differential subsidence and seabed irregularities.This would explain the rooftop smudges as diffused cloudsof methane gas escaping into the water column.

Evolution of Port Phillip

Port Phillip can be compared to other large wave-dominatedestuaries and estuarine facies models in southeasternAustralia (Thom & Roy 1985; Dalrymple et al. 1992; Roy 1984,1998). Like other estuaries, Port Phillip possesses a bay-head delta (the Yarra Delta) composed of a coarsening-upward sedimentary sequence (Coode Island Silt overlainby Port Melbourne Sand). The muddy central basin (PortPhillip) also typifies wave-dominated estuaries, as do theflood-tidal delta (Nepean Bay Bar), and ocean-facing barriercomplex (the Nepean Peninsula) (Figure 7).

The principal difference compared to other estuaries insoutheastern Australia is the preservation of an early last-glacial record of channel cut-and-fill in Port Phillip that canbe correlated stratigraphically to the development of earlylast-glacial phases of quartz–carbonate aeolianite and bar-rier-complex development on the Nepean Peninsula. TheseNepean Peninsula outcrops and subsurface aeolianites arereferred to collectively as the Bridgewater Formation afterthe original definition at Portland by Boutakoff (1963);(Mallett & Holdgate 1985) (Figure 7). Thermoluminescencedating of palaeosols and cemented aeolianites on theNepean Peninsula reveals the succession (from a fewmetres above present sea-level) disconformably overliesolder aeolianites dated at 118 ka and older (Spiers 1992; Zhou

et al. 1994) (Figure 7). Above the disconformity a 1–2 m peatsoil is thought to represent a period of cool wet conditionswith two thermoluminescence dates of 55.2 and 47.1 ka(Zhou et al. 1994). Above the peat soil 35 m of cementedaeolianites occur containing several interbedded dark-brown palaeosols. The aeolianite and palaeosol successionis thought to represent cool dry periods when predominantstrong southwesterly winds prevailed, when carbonate wasactively eroding from Miocene–Pliocene shelfal areas inBass Strait due to a sea-level fall. The central Australiandune systems were also forming during this time (Bowler1976; Spiers 1992). A calcrete layer can be used to divide thissuccession into a lower aeolianite unit (25 m thick) and an upper aeolianite unit (15 m thick). Five thermolumin-escence dates ranging between 57.4 and 47.8 ka wererecorded from the lower aeolianites, whereas the upperaeolianites and terra rossa soils were dated between 22.7and 18.8 ka (Zhou et al. 1994). The Holocene unconsolidateddune sands above the cemented aeolianites yielded thermo-luminescence and 14C dates of 8400 a BP and 5350 a BP,respectively (Bird 1982; Zhou et al. 1994).

Deposition of the three aeolianite units dating from57–47, 22–18, and 8–5 ka would have affected depositionalprocesses in Port Phillip. They were associated with periodsof barrier buildup across the Port Phillip bay mouth andtimes of estuarine deposition in the bay. Therefore, we sug-gest that the seismically defined Sequence A (as confirmedby 14C dates) is contemporaneous with the unconsolidatedNepean dunes. Seismic sequence B infills channels cut dur-ing the last glacial lowstand, and overlaps with the 22–17 kaupper aeolianites. Seismic sequence C also representsestuarine deposition created by ponding in Port Phillip bythe 57–47 ka aeolianites, and seismic sequence D representsdowncut and infill of fluvial channels into the older 118 kasurface. During the more extreme lowstand periods, fluvialdowncutting would have separated the Nepean Bay Barfrom the Nepean Peninsula and allowed fluvial outlets toBass Strait via The Rip to develop. Later filling (both byfluvial and tidal deposition) was incomplete in the CapelSound and South Channel and today this area provides themain deep-water shipping passage across the Nepean BayBar into Port Phillip. Borings on the Nepean Bay Bar con-firm the occurrence of shallow aeolianites beneath a sandveneer (Keble 1950; this study).

Groundwater bores across the Nepean Peninsula indi-cate the Bridgewater Formation aeolianite succession(including palaeosols, shelly sands and peat deposits)extends to 60 m below sea-level (not –130 m as often quotedfrom Keble 1950). This succession is 118 ka or older, with amaximum age based on the underlying shelly sands of theWannaeue Formation, which have been dated byforaminifers as Late Pliocene to possibly Early Pleistocene(Mallett & Holdgate 1985). The 118 ka disconformity correl-ates to the last interglacial sea-level high (Figure 12). Thisimplies that across the Nepean Peninsula and Nepean BayBar the disconformity may have originated as a wave-cutsurface that later became vegetated due to sea-level fall (theoverlying peat soil) dated at 55.2–47.1 ka (Zhou et al. 1994).The older subsurface pre-118 ka aeolianite successionswould also have provided a ponding mechanism fordeposits of earlier estuarine systems such as the Fisher-mans Bend Silt, that in the Yarra Delta has a maximum

Marine geology of Port Phillip, Vic. 453

Page 16: Marine geology of Port Phillip, Victoria

radiometric K/Ar age of 810 ka (Page 1968). Therefore weconclude that the Fishermans Bend Silt beneath PortPhillip is no younger than 118 ka, and that the weatheredsurface on the Fishermans Bend Silt (as recorded innumerous vibrocores) represents a non-depositional recordbetween 118 and 57 ka. The missing depositional recordcould be attributed to lowstand stripping and erosion atapproximately 70 ka based on the significant sea-level fallto –80 m at this time (Figure 12).

The lower and upper ages for the Nepean aeolianites andthe contemporaneous Fishermans Bend Silt suggest depo-sition occurred between 810 and 118 ka, although this mayhave been intermittent, considering the total sedimentthickness. It is also likely that older Tertiary units pro-gressively subcrop below the Fishermans Bend Silt north-ward across Port Phillip, and that the total Quaternarythickness in the bay probably does not exceed 40 m.

Two subsurface peat deposits in the aeolianites wereintersected in bores near Rosebud, and probably representswamp deposition analogous to the overlying present-dayTootgarook Swamp (Figure 7). They also provide evidenceof earlier Pleistocene cool wet conditions. The upper peatoccurs at 10 to 15 m below sea-level, and approximates to aplatform level (benching) on the Bass Strait face of theNepean Peninsula (Keble 1950). This could represent asecond disconformity (wave-cut highstand) followed by peatdeposition as sea-levels fell, analogous to the 118 ka ero-sional event. A lower peat horizon occurs at ~40 m belowsea-level, but appears unrelated to a second (Keble 1950)platform at –30 m. However, any tectonic subsidence of theNepean Peninsula would add to uncertainties concerningthe significance of eustatically derived benching levels.

CONCLUSIONS

Sub-bottom seismic profiling and vibrocoring identify twomajor textural classes in the modern sediment distributionin Port Phillip. They are the peripheral sands and the cen-tral muds. These two classes are controlled by wave-basesorting, are lateral facies equivalents, and were depositedmainly around the post-last glacial marine transgression.Seismic profiling has allowed estimates to be made onregional thicknesses of the central muds, and some esti-mates as to the peripheral sand thicknesses can be derivedfrom the coring results.

Seismic profiling has also demonstrated the existenceof a buried channel complex down the centre of Port Phillipthat delineates the course of the Yarra and Werribee Riversduring periods of sea-level lows in the Late Pleistocene.Seismic stratigraphy suggests that two main channellingepisodes occurred. Radiocarbon dating and correlatedthermoluminescence-dated aeolianites on the NepeanPeninsula suggest the earliest channelling occurred atca 57–47 ka, and was subsequently modified at the peak of the last glaciation during ca 22–8 ka. Infilling of theseLate Pleistocene channels first commenced as fluvial sanddeposits and was subsequently completed during inter-stadial high periods by estuarine and marine mud depo-sition. The deposits are closely related to the developmentof the Nepean barrier and aeolianite formations across themouth of Port Phillip, and are a result of the ponding

effects of this barrier formation. Fluvial downcuttingduring lowstands has probably cut channels between theNepean Peninsula and the Nepean Bay Bar that remainincompletely filled. The infilling sequences exhibited inthese channels provides a modern analogue to sequenceanalysis in a known period of eustatically changing sea-levels.

The phenomena of sea-floor irregularities and inferredgas seepages over the underlying channel features suggestthat the Port Phillip seabed has not as yet reached equil-ibrium. Through differential sediment compaction andincomplete infilling, the bay floor is still in an evolvingstate.

ACKNOWLEDGEMENTS

We wish to acknowledge the former Mines Department ofVictoria (Geological Survey Branch), Ports and HarboursDivision (Public Works Department), and the MarineLaboratory (Port of Melbourne Authority). The bay floorsediment-distribution pattern map was kindly provided byR. W. Buckley, and the expertise of Eddie Frankel is grate-fully acknowledged in operating and obtaining the bestresults from the sub-bottom seismic profiler system. JohnWebb is also thanked for providing references to thermo-luminescence dates on the Nepean Peninsula. The com-ments from journal reviewers J. D. A. Clarke and P. Harrisare acknowledged, for supplying additional references andfor useful improvements to the text.

REFERENCES

ABELE C. 1988. Central coastal basins. In: Douglas J. G. & Ferguson J. A. eds. Geology of Victoria, pp. 303–322. Geological Society ofAustralia, Victorian Division, Melbourne.

AHARON P. & CHAPPELL J. 1986. Oxygen isotope, sea level changes andtemperature history of a coral reef environment in New Guineaover the last 105 years. Palaeogeography, Palaeoclimatology,Palaeoecology 56, 337–379.

BEASLEY A. W. 1966. Port Phillip Survey 1957–1963. The bottom sediments. Memoirs of the National Museum of Victoria 27,69–105.

BEASLEY A. W. 1969. Beach sands of the southern shore of Port PhillipBay, Victoria, Australia. Memoirs of the National Museum ofVictoria 29, 1–21.

BEASLEY A. W. 1971. Safety Beach, Dromana Bay, Port Phillip, Victoria,Victorian Naturalist 88, 291–295.

BIRD E. C. F. 1982. Foundations. In: Hollinshed C. N., Bird E. C. F. &Goss N. eds. Lime, Land, Leisure—Peninsula History in the Shireof Flinders. Melbourne.

BIRD E. C. F. 1993. The Coast of Victoria. Melbourne University Press,Melbourne.

BOUTAKOFF N. 1963. The geology and geomorphology of the Portlandarea. Geological Survey of Victoria Memoir 22.

BOWLER J. M. 1966. Port Phillip Survey 1957–1963. The geology andgeomorphology. Memoirs of the National Museum of Victoria 27,19–59.

BOWLER J. M. 1976. Glacial age aridity in southern Australia.Quaternary Geology 2, 16–25.

BUCKLEY R. W. & CLARK R. G. 1987. Sediments of Port Phillip. MarineLaboratory, Port of Melbourne Authority, Report 87-8-15(unpubl.).

DALRYMPLE R. W., ZAITLIN B. A. & BOYD R. 1992. Estuarine facies models:conceptual basis and stratigraphic implications. Journal ofSedimentary Petrology 62, 1130–1146.

454 G. R. Holdgate et al.

Page 17: Marine geology of Port Phillip, Victoria

DOLAN B. H. 1973. High-resolution seismic profiling survey of PortPhillip Bay near the city of Mordialloc, Victoria. Bureau ofMineral Resources Record 1973/65.

ESSO 1971. Esso Standard Oil (Aust) Limited. Proposed 10 inch EthanePipeline—Port Phillip Bay, Victoria. Report on site investigation.Laboratory Reference Number LA920 (unpubl.).

GEURIN B. 1974a. Present day marine sediments at St Kilda. GeologicalSurvey of Victoria Report 1974/5.

GEURIN B. 1974b. Littoral zone investigations, Rye to Safety Beach. Portsand Harbours Division Report (unpubl.).

GILL E. D. 1968. Radiocarbon dating. Victorian Naturalist 86,161–164.

GILL E. D. 1972. Anadara (Bivalvia) in the Indian and Pacific Watersof Australia. Journal of the Marine Biology Association of India14, 726–731.

GREILACH P., YOU Z., BLACK K. & GORMAN R. 1996. Sediment character-istics in Port Phillip Bay: Port Phillip Environmental Study.CSIRO Marine Research Technical Report 23.

HARRIS G., BATLEY G., FOX D., HALL D., JERNAKOFF R., MURRAY A., NEWELL

B., PARSLOW G., SKYRING G. & WALKER S. 1996. Port PhillipEnvironmental Study Final Report. CSIRO, Dickson.

HOLDGATE G. R. 1977. Notes on the significance of core samples fromthe Port Melbourne (dredged) channel, Port Phillip Bay. VictorianMines Department Report 1977/97 (unpubl.).

HOLDGATE G. R., THOMPSON B. R. & GEURIN B. 1981. Late Pleistocenechannels in Port Phillip, Proceedings of the Royal Society ofVictoria 92, 119–130.

JACKMAN C. H. 1971. Report on additional sub-bottom investigation—Ethane Pipeline Route, Port Phillip Bay, Victoria. ConfidentialReport to Esso (Aust) Limited.

JONES I. S. F. 1980. Tidal and wind-driven currents in Bass Strait.Australian Journal of Marine and Freshwater Research 31,109–117.

KEBLE R. A. 1946. The Sunklands of Port Phillip Bay and Bass Strait.Memoirs of the National Museum of Victoria 14, 69–122.

KEBLE R. A. 1950. The Mornington Peninsula. Memoirs of the GeologicalSurvey of Victoria. Department of Mines, Victoria.

LABEYRIE L. D., DUPLESSY J. C. & BLANC P. L. 1987. Variations in modeof formation and temperature of oceanic deep water masses overthe past 125,000 years. Nature 327, 477–482.

LINK A. B. 1967. Delineating the major depositional environments innorthern Port Phillip Bay, Victoria. Journal of SedimentaryPetrology 37, 931–951.

MCCUTCHEON R. 1979. Mentone to Frankston Beach investigation.Geological Survey of Victoria Report 1979/108 (unpubl.).

MALLETT C. W. & HOLDGATE G. R. 1985. Subsurface Neogene stratigraphyof Nepean Peninsula, Victoria. Department of Mines and EnergySouth Australia, Special Publication 5, 233–245.

NEILSON J. L. 1967. The Physiography of the Melbourne Area.Geological Survey of Victoria Bulletin 59, 47–52.

NEILSON J. L. 1988. Yarra Delta. In: Douglas J. G. & Ferguson J. A. eds.Geology of Victoria, pp. 387–392. Geological Society of Australia,Victorian Division, Melbourne.

PAGE R. W. 1968. Catalogue of radiometric age determinations carriedout on Australian rocks in 1966. Bureau of Mineral ResourcesRecord 1968/30.

ROY R. W. 1984. Holocene sedimentation histories of estuaries in south-eastern Australia. In: Hodgkin E. P. ed. Estuarine Environmentsof the Southern Hemisphere, pp. 23–60. Department of Conservationand Environment (WA) Bulletin 161.

ROY P. S. 1998. Cainozoic geology of the coast and shelf. In: Basden H.ed. Geology of New South Wales—Synthesis Sydney, pp. 361–385.Geological Survey of New South Wales Memoir 13.

SCHLEIGER N. W. 1981. Palaeowind analysis of high and low calcarenitefossil dunes along Southern Australian Coasts. In: Proceedings ofthe first CLIMANZ Conference, Howmans Gap, Victoria, pp. 21–22.Department of Biogeography and Geomorphology, ResearchSchool of Pacific Studies, Australian National University,Canberra.

SECV 1970. Site investigations for the Newport Power Station. StateElectricity Commission of Victoria Report (unpubl.).

SEEDSMAN K. R. & MARSDEN M. A. H. 1980. Sediment distribution andmovement within Corio Bay. Ministry of Conservation, Victoria.Environment Studies Series Publication 312.

SIE 1975. Handbook and Operating Manual for the SIE SubbottomProfiler. Seismic Supplies Australia Ltd, Brisbane (unpubl.).

SMITH G. J. 1981. Nourishment sampling, St Kilda report. Ports andHarbours Report Number 81-3-10 (unpubl.).

SPIERS R. H. 1992. The geological evolution of Quaternary coastal dunesequences in the Sorrento–Portsea area of the Nepean Peninsula,Victoria, Australia. BSc (Hons) thesis, La Trobe University,Melbourne (unpubl.).

STUIVER M., REIMER P. J., BARD E., BECK J. W., BURR G. S., HUGHEN K. A.,KROMER B., MCCORMAC G., VAN DER PLICHT J. & SPURK M. 1998.INTCAL98 radiocarbon age calibration, 24,000–0 cal BP.Radiocarbon 40, 1041–1083.

THOM B. G. & ROY P. S. 1985. Relative sea levels and coastal sedimen-tation in southeast Australia in the Holocene. Journal ofSedimentary Petrology 55, 257–264.

WALKER S. J. 1997. Hydrodynamic models of Port Phillip Bay: PortPhillip Environmental Study. CSIRO Marine Research TechnicalReport 38.

WALKER S. J. & SHERWOOD C. R. 1997. A transport model of Port PhillipBay: Port Phillip Environmental Study. CSIRO Marine ResearchTechnical Report 39.

WALSH R. D. 1978. A survey of continuous seismic reflection profiling.BSc (Hons) thesis, University of Melbourne, Melbourne (unpubl.).

WATKINS J. & WORZEL J. L. 1978. The Serendipity gas seep area, southTexas offshore. American Association Petroleum GeologistsBulletin 62, 1067–1074.

ZHOU L., WILLIAMS A. J. & PETERSON J. A. 1994. Late Quaternaryaeolianites, palaeosols and depositional environments on theNepean Peninsula, Victoria, Australia. Quaternary ScienceReviews 13, 225–239.

Received 6 June 2000; accepted 27 February 2001

Marine geology of Port Phillip, Vic. 455