Present-day sedimentation rates on the southern and southeastern Australian continental margins

The Australian continental margin presents highly contrasted settings depending on (1) the presence or absence of a fluvial sediment supply, (2) the distance from the Australian mainland, and (3) the local hydrological setting. Despite the importance in surface area of the continental margin around the Australian mainland, so far only a few studies have dealt with sedimentation, most of them focusing on the northeastern Australian Shelf. This work presents the first large-scale investigation of modern sedimentation along the southern margin of Australia in the SE Indian Ocean, and the western margin of the Tasman Sea. Sedimentation intensity was assessed on a century timescale using a multi-tracer approach (234Th, 210Pb, 232Th) on interface cores around the 1000 m water depth contour. 234Th (half-life: 24.1 days) in excess was detected in all surface samples, testifying to the occurrence of freshly deposited particles. Sedimentation and mass accumulation rates based on sedimentary 210Pb excess profiles (CF:CS model) range between 0.027 and 0.280 cm y–1 and between 14 and 222 mg cm–2 y–1, respectively. Whereas sedimentation rates are low and associated with carbonates on the western margin, sediments are more influenced by the detrital fraction and organic carbon on the eastern side of the continent. In comparison with northern continental margins (e.g. Timor Sea, Gulf of Papua), the southern Australian margin receives little sediment today, as it is rarely linked to a river system that would otherwise deliver large amounts of sediment, and also because of the presence of the extended shelf south of Australia.


INTRODUCTION
Continental margins, particularly those influenced by upwelling, are among the most productive biological systems in the oceans and are, therefore, of particular interest for both biological and geological processes (Walsh 1991). These margins are characterised by high fluxes of organic carbon, nutrients and other trace elements, which may either be exported to the open ocean or are rapidly deposited and buried on geological scales. During the last decades, numerous studies have documented a large diversity in the intensity of mass fluxes and sedimentation on continental margins, mainly European and North American, depending on their morphologies, the occurrence of fluvial input and hydrological conditions (Nittrouer et al. 1984;Walsh et al. 1988;Biscaye & Anderson 1994;van Weering et al. 2002;Schmidt et al. 2009;Wheatcroft et al. 2013). In addition, the high sediment accumulation rates that characterise many continental margins make them ideal depositional settings for preserving high-resolution records of past climate change.
The Australian continent heralds a fairly large continental shield, such that during glacial maxima in the Quaternary (Yokoyama et al. 2000;Lambeck & Chappell 2001;De Deckker & Yokoyama 2009) when the sea level had dropped by the order of 120 m, the Australian landmass had almost doubled in size. The most extensive shelves are to be found offshore northern Australia, commonly referred to as the Sahul Shelf, which is bordered to the north by the Timor Trough, and in the south the Great Australian Bight is well over 200 km wide and is bordered by an arid region to the north (The Nullarbor Plain, Figure 1) that today lacks active rivers. This is in contrast to the North West Shelf that is bordered by large, monsoonal (and therefore seasonal) rivers systems (Harris et al. 1990;Dunbar & Dickens 2003;Wasson et al. 2010). It is also reasonable to expect an input of aeolian sediment to the shelves, as already documented by Gingele et al. (2001) and De Deckker et al. (2014 for the North West Shelf, and Gingele et al. (2007) based on a core taken in the Murray Canyons Group area offshore the mouth of the Murray River, for the latter part of the Holocene. Nevertheless, this aeolian input is considered to be minimal compared with the hemipelagic clay deposition on the shelves that also contains much biogenic carbonate material. Despite the importance in surface area of the continental margin around the Australian *Corresponding author: patrick.dedeckker@anu.edu.au mainland, there are so far only few studies dealing with present-day sedimentation, with most focused on the northern Australia shelf (Brunskill et al. 2002;Pfitzner et al. 2004). A 3D representation of the southeast Australian margin described it as being narrow, steep and sediment deficient as a result of its tectonic history of asymmetric passive margin rifting (Boyd et al. 2004).

THIS STUDY
The purpose of this investigation is to quantify modern sedimentation (bioturbation, sediment and mass accumulation rates) on the Australian margin on century timescale by using a multi-tracer approach. Here, we report detailed depth profiles of the particle-reactive radionuclides 234 Th (T1/2 D 24.1 days), 210 Pb (T1/2 D 22.3 years) and 232 Th (T1/2 D 1.40.10 10 years) in interface sediments collected around the 1000 m depth contour along the southern and southeastern Australian margins during several cruises ( This study was conducted in parallel with the examination of organic compounds (U k 37 and TEX H 86 ; see Smith et al. 2013) in the same cores discussed here so as to reconstruct sea-surface temperatures for the last few centuries. For this purpose, we chose to collect sediment cores around the 1000 m contour so as to minimise terrigenous input and also for comparison along the Australian continental margin (see below). In another paper, we will use the data acquired here to provide a chronology for each level within samples in all the multicores analysed for organic compounds.

Sampling
The core sampling targeted the 1000 m contour first to avoid reworking along the margin in shallower waters. In addition, sedimentation rates usually decrease with depth; for example, Soetaert et al. (1996) proposed a relationship between water depth (D) and sedimentation rate (S; S D 982 D À1.548 , n D 110, r 2 D 0.66). Despite variability in sedimentation intensity, the 1000 m contour appears the optimal depth to recover interface sediments with little reworking and still measurable sedimentation rates. Each coring site was carefully selected based on previously acquired swath map data and seismic profiles so as to ensure suitable sediments for coring.
Sediments were collected using a multicorer [consisting of a tripod that holds up to eight short (<60 cm long) tubes that permit sampling of undisturbed sediments] in order to recover a well-preserved waterÀsediment interface. Visual inspection of each set of cores confirms the absence of disturbance related to coring or to bioturbation, except in core S5 where worm burrows were observed down to 6À8 cm. Immediately after core retrieval, tubes were carefully extruded with sediment subsampled at 0.5À1 cm intervals and stored on board in a cold room at 4 C.

Analysis of radionuclides
In the laboratory, dry bulk density (DBD) was measured by determining the weight after drying (60 C) of a known volume of wet sediment. Following this procedure, 234 Th, 232 Th, 210 Pb and 226 Ra activities were measured using a very low background, high-efficiency, well-shaped g-detector equipped with a Cryo-Cycle (CANBERRA) . The standards used for the efficiency calibration of the detector are RGU and RGTh gamma standards from IAEA. 210 Pb is measured directly by its peak at 46.5 keV; 226 Ra is measured by selected gamma rays of its progenies 214 Pb (295.2 and 351.9 keV) and 214 Bi (609.3 keV) and 232 Th using the gamma rays emitted by its short-lived decay product 232 Pb (238.6 keV) (Schmidt & Cochran 2010;Reyss et al. 1995). Errors on radionuclide activities are based on 1 standard deviation counting statistics. Excess 234 Th and 210 Pb data were calculated by subtracting the activity supported by their respective parent isotope from the total activity in the sediment, and then by correcting 234 Th values for radioactive decay that occurred between sample collection and counting (this correction is not necessary for 210 Pb owing to its longer half-life). The long-lived 232 Th is usually associated with the detrital fraction, and so activity levels can be an indication of lithological proportions (Stupar et al. 2014).

Complementary analyses
Grain size analyses on wet sediment aliquots were performed with a Malvern Ò Laser diffraction particle sizer. The organic carbon (C org ) and carbonate (CaCO 3 ) contents were determined on bulk and decarbonated sediments by direct combustion in an LECO CS 125 analyser (Schmidt et al. 2010).

Determination of bioturbation, sediment and mass accumulation rates
Sedimentation and mixing rates are generally determined by means of tracers that reach the seafloor in association with particles settling through the water column (Nittrouer et al. 1984;Krishnaswami & Cochran 2008). Taking into account its very short half-life (T1/2 D 24.1 days) and sedimentation rates usually encountered Table 1 Location and water-column depth of the studied cores. Sedimentation and mass accumulation rates (SAR; MAR) calculated from 210 Pb xs profiles. Bioturbation coefficients (D b ) calculated from 234 Th xs profiles. When 234 Th xs profile presents only a single point, D b is assumed to be <0.1 cm 2 y À1 . n.d.: there is some evidence of mixing of the first centimetre owing to sampling and D b was not determined. on continental slopes and shelves (far less than 1 cm y À1 ), 234 Th xs penetration to variable depths indicates efficient mixing of the upper sediments, usually by bioturbation. The simplest way to derive bioturbation rates (D b ) from 234 Th xs profiles is to assume bioturbation as a diffusive process occurring at a constant rate within a surface mixed layer under steady state, according to: where [ 234 Th xs ] 0,z denotes the activities of 234 Th xs , respectively at the waterÀsediment interface and at depth z, and λ is the radioactive decay constant of 234 Th (Aller & DeMaster 1984). We present 234 Th profiles and derived bioturbation rates as an indication of particle input over the last few months of sedimentation. The 210 Pb measurements have been widely used to calculate short-term (years to decades) sediment accumulation rates in continental and oceanic environment over the last 40 years (Kirchner 2011; and references herein). Dating is calculated using excess activity of 210 Pb ( 210 Pb xs ), which decays with time in the sediment column according to its half-life. It is assumed that the specific activity of newly deposited particles at a given site is nearly constant with time. Consequently, sediment accumulation rate can be derived from 210 Pb, based on two assumptions: (1) constant flux, and (2) constant sediment accumulation rate (referred to as the CF:CS method). As a result, the decrease in 210 Pb xs activities with depth is described by the following relation: where [ 210 Pb xs ] 0,z denotes the activities of excess 210 Pb at surface, or the base of the mixed layer and depth z, λ is the decay constant of the nuclide, and S is the sediment accumulation rate. Sediment accumulation rates were calculated from the slope of the 210 Pb profile below the surface mixed layer (SML). The bioturbation effect is not considered, and S corresponds to maximum values. An alternative method is to plot the regression of 210 Pb xs against cumulative mass in order to calculate the mass accumulation rate (MAR), which integrates the compaction effect. Both estimates are given, but the first one is the one used most commonly.

Margin of the western Tasman Sea
For the margin of the western Tasman Sea, surface 234 Th xs activities ranged between 19 and 197 mBq.g À1 (Figure 2) and were generally confined in the first 0.5 cm; only the core S3 showed a penetration down to nearly 2 cm, consistent with a mixed layer in the upper 210 Pb xs profile. Bioturbation rates range between negligible values up to 2.1 cm 2 y À1 . 210 Pb excesses are detected at depth ranging between 4 and 17 cm, depending on the sites (Figure 2), testifying to large differences in sedimentation rates. The highest MARs (125À222 mg cm À2 y À1 ) for the margin of the western Tasman Sea are observed in cores S4 and S5 ( Figure 3) and are associated with the highest content in organic carbon. Off East Tasmania, core S2 records among the lowest sedimentation intensity of this transect associated with coarse surface sediments (121À153 mm) that present also a low C org content (<0.3 % (dry weight)). Core S1 did not penetrate soft sediment and returned only a few gravels. This indicates a location not favourable to sediment deposition owing to a winnowing effect related to local hydrological conditions. Bottom Figure 2 Depth profiles of 210 Pb xs (dark circles) and 234 Th xs (grey triangles) activities for all the multicores studied here. Next to the core label numbers are the water depth at which the cores were collected. Error bars on radionuclides profiles correspond to 1 SD.

Southern Australian margin
Along the 1000 m depth contour of the southern Australian margin, 234 Th xs is mainly detected in the first 0.5 cm, with surface activities ranging between 14 and 59 mBq.g À1 . These are among the lowest values observed in this work (Figure 2). Surface 210 Pb xs activities are also low, comprising between 32 and 279 mBq.g À1 . The lowest 234 Th and 210 Pb excesses are mainly observed on the west side of the southern Australian margin. Cores S11 to S18 are also characterised by the highest CaCO 3 content reported in this work, with coarser surface sediment (mean grain size up to 67 mm) and low detrital content, as revealed by the low 232 Th activities (below 4 mBq g À1 ) (Figure 3 and supplementary data). Bioturbation rates are comprised between negligible values to 0.37 cm 2 y À1 , indicating negligible to low mixing of interface sediments. We failed to recover short cores between S11 and S15, along the southwestern corner of Western Australia ( Figure 1) and assume that this indicates unfavourable area for sediment deposition and very slow sedimentation rates. On the southwestern border of the Australian margin, cores S11 to S18 present low sediment and MARs, varying between 0.027 and 0.046 cm y À1 and between 28 and 52 mg cm À2 y À1 , with low-carbon sediment (0.2À0.5 wt% C org ).
On the eastern side of the southern Australian margin, apart from the Murray Canyon Group area (discussed below), cores S19 and S20 present enhanced Pb xs activities in surface sediments (160À279 mBq g À1 ) and a change in the composition, with enhanced C org and 232 Th contents along with fine sediment (mean grain size around 10À14 mm). However, SAR and MAR do not notably differ in intensity compared with the western side ( Figure 3). Core S21 on the western side of Tasmania is more comparable with core S2 on the eastern side of this island, with high carbonate (> 70 % (dry weight)) and low C org (about 0.3 % (dry weight)) contents.

Submarine canyons area off South of Australia
Core MUC03, recovered at a water depth of 949 m on a platform in the Sprigg Canyon, shows the highest sedimentation rate (0.12 cm y À1 ) recorded along the southern Australian margin (Figure 3; Table 1). The presence of freshly deposited particles, tagged by 234 Th in excess,  Figure 1). The MCG acts as an 'amplifier' of sediment supply through the conduits down to the deep ocean despite the fact that the Murray River today sheds little sediment to the ocean. Our core site MUC03 lies on a promontory within the MCG area and was already the subject of a study by .

Synthesis of the present sedimentation on the Australian continental margin
For the southern margin of Australia (cores S11 to S21, but ignoring MUC3), we note that sediment rates range between 0.027 and 0.048 cm y À1 , with a minor increase in the sedimentation intensity, but mainly a change in sediment nature, eastward. Whereas sedimentation rates are low and associated with carbonates on the western side, sediments are more influenced by the detrital fraction and organic carbon on the eastern side. This change could be related to the continental sediment flux to the coastal zone. Wasson et al. (1996) have estimated the mean annual sediment yield for a 1000 km 2 rural basin to be 1 kt y À1 in the southwest of the continent and between 8 and 22 kt y À1 on the southeastern side of Australia, including the Murray-Darling Basin. The estimates correspond to sediment yield on the mainland with much of the material being transported only over short distances. Nevertheless, this gives an indication of the potential flux of sediment to the coastal zone. The difference between sediment deliveries could account for the changes in sedimentation between the western and eastern margins of southern Australia.
The western margin of the Tasman Sea (cores S2ÀS10) is characterised by a narrow continental shelf, and rivers there have moderate flows compared with the North West Shelf of Australia (Alongi et al. 2013). Consequently, sediment accumulation rates are moderate to high (compared with the other sites investigated in this work): between 0.052 and 0.280 cm y À1 . Unfortunately, we could not retrieve cores offshore Sydney owing to the presence of telephone cables in the area. However, Matthai et al. (2001) previously reported higher sedimentation accumulation rates, ranging between 0.2 and 0.4 cm y À1 , on the middle shelf adjacent to Sydney. It is possible therefore that this area may also register higher sedimentation rates on the upper slope, owing to the presence of abundant sediment that accumulated in the bays around Sydney and that could eventually be delivered to the shelf during periods of heavy river discharges. It is noticeable that cores S4 and S5 off southern New South Wales present the highest sedimentation rates.
The MCG multicore samples, on the other hand, with their extremely high carbonate contents, indicate that, over the last three centuries, there was little supply of River Murray sediments. In fact, over the last six decades, significant impoundment on this fluvial system seriously constrained river flows (Wasson et al. 1996, table 1). This is consistent with the findings of Gingele et al.

Comparison with nearby margins
Sediment and MARs along the 1000 m depth contour along the southern and southeastern Australian margin are in the lower range of values, based on 210 Pb, reported for this oceanic region. For example, Muhammad et al. (2008) reported SAR ranging between 0.05 and 0.23 cm y À1 at equivalent depths in the Gulf of Papua. Locally, high accumulations of modern sediments have also been reported as for the outer Poverty continental margin of New Zealand (Alexander et al. 2010). Further north, the continental margin off southern Taiwan registers high sediment MARs (around 100 mg cm À2 y À1 at 1000 m depth), sustained by high fluvial sediment inputs that lead to efficient trapping of organic carbon (Kao et al. 2006). The bottom part of the Kuroshio Current also transports particles as part of a bottom nepheloid layer that promotes silt to clay deposition (140 mg cm À2 y À1 ) on the upper slope of the East China Sea margin (Oguri et al. 2003;Yamada et al. 2006).
In contrast, the Australian MARs for around the 1000 m depth contour obtained from the multicores at opposite locations (north and southeast) regions reveal that this continent is a poor supplier of terrigenous material to the deep ocean in contrast to the mountainous regions to the north located in the tropics where erosion is also of several orders of magnitude higher. In addition, Australia's sedimentation rates are known to have remained low, even during the Quaternary glacial periods (Helmsath et al. 2000). Australia, with 47% of its surface area benefiting from internal drainage, owing to low rainfall and elevation, contributes little sediment to the coastal ocean (Wasson et al. 1996;Woolfe et al. 1998). Most of the sediment is derived from the tropical drainage along its northeastern coast and Timor (Wasson et al. 2010).
Observed differences in sediment characteristics and in particular sediment accumulation rates are generally interpreted with reference to regional differences in climate and river runoff. For the southern portion of Western Australia, as for the vast Great Australian Bight (over 1500 km wide), with no river reaching the ocean, sedimentation rates are explicably very low. The other rivers for the regions discussed here (Figure 1) are again small compared with European, Asian or American rivers. Even the large Murray-Darling system, which drains over 1 million km 2 , frequently does not discharge to the sea today.

CONCLUSION
The aim of this work was to characterise modern sediment deposition on the southern and southeastern Australian continental margin using a multi-tracer approach based on 234 Th, 210 Pb and 232 Th. These data fill an information gap on a region of the world for which no sedimentation rates have been accurately assessed and would serve as a reference for further studies. Indeed, regarding the importance of the continental margin in general and the number of published studies on European margins by example, the lack of data on sedimentation rates for the Australian margin is wanting. Thus far, the profiles reported here are the first investigations to quantify the present-day sedimentation intensity for deep-water sediments in the Australian continental margin, apart from the work in the MCG area by . MARs overall remain low to moderate, by comparison with the expected values on the upper slope of continental margins (Soetaert et al. 1996), and this is no surprise knowing the overall low relief of the Australian continent and the lack of tectonic activity in the region.