Holocene shoreline progradation and coastal evolution at Guichen and Rivoli Bays,southern Australia

Abstract

Prograded barrier systems record shoreline behaviour and palaeoenvironmental information. The Guichen Bay Holocene embayment fill succession in South Australia has been subject to several prominent studies; however, several important unanswered questions remained regarding the timing of the older ridge sets at this site. Additional Optically Stimulated Luminescence (OSL) dating indicates that progradation commenced in the southeastern corner of the plain ~7300 years ago and was rapid between ~5800 and ~5000 years ago. To augment this record, three OSL dating transects were constructed at nearby Rivoli Bay in the north, central and south. Rapid progradation occurred in the south and then north of the Rivoli plain until ~5000 years ago. Steady progradation occurred in the centre of the plain between ~5000 years ago and present. Rapid shoreline progradation at Guichen and Rivoli Bays before ~5000 years ago was due to the input of sediment from the erosion of Robe and Woakwine Ranges and the inner continental shelf as sea levels rose to present. Raised beach strata imaged with Ground Penetrating Radar (GPR) at Rivoli Bay suggest a sea-level highstand of +2 m above present ~3500 years ago, steadily falling and reaching the present ~1000 years ago. This concurs with evidence from Guichen Bay and may have promoted shoreline progradation. Sediment infilling of Guichen and Rivoli Bays and the fall in sea level restricted the marine corridor between the Woakwine and Robe Ranges to a narrow channel by ~4000 and ~2000 years in the north and south, respectively. Holocene shoreline behaviour was influenced by changing sediment supply and shoreline reorientation with changing wave refraction patterns.

Keywords

coastal dunes coastal evolution foredune ridge plain Ground Penetrating Radar OSL dating sea-level changes

Introduction

Wave-dominated sandy shorelines make up a significant proportion of coastal margins around the world (Davies, 1964). These sandy shorelines are commonly the active margin of a variety of Holocene sandy coastal barrier forms; transgressive barrier islands (e.g. the eastern coast of the US; McBride et al., 2013), embayed barriers of differing morphology formed under stillstand conditions (e.g. the temperate coastal margins of southern Australia and southeastern Brazil; Guedes et al., 2011) or regressive barriers under falling relative sea level (e.g. the uplifted coastline of Japan; Tamura et al., 2008). Of critical importance in these regional settings is whether they are situated within near-, intermediate- or far-field sites with respect to ice sheets influencing glacio-isostatic adjustment (GIA) and relative sea level. In addition, the nature and availability of sediments in the coastal zone, and the incident wave and wind energy, including the influence of storms and cyclones are critical drivers of coastal landscape change. Prograded barriers forming foredune ridge plains or beach ridge plains are one such barrier form where successive elongate low-relief shore-parallel ridges demarcate past shoreline positions. Local and regional studies around the world have used Optically Stimulated Luminescence (OSL) dating and Ground Penetrating Radar (GPR) to further refine the understanding of these prograded Holocene barrier systems (e.g. Brill et al., 2015; Choi et al., 2014; Costas et al., 2016; Dougherty, 2014, 2018a; Dougherty et al., 2019; Fruergaard et al., 2015a, 2015b; Guedes et al., 2011; Hede et al., 2015; Hein et al., 2013, 2014, 2016; Madsen et al., 2007; Nooren et al., 2017; Reimann et al., 2010, 2011; Rémillard et al., 2015; Rink and López, 2010; Sander et al., 2018; Tamura et al., 2012; Timmons et al., 2010). In Australia, similar methods have been adopted demonstrating variable patterns of Holocene sedimentation influenced by the regional setting (e.g. Engel et al., 2015; Forsyth et al., 2010, 2012; Gontz et al., 2014; Jahnert et al., 2012; Murray-Wallace et al., 2002; Nott, 2011; Nott et al., 2015; Nott and Forsyth, 2012; Oliver et al., 2015, 2017b, 2017c, 2019; Oliver and Woodroffe, 2016; Tamura et al., 2018, 2019).

The Australian coastline, with its rich history of coastal studies and many examples of foredune ridge plains, would appear to be an ideal location to evaluate regional trends of Holocene coastal progradation. Recent reviews have tended to focus on sandy ridge formation and the potential to derive palaeoenvironmental information (Goslin and Clemmensen, 2017; Oliver et al., 2017a; Otvos, 2000; Scheffers et al., 2011; Tamura, 2012; Tanner, 1995; Taylor and Stone, 1996). This leaves significant scope for investigating the interplay of fundamental drivers of prograded barrier development and sandy shoreline behaviour such as sea-level changes and variation in sediment supply (Roy and Thom, 1981).

Geochronological datasets for the Holocene ridge plains around Australia reveal that the onset of ridge deposition at many sites is strongly correlated with the time when sea level reached its present level (Lewis et al., 2013). The extensive radiocarbon dating in southeastern Australia in the late 1970s–1980s suggested the onset of ridge deposition occurred ~6500 years ago. In contrast, OSL dating of many of these same sequences, and others, reveals that the most landward ridge was deposited somewhat earlier, between 8000 and 7000 years ago (e.g. Oliver et al., 2015). The role of falling sea level in promoting shoreline progradation around Australia (forced regression) has been discussed recently (Dougherty, 2018b; Oliver et al., 2018a); however, further studies presenting physical evidence for higher sea levels in prograded barrier stratigraphy and morphology are needed. Without numerous well-dated examples, it is premature to infer the influence of sea level on shoreline progradation. ‘Only when comparable data are assembled for all boundary conditions, such as sediment input, and other relevant climate factors can sea level be introduced as a possible explanation’ (Chappell, 1987: 304).

Variation over time in sediment supply to the Australian coast has resulted in significant differences in the pattern of shoreline progradation and barrier development. Where calculated, shoreline progradation rates in temperate southern Australia generally fall within a defined range of 0.1–1 m/yr with a few exceptions, for example, Keppel Bay (Brooke et al., 2008) where rivers augment the supply of sand from the inner continental shelf. These progradation rates appear sufficient to maintain an advancing shoreline with a stable sea level. The volumes of sediment stored in these barrier systems imply that these landforms are significant sediment ‘sinks’ in a sediment budget context (Davies, 1974; Thom et al., 2018). For sites on the eastern coast of Australia where a volume of sediment delivery over the Holocene has been calculated, the values generally fall within 1–2 m³/m/yr (e.g. Oliver et al., 2017c, 2019), although this may underestimate the total volume stored in the barrier as these values refer to volume above present-day mean sea level (MSL). In general, however, values are in accord with the average Holocene supply of sand to the coast reported in Table 1 of Short (2010). Sediment supply has been viewed as an important control on shoreline behaviour along other coastlines (e.g. Billy et al., 2013), and several studies have sought to quantify sediment volumes in coastal barrier successions (Billy et al., 2018; Hein et al., 2016; Kelley et al., 2005; Otvos, 2018).

Table 1.

OSL ages and measurement details for samples collected from Rivoli Bay and Guichen Bay, South Australia.

Sample code	Lab code	K (%)	U (ppm)	Th (ppm)	Cosmic (Gy/ka)	Total dose rate (Gy/ka)^a,^b	Discs used for CAM	CAM De (Gy)	OD (%)	OSL age (years)
Guichen Bay
GB1	UOW-2110	0.16	1.01	1.53	0.17 ± 0.02	0.65 ± 0.04	23	3.27 ± 0.06	8	5010 ± 330
GB2	UOW-2111	0.11	0.97	1.11	0.17 ± 0.02	0.57 ± 0.04	23	2.46 ± 0.11	22	4290 ± 340
GB3	UOW-2112	0.10	1.09	1.40	0.17 ± 0.02	0.61 ± 0.04	24	3.51 ± 0.05	7	5780 ± 390
GB4	UOW-2113	0.09	1.02	0.98	0.17 ± 0.02	0.56 ± 0.04	24	3.14 ± 0.05	7	5650 ± 380
GB5	UOW-2114	0.10	1.04	0.88	0.17 ± 0.02	0.56 ± 0.04	23	4.12 ± 0.06	7	7320 ± 500
GB7	UOW-2116	0.10	1.02	0.92	0.17 ± 0.02	0.56 ± 0.04	16^c	2.74 ± 0.05	6	4910 ± 340
GB8	UOW-2117	0.07	1.04	0.89	0.17 ± 0.02	0.53 ± 0.03	22	2.87 ± 0.05	7	5380 ± 370
Rivoli Bay
RV1	UOW-2069	0.12	0.96	0.94	0.17 ± 0.02	0.57 ± 0.04	23	3.61 ± 0.07	8	6350 ± 430
RV2	UOW-2070	0.10	1.07	0.91	0.17 ± 0.02	0.57 ± 0.04	22	2.93 ± 0.23	11	5120 ± 530
RV3	UOW-2071	0.07	1.12	0.82	0.17 ± 0.02	0.55 ± 0.04	23	3.12 ± 0.14	16	5680 ± 450
RV4	UOW-2072	0.08	1.05	0.78	0.17 ± 0.02	0.54 ± 0.03	23	2.80 ± 0.07	12	5180 ± 370
RV5	UOW-2073	0.14	1.06	0.84	0.17 ± 0.02	0.60 ± 0.04	24	2.93 ± 0.04	6	4890 ± 330
RV6	UOW-2074	0.13	0.96	0.83	0.17 ± 0.02	0.57 ± 0.04	23	2.69 ± 0.08	14	4740 ± 340
RV7	UOW-2075	0.10	1.07	1.88	0.17 ± 0.02	0.63 ± 0.04	24	1.99 ± 0.04	9	3150 ± 210
RV8	UOW-2076	0.12	1.07	0.76	0.17 ± 0.02	0.58 ± 0.04	24	1.21 ± 0.03	9	2100 ± 150
RV9	UOW-2077	0.10	1.09	0.84	0.17 ± 0.02	0.57 ± 0.04	24	2.77 ± 0.06	11	4870 ± 340
RV10	UOW-2078	0.06	1.04	0.76	0.17 ± 0.02	0.52 ± 0.03	24	2.22 ± 0.04	8	4250 ± 300
RV11	UOW-2079	0.11	1.03	0.99	0.17 ± 0.02	0.58 ± 0.04	24	1.82 ± 0.03	7	3160 ± 210
RV12	UOW-2080	0.09	1.02	0.76	0.17 ± 0.02	0.54 ± 0.03	18	1.45 ± 0.12	14	2680 ± 290
RV13	UOW-2081	0.12	1.10	0.93	0.17 ± 0.02	0.60 ± 0.04	23	1.23 ± 0.03	10	2040 ± 140
RV14	UOW-2082	0.11	0.93	0.77	0.17 ± 0.02	0.55 ± 0.04	24	0.99 ± 0.02	9	1810 ± 130
RV15	UOW-2083	0.10	1.08	1.30	0.17 ± 0.02	0.60 ± 0.04	20	0.64 ± 0.01	8	1070 ± 70
RV16	UOW-2084	0.09	1.07	0.74	0.17 ± 0.02	0.55 ± 0.04	21	0.45 ± 0.02	16	810 ± 60
RV17	UOW-2085	0.17	1.03	0.86	0.17 ± 0.02	0.62 ± 0.04	20	0.14 ± 0.01	10	230 ± 20
RV18	UOW-2086	0.13	0.98	0.73	0.17 ± 0.02	0.57 ± 0.04	24	1.96 ± 0.05	13	3440 ± 250
RV19	UOW-2087	0.16	0.93	0.81	0.17 ± 0.02	0.59 ± 0.04	24	1.15 ± 0.02	8	1950 ± 130
RV20	UOW-2088	0.11	0.96	0.75	0.17 ± 0.02	0.54 ± 0.04	23	0.05 ± 0.00	15	90 ± 10
RV21	UOW-2089	0.14	1.02	0.96	0.17 ± 0.02	0.60 ± 0.04	22	3.95 ± 0.07	8	6590 ± 450
RV22	UOW-2090	0.17	0.97	0.80	0.17 ± 0.02	0.60 ± 0.04	24	4.05 ± 0.09	7	6710 ± 460
RV23	UOW-2091	0.13	0.97	0.83	0.17 ± 0.02	0.57 ± 0.04	18	4.01 ± 0.08	8	7070 ± 490
RV24^c	UOW-2092	0.13	1.22	1.21	0.14 ± 0.01	0.62 ± 0.04	17	0.22 ± 0.02	27	350 ± 30
RV25	UOW-2093	0.13	0.98	0.81	0.17 ± 0.02	0.57 ± 0.04	12^c	0.11 ± 0.01	29	190 ± 20
RV26	UOW-2094	0.07	1.09	0.61	0.17 ± 0.01	0.53 ± 0.03	10^c	1.01 ± 0.07	20	1920 ± 190

CAM: central age model; OD: overdispersion; OSL: Optically Stimulated Luminescence.

A water content of 10 ± 5% was assumed for all samples for comparison with Murray-Wallace et al. (2002).

All samples include an internal dose rate contribution of 0.03 ± 0.01 Gy/ka based on measurements made on Australian quartz (Bowler et al., 2003).

Only small amounts of measurable grains were available, hence less discs were measured.

Burial depth for sample RV24 is 240–255 cm.

Given this context, the numerous prograded barriers around Australia represent an important archive of Holocene sea-level changes, the impact and frequency of cyclones and storms, and the role of sediment budget in beach/dune morphology and behaviour. To further the understanding of these important sedimentary repositories, we revisit the Guichen Bay foredune ridge plain in southern Australia, as previous investigations left several important questions unanswered regarding the chronology of the innermost ridges (Murray-Wallace et al., 2002) and the volume of sediment in the plain over the full depositional history (Bristow and Pucillo, 2006). In addition, we present a new chronology and morphostratigraphic data for the Rivoli Bay foredune ridge plain to the southeast of Guichen Bay to compare patterns of shoreline progradation and consider the role of Holocene sea level, sediment supply and variable wave refraction in shaping their depositional history.

Regional setting

Guichen and Rivoli Bays are situated adjacent to the Bonney Shelf on the present-day seaward margin of the Coorong Coastal Plain in southern Australia (Figure 1). The Coorong Coastal Plain is a distinct morphotectonic province, defined by a series of emergent coastal barriers of Pleistocene age, deposited during interglacial and warm interstadial sea-level highstands (Hossfeld, 1952; Murray-Wallace, 2018; Sprigg, 1952, 1979). The coastal barrier landforms have developed within a temperate sedimentary carbonate province and comprise successions of semi-lithified calcarenite (aeolianite) mapped as the Bridgewater Formation (Belperio, 1995; Boutakoff, 1963). The barrier successions have been slowly uplifted by epeirogenic processes throughout the Quaternary (Murray-Wallace, 2018). The regional uplift of 0.07 mm/yr (Murray-Wallace et al., 2001) amounts to 0.8 m over the Holocene (~11,700 years). The modern coastline in the area of Guichen and Rivoli Bays, in part, is represented by cliffs developed on consolidated aeolianite of the emergent Robe Range barrier complex. The aeolianite, representing the most recently formed portions of the Bridgewater Formation, was deposited during the late Pleistocene warm interstadials of MIS 5c and 5a, 105 and 82 ka ago, respectively. Several of the modern outcrops of the barrier correlate with MIS 5c and have been dated by thermoluminescence at 116 ± 6 ka (Huntley et al., 1994). Modern and Holocene dune sands cap the Pleistocene successions. The younger interstadial component (MIS 5a) does not crop out above the present sea level. The late Pleistocene aeolianites of Robe Range are capped by a 1-m thick, strongly indurated calcrete profile which has blanketed the succession and during the last glacial cycle, largely protected the barrier dune complex from denudation.

Figure 1.

The location of Guichen Bay and Rivoli Bay with respect to the regional Quaternary landforms of the Coorong Coastal Plain, southern Australia. Robe Range and Woakwine Range are evident and the corridor between now hosting a series of large lakes. The ‘breaks’ in Robe Range form the embayments of Guichen and Rivoli Bays, which have accumulated Holocene sediments.

Robe Range experienced a rapid phase of erosion along its seaward side in the late stages of Post-Glacial Marine Transgression (PGMT) as the sea level rose to at or slightly above present level approximately 7000 years ago (Lewis et al., 2013). Erosional remnants of Robe Range occur up to 1–2 km offshore as small islands rimmed with modern shore platforms that developed since the attainment of the present sea level. Guichen and Rivoli Bays represent former marine embayments of erosional origin within the Robe Range barrier complex. The erosional development of the bays was followed by their sequential infill by sand derived from Robe Range, the nearshore zone and the inner continental shelf.

Guichen Bay is approximately 10 km long with an embayment fill, foredune ridge plain succession extending 4 km inland to the foot slopes of Woakwine Range. The bay faces west receiving high wave energy from the Southern Ocean. Rivoli Bay, located 45–55 km to the southeast, is approximately 13 km long with a foredune ridge plain some 3 km wide in shore-normal cross-section. The bay faces southwest. The present-day beaches at Guichen and Rivoli Bays are high-energy intermediate-dissipative systems with a well-developed outer bar (Short, 2001). The landward margins of Guichen and Rivoli Bays are defined by the Woakwine Range, a last interglacial (MIS 5e; 125 ka) coastal barrier landform (Murray-Wallace et al., 1999; Short, 1988). The marginal headlands of these embayments are defined by consolidated aeolianite of Robe Range. With the attainment of the Holocene sea-level highstand, a marine corridor existed between the Robe and Woakwine Ranges, and Guichen and Rivoli Bays represented the entrances to the marine corridor before the development of the foredune ridge plains (Cann et al., 1999; Figure 1). The sediment within the marine corridor, represented by Holocene coquina, dominated by bivalve molluscs, slowly aggraded, resulting in the reduction of accommodation space. With continued shoreline progradation and the successive formation of foredune ridges in Guichen and Rivoli Bays, the former marine corridor became isolated from the sea and underwent a transition to a series of shallow lakes characterised by fluctuating salinities.

The Bonney Shelf is a narrow (30–80 km) un-rimmed continental shelf that extends from Cape Jaffa to Cape Northumberland. The shelf experiences the full force of the high wave energy conditions of the Southern Ocean. Waves are dominantly southwesterly with swell at Cape Northumberland exceeding 2 m for 68% of the year. Modal deep-water wave heights are typically >2.5 m, and long-period swell waves (>12 s) have wavelengths of >200 m (Short and Hesp, 1982). Semi-diurnal tides are microtidal with a range consistently <1 m. There is a marked winter to summer change in MSL of up to 1 m. The region experiences a Mediterranean-style climate with hot, dry summers and cool, wet winters. The highest rainfall is concentrated from May to August. Mean annual rainfall recorded at Robe Post Office is 635 mm, and mean annual temperature is 14.7ºC (Climatic Averages Australia: http://www.bom.gov.au/climate/map/climate_avgs/clim_avg1.shtml).

Sedimentation on the Bonney Shelf during the Quaternary, and today, is of temperate carbonate origin, characterised by the deposition of mixed quartz-skeletal carbonate sands. Terrigenous minerals make up <10% of the sediment matrix. The bioclastic constituents are derived from the comminution of marine invertebrates, principally molluscs (bivalves and gastropods), echinoids, coralline algae, bryozoans, foraminifers and sponge spicules. Other sedimentary constituents include stained and abraded relict carbonate grains derived from erosion of the Pleistocene Bridgewater Formation and the Oligo-Miocene Gambier Limestone (James and Bone, 2011).

Methods

Airborne LiDAR processing

Airborne LiDAR data were acquired from South Australian Natural Resources South East as point cloud data for Guichen and Rivoli Bays and processed using ArcGIS 10.2 to derive a bare Earth Digital Elevation Model (DEM). A 5-m resolution bare Earth DEM was also acquired from Geoscience Australia (GA) for the broader Coorong Coastal Plain. Both these elevation datasets enabled visualisation of the Guichen and Rivoli Bay ridge plains and were used for the construction of topographic profiles and calculation of sediment volumes.

OSL dating

Samples for OSL dating were collected from the aeolian dune facies of foredune ridges at Guichen and Rivoli Bays (Figures 2 and 3). All samples were extracted from 95 to 110 cm below the crest of the relict foredunes using a hand auger and steel sampling barrel, except sample RV24, which was from 240 to 255 cm (Table 1). Samples were collected in this manner to ensure that only the aeolian dune facies (commonly >3 m thick) was sampled. At Guichen Bay, sampling targeted areas of the ridge plain not sampled in the transect of Murray-Wallace et al. (2002) and focused on older ridge sets 1 and 2 as identified by Bristow and Pucillo (2006). At Rivoli Bay, prominent ridges were targeted and samples were generally spaced evenly along three separate transects in the north, centre and south of the ridge system. At each sample site, a hole was drilled with a hand auger to the desired depth and a sampling barrel attached to the end of an auger rod was hammered into the base of the hole to extract an undisturbed sample of light-safe grains. This steel sampling barrel was subsampled with a smaller diameter aluminium tube 15 cm long, so that the steel sampling barrel could be reused. The aluminium tube was hammered into the sampling barrel, then removed, capped, taped and labelled.

Figure 2.

DEM derived from LiDAR of Guichen Bay showing the planform configuration of the ridge plain and the position and age estimate of the OSL samples collected in this study. The location of OSL ages of Murray-Wallace et al. (2002) is also shown and the position of the GPR transects of Bristow and Pucillo (2006). Ridge sets 1 to 6 are demarcated according to the analysis of Bristow and Pucillo (2006). Bathymetric contours are shown indicating depth below Highest Astronomical Tide (HAT).

Figure 3.

(a) DEM of Rivoli Bay derived from LiDAR showing the planform configuration of the ridge plain and the position and age estimates of the OSL samples collected in this study. Position of GPR transects are shown and four topographic profiles labelled P1–P4. Bathymetric data are also shown highlighting the presence of several reefs near the middle of the Bay (depth indicated below Lowest Astronomical Tide (LAT). (b) Oblique aerial photo looking SSE over Rivoli Bay with several prominent features demarcated including the reef in the centre of the Bay, cuspate foreland and central OSL dating and GPR transect, and the crest of the high dune in the foreground with a prominent relict blowout. Photo courtesy of Kym Redman.

OSL dating was completed at the University of Wollongong OSL Laboratory following the methods reported in Oliver et al. (2015, 2017b, 2017c). Under dim red light conditions, sample tubes were opened and the outer 2–3 cm of sample was used for measurement of the environmental dose rate. Light-safe grains from the inner part of the sample tube were then subject to standard laboratory preparation procedures using the 212–180 µm fraction of quartz grains. Stimulation, measurement and irradiation of the isolated quartz grains used a modified single-aliquot regenerative-dose (SAR) protocol after Murray and Wintle (2000). Final equivalent dose (De) values for each sample were determined using the central age model (CAM) of Galbraith et al. (1999), which is well suited to well-bleached homogeneous dune sands (Costas et al., 2012). For several samples, less than the full 24 discs were measured because of the paucity of quartz grains in the 212–180 µm fraction (Table 2). The sediments contained a high percentage of calcium carbonate grains, typically greater than 70% for all samples (see also Murray-Wallace et al., 2002) with ~30% of these skeletal carbonate grains being derived from the erosion of Robe and Woakwine Ranges (Murray-Wallace et al., 2002). Other discs were omitted from the CAM as being obviously anomalous.

Table 2.

Sediment volumes supplied to Guichen and Rivoli Bay over the mid- to late-Holocene calculated from a DEM derived from LiDAR and OSL dating from Murray-Wallace et al. (2002) and this study.

Guichen Bay	Ridge set 1 7300–5800 years	Ridge set 2 5800–5400 years	Ridge set 3 5400–5300 years	Ridge set 4 5300–4400 years	Ridge set 5 4400–1800 years	Ridge set 6 1800–present
m³/yr	10,196	128,768	415,566^a	50,483^a	40,973^a	32,281
Rivoli Bay	Ridge set 1 7000–6500 years	Ridge set 2 6500–5000 years	Ridge set 3 5000–3500 years	Ridge set 4 3500–2000 years	Ridge set 5 2000–1000 years	Ridge set 6 1000–present
m³/yr	73,704	26,701^b	46,409^b	28,424^b	25,436	22,386

Ridge sets for Guichen Bay are defined according to Bristow and Pucillo (2006), while for Rivoli Bay they have been drawn with reference to the geographic position of OSL ages presented in this study. Ridge sets 4 and 5 at Rivoli Bay account for the higher sea level inferred from GPR radargrams. Volumes are for sediment above MSL (accounting for a sea-level highstand in ridge sets 4 and 5 at Rivoli Bay) so as to be comparable with the volumes reported by Bristow and Pucillo (2006) for ridge sets 3, 4 and 5 at Guichen Bay.

From Bristow and Pucillo (2006). For ridge set 3 a calculation error was found in Table 2 of Bristow and Pucillo (2006), hence the new value is presented here; however, this is not significant because the ages overlap and are only separated by 100 years.

Ridges are truncated in places, so volume is underestimated.

Dose rates for each sample were estimated from measured concentrations of uranium, thorium and potassium determined using ICP-MS and ICP-OES analysis completed by Intertek Genalysis. These concentrations were converted to dose rates using the conversion values reported in Guérin et al. (2011). Cosmic dose rate contributions were calculated for each sample taking into consideration geographic position, sediment density, altitude and depth of overburden (Prescott and Hutton, 1994). A water content of 10 ± 5% was assumed for all samples which increased the final age by approximately 5% and increased the uncertainty margin on each age by 30–40%. This value is higher than the assumed value of 5 ± 2.5% adopted in recent studies of coastal barriers in southeastern Australia (Oliver et al., 2017b, 2015b, 2019; Oliver and Woodroffe, 2016), but was chosen here so that direct age comparisons can be made with previous results from Guichen Bay (Murray-Wallace et al., 2002).

GPR acquisition and processing

To augment the GPR data of Bristow and Pucillo (2006) at Guichen Bay (see Figure 2 for location), GPR transects were collected at Rivoli Bay along several sandy four-wheel drive vehicle tracks corresponding with the position of the central and southern OSL dating transects (see Figure 3 for locations). GPR data were collected using the Mala ProEx system with a 250-MHz shielded antenna, fixed antenna separation and a sampling interval of 0.05 m measured according to the distance wheel on the ProEx cart apparatus. Topography for the GPR transect was defined with a survey tripod and staff for the southern GPR transect and using interpolated heights from the LiDAR data at specific GPS positions and for the central transect due to the length (1.2 km). GPR processing was completed in RadExplorer Version 1.42 and standard processing routines were applied to the data including desaturation (dewow), time zero correction, horizontal background removal, gain or amplitude correction, bandpass filtering, migration correction and topography adjustment. Default settings were used for each process, with the exception of gain correction where a depth model was constructed to increase gain with increasing depth along the profile. This process improves visual interpretation of deeper reflections, but inhibits the ability to detect variations in reflection strength down trace as trace amplitude values are inflated by the gain curve with increasing depth. However, comparisons along the profile at a single depth are still valid as the same gain curve is applied to all traces along the profile.

Velocity values for depth correction were determined from established values for homogeneous sandy barrier sediments in this region (Bristow and Pucillo, 2006) and prior experience at other locations in southeastern Australia (Oliver et al., 2017c). Considering this, values of 0.14 and 0.07 m/ns were adopted above and below the water table. This is visually expressed with a split depth axis following the procedure of Bristow and Pucillo (2006). Profiles corrected for depth and topography were referenced to MSL by overlaying the topographic survey. Several beach profiles were extracted from LiDAR data proximal to the GPR transect locations for comparison. All heights in the LiDAR data are presented as relative to 0 m Australian Height Datum (AHD) which approximates MSL around the Australian coastline. These profiles were also converted to a split depth axis for comparison of geometry with the processed GPR radargram. GPR data were interpreted following the procedures of Jol and Bristow (2003) and Neal (2004) in reference to sedimentary environments. The contact between the upper beachface and dune was mapped following Dougherty (2014) and Costas et al. (2016) to examine relative sea-level changes. An uncertainty of ±0.3 m was applied to each measurement to account for uncertainty in the GPR radargram due to antenna resolution, topographic correction and natural variability of the beach/dune contact.

Sediment volume calculation

Sediment volumes for the Guichen and Rivoli Bays were determined according to sub-sections of each coastal plain termed ‘ridge sets’. At Guichen Bay, the ridge sets were defined according to the analysis of Bristow and Pucillo (2006) (Figure 2). At Rivoli Bay, the ridge sets were defined by the geographic position of the OSL ages and the location of ridge truncations (Figure 3). Sediment volumes were calculated by multiplying the area of the desired ‘ridge set’ by an average sediment thickness derived from the LiDAR DEM above 0 m AHD for this same area. While this average sediment thickness does not fully account for variations in ridge height and swale depth for a given ridge set, it does eliminate the ‘problem’ of modification or reworking of the original foredune ridge surface, which in places is significant due to human disturbance and the presence of large transgressive dunes (Figure 3). Bristow and Pucillo (2006) calculated sediment volume at Guichen Bay for several ridge sets by deriving the cross-sectional area of sediments above the foreshore-upper beachface contact in the GPR profile and multiplying by the historical shoreline length. This method also averages the topography alongshore, as the two-dimensional cross section and associated ridge and swale heights are only measured at one location. The foreshore-upper beachface contact was approximately 0 m AHD (see Figure 8 of Bristow and Pucillo 2006), hence for calculations of sediment volume in this study, 0 m AHD was also adopted as the base for the sediment thickness value. The sediment thickness calculation for ridge sets 4 and 5 at Rivoli Bay accounts for the higher sea level inferred from GPR radargrams (see below). However, neither the calculations for Guichen Bay nor Rivoli Bay considered the regional uplift rate in order to directly compare with Bristow and Pucillo (2006) whose base level was ~0 m AHD (cf. Figure 8 of Bristow and Pucillo (2006)). As these volume calculations concern the subaerial component (above ~MSL) of the barrier sediment, they do not fully capture the Holocene sediment budget at these two sites. The usefulness of these volumes is therefore limited to inter- and intra-site comparison and does not represent the absolute barrier volume of Guichen or Rivoli Bay. Deeper cores and/or GPR data which penetrate deeper into the barrier stratigraphy are needed to determine the depth of shoreface sediments which may further constrain the relationship between barrier deposition and onshore sediment transport.

Results

OSL dating

The OSL ages for the relict foredune ridges at Guichen and Rivoli Bays reliably constrain the depositional history of these two coastal plains (Table 1). The sediments of both ridge plains are remarkably homogeneous as shown by their environmental dose rates, with an average total dose rate for all samples of 0.6 ± 0.03 Gy/ka (1-sigma uncertainty; Figure 4). The consistent results suggest a common origin for the sediments for both ridge plains. These environmental dose rates are comparable with those reported by Huntley et al. (1993, 1994) and Huntley and Prescott (2001) for the Pleistocene barrier landforms of the Coorong Coastal Plain which ranged between 0.45 and 0.7 Gy/ka.

Figure 4.

Total dose rate (Gy/ka) for all samples dated in this study. Sample codes correspond to Table 1.

The two oldest OSL ages from Guichen and Rivoli Bays are 7320 ± 500 years and 7070 ± 490 years, respectively, suggesting that the onset of coastal progradation coincided closely with the attainment of the present sea level along this coastline (Lewis et al., 2013). At Guichen Bay, a series of strategically placed samples demonstrate the southeastern portion of the ridge plain is the oldest (Figure 2). The southern portion of the ridge plain was deposited ~7300 years ago (ridge set 1, Figure 2), after which a significant shoreline reorientation defines the inner boundary of ridge set 2 (Figure 2) dated to ~5800 years ago. Additional age control in the north confirms the alongshore continuity of ridge set 2, which was deposited between ~5800 and ~5400 years ago (Figure 2). The dating transect of Murray-Wallace et al. (2002) and analysis of Bristow and Pucillo (2006) define the subsequent ridge sets at Guichen Bay. At Rivoli Bay, three transects outline the timing of the Holocene deposition of the ridge plain, with ages generally younging seaward (Figure 3). The oldest ridges are in the southeast of the plain (Figure 3), where progradation began ~7000 years ago. Progradation commenced in the north of the plain ~6500 years ago, while the innermost relict foredune in the centre of the plain formed 4870 ± 340 years ago. A rapid phase of progradation, followed by a slowing of progradation rate, is evident in the north and south of the plain, while in the centre of the plain, ridges steadily built seaward and progradation continued until close to the present (Figure 3). Recent transgressive dune activity in the south is dated to <100 years.

Ridge-plain morphology and sediment volume

Guichen Bay

The relict foredune ridges are very well defined at Guichen Bay and have been divided into a series of ridge sets by Bristow and Pucillo (2006) (Figure 2). Airborne LiDAR reveals that ridge crests are generally lower in the south, between 5 and 8 m AHD, and higher in the north, between 8 and 12 m AHD. The inter-ridge swales are generally 1–3 m below the crest height, although where ridge crests are 11–12 m AHD in the north of the plain, the inter-ridge swale may be up to 4 m below the crest elevation. Ridge spacing is between 30 and 60 m (crest to crest). In some parts of the Guichen Bay ridge plain, higher ridges occur periodically with lower ridges between. These higher ridges are spaced every 150–200 m and are 7–8 m AHD, while the intervening lower ridges are approximately 6 m AHD. This is best expressed along the central OSL dating transect of Murray-Wallace et al. (2002) between the samples dated 4400 ± 220 and 1800 ± 80 (Figure 2). There are several prominent truncations in the planform pattern of the ridges which represent significant shifts in shoreline orientation and length. One example is to the south of ‘Drain L’ (Figure 2), where a ridge dated by OSL to 5780 ± 390 years in this study truncates an older ridge set with the shoreline rotating in a clockwise direction to a new configuration.

Sediment volumes delivered to the Guichen Bay ridge system have been estimated by Bristow and Pucillo (2006) for three of the six ridge sets (3, 4 and 5, Figure 2). Volume estimates for these three ridge sets vary significantly between 400,000 and 50,000 m³/yr and cover the period from ~5400 to ~1800 years ago (Table 2). These volumes account for sediment above ~ 0 m AHD or MSL (cf. Figure 8 of Bristow and Pucillo (2006)). Airborne LiDAR data processed to create a bare Earth DEM combined with the additional OSL ages collected in this study enables an estimate of the sediment volume stored above present-day MSL for ridge sets 1, 2 and 6 (Figure 2), thus completing the Holocene depositional record. The volume of sediment accumulated for ridge set 1 (Figure 2) is ~10,000 m³/yr assuming deposition between ~7300 and ~5800 years ago (Table 2). Ridge set 2 equates to ~130,000 m³/yr assuming deposition between ~5800 and ~5400 years ago (Table 2). For ridge set 3 the bounding OSL ages defining the ridge set overlap in their uncertainty margins and are only separated by 100 years (Table 1). Therefore, the volume is very high (415,566 m³/yr, Table 1). Ridge sets 4 and 5 accreted at a rate of ~50,000 and ~40,000 m³/yr, respectively, according to Bristow and Pucillo (2006). Ridge set six is estimated to be 32,000 m³/yr, assuming deposition over the past 1800 years (Table 2), which equates to 2.8 m³/m/yr for the current shoreline length.

A higher sea level was accounted for by Bristow and Pucillo (2006) (who reported volume calculations for ridge sets 3, 4 and 5) using the lower contact of the beach facies to determine sediment thickness. Therefore, any change in elevation of this contact caused by a change in past sea level was captured. However, this lower contact is very close to the present MSL for most of the GPR data in Bristow and Pucillo (2006), so the calculation of sediment volume for ridge sets 1, 2 and 6 in this study adopted present MSL to estimate sediment thickness.

Rivoli Bay

In planform, the foredune ridges forming the embayment fill at Rivoli Bay can be divided into distinct ridge sets (Figure 3) following the method of Bristow and Pucillo (2006). The ridges at Rivoli Bay vary substantially in height, spacing and orientation alongshore (Figures 3 and 5) and these variations help to define distinct genetically related sediment packages or ridge sets. In the north of the coastal plain, crest height of the ridges steadily increases from 5 to 7 m AHD with inter-ridge swales 1–2 m below the ridge crest elevation. Spacing of the ridge crests varies greatly in this northern portion of the ridge plain ranging between 30 and 90 m. A high single dune containing several prominent blowouts attaining a maximum elevation of 16–18 m AHD is set back around 100 m from the present shoreline position with new ridges formed seaward of this feature near the centre of the plain. The beach sediments beneath this high dune were sampled and dated to 2100 ± 150 years; however, tracing this high dune alongshore, the OSL ages from the central transect suggest a younger age of ~1000 years for this high dune. In the centre of the plain, the ridge crests are higher than in the north, between 9 and 12 m AHD initially, then steadily decreasing to 5–7 m near the present shoreline (Figure 5). Inter-ridge swales are generally more than 2 m below the ridge crest elevation in this area of the plain. Spacing of the ridges from crest to crest is between 30 and 80 m. In the south of the coastal plain the older ridge crests are between 6 and 7 m AHD in elevation, inter-ridge swales are 1–2 m below the crest height and crest-to-crest spacing is 30–60 m. Higher ridges cross-cut this older succession with the younger shoreline rotating anti-clockwise (Figure 3). These higher ridges have crest heights between 10 and 12 m AHD, inter-ridge swales are 2–3 m below the crest elevation and crest-to-crest spacing is between 40 and 80 m. An extensive transgressive dune has developed more recently in the southern sector of the plain and has modified the original ridge topography along a 3.5-km stretch of the modern shoreline (Figure 3). At its widest point, the transgressive dune extends 1.7 km inland and is characterised by a complex morphology of nested parabolic dunes. The parabolic dune walls reach a maximum elevation of 25 m AHD (Figure 3).

Figure 5.

Topographic profiles from Rivoli Bay (see Figure 3 for locations) showing the variation in height of the ridges, the position of the OSL ages and demarcating the phases of depositional history identified for the plain corresponding to Figure 3.

Sediment delivery for ridge set 1 at Rivoli Bay (Figure 3) is estimated to be ~74,000 m³/yr (7000–6500 years), while ridge set 2 is ~27,000 m³/yr (6500–5000 years; Table 2). However, shoreline reconfiguration in this southern area of Rivoli Bay zones means that the sediment volume may be underestimated for ridge set 2. For ridge sets 3–6, sediment volume reduces steadily: 46,000 m³/yr (ridge set 3, 5000–3500 years), 28,000 m³/yr (ridge set 4, 3500–2000), 25,000 m³/yr (ridge set 5, 2000–1000 years) and 22,000 m³/yr (ridge set 6, 1000 years–present; Table 2). The sediment volume in ridge sets 4 and 5 takes into account the higher sea level inferred from GPR profiles at Rivoli Bay by reducing the sediment thickness according to the raised level of the base of the beach facies. During the most recent depositional phases, ridge sets 5 and 6, the volume is likely to have been concentrated in the centre of the plain as the shoreline configuration resembles a cuspate foreland morphology (Figure 3). For the most recent period (1000 years–present), the sediment delivery (22,000 m³/yr, Table 2) represents a value per metre of shoreline of 1.3 m³/m/yr, although this sediment is likely to be disproportionately concentrated towards the centre of the plain where shoreline progradation has been most evident during the past ~1000 years. Due to the uncertainty of the historical shoreline length at Guichen and Rivoli Bays, the volume per metre of beach per year has only been calculated for the modern beach length. It should be restated here that the volumes of sediment presented in this study for Guichen and Rivoli Bays (Table 2) are a subaerial barrier volume only and do not consider the volume of sediment deposited below ~MSL. Therefore, their use is limited to comparison within each site over time and between the two sites, as they do not represent the total sediment volume of the Holocene barrier succession.

GPR radargrams

The GPR datasets corresponding with the outer portion of the central and southern OSL dating transects of Rivoli Bay suggest that periodic storm events caused erosion and recovery of the active beach zone as is presently observed along this coastline (Figure 6). The prominent seaward dipping reflections present in the GPR datasets from the central and southern portions of the barrier correspond with two different facies within the barrier sediments. The upper sections of the profiles correspond with the dune facies, whereas the lower sections correspond with the upper beachface. Mapping the contact between the upper beachface and dune facies, principally for the central GPR transect, and comparing this with the modern elevation of this same facies contact suggest a relative sea-level fall from around +1.7 to 2.3 m above the present sea level (APSL) approximately 3500 years ago, reaching the present level ~1000 years ago (Figure 7).

Figure 6.

GPR radargrams from Rivoli Bay for a portion of the central and southern transects collected.

Figure 7.

Variations in elevation of the upper contact of the beachface over time according to the GPR radargrams.

A comparison was made between the geometry of beach/dune reflections in the GPR radargrams for the central and south transects with proximal modern beach profiles extracted from the LiDAR dataset. This demonstrated that for the southern transect, the geometry was similar for the modern and GPR-imaged profiles, while for the central transect the geometry of the profiles did not conform (Figure 6). This is principally because at the time the LiDAR was flown, the beachface at the southern end of Rivoli Bay had an erosional profile morphology with a narrow steep beachface, while in the central portion of Rivoli Bay, a fairweather beachface morphology was evident. For the central GPR radargram, the low-angle (1.7º) fairweather dissipative beachface observed in the modern LiDAR beach profile proximal to the seaward end of the GPR profile location, appears to be absent from the radargram (Figure 6). At the time of GPR collection at this same location in the centre of the plain, the beach was in an erosive morphological state with a prominent dune scarp (Figure 6) and field measurements of the upper beachface angle were ~6º. This higher angle upper beachface appears more comparable with the preserved reflections in the central GPR radargram.

Holocene deposition at Guichen Bay

The formation of the Guichen Bay ridge plain was first described by Thom et al. (1981) using radiocarbon dating of a single shell species from within the littoral to sub-littoral facies. Samples were extracted by hand auguring to the water table and then pumping up saturated sediments rich in shelly material. Unlike other sites (e.g. Moruya, Oliver et al. (2015)), no deeper cores to bedrock were collected so the basement contour of this site (and Rivoli Bay) is not known. The radiocarbon chronology of Thom et al. (1981) was subsequently compared with OSL dating of quartz sand from the aeolian facies of the relict foredune ridges (Murray-Wallace et al., 2002). Several GPR transects across the embayment fill succession were reported in Bristow and Pucillo (2006), who defined six separate ridge sets based on interpretation of ridge orientation and alongshore continuity (Figure 2). Despite these datasets, several important questions remained regarding the timing of deposition of the innermost ridge sets (1 and 2, Figure 2) in the southeast of the coastal plain. The air photo analysis of Bristow and Pucillo (2006), which identified six ridge sets, suggests the oldest ridges occur in the southeastern corner of the plain. This was also identified by Murray-Wallace et al. (2002), who noted the difference in ridge orientation suggested an earlier period of deposition. The OSL results from this study confirm this hypothesis with several ages between ~5600 and ~5800 years ago and one older age of 7320 ± 500 years from this area of the ridge plain (Table 2). One age of 4290 ± 340 from this region is anomalously young, considering its geographic position (Figure 2). We place more reliance on the older age of 7320 ± 500 years based on the planform arrangement of this ridge crest, the comparable result for the innermost ridge at Rivoli Bay (Figure 3), and the sea-level curve for the region which indicates sea level reached present between ~8000 and ~7500 cal. yr BP (Belperio et al., 2002; Lewis et al., 2013).

The disparity between radiocarbon dating of Thom et al. (1981) and the OSL ages of Murray-Wallace et al. (2002) in the central landward portion of the plain also required further clarification and is addressed by a series of new OSL ages presented in this study (Table 1, Figure 2). Thom et al. (1981) originally cautioned that the inner central portion of the plain (Figure 2), radiocarbon dated to >7200 years old, likely contained reworked transgressive shell material. The OSL chronology of Murray-Wallace et al. (2002) suggested deposition between ~5200 and ~5400 years ago for this region of the plain supporting the notion of sediment reworking foreshadowed by Thom et al. (1981). As the OSL method quantifies the time of deposition of the sediments while the radiocarbon ages define the cessation of radiocarbon uptake in marine invertebrates, some age differences would be expected, even if the radiocarbon analyses were undertaken on in situ (not reworked) carbonate material. As well as the possibility of reworking of transgressive sediment, Murray-Wallace et al. (2002) highlighted the reworking of bioclastic sand from the Robe and Woakwine Ranges as possibly contributing to older radiocarbon ages. However, Thom et al. (1981) state that radiocarbon ages were from a single species of shell from the coarse bioclastic sand and ‘comminuted shell fragments of mixed species were avoided’ (p. 19). The disparity may also be attributed in part to radiocarbon ages and OSL ages being from different facies and on different components of the sediment. The OSL ages are on dune sands, while the radiocarbon ages are for littoral to sub-littoral facies. The precise stratigraphical relationship of these two facies means there may be an offset in their depositional ages. The OSL ages presented in this study confirm that the central, inner portion of the ridge plain was deposited between ~5000 and ~5400 years ago with several samples to the north and south of the central transect of Murray-Wallace et al. (2002) (Figure 2). The innermost ridges in the northern portion of the plain were deposited around this same time, demonstrated by two OSL ages reported in this study: 4910 ± 340 and 5380 ± 370 years.

The seaward progradation of the Guichen Bay shoreline has been an important factor promoting the closure of the Robe-Woakwine marine corridor at its northern end near Robe. The OSL ages from this study suggest that the exchange of marine waters between the corridor and the Southern Ocean was substantially restricted by ~5500 years ago and reduced to a relatively narrow estuary mouth by ~4000 years ago, consistent with the independent evidence for the ages of coquina that formed within the seaway (Cann et al., 1999). The constriction of this seaway in the south by the progradation of Rivoli Bay occurred somewhat later, and thus, it is plausible that there would be a greater level of marine influence within the corridor towards its southern end after ~4000 years BP.

Holocene deposition at Rivoli Bay

The OSL chronology from the three transects at Rivoli Bay shows the general pattern of shoreline progradation spanning the past ~7000 years with six ridge sets identified according to the position of OSL ages (Figure 3). During the initial period after the PGMT between ~7000 and ~6000 years ago, shoreline progradation was limited to the southern portion of Rivoli Bay near the town of Southend (Figure 5) where seven ridges have developed while several others recurve inland and are likely slightly older than ~7000 years (Figures 3 and 5). The deposition of these ridges in the southeast of the plain occurred relatively rapidly demonstrated by the overlap in the uncertainties of the three OSL ages (7070 ± 490 to 6590 ± 450 years; Figure 8). The large area between the southern and central transects (Figure 3, P3 in Figure 5) is also inferred to have been deposited during this time and continued to develop until ~5000 years ago. From ~6500 to ~5000 years ago, progradation was evident in the north of the plain where the innermost ridge formed some 6350 ± 430 years ago (Figure 3; P1 in Figure 5). Shoreline progradation was rapid during this time in the north and the ridges are recurved with differing configurations, probably due to the changing configuration of the entrance of what is now Lake George, but at this time represented an open marine corridor between the Robe and Woakwine Ranges (Cann et al., 1999). Between ~5000 and ~3500 years ago, progradation occurred in the centre of the plain, and one continuous foredune ridge connecting the northern and southern ridge sets is evident (Figure 3). This continuous foredune ridge truncated the inner ridges and the shoreline rotated in a clockwise direction (Figure 3). During this interval, the northern portion of the plain prograded 700–800 m, although at a slower rate than previously (Figure 8). This slower rate of progradation may have resulted in the increased ridge height as there was more time for dune accretion.

Figure 8.

Barrier width as a percentage according to age for Rivoli Bay. The North, Central and South dating transects display distinctly different patterns of progradation. The south first progrades rapidly, then the north progrades rapidly, then the central transect begins and a continuous ridge is formed. The shoreline in the centre of the plain then continues at a steady rate while in the north and south, slower progradation occurs.

Between ~3500 and ~2000 years ago, all three OSL dating transects demonstrate seaward progradation, although at differing rates (Figure 8). The central portion of the barrier continued at a steady rate of 0.38 m/yr, outpacing the northern and southern ends of the embayment (Figure 8). This resulted in a relative straightening of the shoreline. During this time, ridge orientations in the northern portion of the plain suggest that the barrier may have extended towards the present-day town of Beachport (Figure 3). This section of the barrier is now eroded; however, it is probable that the barrier sealed off, or at least significantly restricted, the passage to the marine corridor between the Robe and Woakwine Ranges heralding the formation of what is now Lake George (Figure 3). Further evidence for this is that one of the first lake shoreline ridges which truncate the northern ridge system is dated to 1920 ± 190 years suggesting that a well-defined lake was likely present around this time (Figure 3). This accords with the findings of Cann et al. (1999), who concluded that progradation of the Rivoli Bay shoreline would have closed off the Robe-Woakwine corridor sometime after ~2000 yr BP. Also, during this time, there is some evidence from the GPR radargrams in the central and southern portions of the bay for a fall in relative sea level from a highstand of ~2 m APSL reaching the present level ~1000 years ago (Figure 7), which likely also promoted restriction of marine influence in the Robe-Woakwine corridor. The influence of this sea-level fall on shoreline progradation (forced regression) is complicated by the shoreline reconfigurations evident at this time in the shoreline history, but cannot be discounted.

Between ~2000 and ~1000 years ago, progradation continued steadily in the central portion of the ridge plain where the shoreline advanced approximately 370 m seaward (Figures 5 and 8). In the southern portion of the bay, the shoreline continued to advance towards its present position (Figure 3). If the shoreline continued at the rate of progradation that occurred between ~3500 and ~2000 years ago, it would have attained its present position ~400 years ago; however, it may have advanced further seaward and then eroded to its present position (Figure 8). The seawardmost OSL age of 90 ± 10 years represents the age of a more recent transgressive dune facies.

During the period from ~1000 to ~200 years ago, progradation continued in the centre of Rivoli Bay forming a cuspate foreland in planform (Figure 3). Progradation in the north and south of the bay was limited and sediment is likely to have been transported to the centre of the bay. This redistribution of sediments is due to the influence of the aeolianite reef structures in the centre of the bay (erosional remnants of Robe Range). These reef structures have played an important role in modifying the shoreline configuration through their influence on wave refraction patterns as the shoreline advanced into Rivoli Bay.

The single high foredune at Rivoli Bay in the north of the ridge plain seems to be a relict feature and contains multiple inactive blowouts (Figure 3; P1 in Figure 5). An OSL sample dated 2100 ± 150 years was extracted from the base of a blowout and represents the beach facies underlying this dune. Tracing the crest of this high dune alongshore suggests that the dune activity itself is associated with the sample dated 1070 ± 70 years on the central transect (Figure 3). This makes the mechanism responsible for the formation of this high dune difficult to decipher, but it may relate to a phase of beach retreat, shoreline stability or slowing of shoreline progradation ~1000 years ago. The planform orientation of this high dune suggests it developed as the northern shoreline of Rivoli Bay retreated and the cuspate foreland developed (Figure 3). The seawardmost position of the shoreline in the northern part of the bay is not known; however, the truncations of the ridges near the location of P1 in Figure 3 suggest it may have been around 200–300 m further seaward than its present location.

The semi-permanent closure of the entrance to Lake George, before an artificial opening was cut, appears to have occurred sometime in the past ~400 years indicated by the OSL ages of 350 ± 30 and 190 ± 20 years, respectively. The former is located on the narrow single dune ridge in the northern corner of Rivoli Bay (Figure 3) and is located in a geographic position corresponding with the last opening position of the lake indicated by the recurved lake shoreline ridges suggestive of a small flood tide delta (Figure 3). The younger sample dated 190 ± 20 years is from the outermost lake shoreline ridge indicating constriction of the lake water body.

Shoreline progradation during the past ~200 years is limited to the centre of the ridge plain. The sample dated 230 ± 20 years is approximately 55 m inland of the present shoreline (Figure 3) representing approximately 150 years of shoreline progradation assuming the rate of 0.38 m/yr inferred for the past ~5000 years (Figure 8). Accordingly, the most seaward established dune is likely to have formed between ~60 and ~100 years ago. However, field observations suggest recent erosion of the shoreline at the centre of the bay defined by a sand cliff around 2–3 m high formed in the foremost established dune with slumping at the base and limited beach recovery or incipient dune formation. Several prominent indentations in the shoreline proximal to the northern dating transect (Figure 3) reveal concentrated zones of shoreline erosion. These indentations also correspond with locations of deeper water in the nearshore zone (according to bathymetry), suggesting a reduction in nearshore sediment and opportunity for wave energy to approach the shoreline with less energy dissipation in the surf zone.

Discussion

Sediment delivery to Guichen and Rivoli Bays

Murray-Wallace et al. (2002) inferred a large influx of sediment in the early period of coastal deposition at Guichen Bay derived from the erosion of Robe and Woakwine Ranges, as well as the inner Bonney Shelf, as responsible for the rapid seaward progradation of the shoreline indicated by the OSL dating. The additional OSL ages reported here confirm the timing of the rapid phase of progradation at Guichen Bay between ~5800 and ~5000 years ago. It is evident that a large volume of sediment was deposited during this time at Guichen Bay as ridge sets 2 and 3 have volumes of 128,768 and 415,566 m³/yr, respectively. However, these volume calculations, especially for ridge set 3, are indicative only as the OSL ages overlap in their uncertainty margins and ridge set 3 is only separated by 100 years. Allowing for the maximum time period for ridge set 3 at Guichen Bay according to the OSL age uncertainty margins reduces this value to 74,208 m³/yr. At Rivoli Bay, sediment supply for the older ridge sets was also greater before ~5000 years ago (Table 1), as ridge set 2 is likely significantly underestimated given the prominent ridge truncation (Figure 3, see also P3 in Figure 5).

A substantial portion of the mixed quartz-skeletal carbonate sand deposited within the sand ridge sequences of Guichen and Rivoli Bays is represented by relict carbonate grains derived from the erosion of Robe Range, a coastal barrier landform of late Pleistocene age. The skeletal carbonate grains, which include comminuted molluscs, coralline algae, bryozoans, echinoids and foraminifers are significantly abraded, attesting to a pre-Holocene age and local sediment source. Further evidence for the relict origin of a significant proportion of the carbonate sediment is based on amino acid racemization analyses of the skeletal carbonate sands (Murray-Wallace et al., 2001). The extent of leucine racemization (total hydrolysable amino acids) measured in modern beachface sediment at Guichen Bay, yielded a D/L value of 0.225 ± 0.002, which is significantly higher than noted for biogenic materials of Holocene age (e.g. leucine D/L of 0.09 ± 0.03 on the marine mollusc Katelysia sp. with a radiocarbon age of 7910 ± 140 yr cal BP; Cann et al., 1991). In a similar manner, sediment from a Holocene sand ridge 2.5 km inland from the present-day shoreline of Guichen Bay (where ridges are ~5000 years old), yielded a leucine D/L value of 0.317 ± 0.003, further attesting to a Pleistocene source for the skeletal carbonate sand. At Rivoli Bay, sediment previously collected from the relict foredunes 0.75 km inland from the modern coastline along transect 1 of this study (where ridges are ~2000 years old), yielded a leucine D/L value of 0.369 ± 0.061 (Murray-Wallace et al., 2001) with a conventional radiocarbon age of 13,700 ± 80 yr BP (Beta-104526). The leucine D/L value is too high to represent a Holocene age and reflects the incorporation of a substantial portion of relict carbonate grains from the inner Bonney Shelf. The balance of older relict carbonate grains and biogenic carbonate of Holocene age in the sediments of Guichen and Rivoli Bays requires further investigation. In carbonate-rich settings of Australia such as western Victoria, South Australia, Western Australia as well as places around the world such as parts of the Mediterranean, South Africa and the Caribbean (Brooke, 2001; Davies, 1972), the supply of biogenic carbonate, making up a significant component of the sediment supply to beaches, needs to be better quantified, as sustaining this supply may help to mitigate coastal retreat. Characterisation of the shelf sediments and marginal marine environments in the Robe region has been completed by James et al. (1992) and Boreen et al. (1993), however, it is difficult to determine production rates, abrasion rates and rates of onshore transport.

The inclusion of relict carbonate grains in the sediment matrix and the resulting overestimate of the conventional radiocarbon age, illustrates the value of OSL dating, where the time of deposition of the sediments can be determined. The uniformity of the total dose rate (Figure 4), for the OSL samples in this study, suggests that the sediments of Guichen and Rivoli Bay are mineralogically relatively homogeneous. Furthermore, the absence of secondary carbonates in the Holocene sediments examined in this study precluded the need for more complex dosimetry measurements, which account for interstitial carbonate of diagenetic origin replacing pore water during the formation of Pleistocene aeolianites and beachrock (Jacobs, 2008; Nathan and Mauz, 2008).

Following the PGMT, a large volume of sediment contributed to the rapid phases of progradation evident at Guichen Bay before ~5000 years ago, and at Rivoli Bay in the south and north also before ~5000 years ago (Table 2, Figures 2 and 8). The coincident timing of these rapid phases of progradation at Guichen and Rivoli Bay implies that sediments transported across the Bonney Shelf with rising sea levels during the PGMT and also sediment derived from the marine erosion of Robe and Woakwine Range were significant in promoting initially rapid shoreline progradation. The erosion of Robe and Woakwine Range, and other associated marine deposits, by subaerial processes during the glacial period, when sea level was much lower than the present, may also have resulted in the accumulation of sediments in the vicinity of Guichen and Rivoli Bays. However, the armouring of the Robe and Woakwine Ranges by calcrete caprocks prevents, to a large degree, the scale of erosion found over glacial–interglacial cycles in quartz-dominated settings (e.g. Oliver et al., 2018b). Furthermore, terrestrial aeolian dunes may have developed on the Bonney Shelf during the sea-level lowstand as the last glacial aged dunes are found in other parts of the Coorong Coastal Plain (e.g. Murray-Wallace et al., 2010). Collectively, these accumulated sediments would have been available for onshore transport and deposition during the early phase of barrier development at both the sites.

The differences in sediment delivery to Guichen and Rivoli Bays over the mid- to late-Holocene have impacted directly on the ridge morphology. There is a general tendency for the rapid phase of progradation before ~5000 years ago to have produced lower relief ridges compared with slower phases of progradation where higher ridges are evident. A slower progradation rate enables more time for dune building on each ridge before seaward accretion of the beach leads to a ridge abandoned. This has been shown in recent studies in NSW (e.g. Carvalho et al., 2019; Oliver et al., 2019) and is reflected in recent modelling of beach/shoreface and foredune sediment flux whereby variations therein produce predictable morphological variations (Ciarletta et al., 2019). Moore et al. (2016) have also found this relationship between progradation rate and foredune height, taking into account lateral dune growth rate and vegetation dynamics.

Holocene sea-level change in southern Australia

Holocene sea level in southern Australia reached the present level between ~8000 and ~7500 cal. yr BP with a highstand between 3 and 1 m APSL evident from preserved sedimentary facies in Spencer Gulf and Gulf St Vincent (Figure 1; Belperio et al., 2002; Lewis et al., 2013). The highstand was spatially variable in its magnitude within the two Gulfs with sites further north recording higher relative sea level than those in the south (Belperio et al., 2002; Lewis et al., 2013). Several revisions in the interpretation of these highstand records have been undertaken as the contribution of neotectonism and hydro-isostasy have been considered and modelled (Chappell, 1987; Lambeck and Nakada, 1990; Lewis et al., 2013). The GPR-imaged beach facies preserved at Rivoli Bay suggest a fall in sea level from a highstand of 1.7–2.3 m APSL ~3500 years ago, attaining the present level ~1000 years ago. This record from Rivoli Bay broadly accords with both the proxy evidence (Belperio et al., 2002) and the modelling of Lambeck and Nakada (1990), despite being geographically to the southeast of the reported sites of Belperio et al. (2002) in Spencer Gulf (Figure 1). The age–height relationship of the data presented in this study favours a smooth and steady fall in the relative sea level as also considered most likely by Belperio et al. (2002) and Lewis et al. (2013). The GPR data of Bristow and Pucillo (2006) at Guichen Bay may also suggest a fall in the relative sea level of approximately 1 m sometime after ~3900 years ago based on the elevation of the upper beachface/dune interface. Another factor to consider is the ongoing subtle uplift in this region, cited by Cann et al. (1999) as important in the closure of the marine corridor between the Robe and Woakwine Ranges in the mid- to late-Holocene. The regional uplift of 0.07 mm/yr (Murray-Wallace et al., 2001) means that beach strata ~3500 years old would be expected to be uplifted to approximately 0.25 m. Accounting for this reduces the inferred sea-level highstand at Rivoli Bay ~3500 years ago to 1.45–2.05 m APSL.

The GPR data at Rivoli Bay captures the falling sea-level trend but does not indicate when this highstand first occurred, as only the outer portion of the ridge sequence was imaged with GPR. Further evidence is required from older ridges in the sequence to determine when sea level first attained this higher level. Although the ridge crests appear lower further landward in the sequence to the north and south (Figures 3 and 5), there is less dune sediment capping the ridges at these localities. In fact, at Rivoli Bay, in all topographic profiles except the central transect (P2), ridge heights generally increase (Figure 3), suggesting that the ridge height itself is not a reliable indicator of past sea level at locations where there the dune-capping is significant. The GPR for Guichen Bay may indicate that there was a highstand as early as ~5300 years ago, as the elevation of the upper beachface/dune interface is ~3–4 m APSL for much of the profile (cf. Figure 7 in Bristow and Pucillo (2006)), which is higher than the present. Further investigation of these two sites may yield a more complete record of the Holocene sea level and the results presented here encourage further study at these locations. However, the homogeneity of the sediments in terms of grain size and the potential for spatial variation in the elevation of the upper beach unit in this high-energy setting implies that a cautious approach is needed.

The degree to which a sea-level highstand during the mid-Holocene is observable in ridge plain strata more broadly in Australia is not well established and requires further examination. Some sites, such as Rockingham (Searle et al., 1988; Searle and Woods, 1986), Shark Bay (Jahnert et al., 2012) and Rivoli Bay in this study, point to higher sea levels during the mid-Holocene, while other sites hold the potential for future investigation (e.g. Seven Mile Beach; Dougherty, 2018b; Oliver et al., 2017b, 2018). Throughout much of southeastern Queensland and especially in NSW, evidence for a sea-level highstand has been elusive in ridge stratigraphy, although further examination at specific locations is needed. The review of Sloss et al. (2007) for NSW Holocene sea level suggests a highstand of ~1 m sometime after ~8000 years BP and continuing to ~2000 years BP, after which a fall to the present level is inferred based on examination of back-barrier estuarine successions. As archives which cover this timespan, re-examination of several ridge plains with established chronologies may enable refinement of the sea-level curves (e.g. Brooke et al., 2019).

The large transgressive dunes at Guichen and Rivoli Bays

There are several examples of transgressive dune activity at Rivoli Bay (Figure 3) and Guichen Bay (Figure 2). At Guichen Bay, blowout dunes with deflation hollows occur along much of the shoreline length, increasing in size to the north. Several examples also occur inland of the present shoreline, including a prominent example around 3.5 km east of Robe (Figure 2). Many of these dunes remain active and appear to relate to anthropogenic disturbance from farming, grazing and local sand extraction. Others close to the present shoreline have been active for some time, as can be seen from the 1975 air photo shown in Figure 2 of Murray-Wallace et al. (2002).

At Rivoli Bay, several prominent blowout dunes are evident north of the town of Southend (Figure 3). These blowouts may be the result of dune disturbance from grazing, or alternatively due to acute erosion of the foredune system, such that a low-point in the dunes enables blowout development according to the model of Short and Hesp (1982), or a combination of both factors. A much larger transgressive dune complex is evident between profiles 3 and 4 (Figure 3). This dune has migrated approximately 1.8 km inland at its widest point. The expansion of grazing, vegetation clearing, severe bushfires and disturbance by rabbits (widespread in South Australia by 1890) caused dune destabilisation along much of the southern Australian coastline during the late 1800s and early 1900s (Bourman and Murray-Wallace, 1991; Moulton et al., 2019; Murray-Wallace, 2018). The large transgressive dune at Rivoli Bay has been related to the influence of grazing and not from natural causes (Short and Hesp, 1986). The source of sediment for this dune is from the reworking of the ridge sequence rather than the active beach, as indicated by the location of the deflation hollow (which is set back 70 m from the present shoreline). This transgressive dune appears to have developed independently of the shoreline history, whereas the higher ridge in the northwest of Rivoli Bay (Figure 3) appears to be related to a phase of shoreline stability or retreat ~1000 years ago.

Conclusion

Several new OSL ages reported here for Guichen Bay supplement the dating transect of Murray-Wallace et al. (2002) and demonstrate the southeast corner of the Guichen Bay ridge plain was deposited ~7300 years ago in accord with the ridge sets identified by Bristow and Pucillo (2006). These new OSL ages also confirm the alongshore continuity of the ridge sets and support a rapid phase of progradation prior to ~5000 years ago.

OSL ages for Rivoli Bay demonstrate that several phases of deposition have occurred with shoreline orientation changing significantly. Rapid progradation in the south, and then in the north is evident before ~5000 years ago. After ~5000 years ago, slower progradation ensued in the south and the north while the centre of the plain prograded steadily at a rate of 0.38 m/yr to the present-day shoreline position due to the increased influence of wave refraction around offshore reefs.

At both Guichen and Rivoli Bays, the rapid progradation before ~5000 years ago is attributed to sediments available for onshore transport from the erosion of Robe and Woakwine Range as the sea level rose to the present. Sediment supply through the late-Holocene has decreased and the shorelines in the north and south of Rivoli Bay are presently eroding.

Raised beach strata in the Rivoli Bay ridge sequence imaged with GPR suggests a sea-level highstand of approximately 2 m APSL ~3500 years ago falling steadily and reaching the present ~1000 years ago. The GPR profiles at the Guichen Bay in Bristow and Pucillo (2006) may also indicate a highstand. These findings are consistent with the regional sea-level curves for this region (Lewis et al., 2013).

The new OSL ages from Guichen Bay and those from Rivoli Bay provide independent evidence for the closure of the marine corridor between the Woakwine and Robe Ranges through the mid- to late-Holocene in accord with the timing proposed by Cann et al. (1999). Significant restriction of the exchange of marine waters is likely to have occurred before ~5000 years ago as Guichen and Rivoli Bay rapidly prograded. The inferred fall in sea level, subtle ongoing uplift in the region and the continued progradation of the Guichen and Rivoli Bay shorelines in the mid- to late-Holocene have resulted in complete closure of the corridor and the formation of several shallow lakes.

Footnotes

Acknowledgements

This research was funded by the Australian Research Council, Discovery Project ID: DP150101936. Point Cloud LiDAR data were supplied by South Australian Natural Resources South East and a 5-m DEM from Geoscience Australia via the ELVIS portal. We thank Dr Zenobia Jacobs, Dr Terry Lachlan and Yasaman Jafari for their support in the OSL dating laboratory at the University of Wollongong. Kym Redman and Trevor Handow, local land owners at Rivoli Bay, kindly provided access to their property for OSL sample and GPR collection. In addition, Kym Redman supplied several photos and videos from an aircraft and quadbike showing the morphology of the Rivoli Bay ridge sequence and modern beach after a severe storm.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research was funded by the Australian Research Council, Discovery Project ID: DP150101936.

ORCID iD

Thomas SN Oliver

References

Belperio

(1995) The Quaternary. In: Drexel

Preiss

(eds) The Geology of South Australia (Geological Survey of South Australia, Bulletin 54, vol. 2),. Adelaide: Geological Survey of South Australia, pp. 218–281.

Belperio

Harvey

Bourman

(2002) Spatial and temporal variability in the Holocene sea-level record of the South Australian coastline. Sedimentary Geology 150: 153–169.

Billy

Robin

Certain

et al. (2013) Barrier shoreline evolution constrained by shoreface sediment reservoir and substrate control: The Miquelon-Langlade Barrier, NW Atlantic. Journal of Coastal Research 2: 2089–2094.

Billy

Robin

Hein

et al. (2018) Impact of relative sea-level changes since the last deglaciation on the formation of a composite paraglacial barrier. Marine Geology 400: 76–93.

Boreen

James

Wilson

et al. (1993) Surficial cool-water carbonate sediments on the Otway continental margin, southeastern Australia. Marine Geology 112: 35–56.

Bourman

Murray-Wallace

(1991) Holocene evolution of a sand spit at the mouth of a large river system: Sir Richard Peninsula and the Murray Mouth, South Australia. Zeitschrift Für Geomorphologie Supplementband 81: 63–83.

Boutakoff

(1963) The Geology and Geomorphology of the Portland Area (Victoria Geological Survey, Memoir 22). Melbourne: Geological Survey of Victoria, 172 pp.

Bowler

Johnston

Olley

et al. (2003) New ages for human occupation and climatic change at Lake Mungo, Australia. Nature 421: 837–840.

Brill

Jankaew

Brückner

(2015) Holocene evolution of Phra Thong’s beach-ridge plain (Thailand) – Chronology, processes and driving factors. Geomorphology 245: 117–134.

10.

Bristow

Pucillo

(2006) Quantifying rates of coastal progradation from sediment volume using GPR and OSL: The Holocene fill of Guichen Bay, south-east South Australia. Sedimentology 53: 769–788.

11.

Brooke

(2001) The distribution of carbonate eolianite. Earth-Science Reviews 55: 135–164.

12.

Brooke

Ryan

Pietsch

et al. (2008) Influence of climate fluctuations and changes in catchment land use on Late-Holocene and modern beach-ridge sedimentation on a tropical macrotidal coast: Keppel Bay, Queensland, Australia. Marine Geology 251: 195–208.

13.

Brooke

Huang

Nicholas

et al. (2019) Relative sea-level records preserved in Holocene beach-ridge strandplains – An example from tropical northeastern Australia. Marine Geology 411: 107–118.

14.

Cann

De Deckker

Murray-Wallace

(1991) Coastal Aboriginal shell middens and their palaeoenvironmental significance, Robe Range, South Australia. Transactions of the Royal Society of South Australia 115: 161–175.

15.

Cann

Murray-Wallace

Belperio

et al. (1999) Evolution of Holocene coastal environments near Robe, southeastern South Australia. Quaternary International 56: 81–97.

16.

Carvalho

Oliver

TSN

Woodroffe

et al. (2019) Transition from marine to fluvial-dominated sediment supply at Shoalhaven prograded barrier, southeastern Australia. Geomorphology 341: 65–78.

17.

Chappell

(1987) Late Quaternary sea-level changes in the Australian region. In: Tooley

Shennan

(eds) Sea-Level Changes. Oxford: Basil Blackwell Ltd, pp. 296–331.

18.

Choi

Kim

(2014) Reconstruction of Holocene coastal progradation on the east coast of Korea based on OSL dating and GPR surveys of beach-foredune ridges. The Holocene 24: 24–34.

19.

Ciarletta

Shawler

Tenebruso

et al. (2019) Reconstructing coastal sediment budgets from beach- and foredune- ridge morphology: A coupled field and modeling approach. Journal of Geophysical Research: Earth Surface 124: 1398–1416.

20.

Costas

Ferreira

Plomaritis

et al. (2016) Coastal barrier stratigraphy for Holocene high-resolution sea-level reconstruction. Scientific Reports 6: 38726.

21.

Costas

Jerez

Trigo

et al. (2012) Sand invasion along the Portuguese coast forced by westerly shifts during cold climate events. Quaternary Science Reviews 42: 15–28.

22.

Davies

(1964) A morphogenetic approach to world shorelines. Zeitschrift Für Geomorphologie 8: 127–142.

23.

Davies

(1972) Geographical Variation in Coastal Development. Edinburgh: Oliver and Boyd.

24.

Davies

(1974) The coastal sediment compartment. Australian Geographical Studies 12: 139–151.

25.

Dougherty

(2014) Extracting a record of Holocene storm erosion and deposition preserved in the morphostratigraphy of a prograded coastal barrier. Continental Shelf Research 86: 116–131.

26.

Dougherty

(2018a) Prograded coastal barriers provide paleoenvironmental records of storms and sea level during late Quaternary highstands. Journal of Quaternary Science 33: 501–517.

27.

Dougherty

(2018b) Punctuated transgression (?): Comment on Oliver, T.S.N., Donaldson, P., Sharples, C., Roach, M., and Woodroffe, C.D. ‘Punctuated progradation of the Seven Mile Beach Holocene barrier system, southeastern Tasmania’. Marine Geology 405: 120–130.

28.

Dougherty

Choi

Turney

CSM

et al. (2019) Technical note: Optimizing the utility of combined GPR, OSL, and LiDAR (GOaL) to extract paleoenvironmental records and decipher shoreline evolution. Climate of the past Discussions 15: 389–404.

29.

Engel

May

Scheffers

et al. (2015) Prograded foredunes of Western Australia’s macro-tidal coast-implications for Holocene sea-level change and high-energy wave impacts. Earth Surface Processes and Landforms 40: 726–740.

30.

Forsyth

Nott

Bateman

(2010) Beach ridge plain evidence of a variable late-Holocene tropical cyclone climate, North Queensland, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 297: 707–716.

31.

Forsyth

Nott

Bateman

et al. (2012) Juxtaposed beach ridges and foredunes within a ridge plain – Wonga Beach, northeast Australia. Marine Geology 307–310: 111–116.

32.

Fruergaard

Andersen

Nielsen

et al. (2015a) High-resolution reconstruction of a coastal barrier system: Impact of Holocene sea-level change. Sedimentology 62: 928–969.

33.

Fruergaard

Møller

Johannessen

et al. (2015b) Stratigraphy, evolution, and controls of a Holocene transgressive – Regressive barrier island under changing sea level: Danish North sea coast. Journal of Sedimentary Research 85: 820–844.

34.

Galbraith

Roberts

Laslett

et al. (1999) Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part I, experimental design and statistical models. Archaeometry 41: 339–364.

35.

Gontz

Moss

Wagenknecht

(2014) Stratigraphic architecture of a regressive strand plain, Flinders Beach, North Stradbroke Island, Queensland, Australia. Journal of Coastal Research 295: 575–585.

36.

Goslin

Clemmensen

(2017) Proxy records of Holocene storm events in coastal barrier systems: Storm-wave induced markers. Quaternary Science Reviews 174: 80–119.

37.

Guedes

CCF

Giannini

PCF

Sawakuchi

et al. (2011) Determination of controls on Holocene barrier progradation through application of OSL dating: The Ilha Comprida Barrier example, Southeastern Brazil. Marine Geology 285: 1–16.

38.

Guérin

Mercier

Adamiec

(2011) Dose-rate conversion factors: Update. Ancient TL 29: 5–8.

39.

Harvey

(2006) Holocene coastal evolution: Barriers, beach ridges, and tidal flats of South Australia. Journal of Coastal Research 221: 90–99.

40.

Hede

Sander

Clemmensen

et al. (2015) Changes in Holocene relative sea-level and coastal morphology: A study of a raised beach ridge system on Samso, southwest Scandinavia. The Holocene 25: 1402–1414.

41.

Hein

Fitzgerald

Cleary

et al. (2013) Evidence for a transgressive barrier within a regressive strandplain system: Implications for complex coastal response to environmental change. Sedimentology 60: 469–502.

42.

Hein

Fitzgerald

De Souza

LHP

et al. (2016) Complex coastal change in response to autogenic basin infilling: An example from a sub-tropical Holocene strandplain. Sedimentology 63: 1362–1395.

43.

Hein

FitzGerald

Thadeu

Menezes

et al. (2014) Coastal response to late-stage transgression and sea-level highstand. Geological Society of America Bulletin 126: 459–480.

44.

Hossfeld

(1952) The late Cainozoic history of the south-east of South Australia. Transactions of the Royal Society of South Australia 73: 232–279.

45.

Huntley

Prescott

(2001) Improved methodology and new thermoluminescence ages for the dune sequence in south-east South Australia. Quaternary Science Reviews 20: 687–699.

46.

Huntley

Hutton

Prescott

(1993) The stranded beach-dune sequence of south-east South Australia: A test of thermoluminescence dating, 0–800 ka. Quaternary Science Reviews 12: 1–20.

47.

Huntley

Hutton

Prescott

(1994) Further thermoluminescence dates from the dune sequence in the south-east of South Australia. Quaternary Science Reviews 13: 201–207.

48.

Jacobs

(2008) Luminescence chronologies for coastal and marine sediments. Boreas 37: 508–535.

49.

Jahnert

de Paula

Collins

et al. (2012) Evolution of a coquina barrier in Shark Bay, Australia by GPR imaging: Architecture of a Holocene reservoir analog. Sedimentary Geology 281: 59–74.

50.

James

Bone

(2011) Neritic Carbonate Sediments in a Temperate Realm – Southern Australia. Dordrecht: Springer, 254 pp.

51.

James

Bone

Borch

et al. (1992) Modern carbonate and terrigenous clastic sediments on a cool water, high energy, mid-latitude shelf: Lacepede, southern Australia. Sedimentology 39: 877–903.

52.

Jol

Bristow

(2003) GPR in sediments: Advice on data collection, basic processing and interpretation, a good practice guide. Geological Society, London (Special Publications) 211: 9–27.

53.

Kelley

Barber

Belknap

et al. (2005) Sand budgets at geological, historical and contemporary time scales for a developed beach system, Saco Bay, Maine, USA. Marine Geology 214: 117–142.

54.

Lambeck

Nakada

(1990) Late Pleistocene and Holocene sea-level change along the Australian coast. Palaeogeography, Palaeoclimatology, Palaeoecology 89: 143–176.

55.

Lewis

Sloss

Murray-Wallace

et al. (2013) Post-glacial sea-level changes around the Australian margin: A review. Quaternary Science Reviews 74: 115–138.

56.

McBride

Anderson

Buynevich

et al. (2013) Morphodynamics of barrier systems: A synthesis. In: Shroder

(ed.) Treatise on Geomorphology, vol. 10. Cambridge, MA: Academic Press, pp. 166–244.

57.

Madsen

Murray

Andersen

(2007) Optical dating of dune ridges on Rømø, a barrier Island in the Wadden Sea, Denmark. Journal of Coastal Research 235: 1259–1269.

58.

Moore

Duran Vinent

Ruggiero

(2016) Vegetation control allows autocyclic formation of multiple dunes on prograding coasts. Geology 44: 559–562.

59.

Moulton

MAB

Hesp

Miot da Silva

et al. (2019) Changes in vegetation cover on the Younghusband Peninsula transgressive dunefields (Australia) 1949–2017. Earth Surface Processes and Landforms 44: 459–470.

60.

Murray

Wintle

(2000) Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32: 57–73.

61.

Murray-Wallace

(2018) Quaternary History of the Coorong Coastal Plain, Southern Australia: An Archive of Environmental and Global Sea-Level Changes. Cham: Springer International Publishing, 229 pp.

62.

Murray-Wallace

Banerjee

Bourman

et al. (2002) Optically stimulated luminescence dating of Holocene relict foredunes, Guichen Bay, South Australia. Quaternary Science Reviews 21: 1077–1086.

63.

Murray-Wallace

Belperio

Bourman

et al. (1999) Facies architecture of a last interglacial barrier – A model for Quaternary barrier development from the Coorong to Mount Gambier Coastal Plain, Southeastern Australia. Marine Geology 158: 177–195.

64.

Murray-Wallace

Bourman

Prescott

et al. (2010) Aminostratigraphy and thermoluminescence dating of coastal aeolianites and the later Quaternary history of a failed delta: The River Murray mouth region, South Australia. Quaternary Geochronology 5: 28–49.

65.

Murray-Wallace

Brooke

Cann

et al. (2001) Whole-rock aminostratigraphy of the Coorong Coastal Plain, South Australia: Towards a 1 million year record of sea-level highstands. Journal of the Geological Society, London 158: 111–124.

66.

Nathan

Mauz

(2008) On the dose-rate estimate of carbonate-rich sediments for trapped charge dating. Radiation Measurements 43: 14–25.

67.

Neal

(2004) Ground-penetrating radar and its use in sedimentology: Principles, problems and progress. Earth-Science Reviews 66: 261–330.

68.

Nooren

Hoek

Winkels

et al. (2017) The Usumacinta-Grijalva beach-ridge plain in southern Mexico: A high-resolution archive of river discharge and precipitation. Earth Surface Dynamics 5: 529–556.

69.

Nott

(2011) A 6000 year tropical cyclone record from Western Australia. Quaternary Science Reviews 30(5–6): 713–722.

70.

Nott

Forsyth

(2012) Punctuated global tropical cyclone activity over the past 5,000 years. Geophysical Research Letters 39: 14703.

71.

Nott

Forsyth

Rhodes

et al. (2015) The origin of centennial-to millennial-scale chronological gaps in storm emplaced beach ridge plains. Marine Geology 367: 83–93.

72.

Oliver

TSN

Woodroffe

(2016) Chronology, Morphology and GPR-imaged Internal Structure of the Callala Beach Prograded Barrier in Southeastern Australia. Journal of Coastal Research SI 75: 318–322.

73.

Oliver

TSN

Thom

Woodroffe

(2017a) Formation of beach-ridge plains: An appreciation of the contribution by Jack L. Davies. Geographical Research 55: 305–320.

74.

Oliver

TSN

Donaldson

Roach

et al. (2018a) Reply to: Punctuated Transgression (?): Comment on Oliver, T.S.N., Donaldson, P., Sharples, C., Roach, M., and Woodroffe, C.D. ‘Punctuated progradation of the Seven Mile Beach Holocene barrier system, southeastern Tasmania’. Marine Geology 405: 131–133.

75.

Oliver

TSN

Donaldson

Sharples

et al. (2017b) Punctuated progradation of the Seven Mile Beach Holocene barrier system, southeastern Tasmania. Marine Geology 386: 76–87.

76.

Oliver

TSN

Dougherty

Gliganic

et al. (2015) Towards more robust chronologies of coastal progradation: Optically stimulated luminescence ages for the coastal plain at Moruya, south-eastern Australia. The Holocene 25: 536–546.

77.

Oliver

TSN

Kennedy

Tamura

et al. (2018b) Interglacial-glacial climatic signatures preserved in a regressive coastal barrier, southeastern Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 501: 124–135.

78.

Oliver

TSN

Tamura

Hudson

et al. (2017c) Integrating millennial and interdecadal shoreline changes: Morpho-sedimentary investigation of two prograded barriers in southeastern Australia. Geomorphology 288: 129–147.

79.

Oliver

TSN

Tamura

Short

et al. (2019) The morphodynamic significance of rapid shoreline progradation followed by vertical foredune building at Pedro Beach, southeastern Australia. Earth Surface Processes and Landforms 44: 655–666.

80.

Otvos

(2000) Beach ridges – Definitions and significance. Geomorphology 32: 83–108.

81.

Otvos

(2018) Coastal barriers, northern Gulf – Last Eustatic Cycle; genetic categories and development contrasts. A review. Quaternary Science Reviews 193: 212–243.

82.

Prescott

Hutton

(1994) Cosmic ray contributions to dose rates for luminescence and ESR dating: Large depths and long-term time variations. Radiation Measurements 23: 497–500.

83.

Reimann

Naumann

Tsukamoto

et al. (2010) Luminescence dating of coastal sediments from the Baltic Sea coastal barrier-spit Darss–Zingst, NE Germany. Geomorphology 122: 264–273.

84.

Reimann

Tsukamoto

Harff

et al. (2011) Reconstruction of Holocene coastal foredune progradation using luminescence dating – An example from the Świna barrier (southern Baltic Sea, NW Poland). Geomorphology 132: 1–16.

85.

Rémillard

Buylaert

Murray

et al. (2015) Quartz OSL dating of late-Holocene beach ridges from the Magdalen Islands (Quebec, Canada). Quaternary Geochronology 30: 264–269.

86.

Rink

López

(2010) OSL-based lateral progradation and aeolian sediment accumulation rates for the Apalachicola Barrier Island Complex, North Gulf of Mexico, Florida. Geomorphology 123: 330–342.

87.

Roy

Thom

(1981) Late Quaternary marine deposition in New South Wales and southern Queensland – An evolutionary model. Journal of the Geological Society of Australia 28: 471–489.

88.

Sander

Pejrup

Murray

et al. (2018) Chronology and late-Holocene evolution of Caleta de los Loros, NE Patagonia, Argentina. The Holocene 28: 1276–1287.

89.

Scheffers

Engel

Scheffers

et al. (2011) Beach ridge systems – Archives for Holocene coastal events? Progress in Physical Geography 36: 5–37.

90.

Searle

Woods

(1986) Detailed documentation of a Holocene sea-level record in the Perth region, southern Western Australia. Quaternary Research 308: 299–308.

91.

Searle

Semeniiuk

Woods

(1988) Geomorphology, stratigraphy and Holocene history of the Rockingham-Becher Plain, southwestern Australia. Journal of the Royal Society of Western Australia 70: 89–109.

92.

Short

(1988) The South Australia coast and Holocene sea-level transgression. Geographical Review 78: 119–136.

93.

Short

(2001) Beaches of the Southern Australian Coast and Kangaroo Island. Sydney: Australian Beach Safety and Management Project.

94.

Short

(2010) Sediment transport around Australia – Sources, mechanisms, rates and barrier forms. Journal of Coastal Research 26: 395–402.

95.

Short

Hesp

(1982) Wave, beach and dune interactions in southeastern Australia. Marine Geology 48: 259–284.

96.

Short

Hesp

(1986) Beach and Dune Morphodynamics of the South East Coast of South Australia. Coastal Studies Unit Technical Report 84/1. Sydney: Department of Geography, The University of Sydney.

97.

Sloss

Murray-Wallace

Jones

(2007) Holocene sea-level change on the southeast coast of Australia: A review. The Holocene 17: 999–1014.

98.

Sprigg

(1952) The geology of the south-east province, South Australia, with special reference to Quaternary coastline migrations and modern beach developments. Geological Survey of South Australia, Bulletin 29: 117–120.

99.

Sprigg

(1979) Stranded and submerged sea-beach systems of southeast South Australia and the aeolian desert cycle. Sedimentary Geology 22: 53–96.

100.

Tamura

(2012) Beach ridges and prograded beach deposits as palaeoenvironment records. Earth-Science Reviews 114: 279–297.

101.

Tamura

Cunningham

Oliver

TSN

(2019) Two-dimensional chronostratigraphic modelling of OSL ages from recent beach-ridge deposits, SE Australia. Quaternary Geochronology 49: 39–44.

102.

Tamura

Murakami

Nanayama

et al. (2008) Ground-penetrating radar profiles of Holocene raised-beach deposits in the Kujukuri strand plain, Pacific coast of eastern Japan. Marine Geology 248: 11–27.

103.

Tamura

Nicholas

Oliver

TSN

et al. (2018) Coarse-sand beach ridges at Cowley Beach, north-eastern Australia: Their formative processes and potential as records of tropical cyclone history. Sedimentology 65: 721–744.

104.

Tamura

Saito

Bateman

et al. (2012) Luminescence dating of beach ridges for characterizing multi-decadal to centennial deltaic shoreline changes during Late-Holocene, Mekong River delta. Marine Geology 326–328: 140–153.

105.

Tanner

(1995) Origin of beach ridges and swales. Marine Geology 129: 149–161.

106.

Taylor

Stone

(1996) Beach-ridges: A review. Journal of Coastal Research 12: 612–621.

107.

Thom

Bowman

Gillespie

et al. (1981) Radiocarbon Dating of Holocene Beach-Ridge Sequences in South-East Australia. Duntroon: Department of Geography, University of New South Wales at Royal Military College, 36 pp.

108.

Thom

Eliot

et al. (2018) National sediment compartment framework for Australian coastal management. Ocean & Coastal Management 154: 103–120.

109.

Timmons

Rodriguez

Mattheus

et al. (2010) Transition of a regressive to a transgressive barrier island due to back-barrier erosion, increased storminess, and low sediment supply: Bogue Banks, North Carolina, USA. Marine Geology 278: 100–114.