ArticlePDF Available

Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in northwest Australia

Authors:

Abstract and Figures

This article reports Australia’s first confirmed ancient underwater archaeological sites from the continental shelf, located off the Murujuga coastline in north-western Australia. Details on two underwater sites are reported: Cape Bruguieres, comprising > 260 recorded lithic artefacts at depths down to −2.4 m below sea level, and Flying Foam Passage where the find spot is associated with a submerged freshwater spring at −14 m. The sites were discovered through a purposeful research strategy designed to identify underwater targets, using an iterative process incorporating a variety of aerial and underwater remote sensing techniques and diver investigation within a predictive framework to map the submerged landscape within a depth range of 0–20 m. The condition and context of the lithic artefacts are analysed in order to unravel their depositional and taphonomic history and to corroborate their in situ position on a pre-inundation land surface, taking account of known geomorphological and climatic processes including cyclone activity that could have caused displacement and transportation from adjacent coasts. Geomorphological data and radiometric dates establish the chronological limits of the sites and demonstrate that they cannot be later than 7000 cal BP and 8500 cal BP respectively, based on the dates when they were finally submerged by sea-level rise. Comparison of underwater and onshore lithic assemblages shows differences that are consistent with this chronological interpretation. This article sets a foundation for the research strategies and technologies needed to identify archaeological targets at greater depth on the Australian continental shelf and elsewhere, building on the results presented. Emphasis is also placed on the need for legislation to better protect and manage underwater cultural heritage on the 2 million square kilometres of drowned landscapes that were once available for occupation in Australia, and where a major part of its human history must lie waiting to be discovered.
Content may be subject to copyright.
RESEARCH ARTICLE
Aboriginal artefacts on the continental shelf
reveal ancient drowned cultural landscapes in
northwest Australia
Jonathan BenjaminID
1,2
*, Michael O’Leary
3
, Jo McDonald
4
, Chelsea Wiseman
1
,
John McCarthy
1
, Emma BeckettID
4
, Patrick Morrison
4
, Francis Stankiewicz
1
,
Jerem Leach
1
, Jorg Hacker
1,5
, Paul Baggaley
1,6
, Katarina JerbićID
1
, Madeline FowlerID
1,7
,
John Fairweather
4
, Peter Jeffries
8
, Sean Ulm
2,9
, Geoff Bailey
1,10
1College of Humanities, Arts and Social Sciences, Flinders University, Adelaide, Australia, 2ARC Centre of
Excellence for Australian Biodiversity and Heritage, James Cook University, Cairns, Australia, 3School of
Earth Sciences, University of Western Australia, Perth, Australia, 4Centre for Rock Art Research +
Management, University of Western Australia, Perth, Australia, 5ARA—Airborne Research Australia,
Salisbury South, Australia, 6Wessex Archaeology, Portway House, Old Sarum Park, Salisbury, England,
United Kingdom, 7Archaeology, School of Humanities, University of Southampton, University Road,
Southampton, United Kingdom, 8Murujuga Aboriginal Corporation, Karratha, Australia, 9College of Arts,
Society and Education, James Cook University, Cairns, Australia, 10 Department of Archaeology, University
of York, The King’s Manor, York, England, United Kingdom
*jonathan.benjamin@flinders.edu.au
Abstract
This article reports Australia’s first confirmed ancient underwater archaeological sites from
the continental shelf, located off the Murujuga coastline in north-western Australia. Details
on two underwater sites are reported: Cape Bruguieres, comprising >260 recorded lithic
artefacts at depths down to 2.4 m below sea level, and Flying Foam Passage where the
find spot is associated with a submerged freshwater spring at 14 m. The sites were discov-
ered through a purposeful research strategy designed to identify underwater targets, using
an iterative process incorporating a variety of aerial and underwater remote sensing tech-
niques and diver investigation within a predictive framework to map the submerged land-
scape within a depth range of 0–20 m. The condition and context of the lithic artefacts are
analysed in order to unravel their depositional and taphonomic history and to corroborate
their in situ position on a pre-inundation land surface, taking account of known geomorpho-
logical and climatic processes including cyclone activity that could have caused displace-
ment and transportation from adjacent coasts. Geomorphological data and radiometric
dates establish the chronological limits of the sites and demonstrate that they cannot be
later than 7000 cal BP and 8500 cal BP respectively, based on the dates when they were
finally submerged by sea-level rise. Comparison of underwater and onshore lithic assem-
blages shows differences that are consistent with this chronological interpretation. This arti-
cle sets a foundation for the research strategies and technologies needed to identify
archaeological targets at greater depth on the Australian continental shelf and elsewhere,
building on the results presented. Emphasis is also placed on the need for legislation to bet-
ter protect and manage underwater cultural heritage on the 2 million square kilometres of
PLOS ONE
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 1 / 31
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Benjamin J, O’Leary M, McDonald J,
Wiseman C, McCarthy J, Beckett E, et al. (2020)
Aboriginal artefacts on the continental shelf reveal
ancient drowned cultural landscapes in northwest
Australia. PLoS ONE 15(7): e0233912. https://doi.
org/10.1371/journal.pone.0233912
Editor: Michael D. Petraglia, Max Planck Institute
for the Science of Human History, GERMANY
Received: February 19, 2020
Accepted: May 14, 2020
Published: July 1, 2020
Copyright: ©2020 Benjamin et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
files.
Funding: The Deep History of Sea Country project
team (all authors) were supported by the Australian
Research Council’s Discovery Projects funding
scheme (DP170100812), with supplementary
support from the Murujuga: Dynamics of the
Dreaming Project (LP140100393), Flinders
University and the Hackett Foundation of Adelaide
and ARC Centre of Excellence for Australian
drowned landscapes that were once available for occupation in Australia, and where a
major part of its human history must lie waiting to be discovered.
Introduction
An estimated 20 million km
2
of territory was exposed on the world’s continental shelves dur-
ing the Last Glacial Period (c. 110,000–10,000 cal BP), 2 million km
2
around the Australian
continent alone, increasing its landmass by a third [1,2]. During this period some of the major
transformations of early human history took place, including renewed human dispersals out of
Africa into Europe and Asia, development of seafaring technology, palaeoeconomic diversifi-
cation and intensification including exploitation of marine resources, and entry for the first
time into Australia and the Americas, currently dated at c. 65,000 and c. 20,000 cal BP respec-
tively [39]. These previously exposed territories on the continental shelf likely harboured
favourable environments for hunter-gatherer settlement and dispersal including abundant
water supplies, desirable microclimates, ecological diversity, and the additional potential for
exploitation of marine resources and seaborne travel along and around the inlets and archipel-
agos of its palaeocoastlines [1012]. These conditions in their turn would have created the
potential for relatively high population densities and concentrations of archaeological sites
compared to the more arid hinterlands that prevailed in many regions of the world during the
Last Glacial Period.
Relatively little detail is known about these now-submerged landscapes, their human occu-
pants, their role in patterns of human dispersal and development, or the human impact of a
sustained sea-level rise from a Last Glacial Maximum (LGM) low of 130 m at c. 20,000 cal BP
to reach the present level at c. 7000 cal BP [13].
New investigations are now under way in many parts of the world to explore the role of the
coastal zone in population dispersal, to reconstruct these submerged landscapes and their
palaeocoastlines and palaeoenvironments, and to test their archaeological potential [1,1429].
However, systematic recovery and investigation of underwater archaeological sites, which is
crucial to the evaluation of new hypotheses, is inhibited by powerful and ongoing constraints.
These include the unobtrusive or ephemeral nature of hunter-gatherer material culture, lim-
ited knowledge of the taphonomic conditions under which sites preserve and are accessible to
discovery, and widespread uncertainties that anything remains to be discovered underwater or
that this will be sufficiently intact to make a useful and decisive contribution to new knowl-
edge. There are additional hurdles associated with matters of safety and regulatory compliance
as well as consideration of costs of working in a marine environment, which can vary consider-
ably depending on location and conditions.
The potential for submerged sites on Australia’s continental shelves has long been recog-
nized [3032], but the relatively few attempts made to locate such sites have been unsuccessful
[33,34]. In several areas stone artefact sites and quarries, and stone-walled fish traps, have
been documented in intertidal zones [3538]. However, many of these appear to be extensions
of land-based sites and activities related to the present-day shoreline, and it has proved difficult
to demonstrate that any were features of a pre-inundation landscape occupied at a time of
lower sea level. Underwater artefacts recovered from inland freshwater lakes further under-
score the potential for survival of submerged sites on the continental shelf [39].
The new results presented here demonstrate the existence of in situ archaeological material
on pre-inundation land surfaces associated with periods of lower sea level in Murujuga Sea
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 2 / 31
Biodiversity and Heritage (CE170100015). https://
www.arc.gov.au/. The funders had no role in study
design, data collection and analysis, decision to
publish, or preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
Country (Dampier Archipelago, Western Australia), resulting from a program of predictive
modelling, systematic survey and geoarchaeological analysis, using a suite of remote sensing
techniques including satellite imagery, acoustic survey (sidescan and multibeam), airborne
topographic and bathymetric LiDAR, and in situ diver investigation.
The aims in this article are: to articulate the rationale and research strategy that has
informed the discovery of the underwater sites in this case study region as an exercise in sub-
merged landscape archaeology; to describe the techniques, equipment and sampling proce-
dures used in underwater mapping, site discovery and analysis of finds; to analyse the results
with reference to comparative data from sites on land, with particular attention to the tapho-
nomic and depositional histories of the underwater artefacts and the systematic evaluation of
alternative hypotheses about whether or not they can be considered as evidence of human
activities on a pre-inundation land surface; and to consider the wider implications for archaeo-
logical investigations of the continental shelf in Australia and worldwide.
Regional setting
The Dampier Archipelago (Murujuga) is located in the semi-arid Pilbara region of northwest
Australia (Fig 1), and experiences low and variable rainfall averaging less than 350 mm per
annum. The archipelago is also situated in one of the most cyclone-prone regions in the world.
Thirty-six tropical cyclones crossed the Pilbara coast between 1980 and 2007, and a major
cyclone, Cyclone Veronica, passed over the area in March 2019 towards the end of the project’s
field campaigns. The potential for cyclone activity to cause disturbance, displacement or
destruction of archaeological material in coastal areas is a major factor that needs to be
Fig 1. Location maps of the study area and sites referenced in text. 1) Cape Bruguieres Island; (2) North GidleyIsland; (3) Flying Foam Passage; (4) Dolphin Island; (5)
Angel Island; (6) Legendre Island; (7) Malus Island; (8) Goodwyn Island; (9) Enderby Island.
https://doi.org/10.1371/journal.pone.0233912.g001
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 3 / 31
considered when assessing the integrity or otherwise of coastal archaeological sites and their
post-depositional history [40]. The timing of Cyclone Veronica provided an unusual opportu-
nity to compare the distribution and condition of archaeological materials before and after the
event and to assess the impact of cyclone activity.
The north–south oriented Mermaid Sound separates the archipelago into two island
groups. The eastern islands are an extension of the Burrup Peninsula and are formed from 2.7
billion-year-old rhyodacite (also known as granophyre) and gabbros. The western islands are
formed from similar-age basalts and andesites [41]. Pleistocene-age aeolianites and cemented
beach sediments of mid-to-late Holocene age (calcarenite) are also present around the islands’
coastal fringes and embayments. The former are cemented sand dunes accumulated during
earlier periods of high sea level (MIS 5 or earlier) and are characterised by their reddish colour;
the Holocene calcarenites are cemented beach deposits including beachrock formed in associa-
tion with the establishment of modern sea level and are creamy-white in colour.
The igneous geology has eroded into a rough and complex sheet-fractured nubbin terrain
with ridgelines of massive boulders (often unvegetated), and valleys which form a rectangular
drainage pattern. Freshwater is seasonally available in narrow ephemeral creeks filled by rain-
fall and in springs [41]. This geology provides abundant and ubiquitous material for manufac-
ture of stone tools and artificial stone structures [42]. It also affords numerous boulders and
fractured slabs with surfaces suitable for rock engravings, of which there are estimated to be c.
1 million for the Murujuga rock art province [43]. The slow weathering rates of this geology
[44] have created the ideal conditions for the preservation of a human artistic record that
could have survived as far back as the 50,000 cal BP that humans are known to have occupied
this part of the north-west coast [7].
At the Last Glacial Maximum, the coastline was located 160 km further offshore [45],
exposing a gently seaward-sloping coastal plain composed of marine carbonate and siliciclastic
sediments, dotted with springs and stream channels, and with fringing mangroves and swamps
along palaeocoastlines. Palaeochannels, stranded palaeoshorelines, carbonate reefs, and iso-
lated knolls would have created local relief on the exposed coastal plain. Sea-level rise after the
LGM (c. 18,000 cal BP) progressively drowned this landscape, reaching a mid-Holocene high-
stand of approximately +2 m (MSL) (2 m above modern Mean Sea Level) by 7000 cal BP and
subsequently regressing from approximately 5000 cal BP to present sea level (Fig 2; [46]).
Fig 2. (left) a composite coral growth history from Western Australia, figure modified from [47], blue line (in left insert) represents the minimum age of
inundation. (right) Red Sea deep sea oxygen isotope sea level record (blue line) with lower and upper 95% confidence limits (dotted blue line) data from[48].
https://doi.org/10.1371/journal.pone.0233912.g002
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 4 / 31
The characteristic archaeology of the region is dominated by open-air sites: especially
engraved rock art panels as noted above, but also includes: stone tool assemblages; quarries;
circular or curvilinear stone structures interpreted as hut foundations or terraces to enhance
trapping of sediment; standing stones of probable ceremonial significance; and shell middens,
sometimes forming shell mounds up to 5 m thick [42,49]. Age determinations fall predomi-
nantly within the Holocene (the past ten thousand years), but the sequence of rock art styles
includes extinct animals, demonstrating a longer history of occupation extending back into
the Pleistocene [42,43,45,49,50]. Rockshelters with stratified and dateable deposits are rare
in this type of geology, but one granite overhang on the Burrup Peninsula contains deposits
with evidence of occupation extending from 21,000 to 7000 cal BP [51]; while excavations of
limestone caves on the more distant Barrow Island have yielded a sequence between 50,000
and 8000 cal BP, confirming the Pleistocene time depth of human activity in the region [7].
The lithic raw materials and food remains found within these sites demonstrate their use as
bases for wide-ranging movements into the hinterland and out onto the coastal plain exposed
at lower sea level. As sea level rose and the shoreline moved progressively closer, stratified food
remains and changing rates of artefact discard show changing patterns of site use and move-
ment across the landscape, increased representation of marine foods, and ultimately abandon-
ment and a reconfiguration of land-use patterns adjusted to the modern coastline [51].
Similarly, rock art motifs show an increase in marine animals with progressive sea-level rise.
Research strategy and methods
The team deployed a suite of remote-sensing methods to map and interpret the landscape
through an iterative process conducted over a series of six field campaigns between 2017 and
2019 [28,52]. These were designed to identify submarine features of interest and specific tar-
gets for closer inspection, to locate submerged archaeological sites for diver inspection, and to
recover and analyse geological, geochronological and archaeological samples.
Each of the remote sensing methods and equipment described below has its strengths and
limitations, and methods were chosen because of their complementarity and their suitability
in combination to provide information at a variety of geographical scales and with varying
degrees of resolution and precision. Techniques applied ranged from mapping of topography
and bathymetry at a sub-regional scale to high-precision recording of the positions of individ-
ual stone artefacts on the seabed. By combining different methods in this way over a series of
field campaigns, comparing their results and adjusting subsequent surveys accordingly, the
team established a picture of the submerged landscape and identified prospective targets for
closer investigation.
This iterative process was designed to take into account five variables: (1) locations likely to
have been attractive to the original inhabitants because of proximity to resources such as water
supplies and raw materials for stone-artefact manufacture; (2) locations likely to have pre-
served archaeological materials because of topographic features such as peninsulas and semi-
enclosed basins providing protection from destructive wave action and ocean currents, or rock
overhangs and cliff lines affording concentration and preservation of accumulated sediments;
(3) locations where material was not only likely to have survived sea-level rise but would be
sufficiently exposed to be discovered; (4) local knowledge of community members including
Traditional Owners and fishermen; (5) accessibility for diver investigation.
Because of limitations on the water depth in which some of these techniques can be applied
and their logistical requirements, and taking as a starting point the principle of working from
the known (the present-day land surface and its archaeology) to the unknown (the submerged
landscape), focus is placed on shallow-water conditions (down to depths of c. 20 m) as a first
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 5 / 31
step into the unknown, and travel distances offshore within relatively easy reach of small sup-
port vessels and modern harbour facilities. Investigations at greater depth and further offshore
and the search for evidence buried beneath marine sediments pose different challenges and
require different technologies and equipment, larger support vessels and different principles of
research design and method, a point that is considered further in the final discussion. Fuller
details and the results obtained by remotely sensed mapping are presented elsewhere [28] or
are in preparation.
All necessary permits were obtained for the described study, which complied with all rele-
vant regulations. The project was conducted under Flinders University ethics approval
SBREC7669 and with the approval of the Murujuga Aboriginal Corporation by Circle of Elders
vote (19th January 2017). Mapping and sampling was conducted under a Western Australia
Department of Parks and Wildlife permit under Regulation 4(1) (5th May 2019). Permission
to undertake further analysis of the artefacts was granted by the Murujuga Aboriginal Corpo-
ration. Sampled cultural material was repatriated and remains in the possession of the Tradi-
tional Owners (see supporting information). The specific techniques and methods which
resulted in the successful location of two submerged sites are described in further detail below.
Airborne LiDAR survey
For onshore and offshore aerial survey and mapping of terrestrial and submarine surfaces at a
variety of scales, the team deployed a Diamond Aircraft HK36TTC-ECO Dimona motorglider
with two LiDAR systems mounted in under-wing pods: a Riegl Q680i-S (topographic) and a
Riegl VQ-820-G (topo-bathymetric), each combined with a tactical grade IMU/GPS system
(Novatel SPAN ISA/LCI). A Canon 5D Mk4 was fitted with an EF 24 mm (f/1.4LII USM) lens
and co-mounted with the Q680i-S. Point cloud density ranged between 10 and 20 points/m
2
,
and data was processed and converted to a Digital Elevation Model (DEM) using the Global
Mapper LiDAR module. Airborne mapping offers enormous flexibility in areal coverage and is
the only method that can produce a seamless continuum of images and measurements across
the interfaces between land, the intertidal zone and the adjacent seabed, including measure-
ments of relatively high precision in both the vertical and the horizontal dimension. Its limita-
tion is that it is confined to shallow water, in the study area region down to water depths of c.
12 m.
Marine survey
For more detailed examination of sea-bed surfaces and topographic irregularities and explora-
tion of deeper areas of the seabed, a total of 347 linear km covering approximately 150km
2
were surveyed with an EdgeTech 4125 sidescan sonar system, operated from an 8.5 m support
vessel. Survey areas were gridded using Hypack navigation software, with line spacing ranging
from 200 m as a minimum to 30 m for higher-resolution coverage of areas of particular inter-
est. Parallel transects were run over selected areas to ensure a systematic and comprehensive
coverage of the seabed. Sidescan instrument real-time locations and sonar mosaics were com-
pleted in the SonarWiz processing software. In certain areas where more systematic measure-
ments of seabed bathymetry were required, sidescan imagery was supplemented with
multibeam bathymetric data acquired from EGS Survey and Australian Marine Services.
Diver surveys
Once target features of potential significance were identified, the team deployed archaeologi-
cally trained divers for closer inspection and sample recovery, using standard safety protocols
for scientific diving. Generally, teams of two or four underwater archaeologists worked
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 6 / 31
together on a pre-determined dive plan during any given dive. The navigation system of the
dive support vessel was used to position pre-defined survey lines laid out in a GIS with refer-
ences to aerial and LiDAR basemaps. Survey lines were set using a 100 m leaded line attached to
a shot weight with marker buoys on either end. Dive teams carried a marker float with a Garmin
eTrex GPS to log location, documented any visible archaeological material and made geological
descriptions that included changes in seabed composition (Fig 3). Cameras and GPS were cali-
brated to enable georeferencing of all photographs. The team explored a number of potential
targets in this way before selecting for more detailed investigation the two sites examined here.
Artefact analysis
Underwater artefacts were examined in situ and recorded with as much information as possi-
ble without removal. However, many analyses could not be carried out except in laboratory
Fig 3. (above) Westward facing aerial view of Cape Bruguieres Channel at high tide (Photo: J. Leach); (below) divers record artefacts in the channel (Photos: S. Wright, J.
Benjamin, and M. Fowler).
https://doi.org/10.1371/journal.pone.0233912.g003
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 7 / 31
conditions and a sample of artefacts was removed for that purpose. Each artefact was given a
unique accession number, measured, and photographed. The morphological features recorded
were raw material type, colour and quality; maximum length, breadth and thickness (in cm);
weight (g); artefact type including a range of features specific to flakes and cores; and presence
and nature of retouch. Since the amount of marine growth present often meant that character-
istic features were obscured, comments on each artefact included an assessment of whether it
was a definite, probable or possible artefact (see S1 Table). The nature of the marine growth
was also described. This included corals, sponges, bryozoans, tubeworms, foraminifera and
coralline algae, the presence and relative composition of which changed relative to depth of
submersion. Some of these marine growths were sampled for potential age determination (see
below). A selection of these artefacts was hand drawn and captured in 3D using a Sony
RX100iii and Agisoft Metashape (v 1.6). The team also applied neutron tomography to selected
lithics using the synchrotron at the ANSTO DINGO beam facility in Sydney [53] in order to
remove surface marine concretions digitally to reveal the shape of the artefact more clearly.
Geological sampling
For the analysis of artefact and local igneous rock geochemistry to identify the sources of raw
materials used in artefact manufacture, the team used a handheld Portable X-Ray fluorescence
(pXRF) device–a Niton XL3t GOLDD+; TestAll Geo, standards in SOIL; 99.995 SiO2 (pure sil-
ica standard) using the NIST 2709a soil and sediment standard. This facilitated measurement
in the field as well as in the laboratory. In the laboratory, a 5% v/v solution of HCl was used to
remove carbonate followed by cleaning with DI water to pXRF selected artefacts from the
Cape Bruguieres Channel. For dating and geological analysis of the various aeolianite and cal-
carenite substrates on which artefacts were located, samples were extracted by hand using a
hammer and chisel, or a drill core.
Radiocarbon (
14
C) dating of marine materials
Accelerator Mass Spectrometry (AMS) radiocarbon age determinations on marine shell and
coral embedded in aeolianite and calcarenite were undertaken at the University of Waikato
Radiocarbon Dating Laboratory and the Scottish Universities Environmental Research Centre
(SUERC) Radiocarbon Dating Laboratory. Surfaces of samples were cleaned, washed in an
ultrasonic bath, acid etched in HCl, rinsed and dried. Shells were tested for recrystallization by
Feigl staining [54]. CO
2
was collected and reduced to graphite. Pressed graphite was either
analysed at the Keck Radiocarbon Dating Laboratory, University of California [55] or at
SUERC [56]. Radiocarbon ages were calibrated using OxCal (version 4.3) [57]. Pre-modern
radiocarbon ages were calibrated using the Marine13 dataset [58], with a ΔR of 109±25 [7].
Modern radiocarbon ages (F
14
C%100) were approximately calibrated with reference to post-
AD 1950 F
14
C regional marine concentrations [59].
Aerial drone survey
Where aerial photography of higher resolution was required, low altitude drones were
deployed for mapping of surfaces and features on land and across the intertidal zone when
exposed at low tide. This proved especially valuable in mapping the position and distribution
of artefacts and even smaller items in order to assess the degree of disturbance caused by
Cyclone Veronica. A DJI Phantom 4 Pro and Mavic 2 were flown with automated flight plan-
ning software (Drone Deploy) and employed two survey strategies: single-line transects flown
between 75–20 ft above the ground level (AGL); and large-area surveys flown at 82 ft AGL
with a frontlap of 75% and a sidelap of 70% to produce a ground sample distance of 1 cm.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 8 / 31
Images were imported into Agisoft Metashape (v 1.5.4) to create point cloud data using set-
tings for Highest Accuracy, Ultra High Quality and Aggressive Filtering. Resulting dense
clouds were filtered to achieve 1 cm point spacing, resulting in data sets containing more than
500 million points. These, in turn, were cropped into five 25 x 25 m sample areas and imported
into CloudCompare (v2.11) to facilitate comparison between the two datasets. This approach
drastically reduced vertical distortion between the two datasets and allowed for effective quan-
titative comparison between drone runs conducted over the same surface before and after
Cyclone Veronica.
Results
Detailed underwater surveys and diver investigations identified two underwater locations in
the north of the archipelago assessed to have high archaeological potential, and these were tar-
geted for more detailed investigation. The first is at Cape Bruguieres Channel, where detailed
survey by divers identified 269 underwater lithic artefacts to a depth of 2.4 m (MSL) (i.e. 2.4
m below Mean Sea Level) on the relict Pleistocene aeolianite, which forms the channel floor
between Cape Bruguieres Island and North Gidley Island (Fig 4). The second find location is
in Flying Foam Passage, where a submerged freshwater spring was identified in sonar data at a
depth of 14 m (MSL) mid-channel between Dolphin and Angel Islands. Dive survey at this
location was limited by logistical considerations but one clearly recognisable lithic artefact was
identified and recovered.
Any interpretation of the significance of these two sites must consider them in the archaeo-
logical and geomorphological context of the wider landscape. Two other sites in the area are
referenced for comparative purposes, one a dry-land site of late Holocene date on a calcarenite
beach terrace on the southern shore of the Cape Bruguieres Channel, investigated as part of
the current project, the other a cluster of sites in the intertidal zone on the western shore of
Dolphin Island facing Flying Foam Passage, which have been previously reported (38).
Because the two new underwater sites are relatively close to these onshore sites (at distances
of 10s to 100s of metres), particular attention is paid to the taphonomic and depositional histo-
ries of the underwater artefacts and their geomorphic context in order to examine systemati-
cally a range of alternative hypotheses that might account for their present position. At one
extreme is the hypothesis that all the underwater artefacts are in situ in the locations where
they were originally deposited by human activities on a pre-inundation land surface that pre-
dates Holocene sea-level rise. At the other extreme is the hypothesis that the underwater arte-
facts are all in secondary position, having been eroded out from archaeological sites on land
on the adjacent shoreline and transported to their present positions by a variety of natural
agencies including the action of wind and waves, water currents and downslope movement
under the force of gravity.
Cape Bruguieres channel
This 2.5 km-long tidal channel separates Cape Bruguieres Island (to the north) and North Gid-
ley Island (to the south) and has a maximum width of 150 m and a maximum depth of 2.4 m
(MSL) (Fig 4). At its widest point is a U-bend with the outer (southern) shore of the bend bor-
dered by a calcarenite terrace (cemented beachrock). This terrace extends for approximately
600 m around the channel edge, is 80 m wide at its widest point, and is 2 to 2.3 m in elevation,
with a gentle seaward slope. The terrace is relatively flat and is positioned slightly above Mean
High Water Springs. Elsewhere the channel is flanked by outcrops of igneous rock or Pleisto-
cene aeolianite. Relict Pleistocene aeolianite also forms the channel floor, which is mantled by
a thin modern veneer of mobile sands.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 9 / 31
The calcarenite terrace is interpreted here as an ebb-tidal sand spit that began to form dur-
ing and post the mid-Holocene sea-level highstand interval. This interpretation is consistent
the terrace elevation and with a radiocarbon date of 2446±65 BP (1791–2141 cal BP; Wk-
49709, Table 1) obtained from a shell cemented into the upper surface of the terrace. On the
opposite bank of the U-bend, there is only a minor occurrence of this same calcarenite feature.
According to the tide gauge on Legendre Island, the average tidal range between Mean
High Water Springs (MHWS) and Mean Low Water Springs (MLWS) is 3.6 m, and Mean Sea
Level (MSL) is the mid-point between them. Hence MLWS is 1.8 m (MSL) and MHWS is
+1.8 m (MSL). The maximum tidal range, the range between the highest astronomical tide
(HAT) and the lowest astronomical tide (LAT), is +2.3 m (MSL) to 2.4 m (MSL). However,
Fig 4. Bathymetry, geology and geomorphology of the Cape Bruguieres Channel and surrounding landscape including part of Cape Bruguieres Island to the
north of the channel and North Gidley Island to the south. Note the extensive Holocene dune formations along the west coast of North Gidley Island, the calcarenite
terrace on the south bend of the channel, and the mid-channel sill. The highest elevations are on the granophyre bedrock and the Pleistocene aeolianite. Satellite image
by Sentinel in the public domain.
https://doi.org/10.1371/journal.pone.0233912.g004
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 10 / 31
because of local hydrodynamic effects, observed tidal range in the Cape Bruguieres Channel is
somewhat smaller; with MLWS at 1.1 m (MSL) and LAT at 1.4 m (MSL). This means that
the bed of the channel at 2.4 m is fully submerged at all tidal positions under present-day
conditions except on a sill across the middle of the Channel which, situated at a depth of 0 to
1.4 m (MSL), is partially exposed at low tides.
Radiocarbon-dated marine shells (SUERC 90025, 90026, 90031, Table 1) were collected
from a marine calcarenite that deposited on top of the aeoloanite on the channel floor. These
shells incorporated within the calcarenite are dated in the range 44,700 to 26,600 cal BP. This
is interpreted as a minimum date range for radiocarbon-dead marine shell, most likely of Last
Interglacial (MIS 5e) age, contaminated with modern carbon introduced by penetration of
endolithic marine boring organisms during the past 7000 years. Thus, the channel floor repre-
sents an antecedent surface depression in the landscape, which people could have directly
occupied at any time after first arrival on the north-west coastal plain. This landscape depres-
sion would have become an active tidal channel as soon as the land surface was inundated dur-
ing the mid-Holocene (i.e. after c. 7000 years ago).
Archaeological results. Underwater and onshore archaeological surveys have yielded two
distinct assemblages of worked stone artefacts. The first comprises 269 underwater artefacts
widely distributed across the channel, of which 190 are permanently submerged, with the
remaining 79 artefacts in the intertidal zone nearer the edge of the channel or on the mid-
channel sill (Figs 5and 6). The second assemblage comprises 455 artefacts and associated
stone structures on the surface of the calcarenite terrace. The artefacts both on land and under-
water are made of locally available igneous rocks, and probable quarry sources have been
Table 1. AMS radiocarbon determinations for the Bruguieres channel. Five samples on underwater artefacts returned modern ages, which is inconclusive since active
marine growth is present on most of the underwater artefacts. The three Pleistocene ages from the channel floor are on marine shells cemented in the rock surface, and are
in the range 44.7 to 26.6 ka, which are interpreted as minimum ages. The three modern ages from the channel floor result from dating modern crustose coralline algae,
coral or burrowing mollusc shells attached to the underlying bedrock.
Lab. No. Sample Method δ
13
C‰ % Modern Carbon (F
14
C
%)
Conventional
14
C Age (years
BP)
Calibrated Age BP
(95.4%)
Calibrated Age BP or
AD
Median
Wk-49709 marine shell 14C 1.7±0.4 73.8±0.60 2446±65 1791-2141 1960 cal BP
Wk-50093 coral AMS -1.38
±0.3
105.9±0.32 0±0 NA AD 1960s
Wk-50094 marine shell AMS -2.34
±0.3
104.5±0.33 0±0 NA AD 1960s
Wk-50095 coral AMS NA 101.4±0.31 0±0 NA AD 1960s
Wk-50096 marine shell AMS NA 101.0±0.31 0±0 NA AD 1960s
Wk-50097 coralline
algae
AMS -0.9±0.3 102.7±0.35 0±0 NA AD 1960s
SUERC-
90023
coralline
algae
AMS -0.5 104.4±0.32 0±0 NA AD 1960s
SUERC-
90024
coral AMS -3.3 106.6±0.33 0±0 NA AD 1960s
SUERC-
90025
oyster AMS 2.3 NA 26,676±75 30,220-30,804 30,508 cal BP
SUERC-
90026
marine shell AMS 3.0 NA 44,745±490 46,326-48,808 47,499 cal BP
SUERC-
90030
marine shell AMS 2.1 107.3±0.25 0±0 NA AD 1960s
SUERC-
90031
marine shell AMS 2.2 NA 31,605±117 34,665-35,281 34,959 cal BP
https://doi.org/10.1371/journal.pone.0233912.t001
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 11 / 31
identified to the north on the shoreline of Cape Bruguieres Island. The pXRF analysis shows
that almost all the Cape Bruguieres stone artefacts are made of rhyodacite, which forms the
main bedrock geology on both North Gidley and Bruguieres Islands. Andesite, commonly
found on nearby Lewis and Malus Islands [41,60] comprises 8% (n = 2) of the terrace assem-
blage. One of the submerged artefacts was made from a different raw material that has not yet
been identified.
Fig 5. Survey area and locations of underwater and intertidal stone artefacts recorded in Cape Bruguieres Channel. The base
photomosaic (above) was taken at Lowest Astronomical Tide (LAT) of 2019. Artefacts are labelled as subtidal (below MLWS) and
intertidal (above MLWS and below MHWS). Bathymetric data (below) reflects shoreline at MHWS.
https://doi.org/10.1371/journal.pone.0233912.g005
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 12 / 31
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 13 / 31
The artefacts in the underwater assemblage comprise a range of types including cores and
core tools, retouched flakes, mullers and two possible grinding stones. The onshore artefacts
on the calcarenite terrace include retouched and unretouched flakes and cores. They tend to
cluster around stone arrangements comprising cairns and curvilinear features composed of
slabs and fragments of beachrock. They must postdate the formation of the calcarenite and
therefore represent Aboriginal activities within the last 2000 years associated with the modern
shoreline and its resources, a point confirmed by their association with shell midden deposits
of Tegillarca granosa (syn. Anadara granosa) and Baler shells (Melo sp.). There is a more wide-
spread surface scatter of similar materials across the terrace and adjacent dunes, indicating
that this whole shore zone was a preferred living area and focus of subsistence activities.
Although there are general similarities in artefact types between the underwater and
onshore assemblages, there is also a clear size difference (Table 2). Statistical analysis on the
artefacts was performed in the R statistical language using the ‘stats’ and ‘tidyverse’ packages
[61,62]. Size independence testing was focussed on the 0-20cm size range, which represents
the vast majority of the Cape Bruguieres assemblage and allows comparison with the Dolphin
Island artefact data. A Pearson’s chi-squared test demonstrates the difference in size between
Cape Bruguieres assemblages is statistically significant (X
2
(df = 9, N = 654) = 337.49). The
simulated p-value calculated using 2000 replicates of Monte-Carlo random sampling meets the
very high threshold of <0.001. The effect size, calculated using Crame
´r’s V is 0.71, which dem-
onstrates this difference is strong (<0.5) and practically significant [63]. The same test was
done for differenced in artefact types, which also demonstrated a statistical significance (X
2
(df = 6, N = 351) = 72.27, simulated p<0.001) with a moderate to strong effect size (V = 0.45)
[63]. The submerged artefacts are more massive with more core tools and generally thicker
platforms and less evidence of knapping debitage, in contrast to the terrace assemblage, where
artefacts in the 2–6 cm size category (based on measured maximum length) dominate and the
larger sizes are far fewer (Figs 7and 8).
Taphonomic and depositional history. Given the close proximity of the underwater and
onshore artefact assemblages at Cape Bruguieres Channel, it is critical to address the relation-
ship between the two assemblages. Are they of different ages referring to two separate episodes
of activity separated by a gap of at least 5000 years, where the underwater assemblage refers to
activity in a shallow dry valley at a period of lower sea level, and the onshore assemblage refers
to activity on the modern shoreline? And if that is so, is the proximity of the two assemblages
simply a coincidence resulting from the vagaries of underwater site preservation, visibility and
discovery, or is it due to the relative attractiveness of this locality but for different reasons at
Fig 6. Bathymetric data at Cape Bruguieres Channel (shoreline reflects MSL) with the mid-channel sill clearly visible in the bathymetry. Photos and
drawings of a sample of stone tools and their locations on the map by artefact number. Artefacts can be categorised as: Core / Core Tool (A37, A23); Retouched
flakes (A11, A29, A44, A30); Flaked tool (A10, A20, A34); Hammerstone/Muller (A40). (Artefact Drawings: K. Jerbić).
https://doi.org/10.1371/journal.pone.0233912.g006
Table 2. Summary data of artefact size categories showing the comparison between the submerged Cape Bruguieres channel (underwater, including intertidal)
assemblage where artefact size was confidently recorded (n = 210; note: size determinations were not possible for the remainder of the assemblage which was left in
situ and where adequately scaled photos were not recorded) and the Cape Bruguieres terrace (onshore) assemblage (n = 445). Of the underwater artefacts, 45 were
collected and measured, with the remainder photographed in situ with a scale card.
0–2 cm 2–4 cm 4–6 cm 6–8 cm 8–10 cm 10–12
cm
12–14
cm
14–16
cm
16–18
cm
18–20
cm
20–22
cm
22–24
cm
40–42
cm
42–44
cm
Bruguieres
Channel
0 18 30 44 52 37 13 7 3 1 1 2 1 1
Bruguieres
Terrace
25 228 154 34 7 1 0 0 0 0 0 0 0 0
https://doi.org/10.1371/journal.pone.0233912.t002
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 14 / 31
different sea-level positions? Alternatively, are the assemblages of similar age that refer to
essentially the same episode of activity, where the underwater material has been eroded out
from the calcarenite terrace by wind and wave disturbance and displaced and re-distributed
under water by wave action and water currents?
To answer these questions, it is necessary to identify and evaluate three alternative hypothe-
ses regarding the depositional history of the underwater assemblage:
a. It is in situ and therefore significantly older than the onshore assemblage and associated
with the channel area when it was a terrestrial land surface;
b. It is a re-deposited collection of artefacts that have been transported by natural agencies
from the archaeological site on the adjacent shoreline and is therefore of similar age and
culture to the onshore assemblage;
c. It is a lag deposit representing heavier material originally deposited on a Holocene dune fea-
ture that filled the channel in the final stages of sea-level rise, followed by erosion of the
finer sediment fraction by tidal action, leaving in place the heavier materials, primarily the
Fig 7. Residual plot for the chi-squared test of artefact sizes from the Cape Bruguieres platform (land) and channel
(submerged) assemblages, using the ggcorrplot [64] package in R. Blue circles indicate an over-representation, and red circles
indicate an underrepresentation.
https://doi.org/10.1371/journal.pone.0233912.g007
Fig 8. Residual plot for the chi-squared test of artefact types for the Cape Bruguieres artefacts, using the ggcorrplot package [64] in R. Blue
circles indicate an over-representation, and red circles indicate an under-representation.
https://doi.org/10.1371/journal.pone.0233912.g008
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 15 / 31
stone artefacts. On this hypothesis, the artefacts might be of any age within the period of
Holocene high sea levels during the past 7000 years.
It is necessary to consider hypotheses (a) and (b) together since evidence against one is by
default evidence for the other. Four relevant points are identified below.
1. None of the underwater artefacts show any evidence of rounding or edge-damage consis-
tent with fluvial transport or rolling by wave action.
This criterion alone is not conclusive evidence of in situ location because it is possible that
artefacts made on hard material such as igneous rock might be displaced by water action over
relatively short distances without evidence of such damage, but some of the artefacts are
located more than 50 m from the current shoreline, and some edge-damage might be expected
in those cases.
Further investigation of this issue was conducted using a synchrotron to examine the edges
of 10 of the artefacts collected from the seabed (Fig 9). This demonstrates clearly that beneath
the marine growth these artefacts retain acute edges. This is in marked contrast to other mate-
rial encountered during the surveys by the team, highlighted by a possible artefact of the same
material which was found in an intertidal rock shelter on Goodwyn Island, which shows clear
evidence of edge rounding caused by water rolling, so much so that confident determination
as to whether the object was natural or cultural was not possible (Fig 10).
2. The underwater artefacts are evenly distributed over a large area and at varying depths
and show no patterning consistent with the action of tidal currents or waves
There is no evidence of individual size-sorting or linear distributions along particular depth
contours consistent with areas of higher-energy currents (Fig 11). A linear model calculated
using the ‘stats’ package in R shows a no significant relationship between artefact size and
depth (adjusted R
2
= 0.011, p = 0.08074), with independence comfortably within standard
error and the small effect driven by only by a few exceptionally large outliers (Fig 12). Rather
their distribution appears random, somewhat like the artefact distribution on the calcarenite
terrace (and other terrestrial features), with different-sized artefacts distributed apparently at
random in both the horizontal and vertical dimension, and occasionally clustered at specific
locations. Had the channel artefacts been derived from the onshore material by high energy
wave action, with size differences due to differential sorting by water movement, it would be
expected that the largest material would travel over the shortest distances or remain undis-
turbed while the smaller material would be more likely to travel further offshore. The compar-
ative size analysis (Table 2, Figs 7and 8) demonstrates that this is clearly not the case and
indeed the pattern is the reverse of what would be expected on a hypothesis of differential
water transport.
3. There are statistically significant differences in size and morphology between the under-
water and onshore artefact assemblages.
The submerged artefacts are larger and more massive than those found onshore, and the
flakes have thicker platforms with less core rotation. The absence of the smallest size categories
in the channel could relate to differential visibility or sampling bias. However, there is no rea-
son why the rarity of the larger size categories on the terrace could be explained in this way,
and this size difference is most plausibly explained by technological differences between the
two assemblages. This is further evidence that the underwater assemblage belongs to an earlier
time period than the onshore material and is consistent with observed diachronic size varia-
tion in dated and stratified onshore sites elsewhere in the wider region [65].
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 16 / 31
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 17 / 31
4. Analysis of the pre- and post-cyclone landscape data shows that the only land surface
change was the movement of sand or seaweed across the calcarenite terrace (Fig 13).
Drone imagery was collected over the calcarenite terrace both before and after Cyclone
Veronica, a Category 2 cyclone that crossed the Dampier archipelago on 23–24 March 2019.
The Legendre Island weather station (located 6 km north of the Bruguieres Channel) recorded
a maximum 10-minute mean wind speed of 115 km/h, and a maximum 3-second wind gust of
157 km/h before the weather station went offline [66]. A dGPS-surveyed storm wrack-line at
the base of the dune approximately 30 cm above the back of the calcarenite terrace shows that
the entire terrace was inundated by a storm surge during the cyclone. Despite this, there was
almost no change in the position of the stone features and even small fragments of beach rock
remained in situ. This comparison demonstrates that the beachrock terrace and its associated
archaeology remained relatively stable throughout this cyclone, and it is thus unlikely that
larger artefacts, such as those identified in the channel, would have been transported over the
distances required as a result of extreme weather events of this magnitude.
Hypothesis (c)–the lag hypothesis–requires the consideration of the sorts of geomorpholog-
ical processes associated with Holocene sea-level rise and stabilisation that could have resulted
in a two-step process: first the accumulation of sediments to fill all or a large part of the Bru-
guieres channel and create a living surface for settlement and subsistence activities, and in a
second step the erosion of these sediments and the removal of all material except the stone
artefacts. The main supporting evidence that could be cited for the first step in this process is
the Holocene beach ridge complex along the west coast of North Gidley Island, which formed
as sea-level rise stabilised at about its current position, and which now curves around into the
Bruguieres Channel culminating in the calcarenite terrace (Fig 4). As part of this first step in
the lag hypothesis, it would be necessary to suppose that this beach ridge deposit originally
extended continuously along the western shorelines of North Gidley and Cape Bruguieres
Islands, blocking the channel that separates them today, and was then breached by tidal action,
perhaps during storm surges associated with the mid-Holocene high stand, to create the mod-
ern channel, removing the finer sediment fraction and leaving the heavier lithics to deflate
onto the underlying Pleistocene substrate.
Fig 9. Views of subtidal lithic A23, showing photogrammetric view (A), neutron tomographic slice (B), digitally de-concreted 3D tomography
(C) and digitally half-sectioned 3D tomography.
https://doi.org/10.1371/journal.pone.0233912.g009
Fig 10. Artefacts recovered under water from Cape Bruguieres (A20, A23, A29 and A11) have well-preserved,
acute edges showing no signs of rolling. This is in contrast to material where clear repeated transport and motion has
resulted in rolled edges and a lack of marine growth, demonstrated by a possible (but inconclusive) artefactfrom
Goodwyn Island cave (GC01).
https://doi.org/10.1371/journal.pone.0233912.g010
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 18 / 31
Fig 11. All finds with an associated measurement (as described in Table 2) were plotted on the LiDAR bathymetry in Cape Bruguieres Channel, and the symbols
altered to suit a gradient of smallest to largest recorded artefacts. The distribution of the finds indicates no relationship between depth and artefact size.
https://doi.org/10.1371/journal.pone.0233912.g011
Fig 12. Cape Bruguieres Channel Assemblage size and location data demonstrate the absence of relationship between depth and artefact size.
Plotted using the ggplot package in R [62].
https://doi.org/10.1371/journal.pone.0233912.g012
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 19 / 31
There are three difficulties with this hypothesis. First, there is no evidence, either from
LiDAR DEM or geological surveys, for any remnant of a beach ridge accreted onto the west coast
of Cape Bruguieres Island to the north of the channel, which is what would be expected if a beach
ridge complex had originally extended continuously across the current channel entrance.
Secondly, there is also no geomorphic evidence on the southern side of the channel for
truncation of the beach ridge-calcarenite complex by a channel erosion event. Geomorphic
evidence suggests that soon after the area was flooded, the shallow depression between North
Gidley and Bruguieres Islands became an active tidal channel with currents mobilising sedi-
ments to form an ebb tidal sand spit along the southern edge of the channel. This spit became
a permanent feature and allowed the developing beach ridge complex to accrete onto it while
the strong flow of tidal currents kept the channel open.
Thirdly, the development of an ebb tidal sand spit on the southern side of the channel and
towards its western entrance, and the lack of modern sediment deposition in the channel, sug-
gest that the current speeds are far too strong to allow the accumulation of beach-ridge sedi-
ments to fill the channel, as proposed in step one of the lag hypothesis.
Fig 13. Comparison of the Cape Bruguieres Channel southern beachrock terrace from aerial survey before and after Cyclone Veronica. Area 1 shows the addition
of sand in the dune behind the beachrock and a slight change in sediment depth on the platform. Area 3 shows only a slight change in the sediment at the edge of the
dune and removal of a line of seaweed in the centre of the image. (Figure: E. Beckett and J. Benjamin). Overview aerial image (top right) reproduced by permission of
the Western Australian Land Information Authority.
https://doi.org/10.1371/journal.pone.0233912.g013
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 20 / 31
Diver observations of maximum current flow during spring tides are a further indication
that current speeds are fast enough to transport sand-size particles, inhibiting the accumula-
tion and deposition of substantial sediments of sand and beach sediment. These observations
are also evidence that the artefacts are little disturbed by tidal action and could have remained
in place over long periods of time in this marine environment.
Further evidence against filling of the channel by soft sediments is the presence of extensive
coral communities, which require hard substrates to grow on and cannot tolerate burial by
sediment.
The protected environment of the channel, in the sense that it is protected from the full
force of wave and surf action acting on a shoreline directly facing the open sea, the lack of
cyclone modification, the quantity, condition and random size distribution of the underwater
lithics with respect to natural agencies, the significant size and morphological differences
between the underwater and terrestrial lithic assemblages, and the geomorphological evidence,
all argue in favour of cultural material that has not substantially moved from its original in situ
deposition on a Pleistocene land surface. It is not possible to rule out some slight movement of
material, or the possibility that some of the artefacts on the inner margin of the intertidal zone
have been displaced from the immediate shore edge, but any such movements appear to have
been of very limited extent. All the available evidence rules out any hypotheses arguing for
wholesale secondary deposition given the state of current knowledge of coastal geomorpholog-
ical and climatic processes, and rules in favour of the hypothesis that the underwater assem-
blage is largely in situ on a pre-inundation landscape and belongs to a much earlier time
period and an entirely different pattern of landscape use than the archaeological site on the cal-
carenite terrace.
Turning to consideration of the different patterns of land use associated with the two
assemblages, the attractions of the site on the calcarenite terrace are clearly related to the pres-
ent-day shoreline with its access to marine resources including shellfish, the remains of which
are preserved along with the artefacts. The attractions of the Bruguieres Channel site in a pre-
inundation landscape are less obvious. If late Pleistocene or early Holocene in date, the site
cannot relate to the marine resources of a contemporaneous palaeoshoreline, which would
have been some kilometres distant and perhaps much further. The more likely attractions
relate to the topographic and geological conditions at what would have been the boundary
between the exposed coastal plain extending to the west at lower sea levels, and the outcrop-
ping of the Archean bedrock geology to the east. For late Pleistocene or early Holocene
Aboriginal communities living on the exposed coastal plain and its palaeocoastlines, the Bru-
guieres area would have been the nearest source of raw material for making stone artefacts,
and therefore an obvious target during seasonal movements into the hinterland.
Moreover, the relatively narrow and shallow swale represented by the pre-inundation Bru-
guieres channel, located between more elevated rocky terrain immediately to the north and
the south, with the more hilly and complex volcanic terrain of the Dampier Ranges to the east,
would have afforded ecological and tactical advantages in accessing local patches of sediment,
water and plant food, and animals moving through the wider landscape. However, the imme-
diate attractions of the specific locations where individual artefacts were deposited might relate
to ephemeral and localised features, such as temporary fireplaces, that have left no surviving
trace on the surface of the seabed.
Flying Foam Passage: A submerged freshwater spring
A second submerged find spot was located within a seabed depression located at a depth of
14 m (MSL) along the central axis of Flying Foam Passage, approximately 600 m west of the
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 21 / 31
outer edge of an intertidal site located on the western shoreline of Dolphin Island [38]. The
submarine depression measures 50 m wide x 80 m long (Fig 14). It is one of several locations
along Flying Foam passage inspected by divers. The channel floor around the depression is
formed from limestone rather than crystalline bedrock, with deep erosional notches cut into
the sides of the depression. Multibeam and sidescan sonar imagery over the wider area demon-
strate that this is an isolated depression and not part of a fluvial channel flowing through Fly-
ing Foam Passage during periods of lower sea level. Therefore, one can confidently conclude
that these erosional notches could only have formed through the presence of standing water
forming a permanent or ephemeral spring or billabong. Extensive cobble/boulder fields are
found around the perimeter and base of this submarine freshwater spring but are almost
completely covered in encrusting marine growth, hampering the differentiation between
worked lithic artefacts and natural stone. One confirmed lithic artefact was located and
recorded from the floor of the depression on the final day of the expedition in this area and is
typologically comparable in size, shape and manufacturing technique to the artefacts from
Bruguieres Channel. Given the location of this find at a distance of at least half a kilometre
from the nearest shoreline and the reconstruction of the palaeotopography, the condition of
the artefact, and the direction and relatively weak velocity of current flows once this depression
was inundated by sea-level rise [67], it is highly unlikely that this artefact arrived at its present
position as a result of being eroded out from a terrestrial site on the nearest shoreline and
transported by water action over such a distance. Rather, it is interpreted as evidence of
human activity attracted to a freshwater spring in the pre-inundation landscape.
The depth of this find indicates that people could have been present around this freshwater
source until the area was inundated by sea-level rise c. 8500 cal BP, giving a minimum age of
deposition for the stone tool (Fig 2).
Just how much weight of interpretation can rest on a single artefact is, of course an issue,
but the single artefact is the result of limited diving time and the difficulties of identifying
Fig 14. A single artefact was recovered from Flying Foam Passage (WH-A01) where additional material may be
present rendering the area a high priority target for future investigations. Inspection below marine encrustations
reveals a clear striking platform, a bulb of percussion and dorsal ridges. (Photographs: H. Yoshida; C. Wiseman;
Drawing: K. Jerbić).
https://doi.org/10.1371/journal.pone.0233912.g014
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 22 / 31
artefacts against the background noise of other rocks and obscuring biogenic surface growth,
and the details above assure its status as an in situ specimen. In any case, this site, represents
an attractive target for more detailed investigation. Perhaps the most significant feature of this
find is its association with a submerged freshwater spring. This would have been an obvious
magnet for people living on the pre-inundation landscape. These underwater springs (known
colloquially as ‘wonky holes’) are known to exist elsewhere on the continental shelf and repre-
sent an obvious target for future underwater surveys.
The adjacent Dolphin Island provides additional context (Fig 15). Here an ongoing project
has identified a terrestrial landscape with quarry sites, living areas and rock art panels extend-
ing inland from the shoreline, and artefacts on the shore edge and in the intertidal zone (38).
These include partially submerged assemblages across the intertidal zone and beach with addi-
tional information recovered from the adjacent intertidal mudflats including a rhyodacite
flake recovered by coring, found at a depth of 40 cm below the surface of the alluvial sediment.
The potentially early date of this intertidal material is strongly suggested by the size and
other features of the artefacts, which show differences between intertidal and inland assem-
blages on Dolphin Island that are comparable to the differences between the underwater and
on-shore assemblages at Cape Bruguiere Channel (Figs 16 and 17). Moreover, the rock art
panels show a sequence that includes art styles of an early date belonging to a period when sea-
level was lower than present [38,43].
Investigations to corroborate the early date of the intertidal material and its taphonomic
and depositional history are ongoing [68], but the evidence already available clearly reinforces
Fig 15. Flying Foam Passage where multibeam sonar imaging of the submerged spring shows the location where a lithic artefact was
discovered at 14m (MSL). Distance from the findspot to the Dolphin Island intertidal site is approximately 600 m. Maps used are ©
Commonwealth of Australia (Geoscience Australia), satellite image by Sentinel (in the public domain).
https://doi.org/10.1371/journal.pone.0233912.g015
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 23 / 31
the interpretation of the mid-channel find that this area was an attractive focal point for
human activity with a long history of human occupation extending back into a period of lower
sea level, and a range of sites with the potential to provide insight into the ways in which local
communities responded and adapted to the loss of land and water supplies submerged by sea-
level rise.
Discussion
The past decade has seen interest increasingly focused on the 20 million km
2
of territory which
was available for human occupation on the world’s continental shelves during the low sea-lev-
els of the Last Glacial Period. Exposure of this vast territory and subsequent drowning by post-
glacial sea-level rise must have dramatically impacted patterns of human population dispersal,
cultural interactions and socio-economic development in many parts of the world including
Australia. Archaeological evidence from underwater sites is crucial to understanding these
processes.
The results presented here demonstrate that underwater cultural material can survive inun-
dation by sea-level rise in an Australian context, and that such evidence can be located and
studied using a combination of predictive modelling and an appropriate suite of underwater
Fig 16. Assemblage histogram (%) of Size categories for the four assemblages from CapeBruguieres channel (submerged) and platform (land) and Dolphin
Island intertidal and land sites.
https://doi.org/10.1371/journal.pone.0233912.g016
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 24 / 31
and remote-sensing techniques. This research also shows how ambiguities over the deposi-
tional and taphonomic status of artefacts found in shallow water close to the present-day
shoreline can be resolved through archaeological and geoarchaeological analysis and evalua-
tion of alternative hypotheses taking account of known and identifiable climatic and geomor-
phological processes of coastal disturbance and change. In this respect these results are a first
step to investigating the role of submerged landscapes in the peopling of Australia, and an
encouragement to conduct similar searches throughout the Asia Pacific region in both shallow
and deep water contexts.
Determining the age of cultural material deposited on a sediment-starved seabed, as in
Western Australia, remains a major challenge. The Bruguieres Channel site has a minimum
age of 7000 cal BP, the artefact adjacent to the submerged freshwater spring in Flying Foam
Passage 8500 cal BP, both based on the latest date by when the sites would have been sub-
merged by Holocene sea-level rise SL Figure reference. These submerged sites could be much
older since the terrestrial features with which they are associated would have been available
since the time of first human entry into Australia. However, no absolute dating methods or
other indicators are available that could further constrain the age of these particular sites
beyond the limiting dates and minimum ages presented here.
Late Pleistocene and early Holocene coastal archaeological sequences are rare except where
offshore islands preserve this time slice and provide a window onto earlier use of the sub-
merged landscape [7,69,49]. This reinforces the incentive to extend underwater searches for
earlier material. The minimum age for the Bruguieres and Flying Foam sites is not especially
early, but this reflects the fact that they are sitting on the seabed in relatively shallow water and
can only be given a minimum age derived from the regional sea-level curve. Decisive evidence
for earlier sites will depend on the search for artefacts in stratified and dateable sediments if
they are preserved and can be found, or artefacts on the seabed surface at greater depth and
further offshore.
This poses two challenges. The first is the greater logistical complexity and costs of working
in deeper water. This requires larger vessels equipped with a full range of acoustic equipment
for surface mapping and sub-bottom and seismic equipment for sub-surface mapping as well
Fig 17. Residual plot for the Pearson’s chi-squared test of artefact sizes from the Cape Bruguieres and Dolphin Intertidal
assemblages, using the ggcorrplot package [64] in R. Blue circles indicate an over-representation, and red circles indicate an
under-representation. A Pearson’s chi-squared test shows this difference is statistically significant (X
2
(df = 9, N = 881) = 505.27,
simulated p<0.001) with a moderate to strong effect size (V = 0.44) [63].
https://doi.org/10.1371/journal.pone.0233912.g017
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 25 / 31
as facilities for coring and dredging, remotely operated vehicles and, in the event of discoveries
that need closer inspection, submersibles and diving teams trained in deep diving techniques
and a regulatory compliance environment which allows for scientific diving at those depths.
The second challenge is to develop predictive models that can pinpoint target areas of inter-
est with sufficient precision to focus search and discovery and justify the greater costs of deep-
water investigation. One potential target highlighted by these results is submerged freshwater
springs. These are known to occur in many locations and at varying depths on the continental
shelf, would have been an obvious magnet for human activities, and can be detected by remote
sensing techniques. Another potential target is the palaeoshorelines formed at still stands dur-
ing the Last Glacial, where there is evidence of mangrove swamps and intertidal mudflats,
affording access to marine resources and coastal ecotones that would have attracted concentra-
tions of human activity and settlement. These and other possibilities in this region have been
identified through a variety of predictive models [28,67]. Successful deep-water investigations
have been conducted in other parts of the world [70,21,71] and there is every reason to sup-
pose that they can be applied with equal success in the Australian context.
These findings also reinforce the need for revisions to the existing legislation on the man-
agement and protection of submerged Indigenous heritage in the Australian subtidal jurisdic-
tion. Although the national Historic Shipwrecks Act 1976 has been re-named to be more
inclusive as the Underwater Cultural Heritage Act 2018, the update only automatically protects
shipwrecks and sunken aircraft older than 75 years in Australian waters; however automatic
protection does not extend to other categories of other underwater archaeological sites such as
submerged Aboriginal sites. This is in contrast to those countries that have signed the 2001
UNESCO Convention on the Protection of the Underwater Cultural Heritage (which protects all
categories of UCH sites >100 years old), and member states of the European Union, which are
required to include submerged landscape archaeology in environmental impact assessments in
advance of offshore industrial development [72,73,74]. The findings presented in this article
highlight that this gap in Australian legislation must be addressed. Experience elsewhere shows
that in the wake of such legislation, collaborative arrangements with offshore industries can
flourish to mutual benefit, bringing substantial resources and industrial technology to bear on
underwater archaeological and palaeoenvironmental investigations [7477].
Managing, investigating and understanding the archaeology of the Australian continental
shelf in partnership with Indigenous Traditional Owners is one of the last frontiers in Austra-
lian archaeology, and the results in this article present the first steps in this journey of discov-
ery, in confirming the archaeological potential of Australia’s continental shelves, and in
beginning to fill what has remained until now a major gap in the human history of the
continent.
Supporting information
S1 Fig. A stone artefact recovered from Cape Bruguieres channel (Photograph: S. Wright).
(TIF)
S1 Table. Record of all recovered artefacts from Cape Bruguieres channel (CB), Flying
Foam Passage (FF), and the Flying Foam Passage submerged freshwater spring (WH1).
(XLSX)
Acknowledgments
The Deep History of Sea Country (DHSC) project team recognises the Murujuga Aboriginal
Corporation Council of Elders and Murujuga Land and Sea Rangers Unit as core collaborators
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 26 / 31
on this project. The DHSC project was supported by the Australian Research Council, Flinders
University and the Hackett Foundation. We thank Charles Collins and Anthony Pyne from
EGS Surveys for providing the multibeam data of Flying Foam Passage, and Pilbara Ports
Authority for supplying the LADS survey data for Mermaid Sound and Mermaid Strait. Addi-
tional thanks to Ken Mulvaney, Shakti Chakravarty, Victoria Anderson, Sarah de Koning,
Hiro Yoshida, Kerry Ludwig, Mads Holst, Sam Wright, Annette George, Tom Allardyce and
Graham and Michelle Evans and the AustMarine staff for their support throughout the
project.
Author Contributions
Conceptualization: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Jorg Hacker, Peter
Jeffries, Sean Ulm, Geoff Bailey.
Data curation: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, Emma
Beckett, Francis Stankiewicz, Jerem Leach, Jorg Hacker, Paul Baggaley.
Formal analysis: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, John
McCarthy, Emma Beckett, Patrick Morrison, Francis Stankiewicz, Jerem Leach, Jorg
Hacker, Katarina Jerbić, John Fairweather, Sean Ulm, Geoff Bailey.
Funding acquisition: Jonathan Benjamin, Michael O’Leary, Jorg Hacker, Peter Jeffries, Sean
Ulm, Geoff Bailey.
Investigation: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, John
McCarthy, Emma Beckett, Patrick Morrison, Francis Stankiewicz, Jerem Leach, Jorg
Hacker, Paul Baggaley, Madeline Fowler, Sean Ulm, Geoff Bailey.
Methodology: Jonathan Benjamin, Michael O’Leary, Jo McDonald, John McCarthy, Francis
Stankiewicz, Jerem Leach, Jorg Hacker, Paul Baggaley, Katarina Jerbić, Madeline Fowler,
Sean Ulm, Geoff Bailey.
Project administration: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Sean Ulm,
Geoff Bailey.
Resources: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, Peter Jef-
fries, Sean Ulm, Geoff Bailey.
Software: Jonathan Benjamin, Michael O’Leary, Chelsea Wiseman, John McCarthy, Emma
Beckett, Francis Stankiewicz, Jerem Leach, Paul Baggaley, Sean Ulm.
Supervision: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Peter Jeffries, Sean Ulm,
Geoff Bailey.
Validation: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, John
McCarthy, Emma Beckett, Francis Stankiewicz, Jerem Leach, Jorg Hacker, Katarina Jerbić,
Sean Ulm, Geoff Bailey.
Visualization: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wiseman, John
McCarthy, Emma Beckett, Francis Stankiewicz, Katarina Jerbić.
Writing – original draft: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea Wise-
man, John McCarthy, Emma Beckett, Patrick Morrison, Jorg Hacker, Sean Ulm, Geoff
Bailey.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 27 / 31
Writing – review & editing: Jonathan Benjamin, Michael O’Leary, Jo McDonald, Chelsea
Wiseman, John McCarthy, Emma Beckett, Francis Stankiewicz, Jerem Leach, Jorg Hacker,
Paul Baggaley, Katarina Jerbić, Madeline Fowler, Peter Jeffries, Sean Ulm, Geoff Bailey.
References
1. Bailey GN, Harff J, Sakellariou D, editors. Under the sea: archaeology and palaeolandscapes of the
continental shelf. Cham: Springer; 2017.
2. Williams AN, Ulm S, Sapienza T, Lewis S, Turney CS. Sea-level change and demography during the
last glacial termination and early Holocene across the Australian continent. Quat Sci Rev. 2018; 182:
144–54.
3. Dillehay TD. The settlement of the Americas: a new prehistory. New York: Basic Books; 2000.
4. Clarkson C, Jacobs Z, Marwick B, Fullagar R, Wallis L, Smith M, et al. Human occupation of northern
Australia by 65,000 years ago. Nature. 2017; 547(7663): 306–310. https://doi.org/10.1038/nature22968
PMID: 28726833
5. Maloney T, O’Connor S, Wood R, Aplin K, Balme J. Carpenters Gap 1: A 47,000 year old record of
indigenous adaption and innovation. Quat Sci Rev. 2018; 191: 204–228.
6. McDonald J, Reynen W, Petchey F, Ditchfield K, Byrne C, Vannieuwenhuyse D, et al. Karnatukul (Ser-
pent’s Glen): A new chronology for the oldest site in Australia’s Western Desert. PloS one. 2018; 13(9):
e0202511. https://doi.org/10.1371/journal.pone.0202511 PMID: 30231025
7. Veth P, Ward I, Manne T, Ulm S, Ditchfield K, Dortch J, et al. Early human occupation of a maritime des-
ert, Barrow Island, North-West Australia. Quat Sci Rev. 2017; 168: 19–29.
8. David B, Delannoy JJ, Mialanes J, Clarkson C, Petchey F, Geneste JM, et al. 45,610–52,160 years of
site and landscape occupation at Nawarla Gabarnmang, Arnhem Land plateau (northern Australia).
Quaternary Science Reviews. 2019; 215: 64–85.
9. Davis LG, Madsen DB, Becerra-Valdivia L, Higham T, Sisson DA, Skinner SM, et al. Late Upper Paleo-
lithic occupation at Cooper’s Ferry, Idaho, USA,~ 16,000 years ago. Science. 2019; 365(6456): 891–
897. https://doi.org/10.1126/science.aax9830 PMID: 31467216
10. Faure H, Walter RC, Grant DR. The coastal oasis: ice age springs on emerged continental shelves.
Glob Planet Change. 2002; 33(1-2): 47–56.
11. Bird MI, Taylor D, Hunt C. Palaeoenvironments of insular Southeast Asia during the Last Glacial Period:
a savanna corridor in Sundaland?. Quat Sci Rev. 2005; 24(20-21): 2228–2242.
12. Bailey GN, Flemming NC. Archaeology of the continental shelf: marine resources, submerged land-
scapes and underwater archaeology. Quat Sci Rev. 2008; 27(23-24): 2153–2165.
13. Will M, Kandel AW, Conard NJ. Midden or molehill: The role of coastal adaptations in human evolution
and dispersal. J. World Prehist. 2019; 32(1): 33–72.
14. Mellars P, Gori KC, Carr M, Soares PA, Richards MB. Genetic and archaeological perspectives on the
initial modern human colonization of southern Asia. Proc. Natl. Acad. Sci. 2013; 110(26): 10699–10704.
https://doi.org/10.1073/pnas.1306043110 PMID: 23754394
15. Boivin N, Fuller DQ, Dennell R, Allaby R, Petraglia MD. Human dispersal across diverse environments
of Asia during the Upper Pleistocene. Quat Int. 2013; 300: 32–47.
16. Groucutt HS, Petraglia MD, Bailey G, Scerri EM, Parton A, Clark-Balzan L, et al. Rethinking the dis-
persal of Homo sapiens out of Africa. Evol Anthropol. 2015; 24(4): 149–164. https://doi.org/10.1002/
evan.21455 PMID: 26267436
17. Bailey GN, Meredith-Williams M, Alsharekh A, Hausmann N. The archaeology of Pleistocene coastal
environments and human dispersals in the Red Sea: insights from the Farasan Islands. In Rasul NMA,
Stewart ICF editors. Geological setting, palaeoenvironment and archaeology of the Red Sea. Cham:
Springer; 2019, pp. 583–604.
18. Petraglia MD, Breeze PS, Groucutt HS. Blue Arabia, Green Arabia: examining human colonisation and
dispersal models. In Rasul NMA, Stewart ICF editors. Geological setting, palaeoenvironment and
archaeology of the Red Sea. Cham: Springer; 2019, pp. 675–683.
19. Gaffney VL, Thomson K, Fitch S, editors. Mapping Doggerland: the Mesolithic landscapes of the south-
ern North Sea. Oxford: Archaeopress; 2007.
20. Gaffney VL, Fitch S, Smith DN. Europe’s lost world: the rediscovery of Doggerland. York: Council for
British Archaeology; 2009.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 28 / 31
21. Gaffney V, Allaby R, Bates R, Bates M, Ch’ng E, Fitch S, et al. Doggerland and the Lost Frontiers Proj-
ect (2015-2020). In Bailey G, Harff J, Sakellariou D, editors. Under the sea: archaeology and palaeo-
landscapes of the continentals. Cham: Springer; 2017, pp. 305–319.
22. Benjamin J, Bonsall C, Pickard C, Fischer A, editors. Submerged prehistory. Oxford: Oxbow Books;
2011.
23. Evans AM, Flatman JC, Flemming NC, editors. Prehistoric archaeology on the continental shelf: a
global review. New York: Springer; 2014.
24. Harff J, Bailey G, Lu¨th F, editors. Geology and archaeology: Submerged landscapes of the continental
shelf. Geological Society of London, Special Publications 411; 2016.
25. Ward I, Larcombe P, Carson A, Lane A. Archaeological assessment of coastal and marine development
sites: case study from James Price Point, Western Australia. Journal of the Royal Society of Western
Australia. 2016; 99(2): 31–46.
26. Flemming NC, Harff J, Moura D, Burgess A, Bailey GN, editors. Submerged landscapes of the Euro-
pean continental shelf: Quaternary paleoenvironments. Chichester: John Wiley & Sons; 2017.
27. Larcombe P, Ward IA, Whitley T. Physical sedimentary controls on subtropical coastal and shelf sedi-
mentary systems: Initial application in conceptual models and computer visualizations to support
archaeology. Geoarchaeology. 2018; 33(6): 661–679.
28. Veth P, McDonald J, Ward I, O’Leary M, Beckett E, Benjamin J, et al. A strategy for assessing continuity
in terrestrial and maritime landscapes from Murujuga (Dampier Archipelago), North West Shelf, Austra-
lia. J. Isl. and Coast. Archaeol. 2019; https://doi.org/10.1080/15564894.2019.1572677
29. Bailey G, Galanidou N, Peeters H, Jo
¨ns H, Mennenga M, editors. The archaeology of Europe’s drowned
landscapes. Cham: Springer; 2020. Available from https://link.springer.com/book/10.1007/978-3-030-
37367-2.
30. Jones R. The geographical background to the arrival of man in Australia and Tasmania. Archaeol.
Phys. Anthropol. Ocean. 1968; 3(3): 186–215.
31. Ulm S. Coastal foragers on southern shores: Marine resource use in northeast Australia since the late
Pleistocene. In Bicho N, Haws J, Davis LG, editors. Trekking the shore. New York: Springer; 2011, pp.
441–461.
32. Nutley D. Inundated site studies in Australia. In Evans AM, Flatman JC, Flemming NC, editors. Prehis-
toric archaeology on the continental shelf: A global review. New York: Springer; 2014, pp. 255–273.
33. Flemming NC. Sirius Expedition-Cootamundra Shoals Survey 1982: Expedition Reports.
34. Nutley DM, Coroneos C, Wheeler J. Potential submerged Aboriginal archaeological sites in South West
Arm, Port Hacking, New South Wales, Australia. In Harff J, Bailey G, Lu¨th F, editors. Geology and
archaeology: Submerged landscapes of the continental shelf. Geological Society of London, Special
Publications 411; 2016, pp. 265–85.
35. Rowland MJ, Ulm S. Indigenous fish traps and weirs of Queensland. Queensland Archaeological
Research 2011; 14:1–58.
36. Lewczak C, Wilby C. Aboriginal Quarry Site, Rocky Point, Port Cygnet: Maritime Archaeological investi-
gation. Report for Southeast Tasmania Aboriginal Corporation. Cosmos Archaeology Pty Ltd; 2010.
37. Ulm S. Coastal themes: An archaeology of the southern Curtis Coast, Queensland. ANU E Press;
2006.
38. Dortch J, Beckett E, Paterson A, McDonald J. Stone artifacts in the intertidal zone, Dampier Archipel-
ago: Evidence for a submerged coastal site in Northwest Australia. J. Isl. & Coast. Archaeol. 2019:1–5.
https://doi.org/10.1080/15564894.2019.1666941
39. Dortch C. Prehistory down under: Archaeological investigations of submerged Aboriginal sites at Lake
Jasper, Western Australia. Antiquity. 1997; 71(271): 116–123.
40. Bird MK The impact of tropical cyclones on the archaeological record: An Australian example. Archaeol-
ogy in Oceania. 1992; 27(2): 75–86.
41. Donaldson M. Understanding the rocks: Rock art and the geology of Murujuga (Burrup Peninsula) [with
comments by Bednarik and Donaldson] Rock Art Research. 2011; 28(1): 35–43.
42. McDonald J, Veth P. Dampier Archipelago petroglyphs: Archaeology, scientific values and National
Heritage Listing. Archaeology in Oceania. 2009; 44(S1):49–69.
43. Mulvaney K. Murujuga Marni: rock art of the macropod hunters and mollusc harvesters. CRAR+M
Monograph 1. Perth: UWA Press, 2015.
44. Pillans B, Fifeld K. Erosion rates and weathering history of rock surfaces associated with Aboriginal
rock art engravings (petroglyphs) on Burrup Peninsula, Western Australia, from cosmogenic nuclide
measurements. Quat. Sci. Rev. 2013; 69: 98–106.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 29 / 31
45. Yokoyama Y, De Deckker P, Lambeck K, Johnston P, Fifield LK. Sea–level at the Last Glacial Maxi-
mum: evidence from northwestern Australia to constrain ice volumes for oxygen isotope stage 2.
Palaeogeogr Palaeoclimatol Palaeoecol. 2001; 165(3-4):281–297.
46. Lewis SE, Sloss CR, Murray–Wallace CV, Woodroffe CD, Smithers SG. Post–glacial sea–level
changes around the Australian margin: a review. Quat. Sci. Rev. 2013; 74: 115–138.
47. Solihuddin T, Collins LB, Blakeway D, O’ Leary MJ. Holocene coral reef growth and sea level in a
macrotidal, high turbidity setting: Cockatoo Island, Kimberley Bioregion, northwest Australia. Mar. Geol.
2015; 359: 50–60.
48. Grant KM, Rohling EJ, Bar-Matthews M, Ayalon A, Medina–Elizalde M, Ramsey CB, et al. Rapid cou-
pling between ice volume and polar temperature over the past 50,000 years. Nature. 2012; 491: 744–
747. https://doi.org/10.1038/nature11593 PMID: 23151478
49. McDonald J, Veth P. Assessment of outstanding universal values: Dampier Archipelago, Western Aus-
tralia. Report for Australian Heritage Council. Jo McDonald–CHM report to Department of Sustainabil-
ity, Environment, Water, Population and Communities; 2011.
50. McDonald J. I must go down to the seas again: Or, what happens when the sea comes to you? Muru-
juga rock art as an environmental indicator for Australia’s north–west. Quat Int. 2015; 385: 124–135.
51. McDonald J, Reynen W, Ditchfield K, Dortch J, Leopold M, Stephenson B, et al. Murujuga Rockshelter:
first evidence for Pleistocene occupation on the Burrup Peninsula. Quaternary Science Reviews. 2018;
193: 266–287.
52. Benjamin J, O’Leary M, Ward I, Hacker J, Ulm S, Veth P, et al. Underwater archaeology and submerged
landscapes in Western Australia. Antiquity. 2018; 92(363): https://doi.org/10.15184/aqy.2018.103
53. Garbe U, Randall T, Hughes C, Davidson G, Pangelis S, Kennedy SJ. A new neutron radiography/
tomography/imaging station DINGO at OPAL. Physics Procedia. 2015; 69: 27–32.
54. Friedman GM. Identification of carbonate minerals by staining methods. Journal of Sedimentary
Research. 1959; 29(1): 87–97.
55. Beverly RK, Beaumont W, Tauz D, Ormsby KM, von Reden KF, Santos GM, et al. The Keck Carbon
Cycle AMS laboratory, University of California, Irvine: Status report. Radiocarbon. 2010; 52(2): 301–
309.
56. Dunbar E, Cook GT, Naysmith P, Tripney BG, Xu S. AMS 14C dating at the Scottish Universities Envi-
ronmental Research Centre (SUERC) radiocarbon dating laboratory. Radiocarbon. 2016; 58(1): 9–23.
57. Ramsey CB. Bayesian analysis of radiocarbon dates. Radiocarbon. 2009; 51(1): 337–360.
58. Reimer PJ, Bard E, Bayliss A, Beck JW, Blackwell PG, Ramsey CB, et al. IntCal13 and Marine13 radio-
carbon age calibration curves 0–50,000 years cal BP. Radiocarbon. 2013; 55(4): 1869–1887.
59. Petchey F, Clark G. Tongatapu hardwater: Investigation into the 14C marine reservoir offset in lagoon,
reef and open ocean environments of a limestone island. Quat. Geochronol. 2011; 6(6): 539–549.
60. Fairweather JH, Geological and geochemical characterisation of the Archean Fortescue Group, Rose-
mary Island, and associated archaeological material from the Dampier Archipelago, Western Australia.
Unpublished Geology Honours Thesis, University of Western Australia. 2019.
61. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical
Computing. 2019. Available from: https://www.R-project.org/.
62. Wickham H, Averick M, Bryan J, Chang W, McGowan L, Franc¸ois R, et al. Welcome to the Tidyverse.
Journal of Open Source Software. 2019; 4(43):1686.
63. Wolverton S, Dombrosky J, Lyman RL. Practical significance: ordinal scale data and effect size in
zooarchaeology. Int. J. Osteoarchaeol. 2016; 26(2):255–65.
64. Kassambara, A. ggcorrplot: Visualization of a Correlation Matrix using ’ggplot2’. R package version
0.1.3. 2019. Available from: https://CRAN.R-project.org/package=ggcorrplot
65. Lorblanchet M, Jones RM. The first excavations at Dampier (Western Australia), and their place in Aus-
tralian archaeology (with addendum on radiocarbon dating of Skew Valley midden). Chapter 2, part II.
Technical Reports of the Australian Museum online. 2018; 27: 163–190.
66. Bureau of Meteorology (BOM). Severe Tropical Cyclone Veronica. 2019. Available from: http://www.
bom.gov.au/cyclone/history/veronica.shtml
67. Ward I, Larcombe P. Mulvaney K, Fandry C. The potential for discovery of new submerged archaeolog-
ical sites near the Dampier Archipelago, Western Australia. Quaternary International. 2013;308–309:
216–229.
68. Morrison P, The Geoarchaeology of a Submerged Aboriginal Site in the Intertidal Zone of Dolphin
Island, Murujuga. Unpublished BA(Hons) Thesis, CRAR+M and Archaeology, University of Western
Australia. 2019.
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 30 / 31
69. McDonald J, Berry M. Murujuga, northwestern Australia: when arid hunter–gatherers became coastal
foragers. J. Isl. and Coast. Archaeol. 2016; 12: 24–43.
70. Bailey GN, Devès MH, Inglis RH et al. Blue Arabia: Palaeolithic and underwater survey in SW Saudi
Arabia and the role of coasts in Pleistocene dispersal. Quaternary International. 2015; 382: 42–57.
http://dx.doi.org/10.1016/j.quaint.2015.01.002
71. Braje TJ, Maloney JM, Gusick AE, Erlandson JM, Nyers A, Davis L, et al. Working from the known to
the unknown: Linking the subaerial archaeology and the submerged landscapes of Santarosae Island,
Alta California, USA. Open Quaternary. 2019; 5(1): http://doi.org/10.5334/oq.66
72. Dromgoole S. Continental Shelf Archaeology and International Law. In Bailey G, Galanidou N, Peeters
H, Jo
¨ns H, Mennenga M, editors. The archaeology of Europe’s drowned landscapes. Cham: Springer;
2020, April 10, pp. 495–507. Available from https://link.springer.com/chapter/10.1007/978-3-030-
37367-2_25
73. Pater C. Seabed Infrastructure Projects, Underwater Cultural Heritage and The Environmental Assess-
ment Process: The UK Example. In Bailey G, Galanidou N, PeetersH, Jo
¨ns H, Mennenga M, editors.
The archaeology of Europe’s drowned landscapes. Cham: Springer; 2020, April 10, pp. 509–520.
Available from https://link.springer.com/chapter/10.1007/978-3-030-37367-2_26
74. Peeters H, Bailey G, Galanidou N. The legal, historical and industrial context of underwater heritage:
Introduction. In Bailey G, Galanidou N, Peeters H, Jo
¨ns H, Mennenga M, editors. The archaeology of
Europe’s drowned landscapes. Cham: Springer; 2020, April 10, pp. 485–493. Available from https://
link.springer.com/chapter/10.1007/978-3-030-37367-2_24
75. Moree JM, Sier MM, editors. Interdisciplinary research programme Maasvlakte 2, Rotterdam. BOOR,
Rotterdam; 2015.
76. Tizzard L, Bicket AR, de Loecker D. Seabed prehistory: Investigating the palaeogeography and Early
Middle Palaeolithic archaeology in the southern North Sea. Wessex Archaeology Report 35. Salisbury:
Wessex Archaeology; 2015.
77. Bailey G, Jo
¨ns H. The Baltic and Scandinavia: Introduction. In Bailey G, Galanidou N, Peeters H, Jo
¨ns
H, Mennenga M, editors. The archaeology of Europe’s drowned landscapes. Cham: Springer; 2020,
April 10, pp. 27–38. Available from https://link.springer.com/chapter/10.1007/978-3-030-37367-2_2
PLOS ONE
Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in NW Australia
PLOS ONE | https://doi.org/10.1371/journal.pone.0233912 July 1, 2020 31 / 31
... This study is of particular importance given the current discourse concerning human use of coastal zones throughout our species' evolutionary history. Coastal zones are key to multiple critical developments in human evolution, including migrations out of Africa and across the globe, the development of maritime technologies that supported such movements, variations on subsistence practices that may have supported population increases, the first examples of monumental architecture, and human reactions to climate change (Grøn 2006;Erlandson et al., 2011;Thompsonand Worth, 2011;Astrup 2018;Galili et al., 2019;Benjamin et al., 2020;Bailey and Hardy 2021). Shell middens are often a marker for coastal occupations. ...
... They can generally be divided into studies of coastlines that experience tropical cyclones/hurricanes and those that do not. Studies of regions that experience tropical cyclone/hurricane activity include Western Australia (Benjamin et al., 2020;Wiseman et al., 2020) and Apalachee Bay. Studies of regions that do not experience tropical cycle/hurricane activity include the Channel Islands off the Pacific coast of southern California and the Baltic. ...
... Benjamin, Wiseman and colleagues assessed the submerged sites on the shallow continental shelf of Western Australia for evidence that anthropogenic lithic materials might have been displaced by storm activity (Wiseman et al., 2020). Several studies have determined that horizontal displacement of archaeological deposits appears minimal, including after tropical cyclone impacts on these sites (Benjamin et al., 2020;Wiseman et al., 2020). However, the deposits under study consisted only of lithic deposits instead of intermingled shell, sediment, and lithics, and thus are not direct sedimentological analogs for the Econfina Channel materials. ...
Article
The nature of preservation potential for submerged cultural landscapes on continental shelves remains varied and not easily defined. This is due to two factors: First, relatively few sites have been discovered offshore, creating a small sample size. Second, excavation and analysis of such submerged sites have only begun to mature during the past few decades. We present here results from an exploratory study of one such site that examines impacts from two tropical cyclone systems in successive years. The main feature of this site is a specific type of archaeological feature: a shell midden. These sites are excellent indicators for coastal occupations globally and can retain evidence for major developments in cultural evolution, including reactions to climate change and sea-level rise from the Pleistocene into the Holocene. Further, they are found worldwide. The degree to which middens have been impacted by past and present marine forces, including tropical cyclones, is a key factor in their potential to preserve such evidence. We used a geoarchaeological methodology that quantifies particle size fractions from sediments taken from various zones within the site to compare them before and after each storm. Results have significant implications for preservation of submerged, formerly coastal archaeological sites globally.
... Indeed, the northeast Australian coastline likely hosted large freshwater lagoon systems before sea-level rise breached the continental shelf, with karstified hills along the continental margin from previous reef building episodes during high stands blocking the outlet of rivers to the shelf edge (Woolfe et al., 1998;Dunbar and Dickens, 2003). Benjamin et al. (2020) have also located evidence for the availability of freshwater on the Pleistocene coast in the form of drowned freshwater springs near Cape Bruguieres, Murujuga. ...
... Jeradino, 2016a;Ward et al., 2017Ward et al., , 2018 and underwater archaeology on submerged sites (e.g. Benjamin et al., 2020;Benjamin and Ulm, 2021;McCarthey et al. 2022) but we wish to focus on these three here. ...
... Grøn et al. (2022) have recently critiqued how the application of analogue topographical models, along with their assumptions of productivity, resource distribution through time and inundation, will likely be far more complex than continuity models and thus both the accurate location of submerged sites and their dating in Australia are in their infancy (e.g. Benjamin et al., 2020). ...
Article
There are few archaeological sites that contain records for Pleistocene coastal occupation in Australia, as is the case globally. Two major viewpoints seek to explain why so few sites exist. The first is that the Pleistocene coast was a relatively marginal environment where fluctuating sea levels actively inhibited coastal resource productivity until the mid-to-late Holocene. The second position suggests that the Pleistocene coast (and its resources) was variably productive, potentially hosting extensive populations, but that the archaeological evidence for this occupation has been submerged by sea level rise. To help reconcile these perspectives in Australia, this paper provides a review, discussion, and assessment of the evidence for Australian Pleistocene coastal productivity and occupation. In doing so, we find no reason to categorically assume that coastal landscapes were ever unproductive or unoccupied. We demonstrate that the majority of Pleistocene coastal archaeology will be drowned where dense marine faunal assemblages should only be expected close to palaeo-shorelines. Mixed terrestrial and marine assemblages are likely to occur at sites located >2 km from Pleistocene shorelines. Ultimately, the discussions and arguments put forward in this paper provide a basic framework, and a different set of environmental expectations, within which to assess the results of independent coastal research.
... For example, a blog post from the Deep History of Sea Country (DHSC) project team members indicates it was local knowledge rather than systematic survey (c.f. Benjamin et al., 2020;Wiseman et al., 2021) that directed scientists to the submerged stone features in the Cape Bruguieres channel in the Dampier Archipelago (CRARM, 2020) ( Figure 1). Claims that the Cape Bruguieres site represents the first in situ submerged archaeological site in Australia (Benjamin et al., 2020) have unfortunately not stood up to scientific scrutiny, with the site almost certainly representing a secondary (i.e., reworked) and ponded artefact scatter, i.e., artefacts accumulated in ponded water above lowest tide level (Ward et al., 2022b). ...
... Benjamin et al., 2020;Wiseman et al., 2021) that directed scientists to the submerged stone features in the Cape Bruguieres channel in the Dampier Archipelago (CRARM, 2020) ( Figure 1). Claims that the Cape Bruguieres site represents the first in situ submerged archaeological site in Australia (Benjamin et al., 2020) have unfortunately not stood up to scientific scrutiny, with the site almost certainly representing a secondary (i.e., reworked) and ponded artefact scatter, i.e., artefacts accumulated in ponded water above lowest tide level (Ward et al., 2022b). This reanalysis emphasizes the importance of understanding the evolution of the physical seascape and of past and present physical processes to interpreting site formation Ward et al., 2015;Larcombe et al. 2018) and not emphasizing the significance of a site for merely being under water (Lemke 2020). ...
Article
Full-text available
There is growing awareness of the need for greater acknowledgement of underwater prehistoric cultural resources as part of management and regulation of the seabed around many maritime countries, especially those with large indigenous populations and history such as Australia. Prehistoric cultural places and landscapes inundated by Post-glacial sea-level rise on Australia’s continental shelf remain largely out-of-sight and out-of-mind, hence awareness and hence legal protection of this resource is lacking. There is a clear need for greater integration of archaeology and cultural heritage management within the marine sciences as well as a greater awareness of this resource as part of a common heritage more generally. This paper explores some of the dichotomies between Western and Indigenous cultures in valuing and managing the seabed. We argue that in developing science-policy, an attempt at least needs to be made to bridge both the gap between the nature and culture perspectives, and the jurisdictional divide between land and sea. Part of the answer lies in a convergence of Indigenous knowledge with Western science approaches, focused around our understanding of physical processes impacting past and present coastal landscapes and on the seabed itself. We explore several case studies from northern and Western Australia that are trying to do this, and which are helping to provide a greater appreciation of the inundated landscapes of the inner shelf as part of a common heritage.
... To facilitate accurate comparisons of elevation and ease of understanding, where possible, we have reduced elevation data to one known nationwide fixed datum, the AHD, established in 1971 with the 0.0 m surface fitted to tide gauge measurements of mean sea level (MSL) around the Australian mainland (Granger, 1972; F I G U R E 3 LIDAR-derived digital terrain model (vegetation removed) showing the distribution of underwater and intertidal stone artefacts recorded in the Cape Bruguieres channel using nautical chart tide data (Data: Australian Hydrographic Service, 2017; B2020). Marked tidal elevation lines are relative to Cape Legendre, as listed in Benjamin et al. (2020) Roelse et al., 1975). Small differences do occur between AHD and MSL, and local MSL is adjusted at times to include the analysis of new data and the modern rates of sea-level rise. ...
Article
Full-text available
The absence of known prehistoric underwater cultural heritage (UCH) sites on the Australian inner shelf stands in stark contrast to the thousands of sites revealed elsewhere in the world. Two recent claims—Dortch et al. (D2019) and Benjamin et al. (B2020)—put forward the first in situ (i.e., primary context) UCH sites in the shallow waters of the Dampier Archipelago, North West Australia, each arguing that the stone artefact scatters are at least 7000 years old and are now submerged because of postglacial sea‐level rise. We present new hydrodynamic modelling and data on coastal erosion and bathymetry, and reassess each site's sedimentary setting and archaeological site‐formation history. D2019 and B2020 clearly present lithic cultural artefacts, but the arguments for their sites being of primary context and reflecting early Holocene land surfaces are mistaken. Rather, these sites occur in the intertidal zone, and many or all artefacts are likely to have been reworked. Sites of secondary context, if treated appropriately, can inform our understanding of site‐formation process and change, and may support more powerful contributions to submerged archaeology than attempts to seek the first or the oldest.
... However, recent technical and methodological innovations and new research in diverse geographic areas are providing tantalizing data at an ever-increasing rate (e.g. Astrup et al. 2021;Benjamin et al. 2020;Cook Hale et al. 2021;de Smet, Timothy, and Smith 2019;Faught and Smith 2021;Grøn et al. 2021). Furthermore, global investment in offshore wind farms and other oceanic developments have created a booming industry on the continental shelves, where existing laws require cultural assessments (Bailey et al. 2020). ...
Article
The archaeology of inundated cultural landscape sites is not new and is an important component of the global record, yet these sites are distinct from shipwrecks and other site types underwater. Just as on land, underwater sites are subject to a dynamic range of formation processes, which must be analytically controlled. However, there are lingering misconceptions about underwater sites, specifically how they are formed, how much has been preserved, and their contribution to the broader field of archaeology. This paper discusses issues of preservation, context, and formation processes using misunderstandings of the Pompeii premise in underwater research as a conceptual guide. Ultimately acknowledging that, just as on land, archaeological sites underwater are diverse and unique, with site-specific pre- and post-depositional transformations. Different sites supplement each other, and the unique preservation underwater makes them a particularly valuable complement to the terrestrial record and a vital part of world archaeology.
... The submerged cultural potential of this northern region has been explored first in the Cootamundra Shoals, c. 240 km offshore Darwin (Flemming 1986), and further south in the Dampier Archipelago region (e.g. Ward et al. 2013;Ward, Larcombe, and Veth 2015). This led to more dedicated research in the culturally rich and ancient Dampier Archipelago region and reports of the first in situ marine prehistoric archaeological sites (Dortch et al. 2019;Benjamin et al. 2020). The taphonomy of these finds has, however, come under close scrutiny, with the alternative explanation put forward that they represent reworked and ponded stone artefacts and hence present a different interpretation of past cultural activity . ...
Article
Full-text available
The potential of submerged palaeolandscapes to address questions about global migrations, broad-scale climate and landscape change and human response to this has, to date, been concentrated in the Northern Hemisphere. The Southern Hemisphere has less land, more water and water barriers, higher floral and faunal endemicity and lower population but with indigenous populations that have maintained a connection with coastal and offshore landscapes for at least 40,000 years in Australasia and almost 170,000 years in South Africa. We provide an overview of current knowledge in South America, Southern Africa and Australasia and explore how new palaeogeographic and palaeoecological research, alongside related coastal archaeology, is helping to map out future directions for submerged cultural landscape research in these regions. A common theme across is the need to raise awareness of submerged cultural resources and indigenous knowledge of these as well as the multi-disciplinary approach needed to understand the unique landscapes in which they are preserved.
Article
Over the last 50 years, significant maritime archaeological research, including the excavation of shipwrecks and underwater cultural heritage projects (mainly site inspections and monitoring activities) occurred in Queensland. There is no doubt that there are exceptional historic ship and aircraft sites along the Queensland coastline. Queensland has been the home of an active Maritime Archaeology Department at the Queensland Museum, one of the most successful active avocational groups (the Maritime Archaeology Association of Queensland), and a decade-long university program at James Cook University. However, it seems that after the large-scale excavations of HMS "Pandora" (sunk in 1791), maritime archaeological research, educational outreach, public interest, and eventually, professional positions, dwindled. This paper reviews the past 50 years of shipwreck research, highlighting where and when approaches failed before discussing the future and what lies ahead for maritime archaeological research in Queensland.
Article
There are distinct bodies of cultural knowledge attached to the sea. In this paper we orient the focus towards the nature and extent of cultural framings of sea territories, as inclusive of submerged landscapes, for Indigenous maritime peoples in northern Australia. This approach is distinguished by a pluralist methodology and reorients the primal focus of a human geography and broader geographical scholarship concerning submerged landscapes to begin with an Indigenous perspective. Engaging ethnographic accounts of Indigenous Australian knowledges of sea Country, as inclusive of ancient pre‐inundation landscapes that lie out‐of‐sight on Australia’s continental shelves, highlights the potential for a more expansive vision of human connections to the past and present continental landmass of Australia. Indigenous oral traditions, Dreaming Ancestor narratives and songlines provide extensive detail to assist in understanding these parts of the greater Australian landmass and in this paper are brought into relation with recent sea floor mapping efforts which operate to draw back the water and reveal commensurable geographies upon which to envision possibilities for socialised realms of human emplacement. Both bodies of knowledge generate information of submerged landscapes that call for an expansion of thinking on where the land ends and the sea begins and how submerged terrestrial landscapes are understood across cultures as part of human geography. The approach outlined here calls for a habit of bringing principled systems of understanding to stand together as part of an explanatory schema for a world populated by and yet differentially known by people.
Chapter
Full-text available
Book
Full-text available
This open access volume provides for the first time a comprehensive description and scientific evaluation of underwater archaeological finds referring to human occupation of the continental shelf around the coastlines of Europe and the Mediterranean when sea levels were lower than present. These are the largest body of underwater finds worldwide, amounting to over 2500 find spots, ranging from individual stone tools to underwater villages with unique conditions of preservation. The material reviewed here ranges in date from the Lower Palaeolithic period to the Bronze Age and covers 20 countries bordering all the major marine basins from the Atlantic coasts of Ireland and Norway to the Black Sea, and from the western Baltic to the eastern Mediterranean. The finds from each country are presented in their archaeological context, with information on the history of discovery, conditions of preservation and visibility, their relationship to regional changes in sea-level and coastal geomorphology, and the institutional arrangements for their investigation and protection. Editorial introductions summarise the findings from each of the major marine basins. There is also a final section with extensive discussion of the historical background and the legal and regulatory frameworks that inform the management of the underwater cultural heritage and collaboration between offshore industries, archaeologists and government agencies. The volume is based on the work of COST Action TD0902 SPLASHCOS, a multi-disciplinary and multi-national research network supported by the EU-funded COST organisation (European Cooperation in Science and Technology). The primary readership is research and professional archaeologists, marine and Quaternary scientists, cultural-heritage managers, commercial and governmental organisations, policy makers, and all those with an interest in the sea floor of the continental shelf and the human impact of changes in climate, sea-level and coastal geomorphology.
Chapter
Full-text available
This section deals with some of the largest concentrations of underwater sites in Europe. Because of the complex history of isostatic rebound and periodic damming back of the Baltic Sea associated with the retreat of the Scandinavian ice sheet, these chapters also present a wide range of preservation conditions and palaeoenvironmental changes, ranging from shorelines that have been lifted clear of sea-level rise in Norway and parts of Sweden but without organic preservation, to sinking shorelines on the Baltic coastlines of Denmark and Germany with little overburden of marine sediments and spectacular organic preservation, to open-coast conditions along the North Sea shorelines of Denmark and Germany where sites are mostly buried under thick marine deposits. Many sites in Denmark and Germany have benefited from systematic excavation, demonstrating that the majority of finds and features have been recovered from marine refuse areas in shallow water along the shoreline and that the adjacent settlement areas originally located on dry land have been largely eroded away or badly disturbed by subsequent sea-level rise. The abundance of finds and excellent preservation in many cases reflect the high marine productivity of the Littorina Sea and the concentration of settlements on the shoreline with a heavy emphasis on marine resources, the presence of coastal topography conducive to trapping of fish and sea mammals, the deposition of cultural material directly into shallow water with fine-grained sediments where items were quickly buried and protected from bacterial attack or marine erosion and the occurrence of settlements in relatively shallow water easily accessible to diver investigation. There is considerable scope for new investigations and new discoveries, not only in Denmark and Germany but also in Norway, where study of the numerous Stone Age sites on uplifted shorelines has overshadowed investigation of the many regions with inundated coastlines, in Sweden, where the Early Mesolithic period, almost non-existent on the submerged shorelines of Denmark and Germany, is represented by an extensive and relatively shallow submerged landscape with excellent preservation conditions, and further east in the Baltic on the coastlines of Poland, Lithuania and Latvia, with offshore conditions similar to those of Germany and Denmark and the likelihood that numerous submerged sites of high scientific value await discovery. This chapter is publicly available at https://doi.org/10.1007/978-3-030-37367-2_2
Chapter
Full-text available
Interactions between commercial and industrial exploitation of the seabed and archaeological and scientific investigation have been at the heart of developments in the understanding of Europe’s submerged landscapes and prehistory since at least the early twentieth century. This introduction considers some of the ways in which that relationship has evolved since that time, including the adoption of international laws under the aegis of United Nations Conventions, the development of close relationships between Dutch fishermen operating beam-trawl fishing nets in the North Sea and a network of private collectors specialising in Pleistocene fossils and artefacts, the imposition of European Union regulations on offshore industrial projects to include monitoring of underwater archaeology and palaeoenvironments and most recently the incorporation of seabed mapping and underwater cultural heritage in the European Union’s 2020 Blue Growth agenda. These developments have played an important role in the growth of knowledge about the underwater cultural heritage notwithstanding the potentially destructive effects of offshore industrial activity. The impact of economic growth and industrial exploitation in the coastal zone, coupled with sea-level rise, is likely to intensify the threats to the underwater cultural heritage in the coming decades, posing new challenges as well as opportunities in the further development of relationships between industrial operators, government agencies and scientific and archaeological researchers.
Chapter
Full-text available
This chapter outlines the international legal framework concerning the protection of submerged prehistoric resources from human activities that may cause inadvertent harm. It focuses on what is the core question from a legal perspective: to what extent does a coastal State have the jurisdictional power under international law to regulate such activities in its offshore waters with a view to protecting material of archaeological significance? As will become clear, this question is a complex one, requiring reference to three sources of international law: (i) general principles of international law, (ii) the 1982 UN Convention on the Law of the Sea and (iii) the 2001 UNESCO Convention on the Protection of the Underwater Cultural Heritage. Questions relating to the way that the relevant treaties define underwater cultural heritage, and the potential implications this may have for submerged prehistoric archaeology, though interesting, fall outside the scope of this chapter.
Chapter
Full-text available
This chapter focuses on the legal and regulatory requirements imposed on development projects in the marine offshore areas of England, in particular the requirement to collaborate with professional archaeologists in completing Environmental Impact Assessments that include underwater cultural heritage. This chapter explains the role of Historic England in providing independent advice to developers and regulatory bodies for all aspects of the historic environment. It explains how this advice informs the preparation and approval of planning applications for offshore developments, the implementation of effective survey campaigns that include archaeological and palaeoenvironmental objectives as part of the planning application, mitigation of potential damage, protection of new discoveries and the delivery of training programmes for government administrators, archaeological consultants and industrial partners. The benefits and results of this approach are presented in the context of offshore developments such as wind farms, the laying of electricity interconnector cables, the extraction of gravel deposits and the dredging of shipping channels into ports.
Article
Full-text available
Since the collapse of the Clovis-first model of the peopling of the Americas some 30 years ago, there has been growing interest in the Pacific Coast as a potential early human dispersal corridor. With postglacial eustatic sea level rise inundating most New World paleoshorelines older than ~7000 years, however, locating terminal Pleistocene sites along modern coastlines is challenging. Using the distribution and archaeology of subaerial Paleocoastal archaeological sites on California's Northern Channel Islands as a guide, we developed a Geographic Information Systems (GIS) predictive model to locate and map submerged high probability landforms, which might contain Paleocoastal sites. Our results illustrate how archaeologists can narrow targets in their search for evidence of the first Americans along submerged Pacific Coast paleoshorelines.
Article
Full-text available
The early occupation of America The Cooper's Ferry archaeological site in western North America has provided evidence for the pattern and time course of the early peopling of the Americas. Davis et al. describe new evidence of human activity from this site, including stemmed projectile points. Radiocarbon dating and Bayesian analysis indicate an age between 16,560 and 15,280 years before present. Humans therefore arrived in the Americas before an inland ice-free corridor had opened, so a Pacific coastal route was the probable entry route. The stemmed projectile points closely resemble those found in Upper Paleolithic Japan, also supporting the hypothesis of a coastal route. Science , this issue p. 891
Article
Stone artifacts recently identified in the intertidal zone at Dolphin Island, Dampier Archipelago, suggest Aboriginal Australian occupation before inundation from early to mid-Holocene sea-level rise. If these artifacts do pre-date inundation, they would be the first evidence for a submerged coastal site on the Dampier Archipelago—substantiating persistent Aboriginal use throughout major environmental changes. The find provides supplementary evidence for Early Holocene settlement patterns, site organization and stone artifact production as documented on the outer islands of the Archipelago. It could be argued that the artifacts in the intertidal zone were washed downslope from nearby terrestrial settings where natural stone sources have been quarried. However, chemical analysis of cobbles and artifacts from both the hillslope and intertidal locations indicate that these two source materials are not the same. Furthermore, there is no evidence for reworking: artifact edges are unrounded. The distribution of suitable tool stone across the tidal flats is extensive (as it is elsewhere across this volcanic landscape). Further investigations, including refitting and taphonomic, geomorphological, and sedimentological studies are underway to test the alternative explanations and to provide more detailed context.