Assessment of quartz in Sydney’s rock formations associated with recent engineering infrastructure projects; implications for workplace health and safety

Och, David J.1&2 and Cole, Kate3

1Adj. Assoc. Professor, University of New South Wales, Kensington,, 2Senior Principal Engineering Geoscientist, WSP Australia Pty Ltd, Sydney, NSW., 3A/Director Health & Occupational Hygiene, Sydney Metro, Sydney NSW 2000.

Sydney Metro is Australia’s largest public transport project. This new standalone railway will deliver 31 metro stations and more than 66 kilometres of new metro rail, revolutionising the way Australia’s biggest city travels. As part of the development of Sydney Metro, both historical and current ground data is being collected and compiled to better understand the proportion of quartz that will be encountered along the proposed tunnel alignments.

Tunnelling and excavation work being undertaken in quartz-containing rock creates a dust comprising of a very fine shard particulate due to the impact and grinding reduction of the tunnelling and excavation equipment (i.e. road header, rock breaker and Tunnel Boring Machine (TBM)). As the dominant hard-mineral component is quartz, at the point of excavation, fine particulates of respirable crystalline silica dust can be released which can be suspended into the atmosphere. These fine particulates can be suspended in the air column and may create an occupational health risk resulting in illness and disease such as silicosis and lung cancer.

Geotechnical investigations across Sydney Metro have targeted the Triassic Wianamatta Group, Mittagong Formation and Hawkesbury Sandstone of the Sydney Basin. Quartz was found to make up to 90% (avg. 72%) of the grains in Hawkesbury Sandstone with the other components being siderite and clays. The hard-mineral component in Hawkesbury Sandstone is dominantly quartz detritus with other mineral components being moderate to weak in strength and less susceptible to micro-fracturing due to the mineral’s crystal framework.

This paper will provide an overview of the predicted proportion of quartz along the Triassic Wianamatta Group, Mittagong Formation and Hawkesbury Sandstone formations in the Sydney Basin.


Assoc. Prof David Och is a Fellow of the Geological Society of Australia and a 2019 Winston Churchill Fellow working as a Senior Principal Engineering Geoscientist with WSP Australia.  David’s expertise is demonstrated by performing key roles on large infrastructure projects including Sydney Metro City & Southwest, Sydney Metro West.

Rock Stress in Tasmania

Hills, Peter1

1Pitt & Sherry Operations Pty Ltd, Hobart, Australia

Rock stress can have a profound impact on underground excavations in mining and civil space. Rock stress measurement was first undertaken in Tasmania to inform the design of excavations for the Poatina Power Station in 1960 and led to an innovative trapezoidal section to the crown of station cavern in near horizontal Permian sediments. Since that time stress measurements have been undertaken at 15 sites across western and northern Tasmania and this has provided a database, which in addition to assisting in the management of individual assets, has allowed a number of more general observations to be made[1].

Rock stress is the result of tectonic forces and overburden pressures. The former creates a background stress condition that is widely evident within a given geological terrane. All Tasmania’s stress measurements have been undertaken in the Western Tasmanian Terrane and indicate a major principal stress (σ1) orientation striking WSW-ENE which is an anticlockwise rotation from that observed in Victoria where measurements have largely been undertaken in rocks of the Lachlan Fold Belt. This is thought to reflect upon the key location of Tasmania during the accretion of Gondwanaland and the eventual separation from Antarctica. The latter typically reflects depth below surface with the magnitude of the stress generally increasing with depth. In most cases in Australia, and it is the case in Tasmania, that the orientation of the minor principal stress (σ3) steep and consequently the intermediate principal stress (σ2) is relatively flat. The general stress condition in Tasmania can be described in terms of orientation and magnitude as follows:

  • σ1; 23°/261°         (0.039 x depth) + 10.7 MPa
  • σ2; 15°/164°         035 x depth MPa
  • σ3; 62°/043°         023 x depth MPa

Local geological and geomorphological setting can significantly impact the stress locally. At Cethana Power Station the stress measured in the crown of the station prior to excavation of the cavern was found to be four times the expected magnitude. However it is readily explained by its location in the steep sided Forth River Gorge which was rapidly cut through the Fossey Mountains during relatively recent glaciation. At Hellyer Mine the magnitude of the measured stress is more-or-less consistent with the depth of the measurements beneath the Waratah Plateau, but the orientation of the shallower measurements is strongly influenced by the orientation of the nearby Southwell River Gorge. Renison Mine provides a further example where the orientation of the stress field is strongly influenced by the orientation of the Federal-Bassett Fault and the associated Pine Hill Horst.

Stress measurement in Tasmania has been undertaken for direct engineering construction as in the case of the John Butters Tunnel where is was specifically directed at determination of the length of tunnel lining required. It has also been used extensively to understand the orientation and magnitude of the stress condition in deep mines at Mount Lyell, Renison, Rosebery and Beaconsfield and assist with excavation, and particularly, stope design. Stress change due to extraction has also been monitored at those mines as well as Dolphin and Cleveland.

[1] Hills, P B, 2020. Tasmanian rock stress, Australian Geomechanics, 55(1), 77-111.


Originally graduating as a geologist at UTAS in the early 1980’s, Peter undertook further studies in rock mechanics and Commenced transitioning to mining geomechanics over the following 10-15 years. He now consults in the field geomechanics after a 30 year career in underground mines in Tasmania and Papua New Guinea.

Structural analysis of wavecut platform structures, Shellharbour, New South Wales

Lennox, Paul1, Goff, James1,2, Edwards, David1 & Coates, Ashlie1,3

1The University of New South Wales, Sydney, Sydney, Australia, 2School of Ocean and Earth Sciences, University of Southampton, Southampton, UK, 3Department of Industry, Innovation and Science, Canberra, Australia

Bedrock-sculptured features on the wavecut platform south of Shellharbour, NSW have been structurally assessed as some consider they have been formed by a southeast to northwest directed (mega)tsunami. These features have formed within the up to 120 m thick, Late Permian Upper Bumbo Latite unit of the Gerringong Volcanics.

The research focused on two subareas; the southern side of Bass Point (Bushranger Bay) where apparent s-forms due to tsunami-sculptured rock platforms expose so called “muschelbruche” having a relief of decimeters to several metres and Atcheson Rock one kilometre to the southwest. At Atcheson Rock the orientation of the channel and the so called vortex nearby; a semi-circular depression with raised internal feature were assessed with respect to the orientation of the local jointing, layering and flow foliation.

The muschelbruche vary in size from a few metres across to tens of metres across. The orientation of bedrock-sculptured features south of Bushranger Bay display various relationships to the flow foliation and jointing. In some cases the sides or raised back of the “muschelbruche” are aligned subparallel to either the flow foliation, joint strike or some combination. The long axis of the muschelbruche are not predominantly aligned northwest-southeast as expected if they were generated by a single (mega)tsunami from the southeast.

The cross-cutting ? Cenozoic dyke at Atcheson Rock weathers to a channel in the basalt and is along strike of a similar eroded dyke and channel on the coastline south of Bushranger Bay. The strike of this dyke is subparallel to the strike of flow foliation and the dominant jointing in the basalt.   Some assert that the canyon containing this dyke is not structurally controlled. Normal coastal erosion of this dyke has caused Atcheson Rock headland to almost be cut off from the mainland. The Cenozoic dyke has intruded subparallel to the dominant northeast to southwest striking joints and flow foliation in this region. Some consider that the vortex pit on the south side of Atcheson Rock was eroded by a tsunami from the southeast causing a whirlpool over 10m wide and characterised by a central plug of bedrock 5m high. The presence of a  rounded but north-south elongated depression in the basalt (“vortex”) can be explained by weathering related to the pervasive localised NNE-SSW and ~ NW-SE striking jointing, gently southeast-dipping layering in this locality but not the NE-SW or NW-SE striking flow foliation.

The orientation, scale and development of various coastal erosional features including channels, scour-like structures (muschelbruche) and depressions near Shellharbour, NSW can be explained through normal coastal processes without the necessity to invoke (mega)tsunami-related scouring from the southeast.


Paul Lennox has over thirty years experience teaching undergraduates and undertaking field-based map-based research in Eastern Australia. He has published papers on deformation in granites and their enclosing country rocks in the Lachlan Orogen   and  the  tectonic development of the Southern New England Orogen.

Displacement of mega boulders across coastal rock platforms south of Sydney, Australia, by big storm waves generated by East Coast Lows

Bann, Dr Glen1, Lau, Dr Annie2

1University of Wollongong, Wollongong, Australia, 2University of Queensland, St Lucia, Australia

This research reports on a number of large boulders, or blocks, displaced across coastal rock platforms between Kiama and Snapper Point, south of Sydney, Australia. All boulders are derived from underlying fine to coarse-grained sandstones. A boulder described previously (Bann 2012), weighing  ~10 tonnes, had been transported 42m across the rock platform and rotated 110o from its original in-situ position. Before and after photos show evidence of more recent movements, ~1m sideways and rotating it back 50o in 2016, coinciding with an ‘East Coast Low’ that impacted the whole NSW shoreline in June 2016. It was moved again in 2017 a further 30o and ~30cm sideways with one end lifted ~15cm to rest it on a ledge, then in July 2020 following another East Coast Low, another 10o rotation and ~20cm movement occurred off and away from the ledge. Three smaller boulders nearby of ~1 to 2 tonne each also get tossed around with these events. Two other much larger boulders weighing ~100 and ~180 tonnes also reported movement previously from 2016 (Bann et al. 2018), to the north of Jervis Bay moved ~10m upslope and rotated ~15 degrees for the smaller one and a few metres up slope for the larger one. The smaller boulder has also been flipped over at some stage given by reversed cross-bedding and inverted sedimentary grading. Slight movement of these two has also been detected since 2016. Examples from other localities include the placement of blocks weighing a few tonnes onto an elevated rock platform, then removed at a later date, a block weighing about 1 tonne lifted and moved onto a higher ledge on a cliff about 20m from the platform sea edge and 5m elevation, and a boulder weighing about 10 tonnes on a rock platform lifted with tree logs emplaced and wedged firmly beneath it before settling on them. Satellite imagery was used to identify approximate movement times and then correlated with storm wave data obtained from wave buoys located offshore at Port Kembla and Batemans Bay. This provided timing in addition to wave heights, allowing approximate wave height constraints for wave height formulas. None of the boulders appear to have been pushed or slid across the actual platform, rather, all boulders show evidence of being lifted during displacement, hence by saltation or suspension. Calculations suggest that waves of considerable height and wave length are required to displace these boulders across the platform, either in the one event, or in incremental steps, larger than the actual heights from the wave buoy data show. The findings indicate that large storm waves are capable of shifting very large boulders, such as those generated by East Coast Lows which frequent the coast during the winter months. Although a few previous authors use boulder movements as evidence that the east coast experiences occasional large tsunamis, tsunamis are not required to transport these large boulders on coastal platforms. Clearly, coastal lowland environments south of Sydney are prone to occasional very large storm waves and needs to be prioritized in coastal hazard and sustainable management plans, particularly applicable to climate change and future sea level rise.


This is ongoing research which originated in 2012 when looking for intrusions and tuffs along the coast –  noticing big boulders in places they shouldn’t be, then, noticing these boulders are moving around regularly, without leaving scrape marks, meaning, the boulders are effectively floating…

Real-time tracking of the 2019 pumice raft in the southwest Pacific

Jutzeler, Dr Martin1, Van Sebille, Dr Erik2, Marsh, Prof Robert3

1Centre for Ore Deposit and Earth Sciences, University of Tasmania, Hobart, Australia, 2Institute for Marine and Atmospheric research, Utrecht University, Utrecht, Netherlands, 3School of Ocean and Earth Science, University of Southampton, Southampton, United Kingdom

Pumice rafts can be hazardous to maritime traffic due to their ability to clog engine water intakes, block harbours and divert maritime traffic for weeks. On 7 August 2019, a 195 km2 pumice raft was produced at an unnamed submarine volcano in the Tonga Islands (Southwest Pacific Ocean). The raft quickly expanded in size and got segmented into multiple smaller rafts that reached the Lau and Fiji Islands over the following weeks. Yachts that crossed the raft as early as two days post-eruption sent an alert to the Rescue and Co-Operation Centre New Zealand (RCCNZ), the Maritime Safety Authority of Fiji, who relayed us the information. The coupling of real-time satellite observations with weather reports and oceanographic Lagrangian simulations allowed near-real time forecasting of raft dispersal. The abundance of satellite images allowed us to contrast virtual particle tracking methods with ocean model currents to explore the relative influence of surface currents, wind, and wave action on pumice flotsam dispersal. We produced bi-weekly hazard maps to RCCNZ and key local individuals for dissemination to the yachting, shipping and fishing communities via social media and word of mouth. This strategy successfully prevented further vessels from encountering the pumice raft, and facilitated contact with sailing crews for information on the raft and samples. The dispersal models built for this pumice raft can be used for global maritime hazard mitigation.


Martin Jutzeler is a Senior Research Fellow at the University of Tasmania. Martin uses trans-disciplinary strategies to tackle frontier questions in ancient and modern marine volcanism. His research includes volcanic architecture, submarine volcanism, and dispersal of volcanic material in the oceans.

Tasmanian landslide fatalities and some implications for landslide risk

Roberts, Dr  Nicholas1

1Mineral Resources Tasmania, Rosny Park, Australia

Tasmania experiences frequent, diverse landslides ranging from extremely slow failures that gradually damage structures to extremely fast ones capable of claiming lives. Their impacts in recent decades might give the false impression that Tasmanian landslides threaten only infrastructure and not lives. However, life-loss risk is poorly constrained – and generally underappreciated – because the state’s landslide fatalities have not been fully inventoried. Details from a growing catalogue of Tasmanian landslide fatalities (excluding underground failures) show that deaths, although sporadic, do occur and were surprisingly common during the nineteenth and earliest twentieth centuries. Landslides have injured at least 23 people in the last 100 years. Fourteen of the injuries and four of the five confirmed landslide fatalities from that period occurred in 2001 when a road shoulder collapsed under the weight of a bus. Although sometimes conflated with mass movement, the rainfall-triggered burst of a rock-filled concrete dam above Derby in 1929 that claimed 14 lives was unrelated to landsliding. Tasmania’s at least eight landslide fatalities prior to 1920 provide additional insight into life-loss potential. Deadly landslides occurred during construction of Port Arthur’s Convict Church (ca. 1836), quarrying in Hobart (1848) and Queenstown (1905), railway expansion through Kelly Basin (1889), and open-pit mining at Mount Bischoff (1900). All but the 1848 failure, which killed four, were single-fatality events, providing a stronger historical basis for calibrating quantitative estimates of individual risk compared to group risk. Commonalities between these early events highlight settings and processes of particular concern as well as a population of elevated exposure. Each failure affected very recently excavated slopes, involved extremely rapid sliding and flow of soil-strength materials, and exclusively claimed lives of workers (convicts before 1850, employees thereafter). Coronial Inquests into each of the events from 1848 to 1905 provide more detail about failure metrics and behaviour than is commonly available for landslides of that era. All eight deaths were ruled accidental despite knowledge that many sites were failure-prone. Several close calls prior to 1920 highlight risks of members of the public being engulfed by debris flows or impacted by landsides at home. However, fatalities from hazardous phenomena that commonly influence (bushfires) or accompany landslides (flash flooding) remain more common. Although Tasmanian landslide fatalities superficially appear to have decreased, determining and explaining trends in these records is complicated by data sparsity. Advances over the past century in engineering, workplace safety, and regulation undoubtedly affect risk levels, but evaluating other possible influences such as long-term drought-rain cycles requires further work. Decreases in per-capita landslide fatalities elsewhere are attributed to progressive improvements (e.g. British Columbia) or drastic reforms (e.g. Hong Kong) in local risk management; circumstances in Tasmania are closer to the former, although the low number of deaths complicates comparisons. Notwithstanding these challenges, it is noteworthy that fatal landslide in Tasmania show several commonalities, are generally underestimated, and are possible in the future, particularly as population and development increase.


Nick is a natural hazards geologist and Quaternary geoscientist in the Geological Survey Branch of Mineral Resources Tasmania. A large part of his work centres on characterizing landslides and their impacts across Tasmania through a combination of field investigations, diverse remote sensing techniques, analytical tools, and reviewing historical documents.

About the GSA

The Geological Society of Australia was established as a non-profit organisation in 1952 to promote, advance and support Earth sciences in Australia.

As a broadly based professional society that aims to represent all Earth Science disciplines, the GSA attracts a wide diversity of members working in a similarly broad range of industries.