Australian Geographer

ISSN: 0004-9182 (Print) 1465-3311 (Online) Journal homepage: https://www.tandfonline.com/loi/cage20

A rockfall avalanche in a sandstone landscape,

Nattai North, NSW

David M. Cunningham

To cite this article: David M. Cunningham (1988) A rockfall avalanche in a sandstone landscape,
Nattai North, NSW, Australian Geographer, 19:2, 221-229, DOI: 10.1080/00049188808702961

To link to this article: https://doi.org/10.1080/00049188808702961

Published online: 24 Feb 2007.

Submit your article to this journal

Article views: 59

View related articles

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=cage20
 A ROCKFALL AVALANCHE IN A SANDSTONE LANDSCAPE, NATTAI
NORTH, NSW
DAVID M. CUNNINGHAM*

SUMMARY: Rockfall avalanches are commonly associated with the alpine regions of
Europe, South America and north-western Canada, but modern examples have only
been reported very recently in Australia (Pells et al. 1987). The Nattai North rockfall
avalanche is located on the Burragorang Walls escarpment in the sandstone landscape
of the Sydney Basin. The volume of rock involved in the failure had sufficient
magnitude to enable the resulting mass of debris to flow in the manner of a semivis-
cous fluid. The conventional models of rockslope evolution, involving undercutting
followed by blockfalls, do not apply at this site. Indeed these models do not apply to
most of the large-scale rock collapses in the Sydney Basin. All such rockfalls have
occurred in the vicinity of underground coal mines. Coal mining has affected the
stability of nearby escarpments by altering stress distributions within the rock mass.
The subsequent failures are typically larger and of a different form than those
occurring naturally.

At Nattai North, some 80 km south-west of tion to the great age of many eastern Australian
Sydney (Figure 1), a series of rockfall avalanches landscapes and showed that they evolved far
occurred from the Burragorang Walls escarpment. more slowly than most northern hemisphere coun-
These failures have exposed an area of some 210 terparts. Yet very little is known of the geomorphic
m in height by 775 m in length of the Triassic evolution of this region and there is evidence to
sandstone and interbedded siltstones of the Nar- suggest that rock avalanching has been at times
rabeen Group. A mass of approximately 14 million far more widespread than it is now (Walker 1963;
ma of angular debris has traversed the footslope Young 1977).
for a distance of up to 1 km, reaching the shores This rockfall avalanche exhibits many of the
of Lake Burragorang, part of Sydney's Warragam- characteristics reported from some of the
ba Dam water storage (Plate 1). well-known catastrophic avalanches overseas
The first major collapse of the cliff occurred in (Heim 1932; Shreve 1968; Hsu 1975; Eisbacher
1965, though this was preceded by numerous 1979; Davies 1982; Voight era/. 1983; Eisbacher
smaller rockfalls. Metropolitan Water Sewerage and Clague 1984). Although small by world stan-
and Drainage Board (MWSDB) survey records indi-
cate a failure mass of about 3 million tonnes. It
was not until the mid to late 1970s that the next
major collapse occurred, involving the failure of an . ^Newcastle
estimated 25 million tonnes of rock from the es-
carpment. It was this failure that caused the main My-;
rock avalanche flow which reached the shores of
Lake Burragorang. Further rockfalls occurred in Mr-:
1983 and 1984, bringing the total failure mass to
its present 30 million tonnes.
Warragamba /
This twenty-year history of failures reflects Dam /
changes in the stress conditions within the cliff

WI
face. Coal mining operations commenced beneath
this section of the escarpment in the late 1950s, J» Nattai
with pillar recovery being carried out beyond the f^ North
base of the cliff face, but the exact relationship ' • • • • • . •

between subsidence and stress changes is not Wollongong

obvious at this site. Observations from clifflines &
within this region (Cunningham 1986) suggest that
collapse triggered by mining represents a much ':'•• ' • " • ' '

accelerated process of present-day slope degra- Nowra •-'fc.;.;

dation for the sandstone landscapes of the Syd-
;
ney Basin. Moreover, Young (1983a) drew atten- 0 50 100km
1 1
1 1 1 ' mJ

* School of Earth Sciences, Macquarie University, NSW 2109,

Australia. Figure 1: Location of Nattai North.
Australian Geographer 19 (2), November 1988 221
 Plate 1: The escarpment failure. Note the flow like appearance of the deposit. Photo R. W. Young.

dards, the rockfall avalanche at Nattai North is by joint orientations consist of a set essentially per-
far the largest contemporary failure in the Sydney pendicular to the escarpment face and a second
Basin and is probably one of the largest in Austra- set parallel to the face. These lineations corre-
lia. This paper outlines two important slope pro- spond with the tectonic stress directions recog-
cesses which have not been recorded in detail for nised by Gray (1982) for the Burragorang region.
this geomorphological setting. The proposed Joint continuity is highly variable, though the lon-
mechanisms of cliff retreat for this area are exam- gest joint plane observed was 30 m on the actual
ined with respect to the geology, to form of the failure surface. Mean fracture spacing is 1.5 m.
escarpment and to its proximity to mine workings. Given the approximately horizontal nature of the
Morphological characteristics of the debris depos- bedding, combined with discontinuity orientations,
it are then interpreted in order to define the me- the rock mass is subdivided into prismatic blocks.
chanical processes which may have operated dur-
ing the mass movement.
FAILURE MECHANISMS

LITHOLOGY AND STRUCTURE Cliff failure in the sandstones of the Sydney

Basin has generally been attributed to basal ero-
At Nattai North the Narrabeen Group consists of sion of interbedded siltstones which are suscepti-
four formations. These are, in stratigraphic order, ble to weathering (Waterhouse ef al. 1951; Lang-
the Wombarra Formation (siltstones and clay- ford-Smith 1976; Evans 1981; Kotze 1986). As
stones), the Scarborough Sandstone, the Bulgo these weaker strata are rapidly weathered, their
Sandstone and the Bald Hill Claystone. These support of overlying units is quickly lost, resulting
beds have a mean easterly dip of 5 degrees (i.e. in an increase in the tensile stress field in the
into the escarpment face). Underlying the Narra- overlying rock mass. Evans (1981) suggests that a
been Group are the Permian deposits of the Illa- differential settling of this mass coincides with this
warra Coal Measures. The uppermost unit of the loss of strength in the underlying material. As a
coal measures, the Bulli Coal seam, is currently result, the centre of gravity shifts until it subtends
being mined, whilst the lowest, the split seam of an angle beyond the base, after which toppling
the Tongarra Coal has been mined. The Bulli Coal occurs. Yet, while the process of toppling certain-
seam outcrops about 280 m below the top of the ly takes place in the Sydney Basin, it should not
escarpment and the Tongarra Seam outcrops be assumed that it is the dominant rockslope
some 70 m lower. process (Young 1983b).
A survey of rock fractures revealed two major The concept of basal weathering or undercutting
lineations and two secondary lineations. The major as the main mechanism of cliff collapse for this
222 Australian Geographer 19 (2), November 1988
 Total extraction / Pillar recovery

Lake '.
Burragorang- • ;.'..•'
\== i first
fi
— workings

Burragorang lookout
t

Areas of distal mounding

Figure 2: Avalanche deposit and escarpment in relation to colliery holdings.

region is difficult to reconcile with cliff morphology and are now tilting back into the face. The largest
at many sites in the Sydney region. For example, of these is about 90 m in height with an estimated
major failures such as the Dogface Rockfall (Ka- mass of 20,000 tonnes. These slabs have become
toomba) and the Dombarton Fall (Illawarra) have detached due to extension of tensile cracks which
not been due to undercutting. On the contrary they formed directly behind the escarpment in the
are buttressed at the base. months prior to the main failure (R. Bradley per-
sonal communication 1986).
Subsidence Tension cracks are a common feature in the
The Nattai North failure appears to have result- Sydney Basin where mine subsidence occurs
ed from basal shear failure followed by reverse (Kapp 1985). Pillar recovery directly beneath and
toppling. In the centre of the face large slabs of in front of the escarpment has altered the natural
rock are detached from the main mass (Plate 2) stress distribution within the face as the underly-
Australian Geographer 19 (2), November 1988 223
ing beds collapse into the extracted seam, and feature can be seen in Plate 2 immediately left of
appears to be the most significant factor leading the rock pillars.
to the development of massive failure. Pillar recov-
ery in 1945 to the south of the present mine portal Other mechanisms
(Figure 2) did not extend beyond the escarpment While subsidence clearly affects stress distribu-
line; and the supporting footslope was not affected tions within the rock mass, influences of rainfall
by mine subsidence sufficiently to cause a signifi- (cleft water pressures), seismicity and valley tec-
cant increase in horizontal tensile stress in the tonics cannot be overlooked. Unfortunately, be-
overlying rock mass. Though blockfalls are pres- cause of inaccurate rainfall records and especially
ent in this southern area, a major failure is unlikely because of the absence of exact dates for failure
since the footslope still provides adequate sup- episodes, no correlation could be made, and com-
port. ments are thus limited to general observations.
Figure 3 shows the stages involved in the col- There is often a time delay between rainfall
lapse of the escarpment. Initially the buttress sup- periods and failure. In the Sydney Basin this delay
port fails as pillar recovery progresses. Removal can vary from one day up to several months due to
of the support releases, and effectively reduces, the highly variable nature of rock permeability
the horizontal pressure which in turn introduces a (Evans 1981). Waterhouse ef al. (1951: 79) pro-
tension zone into the upper part of the face. The vide evidence indicating that aquifers within rocks
sandstones in the upper part of the cliff face, in in this region are freely drained following rainfall
fact throughout most of the Sydney Basin (Pells due to the 'super-capillary size' of discontinuities.
1985), are considerably weaker under tension than This conjecture is supported by the fact that no
the normal compressive state, and tension cracks seepage can be observed on the exposed rock
develop immediately behind the escarpment. As a face at Nattai North. An absence of seepage may
consequence of this strength difference, the effec- be a consequence of the underlying coal mine, as
tive cohesion across joints is lost. Basal shear Angel and Hayes (1980) note the potential for
failure occurs as the overlying rock mass is de- underground accessways to drain aquifers.
tached from the cliff face. Seismic records are available for the Burrago-
Using formulae derived by Goodman and Bray rang Valley, but once again an absence of accu-
(1977), backward toppling could be expected to rate dates for the failures precluded any meaning-
occur once the declivity of the rock pillars ful correlation. For the adjacent escarpment,
reaches 13 degrees. Pells (personal communica- above the Nattai North Colliery portal, a more
tion 1987) has termed this a 'ski-jump' failure accurate record of small-scale rockfalls exists.
whereby the rock mass tilts backwards, failing on Kotze (1980) indicates that these small rockfalls
a basal shear plane resembling a ski jump. This frequently occurred during periods of little or no
seismic activity.

Plate 2: Rock towers, now tilting into the face. Older basal failure surface is visible to the left.
224 Australian Geographer 19 (2), November 1988
 table fluctuations would influence stability because
(a) the base of the escarpment is situated about 250
Rnz
m above the water level.

THE ROCK AVALANCHE DEPOSITS

The deposit shown in Plate 1 does not repre-
sent a single avalanche, but rather one major fall
associated with several smaller falls. Because of
(b) this complexity, identification of mass movement
tension crack
processes is not easy. The map of the avalanche
deposit (Figure 2) shows two lobes of the deposit
known from air photos to have occurred as single
massive failures. These two episodes account for
approximately 85 per cent of the total debris
subsidence /'•. ::.:
deposit, and are thus central to the interpretation
of the mechanical processes involved during mass
transport.

Deposit morphology
(c)
& Though there were no eyewitness accounts of
the debris movement, the mechanics of movement
can be inferred from the morphological features of
the debris. There are two distinct forms within the
compression debris deposit, namely a talus slope and a rock
fractures

UiBKBUiB

new tension crack

(d)
,—

T
m
A&i'r-??

Figure 3: Stages of escarpment collapse, (a) Intact coal

seam, (b) As the seam is extracted the supporting
footslope fails, so the tensile stress is increased in the
escarpment face, (c) Further extraction has caused
shearing initiated on joint planes; as a result, additional
stress is placed on basal units, reducing strength, (d)
The remaining pillar of rock cannot be supported by the
basal material, so shear failure occurs in these units,
enabling the pillar to topple backwards.

Gray (1983) attempted to correlate seismic ac-

tivity with the added loading from the rising body
of water in the Warragamba Dam, but found that
the mass of water "had no effect on regional tec-
tonic stress distribution. Fluctuating water levels
were considered as a factor in the Vaiont rock-
slide in Italy (Pariseau and Voight 1979), triggering
alterations of stress distribution. It is uncertain
whether fluctuating water levels influence stress
distribution within the escarpments in the Nattai Plate 3: Sandstone boulder (approx. 12,000 tonnes),
region, but it is very unlikely that corresponding fractured on bedding plane after emplacement.
Australian Geographer 19 (2), November 1988 225
avalanche deposit. The former has resulted from likely that pore pressures of this part of the slope
small-scale denudational processes such as rock- were higher than elsewhere, accounting for the
falls and will not be dealt with in this paper. The limited extent of the slumping.
rock avalanche deposit is most important, not only
because of its size but also because it is a
process not previously recorded in detail for the MECHANISMS OF DEBRIS EMPLACEMENT
sandstone regions of the Sydney Basin.
Rockflow vs. rockslide
The deposit is lobate in shape, with distal Whether mass movements of this type actually
tongues of debris that mainly follow drainage flow or slide is a matter of dispute (e.g. Shreve
channels. At a distance, the surface profile of the 1968; Hsu 1975). However, several distinct mor-
debris mass has a wavy appearance resembling phological features of this avalanche debris sup-
the transverse ridges described from the Black- port the notion of a flowing mass as opposed to a
hawk failure in the USA by Shreve (1968). Block- sliding mass. Firstly, the deposit has a tongue-like
shape is essentially prismatic, being controlled by shape, and its distal lobes are confined to the
bedding planes and joints (Plate 3). Most blocks channels of drainage courses. Secondly, the de-
have volumes greater than 20 m3 and some as bris laps around a small stand of trees at the end
large as 65 m3, though there is no relationship of the deposit, leaving it jutting like a promontory
between size and orientation. Their angularity and into the blockfield. The boulders on either side are
a noted absence of surficial scouring suggests very large and if they had moved as a slide rather
that collisions between blocks may have been than a dispersive flow, they would have flattened
inelastic during the flow, despite the obvious high the trees.
energy which must have been required to transport
blocks of this size. Thirdly, the ridge pushed up at the end of the
avalanche debris would be considerably larger if
A ridge of bulldozed earth, in places up to 7 m the mass had scoured into the underlying slope.
high, flanks the peripheral margins of the flow A s a 1 m wide slice of the ridge has a volume of
deposit. This lateral ridge is composed of sedi- about 35 m3, the average depth of scour over the
ments from the underlying slope and occurs only 800 m from the base of the cliff would be less than
around the larger lobes of the deposit; in areas 4 cm. Indeed, the distal mound is probably an
closer to the escarpment this ridge is not present accumulation of material from quite near the front
(Figure 2). If this distal ridge is characteristic of a of the debris mass.
flowing debris mass, as suggested by Shreve
(1968), then it suggests that the mechanics of Fourthly, given the density of tree cover prior to
movement are probably scale-dependent. The the collapse, one would expect a much higher
downslope face of the distal ridge is eroding to proportion of vegetational debris to have been
such an extent that sedimentation of Lake Burra- displaced to the terminus by a sliding mass. The
gorang has significantly increased, a fact evident absence of such debris is conspicuous not only in
following heavy rainfall. the distal lobes, but also within the main mass of
the deposit. In fact only two large logs were
Apart from the peripheral mounding, the only observed and these were of the Banksia genera
other disruption to the underlying slope is a rota- which hereabouts is found only on the plateau
tional slump, below the main debris deposit (Fig- above the escarpment, indicating that the logs
ure 2). Approximately 1,200 m2 of earth have been were not incorporated into the rock flow from the
here displaced plastically down a small gully. The underlying vegetation, but instead fell with the cliff
downslope margin consists of an exposed face collapses. The absence of vegetation within the
about 3 m high which is eroding severely, adding debris mass shows that the forest was flattened
further to the sedimentation of the lake. This mass and buried by the flowing mass and not incorporat-
of earth has quite obviously failed on a rotational ed in it.
failure plane with the original surface now resting
at an angle of 5 degrees back into the slope. Finally, the relatively intact state of the boulders
Cracks have developed along the margins of the seems to indicate a laminar flow. Breakup of
slump mass as a result of the rotational nature of blocks and, especially the crushing of the angular
movement. The origin of the slumping is problem- edges on them, could be expected in a violent
atic, for the slope has not been undercut, nor has turbulent flow. Yet, as shown in Plate 3, the edges
there been subsidence triggered by mining; the are still well-preserved, and the only major break-
nearest coal seam mined in this area outcrops age is along bedding planes.
some 40 m above the slump. It is possible that
this feature has resulted from the added loading of Air layer lubrication
the avalanche debris on the underlying slopes For a mass of debris to flow, some lubricating
which consist of highly weathered colluvium. Fur- mechanism which overcomes friction on the under-
thermore, this slump feature has occurred in the lying slope must be invoked. Shreve (1968) pro-
upper reaches of a small creek channel above posed that this mechanism was a trapped layer of
which the debris mass has come to rest, and it is air, upon which the mass of debris flowed in a
226 Australian Geographer 19 (2), November 1988
manner similar to that of a hovercraft. For air layer Unfortunately, there is no section through the
entrapment to occur, the mass must be projected debris at Nattai North but it appears that at least
into the air (Shreve 1968) over some change in the surface layers of the debris flowed in a mostly
topography. Although a bench on the underlying
slope is present at Nattai North, there is no evi- (a)
dence to suggest that the mass was projected
into the air, and in the initial stages of movement it
could have traversed the slope in a saltatory
motion. Therefore, the air layer hypothesis needs
to be assessed against field evidence rather than
assumed to apply to Nattai North.
Shreve's description of the morphological fea-
tures associated with both the Elm and Blackhawk
landslides (Shreve 1968) reveals a remarkable
similarity to the Nattai North avalanche. Shreve
suggests that air layer lubrication readily accounts
for several morphological features otherwise diffi-
cult to explain, including the distal ridge of earth,
an extensive dust cloud due to the escape of
compressed air, a 'jigsaw-puzzle' effect whereby
the rock fragments maintain the original strati-
graphic positions, and a surface pattern of trans-
verse ridges on the mass of debris.
At first sight, Shreve's hypothesis seems prom-
ising, for these features are certainly present at
Nattai North. Yet plausible as the case may seem,
none of the features at Nattai North are truly
diagnostic of air layer lubrication, for all of them
have alternative explanations. Moreover, the com-
paratively short runout distance of the debris
tongue is also difficult to reconcile if an air cush-
ion was present. The debris lobe of the main
failure only traversed a distance of some 900 m
from the base of the escarpment, an order of
magnitude less than the Blackhawk landslide
(Shreve 1968). Given that rockflows at both Elm
(Heim 1932) and Blackhawk engulfed the sur-
rounding topography and travelled uphill in the
process, it could be expected that the debris at
Nattai North would have travelled much further
downslope if it had been lubricated by a cushion
of air.

A laminar flow of cohesionless grains

The features found at Nattai North can also be
explained by a flow of cohesionless grains (Hsu
1975). Heim (1932) termed this granular flow a
'mechanical fluidisation' of debris in which momen-
tum is maintained as a result of transferred kinetic
energy. Bagnold (1956) noted that a grain mass
would not flow unless there was some dispersive
force which reduces both the effective normal
pressure of the grains, and the friction coefficient Figure 4: Possible stages in the propagation of debris
of the moving mass. The dispersive force can be flow, (a) Rock units at the base are weakened due to
high stresses, and shear failure is initiated, (b) Almost
derived from media such as air (Kent 1966), dust instantaneously the base of the pillar moves out and it
(Hsu 1975) or water vapour (Goguel 1978). Al- topples backwards, fracturing in the process, (c) The
though the exact nature of this dispersive medium mass has fractured following impact, though the original
stratigraphic order is preserved; the units in the tail of
has been disputed, there seems no reason why the mass maintain momentum via collisions with the
the medium could not vary with site characteris- debris in front, resulting in a transfer of energy in the
tics. downslope direction.

Australian Geographer 19 (2), November 1988 227

dry state because there is no evidence of a mud CONCLUSIONS
matrix between the boulders. The report of a large
dust cloud immediately following a failure in 1983 The rock avalanche at Nattai North is probably
(R. Coulter personal communication 1986), and the largest contemporary failure in Australia. Its
deposits of very fine sediment on top of the larger great magnitude can be related primarily to the
blocks suggest that a dispersive dust medium extent and proximity of mining operations and to
(Hsu 1975) may have played a role in the flow. rock strength characteristics. Because of the
Irrespective of the role of dust in supporting the magnitude of the collapse, the debris moved as a
blocks, the mass must have travelled on a fluidis- laminar flow of cohesionless blocks over an unusu-
ed boundary layer consisting of vegetational de- ally long distance on the footslope. Clearly, there
bris, mud and fractured debris. That conditions in is a need for further research to establish precise-
the underlying slope were moist immediately prior ly what role the rockslope processes outlined in
to the main failure can be seen in places where this paper have contributed to the evolution of the
the distal mound has been plastically deformed Sydney Basin sandstone landscape.
around trees. The escarpments of the Sydney Basin represent
Figure 4 shows the stages envisaged in the a scenic asset and the recent push to extend
propagation of the debris flow. The sandstone mining into these landscapes clearly creates a
towers which have titled back into the face even- conflict of interests. Current restrictions regarding
tually topple backwards as a result of shear failure the limit angle of draw on coal extraction near
near the base. The rock units which fall through rock faces need to be reviewed with respect to
the greatest vertical distance have the greatest local rock characteristics and escarpment mor-
kinetic energy, necessitating a higher downslope phology. More information in this regard may give
velocity than for those units which have only fallen access to large quantities of otherwise sterilised
through a short distance. Therefore collisions from high-grade coal, or conversely, reduce the risk of
distal blocks occur, resulting in a transfer of ener- further large-scale escarpment failures.
gy and thus momentum in the downslope direction.
It is this transfer of momentum which Heim (1932)
observed to be the force providing the impetus for ACKNOWLEDGEMENTS
the flow. In this manner the stratigraphic order in
the cliff face is preserved during the flow. This Original research for this paper was done as
feature was clearly observed in the main lobe of part of an honours degree at the University of
the deposit, but in many areas has been obscured Wollongong under the supervision of Dr R. W.
by smaller blockfalls. A flow of this type transports Young, to whom I am indebted. Thanks also go to
the debris very gently (Johnson 1970) and grain- Dr A. R. M. Young and Dr. G. C. Nanson (University
to-grain collisions are not violent. This would ac- of Wollongong) and to Assoc. Prof. R. J. Blong
count both for the transport of apparently fragile (Macquarie University) for their critical comments
boulders which have clearly fractured after move- and review, and to Rod Bashford for artwork.
ment (Plate 3), and also for the limited degree of MWSDB and Nattai North Colliery kindly provided
debris deformation throughout the entire mass. access to records relating to rockfalls.

REFERENCES
Angel, J. and Hayes, R. (1980) The conflict between underground coal mining and natural lands within the
Sydney Basin, Action Environmental Search, National Parks Association, Sydney.
Bagnold, R. A. (1956) 'The flow of cohesionless grains in fluids', Philosophical Transactions, Royal Society
London, A, 249, 335-97.
Cunningham, D. M. (1986) The Nattai North escarpment failure, unpublished BEnvSc honours thesis,
Department of Geography, University of Wollongong.
Davies, T. R. H. (1982) 'Spreading of rock avalanche debris by mechanical fluidization', Rock Mechanics,
15, 9-24.
Eisbacher, G. H. (1979) 'Cliff collapse and rock avalanches (sturzstroms) in the Mackenzie Mountains,
Northwestern Canada', Canadian Geotechnical Journal, 16, 309-34.
Eisbacher, G. H. and Clague, J. J. (1984) Destructive mass movements in high mountains: hazard and
management. Special Publication No. 84-16, Geological Survey of Canada.
Evans, R. S. (1981) 'An analysis of secondary rock failures — the stress redistribution method', Quarterly
Journal of Engineering Geology, 14, 77-86.
Goguel, J. (1978) 'Scale dependent rockslide mechanisms with emphasis on the role of pore fluid
vaporisation', in Voight, B. (ed.) Rockslides and avalanches, 1 — natural phenomena, Elsevier, New
York, pp. 693-705.

Australian Geographer 19 (2), November 1988 228

Goodman, N. M. and Bray, J. W. (1977) 'Toppling of rockslopes', in Goodman, N. M. and Bray, J. W. (eds)
Engineering for foundations, American Society of Civil Engineering, New York, pp. 201-34.
Gray, N. M. (1982) 'Direction of stress in the southern Sydney Basin', Journal, Geological Society of
Australia, 29, 277-84.
Gray, N. M. (1983) 'Earthquakes and Warragamba Dam storage', Geological Society of Australia, Special
Publication, 11, 109-26.
Heim, A. (1932) Bergsturz und menschenleben, Fretz and Wasmuth Verlag, Zurich.
Hsu, K. J. (1975) 'Catastrophic debris streams (Sturzstroms) generated by rockfalls', Bulletin, Geological
Society of America, 86, 129-40.
Johnson, A. M. (1970) Physical processes in geology, Freeman and Cooper, San Francisco.
Kapp, W. A. (1985) A model for subsidence in NSW, Australia, unpublished PhD thesis, Department of Civil
Engineering, University of Wollongong.
Kent, P. E. (1966) 'The transport mechanism in catastrophic rock falls', Journal of Geology, 74, 79-83.
Kotze, G. P. (1980) Report on the stability of the escarpment above Nattai North Colliery, NSW,
unpublished report by P. J. Burgess and Associates, Consulting Geologists, Sydney.
Kotze, G. P. (1986) Nattai North Colliery — escarpment monitoring, photogrammetry and geological
mapping, unpublished report by P. J. Burgess and Associates, Consulting Geologists, Sydney.
Langford-Smith, T. (1976) 'Excursion V, 13B, Blue Mountains', 25th International Geological Congress,
Sydney.
Pariseau, W. G. and Voight, B. (1979) 'Rockslides and avalanches: basic principles and perspectives in the
realm of civil and mining operations', in Voight, B. (ed.) Rockslides and avalanches — 2, engineering
sites, Elsevier, New York, pp. 1-91.
Pells, P. J. N. (1985) Engineering geology of the Sydney region, Balkema, Rotterdam.
Pells, P. J. N., Braybrooke, J. C., Mong, J. and Kotze, G. P. (1987) 'Cliffline collapse associated with mining
activities', in Walker, B. F. and Fell, R. (eds) Soil slope stability and stabilisation, Balkema,
Rotterdam, pp. 359-85.
Shreve, R. L. (1968) The Blackhawk landslide, Special Paper No. 108, Geological Society of America.
Voight, B., Janda, R., Glicken, H., Douglas, P. M. (1983) 'Nature and mechanics of the Mount St. Helens
rockslide avalanche of 18th May 1980', Geotechnique, 33, 243-73.
Walker, P. H. (1963) 'Soil history and debris avalanche deposits along the Illawarra scarpland', Australian
Journal of Soil Research, 1, 223-30.
Waterhouse, L. L., Browne, W. R., and Moye, D. G. (1951) 'Preliminary geological investigations in
connection with proposed Warragamba Dam, NSW', Institute of Engineers, April-May, 74-84.
Young, A. R. M. (1977) 'The characteristics and origin of coarse debris deposits near Wollongong, NSW,
Australia', Catena, 4, 289-307.
Young, R. W. (1983a) 'The tempo of geomorphological change: evidence from southeastern Australia',
Journal of Geology, 91, 221-230.
Young, R. W. (1983b) 'Blockgliding in sandstones of the southern Sydney Basin', in Young, R. W. and
Nanson, G. C. (eds) Aspects of Australian sandstone landscapes, Australian and New Zealand
Geomorphology Group, Wollongong, pp. 31-8.

Australian Geographer 19 (2), November 1988 229