Content uploaded by Tanziha Mahjabin
Author content
All content in this area was uploaded by Tanziha Mahjabin on Mar 15, 2016
Content may be subject to copyright.
Journal of Coastal Research
SI
75
XX-XX
Coconut Creek, Florida
2016
Factors influencing the occurrence of Dense Shelf Water Cascades
in Australia
Tanziha Mahjabin†*, Charitha Pattiaratchi†, and Yasha Hetzel†
ABSTRACT
Mahjabin, T.; Pattiaratchi, C., and Hetzel, Y., 2016. Factors influencing the occurrence of Dense Shelf Water Cascades
in Australia. In: Vila-Concejo, A.; Bruce, E.; Kennedy, D.M., and McCarroll, R.J. (eds.), Proceedings of the 14th
International Coastal Symposium (Sydney, Australia). Journal of Coastal Research, Special Issue, No. 75, pp. XX-
XX. Coconut Creek (Florida), ISSN 0749-0208.
Transport of inshore waters and suspended material off the continental shelf by Dense Shelf Water Cascades (DSWC)
has important ecological and biogeochemical implications in Australian waters. Because of high rates of evaporation,
denser saline water occurs in the shallow coastal regions around Australia, setting up horizontal density gradients that
can drive DSWC. Ocean glider data available from the Integrated Marine Observing System (IMOS), which is operated
by the Australian National Facility for Ocean Gliders (ANFOG) located at the University of Western Australia, were
used to measure cross-shelf density profiles under varying wind and tide conditions for seven contrasting regions
around the entire continent. Overall 97 sets of spatial and temporal resolution data from year 2008 to 2015 collected
by the ocean gliders and analysed with a subset presented here. Data from 19 transects covering the years 2012 to 2015
for the Pilbara region of Western Australia, indicated that cascades occur during the autumn and winter due to cooling
of the coastal waters which already have higher salinity due to evaporation during the summer months. The cross-shelf
density gradient in this continental shelf was found to be maximum in July with a value of 14.23x10-6 kg m-4.
ADDITIONAL INDEX WORDS: Cascades, gliders, horizontal density gradient.
INTRODUCTION
Dense shelf water is formed in coastal waters either by a
decrease in temperature through cooling or increase in salinity
from evaporation or ice formation (Figure 1). In Australian
waters, high rates of evaporation, up to 2.5m per year, (Figure
2) with negligible rainfall and run-off cause a net loss of fresh
water from the inner continental shelf. This results in coastal
waters having higher salinity than offshore. During the winter
months the shallower coastal waters lose heat due to convective
processes resulting in colder water near the coast. The
combination of higher salinity colder water closer to the coast
results in a horizontal density gradient (d𝜌/dx) with increasing
density from the ocean towards the coast. This gradient drives
a gravitational circulation with the offshore transport of denser
water along the sea bed. This is controlled by vertical mixing
resulting from turbulence generated by the wind and the tide
(Hetzel et al., 2013; Pattiaratchi et al., 2011). Under low wind
and tidal mixing conditions either a bottom gravity current or a
surface plume will be present depending on the sign of the
horizontal density gradient (Figure 1a, 1b) and under strong
wind and tidal mixing conditions the water column is well
mixed (Figure 1c, 1d).
The buoyancy-driven gravity current forms Dense Shelf Water
Cascades: DSWC (Canals et al., 2006; Pattiaratchi et al., 2011;
Shapiro et al., 2003; Shearman and Brink, 2010).
Dense water cascades have been found in over 60 locations
around the world and most of them happen in Polar Regions
due to ice formation (Ivanov et al., 2004). DSWC can play a
major role in transporting terrestrial carbon, nutrients, larvae,
low-oxygen water, sediments and also pollutants from coastal
regions to deeper ocean.
DSWC have been documented previously for in some
locations around Australia by research cruises and using
moorings, usually for individual events during single seasons:
E.g. North-west Australian shelf (Brink and Shearman, 2006;
Shearman and Brink, 2010), Shark Bay (Pattiaratchi and Woo,
2009), Great Australian Bight (Petrusevics et al., 2009),
Spencer Gulf (Bowers and Lennon, 1987).
Seasonal variation of DSWC has been identified as a major
feature through the deployment of ocean gliders in coastal
waters along the Rottnest continental shelf using 13 months of
data (Pattiaratchi et al., 2011). Since these previous studies
wide-ranging ocean glider data have become available, making
it possible to investigate DSWC in other areas around Australia
with different wind and tide conditions (Figure 2). In this paper,
we used high spatial and temporal resolution temperature and
salinity data collected using Teledyne Webb Research Slocum
Gliders (Schofield et al., 2007) to identify DSWC formation in
the Pilbara region of Western Australia and to investigate
DSWC dynamics in general.
____________________
DOI: 10.2112/SI75-XXX.1 received Day Month Year; accepted in
revision Day Month Year.
*Corresponding author: tanziha.mahjabin@research.uwa.edu.au
©Coastal Education and Research Foundation, Inc. 2016
† School of Civil, Environmental and Mining Engineering & UWA
Oceans Institute, The University of Western Australia, Australia.
www.cerf-jcr.org
www.JCRonline.org
XX Mahjabin, Pattiaratchi, and Hetzel
_________________________________________________________________________________________________
Journal of Coastal Research, Special Issue No. 75, 2016
Figure 1. Effects of vertical mixing by wind and tide in the presence of a cross-shelf density gradient. Under low wind and tidal mixing conditions either
a bottom gravity current or a surface plume will be present depending on the sign of the horizontal density gradient (a,b). Under strong vertical mixing
conditions the water column is well mixed (c,d).
Figure 2. Annual evaporation rate in Australia (Yu, 2007) and selected
study sites with tidal range and mean wind speed in summer and winter:
(i) Kimberley; (ii) Pilbara; (iii) Two Rocks; (iv) Investigator Strait; (v)
Port Stephens; (vi) Yamba; (vii) Capricorn Channel .
METHODS
Glider data from seven sites (Figure 2) are available to study
DSWC under contrasting environmental: (i) Kimberley, north-
west Australia: macro-tidal and moderate wind (Shearman and
Brink, 2010); (ii) Pilbara, north-west Australia: macro-tidal
(Holloway, 1983); (iii) Two Rocks, Western Australia: mainly
wind driven (Pattiaratchi et al., 1997) with low tidal range with
diurnal tides (Pattiaratchi and Eliot, 2008); (iv) Investigator Strait,
South Australia: mainly spring-neap tidal cycle driven (Nunes
and Lennon, 1986; 1987); (v) Port Stephens New South Wales:
moderate tide dominated (McPherson et al., 2013) and strong
wind (Geary, 1987); (vi) Yamba, New South Wales: mostly wind
driven and micro-tide (Pritchard et al., 2007) ; (vii) Capricorn
Channel, Queensland: wind and tide driven (Andutta et al., 2011).
This paper focusses on data from the Pilbara site that has
moderate wind and tidal forcing. Analysis of the other sites is the
subject of future work.
We used Teledyne Webb Research Slocum Electric Glider
(Schofield et al., 2007) data which are operated by the Australian
National Facility for Ocean Gliders (ANFOG) located at the
University of Western Australia. The data are publicly available
through the Integrated Marine Observing System (IMOS).
Slocum gliders cover maximum depth range of 200 m. Gliders
can measure data from the surface to up to 5 m above the seabed
with mean speed of 25 km per day (Schofield et al., 2007).
Gliders traverse a saw-tooth pattern using buoyancy control
whilst moving forward to the target destination and navigating to
a series of pre-programmed waypoints using GPS, internal dead
reckoning and altimeter measurements. A Seabird-CTD,
Chlorophyll-a fluorescence measuring sensor, coloured dissolved
organic matter (CDOM) sensor, 660 nm Backscatter WETLabs
BBFL2SLO optical sensor and an Aanderaa Oxygen Optode were
attached to the ocean glider for this study. Slocum Gliders are
small in size (1.8 m), efficient and economical to sample for much
longer periods and higher spatial resolution compared to ships.
For this study, 19 transects from 13 sets of glider missions for
the Pilbara were analysed over the period 2012 to 2015
considering specific tide and wind conditions. This included a
total of 40248 vertical profiles and over 13.5 million data points.
Analysis aimed to identify the presence of DSWC and the effects
of wind and tide on the cascade formation. Quality control of the
data were done after the recovery of each glider and then actual
XX Factors influencing the occurrence of Dense Shelf Water Cascades around Australia XX
_________________________________________________________________________________________________
Journal of Coastal Research, Special Issue No. 75, 2016
vehicle trajectory was transposed onto the Pilbara transects as a
straight line. Each variable was interpolated onto a grid with
vertical and horizontal resolutions of specific time and depth
respectively. Density gradients were calculated by defining
latitudes and longitudes of the starting shallow part and shelf
break of the transects, while the direction of shallow part to shelf
break was considered as positive. Tidal data were predicted using
the TPXO7.2 global database and wind data were obtained from
the European Centre for Medium-Range Weather Forecast
Interim Reanalysis (ECMWF ERA-I) (Dee et al., 2011).
Vertical temperature stratification has been compared with the
local wind speed cubed (W3) and bottom current speed cubed
(|Ub3|). These are proportional to available mixing energy from
the wind and tidal currents respectively (e.g. Nunes and Lennon,
1987; Nunes Vaz et al., 1990).
RESULTS
Analysis of the 19 transects of ocean glider data in the Pilbara,
collected between 2012~2015, indicated that DSWC were a
common occurrence during the winter months when cross-shelf
density gradients formed with denser water near the coast. Figure
3 represents the location of the glider path near Pilbara on July
2012 which is chosen for an example.
Figure 3. Glider path location for Pilbara July 2012 is shown on the Sea
Surface Temperature map.
Cross-shelf transects (Figure 4) indicated that both temperature
and salinity contributed to the dense water formation. The dense
water then flowed along the sea bed and spilled off of the
continental shelf reaching depths of up to 150m (Figure 4). This
DSWC was observed during July 2012 when high salinity water
had accumulated near the coast due to summer evaporation and
subsequently this water was cooled and the gradient became
strong enough to force the inshore water off the shelf. Spatial
patterns of fluorescence, sediment and oxygen closely followed
density, indicating that these properties were influenced by the
DSWC (Figure 5), and were likely transported along the sea bed
from the continental shelf to the open ocean.
Figure 4. Cross-shelf profile of DSWC as measured by a Slocum glider
on the Pilbara coast during July 2012.
Figure 5. Ancillary water properties (Fluorescence, sediment and oxygen)
as measured by a Slocum glider on the Pilbara coast during July 2012.
Driving force for the formation of cascades
Analysis of all 19 transects suggested that winter cooling was
an important contributor to DSWC formation in the Pilbara, with
a majority of the cascades occurring during winter months. In
winter the winds were weaker and the cross-shelf density
gradients were enhanced by cooling (Figure 6).sediment supply
and the concentration of wave energy removes the sand, causing
erosion (Figure 7).
XX Mahjabin, Pattiaratchi, and Hetzel
_________________________________________________________________________________________________
Journal of Coastal Research, Special Issue No. 75, 2016
Figure 6. The monthly density gradient for Pilbara showing the mean
density gradient for all available data for each month. The error bar for
July indicated the range of values calculated using different transects.
Data from multiple seasons indicated that favorable density
gradients formed repeatedly in autumn and winter months of each
year and DSWC occurred whenever the vertical mixing from
wind and tide was weaker.
Using multiple transects acquired over several years we
calculated the average horizontal density gradient for each month
of the year (Figure 6). The calculated positive density gradient
(onshore – offshore) in Pilbara was maximum in July as 14.23x
10-6 kgm-4. The gradient became negative during the summer. The
seasonality of cascade formation in Pilbara closely followed the
density gradient with all observed cascades requiring a positive
density gradient as occurred between April and September.
Control of cascade formation by turbulent vertical mixing
The data suggested that whenever the horizontal density
gradient is strong, we can expect cascade formation. However,
wind mixing and tidal mixing were also capable of inhibiting the
formation of DSWC and the relative importance of these factors
varies around Australia.
Figure 7. Pilbara transect in July 2012 with wind and tidal mixing energy
proxies (windspeed3 and tidal current speed3) with DSWC present.
Under low wind and tidal mixing conditions, the water column
stratifies and water flows offshore along the sea bed; whereas
high mixing inhibits stratification and offshore transport of water.
Tidal amplitudes vary along the Australian coast and influence
cross-shelf water movement. Two contrasting transects from
Pilbara plotted with proxies for wind and tidal mixing
(windspeed3 and tidal current speed3) illustrate the effects of
vertical mixing on DSWC formation (Figure 7 and Figure 8). The
first transect (Figure 7) showed a clear DSWC and weak tidal
currents and wind. Cascades were absent in the second transect
(Figure 8) with higher current speeds and slightly stronger winds.
Figure 8. Pilbara transect in September 2014 showing absence of cascade
and stronger tidal currents and windspeeds.
DISCUSSION
Dense water flows are the result of either intense cooling or
excess evaporation. Analysis of 19 ocean glider transects
collected between 2012 and 2015 in the Pilbara region of Western
Australia indicated that DSWC were a common occurrence,
particularly in Autumn and Winter seasons, even under relatively
strong tide and wind conditions. It has been shown previously that
cascades are controlled by turbulence generated by wind and tide
(Hetzel et al., 2013; Pattiaratchi et al., 2011). However this study
data suggests that the density gradients dominate over mixing
energy most of the time in the Pilbara during Autumn and Winter
seasons. Cascade occurs in this period of time with velocities
between 2~3 cm/s in Pilbara region (Bahmanpour et al., 2016),
which is comparable with other location where it was measured
before. DSWC have been documented previously for a single
season in North-west Australian shelf (Brink and Shearman,
2006; Shearman and Brink, 2010), but here several years data
allowed us to observe the seasonality of cascade formation for
Pilbara region.
The next step will be to analyse the data for other 6 study
regions. If the results also indicate that DSWC occur even where
tides or winds are strong, it will confirm that dense shelf water
cascades are an important process for cross shelf exchange around
the entire Australian continent. The broad range of ocean glider
deployments presents a unique opportunity to examine DSWC
over such a large and varied coastline.
XX Factors influencing the occurrence of Dense Shelf Water Cascades around Australia XX
_________________________________________________________________________________________________
Journal of Coastal Research, Special Issue No. 75, 2016
CONCLUSIONS
An analysis of the dynamics of DSWC formation in the
Pilbara region of Western Australia was completed based on
19 individual transects covering three years (2012-2015). The
formation of DSWC in the Pilbara region of Western Australia
was found to depend on the balance between the cross shelf
density gradient and vertical mixing by wind and tide. When
the density gradient is strong, we can expect DSWC; but when
the mixing is strong enough to make the shallow water
vertically mixed DSWC will be absent. During autumn and
winter the cross shelf density gradient remains positive and
dominates over mixing. As a result DSWC occur in the North-
western Australian coast of Pilbara despite the relatively large
tidal range, with strongest DSWC events occurring during neap
tides and weak winds. Further analysis of 124 remaining
transects will be undertaken for other selected locations around
Australia to determine whether similar relationships exist
between cross-shelf density gradients, vertical mixing, and
DSWC.
ACKNOWLEDGMENTS
All ocean glider data used in this paper are from the Integrated
Marine Observing System (IMOS) which are operated by the
Australian National Facility for Ocean Gliders (ANFOG) located
at the University of Western Australia. IMOS is funded by the
National Collaborative Research Infrastructure Strategy and the
Super Science Initiative. The postgraduate research has been
funded by Scholarship International Research Fees (SIRF) and
University International Stipend.
LITERATURE CITED
Andutta, F.P.; Ridd, P.V., and Wolanski, E., 2011. Dynamics of
hypersaline coastal waters in the Great Barrier Reef. Estuarine,
Coastal and Shelf Science, 94(4), 299-305.
Bahmanpour, M.H.; Pattiaratchi, C., Wijeratne, E.M.S.,
Steinberg, C., and D'Adamo, N., 2016. Multi-year observation
of a boundary current along the shelf edge of North Western
Australia. In: Vila-Concejo, A.; Bruce, E.; Kennedy, D.M., and
McCarroll, R.J. (eds.), Proceedings of the 14th International
Coastal Symposium. Journal of Coastal Research, Special
Issue No. 75.
Bowers, D. G.and Lennon, G. W., 1987. Observations of
Stratified Flow over a Bottom Gradient in a Coastal Sea. Cont
Shelf Res, 7(9), 1105-1121.
Brink, K.H. and Shearman, R.K., 2006. Bottom boundary layer
flow and salt injection from the continental shelf to slope.
Geophys Res Lett, 33(13).
Canals, M.; Puig, P.; de Madron, X. D.; Heussner, S.; Palanques,
A., and Fabres, J., 2006. Flushing submarine canyons. Nature,
444(7117), 354-357.
Dee, D.P., et al., 2011. The ERA-Interim reanalysis:
configuration and performance of the data assimilation system.
Q J Roy Meteor Soc, 137(656), 553-597.
Geary, M. G. (1987), Port Stephens - Salamander Bay, Physical
processes data base Rep. 85045, Public Work Department-
Civil Engineering division.
Hetzel, Y.; Pattiaratchi, C., and Lowe, R., 2013. Intermittent
dense water outflows under variable tidal forcing in Shark Bay,
Western Australia. Cont Shelf Res, 66, 36-48.
Holloway, P. E., 1983. Internal Tides on the Australian Northwest
Shelf - a Preliminary Investigation. J Phys Oceanogr, 13(8),
1357-1370.
Ivanov, V. V.; Shapiro, G. I.; Huthnance, J. M.; Aleynik, D. L.,
and Golovin, P. N., 2004. Cascades of dense water around the
world ocean. Prog Oceanogr, 60(1), 47-98.
McPherson, B.; S. Young; B. Modra; E. Couriel; B. You; D
Hanslow; D.C.; T Baldock, and Nielsen, P., 2013. Penetration
of Tides and Tidal Anomalies in New South Wales Estuaries.
Proceedings of the Coasts and Ports 2013: 21st Australasian
Coastal and Ocean Engineering Conference and the 14th
Australasian Port and Harbour Conference (Sydney, N.S.W),
pp. 537-542.
Nunes, R.A. and Lennon, G.W., 1986. Physical Property
Distributions and Seasonal Trends in Spencer Gulf, South-
Australia - an Inverse Estuary. Aust J Mar Fresh Res, 37(1),
39-53.
Nunes, R.A. and Lennon, G.W., 1987. Episodic Stratification and
Gravity Currents in a Marine-Environment of Modulated
Turbulence. J Geophys Res-Oceans, 92(C5), 5465-5480.
Nunes Vaz, R.A.; Lennon, G.W., and Bowers, D.G., 1990.
Physical Behavior of a Large, Negative or Inverse Estuary.
Cont Shelf Res, 10(3), 277-304.
Pattiaratchi, C. and Eliot, M., 2008. Sea Level Variability in
South-West Australia: From Hours to Decades. In: J. M. Smith
(eds.), Proceedings of the international coastal engineering
conference (New Jersey, London)., pp. 1186-1198.
Pattiaratchi, C. and Woo, M., 2009. The mean state of the
Leeuwin Current system between North West Cape and Cape
Leeuwin. Journal of the Royal Society of Western Australia,
92, 221-241.
Pattiaratchi, C.; Hegge, B.; Gould, J., and Eliot, I., 1997. Impact
of sea-breeze activity on nearshore and foreshore processes in
southwestern Australia. Cont Shelf Res, 17(13), 1539-1560.
Pattiaratchi, C.; Hollings, B.; Woo, M., and Welhena, T., 2011.
Dense shelf water formation along the south-west Australian
inner shelf. Geophys Res Lett, 38(10), L10609.
Petrusevics, P.; Bye, J.A.T.; Fahlbusch, V.; Hammat, J.; Tippins,
D. R., and van Wijk, E., 2009. High salinity winter outflow
from a mega inverse-estuary-the Great Australian Bight. Cont
Shelf Res, 29(2), 371-380.
Pritchard, T.; Holden, C.; Lee, R.; Black, K., and Healy, T., 2007.
Dynamics and Dispersion in the Coastal Boundary Layer off
Coffs Harbour in Eastern Australia. J Coastal Res, 848-857.
Schofield, O., et al., 2007. Slocum gliders: robust and ready. J
Field Robot, 24(6), 473-485.
Shapiro, G.I.; Huthnance, J.M., and Ivanov, V.V., 2003. Dense
water cascading off the continental shelf. J Geophys Res-
Oceans, 108(C12).
Shearman, R.K. and Brink, K. H., 2010. Evaporative dense water
formation and cross-shelf exchange over the northwest
Australian inner shelf. Journal of Geophysical Research,
115(C6), C06027.