Forgot your password?
Document Actions
Kendrick et al 02
Aquatic Botany 73 (2002) 75–87
Changes in seagrass coverage in Cockburn Sound,
Western Australia between 1967 and 1999
Gary A. Kendrick a,∗ , Matt J. Aylward a,c , Bruce J. Hegge b ,
Marion L. Cambridge a , Karen Hillman b ,
Alex Wyllie c , Des A. Lord b
aDepartment of Botany, Faculty of Science, University of Western Australia, 35 Stirling Highway,
Crawley, WA 6009, Australia
b D.A. Lord and Associates Pty Ltd., P.O. Box 3172, LPO Broadway, Nedlands, WA 6009, Australia
c Alex Wyllie and Associates, 10 Kershaw Gardens, Leeming, WA 6149, Australia
Received 20 March 2001; received in revised form 20 December 2001; accepted 20 December 2001
Abstract
Changes in seagrass coverage in Cockburn Sound from 1967 to 1999 were assessed from aerial
photographs using modern mapping methods with the aim of accurately determining the magnitude
of change in hectares of seagrasses between 1967 and 1999 and to set up a baseline for future
monitoring of seagrass loss in Cockburn Sound. Firstly, coverage and assemblages of seagrasses in
Cockburn Sound were mapped using the best available aerial photographs from 1999, rectified to a
common geodesic base with comprehensive groundtruth information, and with a semi-automated
mapping algorithm. Then the same technique was used to map historical seagrass coverage in
Cockburn Sound from aerial photographs taken in 1967, 1972, 1981 and 1994.
The seagrass coverage in Cockburn Sound has declined by 77% since 1967. Between 1967 and
1972, 1587 ha of seagrass, were lost from Cockburn Sound, mostly from shallow subtidal banks on
the eastern and southern shores. By 1981, a further 602 ha had been lost. Since 1981, further seagrass
losses (79 ha) have been restricted to a shallowing of the depth limit of seagrasses, localised losses
associated with port maintenance and a sea urchin outbreak on inshore northern Garden Island.
There has been no recovery of seagrasses on the eastern shelf of Cockburn Sound after nutrient
loads were reduced in the 1980s, suggesting that this shallow shelf environment has been altered
to an environment not suited for large-scale recolonisation by Posidonia species. © 2002 Elsevier
Science B.V. All rights reserved.
Keywords: Mapping; Aerial photography; Seagrass decline; Posidonia spp.; Australia
∗ Corresponding author. Tel.: +61-8-9380-3998; fax: +61-8-9380-1001.
E-mail address: garyk@cyllene.uwa.edu.au (G.A. Kendrick).
0304-3770/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 4 - 3 7 7 0 ( 0 2 ) 0 0 0 0 5 - 0
76 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
1. Introduction
Large-scale losses of seagrass meadows associated with human impacts have occurred
in Australia (Walker and McComb, 1992) and worldwide (Short and Wyllie-Echeverria,
1996). One of the most commonly cited cases of large-scale seagrass loss happened in
Cockburn Sound, Western Australia following increased industrial discharge and expan-
sion of port facilities (Cambridge, 1979; Cambridge and McComb, 1984; Hillman, 1986).
Cambridge and McComb (1984) documented a loss of 3400 ha of seagrass meadows be-
tween 1969 and 1978, with most losses occurring before 1975. Their study has recently
been criticised heavily by local marine industries and management agencies because un-
rectified aerial photographs were used, the mapping area was poorly defined, and the 1969
map of seagrass distribution in the eastern shelf of Cockburn Sound was mostly determined
from field observations, as the aerial photography was of poor quality (Cambridge and
McComb, 1984). This study redresses these limitations, focuses on the years of major his-
torical loss and extends the work of Cambridge and McComb (1984) 20 years to the present
day.
Cockburn Sound has historically been used for recreational boating and commercial fish-
ing, but its relatively protected deep waters also provide an excellent anchorage, and in 1954
it was designated as an industrial harbour for the Perth-Fremantle region (Cambridge and
McComb, 1984). Industrial development commenced with the establishment of an oil refin-
ery in 1955, followed by the addition of iron, steel, alumina and nickel processing and refin-
ing plants, chemical and fertiliser production plants and a bulk grain terminal. In conjunction
with the development of these heavy industries, wharves and groynes were built and chan-
nels dredged for shipping access. A wastewater treatment plant was commissioned in 1966
which discharged into the northern end of the Sound, while at the southern end of the Sound,
a 2.5 km long causeway was constructed between 1971 and 1973 to connect Garden Island
to the mainland (Hicks et al., 1973). This causeway is interrupted by two trestle bridges (one
305 m long and other 610 m long), through which limited exchange of water with the ocean
occurs.
Industrial expansion resulted in a decline in seagrass coverage in Cockburn Sound from
4200 to 900 ha between 1954 and 1978, with most loss occurring between 1969 and 1975
(Cambridge and McComb, 1984). The Western Australian Government funded a 3 year
study (1976–1979; Cambridge, 1979), which identified that the decline in seagrass cover-
age was linked to an increase in nitrogen loading in the Sound. Over 90% of the nitrogen
contributed by industrial effluent and wastewater to the Sound came from two sources:
the outlet shared by the CSBP fertiliser works and the Kwinana Nitrogen Company, and
the outlet of the Woodman Point Wastewater Treatment Plant (Cambridge et al., 1986).
The proposed mechanism for seagrass loss was that increased nitrogen levels led to en-
hanced growth of nuisance epiphytic algae and consequent shading and deterioration of the
seagrasses (Cambridge et al., 1986).
Water quality in Cockburn Sound has improved greatly since the late 1970s (Hillman,
1986; Hillman, unpublished data). The annual loading of nitrogen has decreased from
2000 t N per year in 1978 to less than 500 t N per year in 1997 (Hillman, unpublished
data). The northeastern shore of Cockburn Sound still has high N levels from contaminated
groundwater discharging into the sound. This contaminated groundwater contributed 70%
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 77
of the total nitrogen entering Cockburn Sound in 1995 (Department of Environmental
Protection, 1996).
There has not been a reassessment of seagrass distribution in Cockburn Sound for the
past 20 years since the work of Cambridge and McComb (1984). This paper redresses
this by mapping seagrass coverage in 1981, 1994 and 1999 and revisits the period of
maximum seagrass loss between 1967 and 1972. We map coverage of assemblages of
seagrasses in Cockburn Sound using the best available aerial photographs from 1999, rec-
tified to a common geodesic base with comprehensive groundtruth information (Kendrick
et al., 2000). This is then compared to maps of historical seagrass coverage in Cock-
burn Sound constructed from historical aerial photographs taken in 1967, 1972, 1981
and 1994.
2. Methods
2.1. Mapping of seagrasses
To determine seagrass distribution, submerged vegetation was mapped from recent and
historical aerial photography. Distribution of seagrass assemblages and reef were then de-
termined in 1999 from towed underwater video. We define assemblages of seagrasses as
multi-species assemblages dominated, or characterised, by single or multiple species as
determined from their relative abundance in video footage. In Cockburn Sound, the sea-
grass species Amphibolis antarctica, Amphibolis griffithii, Posidonia australis, Posidonia
coriacea, Posidonia sinuosa, Halophila ovalis, Heterozostera tasmanica and Syringodium
isoetifolium were components of the seagrass assemblages. We did not map to single species,
although a single species assemblage is composed of more than 70% of that species. Results
from mapping of aerial photographs and underwater video footage were then combined in
a GIS to create coverage maps of seagrass assemblages, reef and unvegetated sand. We
describe our maps as coverage maps, rather than maps of seagrass cover to reduce the
confusion between the area covered by a single species of seagrass and the area occu-
pied by an assemblage of seagrasses than is dominated by one or a few species. This is
not presence–absence mapping, as vegetated assemblages do not exclude all unvegetated
habitat (see Section 2.6).
2.2. Study area and mapped regions
Cockburn Sound is a sheltered marine embayment 16 km long × 9 km wide, and consists
of a deep central basin (17–22 m deep) surrounded by shallow platforms. The shallow
platforms vary in width from 50 m to 3 km and are where seagrass meadows are found
(Fig. 1). The seagrass mapping area was delineated as shallow platforms to a depth of 10 m
and covered an area of approximately 3667 ha (Table 1). The mapped area of the present
study is smaller than that used by Cambridge and McComb (1984), as they incorporated an
undefined proportion of Parmelia Bank into their mapped area and never defined a depth limit
for their study. The total area mapped by Cambridge and McComb (1984) was 4200 ha.
78 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Fig. 1. Map of Cockburn Sound showing the extent of seagrass mapping bounded by the coastline and 10 m isobath,
the three mapping regions (Cockburn Sound West, South and East), locations mentioned in the text, and the 1999
mapped distribution of seagrasses and reef.
Historical changes in seagrass coverage of Parmelia Bank have been recently published
(Kendrick et al., 2000) so have not been included in this paper.
Coverage of all submerged vegetation was separately calculated for three regions called
Cockburn Sound East, South and West (Fig. 1, Table 1). Cockburn Sound East encompassed
most of the eastern bank from Woodmans Point to south of James Point. Cockburn Sound
South contained the shallow bank near Rockingham, the southern sand flats and Careening
Bay, Garden Island. Cockburn Sound West encompassed the western bank north of Ca-
reening Bay and shallow waters north of Garden Island. Benthic features are difficult to
resolve from aerial photographs at depths greater than 10 m as seagrasses become sparsely
distributed at these depths. Hence, the 10 m isobath and the coastline are used to delineate
the mapping boundaries within each of these regions.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 79
Table 1
Area of seagrass assemblages in 1999 for each mapped region (Cockburn East, South and West) and for Cockburn
Sound in total
Mapping region Mapping Posidonia Posidonia Posidonia Total Reef Shallow
area (ha) australis sinuosa sinuosa and seagrass (ha) bare sand
(ha) (ha) Posidonia (ha) (ha)
australis (ha)
Cockburn Sound East 2138 0 13 0 13 25 2100
Cockburn Sound South 818 11 279 2 292 0 526
Cockburn Sound West 711 0 356 0 356 21 334
Total 3667 11 648 2 661 46 2960
Data area represented as area in hectares (ha).
2.3. Data sources
In 1999, flights to acquire aerial photographs were purpose-flown in late-February to
early-March when conditions were optimal. These conditions included maximum sea-
grass leaf cover; maximum water clarity; minimum haze from the adjacent industrial zone;
minimum turbidity from dredging, river outflow or industrial activities; minimum cloud
cover; weak prevailing winds; and incident sun angle at 20–30◦ . Colour photography with
a yellow lens filter was obtained at two altitudes. High altitude imagery was collected
at 10,000 m (scale, 1:55,000) for accurate ortho-rectification. The resulting imagery was
used as a rectification base for the low-altitude imagery that was obtained at an altitude of
3,800 m.
Mapping of the historical seagrass coverage in Cockburn Sound was conducted using
rectified and mosaicked imagery obtained in: 1994 (majority of imagery obtained on 4 and
8 January 1994 and some on 6 January 1995); 1981 (imagery from 13 June 1981); 1972
(imagery from 2 May 1972); and 1967 (imagery from 20 March 1967).
2.4. Image geo-referencing and rectification
Images were initially rectified (Datum: WGS-84) using ERMapper-6.1 to a resolution
of 2.0 m (Earth Resource Mapping, 2000). The images were initially scanned into three
colour bands (red, green and blue). The red band of the colour images contained almost
no information for marine benthic habitat, as red light was severely attenuated through the
water column. The green band was better than the red, though images had low contrast due
to atmospheric attenuation. The blue band provided the best contrast between vegetated and
unvegetated habitats and therefore was the only band used for mapping. It was converted
to 256 greyscales.
Rectified imagery was processed using Geographic Resources Analysis Support System
(GRASS) GIS software (Neteler, 1998). GRASS is a raster-based GIS and the 2.0 m cell
size and its spatial location (spatial integrity) was strictly maintained in rectified imagery.
Also the same spatial integrity was kept consistent among the rectified imagery for the
different years mapped (spatial consistency).
80 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
2.5. Automated mapping of submerged vegetation
A computerised, semi-automated, greyscale segmentation mapping method called
Spann–Wilson segmentation (Spann and Wilson, 1985) was employed to map submerged
vegetation. Spann–Wilson combines locally adaptive segmentation (local centroid) with
pre-processing using multi-level quad-tree smoothing. The segmentation method was imple-
mented using the Xite image-processing software (The University of Oslo, 1999). The size of
the moving histogram window and the number of quad-tree smoothing levels were controlled
by the operator, although for this exercise were set to 324 m2 and 3–4. Spann–Wilson seg-
mentation was very effective in defining the boundaries of seagrass meadows reefs and un-
vegetated sand from greyscale images. It reduced the original 256 greyscales from the aerial
photographs to four greyscales. The operator then chose a greyscale which most closely
coincided with the visually interpreted boundary between vegetated and unvegetated habitat.
2.6. Control rules for mapping
The vegetated areas were distinguished from the unvegetated areas as they had a distinct
photo-tone of medium to dark grey. To enable consistent coverage mapping across the study
area, a series of control rules were used (Kendrick et al., 2000). These control rules were
as follows:
• Isolated vegetated patches less than 30 m2 in area were not mapped.
• Vegetated patches, that were greater than 30 m2 and less than 100 m2 were mapped as
separate patches when the distance between one patch and another was greater than the
diameter of the patch.
• Vegetated patches, that were greater than 30 m2 and less than 100 m2 were mapped as
a single meadow when the distance between one patch and another was less than the
diameter of the patch. Unvegetated areas within the meadow with an area greater than
100 m2 were mapped.
• Vegetated patches greater than 100 m2 were mapped and the edges of these areas were
traced accurately. Unvegetated regions within these patches with areas greater than
100 m2 were mapped.
The control rules were applied automatically during mapping with control rules 2–4
processed during the Spann–Wilson segmentation step to a 324 m2 moving window, and
control rule 1 applied after the segmentation step to 1 km2 areas using a PERL script.
2.7. Groundtruth surveys
Detailed groundtruth surveys were conducted between late-February and May 1999.
These groundtruth surveys were undertaken using a differential GPS combined with diver-
operated towed video (manta tow) and a downward-looking surface deployed (drop-down)
video. Following the completion of the formal groundtruth surveys, a series of opportunistic
dives were also undertaken to establish the assemblage type in specific areas of interest. A
total of four manta tow transects (8 km in total) and 114 drop-down video locations were
surveyed during the groundtruthing exercise.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 81
The manta tow videos were analysed by pausing the video at 20 s intervals (corresponding
with the differential GPS waypoints). At each of these video pauses the percent of the
image each seagrass species and habitat type was recorded. For the drop-down videos,
percent representation of species of seagrass and habitats was averaged across the total
video footage obtained from each drop unless the species composition varied noticeably. If
it was variable, the drop was sub-divided to more accurately represent the seagrass coverage.
Seagrass species recognised in the video footage were A. antarctica, A. griffithii, P.
australis, P. coriacea, P. sinuosa, H. ovalis, H. tasmanica and S. isoetifolium. Habitats
other than seagrasses that were recognised were limestone reef and unvegetated sand.
It was not possible to separate the species of P. sinuosa and Posidonia angustifolia from
the video data, as this requires an examination of the rhizome fibres (Cambridge and Kuo,
1979). Groundtruth dives indicated that P. sinuosa was generally more common than P.
angustifolia (greater than 90% except in the Cockburn Sound East region where P. sinuosa
cover was approximately 70%). In the present mapping exercise, these two species were
both mapped as the P. sinuosa assemblage.
2.8. Mapping of seagrass assemblages and benthic habitats
The maps of the 1999 vegetated and unvegetated areas were combined with the groundtruth
data using GRASS GIS to produce a benthic assemblage map. Within the mapped region,
the seagrass assemblages P. sinuosa, mixed species P. sinuosa and P. australis, and P. aus-
tralis were distinguished as these species were the dominants. Single species assemblages
were defined when a single species had greater than 70% representation on videos and the
P. sinuosa and P. australis assemblage was defined when P. sinuosa and P. australis were
equally represented (representation of any one species could range between 30 and 70%).
Shallow unvegetated sand (<10 m depth) and reef were also mapped.
Mapping of historical seagrass coverage in Cockburn Sound was conducted using rectified
and mosaicked imagery obtained for 1967, 1972, 1981 and 1994. Coverage of all seagrass
assemblages combined was calculated for the three regions within Cockburn Sound sepa-
rately. The reef areas which were identified in the 1999 mapping exercise were transposed
onto the historical images from Cockburn Sound, using the assumption that the size and
location of these reef areas had not changed over the mapping period. Previous ground truth
data were collected from a range of sources, including those of Cambridge (1979), Marsh
and Devaney (1978) and Wilson et al. (1978), and used as an aid in interpreting historical
seagrass coverage.
3. Results
3.1. Seagrass coverage in 1999
In 1999, seagrasses occupied an area of 661 ha (18%) of sandbanks less than 10 m in
depth (3667 ha) in Cockburn Sound (Table 1, Fig. 1). Extensive seagrass meadows were
found predominantly on the western margin of Cockburn Sound (Cockburn Sound West,
82 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Table 1) and in the Southern Flats and western Rockingham regions (Cockburn Sound
South, Table 1). The seagrass meadows in 1999 were predominantly composed of P. sinu-
osa assemblages with a small area on the Southern Flats dominated by assemblages of P.
australis (Table 1). Besides these dominant species, small proportions of A. antarctica, A.
griffithii, P. coriacea, H. ovalis, H. tasmanica and S. isoetifolium were observed.
3.2. Time course of seagrass loss
In 1967, seagrasses formed an almost continuous 2929 ha meadow, between 1 and 10 m
depths, that fringed the eastern, southern and western margins of Cockburn Sound (Fig. 2).
Between 1967 and 1981, all but a few isolated patches of seagrass were lost from Cockburn
Sound East and the eastern margins of Cockburn Sound South (Figs. 2 and 3). Seagrass
loss was broad-scale with 1587 ha of seagrasses disappearing between 1967 and 1972,
and 602 ha between 1972 and 1981 (Fig. 3, Table 2). A further 79 ha of seagrasses have
disappeared between 1981 and 1999, mostly from Cockburn Sound West, but the scale of
loss was smaller and more localized.
In Cockburn Sound East, 1750 ha of seagrasses covered 82% of the shallow bank habi-
tat in 1967, but by 1972 only 310 ha (14.5% coverage) survived and by 1981 only 14 ha
(0.6% coverage) remained (Table 2, Fig. 3). In 1999, the few remnants of seagrasses in the
Cockburn Sound East region are associated with relatively shallow waters along its western
margin and have similar coverage to that of seagrasses in 1981.
In Cockburn Sound South, the area of shallow bank was only 38% of Cockburn Sound
East. One hundred and six hectares of seagrass meadows were lost between 1967 and 1972,
and a further 245 ha disappeared between 1972 and 1981 (Table 2). From 1981 to 1999, there
has been a slight increase (9 ha) in seagrass coverage, associated with growth of seagrasses
into sand blowouts on Southern Flats.
Seagrass loss in Cockburn Sound West has been less dramatic than Cockburn Sound
East and South, but has continued up to the present day (Table 2). Forty-one hectares of
seagrass meadows were lost between 1967 and 1972, a further 61 ha disappeared between
1972 and 1981, followed by 71 ha lost between 1981 and 1994, and 16 ha from 1994 and
1999 (Table 2). These losses are substantial, as Cockburn Sound West is only 33% of the
area of Cockburn Sound East.
Table 2
Change in seagrass coverage in Cockburn Sound for 1967, 1972, 1981, 1994 and 1999
Seagrass coverage
Region Mapping area (ha) 1967 1972 1981 1994 1999
ha % ha % ha % ha % ha %
Cockburn East 2139 1750 81.8 310 14.5 14 0.6 26 1.2 13 0.6
Cockburn South 818 634 77.5 528 64.6 283 34.6 291 35.6 292 35.7
Cockburn West 710 545 76.8 504 71.0 443 62.4 372 52.4 356 50.1
Total 3667 2929 79.9 1342 36.6 740 20.2 690 18.8 661 18.0
Data are presented as area in ha and as % of mapping area for each mapped region (Cockburn East, South and
West) and for Cockburn Sound in total.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 83
84 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Fig. 3. Mapped losses of seagrasses between 1967 and 1972, 1972 and 1981, 1981 and 1994 and 1994 and 1999.
4. Discussion
Historical seagrass loss in Cockburn Sound has been re-assessed from 1967 to 1999
using modern semi-automated mapping techniques on a common geodesic mapping base.
Broad-scale seagrass loss of 2087 ha occurred in Cockburn Sound between 1967 and 1981,
with the greatest area lost from the eastern and southern banks. Losses continued between
1981 and 1999 at much reduced spatial scales and mainly on the western margins (total of
79 ha between 1981 and 1999), reflecting more localised impacts, like port developments
at Careening and Sulphur Bays, and urchin outbreaks on north eastern Garden Island. Little
regeneration of seagrass meadows has occurred on the eastern and southern banks over
the two decades since seagrass decline, inferring that the shallow nearshore environment is
altered in some critical way that is counter to seagrass recolonisation.
The time course of seagrass decline in Cockburn Sound between 1967 and 1981 is
similar to that described between 1969 and 1978 by Cambridge and McComb (1984) and
Hillman (1986). The loss of seagrasses from the shallow subtidal bank along the eastern
shores of Cockburn Sound between 1967 and 1972 was extremely rapid in relation to the
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 85
lifespan of the dominant seagrass species, P. sinuosa. Despite P. sinuosa being a dominant
and widespread temperate seagrass, colonisation from seedlings and rhizome spread onto
unvegetated areas is extremely slow (probably on the order of tens of years) (Kirkman and
Kuo, 1990; Kirkman, 1998). The historical decline in seagrasses in Cockburn Sound is
specific to the Sound and the dominant seagrass species P. sinuosa. Over a similar 30 years
period (1965–1995) seagrasses have expanded their distribution by 529 ha on Parmelia
and Success Banks, despite losses caused by shellsand mining and pollution (Kendrick
et al., 2000). Parmelia and Success Banks are shallow sand banks immediately north of
Cockburn Sound. Seagrass expansion has been predominantly in P. coriacea and A. griffithii
assemblages (Kendrick et al., 1999).
The major disappearance of seagrasses recorded between 1967 and 1972 has been at-
tributed to shading of seagrass leaves by excessive growth of epiphytic algae, following
nitrogen enrichment of Cockburn Sound. (Cambridge et al., 1986). This hypothesis is sup-
ported by subsequent studies of epiphyte shading (Silberstein et al., 1986), in situ manipula-
tion of shading levels (Gordon et al., 1994), and photosynthesis-irradiance studies (Masini
et al., 1995; Department of Environmental Protection, 1996). One thousand and four hun-
dred and 40 ha of seagrasses disappeared in 5 years (1967–1972) on Cockburn Sound East,
and tended to overshadow more localised losses and reduction in seagrass depth limits.
These localised losses were attributable to other causes like turbidity, the result of scallop
dredging, harbour construction, dredge spoil dumping, and increased phytoplankton con-
centrations, overgrazing by sea urchins and direct removal of seagrass by channel dredging,
boat moorings and anchor drag (Cambridge and McComb, 1984).
Losses of seagrasses in the southern region of Cockburn Sound were greatest between
1972 and 1981. Losses started on the southeastern shore between 1967 and 1972 and contin-
ued westward between 1972 and 1981. The Garden Island causeway, which was constructed
between 1971 and 1973, appeared responsible for seagrasses disappearing from the South-
ern Flats (Hicks et al., 1973). Since the construction of the causeway, water exchange and
passage of ocean swells between the Indian Ocean and Cockburn Sound has been greatly
reduced (Cambridge et al., 1986). There have also been some increases in seagrass cover-
age associated with the filling by seagrasses of crescentic sand blowouts within meadows.
These blowouts were described by Cambridge (1975), but in 1999 were continuous sea-
grass meadow. Major seagrass losses from the shallow subtidal banks on the western shores
of Cockburn Sound also occurred between 1972 and 1982, including a shallowing of the
maximum depth where seagrasses were found, and seagrasses lost by Australian Navy Port
Developments in Careening (1972–1974) and Sulphur (1976–1978) Bays (Cambridge and
McComb, 1984).
Losses on Southern Flats and western Cockburn Sound after 1981 were small-scale and
included continued retreat of seagrass meadows into shallower water, and loss of seagrasses
caused by maintenance dredging, jetty building, and sea urchin grazing. Losses of seagrasses
from sea urchin grazing are generally well reported, but the effects are restricted in area
when compared to the more insidious and large-scale losses of seagrasses due to increased
light attenuation. For example, only 3 ha of shallow P. sinuosa meadows north of Sulphur
Bay disappeared between 1981 and 1994 as a result of an invasion by the grazing urchins
Heliocidaris erythrogramma and Temnopleuris michaelsenii in 1991 (Bancroft, 1992). Sim-
ilar localised loss of seagrass meadows due to invasions by T. michaelsenii have previously
86 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
been reported near Rockingham in 1972, Careening Bay in 1973, Woodman Point in 1976
and further south in Warnbro Sound in 1978 (Cambridge et al., 1986).
In conclusion, we have successfully mapped the distribution of seagrasses in Cockburn
Sound for the years, 1967, 1978, 1981, 1994 and 1999, to a common geodesic base, and
the seagrass coverage calculated for these years are now directly comparable. We have con-
firmed the extent of major loss in seagrasses in the 1970s as described by Cambridge and Mc-
Comb (1984). Between 1967 and 1972, 46% of seagrass coverage, or 1587 ha of seagrasses,
were lost from Cockburn Sound, mostly from the shallow subtidal banks on the eastern and
southern shores. By 1981, a further 602 ha had been lost from Cockburn Sound. Therefore,
by 1981, 75% of total seagrass coverage in Cockburn Sound recorded from 1967 was lost.
Since 1981, further seagrass losses (79 ha) has been restricted to a shallowing of the depth
limit of seagrasses, localised losses associated with port maintenance and a sea urchin out-
break. That there has been no recovery of seagrasses on the eastern shelf of Cockburn Sound
during two decades since nutrient loads were reduced in the 1980s, suggests that the shallow
shelf environment has been altered to an environment not suited for the recolonisation of
Posidonia species. This contrasts with Success Bank, 8 km north, where 529 ha of shallow
bank have been colonised by P. coriacea and A. griffithii between 1965 and 1995 (Kendrick
et al., 2000). This study now forms an accurate baseline of seagrass change between 1967
and 1999, for further monitoring and management of seagrasses in Cockburn Sound.
Acknowledgements
This project was funded by a consortium of industries and Western Australian Govern-
ment Departments: Cockburn Cement Pty Ltd., Departments of Environmental Protection,
Commerce and Trade and Resources Development, Fremantle Port Authority, James Point
Pty Ltd., Kwinana Industries Council, Royal Australian Navy, Water Corporation of Western
Australia and Waters and Rivers Commission of Western Australia. Peter Radford (Kevron
Aerial Surveys) coordinated the photographic flights, rectification of the high-level aerial
photography and the production of the photographic contact prints. Thanks to Meredith
Campey, Kären Crawley, Jamie Colquhoun, Naomi Telford, AJ Smit, Mark Westera, Chris
Wienczugow and Celeste Wilson for assisting in groundtruth surveys. Boyd Wykes and the
Navy Reserve Divers from HMAS Stirling assisted with the provision of personnel and
an underwater video for some of the groundtruth surveys. The Western Australian Depart-
ment of Conservation and Land Management also provided an underwater video for the
groundtruthing surveys. Thanks to Eric Paling and Di Walker for reviewing early drafts of
this manuscript. Also a heartfelt thanks for Dr. J.E. Vermaat, Co-Editor-in-Chief of Aquatic
Botany, for his detailed review and editorial comments that have greatly improved this
article.
References
Bancroft, K., 1992. Preliminary Investigations of Echinoid Aggregations in the Seagrass Meadow at Garden Island,
Western Australia. The Department of Biological and Environmental Sciences, Murdoch University, Western
Australia, unpublished report.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 87
Cambridge, M.L., 1975. Seagrasses of southwestern Australia with special reference to the ecology of Posidonia
australis Hook f. in a polluted environment. Aquat. Bot. 1, 149–162.
Cambridge, M.L., 1979. Cockburn Sound Environmental Study Technical Report on Seagrass. Report No. 7,
Department of Conservation and Environment, Perth.
Cambridge, M.L., Kuo, J., 1979. Two new species of seagrasses from Australia, Posidonia sinuosa and P.
angustifolia (Posidoniaceae). Aquat. Bot. 6, 307–328.
Cambridge, M.L., McComb, A.J., 1984. The loss of seagrasses in Cockburn Sound, Western Australia. I. The time
course and magnitude of seagrass decline in relation to industrial development. Aquat. Bot. 20, 229–243.
Cambridge, M.L., Chiffings, A.W., Brittain, C., Moore, L., McComb, A.J., 1986. The loss of seagrass in Cockburn
Sound, Western Australia. II. Possible causes of seagrass decline. Aquat. Bot. 24, 269–285.
Department of Environmental Protection, 1996. Southern Metropolitan Coastal Waters Study (1991–1994) Final
Report. Report No. 17, p. 288.
Earth Resource Mapping, 2000. ERMapper 6.1. http://www.ermapper.com.
Gordon, D.M., Chase, S.C., Grey, K.A., Simpson, C.J., 1994. Changes in the structure and productivity of a
Posidonia sinuosa meadow during and after imposed shading. Aquat. Bot. 47, 265–275.
Hicks, A.B., Meagher, T.D., Waterman, P., 1973. Planning of Marine Environmental Investigations for the Garden
Island Causeway, Cockburn Sound. Aust. Soc. Engin. Conf. Papers 1973, pp. 1–11.
Hillman, K., 1986. Nutrient load reduction, water quality and seagrass dieback in Cockburn Sound 1984–1985.
Technical Series 5. Department of Conservation and Environment, Perth, Western Australia.
Kendrick, G.A., Eckersley, J., Walker, D.I., 1999. Landscape-scale changes in seagrass distribution over time: a
case study of Success Bank, Western Australia. Aquat. Bot. 65, 293–309.
Kendrick, G.A., Hegge, B.J., Wyllie, A., Davidson, A., Lord, D.A., 2000. Changes in seagrass cover on Success
and Parmelia Banks, Western Australia between 1965 and 1995. Estuar. Coast Shelf Sci. 50, 341–353.
Kirkman, H., 1998. Pilot Experiments on planting seedlings and small seagrass propagules in Western Australia.
Mar. Pollut. Bull. 37, 460–467.
Kirkman, H., Kuo, J., 1990. Pattern and processes in southern Western Australian seagrasses. Aquat. Bot. 37,
367–382.
Marsh, L.M., Devaney, D.M., 1978. The Benthic Fauna of Cockburn Sound, Western Australia—Parts 2–4. The WA
Department of Conservation and Environment by the Western Australian Museum, Perth, Western Australia,
unpublished report, p. 23.
Masini, R.J., Cary, J.L., Simpson, C.J., McComb, A.J., 1995. Effects of light and temperature on the photosynthesis
of temperate meadow-forming seagrasses in Western Australia. Aquat. Bot. 49, 239–254.
Neteler, M., 1998. Introduction to GRASS GIS Software. University of Hannover, Germany. Based on: Introduction
to GRASS GIS Software Institute of Remote Sensing, Anna University, Chennai, Madras, India.
Short, F.T.T., Wyllie-Echeverria, S., 1996. Natural and human-induced disturbance of seagrasses. Environ. Conserv.
23, 17–27.
Silberstein, K., Chiffings, A.W., McComb, A.J., 1986. The loss of seagrasses in cockburn sound, Western Australia.
III. The effect of epiphytes on productivity of Posidonia australis Hook f. Aquat. Bot. 24, 355–371.
Spann, M., Wilson, R., 1985. A quad-tree approach to image segmentation which combines statistical and spatial
information. Pattern Recog. 18, 257–0269.
The University of Oslo, 1999. XITE—X-based image-processing tools and environment.
http://www.ifi.uio.no/∼blab/Programvare/Xite/.
Walker, D.I., McComb, A.J., 1992. Seagrass degradation in Australian coastal waters. Mar. Pollut. Bull. 25,
191–195.
Wilson, B.R., Kendrick, G.W., Brearley, A., 1978. The Benthic Fauna of Cockburn Sound, Western Australia—Part
1. WA Department of Conservation and Environment, Western Australian Museum, Perth, Western Australia,
unpublished report.
Changes in seagrass coverage in Cockburn Sound,
Western Australia between 1967 and 1999
Gary A. Kendrick a,∗ , Matt J. Aylward a,c , Bruce J. Hegge b ,
Marion L. Cambridge a , Karen Hillman b ,
Alex Wyllie c , Des A. Lord b
aDepartment of Botany, Faculty of Science, University of Western Australia, 35 Stirling Highway,
Crawley, WA 6009, Australia
b D.A. Lord and Associates Pty Ltd., P.O. Box 3172, LPO Broadway, Nedlands, WA 6009, Australia
c Alex Wyllie and Associates, 10 Kershaw Gardens, Leeming, WA 6149, Australia
Received 20 March 2001; received in revised form 20 December 2001; accepted 20 December 2001
Abstract
Changes in seagrass coverage in Cockburn Sound from 1967 to 1999 were assessed from aerial
photographs using modern mapping methods with the aim of accurately determining the magnitude
of change in hectares of seagrasses between 1967 and 1999 and to set up a baseline for future
monitoring of seagrass loss in Cockburn Sound. Firstly, coverage and assemblages of seagrasses in
Cockburn Sound were mapped using the best available aerial photographs from 1999, rectified to a
common geodesic base with comprehensive groundtruth information, and with a semi-automated
mapping algorithm. Then the same technique was used to map historical seagrass coverage in
Cockburn Sound from aerial photographs taken in 1967, 1972, 1981 and 1994.
The seagrass coverage in Cockburn Sound has declined by 77% since 1967. Between 1967 and
1972, 1587 ha of seagrass, were lost from Cockburn Sound, mostly from shallow subtidal banks on
the eastern and southern shores. By 1981, a further 602 ha had been lost. Since 1981, further seagrass
losses (79 ha) have been restricted to a shallowing of the depth limit of seagrasses, localised losses
associated with port maintenance and a sea urchin outbreak on inshore northern Garden Island.
There has been no recovery of seagrasses on the eastern shelf of Cockburn Sound after nutrient
loads were reduced in the 1980s, suggesting that this shallow shelf environment has been altered
to an environment not suited for large-scale recolonisation by Posidonia species. © 2002 Elsevier
Science B.V. All rights reserved.
Keywords: Mapping; Aerial photography; Seagrass decline; Posidonia spp.; Australia
∗ Corresponding author. Tel.: +61-8-9380-3998; fax: +61-8-9380-1001.
E-mail address: garyk@cyllene.uwa.edu.au (G.A. Kendrick).
0304-3770/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 4 - 3 7 7 0 ( 0 2 ) 0 0 0 0 5 - 0
76 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
1. Introduction
Large-scale losses of seagrass meadows associated with human impacts have occurred
in Australia (Walker and McComb, 1992) and worldwide (Short and Wyllie-Echeverria,
1996). One of the most commonly cited cases of large-scale seagrass loss happened in
Cockburn Sound, Western Australia following increased industrial discharge and expan-
sion of port facilities (Cambridge, 1979; Cambridge and McComb, 1984; Hillman, 1986).
Cambridge and McComb (1984) documented a loss of 3400 ha of seagrass meadows be-
tween 1969 and 1978, with most losses occurring before 1975. Their study has recently
been criticised heavily by local marine industries and management agencies because un-
rectified aerial photographs were used, the mapping area was poorly defined, and the 1969
map of seagrass distribution in the eastern shelf of Cockburn Sound was mostly determined
from field observations, as the aerial photography was of poor quality (Cambridge and
McComb, 1984). This study redresses these limitations, focuses on the years of major his-
torical loss and extends the work of Cambridge and McComb (1984) 20 years to the present
day.
Cockburn Sound has historically been used for recreational boating and commercial fish-
ing, but its relatively protected deep waters also provide an excellent anchorage, and in 1954
it was designated as an industrial harbour for the Perth-Fremantle region (Cambridge and
McComb, 1984). Industrial development commenced with the establishment of an oil refin-
ery in 1955, followed by the addition of iron, steel, alumina and nickel processing and refin-
ing plants, chemical and fertiliser production plants and a bulk grain terminal. In conjunction
with the development of these heavy industries, wharves and groynes were built and chan-
nels dredged for shipping access. A wastewater treatment plant was commissioned in 1966
which discharged into the northern end of the Sound, while at the southern end of the Sound,
a 2.5 km long causeway was constructed between 1971 and 1973 to connect Garden Island
to the mainland (Hicks et al., 1973). This causeway is interrupted by two trestle bridges (one
305 m long and other 610 m long), through which limited exchange of water with the ocean
occurs.
Industrial expansion resulted in a decline in seagrass coverage in Cockburn Sound from
4200 to 900 ha between 1954 and 1978, with most loss occurring between 1969 and 1975
(Cambridge and McComb, 1984). The Western Australian Government funded a 3 year
study (1976–1979; Cambridge, 1979), which identified that the decline in seagrass cover-
age was linked to an increase in nitrogen loading in the Sound. Over 90% of the nitrogen
contributed by industrial effluent and wastewater to the Sound came from two sources:
the outlet shared by the CSBP fertiliser works and the Kwinana Nitrogen Company, and
the outlet of the Woodman Point Wastewater Treatment Plant (Cambridge et al., 1986).
The proposed mechanism for seagrass loss was that increased nitrogen levels led to en-
hanced growth of nuisance epiphytic algae and consequent shading and deterioration of the
seagrasses (Cambridge et al., 1986).
Water quality in Cockburn Sound has improved greatly since the late 1970s (Hillman,
1986; Hillman, unpublished data). The annual loading of nitrogen has decreased from
2000 t N per year in 1978 to less than 500 t N per year in 1997 (Hillman, unpublished
data). The northeastern shore of Cockburn Sound still has high N levels from contaminated
groundwater discharging into the sound. This contaminated groundwater contributed 70%
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 77
of the total nitrogen entering Cockburn Sound in 1995 (Department of Environmental
Protection, 1996).
There has not been a reassessment of seagrass distribution in Cockburn Sound for the
past 20 years since the work of Cambridge and McComb (1984). This paper redresses
this by mapping seagrass coverage in 1981, 1994 and 1999 and revisits the period of
maximum seagrass loss between 1967 and 1972. We map coverage of assemblages of
seagrasses in Cockburn Sound using the best available aerial photographs from 1999, rec-
tified to a common geodesic base with comprehensive groundtruth information (Kendrick
et al., 2000). This is then compared to maps of historical seagrass coverage in Cock-
burn Sound constructed from historical aerial photographs taken in 1967, 1972, 1981
and 1994.
2. Methods
2.1. Mapping of seagrasses
To determine seagrass distribution, submerged vegetation was mapped from recent and
historical aerial photography. Distribution of seagrass assemblages and reef were then de-
termined in 1999 from towed underwater video. We define assemblages of seagrasses as
multi-species assemblages dominated, or characterised, by single or multiple species as
determined from their relative abundance in video footage. In Cockburn Sound, the sea-
grass species Amphibolis antarctica, Amphibolis griffithii, Posidonia australis, Posidonia
coriacea, Posidonia sinuosa, Halophila ovalis, Heterozostera tasmanica and Syringodium
isoetifolium were components of the seagrass assemblages. We did not map to single species,
although a single species assemblage is composed of more than 70% of that species. Results
from mapping of aerial photographs and underwater video footage were then combined in
a GIS to create coverage maps of seagrass assemblages, reef and unvegetated sand. We
describe our maps as coverage maps, rather than maps of seagrass cover to reduce the
confusion between the area covered by a single species of seagrass and the area occu-
pied by an assemblage of seagrasses than is dominated by one or a few species. This is
not presence–absence mapping, as vegetated assemblages do not exclude all unvegetated
habitat (see Section 2.6).
2.2. Study area and mapped regions
Cockburn Sound is a sheltered marine embayment 16 km long × 9 km wide, and consists
of a deep central basin (17–22 m deep) surrounded by shallow platforms. The shallow
platforms vary in width from 50 m to 3 km and are where seagrass meadows are found
(Fig. 1). The seagrass mapping area was delineated as shallow platforms to a depth of 10 m
and covered an area of approximately 3667 ha (Table 1). The mapped area of the present
study is smaller than that used by Cambridge and McComb (1984), as they incorporated an
undefined proportion of Parmelia Bank into their mapped area and never defined a depth limit
for their study. The total area mapped by Cambridge and McComb (1984) was 4200 ha.
78 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Fig. 1. Map of Cockburn Sound showing the extent of seagrass mapping bounded by the coastline and 10 m isobath,
the three mapping regions (Cockburn Sound West, South and East), locations mentioned in the text, and the 1999
mapped distribution of seagrasses and reef.
Historical changes in seagrass coverage of Parmelia Bank have been recently published
(Kendrick et al., 2000) so have not been included in this paper.
Coverage of all submerged vegetation was separately calculated for three regions called
Cockburn Sound East, South and West (Fig. 1, Table 1). Cockburn Sound East encompassed
most of the eastern bank from Woodmans Point to south of James Point. Cockburn Sound
South contained the shallow bank near Rockingham, the southern sand flats and Careening
Bay, Garden Island. Cockburn Sound West encompassed the western bank north of Ca-
reening Bay and shallow waters north of Garden Island. Benthic features are difficult to
resolve from aerial photographs at depths greater than 10 m as seagrasses become sparsely
distributed at these depths. Hence, the 10 m isobath and the coastline are used to delineate
the mapping boundaries within each of these regions.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 79
Table 1
Area of seagrass assemblages in 1999 for each mapped region (Cockburn East, South and West) and for Cockburn
Sound in total
Mapping region Mapping Posidonia Posidonia Posidonia Total Reef Shallow
area (ha) australis sinuosa sinuosa and seagrass (ha) bare sand
(ha) (ha) Posidonia (ha) (ha)
australis (ha)
Cockburn Sound East 2138 0 13 0 13 25 2100
Cockburn Sound South 818 11 279 2 292 0 526
Cockburn Sound West 711 0 356 0 356 21 334
Total 3667 11 648 2 661 46 2960
Data area represented as area in hectares (ha).
2.3. Data sources
In 1999, flights to acquire aerial photographs were purpose-flown in late-February to
early-March when conditions were optimal. These conditions included maximum sea-
grass leaf cover; maximum water clarity; minimum haze from the adjacent industrial zone;
minimum turbidity from dredging, river outflow or industrial activities; minimum cloud
cover; weak prevailing winds; and incident sun angle at 20–30◦ . Colour photography with
a yellow lens filter was obtained at two altitudes. High altitude imagery was collected
at 10,000 m (scale, 1:55,000) for accurate ortho-rectification. The resulting imagery was
used as a rectification base for the low-altitude imagery that was obtained at an altitude of
3,800 m.
Mapping of the historical seagrass coverage in Cockburn Sound was conducted using
rectified and mosaicked imagery obtained in: 1994 (majority of imagery obtained on 4 and
8 January 1994 and some on 6 January 1995); 1981 (imagery from 13 June 1981); 1972
(imagery from 2 May 1972); and 1967 (imagery from 20 March 1967).
2.4. Image geo-referencing and rectification
Images were initially rectified (Datum: WGS-84) using ERMapper-6.1 to a resolution
of 2.0 m (Earth Resource Mapping, 2000). The images were initially scanned into three
colour bands (red, green and blue). The red band of the colour images contained almost
no information for marine benthic habitat, as red light was severely attenuated through the
water column. The green band was better than the red, though images had low contrast due
to atmospheric attenuation. The blue band provided the best contrast between vegetated and
unvegetated habitats and therefore was the only band used for mapping. It was converted
to 256 greyscales.
Rectified imagery was processed using Geographic Resources Analysis Support System
(GRASS) GIS software (Neteler, 1998). GRASS is a raster-based GIS and the 2.0 m cell
size and its spatial location (spatial integrity) was strictly maintained in rectified imagery.
Also the same spatial integrity was kept consistent among the rectified imagery for the
different years mapped (spatial consistency).
80 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
2.5. Automated mapping of submerged vegetation
A computerised, semi-automated, greyscale segmentation mapping method called
Spann–Wilson segmentation (Spann and Wilson, 1985) was employed to map submerged
vegetation. Spann–Wilson combines locally adaptive segmentation (local centroid) with
pre-processing using multi-level quad-tree smoothing. The segmentation method was imple-
mented using the Xite image-processing software (The University of Oslo, 1999). The size of
the moving histogram window and the number of quad-tree smoothing levels were controlled
by the operator, although for this exercise were set to 324 m2 and 3–4. Spann–Wilson seg-
mentation was very effective in defining the boundaries of seagrass meadows reefs and un-
vegetated sand from greyscale images. It reduced the original 256 greyscales from the aerial
photographs to four greyscales. The operator then chose a greyscale which most closely
coincided with the visually interpreted boundary between vegetated and unvegetated habitat.
2.6. Control rules for mapping
The vegetated areas were distinguished from the unvegetated areas as they had a distinct
photo-tone of medium to dark grey. To enable consistent coverage mapping across the study
area, a series of control rules were used (Kendrick et al., 2000). These control rules were
as follows:
• Isolated vegetated patches less than 30 m2 in area were not mapped.
• Vegetated patches, that were greater than 30 m2 and less than 100 m2 were mapped as
separate patches when the distance between one patch and another was greater than the
diameter of the patch.
• Vegetated patches, that were greater than 30 m2 and less than 100 m2 were mapped as
a single meadow when the distance between one patch and another was less than the
diameter of the patch. Unvegetated areas within the meadow with an area greater than
100 m2 were mapped.
• Vegetated patches greater than 100 m2 were mapped and the edges of these areas were
traced accurately. Unvegetated regions within these patches with areas greater than
100 m2 were mapped.
The control rules were applied automatically during mapping with control rules 2–4
processed during the Spann–Wilson segmentation step to a 324 m2 moving window, and
control rule 1 applied after the segmentation step to 1 km2 areas using a PERL script.
2.7. Groundtruth surveys
Detailed groundtruth surveys were conducted between late-February and May 1999.
These groundtruth surveys were undertaken using a differential GPS combined with diver-
operated towed video (manta tow) and a downward-looking surface deployed (drop-down)
video. Following the completion of the formal groundtruth surveys, a series of opportunistic
dives were also undertaken to establish the assemblage type in specific areas of interest. A
total of four manta tow transects (8 km in total) and 114 drop-down video locations were
surveyed during the groundtruthing exercise.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 81
The manta tow videos were analysed by pausing the video at 20 s intervals (corresponding
with the differential GPS waypoints). At each of these video pauses the percent of the
image each seagrass species and habitat type was recorded. For the drop-down videos,
percent representation of species of seagrass and habitats was averaged across the total
video footage obtained from each drop unless the species composition varied noticeably. If
it was variable, the drop was sub-divided to more accurately represent the seagrass coverage.
Seagrass species recognised in the video footage were A. antarctica, A. griffithii, P.
australis, P. coriacea, P. sinuosa, H. ovalis, H. tasmanica and S. isoetifolium. Habitats
other than seagrasses that were recognised were limestone reef and unvegetated sand.
It was not possible to separate the species of P. sinuosa and Posidonia angustifolia from
the video data, as this requires an examination of the rhizome fibres (Cambridge and Kuo,
1979). Groundtruth dives indicated that P. sinuosa was generally more common than P.
angustifolia (greater than 90% except in the Cockburn Sound East region where P. sinuosa
cover was approximately 70%). In the present mapping exercise, these two species were
both mapped as the P. sinuosa assemblage.
2.8. Mapping of seagrass assemblages and benthic habitats
The maps of the 1999 vegetated and unvegetated areas were combined with the groundtruth
data using GRASS GIS to produce a benthic assemblage map. Within the mapped region,
the seagrass assemblages P. sinuosa, mixed species P. sinuosa and P. australis, and P. aus-
tralis were distinguished as these species were the dominants. Single species assemblages
were defined when a single species had greater than 70% representation on videos and the
P. sinuosa and P. australis assemblage was defined when P. sinuosa and P. australis were
equally represented (representation of any one species could range between 30 and 70%).
Shallow unvegetated sand (<10 m depth) and reef were also mapped.
Mapping of historical seagrass coverage in Cockburn Sound was conducted using rectified
and mosaicked imagery obtained for 1967, 1972, 1981 and 1994. Coverage of all seagrass
assemblages combined was calculated for the three regions within Cockburn Sound sepa-
rately. The reef areas which were identified in the 1999 mapping exercise were transposed
onto the historical images from Cockburn Sound, using the assumption that the size and
location of these reef areas had not changed over the mapping period. Previous ground truth
data were collected from a range of sources, including those of Cambridge (1979), Marsh
and Devaney (1978) and Wilson et al. (1978), and used as an aid in interpreting historical
seagrass coverage.
3. Results
3.1. Seagrass coverage in 1999
In 1999, seagrasses occupied an area of 661 ha (18%) of sandbanks less than 10 m in
depth (3667 ha) in Cockburn Sound (Table 1, Fig. 1). Extensive seagrass meadows were
found predominantly on the western margin of Cockburn Sound (Cockburn Sound West,
82 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Table 1) and in the Southern Flats and western Rockingham regions (Cockburn Sound
South, Table 1). The seagrass meadows in 1999 were predominantly composed of P. sinu-
osa assemblages with a small area on the Southern Flats dominated by assemblages of P.
australis (Table 1). Besides these dominant species, small proportions of A. antarctica, A.
griffithii, P. coriacea, H. ovalis, H. tasmanica and S. isoetifolium were observed.
3.2. Time course of seagrass loss
In 1967, seagrasses formed an almost continuous 2929 ha meadow, between 1 and 10 m
depths, that fringed the eastern, southern and western margins of Cockburn Sound (Fig. 2).
Between 1967 and 1981, all but a few isolated patches of seagrass were lost from Cockburn
Sound East and the eastern margins of Cockburn Sound South (Figs. 2 and 3). Seagrass
loss was broad-scale with 1587 ha of seagrasses disappearing between 1967 and 1972,
and 602 ha between 1972 and 1981 (Fig. 3, Table 2). A further 79 ha of seagrasses have
disappeared between 1981 and 1999, mostly from Cockburn Sound West, but the scale of
loss was smaller and more localized.
In Cockburn Sound East, 1750 ha of seagrasses covered 82% of the shallow bank habi-
tat in 1967, but by 1972 only 310 ha (14.5% coverage) survived and by 1981 only 14 ha
(0.6% coverage) remained (Table 2, Fig. 3). In 1999, the few remnants of seagrasses in the
Cockburn Sound East region are associated with relatively shallow waters along its western
margin and have similar coverage to that of seagrasses in 1981.
In Cockburn Sound South, the area of shallow bank was only 38% of Cockburn Sound
East. One hundred and six hectares of seagrass meadows were lost between 1967 and 1972,
and a further 245 ha disappeared between 1972 and 1981 (Table 2). From 1981 to 1999, there
has been a slight increase (9 ha) in seagrass coverage, associated with growth of seagrasses
into sand blowouts on Southern Flats.
Seagrass loss in Cockburn Sound West has been less dramatic than Cockburn Sound
East and South, but has continued up to the present day (Table 2). Forty-one hectares of
seagrass meadows were lost between 1967 and 1972, a further 61 ha disappeared between
1972 and 1981, followed by 71 ha lost between 1981 and 1994, and 16 ha from 1994 and
1999 (Table 2). These losses are substantial, as Cockburn Sound West is only 33% of the
area of Cockburn Sound East.
Table 2
Change in seagrass coverage in Cockburn Sound for 1967, 1972, 1981, 1994 and 1999
Seagrass coverage
Region Mapping area (ha) 1967 1972 1981 1994 1999
ha % ha % ha % ha % ha %
Cockburn East 2139 1750 81.8 310 14.5 14 0.6 26 1.2 13 0.6
Cockburn South 818 634 77.5 528 64.6 283 34.6 291 35.6 292 35.7
Cockburn West 710 545 76.8 504 71.0 443 62.4 372 52.4 356 50.1
Total 3667 2929 79.9 1342 36.6 740 20.2 690 18.8 661 18.0
Data are presented as area in ha and as % of mapping area for each mapped region (Cockburn East, South and
West) and for Cockburn Sound in total.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 83
84 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
Fig. 3. Mapped losses of seagrasses between 1967 and 1972, 1972 and 1981, 1981 and 1994 and 1994 and 1999.
4. Discussion
Historical seagrass loss in Cockburn Sound has been re-assessed from 1967 to 1999
using modern semi-automated mapping techniques on a common geodesic mapping base.
Broad-scale seagrass loss of 2087 ha occurred in Cockburn Sound between 1967 and 1981,
with the greatest area lost from the eastern and southern banks. Losses continued between
1981 and 1999 at much reduced spatial scales and mainly on the western margins (total of
79 ha between 1981 and 1999), reflecting more localised impacts, like port developments
at Careening and Sulphur Bays, and urchin outbreaks on north eastern Garden Island. Little
regeneration of seagrass meadows has occurred on the eastern and southern banks over
the two decades since seagrass decline, inferring that the shallow nearshore environment is
altered in some critical way that is counter to seagrass recolonisation.
The time course of seagrass decline in Cockburn Sound between 1967 and 1981 is
similar to that described between 1969 and 1978 by Cambridge and McComb (1984) and
Hillman (1986). The loss of seagrasses from the shallow subtidal bank along the eastern
shores of Cockburn Sound between 1967 and 1972 was extremely rapid in relation to the
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 85
lifespan of the dominant seagrass species, P. sinuosa. Despite P. sinuosa being a dominant
and widespread temperate seagrass, colonisation from seedlings and rhizome spread onto
unvegetated areas is extremely slow (probably on the order of tens of years) (Kirkman and
Kuo, 1990; Kirkman, 1998). The historical decline in seagrasses in Cockburn Sound is
specific to the Sound and the dominant seagrass species P. sinuosa. Over a similar 30 years
period (1965–1995) seagrasses have expanded their distribution by 529 ha on Parmelia
and Success Banks, despite losses caused by shellsand mining and pollution (Kendrick
et al., 2000). Parmelia and Success Banks are shallow sand banks immediately north of
Cockburn Sound. Seagrass expansion has been predominantly in P. coriacea and A. griffithii
assemblages (Kendrick et al., 1999).
The major disappearance of seagrasses recorded between 1967 and 1972 has been at-
tributed to shading of seagrass leaves by excessive growth of epiphytic algae, following
nitrogen enrichment of Cockburn Sound. (Cambridge et al., 1986). This hypothesis is sup-
ported by subsequent studies of epiphyte shading (Silberstein et al., 1986), in situ manipula-
tion of shading levels (Gordon et al., 1994), and photosynthesis-irradiance studies (Masini
et al., 1995; Department of Environmental Protection, 1996). One thousand and four hun-
dred and 40 ha of seagrasses disappeared in 5 years (1967–1972) on Cockburn Sound East,
and tended to overshadow more localised losses and reduction in seagrass depth limits.
These localised losses were attributable to other causes like turbidity, the result of scallop
dredging, harbour construction, dredge spoil dumping, and increased phytoplankton con-
centrations, overgrazing by sea urchins and direct removal of seagrass by channel dredging,
boat moorings and anchor drag (Cambridge and McComb, 1984).
Losses of seagrasses in the southern region of Cockburn Sound were greatest between
1972 and 1981. Losses started on the southeastern shore between 1967 and 1972 and contin-
ued westward between 1972 and 1981. The Garden Island causeway, which was constructed
between 1971 and 1973, appeared responsible for seagrasses disappearing from the South-
ern Flats (Hicks et al., 1973). Since the construction of the causeway, water exchange and
passage of ocean swells between the Indian Ocean and Cockburn Sound has been greatly
reduced (Cambridge et al., 1986). There have also been some increases in seagrass cover-
age associated with the filling by seagrasses of crescentic sand blowouts within meadows.
These blowouts were described by Cambridge (1975), but in 1999 were continuous sea-
grass meadow. Major seagrass losses from the shallow subtidal banks on the western shores
of Cockburn Sound also occurred between 1972 and 1982, including a shallowing of the
maximum depth where seagrasses were found, and seagrasses lost by Australian Navy Port
Developments in Careening (1972–1974) and Sulphur (1976–1978) Bays (Cambridge and
McComb, 1984).
Losses on Southern Flats and western Cockburn Sound after 1981 were small-scale and
included continued retreat of seagrass meadows into shallower water, and loss of seagrasses
caused by maintenance dredging, jetty building, and sea urchin grazing. Losses of seagrasses
from sea urchin grazing are generally well reported, but the effects are restricted in area
when compared to the more insidious and large-scale losses of seagrasses due to increased
light attenuation. For example, only 3 ha of shallow P. sinuosa meadows north of Sulphur
Bay disappeared between 1981 and 1994 as a result of an invasion by the grazing urchins
Heliocidaris erythrogramma and Temnopleuris michaelsenii in 1991 (Bancroft, 1992). Sim-
ilar localised loss of seagrass meadows due to invasions by T. michaelsenii have previously
86 G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87
been reported near Rockingham in 1972, Careening Bay in 1973, Woodman Point in 1976
and further south in Warnbro Sound in 1978 (Cambridge et al., 1986).
In conclusion, we have successfully mapped the distribution of seagrasses in Cockburn
Sound for the years, 1967, 1978, 1981, 1994 and 1999, to a common geodesic base, and
the seagrass coverage calculated for these years are now directly comparable. We have con-
firmed the extent of major loss in seagrasses in the 1970s as described by Cambridge and Mc-
Comb (1984). Between 1967 and 1972, 46% of seagrass coverage, or 1587 ha of seagrasses,
were lost from Cockburn Sound, mostly from the shallow subtidal banks on the eastern and
southern shores. By 1981, a further 602 ha had been lost from Cockburn Sound. Therefore,
by 1981, 75% of total seagrass coverage in Cockburn Sound recorded from 1967 was lost.
Since 1981, further seagrass losses (79 ha) has been restricted to a shallowing of the depth
limit of seagrasses, localised losses associated with port maintenance and a sea urchin out-
break. That there has been no recovery of seagrasses on the eastern shelf of Cockburn Sound
during two decades since nutrient loads were reduced in the 1980s, suggests that the shallow
shelf environment has been altered to an environment not suited for the recolonisation of
Posidonia species. This contrasts with Success Bank, 8 km north, where 529 ha of shallow
bank have been colonised by P. coriacea and A. griffithii between 1965 and 1995 (Kendrick
et al., 2000). This study now forms an accurate baseline of seagrass change between 1967
and 1999, for further monitoring and management of seagrasses in Cockburn Sound.
Acknowledgements
This project was funded by a consortium of industries and Western Australian Govern-
ment Departments: Cockburn Cement Pty Ltd., Departments of Environmental Protection,
Commerce and Trade and Resources Development, Fremantle Port Authority, James Point
Pty Ltd., Kwinana Industries Council, Royal Australian Navy, Water Corporation of Western
Australia and Waters and Rivers Commission of Western Australia. Peter Radford (Kevron
Aerial Surveys) coordinated the photographic flights, rectification of the high-level aerial
photography and the production of the photographic contact prints. Thanks to Meredith
Campey, Kären Crawley, Jamie Colquhoun, Naomi Telford, AJ Smit, Mark Westera, Chris
Wienczugow and Celeste Wilson for assisting in groundtruth surveys. Boyd Wykes and the
Navy Reserve Divers from HMAS Stirling assisted with the provision of personnel and
an underwater video for some of the groundtruth surveys. The Western Australian Depart-
ment of Conservation and Land Management also provided an underwater video for the
groundtruthing surveys. Thanks to Eric Paling and Di Walker for reviewing early drafts of
this manuscript. Also a heartfelt thanks for Dr. J.E. Vermaat, Co-Editor-in-Chief of Aquatic
Botany, for his detailed review and editorial comments that have greatly improved this
article.
References
Bancroft, K., 1992. Preliminary Investigations of Echinoid Aggregations in the Seagrass Meadow at Garden Island,
Western Australia. The Department of Biological and Environmental Sciences, Murdoch University, Western
Australia, unpublished report.
G.A. Kendrick et al. / Aquatic Botany 73 (2002) 75–87 87
Cambridge, M.L., 1975. Seagrasses of southwestern Australia with special reference to the ecology of Posidonia
australis Hook f. in a polluted environment. Aquat. Bot. 1, 149–162.
Cambridge, M.L., 1979. Cockburn Sound Environmental Study Technical Report on Seagrass. Report No. 7,
Department of Conservation and Environment, Perth.
Cambridge, M.L., Kuo, J., 1979. Two new species of seagrasses from Australia, Posidonia sinuosa and P.
angustifolia (Posidoniaceae). Aquat. Bot. 6, 307–328.
Cambridge, M.L., McComb, A.J., 1984. The loss of seagrasses in Cockburn Sound, Western Australia. I. The time
course and magnitude of seagrass decline in relation to industrial development. Aquat. Bot. 20, 229–243.
Cambridge, M.L., Chiffings, A.W., Brittain, C., Moore, L., McComb, A.J., 1986. The loss of seagrass in Cockburn
Sound, Western Australia. II. Possible causes of seagrass decline. Aquat. Bot. 24, 269–285.
Department of Environmental Protection, 1996. Southern Metropolitan Coastal Waters Study (1991–1994) Final
Report. Report No. 17, p. 288.
Earth Resource Mapping, 2000. ERMapper 6.1. http://www.ermapper.com.
Gordon, D.M., Chase, S.C., Grey, K.A., Simpson, C.J., 1994. Changes in the structure and productivity of a
Posidonia sinuosa meadow during and after imposed shading. Aquat. Bot. 47, 265–275.
Hicks, A.B., Meagher, T.D., Waterman, P., 1973. Planning of Marine Environmental Investigations for the Garden
Island Causeway, Cockburn Sound. Aust. Soc. Engin. Conf. Papers 1973, pp. 1–11.
Hillman, K., 1986. Nutrient load reduction, water quality and seagrass dieback in Cockburn Sound 1984–1985.
Technical Series 5. Department of Conservation and Environment, Perth, Western Australia.
Kendrick, G.A., Eckersley, J., Walker, D.I., 1999. Landscape-scale changes in seagrass distribution over time: a
case study of Success Bank, Western Australia. Aquat. Bot. 65, 293–309.
Kendrick, G.A., Hegge, B.J., Wyllie, A., Davidson, A., Lord, D.A., 2000. Changes in seagrass cover on Success
and Parmelia Banks, Western Australia between 1965 and 1995. Estuar. Coast Shelf Sci. 50, 341–353.
Kirkman, H., 1998. Pilot Experiments on planting seedlings and small seagrass propagules in Western Australia.
Mar. Pollut. Bull. 37, 460–467.
Kirkman, H., Kuo, J., 1990. Pattern and processes in southern Western Australian seagrasses. Aquat. Bot. 37,
367–382.
Marsh, L.M., Devaney, D.M., 1978. The Benthic Fauna of Cockburn Sound, Western Australia—Parts 2–4. The WA
Department of Conservation and Environment by the Western Australian Museum, Perth, Western Australia,
unpublished report, p. 23.
Masini, R.J., Cary, J.L., Simpson, C.J., McComb, A.J., 1995. Effects of light and temperature on the photosynthesis
of temperate meadow-forming seagrasses in Western Australia. Aquat. Bot. 49, 239–254.
Neteler, M., 1998. Introduction to GRASS GIS Software. University of Hannover, Germany. Based on: Introduction
to GRASS GIS Software Institute of Remote Sensing, Anna University, Chennai, Madras, India.
Short, F.T.T., Wyllie-Echeverria, S., 1996. Natural and human-induced disturbance of seagrasses. Environ. Conserv.
23, 17–27.
Silberstein, K., Chiffings, A.W., McComb, A.J., 1986. The loss of seagrasses in cockburn sound, Western Australia.
III. The effect of epiphytes on productivity of Posidonia australis Hook f. Aquat. Bot. 24, 355–371.
Spann, M., Wilson, R., 1985. A quad-tree approach to image segmentation which combines statistical and spatial
information. Pattern Recog. 18, 257–0269.
The University of Oslo, 1999. XITE—X-based image-processing tools and environment.
http://www.ifi.uio.no/∼blab/Programvare/Xite/.
Walker, D.I., McComb, A.J., 1992. Seagrass degradation in Australian coastal waters. Mar. Pollut. Bull. 25,
191–195.
Wilson, B.R., Kendrick, G.W., Brearley, A., 1978. The Benthic Fauna of Cockburn Sound, Western Australia—Part
1. WA Department of Conservation and Environment, Western Australian Museum, Perth, Western Australia,
unpublished report.