Assessing Ecosystem Recovery in Transplanted
Posidonia australis at Southern Flats, Cockburn Sound
Ian Dapson
Murdoch University
School of Biological Sciences
2011
ii
Declaration
This thesis is an account of my own research and has not been previously published or
submitted at any tertiary institution, except for where acknowledgement has been made in
the text.
Ian Dapson
November 2011
iii
Abstract
Following on from the large scale loss of seagrass in Cockburn Sound and extensive transplanting
of Posidonia australis which had taken place on Southern Flats, assessment of the recovery of the
seagrass benthic infauna ecosystems was undertaken. Samples from the outer, middle and centre
edge zones of four different density transplant plots (1 m, 0.5 m, 0.25 m and 0.125 m spacing)
located within a larger transplantation meadow were compared against two natural meadows
and a bare sand site. Four years after transplantation the 0.25 and 0.125 m Plots had shoot
densities comparable to those of the natural seagrass sites with a two-way ANOVA revealing
significant effects of site and edge zone on the seagrass shoot density. Total infauna abundance
and infauna assemblages within the 0.25 and 0.125 m Plots had reached equivalent level to the
natural meadows but not at the 1 and 0.5 m Plots. A two-way ANOVA showed a significant
difference in the total infauna abundance between the different sites but no significant edge
effect was detected. Eusiridae, Solecurtidae, Diogenidae, Columbellidae, Fissurellidae, Oweniidae
and Ischnochitonidae were found to occur in the two natural meadows and in the 0.25 and 0.125
m Plots and may be climax or K-species indicating the recovery of the transplanted seagrass to
natural levels. The transplanted seagrass was also found to support small numbers of pipefish,
seahorses and a sea lion. From this study it can be seen that the shoot densities and infauna
abundances and assemblages of the 0.25 and 0.125 m Plots have reached levels comparable the
nearby natural meadows and that those of the 1 and 0.5 m Plots are likely to reach comparable
level another in one to two years.
iv
Acknowledgements
I wish to thank both my supervisors Dr Jennifer Verduin and Dr Mike Van Keulen for suggesting
the project and helping with the field work, as well as their constructive feedback on my work.
I would also like to thank Rhiannon Jones, Steve Goynich, Anka Seidlitz and Mike Taylor for their
assistance with skippering the boat and assisting with the diving (sorry about the cold Steve!). A
special thanks to Rhiannon Jones for her assistance with launching and skippering the boat which
provided much amusement during the very cold and wet field work.
For their assistance with the arduous task of sorting the infauna I would like to express my
gratitude to Aurelie Labbe, Alisia Lampropoulos and Holly Poole. Without their help I would
probably still be in the lab sorting infauna.
My thanks to Andrew Hosie, Stacey Osborne and Genefor Walker-Smith for their assistance with
identifying some of the tricky infauna and putting me on the right track, and additional thanks to
Dr Michael Rule, Dr Glenn Hyndes, Dr Keith Martin-Smith, Dr Anne Brearley and Dr Ryan Admiraal
who offered their advice on how best to tackle the project.
To all my family and friends who offered their support and encouragement throughout the year,
thank you.
v
Contents Declaration.........................................................................................................................................ii
Abstract.............................................................................................................................................iii
Acknowledgements...........................................................................................................................iv
List of Figures...................................................................................................................................viii
List of Tables......................................................................................................................................ix
Chapter 1: Introduction.....................................................................................................................1
1.1 Seagrass Ecosystem Functionality...............................................................................................2
1.1.1 Hydrodynamics.....................................................................................................................2
1.1.2 Sediment Trapping and Stabilisation....................................................................................4
1.1.3 Carbon Sinks.........................................................................................................................6
1.1.4 Food Source..........................................................................................................................7
1.1.5 Nursery Grounds...................................................................................................................9
1.2 Historic Changes in Seagrass Coverage in Cockburn Sound.......................................................10
1.3 Transplantation Efforts in Cockburn Sound...............................................................................13
1.4 Assessment of Ecosystem Functionality....................................................................................16
1.4.1 Global Perspective..............................................................................................................16
1.4.2 Cockburn Sound Perspective..............................................................................................19
1.5 Project Aims...............................................................................................................................20
Chapter 2: Methods.........................................................................................................................21
2.1 Site Description..........................................................................................................................21
2.2 Control Site Selection.................................................................................................................22
2.3 Sampling Methodology..............................................................................................................23
2.3.1 Sample Collection...............................................................................................................23
2.4 Sample Processing.....................................................................................................................24
2.4.1 Infauna Processing..............................................................................................................24
vi
2.4.2 Processing Effectiveness.....................................................................................................26
2.5 Statistical Analysis......................................................................................................................27
Chapter 3: Sampler Considerations.................................................................................................28
3.1 Introduction...............................................................................................................................28
3.2 Method......................................................................................................................................31
3.2.1 Sampling Methodology.......................................................................................................31
3.2.2 Sampler Issues and Considerations....................................................................................32
3.2.3 Statistical Analysis..............................................................................................................34
3.3 Results.......................................................................................................................................35
3.3.1 Diversity and Evenness.......................................................................................................35
3.3.2 Infauna Comparison...........................................................................................................37
3.4 Discussion..................................................................................................................................40
Chapter 4: Comparison of Transplanted and Natural Meadows....................................................43
4.1 Seagrass Shoot Density.............................................................................................................43
4.2 Infauna......................................................................................................................................45
4.2.1 Processing and Sorting Effectiveness.................................................................................45
4.2.2 Infauna Diversity and Evenness.........................................................................................46
4.2.3 Infauna Abundances..........................................................................................................48
Chapter 5: Discussion......................................................................................................................54
5.1 Shoot Density............................................................................................................................54
5.2 Infauna......................................................................................................................................55
5.2.1 Processing and Sorting Effectiveness.................................................................................55
5.2.2 Infauna Abundances..........................................................................................................55
vii
6. Conclusion...................................................................................................................................59
References.......................................................................................................................................60
Appendix 1.......................................................................................................................................71
Appendix 2.......................................................................................................................................72
Appendix 3.......................................................................................................................................75
viii
List of Figures Figure 1: Hjulstrom Curve of erosion and deposition in uniform material (Taken from Beer,
1997).....................................................................................................................................5
Figure 2: Aerial photo of study area on Southern Flats, Cockburn Sound looking North-West. Area
outlined in black shows the 3 hectare area of transplanted seagrass, the yellow outlined
areas show the experimental plots and the red outlined area shows the control sites.
(Image by Jennifer Verduin, taken at 300 m, on 18/4/2010 at 9:19 am)............................22
Figure 3: Layout of where the shoot counts were taken with the 0.25 m2 quadrats, gray shaded
squares indicate the samples where sediment cores were taken......................................25
Figure 4: The two sediment samplers’ trialled for the study. (Left) Venturi suction dredge with air
supplied by the SCUBA tank, (Right) PVC hand corer with serrated edge and rubber
plug.....................................................................................................................................31
Figure 5: Mean log-transformed Heip’s Evenness Index for the Bare Sand and Natural Meadow 1
sites using both the hand corer and venturi suction dredge..............................................36
Figure 6: Mean log of infauna abundances for the Bare Sand and Natural Meadow 1 sites using
both the hand corer and venturi suction dredge................................................................38
Figure 7: MDS plot of the square root transformed infauna abundance data................................39
Figure 8: Mean shoot density of the natural and transplanted seagrass on Southern Flats,
Cockburn Sound..................................................................................................................44
Figure 9: Shoot density in each edge zone for the natural and transplanted seagrass on Southern
Flats, Cockburn Sound........................................................................................................44
Figure 10: Mean Shannon-Wiener Diversity Index for each of the control and experimental plots
on Southern Flats...............................................................................................................49
ix
Figure 11: Mean Heip’s Evenness Index for each of the control and experimental plots on
Southern Flats.....................................................................................................................49
Figure 12: Infauna abundances for the control and experimental sites on Southern Flats.............51
Figure 13: MDS plot of the square root transformed infauna abundance data showing similarities
of the infauna assemblages between each site and edge zone at Southern Flats, Cockburn
Sound..................................................................................................................................52
List of Tables
Table 1: The number of infauna from each family remaining in the tray after the rinsing and
washing process..................................................................................................................47
Table 2: The number of infauna missed during the first sorting......................................................48
Table 3: R statistic outputs from the ANOSIM analysis for the infauna comparisons between the
sites and edge zones. The R statistic ranges from 1 to -1 with values >0.75 indicating that
the infauna assemblages are separate from each other, values >0.5 indicating some
overlap but still forming distinct groups and a values <0.25 indicating that there is no
difference in the infauna assemblages. Significance level is set at 5 % (α=0.05)................53
1
1. Introduction Declines in seagrass have been occurring at alarming rates all over the world in the last 20 years
(Walker et al., 2006). In most instances these declines are the result of human activities such as
eutrophication, dredging and coastal development (Cambridge and McComb, 1984; Short and
Wyllie–Echeverria, 1996). Worldwide there are approximately 60 species of recorded seagrass,
most of which form single species meadows (Short and Coles, 2001; Orth et al., 2006). Of these
just over one third, roughly 26 species, are found within Western Australian waters (Kirkman and
Walker, 1989; Butler and Jernakoff, 1999).
A comprehensive study by Short et al. (2011) examined the risk of extinction of the world’s
seagrasses and found 10 species to be at risk of becoming extinct, three of which qualified for
listing as endangered. With seagrass habitats diminishing, efforts into restoring, rehabilitating and
transplanting seagrass into areas where they formerly occupied, have been increasing (Fonseca et
al., 1982; Kirkman, 1998; Paling et al., 2000; Paling et al., 2001a; Paling et al., 2001b; van Keulen
et al., 2003; Uhrin et al., 2009).
Transplantation of seagrass is vital for the recovery of the various ecosystem functions they
provide, such as alteration of hydrodynamics processes, sediment trapping and stabilisation,
carbon trapping, providing food and acting as a nursery habitat (Butler and Jernakoff, 1999; Duffy,
2006). These ecosystem functions are extremely valuable with estimations for the value of
seagrass habitats ranging from $12,635 to $25,270 ha.-1yr-1 (Lothian, 1999); a more recent study
however has placed the value of seagrass habitats at $34,000 ha.-1yr-1 (Short et al., 2011).
Assessing the recovery of each ecosystem function in transplanted seagrass is vital for the
rehabilitation of lost seagrass meadows, with each ecosystem function providing a ‘piece’ of the
proverbial ‘ecological jig-saw puzzle’; with the full picture not being seen until all the ‘pieces’ are
2
back together. The following section describes each of these ecological function ‘pieces’ and why
they are vital to the seagrass ecosystem.
1.1 Seagrass Ecosystem Functionality
1.1.1 Hydrodynamics
Submerged plants are known for helping prevent bank erosion in rivers and streams by acting as a
buffer against strong currents and waves by reducing the water velocity. A study by Bonham
(1983) revealed that as much as two thirds of boats bow wave energy dissipates after travelling
two meters into aquatic vegetation along river banks. Seagrass provide a similar function within
coastal areas by reducing the force of the currents and waves, thereby reducing their impact on
beaches, shorelines and coastal structures. Research has shown that the majority of the water
velocity is reduced during the first meter from the leading edge of the seagrass meadows (Gambi
et al., 1990; Peterson et al., 2004; Fonseca and Koehl, 2006; Backhaus and Verduin, 2008; Morris
et al., 2008; Lefebvre et al., 2010), and that water flow results in an increase in turbulence above
the seagrass canopy as the water comes into contact with the seagrass leaves (Fonseca and
Fisher, 1986; Gambi et al., 1990; Verduin and Backhaus, 2000; Peterson et al., 2004; Morris et al.,
2008; Lefebvre et al., 2010).
However, depending on the morphological structures of the seagrass, water flow can also be
greater underneath the seagrass canopy, as was found with Amphibolis sp. (Verduin and
Backhaus, 2000; van Keulen and Borowitzka, 2002). The subtle differences in hydrodynamics and
water flow created by these different structures, such as the stems of the Amphibolis species and
the concave surface of Posidonia sinuosa, provide additional niches for fauna. This is supported
by research from Jernakoff and Nielsen (1998) and Trautman and Borowitzka (1999), who
revealed a marked difference in the epiphytic algae and epifauna assemblages associated with
these different seagrass structures and their hydrodynamic characteristics.
3
While the seagrass structure impacts on the water flow and speed, the water dynamics have an
impact upon the seagras structure. The water flow into the seagrass meadows is vital for the
transport of nutrients such as ammonium and nitrates, which the seagrass and their epiphytes
utilize for enhancing their growth (Brun et al., 2003; Cornelisen and Thomas, 2004 and 2006;
Morris et al., 2008). Excessive water flow within seagrass has also been shown to have negative
impacts on their growth, with lower shoot densities occurring in areas of high water movement
compared with sheltered sites (Schanz and Asmus, 2003). This impact on the seagrass is prevalent
at Southern Flats in Cockburn Sound, Western Australia, where the construction of the Garden
Island causeway has restricted water movement into and out of the bay. Water flow into and out
of Cockburn Sound is restricted to two short trestle bridges in the rock wall causeway, and as a
result of the mass movement of water through these narrow sections, the water velocity is
greatly increased, resulting in the scouring of the sea bed and loss of the seagrass (Kendrick et al.,
2002; Cockburn Sound Management Council, 2003).
Hydrodynamic regimes also play a vital role in the seagrass community with marked differences
occurring between tidal and wave dominated areas. Koch and Gust (1999) looked at the effects of
tidal and wave dominated regimes on the seagrass Thalassia testudinum and found marked
differences in the water mixing within the meadow and outside the meadow. These boundaries in
water mixing within tidal dominated areas experiencing unidirectional flow were contributed to
the “skimming flow” or laminar flow experienced above the meadow. This movement of the
water results in the attenuation of the seagrass blades, causing them to blow over and form a
distinct boundary, below which substantially lower water velocities and decreased mixing are
experienced (Fonseca and Fisher, 1986; Gambi et al., 1990; Koch and Gust, 1999).
More recently, research by Carruthers et al. (2007) has shown that seagrass have adapted to
different wave energy environments through morphological features. Reinforcement of above
4
ground structures enable certain seagrass to withstand the battering of the ocean swell, while
deeper rhizome and root penetration, provide a sturdy anchor to prevent being uprooted, but
also to cope with changing sediment burial. Earlier work by Cambridge (1980) also observed
marked zonation in seagrass species across a wave energy gradient with changes in root-rhizome
growth and structure in response to sediment accretion.
1.1.2 Sediment Trapping and Stabilisation
Seagrass sediments are typically characterised by soft sands, often with quantities of fine silt or
mud with a high organic content (van Keulen and Borowitzka, 2003; de Boer, 2007; Bos et al.,
2007; van Katwijk et al., 2010). The reason for the presence of these fine sediments within the
meadows is a result of the change in hydrodynamic processes at the water-seagrass interface. As
the water encounters the seagrass canopy it experiences increased drag as the leaves sway
through the water, reducing the water flow and increasing the turbulence above the seagrass bed
(Gambi et al., 1990; Peterson et al., 2004; Backhaus and Verduin, 2008; Morris et al., 2008;
Lefebvre et al., 2010). Due to the sudden decrease in velocity, the waters’ ability to maintain
particulate matter within the water column decreases, as explained by the Hjulstrom curve in
Figure 1.
Early work by Scoffin (1968) looked at the effects of sediment trapping and transportation by
various plants with the use of an underwater flume. Scoffin’s research reveal that the density and
distance between leaf blades of Thalassia testudinum were important factors influencing the
deposition or erosion of sediments, with dense patches experiencing sediment deposition and
sparse patches, erosion. Such accumulations of sediments are the result of the decreased water
velocity within the meadow (Fonseca and Fisher, 1986; Gacia et al., 1999; Gacia and Duarte,
2001). This reduction in water velocity and subsequent increase in sediment deposition leads to
an increase in the proportion of fine particles within the sediment, which has been observed in
5
many seagrass studies (van Keulen and Borowitzka, 2003; de Boer, 2007; Bos et al., 2007; van
Katwijk et al., 2010).
Figure 1: Hjulstrom Curve of erosion and deposition in uniform material (Taken from Beer, 1997)
While it is generally accepted that seagrass accumulate and trap sediment, research conducted by
Mellors et al. (2002) suggest that this is not entirely true. Their findings indicate that there was no
difference in the accumulation of sediments or nutrients between low biomass ephemeral
seagrass meadows and unvegetated sites, bringing the sediment trapping theory of seagrass into
question. This suggests that the smaller, less dense, seasonal seagrass species do not reduce
water flow enough for sedimentation to occur and that sediment trapping by seagrass may be
species and location specific. Similarly, Paling et al. (2003) observed that dense Amphibolis
transplants were unable to trap and accumulate sediment within a high energy environment and
suggest that sediment trapping is dependent upon the hydrodynamic conditions that the seagrass
is exposed to.
6
In addition to the trapping of sediments, seagrass’ also have the ability to stabilise and prevent
the resuspension and erosion of sand (Gacia and Duarte, 2001; Bos et al., 2007; de Boer, 2007).
The extensive rhizome mats of seagrass bind the sediment and keep it from being eroded, while
the hydrodynamic conditions created by the leaf canopy also aid in preventing sediment
resuspension, due largely to the reduction in turbulence within the meadow (Fonseca and Fisher,
1986; Gacia et al., 1999; Gacia and Duarte, 2001).
1.1.3 Carbon Sinks
As seagrasses grow and photosynthesize they consume CO2 and convert it into complex sugars,
which later get used in the construction of other plant structures (leaves, rhizomes and roots). In
general, the bulk of the biomass for these structures, namely the rhizome and roots, are stored
below-ground (Fourqurean and Zieman, 1991; Mateo and Romero, 1997), however, in some
species, such as Amphibolis sp., the bulk of the biomass is in the above ground structures (Paling
and McComb, 2000). As these structures die, the carbon stored within them becomes ‘trapped’
within the sediment.
Several studies have attempted to estimate the burial of carbon within seagrass habitats (Pollard
and Moriarty, 1991; Gacia et al., 2002; Bouillon et al., 2004; Duarte et al., 2005 and Kennedy et
al., In Press 2010). Values of burial ranging from 182.5 to 1569.5 grams of carbon per square
meter per year were calculated for the seagrasses Enhalus acoroides, Syringodium isoetifolium,
Cymodocea serrulata, Thalassia hemprichii and Cymodocea rotundata within the Gulf of
Carpentaria, Australia (Pollard and Moriarty, 1991), while a value of 198 grams of carbon per
square meter per year was calculated for Posidonia oceanica (Gacia et al., 2002). Duarte et al.
(2005) attempted to calculate the average global carbon burial of vegetated habitats, with
seagrass estimated to contribute 83 grams of carbon per square meter per year. A more recent
study of the global contributions of seagrass burial by Kennedy et al. (In Press 2010) calculated
7
the annual global carbon burial rate at 41 to 66 grams of carbon per year from seagrass derived
sources.
While it is apparent that seagrass contribute directly to the sequestration of carbon from in situ
decomposition, other studies have shown that a major proportion of the carbon from within
seagrass habitats are derived from allochthonous or seston sources (Gacia et al., 2002; Kennedy
et al., In Press 2010). These alternative carbon sources have been shown to contribute 72% (Gacia
et al., 2002) and approximately 50% (Kennedy et al., In Press 2010) of the carbon burial in
seagrass habitat respectively. An analysis of the difference in 13C and phospholipid fatty acids by
Bouillon et al. (2004) in the seagrass and mangrove habitats of Gazi Bay, Kenya, also revealed that
between 21-70% of the sedimentary carbon within the seagrass meadows was derived from the
nearby mangrove habitat, indicating that the seagrass’ act as an important carbon sink.
With issues of increased greenhouse gas emissions and the effects of climate change being
present-day concerns, knowing how much carbon these valuable marine habitats store and for
how long becomes essential. The use of radiocarbon dating within Posidonia oceanica sediments
have shown that carbon trapped within these seagrass habitats can be stored for as long as 3370
years (Mateo et al., 1997), further indicating the importance of seagrass habitats as vital carbon
sinks for the marine environment.
1.1.4 Food Source
Due to the high fibrous content and relatively low nutritional value of the seagrass leaves
(Bjorndal, 1980; Duarte, 1990; Valentine and Heck, 1999), very few organisms feed directly on
seagrass. Those that do, such as Dugongs (Dugong dugon) and Green Sea Turtles (Chelonia
mydas), as well as some fish and invertebrates, account for approximately 10% of the seagrass
consumed in the food web (Valentine and Heck, 1999).
8
Many studies have looked at the contributions seagrass makes through the food web with the use
of carbon and nitrogen stable isotopes (Nichols et al., 1985; Peduzzi and Herndl, 1991;
Kharlamenko et al., 2001; Vizzini et al., 2002; Hyndes and Lavery, 2005; Smit et al., 2005; Leduc et
al., 2006; Nyunja et al., 2009). It is apparent from these studies that the carbon and nitrogen
supplied directly from the seagrass contributes only a relatively minor component of the carbon
and nitrogen within the different trophic levels (Hyndes and Lavery, 2005; Smit et al., 2005) and is
consumed by only a select few invertebrates, such as some copepods, amphipods and polychaete
worms (Hyndes and Lavery, 2005). The majority of the nutrient sources to the seagrass food
network appear to be derived from the consumption of the seagrass detritus and associated
epiphytic organisms (Vizzini et al., 2002; Hyndes and Lavery, 2005; Smit et al., 2005; Nyunja et al,
2009). This is not too surprising as epiphytic algae can account from 40 to 90% of the primary
productivity in some seagrass ecosystems (Pollard and Moriarty, 1991)
A study by Leduc et al. (2006) looked at the seasonal variation of the importance Zostera
capricorni within the food web. Their findings suggest that the seagrass contributes between 24
to 99% of the diets of the consumers in the area with its importance as a food source shifting
during the year, becoming more important during late winter. This suggests that the main food
source of temperate seagrass ecosystems can shift from a detrital food web during the winter
months to an algal/epiphytic based food web during summer.
It has also been found that seagrass not only contributes to the benthic food web but can provide
a food source to the planktonic food web (Thresher et al., 1992). Research by Thresher et al.
(1992) found that nutrients derived from decomposing seagrass wrack that has been transported
offshore provide a carbon source to the microbial community that fuels the food web for the
larval Blue Grenadier (Macruronus novaezelandiae). Another study, conducted by Peduzzi and
Herndl (1991), also found seagrass fuelled the production of free-living marine microbes through
9
monomeric carbohydrates that were leached out from the seagrass leaf wrack. Such productions
of microbial organisms can therefore act as important food sources, but due to their consumption
of seagrass derived carbon can also serve as a carbon sink, as was found in the water column
above seagrass beds during the research by Kaldy et al. (2002).
1.1.5 Nursery Grounds
The sheltered conditions created within the seagrass meadows and highly productive seagrass
and epiphyte community; provide perfect low energy environments for the early life stages of fish
and invertebrate whilst also providing them with a valuable food source (Verweij et al., 2006).
The complex structures created by seagrass also aids in the survival of many juvenile fish and
invertebrate larvae with increased survival and lower predation frequently observed (Wahle et
al., 1992; Rooker et al., 1998). Hyndes et al. (2003) suggested that smaller sized fish would inhabit
seagrass with denser foliage with larger fish occupying less dense meadows, however research by
Bell et al. (1987) and Worthington et al. (1991) showed that increased shoot density made little
impact on the number of juvenile fish that were present with only a significant difference
occurring between seagrass and unvegetated habitats.
Seagrass also plays a pivotal role in the life cycle and subsequent development of many fish and
invertebrate species, providing a source of new recruits to the adult population (Gillanders, 1997;
Vance et al., 1998; Heck et al., 2003; Smith and Sinerchia, 2004). The use of stable carbon
isotopes by Verweij et al. (2008) revealed that 98% of the reef fish Ocyurus chrysurus in the
population would have originated from seagrass habitats.
While it is typically accepted that nursery grounds promote the growth of juvenile and larval
fauna, however the findings from a paper by Grol et al. (2008) on the growth of juvenile reef fish,
found that the fish would have more food, and subsequently better growth if they fed within a
10
reef habitat rather than in seagrass or mangroves. The problem associated with such a statement
is that the fish would be more exposed to predation and have a lower survival rate in reef
habitats, suggesting that the fish have to balance a trade-off between better food sources in reef
habitats and increased survival provided by the shelter from seagrass and mangrove habitats.
1.2 Historic Changes of Seagrass Coverage in Cockburn Sound
In 1954, seagrass in Cockburn Sound covered an estimated area of 4,195 hectares and by 1978;
this had decreased to 889 hectares (Cambridge and McComb, 1984), a decline of approximately
79.8 %. From the 1960’s onward, increased industrial development occurred along the east coast
of the sound, with increased effluent discharge from the CSBP oil refinery, sewage treatment
plant, blast furnace, nitrogen and phosphorous fertiliser plants and the power station (Cambridge
and McComb, 1984). The first large scale losses of seagrass were recorded in 1969 along the
eastern shores before spreading through the rest of the embayment. Cockburn Cement also
commenced shell-sand dredging for lime production at Owen Anchorage, Parmelia and Success
Bank in 1972. From 1994 to 1996, 49 hectares of seagrass was removed by dredging
(Environmental Protection Authority, 1996) and 168 hectares of seagrass during 2002 to 2010
(Oceanica, 2009b).
Construction of the Garden Island causeway after 1970, resulted in seagrass loss on Southern
Flats and also restricted water flushing within Cockburn Sound by much as 40 % (Cambridge and
McComb, 1984; Cockburn Sound Management Council, 2003). By 1999, the estimated
seagrass coverage in Cockburn Sound was 661 hectares (Kendrick et al., 2002), which constitutes
an 84.2 % decrease from 1954.
In 1982, high levels of heavy metals (Talbot and Chegwidden, 1982) and petrochemicals
(Alexander et al., 1982) were found in Cockburn Sound and its associated fauna. This is of
11
concern, as research has shown that heavy metals (Ralph and Burchett, 1998 a; Macinnis-Ng and
Ralph, 2002) and petrochemicals (Cambridge et al., 1986; Ralph and Burchett, 1998 b; Macinnis-
Ng and Ralph, 2003) have negative impacts on the seagrass’ growth and ability to
photosynthesize. While these pollutants would have caused localised death and decreased
growth in some areas (Cambridge and McComb, 1984), Cambridge et al. (1986) indicated that it
was unlikely to be the source of the wide spread loss in Cockburn Sound. However this would
have contributed additional stress to the seagrasses making them more vulnerable to other
stressors.
In an attempt to explain the extensive loss of seagrass which occurred, Cambridge et al. (1986)
conducted several field and laboratory experiments to try and determine the cause. Seagrass
transplant trials were used both in Cockburn Sound and Warnbro Sound to see how the seagrass
survived. The transplants within Warnbro Sounds took hold and grew well, while those within
Cockburn Sound experienced little growth and became matted with large amounts of epiphytes.
Cambridge et al. (1986) concluded that the wide scale losses in seagrass could be the result of
eutrophication, which occurred shortly after the discharge of effluent from the fertilizer factory
commenced in 1969 (Cambridge and McComb, 1984).
Silberstein et al. (1986) examined epiphyte loads on seagrass beds near the effluent outfall and
found epiphyte biomass to be 2-8 times higher than those of unaffected meadows. This was also
supported by Cambridge et al. (2007) through a retrospective analysis which found strong
correlations between the presence of particular epiphytes and the seagrass losses which
occurred. Other small and isolated losses in seagrass have occurred in Cockburn Sound around
Mangles Bay, as well as Warnbro Sound, and at Rottnest Island in boat anchorage areas (Walker
et al., 1989; Hastings et al., 1995). These losses are the result of the scouring of the seabed from
12
mooring chains which create 3-300 m2 circles of devegetated seafloor as the boat and mooring
chain swings around with the changing winds and tides (Walker et al., 1989).
While only relatively small and highly localised areas of seagrass are removed by this process,
once the number of boat moorings present within the area is taken into consideration, the overall
loss of seagrass from this becomes more substantial. In total, 151 of 253 boat moorings were
found within seagrass meadows in Cockburn Sound, resulting in a total loss of 1.8 hectares,
approximately 1.9 % (Walker et al., 1989). While this is only a relatively minor loss, it does
however, increasingly subject seagrass to the effects of waves and swell which can result in
blowouts and increased scouring (Walker et al., 1989; Hastings et al., 1995).
Despite the widespread loss of seagrass coverage in Cockburn Sound, localised recolonisation on
Success and Parmelia Banks has also been recorded (Kendrick et al., 1999; Kendrick et al., 2000).
Research by Kendrick et al. (1999) showed, with the use of aerial photos, that from 1972 to 1993
the seagrass on Success and Parmelia Banks had increased some 20,000 to 30,000 square meters.
A more detailed study revealed that the seagrass on Success Bank had increased from 507
hectares in 1965 to 1036 hectares in 1995 (Kendrick et al., 2000). The same study also showed
that the seagrass on Parmelia Bank experienced little change in coverage with 735 hectares
present in 1965 decreasing to 699 hectares in 1995. It was also observed that the seagrass
increased on the western side of Parmelia Bank and decreased in the east which was a result of
the shell-sand mining which had taken place in the area.
Work by Campbell (2003) into the recruitment of Posidonia australis and P. coriacea propagules
on Success Bank showed that, on average, 55 seagrass propagules established per hectare per
year; however only 69 % of those survived to the end of the 23 month long study. Campbell also
observed that no seagrass seedlings recruited at the site; though at a nearby site, as many as 39
13
seedlings recruited per month, which suggests that recolonisation and recruitment of seagrass
was taking place. While these isolated areas have experienced some natural regrowth the rest of
Cockburn Sound has shown very little and it has been suggest that the embayment had been
modified to a state no longer suitable for natural seagrass recovery (Kendrick et al., 2002).
1.3 Transplantation Efforts in Cockburn Sound
Following the extensive loss of seagrass within Cockburn Sound, substantial efforts were made to
increase their natural recovery and trialling different methods of transplantation, such as manual
(seedlings, plugs and springs) and mechanical (sods) methods, to enhance their survival and
growth. Attempts were made at using seagrass seedlings as a means of replanting the lost
seagrass meadows in Cockburn Sound (Kirkman, 1998). This was done using seedlings and sprigs
of Posidonia australis, P. sinuosa, P. angustifolia and P. coriacea seedlings and Amphibolis
antarctica and A. griffithii seedlings and sprigs, all of which yielded poor survival. In the space of a
year, all the Posidonia seedlings had died and had a dense covering of epiphytes. At the end of
seven months all of the Amphibolis sprigs had died, while the seedlings persisted for 17 months
before dying or being washed away.
In 1993, attempts were made to trial staple and plug transplantation methods with A. griffithii
and P. sinuosa at Carnac Island and to see the effects of stabilising the sediment with plastic mesh
on different sized transplants (van Keulen et al., 2003). It was found that the staple method was
an ineffective way of transplanting the Amphibolis seagrass with all the transplants dying,
regardless of the planting size or the presence of the plastic matting. The plug method on the
other hand showed a significant interaction between the size of the transplanted plugs and the
presence of the sediment stabilizing mat, with larger plug sizes having a greater survival rate
when the plastic mesh was surrounding them (van Keulen et al., 2003). While the plug method of
14
transplantation provided better survival, the P. sinuosa transplants still fared poorly in
comparison to A. griffithii.
Later in 1997 Paling et al. (2000) investigated the survival of A. griffithii plug transplants at
different depths on Success Bank. In all, 580 15 cm diameter plugs were planted at 5, 6, 8 and 10
meter depths and monitored over 14 months. The results indicated that there was no significant
change in transplant survival in response to the different depths, with all the transplants
exhibiting at least a 95 % survival rate during the first few months, before survival decreased
dramatically during the winter storms.
Following the success of the plug transplantation experiments, which showed that larger plugs
survived better than small transplants, mechanical transplantation was also trialled on Success
Bank using the ECOSUB1 described by Paling et al. (2001a). 1,500 “sods” 0.25 m2 in size were
planted using Posidonia sinuosa, P. coriacea and Amphibolis griffithii. Survival varied between the
Posidonia and the Amphibolis transplants with P. sinuosa and P. coriacea having 76.8 % and 75.8
% of transplants survive respectively while A. griffithii experienced 44.3 % over a two year period.
Despite the differences in survival all the transplants exhibited some growth two years after
transplantation (Paling et al., 2001a).
A further study was implemented using the ECOSUB2 (Paling et al., 2001b), a modified version of
the ECOSUB1 described by Paling et al. (2001a). This improved method transplanted 280 “sods”
of 0.55 m2 size in early 2000 and by June that year, all the transplants exhibited a 100 % survival
rate. Continued monitoring of the transplants from Paling et al. (2001a) revealed that the
seagrass was averaging a 70 % survival rate three years after transplantation (Paling et al.,
2001b).
15
Research that followed on from these studies looked at the effects of the transplants spacing on
the seagrass’ survival (Paling et al., 2003). It was found that the spacing of the 0.55 m2 transplants
had no significant effect on the seagrass’ survival with all the transplants experiencing greater
than 90 % survival during the first four months. Survival then decreased to between 9 and 40 %
over the winter months due to mortality from storm events (Paling et al., 2003). Despite the
transplanting method’s initial high survival and recovery rate, its expensive operating costs, in the
order of AU$200 per transplant, made it a non-viable means of seagrass restoration.
The poor survival of transplants on Success Bank seemed to be the result of the highly dynamic
sediments within the high wave energy environment (Paling et al., 2000; Paling et al., 2003).
Campbell and Paling (2003) attempted to test whether the use of an artificial seagrass mat would
increase Posidonia australis transplant survival within this environment. They discovered that
habitat enhancement in the form of sediment stabilisation improved transplant survival by 50 %
in 60 % of the P. australis transplants.
Posidonia sinuosa, as the dominant meadow-forming species within Cockburn Sound, formerly
comprised 80 % of the seagrass coverage (Cambridge and McComb, 1984). Therefore ensuring
the recovery of this species was of vital importance. Paling et al. (2007) conducted research into
assessing the most effective methods and locations for the survival and re-colonization of P.
sinuosa. They trialled both sprig and plug transplantation methods at differing depths and
monitored the seagrass’ survival. The findings indicated that the plug method was the most
successful when compared to the sprig method and that the survival of transplants was greater
for both methods at the shallower three meter depth. While survival was greater in the plug
transplants, the authors indicated it was also the more costly method to implement and
suggested that the sprig method’s cost-effectiveness would outweigh its lower survival rate.
16
Large scale rehabilitation of the seagrass meadows was implemented during the summer of 2004
using the sprig planting method for Posidonia australis on Southern Flats. From 2004 until 2011,
three hectares of manually transplanted P. australis sprigs were planted over the south eastern
corner of Southern Flats (Oceanica, 2011). Both the middle and western areas experienced high
survival rates of more than 85 %, while the eastern hectare exhibited a 23 % survival rate
(Oceanica, 2011); since then the eastern hectare has been replanted with additional sprigs to help
recoup the losses.
1.4 Assessment of Ecosystem Functionality
1.4.1 Global Perspective
As seagrass declines have occurred worldwide a variety of different species have been affected.
To tackle this, a variety of different transplantation methods have been used, with as many
different methods and techniques utilised as there are species which have been affected. Survival
of the transplants varies considerably between the different methods, seagrass species and
hydrodynamic conditions in which they inhabit. As such the time taken for the transplants to
recover to a state comparable to a natural meadow can vary considerably.
In most instances assessing seagrass recovery involves monitoring the shoot density or rate of
horizontal rhizome growth. While monitoring these components of the seagrass is vitally
important, they only provide insight to the recovery of the seagrass’ structural complexity. To
determine whether the transplanted seagrass has fully recovered to a state comparable to
natural meadows, assessment of the recovery of all the seagrass’ ecosystem functions is required;
something which has currently been inadequately studied.
Despite numerous studies which have looked at optimizing the survival and growth of transplants
only a few have tried to assess the recovery of different ecosystem functions. Bell et al. (2008)
17
looked at the recovery of Halodule wrightii transplants and found that while some transplants
obtained shoot densities and biomasses comparable to those of natural meadows, the rate of
seagrass expansion was much less. An earlier study by Sheridan (1998) looked at H. wrightii
transplants and whether certain functions had returned. Sheridan’s findings revealed that after
three to four years the transplant sites structurally resembled nearby natural meadows, as did
the benthic fauna. After three years, the seagrass biomass as well as fish and decapods
abundances matched those of the natural meadows. However, monitoring of the sediment
revealed that the composition was much coarser within the transplant sites than the natural
meadow, indicating that fine sediments had yet to reach levels found in the natural sites. Both
Sheridan (1998) and Bell et al. (2008) expressed the need for monitoring of seagrass recovery to
occur over an extended period of time in order to assess the return of all the seagrass’ ecological
functions.
One such study, which implemented long term monitoring of the seagrass transplants was by
Evans and Short (2005), who monitored the return of ecosystem functions in Zostera marina
transplants over a nine year period. Their aim was to monitor the return of the seagrass
ecosystem functions, then fit trajectory models to them to see if they could predict when
particular functions would return. Their findings indicated that within four years, the biomass,
leaf length, leaf area index and fish diversity had all recovered to levels comparable to the natural
meadows and could be predicted using trajectory models. However, even after nine years, the
sediment composition within the transplants did not resemble that of the natural meadow
controls, although it was within the known ranges for Z. marina. These findings along with those
of Sheridan (1998), indicate that not all ecosystem functions return within the same timeframe,
and can differ both within and between different species. Furthermore, these studies also
highlight the need for long term monitoring of seagrass transplants beyond the normal range of
most projects.
18
In some cases, the recovery of the seagrass and its ability to providing habitat and refuge for
marine organisms is of high interest; such was the case with the research conducted by Smith et
al. (1988). Their research into whether newly transplanted Zostera marina provided suitable
habitat for the scallop Argopecten irradians, a commercially important species, revealed low
numbers of the scallop residing within the transplant site compared to the natural meadow, a
result they attributed to predation due to the patchy coverage which the transplanted seagrass
provided. This indicates that the mere presence of seagrass does not constitute suitable habitat
for organisms and that time is needed for the seagrass to recover before such functions can be
provided.
The recovery of the seagrass is paramount to the survival of many important commercial fish and
invertebrate species, with many of them utilising seagrass for shelter and food; in most cases the
food source that the seagrass provides takes on the form of macrobenthic infauna. Whilst acting
as a food source the infauna also provide valuable insight to other environmental processes
within the seagrass, including water quality and sediment composition (Saether, 1979; Cardoso et
al., 2007). As such, monitoring of the infauna should be of high priority; however studies that
have looked at whether such infaunal communities have recovered to naturally occurring levels
has yielded varying results (Sheridan, 1998; Pranovi et al., 2000; Sheridan et al., 2003; Evans and
Short, 2005).
Pranovi et al (2000) found that 1.5 years after transplantation, the benthic fauna within the
seagrass, Cymodocea nodosa, had obtained levels which matched those of nearby natural
meadows. Sheridan et al (2003), on the other hand, discovered that even three years after
transplantation, the benthic infauna in Halodule wrightii were still noticeably distinct from those
of natural meadows. It has been suggested by Sheridan (1998) that fully restored infauna
communities may be dependent on the sediment composition and the content of fine organics.
19
With different species of seagrass trapping sediment at different rates (Fonseca and Fisher, 1986),
the time taken for the infauna within different meadows to recover would therefore differ.
1.4.2 Cockburn Sound Perspective
Despite the extensive transplantation work which has taken place in Cockburn Sound (Kirkman,
1998; Paling et al., 2001ab; Campbell and Paling 2003; Paling et al., 2003; van Keulen et al., 2003;
Paling et al., 2007; Oceanica, 2011), very little work has looked at whether or not these seagrass
transplants have regained their ecosystem function. In 2006, a preliminary study of the return of
ecosystem functionality in Posidonia sinuosa transplants within Cockburn Sound was conducted
(Kenna et al., 2006). However, due to the lack of replicate sites, the data were not formally
analysed. Despite this, the results from the preliminary study showed that five years after
transplantation the percentage cover, shoot density and leaf length, were very similar between
the transplanted P. sinuosa and the natural meadow.
Sediment trapping was also assessed within different density sprig transplants of Posidonia
australis on Southern Flats, as part of a PhD dissertation by Chisholm (unpublished). The research
indicated that both the higher density 0.25 and 0.125 m spaced transplants showed increased
accretion of sediments while the lower density 0.5 and 1 m spaced transplants experienced more
sediment erosion (Verduin et al., 2007). Experimental manipulation of shoot density within the
natural meadows revealed that densities greater than 50 % cover experience sediment accretion
while no significant change was seen in sediment height at lower densities. This indicates that the
transplanted P. australis is trapping sediment; however it still remains to be seen if it is doing so
at the same rate as that found in natural systems.
Horn et al. (2009) looked at the photosynthetic recovery of sprig transplanted Posidonia sinuosa
within Cockburn Sound using chlorophyll fluorescence. Their findings showed that after three
20
months post-transplantation the maximum electron transport rate and effective quantum yield,
used as proxies for photosynthesis, had fully recovered in relation to the control site. However as
this study only examined individual sprigs in relation to those of a fully functioning meadow, the
recovery of the transplant meadow as a whole would take considerably longer as the
photosynthetic productivity would be dependent on shoot density.
While there has been work done on the macrobenthic communities within Cockburn Sound
(Brearley and Wells, 2000; Oceanica, 2009a), as yet there has been little done within the
transplanted seagrass. It is therefore the purpose of this study to fill a gap in the knowledge
surrounding the transplanted seagrass within Cockburn Sound, focusing on the recovery of the
macrobenthic community within transplanted Posidonia australis on Southern Flats.
1.5 Project Aims
Following on from the extensive rehabilitation work conducted on Southern Flats, this project
aims to assess the ecosystem recovery of the transplanted Posidonia australis sprigs with respect
to the macrobenthic infauna. The primary goals of the project were to:
1) Determine if the infauna present within the transplants resemble those of nearby natural
meadows
2) See if the infauna are present in the same abundances as those in natural meadows
3) Determine if there is any edge effect impacting on the infauna
4) Determine if any of the infauna can be used as potential indicator species to indicate the
recovery of the infauna community within the transplanted seagrass
The secondary goal of the project was to:
5) Compare the sampling effectiveness of two different sediment samplers
21
2. Method 2.1 Site Description
Cockburn Sound is a sheltered coastal embayment located in the south west region of Western
Australia. The area is protected on the western seaward side by Garden and Carnac Island and by
Point Peron to the south. A 4.2 km rock wall causeway extends out from Point Peron northward
to Garden Island’s southern end; the causeway includes two small trestle bridges (613 and 304 m
wide) that allow for restricted water flow in and out of the embayment. The causeway also
provides shelter from prevailing winds and sea swell while shallow areas around Success and
Parmelia Bank in the north provide a buffer against large waves and swell. Despite this the
northern margin of Cockburn Sound is still very open to the wind, with strong north and north-
westerly winds generating wind-waves which make conditions in Cockburn Sound very rough.
Mixing in the embayment is largely wind driven with little impact from the very small semidiurnal
tides, which rarely exceed 0.5 m. The water is very shallow, ranging from 2-9 m deep in areas
such as Parmelia Bank, Success Bank and Southern Flats, and around 20-25 m in the central basin.
The south eastern edge of Southern Flats is situated in relatively shallow water, which ranges
from 2-3 m in depth. The area is comprised of soft sediments colonised by sparse patches of
Posidonia australis with some intermixed P. sinuosa, while the western and northern areas of
Southern Flats are covered by large expanses of Posidonia meadows.
Southern Flats south-eastern end is the location of extensive seagrass restoration effort with
three hectares of hand transplanted P. australis covering the seafloor. The transplanting was
initiated in the western section from 2004 to 2005 with one hectare being planted. During 2005
and 2006 the middle hectare (containing the site for this study) was planted and over 2006 to
22
2007 the eastern hectare was planted. Using seagrass cuttings collected from a donor site at
Success Bank, shoots were planted every 0.5 m. The areas of interest for this study were four 5 x
5 m experimental transplant plots located in the north-western corner of the middle hectare of
the transplant meadow (Figure 2). Plots were planted out at different densities with shoots
planted every 1, 0.5, 0.25 and 0.125 m. In addition to these sites were three control sites,
including a bare sand site, natural fragmented meadow outside of the transplant site (Natural
Meadow 1) and a natural fragmented meadow within the transplant site (Natural Meadow 2).
Figure 2: Aerial photo of study area on Southern Flats, Cockburn Sound looking North-West. Area outlined in black shows the 3 hectare area of transplanted seagrass, the yellow outlined areas show the experimental plots and the red outlined area shows the control sites. (Image by Jennifer Verduin, taken at 300 m, on 18/4/2010 at 9:19 am).
2.2 Control Site Selection
Aerial photos were used to provide estimates of the size and distance of natural seagrass patches
to determine if they could be used as possible control sites for the study. A high resolution,
georeferenced, aerial photo of Southern Flats taken in 2008 (supplied by Oceanica Consulting Pty
23
Ltd) was used in conjunction with a non-georeferenced aerial photo of Southern Flats in 2010.
Three control sites were needed for the study, one on bare sand, one of a natural P. australis
patch outside the transplantation site and one from within. Seagrass patches were only
considered if they met the following three conditions:
1). Were natural Posidonia australis patches
2). Able to fit a 5 X 5 m plot within them
3). Less than 100 m from the four experimental plots
Once control sites had been selected from the aerial photos they were assessed in the field to
determine their suitability. If all the conditions were met then the site was marked out with metal
stakes and roped off.
2.3 Sampling Methodology
The layout of the study area was made prior to the commencement of this project and was
designed for another experiment, so its design was not ideal for this particular project. As a result
it was not possible to have replicate experimental plots and so sub-samples were taken from each
of the seven plots. The sampling was conducted over the winter from the 12th of May until the
22nd of June, 2011, and provides a snapshot in time of how the infauna has recovered compared
to nearby natural meadows.
2.3.1 Sample Collection
Each of the seven 5 x 5 m plots were separated into three zones, the outer zone (1 meter in from
the edge), middle (2 meters in from the edge) and centre (3 meters in from the edge), with 12, 8
24
and 4 shoot count measurements taken from each zone respectively to provide an accurate
representation of each edge zone based on their relative sizes. Each of the shoot counts was done
using a 0.25m2 quadrat by divers on scuba; each quadrat was laid out in the manner shown in
Figure 3. In addition to the shoot counts, sediment cores were also taken using a 55 mm PVC
hand corer with a serrated edge to a depth of 15 cm, labelled and placed into calico bags. Twelve
sediment cores were taken from each site, with 4 samples taken in each of the outer, middle and
centre zones as indicated by the gray shaded squares in Figure 3. Missing and incorrectly labelled
samples were excluded from the analysis. Samples were stored in a freezer at -20°C until they
were needed.
A venturi suction sampler was also compared against the hand corer to determine which method
would be most suitable for this study. Unfortunately due to time constraints and long sample
processing times the hand corer was selected before the samplers relative effectiveness could be
assessed. The impromptu selection of the hand corer over the suction sampler was based on its
ease of use and relatively consistent sample sizes; however a more detailed analysis of the
samplers’ effectiveness is given in the next chapter.
2.4 Sample Processing
2.4.1 Infauna Processing
Sediment samples were thawed out and later transferred into plastic bags for preservation. This
was done by collecting the sediment into one corner of the calico bag then inverting the contents.
Approximately 300 mL of seawater was then poured over the calico bags to remove the
25
remaining sediment and infauna clinging to the sides. 40 mL of 37.5 % formalin was then added to
the samples in the plastic bags to create a 5 % Formalin buffered seawater solution, with 1 mL of
5 % Rose Bengal added to stain the infauna. The samples were then left for a minimum of 24
hours to allow adequate time for the infauna to be fixed and stained before analysis.
Figure 3: Layout of where the shoot counts were taken with the 0.25 m2
quadrats, gray shaded squares indicate the samples where sediment cores were taken.
After fixing and staining, the sediment was tipped into a beaker so that the volume of sediment
could be recorded. Large pieces of shell and seagrass material were removed and placed into a
small dish; the sediment was then left to settle out so an accurate measure of the sediment
volume could be taken. The sediment samples were then tipped into a 500 micron sieve and
26
washed until the bulk of the fine sediments were removed. The contents of the sieve were then
washed into a shallow tray and filled with enough water to submerge the sediment. The tray was
then agitated to get the infauna suspended before pouring them back into the 500 micron sieve
leaving the sediment behind; the tray was then refilled with water and the process repeated.
The contents of the sieve were then washed into a small dish and filled with water. Infauna were
then removed using fine tipped tweezers and placed into 50 mL containers of 70 % ethanol so
they could be later identified. The tray of sediment was then searched thoroughly for any
remaining infauna, which were likewise removed using tweezers and placed into the container of
ethanol. All the invertebrates, where possible, were identified to family level using dissecting and
ocular microscopes and where then enumerated. A comprehensive list of texts and references
used to identify the infauna is given in Appendix 1. Only intact infauna, with identifiable
characteristics were included within the analysis; all fragments and lost limbs were excluded.
2.4.2 Processing Effectiveness
In an attempt to gauge the effectiveness of the processing methodology, 44 samples were split
into two sub-samples. The first sub-sample contained the infauna removed from the tray while
the second sub-sample containing the infauna from the sieve. Separating the samples in this
manner allowed the percentage of different infauna removed by the washing process to be
calculated. This thereby provided an estimate of how effective the washing process was. In
addition to determining what percentages of infauna were removed by the washing process an
additional 15 samples were selected to determine the overall effectiveness of the sample
processing. This was done by having a second person search through the samples after the initial
sorting had taken place and removing any infauna missed by the first attempt.
27
2.5 Statistical Analysis
To determine whether any of the transplanted seagrass plots had recovered in terms of their
overall structural complexity (i.e. shoot density), a one-way ANOVA was used to compare the
shoot densities of the four experimental plots and the two natural meadows. A post hoc Tukey
HSD analysis was also conducted to determine which of the experimental plots had shoot
densities similar to the natural meadows. The diversity and evenness of the benthic fauna in each
of the transplant plots were assessed using the Shannon-Wiener Diversity and Heip’s Evenness
Indices and where compared to each other using a one-way ANOVA and a post hoc Tukey HSD
analysis.
Similarity of the infauna abundances were analysed using the program Primer 6 (Clarke, 1993).
Both MDS plots and an ANOSIM analysis were performed on the data to determine how similar
each of the experimental transplant and control sites were to each other in terms of their infauna
abundances. This was achieved by doing a square root transformation on the infauna abundances
and using the Bray-Curtis similarity index. SIMPER analyses were performed on the data to
determine which of the infauna families were contributing to the bulk of the similarity.
28
3. Sampler Considerations 3.1 Introduction
With a variety of different methods available to sample infauna and with each method having its
own advantages, knowing which one to use becomes an important decision requiring careful
consideration. The different methods of sampling infauna include hand corers, suction samplers
and grabs (e.g., van Veen, Ekman); the aims of the study will determine which method will be
most appropriate.
Consideration is also needed on the size of the sampling device in determining how large an area
the sampling device needs to sample. Lewis and Stoner (1981) examined the effects of using hand
corers of varying diameter on the type and abundance of infauna collected. This study found that
the smaller 55 mm diameter hand corer collected significantly more infauna than 76 or 105 mm
corers and that the two larger corers underestimated the densities of many numerically abundant
infauna species. This was attributed mainly to the difference in the number of samples taken
using each corer, with the 55 mm corer having more samples and therefore having a greater
chance of sampling a dense infauna aggregation (Lewis and Stoner, 1981).
Similar results were also found in a study by Borg et al. (2002), who compared infauna
assemblages using 25, 35 and 45 cm diameter corers within Posidonia oceanica meadows. The
study concluded that smaller diameter corers provide better estimates of infauna abundances
compared to those with larger diameters. Given this, it can then be said that having many small
samples taken further apart allow for patchy distributed infauna to be more accurately
represented. A smaller diameter corer would also be more advantageous in that the processing
29
time of the samples would be shorter due to the smaller volume of sediment in the sample, a
finding also shown by Borg et al. (2002).
While choosing the appropriate sample area or diameter of the sampling device is an important
decision, the depth to which the chosen method samples is just as important. Research has
shown that the majority of infauna occupies the top five centimetres of the substrate (Lie and
Pamatmat, 1965; Lewis and Stoner, 1981; Hines and Comtois, 1985; Weston, 1990; Filgueiras et
al., 2007; Cardoso et al., 2010) and decreases thereafter. It is therefore important to select a
sampling method which will allow for sufficient penetration into the sediment in order to collect a
representative sample of the infauna present; however the appropriate depth needed will vary
depending on the aims and purpose of the study.
Examination of the effectiveness of different Ekman samplers by Blomqvis (1990) indicated that
not all the samplers were reliable at sampling the sediment as many of them produced
inadequate sample sizes due to mechanical flaws (i.e. tilting and sediment resuspension or loss).
An earlier study by Paterson and Fernando (1971) compared the use of Ekman grabs and hand
corers at sampling macrobenthic communities. Their findings showed that the hand corer was
more efficient at capturing infauna than the Ekman grab, however the corer was less effective at
sampling the less common or rare species. As well as being the less efficient sampling method the
Ekman grabs are also restricted to sampling within soft sediment environments as any large rocks,
shell, seagrass or coral would prevent the jaws of the trap from closing shut and result in the loss
of sediment and infauna.
30
Christie (1976) looked at the effectiveness of a diver operated suction sampler and found it to be
85 % effective at sampling both the common and rare infauna. A later study by Stoner et al.
(1983) compared the effectiveness of a sediment corer and suction dredge at sampling infauna in
both vegetated and unvegetated habitats. This research revealed that the hand corer was more
effective at sampling the infauna than the suction dredge. However, there was a difference in the
number of samples taken between the two methods (28 hand cores versus two suction samples),
which would have impacted on the accuracy of the infauna abundances. With substantially more
samples taken with the hand corer the chances of sampling a high abundance infauna patch are
greater and would result in a higher abundance estimate.
While all these sampling methods have their own advantages, only a few would be feasible for
consideration in this study. The grab samplers such as the van Veen and Ekman grabs would not
be viable options for sampling within the seagrass habitats. This is because the seagrass rhizome
would prove too difficult for the grabs to penetrate through and would also obstruct the sampler
when closing shut, resulting in sediment and infauna loss (Short and Coles, 2001; Southwood and
Henderson, 2000).
This chapter looks at assessing two different methods of sampling infauna, the hand corer and a
venturi suction dredge. To ensure a fair assessment of the two sampling methods, an equal
number of samples were collected using both the hand corer and suction dredge. In addition,
both samplers had the same internal diameter and were sampled to the same depth to ensure
that both methods were comparable in all respects. Samplers were compared in a similar manner
to Stoner et al. (1983) in both bare sand and seagrass habitats and assessed on the number and
abundance of infauna families sampled, as well as measures of diversity and evenness.
31
3.2 Method
3.2.1 Sampling Methodology
To compare the hand corer and suction dredge a total of 24 sediment samples (12 hand cores and
12 suction samples) were taken from each of the sites as shown in Figure 3. Missing samples and
incorrectly labelled samples were excluded from the analysis. Sediment samples were taken using
a venturi suction dredge and a PVC hand corer (Figure 4). Both samplers had an internal diameter
of 55 mm and sampled to a depth of 15 cm. For each sample, the hand core and suction dredge
samples were taken as close to each other as possible to minimize any spatial differences in the
infauna abundance and composition between the two sampling methods.
Figure 4: The two sediment samplers’ trialled for the study. (Left) Venturi suction dredge with air supplied by the SCUBA tank, (Right) PVC hand corer with serrated edge and rubber plug.
32
The hand corer was inserted into the sediment to a depth of 15 cm then sealed at the top with a
rubber plug, the sediment core was then removed and transferred into a calico bag and labelled.
A calico bag was attached to the end of the venturi suction dredge to collect the sediment that
was air lifted up and was held in place with an adjustable metal hose clamp. Once the suction
sample had been taken the air to the dredge was turned off and the suction dredge turned upside
down to allow any sediment in the pipe to settle down into the calico bag. The calico bag was
then detached from the suction dredge and labelled. All samples were stored, preserved, stained
and processed in the same manner described in the previous chapter.
3.2.2 Sampler Issues and Considerations:
A number of different issues became apparent in the field when trialling the suction dredge for
collecting the sediment samples. While some of these problems were easily fixed others proved
to be more problematic and compromising to the project. The issues associated with the sampler
and the actions taken to account for them are explained here:
Buoyancy
Due to the trapping of air in the calico bag the suction dredge became positively buoyant and
would lift away from the sediment. To counteract this, a six pound dive weight was attached to
the sampler to help keep it negatively buoyant and in contact with the substrate.
Faulty Equipment
As the suction dredge requires more complicated equipment and parts for it to work the chances
of faults occurring with the equipment are more likely. During the field trials a couple of faults
33
occurred with the suction dredge, the first being leaks from joints and connectors in the hose
which supplied air to the suction dredge. To solve this problem thread tape was used around all
the joints and connectors to provide a more air tight seal. The second problem was with the air
cylinders, as several of the o-rings burst on the tanks resulting in costly delays in the field work
due to having to replace the o-ring seals. As a result, spare equipment was needed on the boat to
ensure that any faults with the gear could be fixed or replaced; however the extra gear ended up
occupying a lot of space on the boat.
Cumbersome
The suction dredge’s bulky size and the added weight of carrying around the air cylinder along
with other sampling gear and sample bags made using the dredge rather difficult. To effectively
sample the sediment the suction sampler required two divers to operate it, compared to the
hand corer which could be used with ease by a single diver.
Area Sampled
As the suction dredge encountered the seagrass rhizome, sediment was drawn into the sampler
from outside the diameter of the dredge pipe and thus sampled sediment from a greater area
than was intended. This meant that it was not possible to directly compare the two samplers
based on the number of infauna per square meter. Instead the abundances were measured as the
number of infauna per unit volume of sediment sampled however it did not completely resolve
the problem. While both methods could be compared based on the volume of sediment sampled
a new problem of having the samplers collecting from different strata within the substratum
arises. When the suction dredge encounters the rhizome mat, it begins to suck sediment in from
34
the sides, drawing in more sediment and infauna from the surface layer, while the hand corer
collects a more even spread of sediment and fauna from each depth.
While the volume of sediment sampled was generally small, the extrapolation of the infauna
abundance to No. m-3 could also lead to unrealistic estimates. This is because infauna can be
rather patchy and locally abundant in particular areas which may lead to over estimation of some
of the abundances. Additional problems arise for both samplers from the use of volume to
estimate the infauna abundances. As the infauna may not be uniformly distributed through the
sediment column some infauna occupying a limited depth range would likely be underestimated
due to the volume of sediment sampled. Caution should then be used when interpreting the
finding of this study, knowing that any differences in infauna abundance between the two
samplers may be a result of the uneven sediment sampling exhibited by the venturi suction
dredge and over and under estimations from over extrapolating the data.
3.2.3 Statistical Analysis
Once the infauna had been identified and counted the abundance was calculated; results were
calculated as the number of infauna m-3 to provide a standardised value which would allow for
the two different methods to be compared. The total number of infauna families was counted
and compared along with the abundance data for both of the sampling methods at each site.
Shannon-Wiener and Heip’s Evenness indices were also calculated for each of the sampling
methods at both sites and compared using a two-way ANOVA. A comparison of total infauna
abundance between the two methods at the different sites was done using a two-way ANOVA
with infauna abundances log-transformed to meet the test’s assumptions. Similarity of infauna
assemblages between the two sampling methods was also compared using SIMPER, ANOSIM and
35
MDS plot analyses using the PRIMER 6 statistical package (Clarke, 1993). This was achieved by
doing a square root transformation on the infauna abundances and using the Bray-Curtis
similarity index.
3.3 Results
3.3.1 Diversity and Evenness
In all, 83 taxa were sampled using the hand corer while the suction dredge collected 93 taxa. A
total of 32 different taxa were collected by both sampling methods at the Bare Sand site while at
the Natural Meadow 1 site 51 taxa were collected by the hand corer and 60 were collected by the
venturi suction dredge. Overall 14 of the taxa sampled were unique to the hand corer while 20
were unique to the suction dredge; a more detailed list of the infauna families and their
abundances is given in Appendix 2.
At both the Bare Sand and Natural Meadow 1 sites the hand corer produced slightly higher values
for the mean Shannon-Wiener Index with 2.075 ± 0.109 Bels at the Bare Sand Site and 3.113 ±
0.158 Bels at the Natural Meadow 1 site. The venturi suction dredge on the other hand had
slightly lower values of 1.991 ± 0.090 Bels and 3.060 ± 0.222 Bels respectively.
Both site and sampling method were included in the two-way ANOVA model to look at their
effect on the Shannon-Wiener Index. The model produced a reasonable fit to the data with an R2
of 0.559, although only the site variable proved to have a significant effect on the Shannon-
Wiener Index (F=51.677, df=1, p<0.001). The sampling method variable did not significantly
improve the predictability of the model (F=0.220, df=1, p=0.641). This indicates that there is no
36
significant difference in the value of the Shannon-Wiener Index obtained using either sampling
method; therefore using either method would yield similar values.
The Heip’s Evenness Index was log transformed to meet the assumptions of the two-way ANOVA.
As with the Shannon-Wiener Index the site and sampling method variables were both included
into the two-way ANOVA model. The model provided a reasonable fit to the data with an R2 of
0.554. Only the site variable was found to significantly improve the model (F=50.706, df=1,
p<0.001); however as with the Shannon-Wiener Index the hand corer produced slightly higher
values for the mean log Heip’s Evenness Index at both the Bare Sand and Natural Meadow 1 sites
(Figure 5)
Figure 5: Mean log-transformed Heip’s Evenness Index for the Bare Sand and Natural Meadow 1 sites using both the hand corer and venturi suction dredge
Retransforming the log Heip’s Evenness Index allowed for easier interpretation of the results and
revealed low values of 0.078 ± 0.011 for the hand corer and 0.068 ± 0.007 for the suction dredge
37
at the Bare Sand site with values of 0.245 ± 0.030 for the hand corer and 0.255 ± 0.048 for the
suction dredge at the Natural Meadow 1 site. These low values indicate that there is a lot of
variation in numbers of individuals within different infauna communities. The results of the two-
way ANOVA showed that sampling method did not significantly improve the model which means
that it was not having a significant effect on the Heip’s Evenness Index. Therefore it can be said
that both sampling methods would provide similar estimates of the Heip’s Evenness Index.
3.3.2 Infauna Comparison
The mean log infauna abundances sampled with the suction dredge were slightly higher than
those taken using the hand corer at the bare sand site with 5.161 ± 0.085 and 5.099 ± 0.072 m-3
respectively (Figure 6). The inverse was observed for samples collected at the Natural Meadow 1
site with the hand corer having a mean log infauna abundance of 5.505 ± 0.064 compared with
5.452 ± 0.150 m-3 for the suction dredge (Figure 6). This change in the mean log infauna
abundances between the two sites when sampled with the different methods indicates a possible
interaction between the sites sampled and the method used.
The results showed that neither the sampling method (F=0.003, df=1, p=0.960) nor the
interaction term (F=0.373, df=1, p=0.545) was having a significant impact on the log infauna
abundance. However the model did reveal a significant difference in response to the different
sites that were sampled (F=13.504, df=1, p=0.001), with the mean log infauna abundance being
significantly higher in the Natural Meadow 1 site.
38
Figure 6: Mean log of infauna abundances for the Bare Sand and Natural Meadow 1 sites using both the hand corer and venturi suction dredge
The comparison between the different sites across the two sampling methods returned a Global R
statistic of 0.631 which indicates that the infauna assemblages collected between these two sites
are sufficiently distinct from one another. The comparison of the hand corer and venturi suction
dredge by means of the two-way ANOSIM gave a low Global R statistic of 0.172 meaning that
there was little difference in the composition of the infauna between the two sampling method.
This is further supported by the MDS plot in Figure 7 which shows clear separation of the samples
taken from the two sites. It can also be seen that the samples have been partitioned based on the
different sampling methods used, however they are not dissimilar enough to form distinct
clusters and hence the low Global R statistic.
39
Figure 7: MDS plot of the square root transformed infauna abundance data
To determine what infauna families contributed most to the dissimilarity between the different
sites and sampling methods a SIMPER analysis was performed. The average dissimilarity between
the two sampling methods was 53.79 % with Tellinidae, Nematoda, Spirorbidae, Rutidermatidae,
Lumbrineridae, Veneridae, Syllidae, Bullidae, Oenonidae and Onuphidae accounting for 50 % of
the dissimilarity. This indicates that there is a reasonable amount of overlap in the type of infauna
collected by both samplers. The average dissimilarity between each of the samples from each
method was 50.20 % for the hand corer and 48.35 % for the suction dredge, indicating that there
is also a reasonable amount of variability in the infauna collected within the different sampling
methods.
Comparisons were also made between the Bare Sand and Natural Meadow 1 sites with an
average dissimilarity of 64.25 %, with 50 % of the dissimilarity attributed to by the Spirorbidae,
Transf orm: Square root
Resemblance: S17 Bray Curtis similarity
SiteBare Sand
Natural Meadow 1
Similarity
50
60
CoreCore
Core
Core
Core
Core
CoreCore
Core
CoreCoreCore
Dredge
Dredge
Dredge
Dredge
Dredge
Dredge
DredgeDredge
Dredge
Dredge
Dredge
CoreCore
Core
Core
Core
Core
Core
CoreCore
Core
Core
DredgeDredge
Dredge
Dredge
Dredge
Dredge Dredge
Dredge
Dredge
Dredge
2D Stress: 0.17
40
Tellinidae, Nematoda, Aoridae, Syllidae, Onuphidae, Rutidermatidae, Veneridae, Lumbrineridae,
Oenonidae and Turbinidae taxa. Comparisons of the individual samples from each site revealed
an average dissimilarity of 53.17 % for the Bare Sand site and 44.76 % for Natural Meadow 1. This,
along with the comparison between the different methods, shows that there is a fair amount of
variability within the samples from each site and method and a distinct difference between
samples from the different sites.
3.4 Discussion
The findings have shown that the venturi suction dredge sampled more taxa with 93 sampled
compared to the 83 taxa sampled by the hand corer. This greater number of taxa collected with
the suction dredge can be attributed to the fact that it is able to sample both the benthic infauna
as well as the epifauna (Short and Coles, 2001). Sampling both the benthic and epifauna would
then provide an additional array of taxa to be sampled in comparison to the hand corer which
predominantly samples just the benthic infauna. Despite the difference in the number of taxa
sampled, both methods provided similar values for the mean Shannon-Wiener and Heip’s
Evenness indices. These values were marginally higher in the hand corer than in the suction
dredge; however they were not statistically significant.
The results also showed no statistically significant difference in the total number of infauna
sampled by each method at either the Bare Sand or Natural Meadow 1 sites. This is in direct
contrast to the findings by Stoner et al. (1983) who found that the suction sampler under-
sampled by as much as 72.8 % in bare sand habitats and 32.6 % within natural seagrass in relation
to the hand corer. These differences in the findings may be attributed to the fact that Stoner et al.
(1983) only took two samples with the suction dredge and 28 hand cores whereas in this study
41
equal numbers of samples were taken using samplers with the same diameter. Such differences
could also be a result of the different seagrass species which were examined, with Stoner et al.
(1983) sampling in Halodule wrightii while this study sampled within Posidonia australis.
Comparisons of infauna abundances through the two-way ANOSIM and MDS plots indicated that
there was a lot of overlap in the infauna assemblages between the two sampling methods
meaning that neither method collected distinctly different infauna assemblages. The results also
showed that there was variability between samples taken by the same sampler, which is
indicative of the patchy nature and localised abundance of infauna (Ramey et al., 2009).
The results have indicated that both sampling methods collect similar abundances of infauna and
sample similar infauna assemblages, therefore either method would be suitable for this project.
The only advantage that the venturi suction dredge appears to have over the hand corer is its
ability to sample a greater number of taxa, which would be useful in determining if all the infauna
associated with a natural meadow has returned to the transplanted seagrass plots. However,
while both infauna and epifauna are collected by the suction dredge there is as yet no way of
being able to separate these different fauna out from the samples (Short and Coles, 2001).
In addition to sampling effectiveness of the samplers, the practicality of the associated sampling
methods also need to be taken into consideration. In this case the simplicity of the hand corer
proves to be more practical and easy to use being small in size relative to the venturi suction
dredge, requiring only one operator to use and not having any mechanical or technical
components which may break or become faulty. Given that both sampling methods yield similar
results in Shannon-Wiener and Heip’s Evenness indices, total infauna abundances and sample the
42
same infauna assemblages; picking the best method would then depend on the samplers’
practicality. Therefore it can be concluded that the hand corer would be the most appropriate
method to conduct the sampling with due to its simplicity, light weight and ease of use.
43
4. Comparison of Transplanted and Natural Meadows 4.1 Seagrass Shoot Density
Similar total shoot densities were measured at Natural Meadow 1, Natural Meadow 2 and the
two higher density 0.25 m and 0.125 m plots, with all sites having a mean shoot density greater
than 500 shoots m-2 (Figure 8). The 0.25 m Plot also had a shoot density which was greater than
either of the two natural meadow sites with a mean of 616.500 ± 13.219 shoots m-2. Both of the
lower density 1 m and 0.5 m Plots had substantially fewer shoots with less than 500 shoots m-2 in
both plots (Figure 8). A one-way ANOVA revealed that the mean shoot density differed
significantly among the different sites (F=30.746, df=5, p<0.001). A post hoc Tukey test showed
that the mean shoot density in the 0.25 m and 0.125 m Plots was significantly higher than in the 1
m and 0.5 m Plots; and significantly higher in the 0.25 m Plot than at all other sites.
These findings indicate that the mean shoot densities in the 0.125 m and Natural Meadows 1 and
2 are not significantly different from each other meaning that the 0.125 m Plot has reached shoot
densities that match those of the natural meadows. The 0.25 m Plot had a mean shoot density
significantly larger than the all other sites, indicating that it has surpassed the mean density of the
natural meadows as well.
Edge effects were also examined in relation to shoot density to see if the sites were denser in the
centre. Figure 9 shows the mean shoot density in the outer, middle and centre zones changing at
each site; such changes indicate that there is a potential interaction occurring between the edge
zone and the sites in relation to the shoot density. To determine if the shoot density was affected
44
Figure 8: Mean shoot density of the natural and transplanted seagrass on Southern Flats, Cockburn Sound
Figure 9: Shoot density in each zone for the natural and transplanted seagrass on Southern Flats, Cockburn Sound.
45
by edge effects at different sites a two-way ANOVA was performed using a model which included
the site, edge zone and the interaction between the site and edge. The model produced a good fit
to the data with an R2 of 0.655 which means that 65.5 % of the data points were explained by the
model. Both the site (F=25.957, df=5, p<0.001) and the interaction between site and edge
(F=4.169, df=10, p<0.001) were significant, meaning that the shoot density in each of the three
edge zones changed in relation to the different sites.
4.2 Infauna
4.2.1 Processing and Sorting Effectiveness
Determining the efficiency to which the infauna were removed from the sorting tray after the
washing and rinsing process is of importance as it provides an indication of how effective the
sorting was but also whether particular infauna were being under estimated. Of the 44 samples
processed 59.70 ± 2.67 % of the infauna were removed by the end of the washing process with
40.29 ± 2.67 % left remaining in the sorting tray. The majority of the infauna remaining in the tray
consisted primarily of taxa possessing heavy shells, exoskeletons or calcified tubes such as the
bivalves, gastropods and polychaetes (Table 1). The five infauna families with the largest
proportions left behind in the sorting trays were the Tellinidae, Veneridae (Venus Clams), Bullidae
(Bubble Shells), Spirobidae and Batillariidae (Creepers) with 69.40, 64.70, 38.81, 31.34 and 22.73
% respectively (Table 1).
Examination of how effective the sorting was at removing all the infauna from the 15 samples
processed revealed that 80.38 ± 3.17 % of all infauna was removed at the end of the first sorting.
It was also noted that those which were removed during the second sorting were generally of
considerably smaller size and difficult to see. A total of 16 different taxa were missed during the
first sorting, with the five taxa with the largest percentages missed belonging to the Nematoda,
Epitoniidae, Rutidermatidae, Batillariidae and Tellinidae with 41.216, 27.333, 18.889, 16.667 and
46
15.347 % respectively (Table 2). The taxa present within Table 2 provide an indication as to how
much the abundance estimates for each family are being underestimated and thereby allow for a
more accurate representation of the infauna abundances within this study.
4.2.2 Infauna Diversity and Evenness
The greatest number of taxa was found at the Natural Meadow 1 site with 50 taxa, 10 of which
were unique to that site. This was followed by the 0.125 m Plot with 46 taxa, nine of which were
unique; and then the Natural Meadow 2 site with 44 taxa and five unique taxa. The 1 m Plot had
35 taxa three of which were unique to that site; 32 taxa were found at the Bare Sand site with
three unique taxa; 31 taxa were found at the 0.25 m Plot with only one unique taxon; and the 0.5
m Plot had the least with 27 taxa with only two being unique to that site. This shows that there is
a great deal of variability in the number of taxa present at each site with no progressive increase
from the Bare Sand site up through the increasing planting density transplants to the higher
density natural meadows. However it should be noted that the 0.125 m Plot did have similar
numbers of taxa which were present and unique compared to those of the two natural meadows.
While no distinct trend was observed in regard to the total numbers of taxa found at each site the
Shannon-Wiener Index tells a different story. The diversity index increased in the higher seagrass
planting densities. The greatest diversity was at Natural Meadow 1 with a Shannon-Wiener Index
of 3.112 ± 0.522 Bels; the lowest was at the Bare Sand site with 2.075 ± 0.378 Bels (Figure 10).
Heip’s Evenness followed the same trend with the lowest value of 0.078 ± 0.038 being recorded
at the Bare Sand site and the highest value of 0.245 ± 0.101 at Natural Meadow 1 (Figure 11).
A significant difference was detected in the mean Shannon-Wiener Index (F=3.930, df=6, p=0.002)
indicating that the mean Shannon-Wiener Index at each site is not the same. The post hoc Tukey
47
Table 1: The number of infauna from each family remaining in the tray after the rinsing and washing process (n=44)
Taxa Min. Max. Mean SE % In Tray
Amphipoda
Aoridae 0 1 0.068 0.038 3.220
Caprellidae 0 1 0.023 0.023 2.273
Cyproideidae 0 1 0.023 0.023 0.758
Ischyroceridae 0 1 0.023 0.023 1.136
Phoxochephalidae 0 1 0.068 0.038 5.682
Cirripeda
Balanidae 0 1 0.045 0.032 4.545
Bivalves
Pectinidae 0 1 0.023 0.023 2.273
Solemyidae 0 2 0.114 0.058 8.333
Solecurtidae 0 1 0.023 0.023 2.273
Tellinidae 0 59 14.591 2.001 69.395
Veneridae 0 10 1.886 0.316 64.697
Decapoda
Diogenidae 0 1 0.045 0.032 4.545
Gastropoda
Batillariidae 0 20 0.886 0.477 22.727
Buccinidae 0 1 0.068 0.038 6.818
Bullidae 0 7 0.955 0.258 38.813
Columbellidae 0 4 0.182 0.099 10.227
Epitoniidae 0 1 0.227 0.064 20.455
Fissurellidae 0 1 0.023 0.023 2.273
Hydatinidae 0 1 0.023 0.023 2.273
Mitridae 0 1 0.045 0.032 4.545
Naticidae 0 4 0.227 0.102 14.773
Olividae 0 1 0.023 0.023 2.273
Terebridae 0 3 0.136 0.083 6.818
Trochidae 0 4 0.364 0.130 18.864
Turbinidae 0 19 0.932 0.457 21.071
Nematoda 0 8 0.727 0.235 5.324
Ostracoda
Order: Podocopida 0 2 0.136 0.062 6.629
Rutidermatidae 0 1 0.023 0.023 0.175
Polychaetes
Lumbrineridae 0 2 0.114 0.058 5.871
Maldanidae 0 1 0.023 0.023 0.758
Oenonidae 0 2 0.091 0.055 3.030
Onuphidae 0 2 0.091 0.064 3.409
Paraonidae 0 1 0.045 0.032 4.545
Spirorbidae 0 133 9.545 4.235 31.344
Syllidae 0 1 0.023 0.023 2.273
Polyplacophora
Ischnochitonidae 0 1 0.045 0.032 3.409
Tanaidacae
Tanaidae 0 1 0.023 0.023 1.136
48
Table 2: The number of infauna missed during the first sorting.
Taxa Min. Max. Mean SE % Missed
Amphipoda
Aoridae 0 1 0.067 0.067 6.667
Ischyroceridae 0 1 0.067 0.067 6.667
Bivalves
Solemyidae 0 1 0.067 0.067 0.952
Tellinidae 0 10 2.267 0.665 15.347
Veneridae 0 3 0.600 0.254 14.365
Copepoda
Order: Harpacticoid 0 1 0.067 0.067 6.667
Gastropoda
Batillariidae 0 1 0.200 0.107 16.667
Epitoniidae 0 3 0.467 0.215 27.333
Naticidae 0 1 0.067 0.067 6.667
Turbinidae 0 1 0.200 0.107 12.222
Nematoda 0 10 3.467 0.703 41.216
Ostracoda
Order: Podocopida 0 1 0.067 0.067 3.333
Rutidermatidae 0 2 0.400 0.190 18.889
Polychaetes
Lumbrineridae 0 3 0.200 0.200 5.000
Spirorbidae 0 1 0.133 0.091 8.889
Syllidae 0 2 0.267 0.153 11.333
test revealed significant differences in the mean Shannon-Wiener Index between the Bare Sand
site and Natural Meadow 1 (p<0.001), and between Natural Meadow 1 and both the 1 m and 0.5
m Plots (p=0.015 and p=0.016 respectively). The same analysis was performed for the Heip’s
Evenness Index with the mean Heip’s Evenness Index found to be significantly different (F=5.042,
df=6, p<0.001). Significant differences between Natural meadow 1 and the Bare Sand site, 1 m
Plot, 0.5 m Plot and the 0.25 m Plot were also observed, with p-values of <0.001, 0.002, 0.003 and
0.016 respectively.
4.2.3 Infauna Abundances
Infauna abundance appeared to increase with the increasing seagrass planting densities with a
mean abundance of 10,592.792 ± 1,777.339 infauna m-2 at the 1 m Plot, increasing to 25,407.395
± 10,829.971 infauna m-2 at the 0.125 m Plot (Figure 12). Despite the increasing abundances the 1
m, 0.5 m and 0.25 m Plots all had means which were lower than that of the Bare Sand site which
49
Figure 10: Mean Shannon-Wiener Diversity Index for each of the control and experimental plots on Southern Flats
Figure 11: Mean Heip’s Evenness Index for each of the control and experimental plots on Southern Flats
50
had a mean of 18,169.093 ± 2,590.927 infauna m-2 (Figure 12). Natural meadow 1 had the
greatest abundance of infauna with a mean of 29,807.772 ± 6267.453 infauna m-2 followed by the
0.125 m Plot and Natural Meadow 2 with mean abundances of 25,407.395 ± 10,829.971 and
21,772.301 ± 3,714.777 infauna m-2 respectively (Figure 12). It should be noted that these
estimates are likely to be underestimates as they only represent 80.38 % of the infauna that were
removed by the sorting process, as indicated previously.
The two-way ANOVA used both site and edge zone variables in the model, which explained 25.9
% of the data points (R2=0.259). The site variable was a significant predictor of the mean log
infauna abundance (F=3.754, df=6, p=0.003) while the edge zone was not (F=1.329, df=2,
p=0.271). A post hoc Tukey test for the site variable indicated that the only significant difference
in the mean log infauna abundance was between Natural Meadow 1 and the 1 m Plot (p=0.002).
Comparisons of infauna assemblages using the MDS plot in Figure 13 showed little separation of
the data points into distinct groups with many of the points from different sites overlapping with
those from other sites. Despite the large amount of overlap there does appear to be some slight
separation of the data points based on the sites, though no separation or grouping is seen for the
different edge zones (Figure 13). The high stress level of the MDS plot (2D stress: 0.26) indicates
that the clustering of the data points are not providing a very reliable representation of the
similarity between the samples and sites.
To gain a better representation of the similarity of the infaunal assemblages between the
different control and experimental seagrass transplants site as well as the different edge zones,
an ANOSIM analysis was performed. A Global R statistic of 0.279 was obtained for the between
site differences indicating that overall there was little separation of the infauna assemblages
between the different sites. The between edge zones also gave a low Global R statistic of 0.145
51
meaning that overall there was no difference in the infauna assemblages in the different edge
zones. Despite the overall lack of separation between the different sites in terms of the infauna,
individual comparisons (shown in Table 3) revealed separation between some sites.
Figure 12: Infauna abundances for the control and experimental sites on Southern Flats.
Comparisons between Natural Meadow 1 and Bare Sand, Natural Meadow 2 and Bare Sand,
Natural Meadow 1 and the 1m Plot, Natural Meadow 1 and the 0.5 m Plot and between Natural
Meadow 1 and Natural Meadow 2 all revealed overlapping but distinctly separate infauna
assemblages (Table 3). Both the 0.25 and 0.125 m Plots gave low R statistics when compared with
Natural Meadow 1 and 2 indicating that there was little to no difference in the infauna
assemblages between the high density seagrass transplants and the natural seagrass meadows
(Table 3).
52
Figure 13: MDS plot of the square root transformed infauna abundance data showing similarities of the infauna assemblages between each site and edge zone at Southern Flats, Cockburn Sound.
The SIMPER analysis indicated that in most cases the five most highly abundant taxa Nematoda,
Tellinidae, Lumbrineridae, Onuphidae and Veneridae generally contributed the greatest amount
to the dissimilarity between the different sites. A one-way ANOVA was performed on these taxa
and found that the abundances of Nematoda (F=1.033, df=6, p=0.411), Tellinidae (F=1.407, df=6,
p=0.224) and Lumbineridae (F=0.899, df=6, p=0.500) were not significantly different between the
sites while Onuphidae (F=10.323, df=6, p<0.001) and Veneridae (F=5.737, df=6, p<0.001) were.
Significant differences in infauna abundances between different sites were also seen in 16 other
taxa, as shown in Appendix 3.
Several infauna taxa were recorded in both of the natural seagrass meadows as well as in some of
the high planting density seagrass transplants. Eusiridae, Solecurtidae, Diogenidae,
Columbellidae, Fissurellidae, Oweniidae and Ischnochitonidae were found at both Natural
Meadow 1 and 2 with Eusiridae also occurring in the 0.125 m Plot and Diogenidae and
Columbellidae both occurring at the 0.25 and 0.125 m Plots.
Transform: Square root
Resemblance: S17 Bray Curtis similarity
SiteBare Sand
Natural Meadow 1
0.25 m Plot
0.125 m Plot
1 m Plot
0.5 m Plot
Natural Meadow 2
Outer Outer
OuterOuter
Middle
Middle
Middle
MiddleCentre
CentreCentreCentre
Outer
Outer
Outer Outer
Middle
Middle
Middle
Centre
Centre
CentreCentre
Outer
Outer Outer
Middle
Middle
Middle
Middle
Centre
CentreCentre
Outer Outer
Outer
Outer
Middle
Middle
Middle
Centre
Centre
Centre
Centre
Outer
Outer
OuterOuter
Middle
Middle
Middle
Middle
CentreCentre
Centre Centre
Outer
Outer
OuterOuter
Middle
Middle
Middle
Middle
CentreCentreCentre
Centre
Outer
Outer
Outer
Outer
Middle
Middle
MiddleCentre
Centre
Centre
Centre
2D Stress: 0.26
53
Table 3: R statistic outputs from the ANOSIM analysis for the infauna comparisons between the sites and edge zones. The R statistic ranges from 1 to -1 with values >0.75 indicating that the infauna assemblages are separate from each other, values >0.5 indicating some overlap but still forming distinct groups and a values <0.25 indicating that there is no difference in the infauna assemblages. Significance level is set at 5 % (α=0.05)
Group V's Group R Statistic Sig. (%)
Site
Bare Sand Natural Meadow 1 0.645 0.1
Bare Sand 0.25 m Plot 0.27 0.7
Bare Sand 0.125 m Plot 0.24 0.4
Bare Sand 1 m Plot 0.257 1.6
Bare Sand 0.5 m Plot 0.333 1.1
Bare Sand Natural Meadow 2 0.519 0.1
Natural Meadow 1 0.25 m Plot 0.241 2.9
Natural Meadow 1 0.125 m Plot 0.256 0.8
Natural Meadow 1 1 m Plot 0.507 0.1
Natural Meadow 1 0.5 m Plot 0.624 0.1
Natural Meadow 1 Natural Meadow 2 0.533 0.1
0.25 m Plot 0.125 m Plot -0.117 84.3
0.25 m Plot 1 m Plot 0.039 35.2
0.25 m Plot 0.5 m Plot 0.128 12.4
0.25 m Plot Natural Meadow 2 0.272 1.7
0.125 m Plot 1 m Plot 0.185 4
0.125 m Plot 0.5 m Plot 0.287 0.2
0.125 m Plot Natural Meadow 2 0.301 2.3
1 m Plot 0.5 m Plot 0.059 30.6
1 m Plot Natural Meadow 2 0.148 9
0.5 m Plot Natural Meadow 2 0.366 0.2
Edge Zone
Outer Middle 0.022 37.6
Outer Centre 0.256 0.2
Middle Centre 0.156 1.9
54
5. Discussion 5.1 Shoot Density
Four years after their initial transplantation onto Southern Flats the two high planting density P.
australis transplants, 0.25m and 0.125 m Plots, had reached structurally equivalent levels
compared to nearby natural meadows. It was observed that the shoot density in the 0.25 m Plot
was significantly greater than any of the other sites with 616.500 ± 13.219 shoots m-2 with
differences also observed in the different edge zones at different sites.
Bivalves have been shown to increase the nitrogen and phosphorus content of the sediment,
creating a mosaic of nutrient rich patches (Peterson and Heck Jr., 1999; 2001ab). Bivalve density
manipulation experiments by Peterson and Heck Jr. (2001ab) have shown that the presence of
bivalves within seagrass meadows enhance the growth and productivity of the seagrass which in
turn increases survivorship of the bivalves. It is therefore possible, that the presence of bivalves
within the transplanted P. australis were responsible for the different shoot densities observed at
different sites and edge zones. Examination of the total number of bivalves and of individual
bivalve families however found no connection between the bivalves and the shoot densities
observed.
Another explanation for the higher shoot density observed in the 0.25 m Plot could be because
that planting the shoots out at 0.25 m intervals may be the optimal planting density for P.
australis, with more sprigs being planted per meter while still having enough space for them to
expand. Planting at 0.125 m intervals may have resulted in overcrowding and hindered growth of
the transplants, hence the lower shoot density in the 0.125 m Plot. The initial planting of the high
density transplant plots in 2007 was undertaken by more experienced divers while the lower
density plots were planted by less experienced divers which may have accounted for some of the
differences in the shoot density (Van Keulen, Murdoch University, pers. com.).
55
Rehabilitation efforts conducted at Oyster and Princess Royal Harbour using P. australis and
P. sinuosa showed transplants recovering to levels comparable to natural meadows five years
post transplantation (Cambridge et al., 2002; Bastyan and Cambridge, 2008). Compared with this
study, the P. australis transplants in Cockburn Sound showed structural equivalence to natural
meadows in the 0.25 and 0.125 m Plots after four years; this is due to the different planting
densities used in the Oyster and Princess Royal Harbour studies which were planted out at 1 m
intervals. Based on the findings from Cambridge et al. (2002) and Bastyan and Cambridge (2008)
it can be anticipated that the 1 m and 0.5 m Plots in Cockburn Sound will be structurally
equivalent to natural meadows in another one to two years.
5.2 Infauna
5.2.1 Processing and Sorting Effectiveness
Overall the effectiveness of the infauna sorting was good, with up to 80.38 % of the infauna being
removed from the sediment. The underestimation of the Nematoda and families of polychaetes
was largely the result of large quantities of seagrass material being present within the samples,
with many of the nematodes and polychaetes getting in among the seagrass fibres making them
difficult to find during the first sorting.
5.2.2 Infauna Abundances
Natural Meadow 1 and Natural Meadow 2 differed from each other in regards to their infauna
abundances as well as their assemblages, as indicated by the ANOVA and ANOSIM. Such
differences between the two can be explained by their exposure to different hydrodynamic
conditions. As Natural Meadow 2 is located within the transplantation meadow the water velocity
and turbulence would be much lower (Backhaus and Verduin, 2008; Morris et al., 2008; Lefebvre
et al., 2010). As the water velocity decreases with distance into the meadow any infauna recruits
being transported by the water would settle out before reaching Natural Meadow 2 (Macreadie
56
et al., 2010; Murphy et al., 2010; Smith et al., 2011). Natural Meadow 1, on the other hand, was
outside the transplantation meadow, so any transported infauna recruits would settle out onto
the meadow, giving it greater infauna abundance.
Both the 0.25 m Plot and 0.125 m Plots had infauna abundances and assemblages comparable to
Natural Meadow 1 and Natural Meadow 2 within four years after their initial transplantation. This
recovery falls in line with other estimations of benthic infauna recovery recorded for
Halodule wrightii which had recovery times of three to five years (Sheridan, 1998; Sheridan et al.,
2003; Sheridan, 2004). Recovery times did vary considerably with other studies and seagrasses,
with 1.5 years in Cymodocea nodosa (Paranovi et al., 2000), two years in Zostera marina (Evans
and Short, 2005) and 1.8 years in a study by Fonseca et al. (1996) with H. wrightii. Such variation
in the recovery times of the infauna can be attributed in part to the different growth rates
exhibited by the different seagrass’, with Z. marina and C. nodosa having fast growth rates
(Olesen and Sand-Jensen, 1994; Vidondo et al., 1997) and hence the faster time for the infauna to
reach comparable levels to natural meadows.
The results also showed no significant differences in the outer, middle or centre edge zones on
the infauna abundances or assemblages which is in contrast to findings from other studies
(Tanner, 2005; Warry et al., 2009; Macreadie et al., 2010; Murphy et al., 2010; Smith et al., 2011).
The main reason for this is that these other studies examined individual infauna families across
seagrass patches surrounded by sand, while this study had different density plots located within a
transplantation meadow and looked at the overall infauna abundances and assemblages. The
work by Tanner (2005) showed that only certain infauna respond to edge effects, with most
bivalves and polychaetes not being impacted. Murphy et al. (2010) also stated that edge effects
could not be generalised across seagrass habitats with the effects differing from taxon to taxon.
57
Eusiridae, Solecurtidae, Diogenidae, Columbellidae, Fissurellidae, Oweniidae and
Ischnochitonidae were all found to occur within the two natural meadows with Eusiridae,
Diogenidae and Columbellidae in the higher planting density transplants. As these infauna were
only found in natural meadows and transplanted sites which had attained comparable levels of
shoot density and infauna assemblages, it can be suggested that these families may represent
climax or K-species, indicating the transition to a state comparable to natural seagrass meadows.
However as this study was only a snapshot of the recovery of transplanted seagrass, long term
monitoring would be required to see if they persist within the 0.25 and 0.125 m plots as well as
the natural meadows. Monitoring for their presence would also be required within the 1 and 0.5
m Plots to determine if they occur once the shoot density has reached comparable levels to the
natural meadows.
The presence of other infauna within the transplanted seagrass also gives an indication of how
well the ecosystem is developing. In particular were the presences of harpacticoid copepods,
including individuals from the family Peltitiidae. Research has shown that the harpacticoid
copepods form a large proportion of the diet for King George Whiting (Sillaginodes punctatus), a
valuable commercial and recreational fish species (Jenkins et al., 2011). The presence of the
Harpacticoida copepods indicates that the transplanted seagrass is capable of providing a vital
food source as well as foraging areas for the valuable King George Whiting.
The Western Australian Seahorse (Hippocampus subelongata) and Wide-bodied Pipefish
(Stigmatopora nigra) were both observed within the natural seagrass sites and experimental
transplant plots as well as in the surrounding transplant meadow. With both of these species
feeding on copepods and H. subelongatus on nematodes and polychaetes (Kendrick and Hyndes,
2005), the copepods, nematodes and polychaetes may have sufficiently recovered to be able to
support small numbers of these higher order predators.
58
A dietary study by Smith et al. (2011) on Stigmatopora nigra revealed that on average 89.4 % of
the fish ingested between 12 to 17 Harpacticoid copepods. Examination of the number of
harpacticoid copepods found in each site from this study revealed that with the exception of the
0.5 m Plot, which had none, the number of Harpacticoid copepods ranged from 35.075 to
153.057 copepods m-2. This then indicates that the seagrass is able to provide sufficient
harpacticoid copepod prey to supply a food source for these higher order predators.
In addition a Sea lion (Neophoca cinerea) was observed feeding on an adult Blue-Manna Crab
(Portunus pelagicus) in nearby transplanted seagrass during the study; this indicates that the
transplanted seagrass is currently providing food and foraging grounds for an array of higher
order predators. However it could also be said that the transplanted seagrass meadow (planted
out at 0.5 m intervals) is not providing sufficient protection to the associated marine
invertebrates from predators, with research having shown that survival of invertebrates increases
with increasing shoot density (Hovel and Lipcius, 2001; Peterson and Heck Jr., 2001ab; Hovel,
2003).
59
6. Conclusion It is evident that after four years post transplantation, the Posidonia australis seagrass in the 0.25
and 0.125 m Plots have attained structurally equivalent levels of shoot densities, as well as having
infauna abundances and assemblages, equivalent to those of Natural Meadows 1 and 2. While
not currently at levels comparable to Natural Meadow 1 and 2, the 1 m and 0.5 m Plots are likely
to reach equivalent levels within the next one to two years. To ensure that the 1 and 0.5 m Plots
attain equivalent levels of shoot density and infauna abundances and assemblages, long term
monitoring of these sites throughout the year would be advisable.
Monitoring would also give insight to the seasonal variability in the infauna communities and
provide an indication of the importance of the Eusiridae, Solecurtidae, Diogenidae,
Columbellidae, Fissurellidae, Oweniidae and Ischnochitonidae as possible indicators of succession
to a state comparable to the natural P. australis meadows. Future monitoring would also benefit
from looking at the succession in the larger macrobenthic invertebrates including iconic seagrass
species such as Razor Clams (Pinna bicolour), Blue-Manna Crabs (Portunus pelagicus) and
cephalopods.
60
References Alexander, R; Cumbers, M; Kagi, R; Offer, M and Taylor, R, 1982; Petroleum contamination in Cockburn Sound, Western Australia, Toxicological and Environmental Chemistry, vol.5, pp.251-275 Backhaus, J.O and Verduin, J.J, 2008; Simulating the interaction of seagrass with their ambient flow, Estuarine, Coastal and Shelf Science, vol.80, pp.563-572 Bastyan, G.R and Cambridge, M.L, 2008; Transplantation as a method for restoring the seagrass Posidonia australis, Estuarine, Coastal and Shelf Science, vol.79, pp.289-299 Beer, T, 1997; Chapter 2: Shore Processes, In: Environmental Oceanography, 2nd Ed., CRC Press Bell, J.D; Westoby, M and Steffe, A.S, 1987; Fish larvae settling in seagrass: do they discriminate between beds of different leaf density?, Journal of Experimental Marine Biology and Ecology, vol.111, pp.133-144 Bell, S.S; Tewfik, A, Hall, M.O and Fonseca, M.S, 2008; Evaluation of seagrass planting and monitoring techniques: Implications for assessing restoration success and habitat equivalency, Restoration Ecology, vol.16, pp.407-416 Bjorndal, K.A, 1980; Nutrition and grazing behaviour of the green turtle Chelonia mydas, Marine Biology, vol.56, pp.147-154 Blomqvis, S, 1990; Sampling performance of Ekman grabs - in situ observations and design Improvements, Hydrobiologia, vol.206, pp.245-254 Bonham, A.J, 1983; The management of wave-spending vegetation as bank protection against boat wash, Landscape Planning, vol.10, issue.1, pp.15-30 Borg, J.A; Attrill, M.J; Rowden, A.A; Schembri, P.J and Jones, M.B, 2002; A quantitative technique for sampling motile macroinvertebrates in beds of the seagrass Posidonia oceanica (L.) Delile, Scientia Marina, vol.66, issue.1, pp.53-58 Bos, A.R; Bouma, T.J; de Kort, G.L.J and van Katwijk, M.M, 2007; Ecosystem engineering by annual intertidal seagrass beds: Sediment accretion and modification, Estuarine, Coastal and Shelf Science, vol.74, pp.344-348 Bouillon, S; Moens, T and Dehairs, F, 2004; Carbon sources supporting benthic mineralization in mangrove and adjacent seagrass sediments (Gazi Bay, Kenya), Biogeosciences, vol.1, pp.71-78 Brearley, A and Wells, F.E, 2000; Invertebrate fauna in seagrass on Success Bank, Western Australia, Biologia Marina Mediterranea, vol.7, pp.199-202 Brun, F.G; Perez-Llorens, J.L; Hernandez, I and Vergara, J.J, 2003; Patch distribution and within-patch dynamics of the seagrass Zostera noltii Hornem. in Los Torunos salt-marsh, Cadiz Bay, Natural Park, Spain Butler, A and Jernakoff, P, 1999; Seagrass in Australia: Strategic Review and Development of an R and D Plan, CSIRO Publishing
61
Cambridge, M.L, 1980; Ecological studies on seagrass of South Western Australia with particular reference to Cockburn Sound, PhD dissertation, University of Western Australia Cambridge, M.L; Bastyan, G.R and Walker, D.I, 2002; Recovery of Posidonia meadows in Oyster Harbour, Southwestern Australia, Bulletin of Marine Science, vol.72, pp.1279-1289 Cambridge, M.L; Chiffings, A.W; Brittan, C; Moore, L and McComb, A.J, 1986; The loss of seagrass in Cockburn Sound, Western Australia. II. Possible causes of seagrass decline, Aquatic Botany, vol.24, pp.269-285 Cambridge, M.L; How, J.R; Lavery, P.S and Vanderklift, M.A, 2007; Retrospective analysis of epiphyte assemblages in relation to seagrass loss in a eutrophic coastal embayment, Marine Ecology Progress Series, vol.346, pp.97-107 Cambridge, M.L and McComb, A.J, 1984; The loss of seagrass in Cockburn Sound, Western Australia. I. The time course and magnitude of seagrass decline in relation to industrial development, Aquatic Botany, vol.20, pp.229-243
Campbell, M.L, 2003; Recruitment and colonisation of vegetative fragments of Posidonia australis and Posidonia coriacea, Aquatic Botany, vol.76, pp.175-184 Campbell, M.L and Paling, E.I, 2003; Evaluating vegetative transplant success in Posidonia australis: A field trial with habitat enhancement, Marine Pollution Bulletin, vol.46, pp.828-834 Cardoso, P.G; Bankovic, M; Raffaelli, D and Pardal, M.A, 2007; Polychaete assemblages as indicators of habitat recovery in a temperate estuary under eutrophication, Estuarine, Coastal and Shelf Science, vol.71, pp.301-308 Cardoso, I; Granadeiro, J.P and Cabral, H, 2010; Benthic macroinvertebrates’ vertical distribution in the Tagus estuary (Portugal): The influence of tidal cycle, Estuarine, Coastal and Shelf Science, vol.86, pp.580-586 Carruthers, T.J.B; Dennison, W.C; Kendrick, G.A; Waycott, M; Walker, D.I and Cambridge, M.L, 2007; Seagrasses of south–west Australia: A conceptual synthesis of the world's most diverse and extensive seagrass meadows, Journal of Experimental Marine Biology and Ecology, vol.350, pp.21-45 Chisholm, W, unpublished; Sediment movement and stability, a study of natural and transplanted seagrasses, PhD dissertation, Murdoch University Christie, N.D, 1976; The efficiency and effectiveness of a diver operated suction sampler on a homogeneous macrofauna, Estuarine, Coastal and Shelf Science, vol.4, pp.687-693 Clarke, KR, 1993; Non-parametric multivariate analyses of changes in community structure, Australian Journal of Ecology, vol.18, pp.117-143. Cockburn Sound Management Council, 2003, The influence of the Garden Island causeway on Mangles Bay, Cockburn Sound, Community Summary Paper
62
Cornelisen, C.D and Thomas, F.I.M, 2004; Ammonium and nitrate uptake by leaves of the seagrass Thalassia testudinum: impact of hydrodynamic regime and epiphyte cover on uptake rates, Journal of Marine Systems, vol.49, pp.177-194 Cornelisen, C.D and Thomas, F.I.M, 2006; Water flow enhances ammonium and nitrate uptake in a seagrass community, Marine Ecology Progress Series, vol.312, pp.1-13 de Boer, W.F, 2007; Seagrass–sediment interactions, positive feedbacks and critical thresholds for occurrence: a review, Hydrobiologia, vol.591, pp.5-24 Duarte, C.M, 1990; Seagrass nutrient content, Marine Ecology Progress Series, vol.67, pp.201-207 Duarte, C.M; Middlburg, J.J and Caraco, N, 2005; Major role of marine vegetation on the oceanic carbon cycle, Biogeosciences, vol.2, pp.1-8 Duffy, J.E, 2006; Biodiversity and the functioning of seagrass ecosystems, Marine Ecology Progress Series, vol.311, pp.233-250 Environmental Protection Authority, 1996; Short-term shell-sand dredging, Success Bank, Owen Anchorage; Bulletin 833 November 1996 Evans, N.T and Short, F.T, 2005; Functional trajectory models for assessment of transplanted eelgrass, Zostera marina L., in the Great Bay Estuary, New Hampshire, Estuaries, vol.28, pp.936-947 Filgueiras, V.L; Campos, L.S; Lavrado, H.P; Frensel, R and Pollery, R.C.G, 2007, Vertical distribution of macrobenthic infauna from the shallow sublittoral zone of Admiralty Bay, King George Island, Antarctica, Polar Biology, vol.30, pp.1439-1447 Fonseca, M.S and Fisher, J.S, 1986; A comparison of canopy friction and sediment movement between four species of seagrass with reference to their ecology and restoration, Marine Ecology Progress Series, vol.29, pp.15-22 Fonseca, M.S; Kenworthy, W.J and Thayer, G.W, 1982; A low cost transplanting procedure for sediment stabilisation and habitat development using Eelgrass (Zostera marina), wetlands, vol.2, pp.138-151 Fonseca, M.S and Koehl, M.A.R, 2006; Flow in seagrass canopies: The influence of patch width, Estuarine, Coastal and Shelf Science, vol.67, pp.1-9 Fonseca, M.S; Meyer, D.L and Hall, M.O, 1996; Development of planted seagrass beds in Tampa Bay, Florida, USA. II. Faunal components, Marine Ecology Progress Series, vol.132, pp.141-156 Fourqurean, J.W and Zieman, J.C, 1991; Photosynthesis, respiration and whole plant carbon budget of the seagrass Thalassia testudinum, Marine Ecology Progress Series, vol.69, pp.161-170 Gacia, E and Duarte, C.M, 2001; Sediment Retention by a Mediterranean Posidonia oceanica Meadow: The Balance between Deposition and Resuspension, Estuarine, Coastal and Shelf Science, vol.52, pp.505-514
63
Gacia, E; Duarte, C.M and Middleburg, J.J, 2002; Carbon and nutrient deposition in a Mediterranean seagrass (Posidonia oceanica) meadow, The American Society of Limnology and Oceanography, vol.47, pp.23-32 Gacia, E, Granata, T.C and Duarte, C.M, 1999; An approach to the measurement of particle flux and sediment retention within seagrass (Posidonia oceanica) meadows. Aquatic Botany, vol.65, pp.255-268. Gambi, M.C; Nowell, A.R.M and Jumars, P.A, 1990; Flume observations on flow dynamics in Zostera marina (eelgrass) beds, Marine Ecology Progress Series, vol.61, pp.159-169 Gillanders, B.M, 1997; Patterns of abundance and size structure in the blue groper, Achoerodus viridis (Pisces, Labridae): evidence of links between estuaries and coastal reefs, Environmental Biology of Fishes, vol.49, pp.153-173 Grol, M.G.G; Dorenbosch, M; Kokkelmans, E.M.G; Nagelkerken, I, 2008; Mangroves and seagrass beds do not enhance growth of early juveniles of a coral reef fish, Marine Ecology Progress Series, vol.366, pp.137-146 Hastings, K; Hesp, P and Kendrick, G.A, 1995; Seagrass loss associated with boat moorings at Rottnest Island, Western Australia, Ocean and Coastal Management, vol.26, pp.225-246 Hines, A.H and Comtois, K.L, 1985; Vertical distribution of infauna in sediments of a subestuary in Central Chesapeake Bay, Estuaries, vol.8, pp.296-304 Horn, L.E; Paling, E.I and van Keulen, M, 2009; Photosynthetic recovery of transplanted Posidonia sinuosa, Western Australia, Aquatic Botany, vol.90, pp.149-156 Hovel, K.A, 2003; Habitat fragmentation in marine landscapes: relative effects of habitat cover and configuration on juvenile crab survival in California and North Carolina seagrass beds, Biological Conservation, vol.110, pp.401-412 Hovel, K.A and Lipcius, R.N, 2001; Habitat fragmentation in a seagrass landscape: patch size and complexity control blue crab survival, Ecology, vol.82, pp.1814-1829
Hyndes, G.A; Kendrick, A.J; MacArthur, L.D and Stewart, E, 2003; Differences in the species- and size-composition of fish assemblages in three distinct seagrass habitats with differing plant and meadow structure, Marine Biology, vol.142, pp.1195-1206
Hyndes, G.A and Lavery, P.S, 2005; Does transported seagrass provide an important trophic link in unvegetated, nearshore areas?, Estuarine, Coastal and Shelf Science, vol.63, pp.633-643 Jenkins, G.P; Syme, A and Macreadie, P.I, 2011; Feeding ecology of King George whiting Sillaginodes punctatus (Perciformes) recruits in seagrass and unvegetated habitats. Does diet reflect habitat utilization?, Journal of Fish Biology, vol.78, pp.1561-1573 Jernakoff, P and Nielsen, J, 1998, Plant–animal associations in two species of seagrasses in Western Australia, Aquatic Botany, vol.60, pp.359-376
64
Kaldy, J.E; Onuf, C.P; Eldridge, P.M and Cifuentes, L.A, 2002; Carbon Budget for a Subtropical Seagrass Dominated Coastal Lagoon How Important are Seagrasses to Total Ecosystem Net Primary Production?, Estuaries, vol.25, pp.528-539 Kenna, R, Hyndes, G and Lavery, P, 2006; Evaluation of the return of ecological function in transplanted Posidonia australis in Oyster Harbour over 36 months, Report for the Seagrass Research and Revegetation Plan, September 2006, Report No. 2006-08 Kendrick, A.J and Hyndes, G.A, 2005; Variations in the dietary compositions of morphologically diverse syngnathid fishes, Environmental Biology of Fish, vol.72, pp.414-427 Kendrick, G.A; Aylward, M.J; Hegge, B.J; Cambridge, M.L; Hillman, K; Wyllie, A and Lord, D.A, 2002; Changes in seagrass coverage in Cockburn Sound, Western Australia between 1967 and 1999, Aquatic Botany, vol.73, pp.75-87 Kendrick, G.A; Eckersley, J and Walker, D.I, 1999; Landscape-scale changes in seagrass distribution over time: a case study from Success Bank, Western Australia, Aquatic Botany, vol.65, pp.293-309 Kendrick G.A; Hegge, B.J; Wyllie, A; Davidson, A and Lord, D.A, 2000; Changes in seagrass cover on Success and Parmelia Banks, Western Australia between 1965 and 1995, Estuarine, Coastal and Shelf Science, vol.50, pp.341-353 Kennedy, H; Beggins, J; Duarte, C.M; Fourqurean, J.W; Holmer, M; Marba, M and Middleburg, J.J, In Press 2010; Seagrass sediments as a global carbon sink: Isotopic constraints, Global Biogeochemical Cycles, vol.24 Kharlamenko, V.I; Kiyashko, S.I; Imbs, A.B and Vyshkvartzev, D.I, 2001; Identification of food sources of invertebrates from the seagrass Zostera marina community using carbon and sulfur stable isotope ratio and fatty acid analyses, Marine Ecology Progress Series, vol.220, pp.103-117 Kirkman, H, 1998; Pilot experiments on planting seedlings and small seagrass propagules in Western Australia, Marine Pollution Bulletin, vol.37, pp.460-467 Kirkman, H and Walker, D.I, 1989; ‘Regional studies- Western Australian seagrasses’ In: Biology of seagrasses: A treatise on the biology of seagrasses with special reference to the Australian region, eds. A.W.D Larkum, A.J. McComb & S.A. Shepherd, Else Koch, E.W and Gust, G, 1999; Water flow in tide- and wave-dominated beds of the seagrass Thalassia testudinum, Marine Ecology Progress Series, vol.184, pp.63-72 Leduc, D; Probert, P.K; Frew, R.D and Hurd, C.L, 2006; Macroinvertebrate diet in intertidal seagrass and sandflat communities: a study using C, N, and S stable isotopes, New Zealand Journal of Marine and Freshwater Research, vol.40, pp.615-629 Lefebvre, A; Thompson, C.E.L and Amos, C.L, 2010; Influence of Zostera marina canopies on unidirectional flow, hydraulic roughness and sediment movement, Continental Shelf Research, vol.30, pp.1783-1794 Lewis, F.G and Stoner, A.W, 1981; An examination of methods for sampling macrobenthos in seagrass meadows, Bulletin of Marine Science, vol.31, issue.1, pp.116-124
65
Lie, U and Pamatmat, M.M, 1965; Digging characteristics and sampling efficiency of the 0.1 m super(2) van Veen grab, Limnology and Oceanography, vol.10, issue.3, pp.379-384 Lothian, A, 1999; Application of environmental evaluation in South Australia. Report of the Environmental Valuation Working Group to the Natural Resources Council. Department of Environment, Heritage and Aboriginal Affairs. Adelaide Macinnis-Ng, C.M.O and Ralph, P.J, 2002; Towards a more ecologically relevant assessment of the impact of heavy metals on the photosynthesis of the seagrass, Zostera capricorni, Marine Pollution Bulletin, vol.45, pp.100-106 Macinnis-Ng, C.M.O and Ralph, P.J, 2003; In situ impact of petrochemicals on the photosynthesis of the seagrass Zostera capricorni, Marine Pollution Bulletin, vol.46, pp.1395-1407 Macreadie, P.I; Connolly, R.M; Jenkins, G.P; Hindell, J.S and Keough, M.J, 2010; Edge patterns in aquatic invertebrates explained by predictive models, Marine and Freshwater Research, vol.61, pp.214-218 Mateo, M.A and Romero, J, 1997; Detritus dynamics in the seagrass Posidonia oceanica: elements for an ecosystem carbon and nutrient budget, Marine Ecology Progress Series, vol.151, pp.43-53 Mateo, M.A; Romero, J; Perez, M; Littler, M.M and Littler, D.S, 1997; Dynamics of millenary organic deposits resulting from the growth of the Mediterranean seagrass Posidonia oceanica, Estuarine, Coastal and Shelf Science, vol.44, pp.103-110 Mellors, J; Marsh, H; Carruthers, T.J.B and Waycott, M, 2002; Testing the sediment-trapping paradigm of seagrass: do seagrasses influence nutrient status and sediment structure in tropical intertidal environments?, Bulletin of Marine Science, vol.71, issue.3, pp.1215-1226 Morris, E.P; Peralta, G; Burn, F.G; van Duren, L; Bouma, T.J and Perez-Llorens, J.L, 2008; Interaction between hydrodynamics and seagrass canopy structure: Spatially explicit effects on ammonium uptake rates, American Society of Limnology and Oceanography, vol.53, issue.4, pp.1531-1539 Murphy, H.M; Jenkins, G.P; Hindell, J.S and Connolly, R.M, 2010; Response of fauna in seagrass to habitat edges, patch attributes and hydrodynamics, Austral Ecology, vol.35, pp.535-543 Nichols, P.D; Klumpp, D.W and Johns, R.B, 1985; A study of food chains in seagrass communities III. Stable carbon isotope ratios, Australian Journal of Marine and Freshwater Research, vol.36, pp.683-690 Nyunja, J; Ntiba, M; Onyari, J; Mavuti, K, Soetaert, K and Bouillon, S, 2009; Carbon sources supporting a diverse fish community in a tropical coastal ecosystem (Gazi Bay, Kenya), Estuarine, Coastal and Shelf Science, vol.83, pp.333-341 Oceanica Consulting Pty Ltd., 2009a; Perth Metropolitan Desalination Plant - Cockburn Sound Benthic Macrofauna Community and Sediment Habitat; Repeat Macrobenthic Survey, Report No. 604_001/1; June 2009
66
Oceanica Consulting Pty Ltd., 2009b; Long-term shellsand dredging, Owen Anchorage - Dredging and Environmental Management Plan, Stage 2 – West Success Bank, Report No. 334_575/1; October 2009 Oceanica Consulting Pty Ltd., 2011; Seagrass Research and Rehabilitation Plan, Synthesis Report July 2003 to June 2010, Report No. 520_001/3 August 2011 Olesen, B and Sand-Jensen, K, 1994; Patch dynamics of eelgrass Zostera marina, Marine Ecology Progress Series, vol.106, pp.147-156 Orth, R.J; Carruthers, T.J.B; Denisson, W.C; Duarte, C.M; Fourqurean, J.W; Heck. K.L Jr; Hughes, A.R; Kendrick, G.A; Kenworthy, W.J; Olyarnik, S; Short, F.T; Waycott, M and Williams, S.L, 2006, A global crisis for seagrass ecosystems, Bioscience, vol.56, issue.12, pp.987-996 Paling, E.I and McComb, A.J, 2000; Autumn biomass, below-ground productivity, rhizome growth at bed edge and nitrogen content in seagrasses from Western Australia, Aquatic Botany, vol.67, pp.207-219 Paling, E.I; van Keulen, M and Tunbridge, D.J, 2007; Seagrass transplanting in Cockburn Sound, Western Australia: A comparison of manual transplantation methodology using Posidonia sinuosa Cambridge et Kuo, Restoration Ecology, vol.15, pp.240-249 Paling, E.I; van Keulen, M; Wheeler, K.D; Phillips, J and Dyhrberg, R, 2001a; Mechanical seagrass transplantation in Western Australia, Restoration Ecology, vol.16, pp.331-339 Paling, E.I; van Keulen, M; Wheeler, K.D; Phillips, J and Dyhrberg, R, 2003; Influence of spacing on mechanically transplanted seagrass survival in a high wave energy regime, Restoration Ecology, vol.11, pp.56-61 Paling, E.I; van Keulen, M; Wheeler, K.D; Phillips, J; Dyhrberg, R and Lord, D.A, 2001b; Improving mechanical seagrass transplantation, Ecological Engineering, vol.18, pp.107-113 Paling, E.I; Van Keulen, M; Wheeler, K and Walker, C, 2000; Effects of depth on manual transplantation of the seagrass Amphibolis griffithii (J. M. Black) Den Hartog on Success Bank, Western Australia, Pacific Conservation Biology, vol.5, pp.314-320 Paterson, C.G and Fernando, C.H, 1971; Comparison of a simple core and an Ekman grab for sampling shallow water benthos, Journal of the Fisheries Research Board of Canada, vol.28, pp.365-368 Peduzzi, P and Herndl, G.J, 1991; Decomposition and significance of seagrass leaf litter (Cymodocea nodosa) for the microbial food web in coastal waters (Gulf of Trieste, Northern Adriatic Sea), Marine Ecology Progress Series, vol.71, pp.163-174 Peterson, B.J and Heck Jr., K.L, 1999; The potential for suspension feeding bivalves to increase seagrass productivity, Marine Ecology Progress Series, vol.240, pp.37-52 Peterson, B.J and Heck Jr., K.L, 2001a; An experimental test of the mechanism by which suspension feeding bivalves elevate seagrass productivity, Marine EcologyProgress Series, vol.218, pp.115-125
67
Peterson, B.J and Heck Jr., K.L, 2001b; Positive interactions between suspension-feeding bivalves and seagrass—a facultative mutualism, Marine Ecology Progress Series, vol.213, pp.143-155 Peterson, C.H; Luettich, R.A Jr; Micheli, F and Skilleter, G.A, 2004; Attenuation of water flow inside seagrass canopies of differing structure, Marine Ecology Progress Series, vol.268, pp.81-92 Pollard, P.C and Moriarty, D.J.W, 1991; Organic carbon decomposition, primary and bacterial productivity, and sulphate reduction, in tropical seagrass beds of the Gulf of Carpentaria, Australia, Marine Ecology Progress Series, vol.69, pp.149-159 Pranovi, F; Curiel, D; Rismondo, A; Marzocchi, M and Scattolin, M, 2000; Variations of the macrobenthic community in a seagrass transplanted area of the Lagoon of Venice, Scientia Marina, vol.64, pp.303-310 Ralph, P.J and Burchett, M.D, 1998(a); Photosynthetic response of Halophila ovalis to heavy metal stress, Environmental Pollution, vol.103, pp.91-101 Ralph, P.J and Burchett, M.D, 1998(b); Impact of Petrochemicals on the Photosynthesis of Halophila ovalis Using Chlorophyll Fluorescence, Marine Pollution Bulletin, vol.36, pp.429-436 Ramey, P.A; Grassle, J.P; Grassle, J.F and Petrecca, R.F, 2009; Small-scale, patchy distributions of infauna in hydrodynamically mobile continental shelf sands: Do ripple crests and troughs support different communities?, Continental Shelf Research, vol.29, pp.2222-2233 Rooker, J.R; Holt, G.J and Holt, S.A, 1998; Vulnerability of newly settled red drum (Sciaenops ocellatus) to predatory fish: is early-life survival enhanced by seagrass meadows?, Marine Biology, vol.131, pp.145-151 Saether, O.A, 1979; Chironomid communities as water quality indicators, Holarctic Ecology, vol.2, pp.65-74 Schanz, A and Asmus, H, 2003; Impact of hydrodynamics on development and morphology of intertidal seagrasses in the Wadden Sea, Marine Ecology Progress Series, vol.261, pp.123-134 Scoffin, T.P, 1968; An underwater flume, Journal of Sedimentary Petrology, vol.38, pp.244-246 Sheridan, P, 1998; Trajectory for structural equivalence of restored and natural Halodule wrightii beds in Texas, Gulf Research Reports, vol.10, pp.81-82 Sheridan, P, 2004; Comparison of restored and natural seagrass beds near Corpus Christi, Texas, Estuaries, vol.27, pp.781-792 Sheridan, P; Henderson, C and McMahan, G, 2003; Fauna of natural seagrass and transplanted Halodule wrightii (Shoalgrass) beds in Galveston Bay, Texas, Restoration Ecology, vol.11, pp.139-154 Short, F.T and Coles, R.G, 2001; Global Seagrass Research Methods, Chapter 12, Elsevier Science B.V
68
Short, F.T and Wyllie–Echeverria, S, 1996; Natural and human induced disturbance of seagrass, Environmental Conservation, vol.23, pp.17-27 Short, F.T; Polidoro, B; Livingstone, S.R; Carpenter, K.E; Bandeira, S; Bujang, J.S; Calumpong, H.P; Carruthers, T.J.B; Coles, R.G; Dennison, W.C; Erftemeijer, P.L.A; Fortes, M.D; Freeman, A.S; Jagtap, T.G; Kamal, A.H.M; Kendrick, G.A; Kenworthy, W.J; La Nafie, Y.A; Nasution, I.M; Orth, R.J; Prathep, A; Sanciangco, J.C; van Tussenbroek, B; Vergara, S.G; Waycott, M; Zieman, J.C, 2011; Extinction risk assessment of the world’s seagrass species, Biological Conseration, vol.144, pp.1961-1971 Silberstein, K; Chiffings, A.W and McComb, A.J, 1986; The loss of seagrass in Cockburn Sound, Western Australia. III. The effect of epiphytes on productivity of Posidonia australis Hook. f, Aquatic Botany, vol.24, pp.355-371 Smit, A.J; Brearley, A; Hyndes, G.A; Lavery, P.S and Walker, D.I, 2005; Carbon and nitrogen stable isotope analysis of an Amphibolis griffithii seagrass bed, Estuarine, Coastal and Shelf Science, vol.65, pp.545-556 Smith, K.A and Sinerchia, M, 2004; Timing of recruitment events, residence periods and post-settlement growth of juvenile fish in a seagrass nursery area, south-eastern Australia, Environmental Biology of Fishes, vol.71, pp.73-84 Smith, T.M; Hindell, J.S; Jenkins, G.P; Connolly, R.M and Keough, M.J, 2011; Fine-scale spatial and temporal variations in diets of the pipefish Stigmatopora nigra within seagrass patches, Journal of Fish Biology, vol.78, pp.1824-1832 Southwood, T.R.E and Henderson, P.A, 2000; Ecological Methods, Chapter 5, 3rd Ed., Blackwell Science Ltd Stoner, A.W; Greening, H.S; Ryan, J.D and Livingston, R.J, 1983; Comparison of Macrobenthos Collected with Cores and Suction Sampler in Vegetated and Unvegetated Marine Habitats, Estuaries,vol.6, pp.76-82 Talbot, V and Chegwidden, A, 1982; Cadmium and other Heavy Metal Concentrations in Selected Biota from Cockburn Sound, Western Australia, Australian Journal for Marine and Freshwater Research, vol.33, pp.779-788 Tanner, J.E, 2005; Edge effects on fauna in fragmented seagrass meadows, Austral Ecology, vol.30, pp.210-218 Thresher, R.E; Nichols, P.O; Gunn, J.S; Bruce, B.D; Furlani, D.M, 1992; Seagrass detritus as the basis of a coastal planktonic food chain, Limnology and Oceanography, vol.37, pp.1754-1758 Trautman, D.A and Borowitzka, M.A, 1999, Distribution of the epiphytic organisms on Posidonia australis and P. sinuosa, two seagrasses with differing leaf morphology, Marine Ecology Progress Series, vol.179, pp.215-229 Uhrin, A.V; Hall, M.O; Merello, M.F and Fonseca, M.S, 2009; Survival and expansion of mechanically transplanted seagrass sods, Restoration Ecology, vol.17, pp.359-368 Valentine, J.F and Heck Jr, K.L, 1999; Seagrass herbivory: Evidence for the continued grazing of marine grasses, Marine Ecology Progress Series, vol.176, pp.291-302
69
Vance, D.J; Haywood, M.D.E; Heales, D.S; Kenyon, R.A and Loneragan, N.R, 1998; Seasonal and annual variation in abundance of postlarval and juvenile banana prawns Penaeus merguiensis and environmental variation in two estuaries in tropical northeastern Australia: a six year study, Marine Ecology Progress Series, vol.163, pp.21-36 van Katwijk, M.M, Bos, A.R; Hermus, D.C.R and Suykerbuyk, W, 2010; Sediment modification by seagrass beds: Muddification and sandification induced by plant cover and environmental conditions, Estuarine, Coastal and Shelf Science, vol.89, pp.175-181 Van Keulen, M and Borowitzka, M, 2002; Comparison of water velocity profiles through morphologically dissimilar seagrass measured with a simple and inexpensive current meter, Bulletin of Marine Science, vol.71, issue.3, pp.1257-1267 Van Keulen, M and Borowitzka, M, 2003; Seasonal variability in sediment distribution along an exposure gradient in a seagrass meadow in Shoalwater Bay, Western Australia, Estuarine, Coastal and Shelf Science, vol.57, pp.587-592 Van Keulen, M; Paling, E.I and Walker, C.J, 2003; Effect of planting unit size and sediment stabilization on seagrass transplants in Western Australia, Restoration Ecology, vol.11, pp.50-55 Verduin, J.J and Backhaus, J.O, 2000; Dynamics of plant–flow interactions for the seagrass Amphibolis antarctica: Field Observations and Model Simulations, Estuarine, Coastal and Shelf Science, vol.50, pp.185-204 Verduin, J. J., Horn, L. E., Paling, E. I. & Van Keulen, M. 2007, Seagrass rehabilitation studies for the seagrass research and rehabilitation plan, project 3 - Annual Report 2007, Prepared by Marine and Freshwater Research Laboratory, Perth, Western Australia, June 2007. Verweij, M.C; Nagelkerken, I; de Graaff, D; Peeters, M; E. J. Bakker, E.J and van der Velde, G, 2006; Structure, food and shade attract juvenile coral reef fish to mangrove and seagrass habitats: a field experiment, Marine Ecology Progress Series, vol.306, pp.257-268 Verweij, M.C; Nagelkerken, I; Hans, I; Ruseler, S.M and Mason, P.R.D, 2008; Seagrass nurseries contribute to coral reef fish populations, Limnology and Oceanography, vol.53, pp.1540-1547 Vidondo, B; Duarte, C.M; Middleboe, A.L; Stefansen, F; Lutzen, T and Nielsen, S.L, 1997; Dynamics of a landscape mosaic: size and age distributions, growth and demography of seagrass Cymodocea nodosa patches, Marine Ecology Progress Series, vol.158, pp.131-158 Vizzini, S; Sarà, G; Michener, R.H and Mazzola, A, 2002; The role and contribution of the seagrass Posidonia oceanica (L.) Delile organic matter for secondary consumers as revealed by carbon and nitrogen stable isotope analysis, Acta Oecologica, vol.23, pp.277-285 Wahle, R.A and Steneck, R.S, 1992; Habitat restrictions in early benthic life: experiments on habitat selection and in situ predation with the American lobster, Journal of Experimental Marine Biology and Ecology, vol.157, pp.91-114 Walker, D.I; Kendrick, G.A; McComb, A.J, 2006; Decline and recovery of seagrass ecosystems-The dynamics of change. In. Larkum, A.W.D; Orth, R.J; Duarte, C.M (eds.) Seagrass: Biology, Ecology and Conservation, Springer, The Netherlands, pp.551-565
70
Walker, D.I; Lukatelich, R.J; Bastyan, G and McComb, G, 1989; Effects of boat moorings on seagrass beds near Perth, Western Australia, Aquatic Botany, vol.36, pp.69-77 Warry, F.Y; Hindell, J.S; Macreadie, P.I; Jenkins, G.P and Connolly, R.M, 2009; Integrating edge effects into studies of habitat fragmentation: a test using meiofauna in seagrass, Oecologia, vol.159, pp.883-892 Weston, D.P, 1990; Quantitative examination of macrobenthic community changes along an organic enrichment gradient, Marine Ecology Progress Series, vol.61, pp.233-244 Worthington, D.G; Westoby, M and Bell, J.D, 1991; Fish larvae settling in seagrass: Effects of leaf density and an epiphytic alga, Australian Journal of Ecology, vol.16, pp.289-293
71
Appendix 1 References used for infauna identification
Books
Edgar, G.J, 2008; Australian Marine Life: The plants and animals of temperate waters, 2nd Ed.,
Sydney, New Holland Publishers
Jones, D and Morgan, G, 2002, A Field Guide to Crustaceans of Australian Waters, 2nd Ed., New
Holland Publishers
Wells, F.E and Bryce, C.W, 1993; Seaslugs of Western Australia, Western Australian Museum
Wilson, B.R, 2002; A Handbook to Australian Seashells: On Seashores East to West and North to
South, New Holland Publishers
Journals
Clark, W.C, 1963; Australian Pycnogonida, Records of the Australian Museum, vol.26, pp.1–82.
King, P.E, 1986; Sea Spiders: A revised key to the adults in littoral Pycnogonida in the British Isles,
Field Studies, vol.6, pp.493-516
Software
IntKey for Windows, Version 5.11
Dallwitz, M.J, 1980; A general system for coding taxonomic descriptions, Taxon, vol.29, pp.41–46
Dallwitz, M.J; Paine, T.A and Zurcher, E.J, 1993; User’s guide to the DELTA System: a general
system for processing taxonomic descriptions, 4th Ed., http://delta-intkey.com
Dallwitz, M.J; Paine, T.A and Zurcher, E.J, 1995; User’s guide to Intkey: a program for interactive
identification and information retrieval, http://delta-intkey.com
Dallwitz, M.J; Paine, T.A and Zurcher, E.J, 2000; Principles of interactive keys. http://delta-
intkey.com
Web Sites
http://home.comcast.net/~fireflea2/OstracodeKeyindex.html
http://www.marinespecies.org/cumacea/KeyStart.php
72
Appendix 2 Mean abundances of taxa from different sites and sampling methods in Southern Flats,
Cockburn Sound
Taxa
Corer Dredge
Bare Sand Natural Meadow 1 Bare Sand Natural Meadow 1
Amphipoda N Mean SE N Mean SE N Mean SE N Mean SE
Ampithoidae 12 0 0
11 857 575
11 0 0
10 373 373
Aoridae 12 497 335
11 10690 1639
11 0 0
10 11884 2934
Caprellidae 12 741 529
11 2875 1244
11 0 0
10 11282 3320
Ceradocopsis Group 12 0 0
11 0 0
11 0 0
10 3338 1510
Cyproideidae 12 0 0
11 3008 2140
11 239 239
10 7786 3419
Dexaminidae 12 271 271
11 1826 977
11 0 0
10 1095 1095
Eusiridae 12 0 0
11 1285 668
11 0 0
10 1070 546
Isaeidae 12 0 0
11 0 0
11 0 0
10 917 917
Ischyroceridae 12 303 303
11 6290 2879
11 201 201
10 1255 1255
Leucothoidae 12 287 287
11 0 0
11 0 0
10 365 365
Lysianassidae 12 303 303
11 0 0
11 0 0
10 0 0
Phoxochephalidae 12 265 265
11 2150 927
11 834 456
10 2311 890
Platyscelidae 12 0 0
11 411 411
11 0 0
10 365 365
Sebidae 12 0 0
11 0 0
11 0 0
10 365 365
Stenothoidae 12 0 0
11 568 568
11 293 293
10 459 459
Thoriella Group 12 0 0
11 568 568
11 0 0
10 0 0
Unidentified 12 0 0
11 0 0
11 0 0
10 1210 1210
Cirripedia
Balanidae 12 0 0
11 352 352
11 0 0
10 365 365
Bivalves
Pectinidae 12 0 0
11 478 478
11 0 0
10 0 0
Solemyidae 12 1101 472
11 831 565
11 1688 657
10 2924 1317
Solecurtidae 12 0 0
11 1153 801
11 0 0
10 0 0
Tellinidae 12 38419 7194
11 61845 8986
11 68160 15914
10 117404 26824
Veneridae 12 3626 977
11 11556 2544
11 8812 1857
10 6088 2221
Copepoda
Epacteriscidae 12 0 0
11 0 0
11 0 0
10 313 313
Peltitiidae 12 0 0
11 411 411
11 0 0
10 9910 3088
Order: Harpacticoid 12 298 298
11 1005 681
11 0 0
10 6034 2258
Cumacean
Diastylidae 12 0 0
11 0 0
11 0 0
10 1094 557
Gynodiastylidae 12 0 0
11 916 615
11 0 0
10 1457 971
Nannastacidae 12 0 0
11 3005 1573
11 587 587
10 5873 2261
Decapoda
Diogenidae 12 0 0
11 903 607
11 0 0
10 187 187
73
Majidae 12 0 0
11 0 0
11 0 0
10 187 187
Pilumnidae 12 0 0
11 957 957
11 0 0
10 0 0
Portunidae 12 0 0
11 437 437
11 0 0
10 738 492
Echinodermata
Diadematidae 12 0 0
11 0 0
11 293 293
10 0 0
Ophiodermatidae 12 298 298
11 0 0
11 0 0
10 418 418
Gastropoda
Aplysiidae 12 0 0
11 0 0
11 293 293
10 0 0
Batillariidae 12 278 278
11 8596 7604
11 310 310
10 6055 2352
Buccinidae 12 581 392
11 0 0
11 0 0
10 364 364
Bullidae 12 4016 1561
11 0 0
11 6709 2722
10 2475 582
Columbellidae 12 0 0
11 2746 1462
11 0 0
10 1455 1455
Epitoniidae 12 303 303
11 382 382
11 768 409
10 1971 663
Fissurellidae 12 0 0
11 1007 681
11 0 0
10 187 187
Hydatinidae 12 0 0
11 0 0
11 0 0
10 459 459
Mitridae 12 0 0
11 0 0
11 0 0
10 782 523
Naticidae 12 288 288
11 0 0
11 783 600
10 3306 1640
Olividae 12 0 0
11 0 0
11 0 0
10 418 418
Terebridae 12 0 0
11 0 0
11 0 0
10 2578 1514
Trochidae 12 850 445
11 1210 845
11 1655 860
10 2131 1810
Turbinidae 12 515 348
11 6213 3009
11 598 404
10 12713 7369
Insecta
Chironomidae 12 0 0
11 0 0
11 0 0
10 1296 909
Isopoda
Antheluridae 12 0 0
11 819 551
11 0 0
10 0 0
Holidoteidae 12 0 0
11 0 0
11 0 0
10 4449 2180
Paranthuridae 12 0 0
11 425 425
11 0 0
10 0 0
Mysidacea
Mysidae 12 0 0
11 0 0
11 0 0
10 1925 847
Nematoda 12 68457 15708 11 39067 8684 11 63206 14301 10 43267 10080
Ostracoda
Order: Halocypridina 12 0 0
11 0 0
11 239 239
10 0 0
Order: Podocopida 12 0 0
11 5811 2312
11 538 364
10 4900 1751
Cypridinoidae 12 0 0
11 382 382
11 0 0
10 1427 583
Rutidermatidae 12 6108 2378
11 11134 3524
11 7040 2735
10 12215 3097
Polychaeta
Apistobranchidae 12 271 271
11 0 0
11 0 0
10 560 560
Cirratulidae 12 0 0
11 1136 1136
11 0 0
10 0 0
Dorvilleidae Group 3 12 0 0
11 2027 1430
11 490 490
10 914 730
Eunicidae 12 0 0
11 1435 1435
11 0 0
10 0 0
Hartmaniellidae 12 0 0
11 0 0
11 201 201
10 0 0
Lumbrineridae 12 4592 1251
11 16345 4692
11 1611 602
10 1498 819
Magelonidae 12 271 271
11 0 0
11 239 239
10 0 0
Maldanidae 12 0 0
11 1337 1337
11 402 402
10 0 0
74
Nereididae 12 0 0
11 1145 791
11 0 0
10 0 0
Oenonidae 12 5101 1396
11 7447 4365
11 4003 1161
10 2390 1230
Onuphidae 12 1824 1097
11 18453 3083
11 579 389
10 2235 1122
Oweniidae 12 0 0
11 1459 1008
11 0 0
10 0 0
Paraonidae 12 559 377
11 1795 1005
11 473 331
10 1741 1384
Phyllodocidae 12 574 388
11 0 0
11 201 201
10 0 0
Poecilochaetidae 12 271 271
11 0 0
11 463 315
10 0 0
Sigalionidae 12 0 0
11 0 0
11 0 0
10 365 365
Spirorbidae 12 0 0
11 94198 61907
11 918 644
10 74367 41819
Syllidae 12 2300 994
11 20177 4425
11 174 174
10 3004 1546
Terebellidae 12 0 0
11 0 0
11 0 0
10 1866 1866
Polyplacophora
Ischnochitonidae 12 0 0
11 1309 986
11 0 0
10 0 0
Pycnogonida
Callipallenidae 12 244 244
11 0 0
11 0 0
10 0 0
Tanaidacae
Neotenaidae 12 0 0
11 561 561
11 0 0
10 0 0
Tanaidae 12 303 303 11 1705 969 11 0 0 10 1668 586
75
Appendix 3 Abundances of taxa from different sites with one-way ANOVA analyses
On
e-W
ay A
NO
VA
P-v
alu
e
0.0
44
< 0
.00
1
0.3
74
0.4
11
0.2
79
0.1
72
0.6
32
0.2
05
0.4
11
0.0
22
0.3
81
0.5
64
0.4
72
0.4
11
0.4
11
0.4
11
0.4
11
0.4
11
0.0
42
0.1
58
0.2
24
< 0
.00
1
F st
atis
tic
2.2
95
10
.32
3
1.0
94
1.0
33
1.2
76
1.5
57
0.7
24
1.4
57
1.0
33
2.6
68
1.0
82
0.8
12
0.9
39
1.0
33
1.0
33
1.0
33
1.0
33
1.0
33
2.3
23
1.6
06
1.4
07
5.7
37
Bar
e Sa
nd
SE
0.0
00
47
.29
6
75
.52
5
0.0
00
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
35
.07
5
35
.07
5
35
.07
5
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
59
.82
5
0.0
00
82
7.8
78
12
1.0
44
Mea
n
0.0
00
70
.15
1
10
5.2
26
0.0
00
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
35
.07
5
35
.07
5
35
.07
5
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
14
0.3
02
0.0
00
47
70
.26
4
45
5.9
81
N
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
Nat
ura
l Mea
do
w 1
SE
51
.33
7
17
4.5
11
10
4.0
90
0.0
00
11
8.5
57
59
.27
9
0.0
00
59
.27
9
0.0
00
24
7.9
80
0.0
00
0.0
00
64
.02
8
38
.26
4
38
.26
4
38
.26
4
0.0
00
38
.26
4
51
.33
7
51
.33
7
82
1.3
88
22
5.7
26
Mea
n
76
.52
8
88
0.0
75
22
9.5
85
0.0
00
19
1.3
21
11
4.7
92
0.0
00
11
4.7
92
0.0
00
53
5.6
98
0.0
00
0.0
00
15
3.0
57
38
.26
4
38
.26
4
38
.26
4
0.0
00
38
.26
4
76
.52
8
76
.52
8
50
89
.13
2
91
8.3
40
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Nat
ura
l Me
ado
w 2
SE
0.00
0
51.3
37
51.3
37
38.2
64
38.2
64
0.00
0
38.2
64
76.5
28
0.00
0
85.5
61
0.00
0
38.2
64
0.00
0
0.00
0
0.00
0
0.00
0
38.2
64
0.00
0
160.
527
38.2
64
1290
.926
392.
464
Me
an
0.00
0
76.5
28
76.5
28
38.2
64
38.2
64
0.00
0
38.2
64
76.5
28
0.00
0
153.
057
0.00
0
38.2
64
0.00
0
0.00
0
0.00
0
0.00
0
38.2
64
0.00
0
420.
906
38.2
64
8226
.792
2448
.906
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
1 m
Plo
t
SE
0.00
0
54.9
53
81.9
19
0.00
0
35.0
75
0.00
0
105.
226
0.00
0
0.00
0
47.2
96
47.2
96
0.00
0
70.1
51
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
107.
851
0.00
0
1240
.015
303.
578
Me
an
0.00
0
105.
226
210.
453
0.00
0
35.0
75
0.00
0
105.
226
0.00
0
0.00
0
70.1
51
70.1
51
0.00
0
70.1
51
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
140.
302
0.00
0
4138
.905
982.
113
N
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.5
m P
lot
SE
0.00
0
54.9
53
81.9
19
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
47.2
96
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
273.
335
0.00
0
711.
011
304.
314
Me
an
0.00
0
105.
226
210.
453
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
70.1
51
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
771.
660
0.00
0
5331
.471
1052
.264
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12 12
12
12
12
12
0.25
m P
lot
SE
0.00
0
109.
579
42.0
91
0.00
0
56.1
21
42.0
91
0.00
0
0.00
0
0.00
0
42.0
91
42.0
91
0.00
0
64.2
94
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
221.
393
0.00
0
1444
.496
236.
857
Mea
n
0.00
0
294.
634
42.0
91
0.00
0
84.1
81
42.0
91
0.00
0
0.00
0
0.00
0
42.0
91
42.0
91
0.00
0
126.
272
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
462.
996
0.00
0
4629
.962
631.
358
N 10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10 10
10
10
10
10
0.12
5 m
Plo
t
SE
0.00
0
85.5
61
51.3
37
0.00
0
76.5
28
38.2
64
0.00
0
38.2
64
38.2
64
85.5
61
0.00
0
0.00
0
76.5
28
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
144.
190
0.00
0
1380
.694
232.
752
Mea
n
0.00
0
153.
057
76.5
28
0.00
0
76.5
28
38.2
64
0.00
0
38.2
64
38.2
64
153.
057
0.00
0
0.00
0
76.5
28
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
382.
641
0.00
0
5127
.396
765.
283
N 11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Taxa
Am
ph
ipo
da
Am
pit
hoi
dae
Ao
rid
ae
Cap
relli
dae
Cer
ado
cop
sis
Gro
up
Cyp
roid
eid
ae
Dex
amin
idae
Exo
edic
ero
tid
ae
Eusi
rid
ae
Hya
lidae
Isch
yroc
erid
ae
Leu
coth
oid
ae
Lysi
anas
sid
ae
Ph
oxo
chep
hal
idae
Pla
tysc
elid
ae
Sten
oth
oid
ae
Tho
riel
la G
rou
p
Uro
hau
sto
riid
ae
Biv
alve
s
Pec
tin
idae
Sole
myi
dae
Sole
curt
idae
Telli
nid
ae
Ven
erid
ae
76
On
e-W
ay A
NO
VA
P-v
alu
e
0.2
47
0.8
14
0.4
11
0.0
44
0.4
11
0.0
16
0.2
78
0.5
00
0.4
11
0.4
11
0.5
64
0.0
44
< 0
.00
1
0.3
74
0.4
11
F st
atis
tic
1.3
50
0.4
90
1.0
33
2.2
95
1.0
33
2.8
13
1.2
78
0.8
99
1.0
33
1.0
33
0.8
12
2.2
95
10
.32
3
1.0
94
1.0
33
Bar
e Sa
nd
SE
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
35
.07
5
47
.29
6
19
2.6
98
Mea
n
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
35
.07
5
70
.15
1
49
1.0
57
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
Nat
ura
l Mea
do
w 1
SE
38
.26
4
51
.33
7
38
.26
4
51
.33
7
0.0
00
15
3.0
57
51
.33
7
76
.52
8
38
.26
4
0.0
00
0.0
00
0.0
00
76
1.4
47
0.0
00
0.0
00
Mea
n
38
.26
4
76
.52
8
38
.26
4
76
.52
8
0.0
00
26
7.8
49
76
.52
8
76
.52
8
38
.26
4
0.0
00
0.0
00
0.0
00
84
1.8
11
0.0
00
0.0
00
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Nat
ura
l Mea
do
w 2
SE
0.0
00
76
.52
8
0.0
00
0.0
00
0.0
00
0.0
00
11
7.3
16
0.0
00
0.0
00
38
.26
4
0.0
00
76
.52
8
19
48
.84
4
0.0
00
38
.26
4
Mea
n
0.0
00
76
.52
8
0.0
00
0.0
00
0.0
00
0.0
00
15
3.0
57
0.0
00
0.0
00
38
.26
4
0.0
00
76
.52
8
30
61
.13
2
0.0
00
38
.26
4
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
1 m
Plo
t
SE
0.00
0
105.
226
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
96.3
49
0.00
0
81.2
33
Me
an
0.00
0
105.
226
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
175.
377
0.00
0
175.
377
N
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.5
m P
lot
SE
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
81.9
19
Me
an
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
210.
453
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.25
m P
lot
SE
0.00
0
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
1335
.818
0.00
0
56.1
21
Mea
n
0.00
0
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
1473
.170
0.00
0
84.1
81
N 10
10 10 10
10
10 10
10
10 10
10
10 10
10
10
0.12
5 m
Plo
t
SE
82.0
67
117.
316
0.00
0
0.00
0
38.2
64
51.3
37
271.
109
38.2
64
0.00
0
0.00
0
38.2
64
0.00
0
1048
4.37
7
0.00
0
51.3
37
Mea
n
114.
792
153.
057
0.00
0
0.00
0
38.2
64
76.5
28
344.
377
38.2
64
0.00
0
0.00
0
38.2
64
0.00
0
1048
4.37
7
0.00
0
76.5
28
N 11
11 11 11
11
11 11
11
11 11
11
11 11
11
11
Taxa
Co
pep
od
a
Pel
titi
idae
Ord
er: H
arp
acti
coid
a
Cir
rip
edia
Bal
anid
ae
Cu
mac
ean
Gyn
odi
asty
lidae
Lam
pro
pid
ae
Nan
nas
taci
dae
Dec
apo
da
Dio
gen
idae
Pilu
mni
dae
Po
rtu
nid
ae
Ech
ino
der
mat
a
Dia
dem
atid
ae
Op
hio
der
mat
idae
Cu
cum
ariid
ae
Gas
tro
po
da
Bat
illar
iidae
Bu
ccin
idae
Bu
llid
ae
77
On
e-W
ay A
NO
VA
P-v
alu
e
0.2
79
0.1
72
0.6
32
0.2
05
0.4
11
0.0
22
0.3
81
0.5
64
0.4
72
0.4
11
0.4
11
0.4
11
0.4
11
0.4
11
0.4
11
0.0
42
F st
atis
tic
1.2
76
1.5
57
0.7
24
1.4
57
1.0
33
2.6
68
1.0
82
0.8
12
0.9
39
1.0
33
1.0
33
1.0
33
1.0
33
1.0
33
1.0
33
2.3
23
Bar
e Sa
nd
SE
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
54
.95
3
47
.29
6
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
19
59
.43
7
0.0
00
Mea
n
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
10
5.2
26
70
.15
1
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
86
98
.71
6
0.0
00
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
Nat
ura
l Mea
do
w 1
SE
87
.25
6
38
.26
4
51
.33
7
0.0
00
0.0
00
0.0
00
0.0
00
82
.06
7
20
0.2
94
0.0
00
51
.33
7
0.0
00
38
.26
4
0.0
00
60
3.5
56
22
5.7
26
Mea
n
19
1.3
21
38
.26
4
76
.52
8
0.0
00
0.0
00
0.0
00
0.0
00
11
4.7
92
45
9.1
70
0.0
00
76
.52
8
0.0
00
38
.26
4
0.0
00
30
61
.13
2
49
7.4
34
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Nat
ura
l Mea
do
w 2
SE
82
.06
7
85
.56
1
38
.26
4
0.0
00
0.0
00
38
.26
4
51
.33
7
51
.33
7
12
4.5
79
0.0
00
51
.33
7
0.0
00
0.0
00
0.0
00
61
6.7
54
12
8.0
56
Mea
n
11
4.7
92
15
3.0
57
38
.26
4
0.0
00
0.0
00
38
.26
4
76
.52
8
76
.52
8
34
4.3
77
0.0
00
76
.52
8
0.0
00
0.0
00
0.0
00
19
89
.73
6
30
6.1
13
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
1 m
Plo
t
SE
0.00
0
70.1
51
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
35.0
75
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
586.
925
47.2
96
Me
an
0.00
0
70.1
51
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
35.0
75
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
1964
.226
70.1
51
N
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.5
m P
lot
SE
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
638.
755
70.1
51
Me
an
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
2525
.434
70.1
51
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.25
m P
lot
SE
56.1
21
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
84.1
81
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
597.
890
0.00
0
Mea
n
84.1
81
42.0
91
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
84.1
81
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
3030
.521
0.00
0
N 10
10
10
10
10
10
10
10
10 10 10
10
10 10
10 10
0.12
5 m
Plo
t
SE
76.5
28
38.2
64
0.00
0
38.2
64
38.2
64
0.00
0
0.00
0
38.2
64
38.2
64
38.2
64
51.3
37
38.2
64
0.00
0
0.00
0
557.
134
87.2
56
Mea
n
76.5
28
38.2
64
0.00
0
38.2
64
38.2
64
0.00
0
0.00
0
38.2
64
38.2
64
38.2
64
76.5
28
38.2
64
0.00
0
0.00
0
2295
.849
191.
321
N 11
11
11
11
11
11
11
11
11 11 11
11
11 11
11 11
Taxa
Gas
tro
po
da
Co
lum
bel
lidae
Epit
oni
idae
Fiss
ure
llid
ae
Hyd
atin
idae
Mit
rid
ae
Nat
icid
ae
Tere
bri
dae
Tro
chid
ae
Turb
inid
ae In
sect
a
Ch
iro
nom
idae
Iso
po
da
An
thel
uri
dae
Ido
teid
ae
Par
anth
urid
ae
Neb
alia
cea
Neb
aliid
ae
Nem
ato
da
Ost
raco
da
Ord
er: P
od
oco
pid
a
78
On
e-W
ay A
NO
VA
P-v
alu
e
0.1
58
0.2
24
< 0
.00
1
0.2
47
0.8
14
0.0
44
0.4
11
0.0
16
0.2
78
0.5
00
0.4
11
0.4
11
0.5
64
0.0
44
< 0
.00
1
0.3
74
0.4
11
0.2
79
0.1
72
0.6
32
F st
atis
tic
1.6
06
1.4
07
5.7
37
1.3
50
0.4
90
2.2
95
1.0
33
2.8
13
1.2
78
0.8
99
1.0
33
1.0
33
0.8
12
2.2
95
10
.32
3
1.0
94
1.0
33
1.2
76
1.5
57
0.7
24
Bar
e Sa
nd
SE
0.0
00
31
2.8
32
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
14
9.5
62
35
.07
5
0.0
00
0.0
00
16
7.8
83
15
0.6
80
0.0
00
0.0
00
47
.29
6
47
.29
6
0.0
00
Mea
n
0.0
00
80
6.7
36
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
0.0
00
56
1.2
08
35
.07
5
0.0
00
0.0
00
63
1.3
58
24
5.5
28
0.0
00
0.0
00
70
.15
1
70
.15
1
0.0
00
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
Nat
ura
l Mea
do
w 1
SE
38
.26
4
34
5.6
50
0.0
00
0.0
00
76
.52
8
11
7.3
16
11
4.7
92
0.0
00
0.0
00
33
4.8
93
0.0
00
11
4.7
92
51
.33
7
24
6.2
02
29
5.4
03
0.0
00
82
.06
7
85
.56
1
0.0
00
0.0
00
Mea
n
38
.26
4
95
6.6
04
0.0
00
0.0
00
76
.52
8
15
3.0
57
11
4.7
92
0.0
00
0.0
00
11
86
.18
9
0.0
00
11
4.7
92
76
.52
8
49
7.4
34
15
68
.83
0
0.0
00
11
4.7
92
15
3.0
57
0.0
00
0.0
00
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Nat
ura
l Mea
do
w 2
SE
51
.33
7
13
6.8
98
0.0
00
0.0
00
0.0
00
76
.52
8
0.0
00
0.0
00
11
8.5
57
14
8.1
96
0.0
00
0.0
00
0.0
00
16
1.4
37
21
4.4
16
0.0
00
38
.26
4
82
.06
7
38
.26
4
38
.26
4
Mea
n
76
.52
8
49
7.4
34
0.0
00
0.0
00
0.0
00
76
.52
8
0.0
00
0.0
00
19
1.3
21
49
7.4
34
0.0
00
0.0
00
0.0
00
30
6.1
13
68
8.7
55
0.0
00
38
.26
4
11
4.7
92
38
.26
4
38
.26
4
N
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
1 m
Plo
t
SE
35.0
75
101.
438
0.00
0
35.0
75
35.0
75
75.5
25
0.00
0
35.0
75
0.00
0
75.5
25
0.00
0
0.00
0
47.2
96
141.
493
193.
855
0.00
0
0.00
0
35.0
75
0.00
0
35.0
75
Me
an
35.0
75
350.
755
0.00
0
35.0
75
35.0
75
105.
226
0.00
0
35.0
75
0.00
0
105.
226
0.00
0
0.00
0
70.1
51
175.
377
420.
906
0.00
0
0.00
0
35.0
75
0.00
0
35.0
75
N
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.5
m P
lot
SE
0.00
0
156.
505
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
81.9
19
0.00
0
0.00
0
0.00
0
70.1
51
132.
090
0.00
0
0.00
0
79.1
41
35.0
75
0.00
0
Me
an
0.00
0
526.
132
0.00
0
0.00
0
0.00
0
35.0
75
0.00
0
0.00
0
0.00
0
210.
453
0.00
0
0.00
0
0.00
0
70.1
51
631.
358
0.00
0
0.00
0
140.
302
35.0
75
0.00
0
N 12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
0.25
m P
lot
SE
0.00
0
89.8
37
0.00
0
0.00
0
0.00
0
56.1
21
0.00
0
0.00
0
84.1
81
206.
201
0.00
0
0.00
0
0.00
0
112.
242
266.
574
0.00
0
0.00
0
42.0
91
0.00
0
0.00
0
Mea
n
0.00
0
126.
272
0.00
0
0.00
0
0.00
0
84.1
81
0.00
0
0.00
0
84.1
81
336.
725
0.00
0
0.00
0
0.00
0
252.
543
1388
.989
0.00
0
0.00
0
42.0
91
0.00
0
0.00
0
N 10
10 10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0.12
5 m
Plo
t
SE
38.2
64
117.
316
0.00
0
0.00
0
0.00
0
373.
345
38.2
64
0.00
0
0.00
0
279.
616
76.5
28
0.00
0
0.00
0
85.5
61
301.
778
76.5
28
0.00
0
38.2
64
114.
792
0.00
0
Mea
n
38.2
64
267.
849
0.00
0
0.00
0
0.00
0
573.
962
38.2
64
0.00
0
0.00
0
573.
962
76.5
28
0.00
0
0.00
0
267.
849
1415
.773
76.5
28
0.00
0
38.2
64
114.
792
0.00
0
N 11
11 11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Taxa
Ost
raco
da
Cyp
rid
ino
idae
Ru
tid
erm
atid
ae
Po
lych
aeta
Ap
isto
bra
nch
idae
Cap
itel
lidae
Cir
ratu
lidae
Do
rvill
eid
ae G
rou
p 3
Eun
icid
ae
Go
nia
did
ae
Lop
ado
rhyn
chid
ae
Lum
brin
erid
ae
Mag
elo
nid
ae
Mal
dan
idae
Ner
eid
idae
Oen
on
idae
On
up
hid
ae
Orb
iniid
ae
Ow
eniid
ae
Par
aoni
dae
Ph
yllo
do
cid
ae
Pila
rgid
ae
79
On
e-W
ay A
NO
VA
P-v
alu
e
0.2
05
0.4
11
0.0
22
0.3
81
0.5
64
0.4
72
0.4
11
0.4
11
0.4
11
0.4
11
0.4
11
0.4
11
0.0
42
F st
atis
tic
1.4
57
1.0
33
2.6
68
1.0
82
0.8
12
0.9
39
1.0
33
1.0
33
1.0
33
1.0
33
1.0
33
1.0
33
2.3
23
Bar
e Sa
nd
SE
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
11
9.6
50
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
35
.07
5
Mea
n
0.0
00
35
.07
5
0.0
00
0.0
00
0.0
00
0.0
00
28
0.6
04
0.0
00
0.0
00
0.0
00
35
.07
5
0.0
00
35
.07
5
N 12
12
12
12
12
12
12
12
12
12
12
12
12
Nat
ura
l Mea
do
w 1
SE
0.0
00
0.0
00
0.0
00
0.0
00
56
19
.96
0
0.0
00
24
3.2
11
0.0
00
0.0
00
82
.06
7
0.0
00
38
.26
4
85
.56
1
Mea
n
0.0
00
0.0
00
0.0
00
0.0
00
82
65
.05
6
0.0
00
14
92
.30
2
0.0
00
0.0
00
11
4.7
92
0.0
00
38
.26
4
15
3.0
57
N
11
11
11
11
11
11
11
11
11
11
11
11
11
Nat
ura
l Mea
do
w 2
SE
0.0
00
38
.26
4
0.0
00
0.0
00
10
2.6
73
0.0
00
19
1.3
21
0.0
00
0.0
00
38
.26
4
0.0
00
0.0
00
0.0
00
Mea
n
0.0
00
38
.26
4
0.0
00
0.0
00
15
3.0
57
0.0
00
61
2.2
26
0.0
00
0.0
00
38
.26
4
0.0
00
0.0
00
0.0
00
N
11
11
11
11
11
11
11
11
11
11
11
11
11
1 m
Plo
t
SE
0.00
0
0.00
0
35.0
75
0.00
0
75.5
25
0.00
0
281.
201
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
Me
an
0.00
0
0.00
0
35.0
75
0.00
0
105.
226
0.00
0
385.
830
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
35.0
75
N
12
12
12
12
12
12
12
12
12
12
12
12
12
0.5
m P
lot
SE
35.0
75
47.2
96
0.00
0
0.00
0
0.00
0
0.00
0
141.
887
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
Me
an
35.0
75
70.1
51
0.00
0
0.00
0
0.00
0
0.00
0
210.
453
35.0
75
0.00
0
0.00
0
0.00
0
0.00
0
0.00
0
N 12
12
12
12
12
12
12
12
12
12
12
12
12
0.25
m P
lot
SE
0.00
0
0.00
0
0.00
0
0.00
0
56.1
21
42.0
91
146.
479
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
42.0
91
Mea
n
0.00
0
0.00
0
0.00
0
0.00
0
84.1
81
42.0
91
462.
996
0.00
0
0.00
0
0.00
0
0.00
0
42.0
91
42.0
91
N 10
10
10
10
10
10
10
10
10 10 10 10
10
0.12
5 m
Plo
t
SE
0.00
0
0.00
0
0.00
0
38.2
64
38.2
64
38.2
64
221.
800
0.00
0
38.2
64
0.00
0
0.00
0
0.00
0
0.00
0
Mea
n
0.00
0
0.00
0
0.00
0
38.2
64
38.2
64
38.2
64
573.
962
0.00
0
38.2
64
0.00
0
0.00
0
0.00
0
0.00
0
N 11
11
11
11
11
11
11
11
11 11 11 11
11
Taxa
Po
lych
aeta
Pis
ioni
dae
Po
ecilo
chae
tid
ae
Sib
ogl
inid
ae
Siga
lion
idae
Spir
orb
idae
Serp
ulid
ae
Sylli
dae
Tere
bel
lidae
Tric
ho
bran
chid
ae
Po
lyp
laco
ph
ora
Isch
no
chit
on
idae
Pyc
no
gon
ida
Cal
lipal
len
idae
Tan
aid
acae
Neo
ten
aid
ae
Tan
aid
ae