This is the published version:   Hirst,Alastair,Ball,David,Heislers,Simon,Blake,SeanandCoots,Allister2008,Baywideseagrassmonitoringprogram:milestonereportno.1(2008)DepartmentofPrimaryIndustries,Queenscliff,Vic.

Available from Deakin Research Online:  http://hdl.handle.net/10536/DRO/DU:30059138Reproducedwiththekindpermissionofthecopyrightowner.Copyright:2008,DepartmentofPrimaryIndustries

Baywide Seagrass Monitoring Program Milestone Report No. 1 (2008).

No. 25

November 2008

Baywide Seagrass Monitoring Program

Milestone Report No. 1 (2008).

Alastair Hirst, David Ball, Simon Heislers, Sean Blake and Allister Coots.

Fisheries Victoria

Technical Report Series No. 25

November 2008

ii

Published: Fisheries Victoria Department of Primary Industries, Queenscliff Centre PO Box 114, Queenscliff, Victoria 3225 Australia.

General disclaimer This publication may be of assistance to you but the State of Victoria and its employees do not guarantee that the publication is without flaw of any kind or is wholly appropriate for your particular purposes and therefore disclaims all liability for any error, loss or other consequence which may arise from you relying on any information in this publication.

Copyright The State of Victoria, Department of Primary Industries, 2008.

This publication is copyright. No part may be reproduced by any process except in accordance with the provisions of the Copyright Act 1968.

Authorised by the Victorian Government, GPO Box 4440, Melbourne 3001.

Printed by Fisheries Victoria, Queenscliff, Victoria

Preferred way to cite: Hirst A, Ball D, Heislers S, Blake S, Coots A (2008) Baywide Seagrass Monitoring Program, Milestone Report No. 1 (2008). Fisheries Victoria Technical Report No. 25, November 2008, 30 pp. Department of Primary Industries, Queenscliff, Victoria, Australia.

ISSN 1835-4785

ISBN

iii

Executive Summary

Seagrass is an important habitat in Port Phillip Bay (PPB). Seagrasses are highly productive ecosystems, supporting diverse faunal assemblages, many of commercial importance. Seagrass plants filter and retain nutrients, stabilise sediments and baffle wave energy, protecting adjacent coastal shorelines from erosion.

The Baywide Seagrass Monitoring Program comprises three main elements: 1) large-scale mapping of seagrass extent, 2) small-scale assessment of seagrass health in the field and 3) monitoring of environmental factors that are known to influence seagrass health. Large-scale mapping of seagrass using aerial photography flown in April 2008 at the Blairgowrie region was presented in Progress Report No. 1. Mapping for all regions will be presented in Milestone Report No. 2.

The objective of this program is to detect changes in seagrass health in PPB outside expected variability.

This milestone report covers the first seagrass health assessment completed in autumn (April/May) 2008 and includes a detailed assessment of 1) seagrass cover, stem density and canopy height for intertidal and subtidal seagrass plots, and 2) factors that are known to influence seagrass health (light, turbidity, nutrients and epiphyte cover), at six regions located in the south and west of PPB. Comparisons were made against historical data collected between 2004 and 2007 at three of the six regions.

Seagrass cover, density and canopy height varied between regions and depths. Subtidal seagrass beds monitored in this study consisted entirely of a single seagrass species Heterozostera nigricaulis. Intertidal seagrass beds usually comprised Zostera muelleri, although Lepilaena marina was also present at two of the regions monitored.

Seagrass cover and stem density, and to a lesser extent canopy height, were highest at subtidal plots in the southern part of PPB (i.e. Blairgowrie, Mud Islands and Swan Bay). Seagrass cover was low at subtidal plots located in the western part of PPB. These plots were dominated primarily by stems without leaf shoots. Historical data at Point Richards indicated that seagrass cover had been higher in the past and losses at this plot were

part of an ongoing trend for this region. Seagrass loss at the Kirk Point plot appeared to be recent, with most of the loss occurring between April 2007 and April 2008.

Environmental light conditions were monitored at each region using light sensors deployed between April and July 2008. Benthic light availability exceeded conservative environmental requirements for seagrasses in southern PPB at all regions except Swan Bay. The data at Swan Bay was thought to be unreliable due to problems with the logger deployment.

Turbidity levels adjacent to the seagrass assessment regions were low and were well within the limits specified in the CDP Environmental Management Plan.

Epiphytic algae were more abundant on subtidal than intertidal seagrass plants. Epiphytic algal loads on subtidal seagrasses were patchy in distribution and were only rarely high. In the cases where epiphytic algal loads were high, comparisons with historical data indicated that current levels were similar to those recorded in the recent past at these regions.

Conclusions Subtidal seagrass cover, height and density in autumn 2008 were higher in the southern part than in the western part of PPB. This suggests that seagrass in the south was relatively healthy; although in the absence of previous information about the past condition of seagrasses at these regions it is difficult to be definitive regarding this assessment. Historical information indicates that seagrass cover in the west was higher in the past. Few conclusions can be drawn regarding the role various environmental variables play in influencing seagrass health at this stage of the program.

v

Table of Contents

Executive Summary............................................................................................. iii

Introduction............................................................................................................ 1

Purpose of this report................................................................................................................................................1

Project Design and Methods ............................................................................... 2

Sampling Regions......................................................................................................................................................2

Field Sampling ...........................................................................................................................................................2 Establishment of plots for small-scale monitoring.............................................................................................2

Measurement of seagrass health...........................................................................................................................2

Measuring seagrass boundaries............................................................................................................................2

Light, turbidity, nutrients and epiphyte monitoring .........................................................................................3

Comparisons with Historical Data .........................................................................................................................4

Other Factors...............................................................................................................................................................4 Grazing by swans....................................................................................................................................................4

Desiccation stress ....................................................................................................................................................4

Spadices....................................................................................................................................................................4

Drift algae.................................................................................................................................................................4

Other variables ........................................................................................................................................................4

Data Management QA/QC.......................................................................................................................................4

Exceptions to Detailed Design ................................................................................................................................4

Results...................................................................................................................... 7

Seagrass Health ..........................................................................................................................................................7 Subtidal ....................................................................................................................................................................7

Intertidal...................................................................................................................................................................7

Light, Turbidity and Epiphytes...............................................................................................................................9 Light attenuation (Kd), % surface irradiance and turbidity...............................................................................9

Epiphytes ...............................................................................................................................................................15

Other Factors .........................................................................................................................................................15

Comparisons against historical data ....................................................................................................................18 Seagrass health ......................................................................................................................................................18

Seagrass epiphyte cover .......................................................................................................................................18

Discussion............................................................................................................. 22

Seagrass cover...........................................................................................................................................................22

vi

Factors that affect seagrass health ........................................................................................................................ 22

Conclusions.............................................................................................................................................................. 23

Acknowledgements ............................................................................................ 24

References ............................................................................................................. 25

Appendix 1............................................................................................................ 26

Appendix 2............................................................................................................ 27

Appendix 3............................................................................................................ 29

vii

List of Tables Table 1. Monitoring regions and assessment methods for the different components of the CDBMP

Seagrass Monitoring Program.......................................................................................................................... 5

Table 2. Mean daily light attenuation coefficients (Kd) and % surface irradiance at depths of shallow (2 m) and deep plots (5 m) from 10am–2pm calculated for each region and for varying intervals of time at different sites during the reporting period April-July 2008 (see Figs 6-11 for actual reporting dates).. 9

Table 3. Diver precision: Mean standard deviation (SD), % coefficient of variation (CV), range of diver observations and differences between diver pairs (n = 5 divers) for the QC/QA exercise (n = 10 quadrats). .......................................................................................................................................................... 27

Table 4. Diver accuracy and bias for the QA/QC exercise (n = 10 quadrats). .................................................. 28

List of Figures Figure 1 Location of seagrass monitoring regions in Port Phillip Bay. .............................................................. 6

Figure 2. Location of light loggers, EPA water quality monitoring sites and PoMC turbidity monitoring stations................................................................................................................................................................. 6

Figure 3. Mean (± se) seagrass cover (%) and shoot length (cm) for subtidal H. nigricaulis assemblages sampled at shallow and deep plots at six regions around PPB (n=12 quadrats) during the reporting period (April/May 2008); n/a indicates where no data were available....................................................... 7

Figure 4. Mean (± se) shooting stem and non-shooting stem density (count per 0.0625 m2 quadrat) for subtidal H. nigricaulis assemblages sampled at shallow and deep plots at six regions around PPB during the reporting period (April/May 2008); n/a indicates where no data were available.................. 8

Figure 5. Mean (± se) a) composition (%) by species (Z. muelleri or L. marina), b) cover (%), c) shooting stem density (count per 0.0625 m2 quadrat) and d) shoot length (cm) for intertidal plots sampled at four regions around PPB during the reporting period (April/May 2008).................................................. 8

Figure 6. Mean (± se) light attenuation coefficients (m-1), calculated daily between 10am and 2pm, and turbidity at St Leonards. ................................................................................................................................. 10

Figure 7. Mean (± se) light attenuation coefficients (m-1), calculated daily between 10am and 2pm, and turbidity at Mud Islands (note: logger data for the north west pile after 7 June may have been affected by fouling of the wiper system, and loggers at the south east pile failed to record usable data for most of the deployment period). ............................................................................................................. 11

Figure 8. Mean (± se) light attenuation coefficients, calculated daily between 10am and 2pm, at Swan Bay (note: different Y axis scale to other light attenuation figures).................................................................. 12

Figure 9. Mean (± se) light attenuation coefficients (m-1), calculated daily between 10am and 2pm, at Kirk Point................................................................................................................................................................... 12

Figure 10. Mean (± se) light attenuation coefficients (m-1), calculated daily between 10am and 2pm, and turbidity at Blairgowrie................................................................................................................................... 13

Figure 11. Mean (± se) light attenuation coefficients (m-1), calculated daily between 10am and 2pm, at Point Richards. ................................................................................................................................................. 14

Figure 12. Mean (± se) epiphytic turfing and encrusting algal cover (%) of H. nigricaulis leaf area at shallow and deep subtidal plots in PPB during the reporting period (April/May 2008)....................... 15

Figure 13. Mean (± se) epiphytic macroalgal cover (%) of quadrats sampled at shallow and deep subtidal plots in PPB during the reporting period (April/May 2008). ..................................................................... 16

Figure 14. Mean (± se) turfing and encrusting epiphytic algal cover (%) of Z. muelleri shoot area at intertidal plots in PPB during the reporting period (April/May 2008)..................................................... 16

viii

Figure 15. Mean (± se) epiphytic macroalgal cover (%) at intertidal plots in PPB during the reporting period (April/May 2008).................................................................................................................................. 17

Figure 16. Mean (± se) cover (%) of drift macroalgae recorded for subtidal and intertidal plots in PPB during the reporting period (April/May 2008)............................................................................................. 17

Figure 17. Mean (±se) cover (%), seagrass length and stem density for H. nigricaulis seagrass at Kirk Point and Point Richards between November 2004 and April 2007 (Ball et al. in prep.) and the first Baywide seagrass monitoring field assessment in April 2008 (depicted in grey); n/a indicates where no data were available ................................................................................................................................................... 19

Figure 18. Mean (±se) cover (%), shoot length (cm) and density for intertidal seagrasses at Point Richards and Swan Bay between November 2004 and April 2007 (Ball et al. in prep) and the first Baywide seagrass field assessment in May 2008 (grey shading for Z. muelleri and no shading for L. marina); n/a indicates where no data were available ........................................................................................................ 20

Figure 19. Mean (±se) turfing, encrusting and macroalgal epiphytic cover (%) for subtidal H. nigricaulis at Kirk Point and Point Richards between April 2005 and April 2007 (Ball et al. in prep.) and the first Baywide seagrass field assessment in April 2008 (depicted in grey)........................................................ 21

1

Introduction

Seagrass is an important habitat in Port Phillip Bay (PPB). Seagrasses are highly productive ecosystems, supporting diverse faunal assemblages, many of commercial importance. Seagrass plants filter and retain nutrients, stabilise sediments and baffle wave energy, protecting adjacent coastal shorelines from erosion.

More than 95% of the seagrass in PPB is Heterozostera nigricaulis and Zostera muelleri (eel grasses belonging to the family Zosteraceae). The total area of Zosteraceae in PPB was estimated to be approximately 59 km2 in 2000 (Blake and Ball 2001).As PPB has a restricted tidal range (approximately 1 m) and a small intertidal zone, most of this seagrass is the subtidal H. nigricaulis.

Seagrass ‘health’ is affected by a range of natural and human influences that occur at a range of spatial and temporal scales. These factors include: • Habitat availability – substratum type and

stability • Water quality – nutrients, turbidity and

temperature • Hydrodynamics and coastal processes, such

as tides, currents and sediment transport • Trophic interactions between epiphytes,

grazers and predators.

Due to this breadth of factors and the various scales of influence, seagrass beds can be highly dynamic in both space and time.

The Baywide Seagrass Monitoring Program in PPB consists of three main elements (Table 1): • Annual large-scale monitoring of seagrass

coverage at nine regions using aerial mapping and video ground-truthing

• Small-scale monitoring of seagrass health for six of the nine regions at representative field assessment plots sampled quarterly (frequency of sampling to be reviewed after two years)

• Monitoring of key parameters that are known to affect seagrass health (including light and epiphyte abundance).

This program is described in the Port of Melbourne Corporation (PoMC) Channel Deepening Baywide Monitoring Programs (CDBMP) Seagrass Monitoring Detailed Design Rev2 (PoMC 2008a).

The objective of this program is to detect changes in seagrass health in PPB outside expected variability.

Purpose of this report This milestone report presents: • A summary of results for the small-scale

monitoring of seagrass health undertaken in autumn (April/May) 2008

• A summary of measurements for primary factors influencing seagrass health, including light, turbidity and epiphytes (for varying intervals of time at different sites during the reporting period April-July 2008)

• A discussion of relevant observations for other factors considered to influence seagrass health

• A discussion of trends in the data observed, along with comparisons against available historical data (2004–2007)

• Discussion of QA/QC issues and any peculiarities along with any associated implications for the collected data.

This milestone report does not include a summary of large-scale monitoring of seagrass beds with aerial mapping. Processing of the aerial photography captured in April/May 2008 is underway, and this will be presented in Milestone Report No. 2. Aerial mapping at the Blairgowrie region was presented in Progress Report No. 1.

2

Project Design and Methods

Project design and methods for this study are outlined in PoMC (2008a).

Sampling Regions Seagrass in PPB is being monitored at several spatial scales. Aerial photography of broad stretches of the coastline flown annually during April/May will allow characterisation of large areas of seagrass in PPB. More detailed assessments are undertaken through mapping seagrass distribution from aerial photography in approximately 1 km2 mega-quadrats at nine intertidal/shallow regions in the Bay (Table 1, Figure 1; see also Exceptions, ER2008#13). These regions were selected to be representative of the major seagrass areas in PPB.

Measurement of seagrass health at a smaller spatial scale is being undertaken at six of the nine detailed field assessment regions (i.e. Kirk Point, Point Richards, St Leonards, Swan Bay, Mud Islands and Blairgowrie) and began in autumn (April/May) 2008 (Table 1, Figure 1). These six regions will be monitored quarterly for the first two years of the program, with the frequency of monitoring to then be reviewed.

Field Sampling Establishment of plots for small-scale monitoring Seagrass health was measured at the six seagrass field assessment regions in three, fixed, 10 m diameter plots, located in intertidal, shallow subtidal (1–2 m) and deep subtidal (3–5 m) seagrass. Differences in bathymetry and seagrass distribution meant that seagrass was not found at all depths at each region. Consequently, not all plot types were established at each region (Table 1). Intertidal seagrass was only present at Point Richards (Bellarine Bank), St Leonards, Swan Bay and Mud Islands.

Twelve, 0.25 m2 (0.5 m x 0.5 m) quadrats were initially randomly located within each plot. Twelve replicates were chosen based on power analysis of shoot density (PoMC 2008a). Quadrat positions were fixed in the shallow subtidal and deep subtidal plots. Fixed quadrats enable greater detection of temporal change by minimising variation associated with small-scale, spatial patchiness.

Quadrats were not fixed in the intertidal plots because these habitats were exposed at low tide and could not be marked without attracting possible interference from the public. As a consequence, quadrats were randomly positioned upon each visit.

Measurement of seagrass health Seagrass cover, height and shoot, and non-shooting stem density were measured in situ. Percentage seagrass cover was estimated visually. Seagrass height (80th percentile) was measured with a ruler. Photographs of each quadrat were taken prior to sampling to provide a permanent record of the seagrass cover for future reference.

The number of H. nigricaulis primary vertical stems (with and without shoots) and Z. muelleri shoots were measured within a smaller 0.0625 m2 (0.25 m x 0.25 m) quadrat placed within the south-western corner of the larger quadrat (i.e. a fixed sub-sample). Z. muelleri shoots originate directly from the horizontal rhizome, whereas H. nigricaulis shoots arise from a vertical stem (i.e. a modified rhizome).

The selection of 12 replicate 0.0625 m2 quadrats was based on power analysis of shoot density data from 0.02 m2 cores collected in a previous study (Ball et al. 2006). The adoption of 0.0625 m2 quadrats is expected to have little impact on statistical power as the area sampled is now larger than in the previous study (0.0625 versus 0.02 m2). Consequently, the variance between samples is expected to be similar or less than that observed for 0.02 m2 cores.

Measuring seagrass boundaries Upper boundary of intertidal seagrass The visible upper boundary of intertidal seagrass was measured at regions where intertidal seagrass occurred (Table 1). A ‘Thales’ mobile mapper, which incorporates a DGPS and mapping software, was used to record the position of a line along the upper limit of the intertidal seagrass walked by the holder of the mobile mapper (accuracy ±2 m). At least three lines, each 10–20 m long and approximately 50 m apart, were mapped and will be monitored at the same frequency as the seagrass health assessment.

3

The upper limits of intertidal seagrass were not able to be mapped at Swan Bay because the intertidal zone at this region is often covered with drift seagrass and algae (i.e. cast onto the shoreline) (see also Exceptions, ER 2008#13).

Outer boundary of subtidal seagrass The outer (deeper) edge of seagrass beds within or adjoining each region was to be monitored by establishing permanent markers in the sand. This method was found to be impractical since it relied on the existence of a clear boundary at the maximum depth limit of the seagrass. The deeper subtidal H. nigricaulis beds were very sparse and patchy with no clearly recognisable boundary between the seagrass and bare sediment. As a consequence, this task was not undertaken and an alternative method of monitoring the depth limit of the deep seagrass using underwater video will commence during the spring sampling (see also Exceptions, ER2008#13).

Light, turbidity, nutrients and epiphyte monitoring In addition to monitoring changes in seagrass health, other factors that influence seagrass health were monitored at each of the six seagrass field assessment regions.

Light Light is a primary determinant of seagrass depth and distribution. Light intensity was continuously measured at each region using Odyssey light loggers fitted with an automated wiper system to reduce fouling. A pair of light loggers was placed on the nearest suitable navigation pile at the Kirk Point, Point Richards and Swan Bay regions, and an additional pair of loggers were deployed at the Blairgowrie, Mud Islands and St Leonards regions (Figure 2). The closest pile to the Kirk Point region was at Long Reef approximately 4.5 km to the northeast. Each pair of loggers were placed underwater and separated by at least 1 m, and oriented to the north to limit shading from the pile. Data were uploaded after three months of deployment.

Data from each pair of sensors were converted to ‘light attenuation’ coefficients (Kd) using Beers Law and daily means for the period 10am–2pm plotted. The light attenuation coefficient (Kd) measures the proportion of PAR (photosynthetically active radiation) light absorbed as light travels through the water. Larger attenuation coefficients indicate increased attenuation and reduced PAR light availability as a result of absorption and scattering in the water column. Generally, Kd coefficients >1 m-1 are

indicative of turbid waters, whereas coefficients 0.2–0.3 m-1 are representative of clearer waters (Carruthers et al. 2001). In clear, shallow waters in the southern part of PPB, Kd coefficients are typically about 0.2 m-1 (Longmore et al. 1996).

The percentage of surface irradiance at the depths of the shallow (2 m) and deep (5 m) plots were calculated from the mean daily light attenuation Kd coefficients by transposing Beer’s Law. The calculation of light attenuation did not account for changes in tidal height.

Background light attenuation in clear seawater is approximately 0.1 m-1 and attenuation coefficient values that were less that this value were treated as data anomalies and excluded from the calculations.

Turbidity High water-column turbidity reduces the penetration of light to the benthos. Data from the PoMC turbidity conformance monitoring locations near the seagrass regions in southern PPB (i.e. Coles Channel (St Leonards), Mud Islands and Camerons Bight (Blairgowrie) (Figure 2) were provided by the PoMC as 6-hourly exponentially weighted moving averages (EWMA) (PoMC 2008b). The CDP Environmental Management Plan (POMC 2008c) identified an environmental limit for these locations of 25 NTU not to be exceeded as a 6 hourly EWMA. The turbidity data was plotted against light attenuation and was used to aid interpretation of trends in light attenuation.

Nutrients Nutrient data are collected by EPA near some of the seagrass regions for the CDBMP Water Quality program. Relationships between nutrients and seagrass, if present, will be difficult to detect until changes in time series of both have been assessed (at least two years of data required). Nutrient impacts are most likely to be expressed as epiphytic growth.

Epiphytes High nutrient levels may stimulate epiphytic algal growth on seagrass, leading to excessive shading and potential loss of seagrass. Epiphytes commonly occur on several different parts of the seagrass plant (Ball et al. in prep.). Epiphyte growth on seagrass leaves was assessed as this potentially has the greatest influence on seagrass health.

Epiphytic algae can be broadly split into three functional forms: • Encrusting coralline algae

4

• Turfing algae characterised as a fine “fuzzy” layer of filamentous algae growing on leaf surfaces typically <1 cm long

• Macroalgae, usually > 1 cm long.

Encrusting and turfing algal abundance were estimated as percentage cover of visible leaf area. Foliose macroalgae, due to greater shading (i.e. >1 seagrass leaf), were estimated as the percentage cover of each quadrat.

Comparisons with Historical Data Historical seagrass data were available for three of the six subtidal field assessment plots: Kirk Point, Point Richards and Swan Bay; and two of the intertidal plots: Point Richards and Swan Bay (Ball et al. in prep.). Each of these regions has information on seagrass cover, stem/shoot density and length, and epiphyte cover from spring (November) and autumn (April/May) between November 2004 or April 2005 through to April 2007 (typically five sampling events).

Other Factors Evidence of other factors which influence seagrass health was also recorded as field notes.

Grazing by swans Swan grazing on seagrass rhizomes influences tidal flat topography through the creation of small pits dug during grazing. Where grazing is observed, the % cover of small pits within each quadrat will be estimated. Swan grazing is greatest during summer when swans are most abundant in shallow marine areas and is not reported here.

Desiccation stress High air temperatures and strong drying winds may kill seagrass plants exposed to extreme conditions at low tide. Differences in desiccation stress can be assessed by comparing the extent of brown seagrass leaves (i.e. dead tissue) over time. The relative proportions of green and brown leaves occurring within quadrats were estimated during each field survey. Change in the proportion of brown leaf area will be used as an index of desiccation stress, primarily during summer and in the intertidal plots, when and where seagrass beds are most susceptible to desiccation. No desiccation data is reported here.

Spadices Traditionally seagrass regeneration has been considered to occur primarily through asexual growth, as they spread using rhizomes. However, seagrass seeds may play an important

role in meadow regeneration. A measure of inter-annual and regional differences in seed production will be obtained by taking field notes on the abundance of spadices during fertile periods: • Z. muelleri from October to March • H. nigricaulis from September to February.

No data on spadices is reported here.

Drift algae The presence of large amounts of drift algae can shade and smother seagrass. Field-notes were taken to describe the presence and cover of any drift algae in the plots.

Other variables H. nigricaulis also reproduces asexually (Cambridge et al. 1983), and produces vegetative propagules consisting of a shoot, horizontal rhizome and roots. The number of vegetative propagules of H. nigricaulis attached to the rhizomes was counted in each quadrat.

A range of organisms other than epiphytic algae routinely foul seagrass leaves. The percentage cover of common seagrass epibiota such as sponges, ascidians Pyura sp. and the encrusting bivalve Electroma georgiana were estimated where they were encountered within the 0.0625 m2 quadrat. Common grazing gastropods, such as Austrocochlea sp. and members of the family Dialidae, were also counted where present.

Data Management QA/QC. Data Quality Assurance and Quality Control procedures employed in this study are detailed in Appendix 2.

Exceptions to Detailed Design Exceptions to the Detailed Design (PoMC 2008a) for the reporting period are documented in Exception Report (ER2008#13) and summarised as follows: • The location of seagrass field and/or aerial

assessment regions at St Leonards (intertidal, shallow and deep), Blairgowrie (deep) and Mud Islands (deep) were changed

• The presence of intertidal seagrass was confirmed at the Mud Islands region and a field-assessment plot was established in this seagrass. No intertidal seagrass was present at the Kirk Point and Blairgowrie regions

• Fixed markers were not installed to monitor the outer (deeper) edge of seagrass beds at the regions with deep (approximately 5 m)

5

seagrass plots (i.e. Point Richards, St Leonards, Mud Islands and Blairgowrie)

• The upper boundary of the intertidal seagrass was not measured at the Swan Bay region

• The location of the shallow field-assessment plot established at Swan Bay in April 2008

did not match the position of the shallow plot used by Ball et al. (in prep.)

• This milestone report was not delivered in the specified timelines.

Table 1. Monitoring regions and assessment methods for the different components of the CDBMP Seagrass Monitoring Program.

Region Assessment Method Field Assessment Plots*

Aerial (Annual) Field (Quarterly)

Intertidal Shallow (1–2 m) Deep (2–5 m)

Altona √

Kirk Point √ √ √

Point Henry West √

Curlewis Bank √

Point Richards √ √ √ √ √

St Leonards √ √ √ √ √

Swan Bay √ √ √ √

Mud Islands √ √ √ √ √

Blairgowrie √ √ √ √

*n = 12 replicate quadrats/plot

6

Figure 1 Location of seagrass monitoring regions in Port Phillip Bay.

Figure 2. Location of light loggers, EPA water quality monitoring sites and PoMC turbidity monitoring stations.

Note: The closest pile for deployment of light loggers at the Kirk Point region was located at Long Reef.

7

Results

Seagrass Health The information reported in this milestone report includes the first seagrass health assessment undertaken in autumn 2008 (April & May). Results are reported separately for subtidal plots (shallow and deep) containing H. nigricaulis, and intertidal plots, typically dominated by Z. muelleri.

Subtidal H. nigricaulis seagrass cover and canopy height (shoot length) varied between regions and depths within a region (Figure 3). Seagrass cover was greatest at Mud Islands (both plots), Swan Bay and the shallow Blairgowrie plot. Seagrass cover was higher in shallow plots than in deep plots, except at St Leonards. Seagrass canopy height varied between regions and depths. Shoot length was greatest at Mud Islands and Swan Bay.

Shooting stem density varied between regions (Figure 4). Densities >70 shooting stems/quadrat were recorded at Blairgowrie, Mud Islands, and Swan Bay. Point Richards and the deeper plots at

Blairgowrie and St Leonards were dominated by stems without shoots.

Intertidal Intertidal seagrass beds were present at four of the six regions: Mud Islands, Point Richards, Swan Bay and St Leonards. Z. muelleri dominated the Point Richards and St Leonards plots. The Swan Bay plot was dominated by the aquatic macrophyte Lepilaena marina. The intertidal plot at Mud Islands consisted of a mixture of Z. muelleri and L. marina (Figure 5a).

Seagrass cover (Z. muelleri and L. marina combined) varied between regions (Figure 5b). Seagrasses covered at least 80% of the intertidal benthos at Swan Bay and St Leonards (Figure 5b). Shooting stem density varied between intertidal regions. Shooting stem density was greatest at St Leonards and Swan Bay (>500 shoots/quadrat) and least at Mud Islands (Figure 5c). Z. muelleri shoot length was similar at all regions (Figure 5d) regardless of differences in shoot density.

Blairgowrie

Mud Islands

Swan Bay

St Leonard

s

Pt Richard

s

Kirk P

oint0

20

40

60

80

100

% c

over Deep (2-5 m)

Shallow (1-2 m)

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

15

30

45

60

75

sh

oot

l en

gth

(cm

)

Figure 3. Mean (± se) seagrass cover (%) and shoot length (cm) for subtidal H. nigricaulis assemblages sampled at shallow and deep plots at six regions around PPB (n=12 quadrats) during the reporting period (April/May 2008); n/a indicates where no data were available.

n/a n/a n/a n/a

8

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

75

150

225

300

sh

ootin

g s

tem

co

un

t

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

75

150

225

300

no

n- s

ho

otin

g s

tem

co

un

t

Deep (2-5 m)Shallow (1-2 m)

Depth

Figure 4. Mean (± se) shooting stem and non-shooting stem density (count per 0.0625 m2 quadrat) for subtidal H. nigricaulis assemblages sampled at shallow and deep plots at six regions around PPB during the reporting period (April/May 2008); n/a indicates where no data were available.

Mud Islands

Swan Bay

St Leonard

s

Pt Rich

ards0

20

40

60

80

100

% c

om

po

sitio

n

LepilaenaZostera

Mud Islands

Swan Bay

St Leonard

s

Pt Richard

s0

20

40

60

80

100

% c

ove

r

Mud Islands

Swan Bay

St Leonard

s

Pt Richard

s0

300

600

900

1200

1500

sh

oo

tin

g s

tem

cou

nt

Mud Islands

Swan Bay

St Leonard

s

Pt Richard

s0

5

10

15

20

25

Sh

oo

t le

ng

th (

cm

)

Figure 5. Mean (± se) a) composition (%) by species (Z. muelleri or L. marina), b) cover (%), c) shooting stem density (count per 0.0625 m2 quadrat) and d) shoot length (cm) for intertidal plots sampled at four regions around PPB during the reporting period (April/May 2008).

a b

c d

n/a n/a n/a n/a

9

Light, Turbidity and Epiphytes Light attenuation (Kd), % surface irradiance and turbidity Mean daily Kd values are presented in Figures 6–11. Where turbidity data (6-hourly exponentially weighted moving averages) were available from a nearby PoMC monitoring station the value from 12 noon was overlaid on the light attenuation data.

Turbidity data were only available for the St Leonards, Mud Islands and Blairgowrie regions (Figures 6, 7 and 10). Turbidity levels at these regions were generally low throughout the first assessment period (April-July).

Attenuation coefficients for regions in the southern part of the Bay were typically in the range of 0.2–0.5 m-1. Attenuation coefficients at these regions rarely exceeded 0.5 m-1. The periods where attenuation values exceeded 0.5 m-1 were mostly attributed to problems with the operation of the loggers (Appendix 3).

Mean daily attenuation coefficients at Swan Bay typically exceeded 1.0 m-1 with most observations falling within the range of 1–4 m-1 (Figure 8).

Percentage surface irradiance calculated at the depths of the shallow (2 m) and deep (5 m) plots is summarised in Table 2.

Table 2. Mean daily light attenuation coefficients (Kd) and % surface irradiance at depths of shallow (2 m) and deep plots (5 m) from 10am–2pm calculated for each region and for varying intervals of time at different sites during the reporting period April-July 2008 (see Figs 6-11 for actual reporting dates).

Region (light logger) Lower

logger

depth (m)

Distance

to shallow

plot (km)

Distance

to deep

plot

(km)

Mean

daily Kd

(m-1)

Mean %

daily

irradiance

at 2 m

Mean %

daily

irradiance

at 5 m

Total

days

Point Richards (Aquaculture zone pile)

4.0 1.3 0.07 0.2 66 37 60

Kirk Point (Long Reef) 3.0 4.5 NA 0.3 55 NA 54

Blairgowrie (speed restriction pile)

2.7 0.7 0.08 0.2 65 36 47

Blairgowrie (Sorrento Channel No. 10)

5.4 0.7 0.5 0.2 67 38 55

Mud Islands (North West MNP pile)

3.2 1.2 5 0.2 63 33 61

Mud Islands (South East MNP pile)*

2.0 2.5 2.4 0.2 69 40 6

St. Leonards (Coles Channel No. 5)

5.0 0.8 0.4 0.4 51 23 66

Swan Bay (MNP pile south east of jetty)

1.3 0.5 NA 2.3 8 NA 66

* not a complete time-series

10

St L

eo

na

rds - C

ole

s C

ha

nne

l No

. 5

0

0.5 1

1.5 2

2.5 3

14-Apr-08

16-Apr-0818-Apr-08

20-Apr-0822-Apr-08

24-Apr-0826-Apr-08

28-Apr-0830-Apr-08

02-May-0804-May-08

06-May-0808-May-08

10-May-0812-May-08

14-May-0816-May-08

18-May-0820-May-08

22-May-0824-May-08

26-May-0828-May-08

30-May-0801-Jun-08

03-Jun-08

05-Jun-0807-Jun-08

09-Jun-0811-Jun-08

13-Jun-0815-Jun-08

17-Jun-0819-Jun-08

21-Jun-0823-Jun-08

25-Jun-0827-Jun-08

29-Jun-0801-Jul-08

03-Jul-0805-Jul-08

07-Jul-0809-Jul-08

11-Jul-0813-Jul-08

15-Jul-0817-Jul-08

19-Jul-08

Da

y

Kd (m-1

)

0 5 10

15

20

25

Turbidity (NTU)

Mea

n L

ight A

ttenuatio

nP

OM

C T

urb

idity

Fig

ure

6. M

ea

n (±

se) lig

ht a

tten

ua

tion

coe

fficien

ts (m-1), calcu

late

d d

aily b

etw

een 1

0a

m a

nd

2p

m, a

nd

tu

rbid

ity a

t St L

eo

na

rds.

11

Mud

Isla

nds N

orth

We

st

0

0.5 1

1.5 2

2.5 3

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

23-Jun-08

25-Jun-08

27-Jun-08

29-Jun-08

01-Jul-08

03-Jul-08

05-Jul-08

07-Jul-08

Day

Kd (m-1

)

0 5 10

15

20

25

Turbidity (NTU)

Mea

n L

ight A

tten

uatio

nP

OM

C T

urb

idity

M

ud

Isla

nd

s S

outh

Ea

st

0

0.5 1

1.5 2

2.5 3

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

23-Jun-08

25-Jun-08

27-Jun-08

29-Jun-08

01-Jul-08

03-Jul-08

05-Jul-08

07-Jul-08

Da

y

Kd (m-1

)

0 5 10

15

20

25

Turbidity (NTU)

Me

an L

ight A

ttenua

tion

PO

MC

Turb

idity

Fig

ure

7. M

ea

n (±

se) lig

ht a

tten

ua

tion

coe

fficien

ts (m-1), calcu

late

d d

aily b

etw

een 1

0a

m a

nd

2p

m, a

nd

tu

rbid

ity a

t Mu

d Isla

nd

s (no

te: lo

gg

er d

ata

for th

e n

orth

we

st pile

afte

r 7 Ju

ne

ma

y h

av

e b

ee

n a

ffecte

d

by

fou

ling

of th

e w

ipe

r syste

m, a

nd

log

ge

rs at th

e so

uth

ea

st pile

failed

to re

cord

usa

ble d

ata

for m

ost

of th

e d

ep

loy

me

nt p

erio

d).

12

Sw

an B

ay

0 1 2 3 4 5 6 7

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

23-Jun-08

25-Jun-08

27-Jun-08

29-Jun-08

01-Jul-08

03-Jul-08

05-Jul-08

07-Jul-08

09-Jul-08

11-Jul-08

13-Jul-08

15-Jul-08

17-Jul-08

Da

y

Kd (m-1

)M

ea

n L

ight A

ttenua

tion

Fig

ure

8. M

ea

n (±

se) lig

ht a

tten

ua

tion

coe

fficien

ts, calcula

ted

da

ily b

etw

ee

n 1

0am a

nd

2p

m, a

t Sw

an

B

ay

(no

te: d

iffere

nt Y

axis sca

le to

oth

er lig

ht a

tten

uatio

n fig

ure

s).

Kirk

Po

int

0

0.5 1

1.5 2

2.5 3

16-Apr-08

18-Apr-08

20-Apr-08

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

Day

Kd (m-1

)

Me

an L

ight A

ttenua

tion

Fig

ure

9. M

ea

n (±

se) lig

ht a

tten

ua

tion

coe

fficien

ts (m-1), calcu

late

d d

aily b

etw

een 1

0a

m a

nd

2p

m, a

t K

irk P

oin

t.

13

Bla

irgo

wrie

Sp

ee

d R

estric

tion P

ile (n

ea

r DQ

plo

t)

0

0.5 1

1.5 2

2.5 3

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

23-Jun-08

25-Jun-08

27-Jun-08

29-Jun-08

01-Jul-08

03-Jul-08

05-Jul-08

07-Jul-08

Da

y

Kd (m-1

)

0 5 10

15

20

25

Turbidity (NTU)

Mea

n L

ight A

ttenua

tion

PO

MC

Turb

idity

S

orre

nto

Cha

nne

l No

. 10

(oth

er s

ide

of c

ha

nne

l to B

lairg

ow

rie D

Q p

lot)

0

0.5 1

1.5 2

2.5 3

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

23-Jun-08

25-Jun-08

27-Jun-08

29-Jun-08

01-Jul-08

03-Jul-08

05-Jul-08

07-Jul-08

Da

y

Kd (m-1

)

0 5 10

15

20

25

Turbidity (NTU)

Me

an L

ight A

ttenua

tion

PO

MC

Turb

idity

Fig

ure

10

. Me

an

(± se) lig

ht a

tten

ua

tion

coe

fficien

ts (m-1), ca

lcula

ted

daily

be

twe

en

10a

m a

nd

2p

m, a

nd

tu

rbid

ity a

t Bla

irgo

wrie

.

14

Po

int R

icha

rds (B

ella

rine

Ba

nk)

0

0.5 1

1.5 2

2.5 3

16-Apr-08

18-Apr-08

20-Apr-08

22-Apr-08

24-Apr-08

26-Apr-08

28-Apr-08

30-Apr-08

02-May-08

04-May-08

06-May-08

08-May-08

10-May-08

12-May-08

14-May-08

16-May-08

18-May-08

20-May-08

22-May-08

24-May-08

26-May-08

28-May-08

30-May-08

01-Jun-08

03-Jun-08

05-Jun-08

07-Jun-08

09-Jun-08

11-Jun-08

13-Jun-08

15-Jun-08

17-Jun-08

19-Jun-08

21-Jun-08

Da

y

Kd (m-1

)M

ea

n L

ight A

ttenua

tion

Fig

ure

11

. Me

an

(± se) lig

ht a

tten

ua

tion

coe

fficien

ts (m-1), ca

lcula

ted

daily

be

twe

en

10a

m a

nd

2p

m, a

t P

oin

t Rich

ard

s.

15

Epiphytes During the reporting period (April/May 2008), epiphytic algal cover on subtidal H. nigricaulis plots varied between regions and depths (Figure 12). For most regions, turfing and encrusting epiphytic algae covered <10% of leaf area and macroalgae covered <20% of the quadrat.

Encrustring coraline algae covered >50% and >20% respectively of the seagrass leaf area at the shallow Swan Bay and Mud Islands plots. Epiphytic turfing algae covered <5% of leaf area at all regions except the shallow Mud Islands region. Epiphytic macroalgae covered >30% of the quadrat at the shallow Mud Islands plot (Figure 13).

Epiphytic algal cover on Z. muelleri plants surveyed in the intertidal plots was less than 3% of leaf area or quadrat area (Figures 14 and 15).

Other Factors Drift Algae During the reporting period (April/May 2008), drift macroalgae covered >30% of the quadrat area at the shallow Swan Bay plot (Figure 16). Elsewhere drift macroalgae comprised <10% of subtidal plots and <2% of intertidal plots sampled.

Other Epiphytes Non-algal epiphytes such as the encrusting bivalve E. georgiana were patchily distributed, contributing only a small percentage of the epiphytic cover on seagrass in PPB (data not shown).

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

20

40

60

80

% c

over

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

20

40

60

80

DeepShallow

Epiphytic turf algal cover Encrusting algal cover

Figure 12. Mean (± se) epiphytic turfing and encrusting algal cover (%) of H. nigricaulis leaf area at shallow and deep subtidal plots in PPB during the reporting period (April/May 2008).

16

Blairgowrie

Mud Is

lands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

20

40

60

80

% c

over

DeepShallow

Figure 13. Mean (± se) epiphytic macroalgal cover (%) of quadrats sampled at shallow and deep subtidal plots in PPB during the reporting period (April/May 2008).

10.0 10.0Mud Is

lands

Swan Bay

St Leonard

s

Pt Richard

s0.0

2.5

5.0

7.5

10.0

% c

over

Mud Islands

Swan Bay

St Leonard

s

Pt Rich

ards

0.0

2.5

5.0

7.5

10.0Turfing algal cover Encrusting algal cover

Figure 14. Mean (± se) turfing and encrusting epiphytic algal cover (%) of Z. muelleri shoot area at intertidal plots in PPB during the reporting period (April/May 2008).

17

Mud Islands

Swan Bay

St Leonard

s

Pt Rich

ards

0.0

2.5

5.0

7.5

10.0

% c

ove

r

Figure 15. Mean (± se) epiphytic macroalgal cover (%) at intertidal plots in PPB during the reporting period (April/May 2008).

Blairgowrie

Mud Islands

Swan Bay

St Leonard

s

Pt Rich

ards

Kirk P

oint0

20

40

60

80

% c

over

DeepShallow

M

ud Islands

Swan Bay

St Leonard

s

Pt Rich

ards

0.0

2.5

5.0

7.5

10.0

% c

over

Figure 16. Mean (± se) cover (%) of drift macroalgae recorded for subtidal and intertidal plots in PPB during the reporting period (April/May 2008).

Subtidal Intertidal

18

Comparisons against historical data Seagrass health The historical data indicates that subtidal seagrass cover, seagrass length and shooting stem density were higher at Kirk Point and Point Richards prior to April 2008 (Figure 17). The shallow plot at Point Richards exhibited a long-term trend in seagrass abundance rather than short-term seasonal patterns (i.e. a visible contrast between spring and autumn samples). Seagrass cover has been declining at the Point Richards plot since November 2006 (Figure 17), whilst most of the seagrass loss at the Kirk Point plot occurred between April 2007 and April 2008.

Intertidal seagrass was present at only two of the regions where historical data was available: Point Richards and Swan Bay. When the intertidal plot in Swan Bay was established in January 2005 this plot was dominated by Z. muelleri, but since this date Z. muelleri cover has declined and L. marina cover has increased to cover almost all the benthos (Figure 18). Z. muelleri cover at Point

Richards increased between April 2007 and April 2008 following two years of decline (Figure 18).

There are some inherent difficulties in comparing data collected from different studies using different methods. There were a number of differences between the methods used in the current project and previous studies (Ball et al. 2006). These include the previous use by Ball et al. (2006) of random rather than fixed quadrats, fewer replicates (n = 5 versus 12) and the use of destructive core sampling to estimate shoot/stem densities. In relation to the latter, appendix 2 indicates that the current visual assessment methods are likely to underestimate shoot/stem counts by up to 20%. These differences are likely to influence the variances to a greater extent than the means (and hence the trends observed).

Seagrass epiphyte cover Epiphyte cover varied over time in the subtidal seagrass plots (Figure 19). In April 2007 macroalgal cover was > 40% at Kirk Point, and was <10% in April 2008. Turfing algal cover was >80% at Kirk Point in April 2005 compared with <10% in subsequent years.

19

Nov 04

Apr 05

Nov 05

May 0

6

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% s

ea

gr a

ss c

ove

r

n /a

Kirk Pt

Nov 04

Apr 05

Nov 05

May 0

6

Nov 06

Apr 07

Apr 08

0

20

40

60

80

se

ag

rass le

ngth

(cm

)

Nov 04

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

100

200

300

400

No

. ste

ms 0

.06

25

m-2

n/a

Nov 04

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% s

ea

gra

ss c

ove

r

B

Nov 04

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

15

30

45

60

se

ag

rass le

ngth

(cm

)

Nov 04

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

100

200

300

400

No

. ste

ms 0

.06

25

m-2

n/a n/a

Figure 17. Mean (±se) cover (%), seagrass length and stem density for H. nigricaulis seagrass at Kirk Point and Point Richards between November 2004 and April 2007 (Ball et al. in prep.) and the first Baywide seagrass monitoring field assessment in April 2008 (depicted in grey); n/a indicates where no data were available

Pt Richards

n/a n/a

n/a

20

Nov 04

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0

25

50

75

100%

se

ag

rass c

ove

r

n/a n /a

Pt Richards

Nov 04

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0

5

10

15

20

se

ag

rass le

ngth

(cm

)

n /a

Nov 04

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0

200

400

600

800

No

. sh

oo

ts 0

.06

25

m-2

n/a

Jan 05

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0

20

40

60

80

100

% s

ea

gra

ss c

ove

r

B

n /a

Jan 05

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0.0

3.5

7.0

10.5

14.0

17.5se

ag

rass le

ngth

(cm

)

n /a

Jan 05

Apr 05

Nov 05

Apr 06

Nov 06

Apr 07

May 08

0

500

1000

1500

No

. sh

oo

ts 0

.06

25

m-2

ZosteraLepilaena

n/a n /a n /a

Figure 18. Mean (±se) cover (%), shoot length (cm) and density for intertidal seagrasses at Point Richards and Swan Bay between November 2004 and April 2007 (Ball et al. in prep) and the first Baywide seagrass field assessment in May 2008 (grey shading for Z. muelleri and no shading for L. marina); n/a indicates where no data were available

Swan Bay

21

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% t

urf

ing

alg

al co

ve

r

Kirk Pt

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% e

ncru

stin

g a

lga

l co

ve

rApr 0

5

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% e

pip

hytic m

acro

alg

al co

ve

r

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% t

urf

ing

alg

al co

ve

r

Pt Richards

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% e

ncru

stin

g a

lga

l co

ve

r

Apr 05

Nov 05

May 06

Nov 06

Apr 07

Apr 08

0

25

50

75

100

% e

pip

hytic m

acro

alg

al co

ve

r

Figure 19. Mean (±se) turfing, encrusting and macroalgal epiphytic cover (%) for subtidal H. nigricaulis at Kirk Point and Point Richards between April 2005 and April 2007 (Ball et al. in prep.) and the first Baywide seagrass field assessment in April 2008 (depicted in grey).

22

Discussion

The majority of seagrass in PPB is found in Corio Bay, along the southern shore of the Geelong Arm and in Swan Bay (Blake and Ball 2001). More than 95% of this is H. nigricaulis and Z. muelleri. Another species, Amphibolis antarctica, is restricted to more exposed sediments around PPB Heads.

The combined area of H. nigricaulis and Z. muelleri estimated from aerial photos was 59 km2 in 2000 (Blake and Ball 2001). Previous seagrass mapping projects indicate there has been an overall decline in seagrass abundance since the 1960s. At a finer spatial scale this pattern varied. Blake and Ball (2001) examined change in seagrass cover in four areas where good historical aerial photo records were available. One area showed a small decrease since the 1940s, two areas a probable increase since the 1950s, and one area an overall increase since the 1960s.

The aims of this study were to: 1) map seagrass extent within areas

representative of seagrass in PPB using aerial photos

2) monitor seagrass health, quarterly, at a finer spatial scale within each of these areas

3) monitor the likely drivers of seagrass change by collecting information on light, turbidity and epiphytes where possible to link observed changes in seagrass to these explanatory variables

4) contrast current seagrass extent and health with historical information where available, to compare seagrass extent and health with that measured in the past.

This first milestone report presents results pertaining to aims 2–4.

Seagrass cover Seagrass beds are habitats capable of rapid expansion and contraction over relatively short periods. Ball et al. (in prep.) estimated from aerial photographs that seagrass area at Point Richards contracted by 70% between 2000 and 2005. During the same period seagrass at Kirk Point doubled in area, although seagrass cover appears to have recently diminished inshore where the field assessment plot is based.

H. nigricaulis cover at Kirk Point, Point Richards and St Leonards was low compared with plots in the southern part of PPB. Seagrass at these plots was dominated by stems without shoots. Analysis of historical changes between 2004 and April 2008 indicated that cover at the Kirk Point and Point Richards shallow plots had been greater in the past. By comparison, intertidal Z. muelleri cover and density increased at Point Richards in May 2008 following a decline in cover between November 2005 and April 2007.

Historical data were not available for three of the six field assessment regions (Blairgowrie, St Leonards and Mud Islands) considered in this first milestone report, although previous sampling of seagrass at Mud Islands and Blairgowrie was undertaken by Edmunds et al. (2006) as part of the Supplementary Environmental Effects Statement at different plots to those in the present study. The absence of previous information on the health of seagrasses precludes an assessment of the health of seagrass relative to past conditions at these plots. However, factors known to influence seagrass health were directly assessed.

Factors that affect seagrass health Based on evidence from the literature and investigations in PPB, a light requirement of 15% of surface light appeared to be a realistic annual light requirement for Zosteraceae in the south of the Bay (CEE 2007).

The percentage of surface irradiance reaching seagrass plants may be reduced by increases in water column turbidity, phytoplankton blooms and shading from epiphytic algae. Seagrasses are also subject to self-shading (Zimmerman 2006).

All regions, except Swan Bay, recorded mean daily benthic light levels that exceeded 15% of surface irradiance. It appears that the data recorded at Swan Bay was unreliable (Appendix 3).

Swan Bay is a shallow, silty embayment that supports extensive seagrass beds. Although no turbidity data was available for Swan Bay, it would be expected to become periodically turbid. Longmore et al. (2002) recorded Kd < 0.5 m-1 at a depth of 4 m at a site in southern Swan Bay over a 6-month period in 2001.

23

Turbidity levels adjacent to the Blairgowrie, Mud Islands and St Leonards regions were low.

Seagrasses are important sites for attachment of biota, including epiphytic algae and encrusting sessile invertebrates. Epibiota, particularly algae, are important sources of food for mobile invertebrate grazers, which in turn are food for resident fish, and increase the overall habitat complexity of these environments. In high abundance, epiphytic algae may cause excessive shading of seagrass leaves leading to seagrass death. Insufficient data are available to predict at what point epiphyte loads will impair seagrass health in PPB.

Epiphytes were far more abundant on subtidal compared to intertidal seagrass plants. Epiphytic algal loads on subtidal seagrasses were patchily distributed, but mostly low. Current epiphyte loads are similar to those observed in the past for plots where information was available. Drift algae were abundant at Swan Bay (>30% cover).

Conclusions Seagrass extent in PPB varies spatially and temporally. This reports details the data gathered from the first field season of a program which will be conducted quarterly for the first two years. The report describes seagrass health in PPB in the autumn of 2008. Limited conclusions can be drawn regarding the role various environmental variables play in influencing seagrass health at this stage of the program.

Seagrass cover at Kirk Point and Point Richards in western PPB was low and dominated by stems without shoots. Historical data at Point Richards indicated that seagrass cover had been higher in the past and losses at this plot were part of an ongoing trend for this region. Seagrass loss at the Kirk Point plot appears to be recent, with most of the loss occurring between April 2007 and April 2008.

Subtidal seagrass cover, canopy height and shoot density at Mud Islands and Swan Bay were high compared with plots in the western part of PPB. The absence of historical data prevents an assessment of past trends at these plots.

Benthic light levels exceeded conservative minimum light requirements for seagrass outlined in the CDP Turbidity Detailed Design (PoMC 2008b) at five of six field assessment regions. The validity of the light data at Swan Bay is questionable due movement of the upper logger during the deployment, and the close proximity of the lower logger to the benthos at this region.

Turbidity levels were well within the limits outlined in the CDP Environmental Management Plan (PoMC 2008c).

The results presented in this report indicate that the seagrass cover and extent in autumn 2008 were within the expected variability for PPB.

24

Acknowledgements

FRB staff who participated in the SCUBA diver surveys were David Reilly, Silvana Acevedo, Brent Womersley, Kade Mills and Guy Werner. David Hatton assisted in setting up the light loggers and wiper systems.

Some of the field work was undertaken with the charter vessel ‘Reel Easy’ skippered by Ian Garland.

25

References

Ball, D., Parry, G., Heislers, S., Blake, S., and Werner, G. (in prep.). Analysis of Victorian seagrass health at a multi-regional level 2004-07. NHT Statewide Project No 202243. Department of Primary Industries, Queenscliff, Victoria, Australia.

Blake, S., and Ball, D. (2001). Victorian marine habitat database: Seagrass mapping of Port Phillip Bay, Marine and Freshwater Resources Institute Report no. 36. Marine and Freshwater Resources Institute, Queenscliff.

Cambridge, M.I., Constans, S.A., and Kuo, J. (1983). An unusual method of vegetative propagation in Australian Zosteraceae. Aquatic Botany 15, 201-203.

Carruthers, T.J.B., Longstaff, B.J., Dennison, W.C., Abal, E.G., and Aioi, K. (2001). Measurement of light penetration in relation to seagrass. In Global Seagrass Research Methods, (Eds Short FT, Coles RG) pp. 370-392. (Elsevier Science).

CEE. (2007) PPB Channel Deepening Project Supplementary Environmental Effects Statement, Technical Appendix 50 Overview impact assessment. Seagrass. Consulting Environmental Engineers, Port of Melbourne Corporation, Melbourne.

Edmunds, M., Pickett, P., and Stewart, K. (2006). PPB Channel Deepening Project Supplementary Environmental Effects Statement – Marine Ecology Specialist Studies. Volume 6 Seagrass Beds, Australian Marine Ecology Report No. 354, Melbourne.

Harris et al. (1996) Port Phillip Bay Environmental Study Final Report. CSIRO, Canberra, Australia.

Longmore, A.R., Cowdell, R.A., and Flint, R. (1996). Nutrient status of the water in Port Phillip Bay. Technical Report No. 24, CSIRO Port Phillip Bay Environmental Study, CSIRO Canberra.

Longmore, A.R., Nicholson, G.J., and Abbott, B. (2002). Causes of seagrass loss in Swan bay and prospects for recovery. Internal Report No. 34, Marine and Freshwater Resources Institute, Queenscliff.

Parry, G.D., Heislers, S., Chan, K., Cleaver, J., and Werner, G.F. (2005). Seasonality of flowering and fruiting of the seagrass Zostera muelleri and Zostera (formerly Heterozostera) tasmanica. Primary Industries Research Victoria, Queenscliff.

PoMC (2008a). Seagrass Monitoring Program – Detailed Design - CDP_ENV_MD_022 Rev 2. Port of Melbourne Corporation, Melbourne.

POMC (2008b). Turbidity Detailed Design, CDP_ENV_MD_024 Rev 2, Port of Melbourne Corporation, Melbourne.

POMC (2008c). Channel Deepening Project Environmental Management Plan CDP_IMS_PL_OO4 Rev4, POMC, Melbourne.

Zimmerman, R.C. (2006). Light and photosynthesis in seagrass meadows. In ‘Seagrasses: Biology, Ecology and Conservation’. (Eds AWD Larkum, RJ Orth, CM Duarte) pp. 303-321. (Springer, Netherlands).

26

Appendix 1

Electronic files are as follows:

• Seagrass health observations at plots and quadrats: CDP_Seagrass_database.xls

• Intertidal seagrass upper limit boundaries: CDP_Seagrass_UL_MGA55.shp

• Light logger data: Logger_data_April-July08.xls

.

27

Appendix 2

Quality assurance/Quality control procedures

Seagrass health observations in the field The quality of the raw data generated by the field monitoring component of the seagrass monitoring program is influenced by two factors: 1) accuracy and 2) precision of measurements in the field. Accuracy is the degree of closeness of the measured value to the actual (true) value. Precision is the degree to which repeated measurements display the same or similar results (in this case between divers). Accuracy may be an issue for the Baywide Seagrass Monitoring Program because visual assessments are used in the field. Photo quadrats or destructive sampling are likely to be more accurate but require additional and more time-consuming processing in the lab. Precision is also likely to be an important issue for this project as a pool of divers (>5) was used for the seagrass fieldwork and therefore some variation between diver estimates is expected.

Methods/Results The accuracy and precision of divers in the field was quantified through an exercise specifically designed to assess these parameters. After the first sampling event, a shallow seagrass bed near the St Leonards monitoring plot was sampled using the same quadrat methods applied to the Baywide Seagrass Monitoring Program. Five FRB divers involved in the monitoring program participated in the exercise. At the end of the exercise the seagrass in each quadrat was physically excavated and a shoot count was undertaken in the laboratory

Precision between divers was quantified by comparing estimates of % cover, shoot height and shoot density for H. nigricaulis across the divers at each of 10 quadrats in the field. A photograph of each quadrat was taken and accuracy was assessed for % cover, by comparing diver estimates with cover estimates generated using photo point quadrat analysis (Coral Point software), and for shoot density, by comparing diver estimates with actual counts of shoots processed in the laboratory.

Precision between divers was examined by computing the standard deviation (SD), % coefficient of variation (SD/mean) and range (max.–min.) for each quadrat (n = 5 divers). The means of these parameters are shown for % cover, shoot height and density in Table 3. The mean coefficient of variation (CV) was less than 12% for % cover, shoot height and shoot density. That is the standard deviation of estimates was on average within 12% of the mean. CVs of about 10% are quite acceptable given the inherent variability expected between five different divers in the field. During monitoring, where only two divers are operating at any one time, this variation is likely to be less. This is supported by the finding that differences between pairs of divers calculated at random were often less than half the range of values encountered by five divers collectively (Table 3).

Table 3. Diver precision: Mean standard deviation (SD), % coefficient of variation (CV), range of diver observations and differences between diver pairs (n = 5 divers) for the QC/QA exercise (n = 10 quadrats).

Seagrass variable % cover

shoot ht (cm)

shoot density

SD (obs) 5.1 2.2 4.0

% CV (obs) 12.2 11.7 11.1

Range (obs) 13.2 5.4 10.0

Diver pairs (mean diff.) 5.9 2.3 4.7

This exercise also found that divers underestimated % cover and shoot density by on average 11% and 18% respectively (Table 4), but with consistent bias in terms of estimation. Most sampling procedures have some bias, requiring calibration, and it is important to quantify the direction and magnitude of this bias.

28

Table 4. Diver accuracy and bias for the QA/QC exercise (n = 10 quadrats).

Seagrass variable % cover

shoot density

Accuracy (mean(obs)-actual value*) -7.5 -8.5

% bias (accuracy/actual value*100) -11.2 -17.8

*derived through point quadrat photo analysis for % cover and counts of shoots in the lab for shoot density

Conclusions Overall this exercise indicated that precision between divers was acceptable, but accuracy was low, particularly, for shoot counts. This is not unexpected given the difficulty of this task in the field, particularly where shoot densities

are high and shoots are difficult to distinguish one from another. A major concern for environmental monitoring projects is that the source of variation associated with differences between operators (divers in this case) is minimised (particularly given all the other natural sources of variation encountered in monitoring designs). This can be achieved through the use of standardized sampling protocols and sufficient training of staff in the field.

In addition, precision is routinely assessed in the field by incorporating random duplicate quadrats during seagrass monitoring (generally two per plot). Diver estimates are compared in the field to ensure concordance between divers during field monitoring.

29

Appendix 3

Light logger assessment Inconsistencies in some of the light attenuation data were recorded during this initial deployment of the light loggers. The primary causes of the problems were: • excessive build up of marine fouling on the

loggers and/or wiper systems • failure or ineffective operation of wiper

systems • failure of straps securing the logger/wiper

units to navigation piles.

The following provides a summary of the performance of the light loggers and wiper systems from the first deployment.

Blairgowrie (speed restriction pile) This pile was closest to the deep seagrass monitoring plot at Blairgowrie. The attenuation coefficients from the first month of the April-July deployment matched the pattern of the PoMC turbidity data from nearby Camerons Bight (Figure 10). The loggers failed to record usable data from late May to early June. The attenuation coefficients after this period were more variable and were only partially matched by changes in turbidity and this data is questionable. While both wiper systems were still working when the loggers were retrieved there was fouling evident on the black discs surrounding the light sensors.

Blairgowrie (Sorrento Channel No. 10) This logger was established as a backup to the loggers at the speed restriction pile. The first month of the attenuation coefficients were similar to values at the speed restriction pile, but became more erratic after 22 May (Figure 10). The erratic attenuation coefficients in late May to mid-June were not matched by changes in the turbidity data from Camerons Bight. There was heavy fouling on the wipers when they were retrieved in July which may have impacted the data.

Mud Islands (north west MNP pile) This pile was closest to the shallow seagrass plot at Mud Islands. The attenuation coefficients for the first month of the April-July deployment matched the PoMC turbidity data (Figure 7). The attenuation coefficients become erratic after 8 June with large spikes in the attenuation values without corresponding changes in the PoMC turbidity data. The loggers at this location had clean light sensors when they were retrieved, but

algae had grown on the wiper system which may have impacted the data.

Mud Islands (south east MNP pile) The loggers at this pile were established as a backup to the loggers at the north west pile, and were also closest to the deep seagrass plot at Mud Islands. The first deployment at this region did not commence until 9 May and only recorded usable data until 13 May (Figure 7). Both loggers had heavy fouling when they were retrieved in July, but it was unclear why this pair of loggers only recorded a few days of usable data. The loggers from this region were replaced with new loggers and will not be used in future deployments.

St Leonards (Coles Channel No. 5) The loggers at Coles Channel No. 5 are closest to the main St Leonards 1 deep plot. The attenuation coefficients recorded during the first deployment up to11 May followed a similar pattern to the PoMC turbidity data (Figure 6). No usable data was recorded during 17–30 May. The observed peaks in light attenutation recorded at Coles Channel No. 5 logger after 13 June appear to match pulses in turbidity recorded by the nearby PoMC monitoring station.

St Leonards (Coles Channel No. 3) A secondary deep seagrass plot was set up at the deep seagrass bed being used in the fish in seagrass monitoring program. Light loggers were deployed at Coles Channel No. 3 which is the closest pile to this region and is also close to the location of the PoMC turbidity monitoring station.

The straps securing the loggers at Coles Channel No. 3 came loose during the deployment and the loggers were found rotated approximately 20° away from the north when they were serviced in July. This caused the attenuation values to be unreliable due to an increased amount of shading by the pile. Based on a comparision with the loggers at Coles Channel No. 3, it is likely that the loggers came loose by mid-May. The data from this deployment was omitted from this report.

Swan Bay Light attenuation coefficients in Swan Bay for the May-July deployment varied and were consistently above 1.0 m-1, indicating turbid water (Figure 8). The straps securing the upper logger were found to have come loose when it

30

was retrieved in June. This allowed the logger to rotate away from the north, and this would have led to unreliable light values being recorded at this region.

Swan Bay is shallow (predominantly <2 m) and the maximum depth at the base of the pile for the lower logger was only 1.3 m and was positioned close to the silty sediments at the seabed The dense beds of seagrass in Swan Bay also mean that the lower logger may periodically be affected by drifting rafts of dead seagrass leaves and macroalgae. These two factors may have also impacted the quality of the light data recorded at Swan Bay during May-July.

It therefore appears that the May-July data is unreliable due to movement of the upper logger during the deployment and physical conditions at the logger site.

Kirk Point The Kirk Point attenuation coefficients for April-June varied between about 0.1–0.6 m-1. Some fouling was evident on the black disc surrounding the light sensors indicating that the wiper system had not been able to keep the top of the logger completely clean. This may have caused the variability in the attenuation coefficients, although the PoMC turbidity data from nearby Long Reef was not yet available to compare with the logger data. The poor performance of the wiper systems appeared to be due to the increased height of the lip around the top of the new logger housing which reduced the effectiveness of the brush in keeping the sensor and top of the logger clean.

Point Richards (Bellarine Bank) The attenuation coefficients at Point Richards mostly varied between about 0.1–0.5 m-1 (Figure 10). Light attenuation coefficient values of <0.1 m-1 were recorded between 23 May and 6 June, and these were below expected background light attenuation for clear seawater. The light sensors at this region were clean when they were retrieved in late June, but fouling on the wiper arms may have affected the light readings in the latter half of the deployment period.

Logger improvements Following the first deployment of the light loggers, efforts have been made to improve the reliability of the equipment in future deployments. Changes to the operating procedure for future deployments include reducing the length of deployment times, adding copper tape to the logger housings and wiper ams to inhibit fouling, using an alternative type of straps for securing the loggers to the piles, installing more powerful servo motors in the wiper systems and replacing any loggers that failed to record data. The loggers are also now being visually checked during the seagrass sampling events.