THE EFFECT OF SEWAGE EFFLUENT ON ROCKY REEF FISH IN NEW SOUTH WALES, AUSTRALIA.

Adam Kingsley Smith

A thesis for the degree of Doctor of Philosophy, University of New South Wales, 1998. U N S W 2 1 SEP 201)0 LIBRARY TABLE OF CONTENTS

ACKNOWLEDGEMENTS i

ABSTRACT ii

LIST OF TABLES iv

LIST OF FIGURES xi

CHAPTER 1.

GENERAL INTRODUCTION 1

SEWAGE OUTFALLS IN NSW 1

TEMPERATE ROCKY REEF FISH IN NSW 3

THE IMPACTS OF SEWAGE POLLUTION ON FISH 5

OUTLINE OF RESEARCH 6

PUBLICATION OF RESULTS 7

REFERENCES 8

CHAPTER 2.

IMPACTS OF SEWAGE POLLUTION ON AQUATIC HABITATS,

FISH AND FISHERIES - A REVIEW 17

ABSTRACT 17

INTRODUCTION 18

OVERVIEW OF LARGE-SCALE IMPACTS OF SEWAGE POLLUTION 20

IMPACTS OF SEWAGE DISPOSAL ON AQUATIC HABITATS 20

Physico-chemical impacts on water 21

Soft sediment biota and habitats 22 2

Rock and coral habitats 23

Vegetated biota and habitats 23

IMPACTS OF SEWAGE DISPOSAL ON FISH 25

Mortality of fish 26

Physiological effects on fish 27

Effects on fish communities 28

IMPACTS OF SEWAGE DISPOSAL ON FISHERIES 29

Commercial and recreational fisheries 29

Aquaculture 32

CASE STUDY- SEWAGE OUTFALLS IN NSW, AUSTRALIA 33

Description 33

Monitoring 33

Detection of effects 34

Reporting 34

Management of the environmental health 34

RESEARCH ISSUES 35

Recommended research to detect impacts of sewage pollution 37

Aquatic habitats 37

Fish 37

Fisheries 37

MANAGEMENT ISSUES 38

CONCLUSIONS 40

REFERENCES 41 3

CHAPTER 3.

A BASELINE SURVEY OF FISH AND MACROBENTHIC

ASSEMBLAGES AT TWO SEWAGE OUTFALLS IN THE

ILLAWARRA COASTAL ZONE 56

ABSTRACT 56

INTRODUCTION 57

Ecology of fish and macrobenthos from rocky reefs 57

Impacts of sewage on fish and macrobenthos 59

METHODS 60

Site descriptions 60

Bellambi study 61

Port Kembla study 61

Survey methodology 61

Analysis of fish 63

Analysis of macrobethos 64

RESULTS 64

Fish assemblages in the Bellambi study 64

Fish assemblages in the Port Kembla study 64

Macrobenthic assemblages in the Bellambi study 65

Macrobenthic assemblages in the Port Kembla study 66

DISCUSSION 66

REFERENCES 69 4

CHAPTER 4.

A BEYOND BACI SURVEY OF THE IMPACT OF SEWAGE ON

FISH ASSEMBLAGES AND A SEA URCHIN IN THE HUNTER

COASTAL ZONE. 75

ABSTRACT 75

INTRODUCTION 76

MATERIALS AND METHODS 77

Fish surveys 77

Multivariate analyses 78

Univariate analyses 78

RESULTS 79

Multivariate analyses of fish communities • 79

Univariate analyses of fish communities 80

DISCUSSION 81

Comparison with other studies 82

Management implications 83

CONCLUSIONS 85

REFERENCES 86

CHAPTER 5.

EFFECTS OF THE SHUTDOWN OF A NEARSHORE SEWAGE

OUTFALL IN SYDNEY ON ASSEMBLAGES OF TEMPERATE

REEF FISHES 92

ABSTRACT 92

INTRODUCTION 92 5

METHODS 95

Study designs 95

Fish surveys . 95

Statistical analyses 96

RESULTS 97

Abundance and richness of fish 97

Multivariate analyses of fish assemblages 97

Univariate analyses of fish assemblages and species 98

DISCUSSION 99

Differences between outfall and control locations 99

Comparison to other studies 100

Future research and fisheries management 102

REFERENCES 103

CHAPTER 6.

IMPACTS OF SEWAGE EFFLUENT DISCHARGE ON

ABUNDANCE AND BIOMASS OF FISH ASSEMBLAGES 110

ABSTRACT 110

INTRODUCTION 111

MATERIALS AND METHODS 112

Fish surveys 112

Estimation of average weight of species 113

Multivariate analyses of fish assemblages 114

RESULTS 114

Abundance of fish 114

Biomass of fish 116

DISCUSSION 117 6

Abundance verses biomass measures 117

Indicators of sewage pollution 117

Comparison with other studies 119

Research implications 120

Management implications 121

REFERENCES 121

CHAPTER 7.

CONDITION OF TWO SPECIES OF RESIDENT REEF FISH

EXPOSED TO SEWAGE EFFLUENT DISCHARGE 128

ABSTRACT 128

INTRODUCTION 128

MATERIALS AND METHODS 130

Species, collection and measurements of fish 130

Statistical analyses 131

RESULTS 132

Fish characteristics 132

Condition index 132

Organ indices 133

Long-term trends 133

DISCUSSION 134

Morphometric indices in detecting sewage pollution effects 134

Organ indices in detecting sewage pollution effects 135

Research and management implications 136

REFERENCES 138 7

CHAPTER 8.

EFFECTS OF SEWAGE EFFLUENT DISCHARGE ON THE

ABUNDANCE, CONDITION AND MORTALITY OF HULAFISH,

Trachinops taeniatus (PLESIOPIDAE). 144

ABSTRACT 144

INTRODUCTION 144

METHODS 146

Abundance of T. taeniatus 146

Condition of T. taeniatus 147

Mortality of T. taeniatus 147

RESULTS 148

Abundance of T. taeniatus 148

Condition of T. taeniatus 149

Mortality of T. taeniatus 149

DISCUSSION 150

Abundance of fish impacted by sewage 150

Condition of fish impacted by sewage 150

Mortality of fish impacted by sewage 151

Future research and management 152

REFERENCES 153

CHAPTER 9.

GENERAL CONCLUSIONS 158

SURVEYS OF THE ABUNDANCE OF FISH ASSEMBLAGES 159

CONDITION OF THREE SPECIES OF FISH 161

EXPERIMENTAL MANIPULATION OF ONE SPECIES OF FISH 162 8

A MODEL OF THE IMPACTS OF SEWAGE ON TEMPERATE REEF FISH 163

FUTURE RESEARCH AND MANAGEMENT 165

REFERENCES 168

APPENDICES 172

ACKNOWLEDGEMENTS

Dr Iain Suthers, my supervisor, encouraged me to upgrade from a MSc to a PhD and to focus on the biology of potentially stressed species of fish. His enthusiasm and ideas for additional spatial and temporal replicates and for cajoling me to take study leave from a busy job were instrumental in the completion of this thesis. I would also like to acknowledge my co-supervisor Marcus Lincoln Smith, Director of the Ecology Lab Pty Ltd, he employed me to survey fish adjacent to sewage outfalls and provided the opportunity to use this information to commence my degree. Marcus while extremely busy with his own thesis and business, has provided invaluable discussion and friendship.

I would like to thank my present employer, NSW Fisheries, for granting study leave and providing me with a range of support such as library searches, permits for collecting fish and travel to conferences. I am particularly grateful to Carolyn Bland and Kathy Bown, the librarians at NSW Fisheries. The Sydney Water Board gave permission for me to use the data I collected on fish in the Sydney and lllawarra regions. The Seaworld Research and Development Corporation (Brisbane) provided financial assistance for research on the hulafish. Pat Tully designed the title page and painted the fish illustrations.

Several of my chapters have been written as scientific papers and I would like to acknowledge my co­ authors: Danny Roberts and Penelope Ajani (Environment Protection Authority - Chapter 4), Marcus Lincoln Smith (The Ecology Lab Pty Ltd - Chapter 5), and Iain Suthers (University of NSW - Chapter 8). A number of people provided comments on draft chapters (papers), Dr David Pollard, is particularly thanked for his meticulous attention to detail. I would also like to thank Dr Tony Miskiewicz (Sydney Water), Dr Michael Kingsford (Sydney University), Dr Rodney James (NSW Fisheries), Duncan Leadbitter (Ocean Watch) and Dr Peter Scanes and Martin Krogh (Environment Protection Authority).

A large number of people assisted with field and laboratory work and I would particularly like to thank Marcus Lincoln Smith, Phillip Hawes, Danny Roberts, Penelope Ajani, Ed Castro, Mitch Carpenter and Richard Piola.

I would like to thank my father and mother, David and Bernadette, for developing my lifetime appreciation of the ocean and its fish and facilitating my search for knowledge. Finally, I would like to thank my wife, Elizabeth Lynch, for her encouragement, love and support. ABSTRACT

Over one hundred scientific papers and reports related to the impacts of sewage on aquatic habitats, fish and fisheries are published each year, yet they are often descriptive, specific to an area or issue or suffer from poor experimental design. Research and management of sewage in NSW waters has been extensive but short-term on very large outfalls and limited on the smaller outfalls. Relocation of some shoreline outfalls in Sydney to deepwater appears to have resolved many public concerns, but there is ongoing community dissatisfaction with management of sewage.

Underwater visual surveys of rocky reef fish were conducted at ocean sewage outfalls and multiple control locations within the lllawarra, Hunter and Sydney regions on the central coast of New South

Wales. In the lllawarra region, fish and macrobenthos at controls and two outfalls were surveyed four times over one year. A lack of data from prior to commissioning of the outfalls prevented a rigorous test of the effects of the outfalls. In the Hunter region, however, a Before/After/Control/Impact (Beyond

BAC1) experimental design could be used, providing an optimal design to detect the effects of sewage pollution contrasted against natural spatial and temporal variability in fish and a sea urchin. Both the lllawarra and Hunter studies indicated that sewage has significant impacts on assemblages and populations of rocky reef fish and selected macrobenthos. Generally there were greater abundances of

Cheilodactylus fuscus (Cheilodactylidae), smaller abundances of Trachinops taeniatus (Plesiopidae),

Hypoplectrodes mccullochi (Serranidae) and the sea urchin Centrostephanus rodgersii at outfalls compared to control locations.

A survey of the status of fish populations following the shutdown of a large sewage outfall also indicated that fish assemblages had been significantly different at the outfall compared to cotltrol locations. It also provided information on “recovery” over a 13 month period. For example, large numbers of Acanthopagrus australis (Sparidae) were observed at the outfalls but declined after about 10 weeks. iii

Total biomass of fish was estimated from data at the three regions and showed that sewage outfalls generally contained less fish but a greater biomass than the controls. This pattern was attributed to a greater proportion of large species such as Achoerodus viridis (Labridae), Acanthopagrus australis and

Cheilodactylus fuscus at the outfalls, and a greater proportion of small fish such as Trachurus novaezelandiae (Carangidae) and Trachinops taeniatus at the controls.

Cheilodactylus fuscus, Parma microlepis (Pomacentridae) and Trachinops taeniatus, were consistently more or less abundant adjacent to outfalls compared to controls. These species were collected from outfall and control locations in the Sydney region to investigate impacts on fish condition that could be attributed to sewage pollution. C. fuscus and P. microlepis occurring adjacent to sewage outfalls had a greater proportion of‘unhealthy’ liver tissue and greater gonadosomatic indices compared to controls. A comparison of C. fuscus collected in the 1980s with the present study indicated a 10% increase of liver index. The mean total length of T. taeniatus was consistently smaller from one outfall location. Female

T. taeniatus from outfalls had a larger number of eggs and smaller diameter of eggs than fish from control locations. A caging experiment was conducted, and although mortalities of up to 73% of T. taeniatus per cage were recorded at one outfall, there was no overall significant mortality due to sewage.

The surveys of fish at three regions in NSW suggest several bioindicators that could be used to measure the effects of sewage pollution. A model of the ecological impacts of sewage on rocky reef fish assemblages indicates that species number is resilient to pollution, while abundance is likely to decline with increasing pollution and biomass may decline initially but then may increase (as large individuals dominate). iv

LIST OF TABLES Page. Table 1.1. Characteristics of sewage outfalls investigated in three regions in NSW and previous ecological findings. ADWF - average dry weather flow, N - no effect reported, 1 - indeterminate effect reported, E - effect reported, * indicates the year the outfall was shutdown, na - not applicable. 3

Table 1.2. Some examples of ecological research on species of rocky reef fish in NSW waters. 4

Table 1.3. Chapters, subject and the name of the scientific journal targeted for publication. P - published, S- submitted. 7

Table 2.1. Number of articles in the Aquatic Science and Fisheries Abstracts published between

1988 and 1997 which contain the keywords ‘sew*’ and ‘habitat’ or ‘fish’ or ‘fishery’ and the year. Note: sew* is a search term for words which begin with sew, e.g. sewer, sewerage, sewage. 18

Table 2.2. Summary of the increase and decrease in several parameters as a result of sewage pollution on aquatic habitats, fish and fisheries and examples of a scientific reference which describes this impact. 22

Table 2.3. Overview of ocean sewage outfall characteristics and extent of monitoring that has been conducted in NSW (derived from MHL 1997 and EPA 1996). ML- Megalitres a day dry flow. NA - Not applicable, E - Effect reported, I - Indeterminate effect reported, N - No effect reported. 33

Table 2.4. Examples of rigorous scientific research which have utilised appropriate experimental design to detect the impacts of sewage on aquatic habitats, fish and fishes. B- Before, A- After,

C- Control, I- Impact, Exp- experiment, E - Effect reported, N - No effect reported, na- not applicable. * - pollution gradient design. The number in brackets indicates the outfalls in the study, although these locations were not directly sampled. 36 V

Table 3.1. Volume, treatment and characteristics of Bellambi outfall and Port Kembla outfall

(from EMU (1991b) and GHD (1990a, 1990b). EP = Equivalent Persons, MLD: Million Litres a day. 61

Table 3.2. Numbers and type of fish, percentage cover (%) of alga, reef complexity and numbers of macrobenthos from two sewage outfalls and four control locations in the lllawarra region. M - mobile, C - cryptic, BE - Bellambi, BU - Bulli, SH - Shell Harbour, PK - Port

Kembla, FI - Flinders Island, AR - Atchinsons Rock, O - Outfall, C - Control, * - economically important species, T - Tropical species. 64

Table 3.3. R-statistics (Clarke 1993) from one-way ANOSIM and pairwise comparisons of fish assemblages at Bellambi and Port Kembla. B = Bellambi, BU = Bulli, SH = Shellharbour, PK =

Port Kembla, FI = Flinders Island, AR = Atchinsons Rock. Number of permutations = 5000. ns

= not significant, * = significant at a<0.05, **= significant at a<0.025 - alphas adjusted for multiple comparisons. 64

Table 3.4. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Bellambi, Bulli and Shellharbour. O vs C, outfall versus controls; nt, no test; *, significant (P<0.05); **, highly significant (P<0.01). 64

Table 3.5. Relative mean abundances and cumulative percentage (Cum %) of the five species that were most responsible for differences between fish assemblages in the boulder habitat at the outfall location at Port Kembla and the two control locations at Flinders Island (FI) and

Atchinsons Rock (AR). 64

Table 3.6. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Port Kembla, Flinders Island and Atchinsons Rock. O vs C, outfall versus controls; *, significant (P<0.05); **, highly significant (P<0.01). 65 vi

Table 3.7. Associations between (a) abundance and (b) richness of fishes, and macrobenthos from six sites from the Bellambi study. Significance is based on regression at the P < 0.05 level and P < 0.0086# (Dunn-Bonferroni Procedure to correct for type I error). Statistically significant positive (+ve), negative (-ve) and no significant associations (-) are shown. 66

Table 3.8. Associations between (a) abundance and (b) richness of fishes, and macrobenthos from boulder habitat at three locations from the Port Kembla study. Significance is based on regression at the P < 0.05 level and P < 0.01# (Dunn-Bonferroni Procedure to correct for type I error). Statistically significant positive (+ve), negative (-ve) and no significant associations (-) are shown. 66

Table 4.1. Sampling details for fish at three time periods and twelve times. 77

Table 4.2. Mean abundance of fish at outfall (O) and control (C) locations from three periods: PI

- Period 1 (before commissioning), P2 - Period 2 (immediately post commissioning), P3 - Period

3 (one year post-commissioning),. M - mobile, C - cryptic, t - tropical, * - commercial/recreational importance. 79

Table 4.3. Summary of one-way analysis of similarities (ANOS1M) comparison of fish assemblages at each location and time period. TH - Tomaree Head, PS - Point Stephens, BB -

Boulder Bay, ns - not significant, * significant (P<0.05). 80

Table 4.4a. Relative abundances of the five species that were most responsible for differences between fish assemblages pre and post-commissioning of the sewage outfall. 81

Table 4.4b. Relative abundances of the five species that were most responsible for differences between fish assemblages at the control locations and post-commissioning (Periods 2 and 3) of the sewage outfall. 81 vii

Table 4.5a. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Tomaree Head, Point Stephens and Boulder Bay. OvsC, outfall versus controls; ns, not significant; *, significant (P<0.05); **, highly significant (P<0.01); F- ratios in bold have been calculated after pooling. 81

Table 4.5b. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Tomaree Head, Point Stephens and Boulder Bay. OvsC, outfall versus controls; ns, not significant; *, significant (P<0.05); **, highly significant (P<0.01); F- ratios in bold have been calculated after pooling. 81

Table 5.1. Survey number, start and completion of survey, weeks since North Head cliff-face outfall ceased operation and season. 95

Table 5.2. Total abundance of fish at outfall and control locations in boulder and wall habitats,

M - mobile, C - cryptic, NN - North Head North (outfall), NS - North Head South (outfall), G -

The Gap, D - Dover Heights. 98

Table 5.3. R-statistics (Clarke 1993) from one-way ANOSIM and pairwise comparisons of mobile and cryptic fish at boulder and wall habitats (4 analyses). NN = North Head North outfall, NS = North Head South outfall, G = Gap control, DH = Dover Height Control. Number of permutations = 5000. ns = not significant, R (* = significant at a<0.05), Pairwise comparisons

(**= significant at a<0.016 - alphas adjusted for multiple comparisons). 99

Table 5.4. Relative abundances and cumulative percentage (Cum %) of the five mobile species that were most responsible for differences between fish assemblages in the boulder habitat at the outfall location at North Head North and the two control locations at The Gap and Dover

Heights (DH). 99 viii

Table 5.5. Summary of mean square (m.s) and F ratios from analyses of variance of fish in the boulder habitat, ns - not significant, * p<0.05, ** p<0.01. 100

Table 6.1. Estimated average biomass (g) of individual fish for the 107 species surveyed during summer at three regions on the central coast of NSW in 1991-95. 1 12

Table 6.2. R-statistics (Clarke 1993) from two-way crossed ANOSIM for comparisons of outfall and control locations and three regions for untransformed and double square root transformed data at abundance and biomass levels. ***P <0.001. 117

Table 6.3. Summary of results of SIMPER analyses for abundance and biomass data

(untransformed data). The species contributing the most to the average dissimilarity between outfall and control locations are listed together with the mean, percentage contribution (Cont. %) and cumulative contribution (Cum. %) to the overall dissimilarity. * - indicates changes between important species between untransformed and VV (x+1) analyses. 117

Table 6.4. Summary of results of SIMPER analyses for abundance and biomass data (W (x+1) transformed data). The species contributing the most to the average dissimilarity between outfall and control locations are listed together with the mean, percentage contribution (Cont. %) and cumulative contribution (Cum. %) to the overall dissimilarity. * - indicates changes between important species between untransformed and VV (x+1) analyses. 117

Table 7.1 . Mean (+SE) measurements, condition factor, organ indices and sex ratios for white ear Parma microlepis and red morwong Cheilodactylus fuscus from sewage outfall (n=40) and control locations (n=40) in the Sydney region. 132 IX

Table 7.2. Red morwong Cheilodactylus fuscus. Results of one way analysis (3,76 df) of variance for fork length, weight, Condition, liver health, Hepatosomatic Index, Stomach Weight

Index and Gonadosomatic Index for males (3,33 df) and females (3,37 df).NH = North Head,

RG = Rosa Gully outfall, GB = Gordons Bay, PP = Potters Point outfall. * < 0.05, ** < 0.01

(untransformed data), df = degrees of freedom. 133

Table 7.3. White ear Parma microlepis. Results of one way analysis (3,76 df) of variance table for fork length, weight, Condition, liver health, Hepatosomatic Index, Stomach Weight Index and Gonadosomatic Index for males (3,38 df) and females (3,32 df). NH = North Head, RG =

Rosa Gully, GB = Gordons Bay, PP = Potters Point. * < 0.05, ** < 0.01 (untransformed data). dof = degrees of freedom. 133

Table 7.4. R-statistics (Clarke 1993) from one-way ANOSIM and pairwise comparisons of four sites for morphometric and organ indices of male and females of two species offish, red morwong Cheilodactylus fuscus and white ear Parma microlepis (8 analyses). NH = North Head control, RG = Rosa Gully outfall, GB = Gordons Bay control, PP = Potter Point outfall. Data was standardised. Number of permutations = 5000. ns = not significant, R (* = significant at oc<0.05), Pairwise comparisons (**= significant at cc<0.008 - alphas adjusted for multiple comparisons). 134

Table 7.5. Mean (+ SE) (+ SD for 1989 data) fork length, weight, liver weight and liver index of red morwong Cheilodactylus fuscus from this study (1996) compared to fish captured at the same sites during 1989 (Andrijanic 1991) and 1987 (Lincoln Smith and Mann 1989b). na- not applicable as fish not sampled. 134

Table 8.1. Summary of Beyond BACI analysis of variance of the abundance of Trachinops taeniatus from the Hunter region at 1 outfall and 2 control locations, 3 periods (before, immediately after, 1 year after - 4 times in each period) and with 4 replicates, ns = p> 0.05; *=p<

0.05; **=p< 0.01; ***=p< 0.001. 149 X

Table 8.2. Summary of analysis of variance of the abundance of Trachinops taeniatus from the

Sydney region at 2 outfall and 2 control locations, 2 times and with 4 replicates, ns = p> 0.05;

*=p< 0.05; **=p< 0.01; ***=p< 0.001. 149

Table 8.3. Summary of analysis of variance of the abundance of Trachinops taeniatus from the lllawarra region at 1 outfall and 2 control locations, 4 times and with 4 replicates, ns = p> 0.05;

*=p< 0.05; **=p< 0.01; ***=p< 0.001. 149

Table 8.4. Summary of analysis of variance of condition (length and reproduction) indices for

Trachinops taeniatus at 2 outfall and 2 control locations in the Sydney region, ns = p> 0.05;

*=p< 0.05; **=p< 0.01; ***=p< 0.001. 150

Table 8.5. Number of Trachinops taeniatus (out of a total of 15 per cage) which survived experimental caging at control and outfall locations in the Sydney region (x = missing cage). 150

Table 8.6. Analysis of variance of the survivorship of Trachinops taeniatus in cages. O - Outfall sites (Potter Point and Rosa Gully), C - Control sites (North Head and Gordons Bay). 150

Table 9.1. Percentages change in species, abundance and biomass (kg) of fish assemblages from three regions and three categories of pollution. ‘Unpolluted’ refers to control locations. ‘Slightly polluted’ refers to outfall locations in the lllawarra and Hunter regions and control locations in the Sydney region. ‘Polluted’ refers to outfall locations in the Sydney region. S - Species, A-

Abundance, B - Biomass. All times are in summer; time 12 is one year after the Hunter outfall was turned on, time 1 is immediately after the Sydney outfall was shutdown. * All numbers are means except for the single outfall location in the Hunter region. 164 xi

LIST OF FIGURES Page.

Figure 1.1. Location of sewage outfalls in the Hunter, Sydney and Illawarra regions that were investigated. 2

Figure 2.1. General form of longitudinal downstream changes in some physico-chemical

parameters following discharge (outfall) of organic material (from Bayly and Williams 1973). 21

Figure 2.2. Water quality conditions along the Illinois River, 1913-1965. Pollutional conditions indicate a predominance of pollution tolerant aquatic life (Colten 1992). 22

Figure 2.3. Generic models for change in benthic communities in response to sewage disposal

(from Pearson and Rosenberg 1978). 23

Figure 2.4. Hypothesised shifts in primary production with eutrophication status and nutrient loading rates from coastal plain estuaries such as Chesapeake Bay (from Boynton 1997). 24

Figure 2.5. Seagrass loss in Cockburn Sound, Australia. Each map shows Cockbum Sound surrounded by the coast of the mainland to the right, and Garden Island to the left. The 10m contour line is indicated. The shading shows the area of seagrass present at different times (from

Cambridge and McComb 1984). 25

Figure 2.6. Generalised responses of fish to chronic environmental stress. This figure emphasises the acute response resulting in direct mortality of fish, direct and indirect responses on fish physiology (toxicological and energy availability) and community levels (Adams 1990). 26

Figure 2.7. Summary of the relationship between increasing eutrophy and yields of coregonids, percids and cyprinids from 17 European lakes (from Leach et al., 1977). 31 xii

Figure 3.1. Sampling locations for fish and macrobenthos at the Bellambi sewage outfall (•) and control locations (O) and the Port Kembla sewage outfall (■) and controls (□) in the lllawarra region. 60

Figure 3.2. MDS plots for the abundance of fish at Bellambi sewage outfall and control locations

(stress = 0.06) (top) • - Bellambi, A - Bulli, □ - Shell Harbour, and Port Kembla sewage outfall and control locations (stress = 0.1) (bottom). • - Port Kembla, A - Flinders Island, □ - Atchinsons

Rock. 64

Figure 3.3. Mean abundance (+ 1 SE) of fishes from the Bellambi study (a) abundance of mobile fish, (b) abundance of cryptic fish, (c) Trachinops taeniatus, (d) Cheilodactylus fuscus, (e) Parma unifasciata, (f) Norfolkia clarkei. □ - Bellambi, ♦ - Bulli, - Shellharbour. 64

Figure 3.4. Mean richness or abundance (+ 1 SE) of fishes from the Port Kembla study (a) richness of cryptic fishes, (b) Crinodus lophodon (c) Cheilodactylus fuscus, (d) Hypoplectrodes mccu/lochi, (e) Trachinops taeniatus. □ - Port Kembla, ♦ - Controls. 65

Figure 3.5. Mean abundance of macrobenthos from the Bellambi study (a) Ecklonia radiata, (b) encrusting algae, (c) Caulerpa filiformis, (d) Centrostephanus rodgersii, (e) Heliocidaris tuberculata, (f) Australium tentoriforme. □ - Bellambi, ♦ - Bulli, <£> - Shellharbour. 66

Figure 3.6. Mean abundance of macrobenthos from the Port Kembla study (a) Ecklonia radiata,

(b) Centrostephanus rodgersii, (c) Heliocidaris erythrogramma, (d) Australium tentoriforme. □ -

Port Kembla, ♦ - Flinders Island, 0 - Atchinsons Rock. 66

Figure 4.1. The study locations in the Hunter region, NSW, Australia. Boulder Bay is the impacted location with the sewage outfall. 11' xiii

Figure 4.2. Horizontal visibility at the three locations over time. • - Controls, • -. Outfall. Refer to Table 4.1 for relationship between time and periods. 79

Figure 4.3. MDS plot for the abundance of species at each location and period (stress = 0.18).

• - precommissioning at outfall, • - post-commissioning at outfall, • - Tomaree Head, <-Point

Stephens. Note: symbol size increases from Period 1 to 2 to 3. 80

Figure 4.4. Mean abundance or species richness of derived variates from outfall and control locations: (a) total abundance, (b) total number of species, (c) mobile species abundance and (d) cryptic species abundance from outfall and control locations. • - Controls, • -. Outfall. Refer to

Table 4.1 for relationship between time and periods. 81

Figure 4.5. Mean abundance of individual species from outfall and control locations, (a)

Trachinops taeniatus, (b) Atypichthys strigatus, (c) Cheilodactylus fuscus, (d) Ophthalmolepis lineo/atus, (e) Hypoplectrodes mccullochi, (f) Parma microlepis, (g) Pempheris compressus, and

(h) Centrostephanus rodgersii. • - Controls, • -. Outfall. Refer to Table 4.1 for relationship between time and periods. 81

Figure 5.1. The study locations in the Sydney region, NSW, Australia. NHN - North Head North and NHS - North Head South are the sewage outfall locations. The Gap and DH - Dover Heights are the control locations. 95

Figure 5.2. MDS plots for the abundance of mobile and cryptic species at boulder and wall habitats (a) mobile fish in boulder habitat (b) mobile fish in wall habitat, (c) cryptic fish in boulder habitat, and (d) cryptic fish in wall habitat. O - North Head North, • - North Head South, A - The

Gap, □ - Dover Heights. 99 xiv

Figure 5.3. MDS plots for the abundance of mobile species at the boulder habitat over time. O -

sewage outfall locations, □ - control locations. The size of the symbols increases over time from

Times 1 to 5 (see Table 5.1). 99

Figure 5.4. Mean abundance from outfall and control locations: (a) total abundance of cryptic fish,

(b) species richness of cryptic fish, (c) Trachinops taeniatus, (d) Pempheris multiradiata, (e)

Parma unifasciata, and (f) Cheilodactylus fuscus. 100

Figure 5.5. Mean abundance from outfall and control locations of (a) total abundance of mobile

fish, (b) species richness of mobile fish, (c) Trachurus novaezelandiae, (d) Atypichthys strigatus,

(e) Acanthopagrus australis (f) Achoerodus viridis and (g) Hypoplectrodes mccullochi. Refer to

Table 5.1 for sampling periods. 100

Figure 6.1. Study locations for surveys of reef fish in the (a) Hunter, (b) Sydney and (c) Illawarra

regions. The sewage outfall locations are represented by filled symbols. 112

Figure 6.2. Mean (+ SE) for (a) species abundance, (b) biomass and (c) average weight, of rocky

reef fish assemblages at sewage outfall (dark) and control locations (striped) from three regions (H

= Hunter, S = Sydney, I = Illawarra, o = outfall, c = control) (all samples combined). 114

Figure 6.3. MDS ordinations of rocky reef fish assemblages at sewage outfall and control locations

using untransformed, double square root transformed, logarithmically transformed and binary data at abundance and estimated biomass measures. The Hunter, Sydney and Illawarra regions are

represented by circles, squares and triangles, respectively. The sewage outfall locations are represented by filled symbols. 117 XV

Figure 7.1. Location of sampling sites for collection of white ear Parma microlepis and red

morwong Cheilodactylus fuscus at sewage outfall locations (■) and control locations (O) in the

Sydney region. 130

Figure 7.2. Mean (+SE) length, weight, condition, liver health, hepatosoniatic index, stomach

weight index and gonadosomatic index for (a) white ear Parma microlepis and (b) red morwong

Cheilodactylus fuscus from two control and two outfall sites in the Sydney region. 133

Figure 7.3. Mean (+SE) length, weight, condition, liver health, hepatosomatic index, stomach

weight index and gonadosomatic index for (a) male and female white ear Parma microlepis and

(b) male and female red morwong Cheilodactylus fuscus from two control and two outfall sites in

the Sydney region. 133

Figure 7.4. MDS ordinations of (a) morphometric variables for male red morwong Cheilodactylus fuscus (stress = 0.02), and (b) organ indices for female C. fuscus (stress =0.06 ), at four sites in the

Sydney region. The North Head and Gordons Bay control sites are represented by small and large

circles respectively. The Rosa Gully and Potter Point sewage outfall sites are represented by small

and large squares respectively. Significant differences are represented by filled symbols. 134

Figure 8.1. Study locations for surveys of T. taeniatus in the Hunter, Sydney and Illawarra

regions. The locations in the Sydney region marked with A were also used for the condition and

mortality studies. 146

Figure 8.2. Mean abundance (+ SE) of T. taeniatus (per 120 m2) at outfall and control sites in the

(a) Hunter, (b) Sydney, and (c) Illawarra regions. Sewage outfalls are indicated by solid bars. Pre -

pre-commissioning, post - post-commissioning of the sewage outfall in the Hunter region. 149

Figure 8.3. Length frequency distribution (total length - mm) of T. taeniatus from four locations in

the Sydney region and five times (5 mm increments). Note the scale of the y-axis varies. 150 Figure 8.4. Average total length (+ SE) of T. taeniatus from four locations in the Sydney region and five times. 150

Figure 8.5. Reproductive indices (+ SE) of female T. taeniatus from control and outfall locations in the Sydney region (a) mean Gonadosomatic index, (b) mean abundance of eggs, and (c) mean diameter of eggs. 150

Figure 9.1. Models of types of impacts on biological communities with increasing distance from a source of pollution. Model at top of page is a conceptual model of species, abundance and biomass of macrofauna (from Pearson and Rosenberg 1978). Model at bottom of page graphs data on species (S □), abundance (A O) and biomass (B A) of fish assemblages from three regions (see

Table 9.1 for data). 164 CHAPTER 1 GENERAL INTRODUCTION

CHAPTER 1

GENERAL INTRODUCTION

The most common cause of point source pollution to the world’s waterways is the disposal of sewage

(Dubinsky and Stambler 1996). The degree of impact of sewage pollution varies from area to area, reflecting site conditions and the characteristics of the sewage (GESAMP 1990). The impact of sewage pollution is costly in terms of lost resources and degraded amenity, yet continuing growth of human populations will undoubtedly result in future deterioration of aquatic ecosystems and associated activities, such as fisheries and tourism (Scanes et al. 1995, Jones and Kaly 1996). There are some encouraging reports that sewage pollution is decreasing in some aquatic areas, but the problem is often transferred to other areas e.g from nearshore to offshore waters (Koop and Hutchings 1996). Scientists continue to monitor the impacts of sewage pollution and fish are and often regarded as suitable indicators of environmental change as they are relatively diverse, abundant and sedentary (Stephens et at. 1988, Warwick 1993, Lincoln Smith and Jones 1995). Various sources of stochastic variation can make it difficult to distinguish natural variation from that imposed by human activities (Holbrook et al.

1994).

The following sections research the impact of sewage outfalls in NSW waters with particular emphasis on temperate rocky reef fish in NSW. I outline the four components of research in this study: abundance, biomass, condition and mortality of fish at outfall and control locations.

SEWAGE OUTFALLS IN NSW

The first coastal sewage outfall constructed in NSW was the Bondi outfall in 1889, and currently there are 35 sewage outfalls discharging effluent to the coastal waters of NSW (EPA 1997a, MHL 1997).

There is a range of different types of ocean outfalls in NSW: shorebased, nearshore, offshore and deepwater outfalls. Shorebased outfalls discharge from the shoreline, either above or below low tide PORT STEPHENS • Boulder Bay

SYDNEY

SYDNEY 1 HARBOUR North Head

Rosa Gully

New South Wales

BOTANY BAY

500 1cm

Potter Point

10 ILLAWARRA

• Beilambi

WOLLONGONG

• Port Kembla

Figure 1.1. Location of sewage outfalls in the Hunter, Sydney and lllawarra regions that were investigated. CHAPTER 1 GENERAL INTRODUCTION 2 level, and the shoreline outfalls investigated in this study were located at North Head, Rosa Gully,

Potter Point and Port Kembla (Figure 1.1, Table 1.1). Nearshore outfalls discharge out from the shoreline in water depths of less than 10 metres, such as that investigated at Bellambi (Figure 1.1,

Table 1.1). Offshore outfalls discharge some distance from shore in depths greater than 10 metres, such as the outfall at Boulder Bay (Figure 1.1, Table 1.1). No deepwater outfalls were investigated in this study.

About 82% of the effluent produced in coastal NSW (or about 1270 Megalitres (ML)/day) is discharged to ocean waters. Approximately half the outfalls are very small with an average dry weather flow (AWDF) of less than 5 ML/day (MHL 1997). The Boulder Bay and Rosa Gully outfalls arre small outfalls; Bellambi, Port Kembla and Potter Point are medium sized outfalls with a discharge rate in the range of 15 to 50 ML/day; and, prior to shutdown, North Head was a very large outfall with a discharge rate of 300 ML/day (Table 1.1). The average volume discharged from outfalls may increase substantially (2 to 5 times) during wet weather.

The contents of sewage include organic matter, bacteria, nutrients, oil and grease, chemicals, metals and litter (Scanes et al., 1995). Most sewage is treated to some degree and then discharged through licensed outfalls or dumped as concentrated sludge. Sewage in NSW may be discharged as raw, primary, secondary or tertiary treated (MHL 1997). Raw sewage is untreated. Primary treatment is merely the mechanical removal of solid material. Secondary treatment may result in the removal of up to 90 percent of degradable organic wastes. Tertiary treatment involves chemical removal of nutrients and other contaminants. The sewage outfalls in this study discharged a range of effluents. Rosa Gully outfall discharges raw sewage. Potters Point, Bellambi and Port Kembla outfalls discharge primary treated effluent and North Head discharged primary effluent before it was shutdown. The Boulder Bay outfall discharges secondary treated effluent (Table 1.1). None of the outfalls studied here discharged tertiary treated effluent.

There has been limited research on algae, benthos and fish at North Head, Bellambi and Port Kembla sewage outfalls (Table 1.1). No previous research on these variables has been undertaken at Potter Table 1.1. Characteristics of sewage outfalls investigated in three regions in NSW and previous ecological findings. ADWF - average dry weather flow, N - no effect reported, 1 - indeterminate effect reported, E - effect reported, * indicates the year the outfall was shutdown, na - not applicable.

REGION Year Release ADWF Treatment Ecological Reference Outfall Built Type ML/day Findings HUNTER Boulder Bay 1993 offshore 4.3 secondary E algae Ajani et al. (submitted) E benthos Roberts et al. (1998) E fish Smith etal. (1998) N bioaccum. EPA (1995a)

SYDNEY North Head 1917 shore 300 primary E algae Fairweather (1990) 1990 * I benthos Chapman et al. (1995) N benthos Underwood & Chapman (1996) E fish TEL (1994) E bioaccum. Scanes(1996, 1997) E bioaccum. Krogh & Scanes(1996) E bioaccum. Lincoln Smith & Mann (1989)

Rosa Gully 1930s shore 5.1 raw algae na benthos na E fish Lincoln Smith (1985) bioaccum. na

Potter Point ? shore 46 primary algae na benthos na fish na bioaccum. na ILLAWARRA Bellambi 1950s nearshore 20 primary E algae TEL (1994) E benthos TEL (1994) E benthos EPA (1995b) E fish TEL (1994) E bioaccum. EPA (1995b)

Port Kembla 1950s shoreline 16.8 primary E algae TEL (1994) E benthos TEL (1994) E benthos EPA (1995b) E fish TEL (1994) E bioaccum. EPA (1995b) CHAPTER 1 GENERAL INTRODUCTION 3

Point, and only fish research has been undertaken at Rosa Gully. Most of the previous research on sewage at these outfalls has been published in reports by Government agencies or environmental consultants, although several scientific papers have also been published on the impacts of sewage on biota at North Head and Boulder Bay (Table 1.1). The methodology for research on algae and benthos has varied widely, including the use of transects, photoquadrats, collection of biota from kelp holdfasts etc., which makes comparisons difficult (TEL 1994, EPA 1995a, b, Roberts et al. 1998). The methodology for fish surveys has consistently used underwater visual census and may be comparable between studies. There has also been limited research on the bioaccumulation of contaminants such as metals and organochlorines in aquatic biota at outfalls in the Hunter, Sydney and lllawarra regions

(Table 1.1). The species selected for bioaccumulation studies include fish, molluscs and crustaceans, however, red morwong Cheilodactylus fuscus have been collected in the Sydney and lllawarra regions and Sydney rock oysters Crassostrea commercialism have been deployed in the Sydney and Hunter regions.

Significant effects of sewage have been detected for algae, benthos, fish and/or bioaccumulation at all of the selected outfalls where research has been undertaken to date (Table 1.1). No effect has been detected for bioaccumulation at Boulder Bay or benthos at North Head sewage outfalls (Table 1.1).

TEMPERATE ROCKY REEF FISH IN NSW

There are at least 1000 species of fish in the waters of southern Australia (Hutchins and Swainston

1986, Kuiter 1993). Many are endemic to these waters, although there is a gradual biogeographic change from tropical species in the north to temperate species in the south. Many species are readily identified although some species such as the wrasses (F. Labridae) can differ dramatically in colour and shape between juvenile and adult or between sexes. Some fish are sedentary whilst others are highly mobile and many show complex behaviour and social organisation (Lincoln Smith and Jones

1995).

Most of the research on rocky reef fishes in NSW waters has assessed fish assemblages at impacted and reference locations (e.g. Lincoln Smith 1985, Lincoln Smith et al. 1991, 1992, TEL 1993, 1994, EPA JD H ■8 1.2. Some exam ples o fecological research on species o fro cky re e ffish in NSW wat< -o _o X 60 o o 1 ' ‘ E fo o O *3 £? — g O 0 5 cx O <3 § a 60 c 60 01

n ) -s: U "O x D Q _o -a O O 60 b eg a o c 60 e /5 3 * n n n .s !X •8 1 3 -2 -2 ■2 _o '5b _c o So c/o X X C/0 OO /- ON >N u o 60 o s: cd 60 60 o — £ s

X X X C3 -s :§ '•8 2 3 1 ^ X oo CQ ✓ ^ O o cd eo w <0 > 3 0 o c 60 Q c — w n s -a •2 ■j? _c X T •8 .o D a _cd 5 a ~o /- I § E E cd o o 3 £ 60 >-> C 60 c/5 -- 3

s x _C I x •E U U 'u! 2 •8 •5 H O ON / E E £ a> s 3 5 § £ CJ « £ e/5 o £ c o « c o -- n s ” 5 -2 •§ •o _C U 1 '5 o^ m ON 8 E o e 60 o aj 60 o 0 c cd 60 60 o b £ c/5 3 b c c (U £ -- N X _a> I o •2 D •a DC ■a =<3 gs < ON ON E c 60 o

1996) or fisheries (Lincoln Smith et al. 1989, Steffe et al. 1996). Basic ecological research has only been undertaken on about 10 species of rocky reef fish in NSW waters (Table 1.2). This level of research appears very limited in the context that approximately 300 species of fish are exploited for food and sport in the coastal waters of NSW (Lincoln Smith et al. 1989, Kingsford et al. 1991, NSW

Fisheries 1998).

Fish assemblages may vary according to the physical and biological structure of the habitat (Jones

1988, Holbrook et al. 1994). In NSW waters the depth of water, presence of kelp Ecklonia radiata

(Underwood et al. 1991) and topographic complexity are factors which may influence the distribution and abundance of fish (TEL 1993, 1994; Gillanders 1997). More species and numbers of fish are often associated with more complex habitats (Bodkin 1988, Hair and Bell 1992, Anderson 1994). Factors influencing patterns of distribution and abundance are likely to be species-specific (Jones 1988, Levin

1994) . For reef fish in temperate marine regions, components of local assemblage diversity (i.e. within a reef) such as species richness, total fish density and rank order of abundance can remain relatively constant through long periods (decades) of time (Holbrook et al. 1994). The abundance of some fish, including pomacentrids, cheilodactylids and labrids may remain relatively constant over time (Jones

1988, McCormick 1989, Lincoln Smith and Jones 1995). Unforeseen changes to fish assemblages may also occur due to large-scale processes such as climate change (Holbrook et al. 1997). One of the greatest challenges in the study of reef fishes is to understand the causes of the huge natural variation in abundance among different locations and times (Lincoln Smith et al. 1991, Lincoln Smith and Jones

1995) .

Surveying reef fish in shallow water is commonly undertaken using underwater visual census (Lincoln

Smith 1988, 1989, Sale 1994, Thompson and Mapstone 1997). This technique provides rapid estimates of relative abundance, biomass and length frequency distributions of fish, which are non-destructive, repeatable and fishery-independent (Samoilys 1997).

The disadvantages of this technique include limited survey time, depth restrictions, poor visibility (at night or turbid water), observer bias, restriction to species that are visually obvious, diurnal and not CHAPTER 1 GENERAL INTRODUCTION 5 repulsed by divers, attraction of some species to divers, difficulty in counting cryptic and shy species or in counting and identifying large numbers of species and individuals simultaneously (Jennings and

Polunin 1995, Stanley and Wilson 1995, Samoilys 1997).

The effectiveness of visual surveys can be improved by diver training and by comparison with

destructive surveys (Lincoln Smith 1988, 1989, Samoilys 1997). Bias due to habitat structure can be

reduced by stratified sampling (McCormick and Choat 1987). Alternative techniques for surveying fish

include trawling, netting, cages, long-lining, poisons, explosives, video and acoustics, all of which also

have advantages and disadvantages. Underwater visual census was the technique that was selected for

the quantitative studies of fish assemblages and spears, nets and cages were used for collection of fish

and manipulative studies.

IMPACTS OF SEWAGE POLLUTION ON FISH

The current state of knowledge of the impact of sewage pollution on aquatic habitats, fish and fisheries

from throughout the world is reviewed in Chapter 2. There has been no general review of the

ecological impacts of sewage on aquatic ecosystems since Tsai (1975). Approximately 100 studies are

published each year on the ecological effects of sewage on fish although many are descriptive, specific

to an area or issue or suffer from poor experimental design.

Scientific research of the impacts of sewage on fish has generally focused on demersal or site attached

species which are commercially or recreationally important. In NSW coastal waters the species that are

commonly studied include Cheilodactylus fuscus (Cheilodactylidae) and Pagrus auratus (Sparidae)

(Lincoln Smith and Mann 1989, Andrijanic 1991, Otway et al. 1996, EPA 1996). Fish may exhibit a

range of responses to sewage at a range of different scales including communities, individuals and

organs (Pastorok and Bilyard 1985, Gray 1989, Adams 1990, Adams et al. 1993, EPA 1993, Grigg

1994, Lye et al. 1997). Recent research on the impacts of sewage outfalls on fish in NSW waters has

been published by EPA (1996), Otway et al. (1996) and MHL (1997), however, most of this research

has been on the deepwater outfalls in the Sydney region. CHAPTER 1 GENERAL INTRODUCTION 6

OUTLINE OF RESEARCH

The scientific literature abounds with studies seeking to determine and describe the effects of impact on aquatic communities (Lincoln Smith 1991, Fairweather and Lincoln Smith 1994). The way that researchers assess impact can vary from total reliance on existing information to detailed pre-impact surveys and post-impact monitoring. In this research on sewage outfalls, I report on several studies which were based on post-impact monitoring (because it is difficult to ask authorities to construct or turn-off sewage outfalls for a scientific experiment!) and one study which used the best available

approach of detailed pre-impact surveys and post-impact monitoring. A multidiscipinary approach which investigates ecology, physiology and also experimental manipulation provides the best possible

approach for assessing impacts (Liu and Morton 1998) and this approach was adopted to investigate

the effects of sewage pn rocky reef fish in NSW waters.

The initial focus of this research was to describe the effects of sewage discharge on fish assemblages at

several locations in NSW waters. Three separate studies were completed. The first study in the

Illawarra region investigated the effects of sewage on fish and selected macrobenthos at multiple

control and outfall locations. The study was short-term (one year) and no ‘before’ data were available.

The second study was located in the Hunter region and used a Beyond BACI

(Before/After/Control/Impact) experimental design (Underwood 1991, 1992, 1993, 1994), which

involved surveys of fish at multiple times before and after the sewage outfall was operational. This

provides an optimal design to detect the effects of an impact contrasted against natural spatial and

temporal variability (Underwood 1997), although there are alternatives such as gradient analysis

(Keough and Mapstone 1995). The third study was in the Sydney region and involved a study of fish

assemblages in response to the shutdown of a large shoreline sewage outfall. Fish were monitored at

the outfall and control locations over a 13 month period.

The results of the above studies indicated that sewage may significantly impact fish abundance, and

some common species showed consistently increased or decreased abundances (Chapters 3,4,5,6). In

order to investigate why the abundance changed, the condition of three species was measured from

control and outfall locations (Chapters 7,8). Condition was measured using several morphometric and Table 1.3. Chapters, subject and the name of the scientific journal targeted for publication. P - published, S- submitted.

Chapter Subject Targeted journal

2. Review of impacts of sewage on fish Reviews in Fish Biology and Fisheries S

3. Effects of sewage on fish in the Illawarra region Marine and Freshwater Research

4. Effects of sewage on fish in the Hunter region Marine Environmental Research P

5. Effects of sewage on fish in the Sydney region Marine Pollution Bulletin

6. Effects of sewage on abundance and biomass Australian Journal of Ecology

7. Effects of sewage on fish condition Marine Pollution Bulletin

8. Effect of sewage on hulafish Environmental Pollution S CHAPTER 1 GENERAL INTRODUCTION 7 organ indices such as length, weight, hepatosomatic, gonadosomatic and stomach weight indices, egg number and egg diameter. Sewage outfalls may also increase mortality of fish. Therefore, an experiment placed fish in cages at multiple control and sewage outfall locations to investigate mortality

(Chapter 8).

The general null hypotheses that were tested in this study were:

Study 1: There is no difference in the abundance of fish assemblages at sewage outfall and control locations (Chapters 3, 4 and 5).

Study 2: There is no difference in the biomass of fish assemblages at sewage outfall and control locations (Chapters 6).

Study 3: There is no difference in the condition of Cheilodactylus fuscus, Parma microlepis and

Trachinops taeniatus collected at sewage outfall and control locations (Chapters 7 and 8).

Study 4: There is no difference in the mortality of Trachinops taeniatus deployed in cages at sewage outfall and control locations (Chapter 8).

PUBLICATION OF RESULTS

The structure of this thesis has been designed to facilitate publication of scientific papers. It is widely recognised that scientists must publish the results of scientific research in internationally recognised scientific journals and other publications or academically perish (Moxley 1992, Day 1996). The chapters in this thesis have generally been targeted (and formatted) for publication in specific scientific journals, and in some cases they have already been submitted and/or accepted for publication (Table

1.3).

There are clear linkages between the chapters. The descriptive studies (Chapters 3, 4, 5) are the basis for broader generalisations in Chapter 6, and focus for selecting indicator species for further study of condition in Chapters 7 and 8 and manipulative experiments in Chapter 8. The General Conclusions CHAPTER 1 GENERAL INTRODUCTION 8 attempts to synthesise the research from all chapters into a model and provides recommendations for future research and management.

In addition to the above papers, my contribution to research on sewage has also been presented in reports (Smith et al. 1993, TEL 1993, 1994) while I was employed as an environmental consultant with

The Ecology Lab Pty Ltd. Associated outcomes of my research include scientific papers on the effects of sewage on aquatic biota (Roberts et al. 1998, Smith et al. 1998, Ajani et al. submitted), and scientific papers and articles on temperate reef fish (Henrisson and Smith 1994, Smith 1994, 1995a, b,

1997a, b, Smith and Lugg 1994, Smith and Puckeridge 1994a, b, Pollard et al. 1996, Smith et al. 1996,

Smith and Pollard 1996, 1997, Byrnes and Smith 1997a, b), contributions to the NSW State of the

Environment Report (EPA 1995c, 1997a), submissions to a public inquiry into the management of sewage by-products in the coastal zone (EPA 1997b) and environmental policy (NSW Fisheries 1998).

REFERENCES

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium 8,

Bethesda, Maryland. 191 pp.

Adams, S.M., Brown, A.M., Goede, R.W. (1993). A quantitative health assessment index for rapid

evaluation of fish condition in the field. Trans. Am. Fish. Soc., 122, 63-73.

Ajani, P., Roberts, D., Smith, A., Krogh, M. (submitted). The effect of sewage discharge on two

bioindicators at Port Stephens, New South Wales, Australia. Ecol. App.

Anderson, T.W. (1994). Role of macroalgal structure in the distribution and abundance of a temperate

reef fish. Mar. Ecol. Prog. Ser. 113, 279-290.

Andrew, N.L., Jones, G.P. (1990). Patch formation by a herbivorous fish in a temperate kelp forest.

Oecologia 85, 57-68.

Andrijanic, S. (1991). Organochlorine residues in fish: Bioaccumulation in red morwong

(Cheilodactylus fuscus) in the Sydney region. Progress report. Sydney Water Board. 15 pp.

Bell, J.D. (1979). Observations on the diet of red morwong, Cheilodactylus fuscus Castelnau (Pisces :

Cheilodactylidae). Aust. J. Mar. Freshwater Res. 30, 129-33. CHAPTER 1 GENERAL INTRODUCTION 9

Bell, J.D., Burchmore, J.J., Pollard, D.A. (1980). The food and feeding habits of the rock blackfish,

Girella elevata Macleay (Pisces: Girellidae), from the Sydney region, New South Wales. Aust.

Zool. 20, 391-405.

Bodkin, J.L. (1988). Effects of kelp forest removal on associated fish assemblages in central California.

J. Exp. Mar. Biol. Ecol. 117,227-238.

Byrnes, T., Smith, A. (1997a). Protected marine fish species sighting sheet program - status report

1997. NSW Fisheries Fishnote DF\69. December. 6 pp.

Byrnes, T., Smith, A. (1997b). Protected species sighting sheet program - a status report. New South

Wales. Regional Ripples 4(4), 5.

Chapman, M.G., Underwood, A.J., Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

J. Exp. Mar. Biol. Ecol. 189, 103-122.

Day, A. (1996). How to get research published in journals. Gower, Hampshire, U.K.

Dubinsky, Z., Stambler, N. (1996). Marine pollution and coral reefs. Global Change Biol. 2, 511-526. *

EPA (1993). Deformities and associated sublethal effects in fish exposed to sewage-borne

contaminants literature review. Environment Protection Authority, Sydney, Australia, 93/72.

August.

EPA (1995a). Hunter Environmental Monitoring Program 1992-1994. Environment Protection

Authority, Chatswood, Sydney.

EPA (1995b). Effects of sewage outfalls on marine organisms at Port Kembla and Bellambi, Illawarra

region, NSW. Environment Protection Authority, Chatswood, Sydney.

EPA( 1995c). New South Wales State of the Environment 1995. Environment Protection Authority,

Chatswood, Sydney.

EPA (1996). Sydney Deepwater Outfalls Environmental Monitoring Program. Final Report Series.

Volume 1. Environment Protection Authority, Bankstown, Sydney. 194 pp.

EPA (1997a). New South Wales State of the Environment 1997. Environment Protection Authority,

Chatswood, Sydney. 500 pp.

EPA (1997b). Public Inquiry into the Management of Sewage and Sewage By-products in the NSW

Coastal Zone. Final report prepared by the Environment Protection Authority, Sydney. 299 pp. CHAPTER 1 GENERAL INTRODUCTION 10

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environ. Monit. Assess. 14, 197-210.

Fairweather, P.G., Lincoln Smith, M.P. (1994). The difficulty of assessing environmental impacts

before they have occured: a perspective from Australian consultants, pp 121-120 in Battershil 1,

C.N (ed) Proceedings of the second international temperate reef symposium, 7-10 January 1992,

Auckland, New Zealand, NIWA Marine, Wellington. 252p.

GESAMP (1990). The state of the marine environment. 1MO / FAO / Unesco / WHO / IAEA / UN /

UNEP Joint Group of Experts on the Scientific Aspects of Marine Pollution. UNEP Regional

Seas Reports and Studies No. 115. UNEP.

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 reef. Environ. Biol. Fish.

49,153-173.

Glasby, T.M., Kingsford, M.J. (1994). Atypichthys strigatus (Pisces: Scorpididae): An opportunistic

planktivore that responds to benthic disturbances and cleans other fishes. Aust. J. Ecol. 19, 385-

394.

Gray, J.S. (1989). Ecological theory and marine pollution monitoring. Mar. Poll. Bull. 16, 224-227.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Mar. Ecol. Prog. Ser. 103, 25-34.

Hair, C.A., Bell, J.D. (1992). Effects of enhancing pontoons on abundance of fish: initial experiments

in estuaries. Bull. Mar. Sci. 51, 30-36.

Henrisson, C., Smith, A. (1994). Black rock cod - a protected fish. NSW Fisheries Fishnote DF/37. 2

pp.

Holbrook, S.J., Kingsford, M.J., Schmitt, R.J., Stephens, J.S. (1994). Spatial and temporal patterns in

assemblages of temperate reef fish. Amer. Zool. 34, 463-475.

Holbrook, S.J., Schmitt, R.J., Stephens, J.S. (1997). Changes in an assemblage of temperate reef fishes

associated with a climate shift. Ecol. App. 7, 1299-1310.

Hutchins, B., Swainston, R. (1986). Sea Fishes of Southern Australia. Swainston Publishing, Perth,

180 pp. CHAPTER 1 GENERAL INTRODUCTION 11

Jennings, S., Polunin, N.V.C. (1995). Biased underwater visual census biomass estimates for target-

species in tropical reef fisheries. J. Fish. Biol. 47, 733-736.

Jones, G.P. (1988). Ecology of rocky reef fish of north-eastern New Zealand: a review. N.Z. J. Mar.

Freshwat. Res. 22, 445-462.

Jones, G.P., Kaly, U.L (1996). Criteria for selecting marine organisms in biomonitoring studies, pp.

29-48 In Detecting Ecological Impacts: Concepts and Applications in Coastal Habitats. Schmitt,

R.J., Osenberg, C.W (Eds.). Academic Press, San Diego.

Keough, M.J., Mapstone, B.D. (1995). Protocol for designing marine ecological monitoring programs

associated with BEK mills. National Pulp Mills Research Program Technical Report No. 11.,

Australia. 150 pp.

Kingsford, M.J., Underwood, A.J., Kennedy, S.J. (1991). Humans as predators on rocky reef in New

South Wales, Australia. Mar. Ecol. Prog. Ser. 72, 1-14.

Koop, K., Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Mar. Poll. Bull.

33, 121-123.

Krogh, M., Scanes, P.R. (1996). Organochlorine compound and trace metal contaminants in fish near

Sydney’s ocean outfalls. Mar. Poll. Bull. 33, 213-225.

Kuiter, R. (1993). Coastal fishes of south-eastern Australia. Crawford House Press, Bathurst. 437 pp.

Levin, P.S. (1994). Small-scale recruitment variation in a temperate fish: the roles of macrophytes and

food supply. Environ. Biol. Fish. 40, 271-281.

Lincoln Smith, M.P. (1985). The development and application of visual survey procedures for fish

communities on shallow rocky reefs. MSc thesis, University of Sydney, Australia, 187 pp.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate rocky

reef fish assemblages. Mar. Ecol. Prog. Ser. 43, 223-231.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Env. Biol. Fish. 26, 29-37.

Lincoln Smith, M.P. (1991). Environmental impact assessment: the roles of predicting and monitoring

the extent of impacts. Aust. J. Mar. Freshwater Res. 42, 603-14. CHAPTER 1 GENERAL INTRODUCTION 12

Lincoln Smith, M.P., Bell, J.D., Hair, C.A. (1991). Spatial variation in abundance of recently settled

rocky reef fish in southeastern Australia: implications for detecting change. Mar. Ecol. Prog. Ser.

77, 95-103.

Lincoln Smith, M.P., Hair, C.A., Bell, J.D. (1992). Jervis Bay marine ecological study. Project 4: fish

associated with natural rocky reefs and with artificial breakwaters, final report. Fisheries Research

Institute, Cronulla.

Lincoln Smith, M.P., Bell, J.D., Pollard, D.A., Russell, B.C. (1989). Catch and effort of competition

spearfishermen in south-eastern Australia. Fish. Res. 8, 45-61.

Lincoln Smith, M.P., Jones, G.P. (1995). Fishes of shallow coastal habitats, pp. 240-253 in

Underwood, A.J., Chapman, M.G. (Eds) Coastal marine ecology> of temperate Australia. UNSW

Press, Sydney.

Lincoln Smith, M.P., Mann, R.P. (1989). Bioaccumulation in nearshore marine organisms. II.

Organochlorine compounds in the red morwong Cheilodactylus fuscus, around Sydney’s three major

sewage ocean outfalls. A report prepared by State Pollution Control Commission and the Ecology

Lab Pty Ltd, Sydney.

Liu, J.H. & Morton, B. (1998). The impacts of pollution on the growth, reproduction and population

structure of Hong Kong limpets. Mar. Poll. Bull. 36, 152-158.

Lowry, M.B., Suthers, I.M. (in press). Home range, activity and distribution patterns of a temperate rocky-

reef fish Cheilodactylus fuscus. Mar. Biol.

Lye, C.M., Frid, C.L.J., Gill, M.E., McCormick, D. (1997). Abnormalities in the reproductive health of

flounder Platichthys flesus exposed to effluent from a sewage treatment works. Mar. Poll. Bull.

34, 34-41.

McCormick, M.I., Choat, J.H. (1987). Estimating total abundance of a large temperate-reef fish using

visual strip-transects. Mar. Biol. 96, 469-478.

McLaughlin, R.H., O’Gower, A.K. (1971). Life history and underwater studies of a heterodont shark.

Ecol. Monog. 41, 271-289.

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Report MHL769 by

Manly Hydraulics Laboratory and the Ecology Lab Pty Ltd, Sydney. 65 pp. CHAPTER 1 GENERAL INTRODUCTION 13

Miskiewicz, A.G. (1992). Proceedings of a Bioaccumulation Workshop. Published by the Water Board

and Australian Marine Sciences Association Inc. Sydney. 334 pp.

Moxley, J.M. (1992). Publish, don 7 perish: the scholar's guide to academic writing and publishing.

Westport, USA.

NSW Fisheries.(1998). Policy and guidelines: Aquatic habitat management and fish conservation.

Smith, A.K., Pollard, D.A. (Eds). Sydney. 76 pp.

Otway, N.M, Gray, C.A., Craig, J.R., McVea, T.A., Ling, J.E. (1996). Assessing the impacts of

deepwater sewage outfalls on spatially and temporally variable marine communities. Mar.

Environ. Res. 41,45-71.

Pastorok, R., Bilyard, G.R. (1985). Effects of sewage pollution on coral-reef communities. Mar. Ecol.

Prog. Ser. 21, 175-189.

Pollard, D.A., Lincoln Smith, M., Smith, A. (1996). The biology and conservation status of the grey

nurse shark (Carcharias taurus, Odontaspidae), a protected species in New South Wales, south­

eastern Australia. Aquatic Conserv. 6, 1-20.

Roberts, D.E., Smith, A., Ajani, P., Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACE approach. Mar.

Ecol. Prog. Ser. 163,213-224.

Sale, P. (1994). Ecology of Coral Reef Fish. Academic Press, New York.

Samoilys, M. (Ed.) (1997). Manual for assessing fish stocks on Pacific coral reefs. Department of

Primary Industries, Queensland. 75 pp.

Scanes, P.R. (1996). ‘Oyster Watch’: monitoring trace metal and organochlorine concentrations in

Sydney’s coastal waters. Mar. Poll. Bull. 33, 226-238.

Scanes, P.R. (1997). Uptake and depuration of organochlorine compounds in Sydney rock oysters

(Saccostrea commercialis). Mar. Freshwater Res. 48, 1-6.

Scanes, P.R., Scanes, K., Otway, N.M. (1995). Environmental problems due to disposal of wastes

p297-310 In Underwood, A.J., Chapman, M.G. (Eds) (1995) Coastal marine ecology of

temperate Australia. UNSW Press, Sydney, Australia.

Schroeder, A., Lowry, M., Suthers, I. (1994). Sexual dimorphism in the red morwong (Cheilodactylus

fuscus). Aust. J. Mar. Freshwat. Res. 45, 1173-1180. CHAPTER 1 GENERAL INTRODUCTION 14

Smith, A.K. (1987). Genetic variation and dispersal of the yellowtail kingfish, Seriola lalandi from

New South Wales waters. Honours thesis. University of New South Wales.

Smith, A. (1994). An unusual fish. The silverspot Threpterius maculosus. Australian Freediving and

Spearfishing News. Issue 3. Winter.

Smith, A. (1995a). An unusual fish. The bluefish Girella cyaena. Australian Freediving and

Spearfishing News. Issue 5. Summer.

Smith, A. (1995b). An unusual fish. The pennantfish Alectes ciliaris. Australian Freediving and

Spearfishing News. Issue 6. Autumn

Smith, A. (1997a). Is the eastern hula a potential indicator of sewage pollution? Abstract from 23rd

Annual Workshop of Australian Society for Fish Biology and Fish and Aquatic Resource

Management Association of Australia, Darwin, NT, Australia. 41 pp.

Smith, A. (1997b). Book review. Coastal fishes of south-eastern Australia. Australian Freediving and

Spearfishing News. Issue 13. Winter.

Smith, A.K., Ajani, P.A., Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Mar. Environ. Res. 47, 1-20.

Smith, A.K., Inglis, G., Lincoln Smith, M. (1993). An assessment of impacts on marine flora and fauna

of the proposed Pacific Palms sewerage scheme. Report to Manidis Roberts.

Smith, A., Lugg, A. (1994). Fish fauna. p26-32. In Surface water - New South Wales South East

Region Water Management Strategy. Water Planning for the Future (Kathryn Miller, Ed.). NSW

Water Resources Council. 95 pp.

Smith A., Pollard, D. (1996). The best available information - some case studies from NSW, Australia

of conservation related management responses which impact on recreational fishers. Mar. Pol. 20,

261-267.

Smith A., Pollard, D. (1997). Protection of the great white shark in NSW waters. NSW Fisheries

Fishnote DF/68. 2 pp.

Smith A., Pollard, D., Gillanders, B. (1996). Blue groper in NSW. Fishnote DF/63, NSW Fisheries,

Sydney, 2 pp.

Smith, A., Puckeridge, I. (1994a). Fish profile - mulloway. Australian Freediving and Spearfishing

News. Issue 2. Autumn. CHAPTER 1 GENERAL INTRODUCTION 15

Smith, A., Puckeridge, I. (1994b). Fish profile - kingfish. Australian Freediving and Spearfishing

News. Issue 3. Winter.

Stanley, D.R., Wilson, C.A. (1995). Effects of scuba divers on fish density and target strength

estimates from stationary dual-beam hyroacoustics. Trans. Am. Fish. Soc. 124, 946-949.

Steffe, A.S., Murphy, J.J., Chapman, D.J., Tarlinton, B.E., Gordon, G.N.G., Grinberg, A. (1996). An

assessment of the impact of offshore recreational fishing in New South Wales waters on the

management of commercial fisheries. Final report to The Fisheries Research and Development

Corporation from the Fisheries Research Institute, Sydney. 139 pp.

Stephens, J.S, Jr., Hose, J.E., Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In Marine Organisms as Indicators. Soule, D.F., Kleppel G.S.

(Eds). Springer-Verlag NY. pp. 91-105.

TEL (1993). Changes to assemblages of reef fish adjacent to the North Head and Bondi cliff face

outfall following decommissioning. Final report by The Ecology Lab for the Water Board,

Sydney. 49 pp.

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final report by The Ecology Lab for the Water Board, Sydney. 29 pp.

Thompson, A.A., Mapstone, B.D. (1997). Observer effects and training in underwater visual surveys of

reef fishes. Mar. Ecol. Prog. Ser. 154, 53-63.

Tricklebank, K.A. (1994). The distribution and effects of aldrin in the damselfish Parma microlepis: an

experimental field study. International symposium on aquatic animal health: program and

abstracts, University of California, School of veterinary medicine.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA. 229 pp.

Underwood, A.J. (1991). Beyond BACI: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Mar. Freshwat. Res. 42, 569-587.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. J. Exp. Mar. Biol. Ecol. 161, 145-178. CHAPTER 1 GENERAL INTRODUCTION 16

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Aust. J. Ecol. 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15.

Underwood, A.J. (1997). Experiments in Ecology. Their Logical Design and Interpretation using

Analysis of Variance. Cambridge University Press. 504 pp.

Underwood, A.J., Chapman, M.G. (1996). Subtidal assemblages on rocky reefs at a cliff-face sewage

outfall (North Head, Sydney, Australia): What happened when the outfall was turned off? Mar.

Poll. Bull. 33, 293-302.

Underwood, A.J., Kingsford, M.J., Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Aust. J. Ecol. 6, 231-249.

Warwick, R.M. (1993). Environmental impact studies on marine communities: Pragmatical

considerations. Aust. J. Ecol. 18, 63-80.

Webb, R.O., Kingsford, M.J. (1992). Protogynous hermaphroditism in the half-banded seaperch,

Hypoplectrodes mccullochi (Serranidae). J. Fish. Biol. 40, 951-961. CHAPTER 2 LITERATURE REVIEW 17

CHAPTER 2

IMPACTS OF SEWAGE POLLUTION ON AQUATIC HABITATS,

FISH AND FISHERIES - A REVIEW

ABSTRACT

This review summarises information on the impacts of sewage pollution on aquatic habitats, fish and fisheries throughout the world. The scientific research that is available on sewage is voluminous (over one hundred published papers per year) but, it is often qualitative, specific to an area or issue, or suffers from poor experimental design. This review distinguishes well designed research which has detected (or did not detect) impacts that could be attributed to sewage from studies that cannot satisfactorily achieve this aim. Sewage contributes to major pollution problems adjacent to large human populations in many parts of the world. On a global scale, sewage pollution problems have remained relatively stable in developed regions such as Europe but are increasing in developing countries. On a habitat scale, sewage disposal is particularly damaging to freshwater, estuarine or small, enclosed waterbodies and to aquatic habitats such as coral reefs, seagrasses and macroalgae.

Sewage may affect fish at a number of levels, including through increased mortality, physiological effects or changes to communities. The relationship between sewage disposal and fish is species specific. The impacts of sewage on commercial and recreational fishers and aquaculturists include

1 changes in fish abundance, species composition, catchability, size and quality. A case study in relation to research and management of sewage pollution in NSW waters describes issues of monitoring, effect, reporting and fisheries. Recommendations are made for future research to detect the impacts of sewage on aquatic habitats, fish and fisheries to facilitate comparisons between studies. Management of sewage discharge is complex, however, the main issues affecting fish relate to human health, environmental health, fisheries and economics. Table 2.1. Number of articles in the Aquatic Science and Fisheries Abstracts published between 1988 and 1997 which contain the keywords ‘sew*’ and ‘habitat’ or ‘fish’ or ‘fishery’ and the year. Note: sew* is a search term for words which begin with sew, e.g. sewer, sewerage, sewage.

Year Habitat Fish Fishery

1988 34 91 26

1989 19 100 30

1990 15 103 22

1991 20 83 26

1992 34 85 19

1993 23 63 28

1994 35 78 22

1995 24 81 17

1996 74 60 15

1997 31 86 29

Average 31 78 23 CHAPTER 2 LITERATURE REVIEW 18

INTRODUCTION

Environmental pollution is a growing, world-wide concern because of deleterious impacts on aquatic resources (Clark, 1995; Earle, 1995; Baird, 1996; Koop and Hutchings, 1996). The most common cause of point source pollution to the world’s waterways is the disposal of sewage (GESAMP, 1990;

Boates and Russell, 1992; Caine, 1995; Dubinsky and Stambler, 1996).

Sewage is defined as ‘matter conveyed in sewers’ or ‘manure’ (concise Oxford dictionary) and it contains the waste products of human activity, both domestic and industrial, diluted by considerable quantities of water (Williams, 1985; Quentin and De Rouville, 1986). The contents of sewage include organic matter, bacteria, nutrients, oil and grease, chemicals, metals and litter (Scanes et al., 1995). The amount of sewage disposed of in undeveloped countries is largely unknown, but in developed countries most sewage is treated to some degree and then discharged through licensed outfalls or dumped as concentrated sludge. It is estimated that on a global basis 15 million tonnes of sewage sludge is dumped at sea annually (GESAMP, 1990). An unknown quantity of sewage is pumped from vessels, overflows from stormwater drains during wet weather conditions or is discharged illegally.

Many reports and environmental impact studies on the impacts of sewage are produced as ‘grey’ literature, and are difficult to obtain. This review focuses on the scientific literature and the impacts of sewage on aquatic habitats, fish and fisheries. Over 100 papers on these topics have been published per year since 1988 (Table 2.1). There has been no general review of the ecological impacts of sewage on aquatic ecosystems since that of Tsai (1975). Specific reviews have been published on deformities in fish exposed to sewage (EPA, 1993a), toxicity of sewage sludge to fish (Costello and Read, 1994), effects of sewage on coral-reef communities (Pastorok and Bilyard, 1985) and benthos (Reish and

Bellan, 1995). Assessment of the impacts of sewage on benthic macro invertebrate communities and also the bioaccumulation of contaminants are not considered in detail in this review because of the intended emphasis is on ecological impacts on fish (however see papers by Anderlini et al., 1992;

Miskiewicz, 1992; Diener et al., 1995; Reish and Bellan 1995; Avery et al., 1996; Krogh and Scanes,

1996; Scanes 1997). CHAPTER 2 LITERATURE REVIEW 19

Generally, increasing sewage pollution may result in a sequence of changes such as: (a) enhanced primary productivity, (b) changes in plant species composition, (c) very dense, often toxic blooms, (d) anoxic conditions, (e) adverse impacts on fish and invertebrates, (f) impact on amenity, (g) changes in structure of benthic communities (GESAMP, 1990). Not all of these features are observed in every case.

The spatial scales of sewage effects on aquatic biota have been described as ranging from negligible

(Smith, 1994; Grigg and Dollar, 1995), hundred of metres (Pastorok and Bilyard, 1985; Scanes et al.,

1995; MHL 1997), kilometres (Tsai, 1968; Ferraro et al., 1991; Brown et al., 1990; Gray, 1996) or hundreds of kilometres (Tsai, 1975; Colton, 1992; Anon., 1998). Temporal scales of sewage effects are generally characterised by a “press” or long-term disturbance (Scanes et al., 1995), however, “pulse” or short-term disturbances may also occur from sewage overflow events during dumping of sludge, seasons, storms, low flows in rivers or accidents (Seager and Abrahams, 1990; Gappa et al., 1993;

Tegner et al., 1995; Lemly, 1996; Pritchard et al., 1997).

Commonly, three levels of sewage treatment are recognised: primary, secondary and tertiary'. Primary treatment is merely the mechanical removal of solid material. Secondary treatment is mostly biological and may result in the removal of up to 90 percent of degradable organic wastes. Tertiary treatment generally involves chemical removal of nutrients and other contaminants. Untreated sewage is termed

‘raw’ sewage. In most countries of the world, primary or secondary treated sewage is the most common method of disposal to waterways.

The purpose of this review is to investigate the impacts of sewage on aquatic habitats, fish and fisheries and to use research and management of sewage impacts in New South Wales, Australia as a case study.

A desirable outcome is to identify well designed research which detected (or did not detect) impacts that could be attributed to sewage pollution. CHAPTER 2 LITERATURE REVIEW 20

OVERVIEW OF LARGE-SCALE IMPACTS OF SEWAGE POLLUTION

Globally, present inputs of nutrients from human activities are at least as great as those from natural processes (GESAMP, 1990). The inputs in different localities vary widely, depending on a range of factors including population density, land use, effluent treatment and dispersal rate (GESAMP, 1990).

Areas of critical global pollution include Europe followed by the Asia-Pacific, with the Latin American region regarded as having less but increasing pollution concerns and the polar regions a low concern for pollution (UNEP, 1997). Pollution trends are relatively stable in North America and the polar regions but were increasing in freshwater and/or coastal zones throughout the rest of the world (UNEP,

1997). Maps identify the Aral, Baltic, Black Mediterranean and North Seas as areas of severe coastal pollution (FAO, 1971; Lean and Hinrichsen, 1992). All seas in Europe, except for the northern seas, are facing eutrophication problems and the Black Sea is the most affected (EE, 1995).

Most of the world’s large rivers are polluted to some degree, however the worst pollution in European rivers probably occurred in the 1960s and water quality has since improved. Severe pollution is presently occurring in rapidly developing countries such as Brazil and India (Arthington and

Welcomme 1995). Nutrient concentrations in European rivers and lakes are highest in a belt from the southern UK to the Balkans and Ukraine, and phosphorus concentrations have declined but nitrogen concentrations continue to increase (EE, 1995). In 1990 about 60% of the European population, in the

24 countries for which data were available, was served by sewage treatment of some nature (EE,

1995). There were large variations between countries; raw sewage is discharged in Albania, less than

50% of the population in Portugal and Greece has sewage treatment but over 98% of the population of

Denmark is served by wastewater plants (EE, 1995).

IMPACTS OF SEWAGE DISPOSAL ON AQUATIC HABITATS

It is widely believed that the impact of sewage may be large in confined waterways such as rivers, coastal

lagoons and some estuaries (NOAA, 1990; Johnston et al., 1993), and that it is relatively minor in marine waters. This generalisation is based on the concept of the assimilative capacity or resistance of the following Figure

2.1. discharge e values General sewage discharge

form (outfall)

of

longitudinal of oxygen organic

material downstream

(from flow suspended

changes

Bayly

downstream

and in

some

solids Williams

physico-chemical

1973).

parameters CHAPTER 2 LITERATURE REVIEW 21 different systems (EPA, 1992; Johnston et al., 1993; Scanes et al., 1995; Scanes et al., 1997) or more simply ‘dilution is the solution to pollution’ (FAO, 1971).

Physico-chemical impacts on water

The direct impact of sewage is on the physical and chemical characteristics of the water, which may include changes to nutrients, turbidity, salinity, dissolved oxygen content, sedimentation and levels of chemicals and metals (Bayly and Williams 1973). A summary of the increased or decreased impact of sewage on several of these factors is contained in Table 2.1. Sewage generally results in increased nutrients, metals, chemicals and turbidity, however, decreased levels of salinity and dissolved oxygen may result from sewage pollution. A summary of the impact of sewage on the physical and chemical parameters is given in Table 2.2. Changes to any of these factors compared to natural levels, and particularly the additive or synergistic impacts from a number of factors, may result in a range of impacts to aquatic habitats.

The degree of physico-chemical impact of sewage pollution upon water depends on the type of pollutant, the volume and frequency with which it is discharged, the resilience of the habitat and the hydrography of the receiving waters. Some pollutants (most organic materials for example) may over time be decomposed by normal biological processes, but others such as some chemicals and metals may persist for a longer time. The persistent pollutants accumulate in the water, habitats and biota. A generalised pattern of changes in physico-chemical parameters following discharge of sewage in rivers is large increases in suspended solids and Biological Oxygen Demand (BOD) followed by a decrease in oxygen concentration. These variables return to normal with increasing distance from the source

(Figure 2.1; Bayly and Williams, 1973).

An increased flow of sewage into a waterbody may lead to decreases in water transparency and hypolimnetic oxygen and increases in nutrients and total dissolved solids (Leach et al., 1977; Marneffe et al., 1996). Surface sewage plumes off Sydney, Australia, were characterised by a lens of discoloured, turbid, low-salinity water 1 to 5 m deep, overlaying clear high-salinity marine water

(Gray, 1996). The individual sewage plumes extended up to 5 km seaward and 8 km along the coast Table 2.2. Summary of the increase and decrease in several parameters as a result of sewage pollution on aquatic habitats, fish and fisheries and examples of a scientific reference which describes this impact.

INCREASED DECREASED REFERENCE

AQUATIC nutrients Mameffe et al., (1996)

HABITATS sedimentation Pastorok and Bilyard (1985)

turbidity Grey (1996)

chemicals Abamou and Miossec (1992)

metals Bothner et al., (1994)

freshwater Sherwin and Jonas (1994)

algal blooms GESAMP (1990)

introduced species Chisholm et al., (1997)

macroalgae macroalgae Hardy et al., (1993)

dissolved oxygen Tsai et al., (1991)

seagrasses Short and Wyllie-Echeverria (1996)

sponges Roberts et al., (1998)

corals Pastorok and Bilyard (1985)

FISH mortality Kakuta and Murachi (1997)

bioaccumulation Miskiewicz (1992)

deformities Kingsford and Gray (1996)

abundance abundance Smith etal., (1998)

diversity Potter et al., (1988)

reproduction Lye etal., (1997)

movement Tsai (1968)

FISHERIES aquaculture Easa et al., (1995) Table 2.2 contd

aquaculture FAO (1971)

bacteria Krogh and Robinson (1996)

closures Leadbitter (1992)

catch catch Chang(1993)

species species Caddy (1995)

quality Montgomery and Needelman (1997)

money Collins etal., (1998)

taste Anttila (1973)

human health Leadbitter (1992) LaSalle Illinois River Pollution Conditions

LaSalle Peoria

LaSalle Peoria

..•^Beardstown 1913-15 Peoria Beardstown 1920-25

— Clean Beardstown __Subpollufional 1964-65 — Pollutional

Figure 2.2. Water quality conditions along the Illinois River, 1913-1965. Pollutional conditions indicate a predominance of pollution tolerant aquatic life (Colten 1992). CHAPTER 2 LITERATURE REVIEW 22 from the point of discharge. The dispersion and therefore impact of sewage plumes also depends upon the strength of tidal currents and the degree of stratification (Sherwin and Jonas, 1994). For example, sewage pollution contributes to the stratified anoxic and hypoxic system which occurs seasonally in

Chesapeake Bay, USA (Boynton, 1997).

Individual sewage outfalls may have a relatively small impact on aquatic habitats, but the cumulative impacts of sewage discharge in the River Hull, England, resulted in a zone where neither plants nor animals could live, extending for over 100 miles out of a total of 550 miles of the river (Tsai, 1975).

Figure 2.2 illustrates the increasing pollution of the Illinois River, from a predominantly clean river to an almost totally polluted state over a period of 50 years.

Soft sediment biota and habitats

Soft sediment habitats include mud, silt, clay and sand and may contain a diverse biota. The impacts of sewage on the benthic animals which live on or in these habitats has been extensively studied and the reader is referred to a recent review by Reish and Bellan (1995) which presents case studies of benthic monitoring off southern California and the French Mediterranean coast. A model of benthic communities near a sewage outfall suggests that there is a transition ranging from an abiotic region closest to the sewage, to high numbers of a few opportunistic species, to unstable numbers of most of the common species, to a normal community some distance away (Figure 2.3; Pearson and Rosenberg,

1978). However, this model was developed for semi-enclosed low energy environments and when it was tested by Maurer et al., (1993) in open ocean environments it was not found to be applicable because no observations were made of (1) sharp declines in Species Abundance Biomass (SAB) curves, (2) SAB curves displaced away from the pollution, and (3) dominant species excluding rare species.

Sewage, particularly when raw or primary treated, may contain sludge, silt, rubbish and contaminants which may alter existing soft sediment habitats to silt or mud habitats. The extent of habitat alteration will depend on the hydrography of the receiving waters and will generally be large in enclosed waters and small in ocean waters. Some of the particulate matter in a sewage plume is decomposed or Increasing Pollution

Number of Species

Biomass

Total Abundance

NORMAL TRANSITION POLLUTED GROSSLY POLLUTED

Increased Abundance Pollution Indicator No Macrofauna Species

Figure 2.3. Generic models for change in benthic communities in response to sewage disposal (from

Pearson and Rosenberg 1978). CHAPTER 2 LITERATURE REVIEW 23 suspended in the water column, but most of it settles to the bottom (Bothner et al., 1994). Sewage disposal is responsible for increased levels of metal and chlorohydrocarbon contaminants in the soft sediment habitats in the estuarine and oceanic waters adjacent to Sydney (Mortimer and Connell, 1995;

Birch, 1996). However, the maximum deposition rates of sewage particulate matter were approximately 1 gm 2day'' which may be difficult to distinguish from natural deposition (Bickford,

1996). The construction of outfall pipes over flat, soft sediment habitats in Hawaii provided an attachment site and refuge for a diversity of organisms and essentially changed the habitat to reef

(Russo, 1989; Grigg, 1994). The nearshore disposal of sediment from construction of a sewage outfall resulted in the long-term smothering of a rocky reef habitat and creation of a temporary beach in

Sydney (EPA, 1993b).

Rock and coral habitats

Habitats such as rocky and coral reefs may be affected by suspended sediments in sewage (Roberts et al., 1998). The settlement of sediments can smother or alter the substratum type (e.g. gravel to mud, reef to silt). Biological components of these habitats, such as sponges, corals and algae, are susceptible to sewage pollution and may suffer high mortalities due to smothering or shading, recruitment failure or rapid growth of competing species (Chapman et al., 1993; Roberts, 1996; Roberts et al., 1998;

Bellgrove et al., 1997).

The three components of sewage effluent most detrimental to coral habitats are nutrients, sediments and toxic substances (Pastorok and Bilyard, 1985). Sewage disposal invariably results in nutrient enrichment, which stimulates proliferation of oxygen-consuming microbes which may kill aquatic biota either directly by anoxia, or by related hydrogen sulfide production (Dubinsky and Stambler, 1996). For example, disposal of sewage in Kaneohe Bay, Hawaii, affected benthic habitats resulting in decreasing coral cover, taxonomic richness and net calcification rates, and increased biomass of the bubble alga Dictyosphaeria cavernosa, sponges and zoanthids (Pastorok and Bilyard, 1985).

Vegetated biota and habitats

The disposal of sewage into vegetated habitats generally results Primary Production

800 -

Seagrass Loss

Seaweed Shifts

Nuisance Loss of Phytoplankton Benthic Blooms Autotrophs

Oligo- Meso- Eu- Dys­ trophic trophic trophic trophic

Figure 2.4. Hypothesised shifts in primary production with eutrophication status and nutrient loading rates from coastal plain estuaries such as Chesapeake Bay (from Boynton 1997). CHAPTER 2 LITERATURE REVIEW 24

in increased phytoplankton ‘algal blooms’ (Vukadin, 1992; Lee and Chah, 1996; EPA, 1997), increased biomass of macroalgae and decline of seagrass beds (Short and Wyllie-Echeverria, 1996; Udy and

Dennison, 1997). The impact of sewage on vegetated estuarine habitats may result in a succession characterised by an initial increase in primary productivity, which then decreases due to seagrass losses, and then increases with seaweed and phytoplankton blooms (Figure 2.4; Boynton, 1997). Algal blooms have increased in frequency throughout the world as a result of long-term nutrient enrichment and many species exhibit toxic blooms or contribute to increased hypoxic/anoxic episodes (GESAMP, 1990;

NOAA, 1991).

There is a general trend for decreased diversity of brown (Phaeophyta) and red (Rhodophyta) algae and

increased diversity of green (Chlorophyta) and blue-green (Cyanophyta) algae in areas exposed to sewage disposal (Hardy et al., 1993; Bokn et al., 1996). A long term (60 year) study of macroalgal habitats of three estuaries in England exposed to sewage reported the above trends together with the

loss or restricted distribution of several previously conspicuous species (Hardy et al., 1993).

Opportunistic macroalgae species such as Ulva, Entermorpha, Cladophora, Gracilaria and Codium may greatly increase in abundance as a result of increased eutrophication from sewage disposal. The rapid proliferation of benthic green algae and the displacement of other biota is a common impact of sewage disposal (Litter and Murray, 1975; May, 1985; Pastorok and Bilyard, 1985; Fairweather, 1990). The disposal of sewage in the Mediterranean Sea (France) has resulted in a rapid increase in the introduced alga Caulerpa and mortality of the seagrass Posidonia (Chisholm et al., 1997).

Declines in seagrass often result from decreased penetration of light into the water column caused by eutrophication (Short and Wyllie-Echeverria, 1996). The cumulative impact of sewage discharge resulted

in the decline of seagrass in Cockbum Sound, Western Australia. Approximately 90% of Posidonia

seagrass beds that once flourished in the bay died relatively suddenly (i.e. over a period of several years)

after almost 20 years of continuous waste discharge (Figure 2.5; Cambridge and McComb, 1984; EPA,

1992). This delay may be explained as a cumulative effect of eutrophication which may reduce seagrass

root structure and undermine the bed over several years (Madden and Kemp, 1996). The loss of Posidonia

seagrass is essentially irreversible (Smith et al., 1997). 1954 — 1962 (b)

Figure 2.5. Seagrass loss in Cockbum Sound, Australia. Each map shows Cockbum Sound surrounded by the coast of the mainland to the right, and Garden Island to the left. The 10m contour line is indicated. The shading shows the area of seagrass present at different times (from Cambridge and

McComb 1984). CHAPTER 2 LITERATURE REVIEW 25

Sewage may cause direct mortality of vegetated habitats close to the discharge point, and indirect sub- lethal effects, such as the inhibition of growth or reproduction further away. San Diego’s sewage outfall failed in 1992, spilling 710 ML/day of treated sewage into a kelp Macrocystis pyrifera forest for a two month period (Tegner et al., 1995). Damage to kelp occurred in the immediate vicinity of the accident, apparently as a result of low light and high nutrient conditions. A short-term reduction in kelp germination also occurred. However, after the outfall was repaired kelp germination and growth rapidly resumed and there were no lasting effects on the aquatic community (Tegner et al., 1995).

Vegetated habitats such as algae, mangroves and other wetlands have been utilised as biological filters of sewage. A saltmarsh community in Massachusetts, USA, was used as a site for disposal of sewage, which resulted in an accretion rate of several centimetres of sewage sludge per year, retention of 80-

96% of the nitrogen, and the accumulation of toxic materials (Haines, 1980).

IMPACTS OF SEWAGE DISPOSAL ON FISH

Fish may exhibit a range of responses to sewage pollution at a range of different scales, including at the community, individual and tissue levels (Figure 2.6; Underwood and Peterson, 1988; Gray, 1989a;

Adams et al., 1989; EPA, 1993a,). An environmental stress, such as sewage pollution may result in an acute or chronic impact on fish (Figure 2.6). If the impact is acute the fish dies, alternatively chronic impacts result in direct or indirect changes to the individual through physiological or molecular effects, or changes to the ecosystem such as food availability. Impacts may also be detected in a fish’s growth or reproduction, population or community level (Figure 2.6). Some of these impacts will be discussed below in the sections on mortality, physiology and communities.

Scientific research on the impacts of sewage on fish has generally focused on demersal or site-attached species which are commercially or recreationally important. These include freshwater groups such as trouts and other salmonids (F. Salmonidae), carps (F. Cyprinidae), Australian bass (Macquaria novemaculeata, Percichthyidae) (Antilla, 1973; Tsai, 1975; Growns et al., 1998; Kakuta, 1997; Kakuta and Murachi, 1997), and estuarine and coastal species such as flounders (F. Pleuronectidae), red ENVIRONMENTAL ACUTE DIRECT MORTALITY STRESS RESPONSE

CHRONIC RESPONSES OIRECT INDIRECT

TOXICOLOGICAL ENERGY (FOOD) EXPOSURE/EFFECTS AVAILABILITY

PHYSIOLOGICAL MOLECULAR/ PATHOLOGICAL METABOLISM DEVELOPMENT PERFORMANCE MUTAGENS ( SPRING I IMMUNE SYSTEM CARCINOGENS COMPETENCE NEOPLASIA REPAIR

ENERGY ORGANISM NUTRIENT TISSUE ELABORATION GROWTH REPRODUCTION STORAGE

POPULATION

COMMUNITY

Figure 2.6. Generalised responses of fish to chronic environmental stress. This figure emphasises the acute response resulting in direct mortality of fish, direct and indirect responses on fish physiology

(toxicological and energy availability) and community levels (Adams 1989). CHAPTER 2 LITERATURE REVIEW 26 morwong (Cheilodactylus fuscus, Cheilodactylidae) and Australian snapper (Pagrus auratus, Sparidae)

(Lincoln Smith and Mann, 1989; Otway et al., 1996a; EPA, 1996; Lye et al., 1997). There are few studies of the impacts of sewage on large, pelagic or highly migratory species such as sharks, tunas or anguillid eels.

Mortality of fish

The death of large numbers of fish can be a very dramatic outcome of water pollution. In 1970 the death of approximately 20 million fish was directly attributed to sewage pollution in the USA (Tsai,

1975). Municipal sewage pollution was the highest source of fish kills in Germany and the second largest source of fish kills after industrial wastes in the USA (Tsai, 1975). The developing fish embryo or larva is considered the most sensitive stage to pollution in the life history of a fish (Costello and

Gamble, 1992; Kingsford and Gray, 1996). Shrimp larvae were up to 500 times more sensitive to sewage sludge than adults (Franklin, 1983). Migratory species with high dissolved oxygen requirements are also susceptible to sewage (Arthington and Welcomme, 1995). The reason that most fish die from sewage is probably because of low dissolved oxygen levels, although suspended solids may also cause abrasive injuries and clogging of the gills, and toxic chemicals such as chlorine can also cause mortalities (Tsai, 1975; Abarnou and Miossec, 1992). The cumulative or synergistic impact of sewage pollution and cold water has been described as Winter Stress Syndrome, which may be responsible for eliminating 30% of a year class of fish such as bluegill sunfish (Lepomis macrochirus,

Centrarchidae) (Lemly, 1996).

Experimental approaches for measuring mortality of fish exposed to sewage have been described in a number of papers (Tsai, 1975; Birtwell et al., 1983; Mitz and Giesy, 1985; Seager and Abrahams,

1990; Costello and Gamble, 1992; Kakuta and Murachi, 1997). All carp (Cyprinus carpio, Cyprinidae) exposed to 100% sewage effluent died within 6 h, and 60% and 10% of carp died within 48 h exposure to 20% and 10 % of sewage, respectively (Kakuta and Murachi, 1997). Acute mortality of channel catfish (Ictalurus punctatus, Ictaluridae) occurred at sites 300 and 500 m downstream from a 76 ML/d sewage treatment plant outfall on the Flint River, Michigan (Mitz and Giesy, 1985). Concentrations of

0.1% sewage sludge caused significant toxic effects on the embryos and larvae of herring (Clupea CHAPTER 2 LITERATURE REVIEW 27

harengus, Clupeidae) and cod (Gadus morhua, Gadidae) (Costello and Gamble, 1992). Mortality of juvenile chinook salmon (Oncorhynchus tshawytscha, Salmonidae) occurred at all experimental sites

within 4.4 km of a 1530 ML/d sewage treatment plant outfall which discharged to the Fraser River,

British Columbia (Birtwell et al., 1983). Mortality was rapid, and fish placed 2.2 km from the outfall

died within 9 minutes as a result of low dissolved oxygen (Birtwell et al., 1983). However, three

species of rock Fish were placed in cages in a sewage effluent zone in San Francisco Bay and no

significant mortality could be demonstrated on fish after 96 hours of exposure to 1% effluent (Tsai,

1975).

Physiological effects on fish

Sewage may result in physiological effects on fish such as stress, metabolism, growth rate, fecundity,

deformities, neoplasia (cancers), contaminant levels, parasites or diseases (Figure 2.6; Weis et al,

1990; EPA, 1993a; Avery et al., 1996; Moore et al., 1996; Kakuta 1997). Different species of fish and

even fish of the same species, population, age and sex may respond differently to sewage pollution due

to differences in their general health status, which can be affected by habitat, diet and season.

Decreases in growth and reproduction and increased incidence of diseases or deformities are common

measures of the effects of sewage on fish (Table 2.2). Relatively brief exposures (24 - 96 h) to

pollution plumes have resulted in reduced viability of larvae, reduced growth and deformities

(Kingsford and Gray, 1996). In Santa Monica Bay, California, it was reported that spotted turbot

(Pleuronichthys ritteri, Pleuronectidae) from the vicinity of a sewage outfall weighed significantly less

than those caught at control sites and many fish had tumours or lesions. In the New York Bight,

bacteria associated with sewage pollution were implicated in the infection of up to 8% of bluefish

(Pomatomus saltatrix, Pomatomidae) and 4% of flounder (F. Pleuronectidae) with fin rot disease, skin

haemorrhages, skin ulcers and abnormal necrotic kidneys (Tsai, 1975). Increased liver weight (which

may be associated with enhanced detoxification activities) and abnormalities in the reproductive health

of up to 53% of male flounder (Platichthys flesus, Pleuronectidae) was attributed to sewage disposal in

the Tyne estuary, England (Lye et al., 1997). Sand gobies (Pomatoschistus minutus, Gobiidae) exposed

to 0.1 % sewage sludge produced a decreased number of eggs and larvae, and 60-70% of these larvae CHAPTER 2 LITERATURE REVIEW 28 were lighter than those from non-exposed parents (Waring et al., 1996). The reproduction potential of burbot (Lota lota, Gadidae) in the Rybinskoe Reservoir, Russia, has been reduced by approximately

50% due to sewage disposal (Volodin, 1994). Exposure of trout (F. Salmonidae) to sewage effluent resulted in a very pronounced increase (500 to 100,000 fold) in the plasma vitellogenin concentration, which may result in the occurrence of hermaphiditic fish (Purdom et al., 1994).

Effects on fish communities

A range of simple classifications or models of the impacts of sewage pollution on fish community structure has been proposed over the past 60 years. Four zones, a recent pollution zone, active decomposition zone, recovery zone and cleaner water zone were described for a stream fish community (Tsai, 1975). In the zone of recent pollution, the sewage discharged into the water is still

‘fresh’ and fish frequently gather at the vicinity of the outfall. In the zone of active decomposition, dissolved oxygen (DO) levels are low and fish abundances are low. In the zone of recovery, DO gradually increases and certain fish, particularly carp are found. In zones of clean water, ‘game’ fish are more commonly found. A similar zonation describes five zones, an active bacterial decomposition zone, intermediate zone, fertile zone, game fish zone, and biologically poor zone. In the zone of active bacterial decomposition immediately below a sewage outfall on the Ohio River, USA, only a few species of fish such as carp were reported. When desirable enrichment changes to undesirable pollution, the effects on a fish community are first a reduction in the number of species, then a reduction in the total biomass, and finally a reduction in the number of individuals (Tsai, 1975; Weng,

1990).

The relationship between sewage disposal and the ecology of fish is species specific and sometimes sites (Hurley and Christie, 1977; Leach et al, 1977; Harmelin-Vivien, 1992; Pilanowski, 1992; TEL,

1993, 1994; Kingsford and Gray, 1996; Otway et al., 1996b). For example, numbers of the snapper

(Lutjanus kasmira, Lutjanidae) and the relative numbers of herbivorous fishes increased adjacent to deepwater outfalls off Hawaii (Russo 1982, 1989; Grigg 1994). Localised increases in the abundance of bait fish were correlated with the disposal of effluent from a sewage outfall in the Mediterranean Sea

(Bell and Harmelin-Vivien 1982). Increases in the abundances of some species of demersal fish, CHAPTER 2 LITERATURE REVIEW 29 including gurnard (Lepidotrigla mulhalli, Triglidae) and flathead (F. Platycephalidae) and decreases in the abundances of a range of species including Australian snapper, leatherjacket (F. Monacanthidae) and stingray (F. Urolophidae), were observed in collections made adjacent to deepwater ocean outfalls off Sydney (Otway, 1995; EPA, 1996; Otway et al., 1996a).

The reasons for these differences in fish communities are complex. There is evidence that adaptation to environmental stress may allow some species to tolerate sewage pollution e.g. the Indian catfish

(Pangasins pangasius) appears to be unaffected by sewage pollution in India, because of its ability to store air in the buccal cavity and undergo subcutaneous respiration (Tsai, 1975). Other possible reasons for an increased abundance of fish at a sewage outfall include: food supply (Hurley and Christie, 1977;

Love et al., 1987; Stephens et al., 1988; Russo, 1989; Steimle, 1994; Hall et. al. 1997), avoidance of predators in turbid water (Lenanton et al., 1984) and provision of additional habitat (Grigg, 1994,

Bolotova et al., 1996). Decreased abundance of fish at a sewage outfall may be due to mortality, avoidance of turbid or toxic water, decreased recruitment, behaviour and competition (Tsai, 1968;

Adams, 1990; Gray et al., 1992; Otway, 1995). The zone below sewage discharge points or chlorination in rivers may be a barrier to migratory fish either up or downstream (Tsai, 1968; Paller et ai, 1988). Degradation of fish diversity from an Illinios, USA, stream was severe during secondary treatment with chlorination, but improvements occurred in species number when sewage was upgraded to secondary treatment without chlorination (Paller et al., 1988).

IMPACTS OF SEWAGE DISPOSAL ON FISHERIES

Commercial and recreational fisheries

Nutrient enrichment and overfishing have similar and synergistic effects: a decline in diversity, an initial increase in productivity of benthic/demersal and pelagic food webs, (followed by a seasonal, and finally permanent anoxia of bottom water), and the progressive dominance of the production system by short-lived, especially pelagic species (Caddy, 1995). Sewage pollution may impact on resources harvested by commercial and recreational fishers, aquaculturists and biota that are important to non­ consumptive users such as SCUBA divers. The discharge of sewage effluents may influence fisheries directly or indirectly, including the reduction of stocks by mass mortalities, gradual change in CHAPTER 2 LITERATURE REVIEW 30 composition of communities or ecosystems, fish size, catchability, increased incidence of diseases, and decreased quality, edibility and market potential.

A major impact of sewage pollution may be the total collapse of affected fisheries. For example, discharge of sewage led to an anaerobic condition in the stratified waters of Osloljord and destroyed the commercial fishery for prawns (Tsai, 1975). Changes to the composition of fishery catches are a more common impact of sewage. There is often a replacement of valuable species (gadoids, flatfishes and lobsters in marine areas, and salmonids and sturgeon in freshwater estuarine habitats), by small, lower-value pelagic fish and perciforms respectively (Caddy, 1995). This generalisation is supported by analysis of fisheries abundance data based on catch per unit effort of bottom trawl surveys, which indicated that abundances of hakes (Merluccius bilinearis, Merlucciidae and Urophycis chuss,

Gadidae), summer flounder (Paralichthys dentatus, Pleuronectidae), goosefish (Lophius americanus,

Lophiidae) and black sea bass (Centropristis striata, Serranidae) declined; however, abundance of spiny dogfish (F. Squalidae), skates (F. Rajidae) and some pelagic species increased (Chang, 1993) around a sewage disposal site from 1982 to 1990 in the USA.

In many rivers in Europe, such as the Rhine (France), Thames, Severn and Hull (England), Vistula

(Poland) and Bidasoa (Spain), declining catches of salmon and trout have been attributed to sewage pollution (Tsai, 1975; Arthington and Welcomme, 1995). The Rhine supported 400 commercial fishers in 1850, which declined to 60 fishers by 1900 (Arthington and Welcomme, 1995). Sewage in the

Helsinki area resulted in declining catches of Lota lota, brown trout (Salmo trutta, Salmonidae), whitefish (Coregonus lavaretus, Coregonidae), and ide (Leuciscus idus, Cyprinidae); however, catches of pike-perch {Lucioperca lucioperca, Esocidae) remained stable (Antilla, 1973). The fisheries yields of several groups of freshwater fishes from European lakes suggest that increasing eutrophication initially results in increased catches of corengoids and percids which then decline, followed by very large yields of cyprinids under eutrophic conditions (Figure 2.7, Leach et al., 1977).

It has been suggested that sewage effluents may also improve fisheries in some cases (Boddeke and

Hagel, 1995; Tsai et al., 1991). Commercial landings of demersal fish in the North Sea rose after 1963 C YPRI NIDS

O 30

C0REG0NIDS PERCIDS

EU T ROPHY TRANSPARENCY (m)

10 12 0 Plot (m g • m 3)

Figure 2.7. Summary of the relationship between increasing eutrophy and yields of corregonids, percids and cyprinids from 17 European lakes (from Leach et al., 1977). CHAPTER 2 LITERATURE REVIEW 31 to more than 1 million tonnes from a rather constant level of 400,000 tonnes since 1909, and this rise and the recent fall has been linked to eutrophication (of which sewage discharge and treatment is one factor), resulting in increased primary and secondary productivity and fish growth (Boddeke and

Hagel, 1995). A similar causal relationship between sewage nutrients, fertility of striped bass (Morone saxatilis, Percichthyidae) and commercial fisheries was suggested to explained a rise in nutrients, fertility and catches from the 1940s through the 1960s, followed by declines in the 1970s due to improvements to sewage treatment (Tsai et al., 1991).

A sewage outfall in Ohio resulted in a positive effect on a recreational fishery and this was attributed to the effluent supporting populations of microcrustaceans which attracted minnows and then white bass

(Tsai, 1975). Large numbers of recreational fishers may also fish adjacent to some sewage outfalls in

Australia, because of large catches (TEL, 1993). A survey of recreational fishers in Sydney estuaries found that some people go fishing during sewer overflow events (rainy days) and catch large numbers of several species, including yellowfm bream (Acanthopagrus australis, Sparidae), luderick (Girella tricuspidata, Girellidae), Pagrus auratus, Pomatomus saltatrix and yellowtail (Trachurus novaezelandiae, Carangidae) (AWT, 1997).

Most recreational fishers may not fish adjacent to sewage outfalls because the quality of the recreational experience would be compromised by the discoloured water and putrid odour. Another negative aspect of the capture of fish near sewage outfalls is concern about the edibility of fish, which may be affected by tainting, pathogens, parasites or contaminants such as toxic heavy metals and organochlorines (Beder, 1989; McLean et al., 1991; Scanes and Philip, 1995; Gibbs and Miskiewicz,

1995; Krogh and Robinson, 1996; Montgomery and Needelman, 1997). One in three people who caught fish near a sewage outfall near Helsinki reported that the taste was tainted by oil, mud, dung or petroleum (Anttila, 1973). The bioaccumulation of contaminants in fish captured near sewage outfalls

(Lincoln Smith and Mann, 1989) and concerns about public health (Beder, 1989), resulted in a prohibition of fishing within 500 m of the three major shoreline outfalls in Sydney and also economic losses for the fishing industry (Leadbitter, 1992). Risk assessment of humans consuming

Cheilodactylus fuscus from sewage-affected locations near Sydney was estimated at 2 to 8 chances in CHAPTER 2 LITERATURE REVIEW 32

10,000 of contracting a cancer, although this was an underestimate because PCBs were present and not included in the analyses (Leadbitter, 1992). Fisheries interests were important in the decision to relocate large-scale sewage disposal from nearshore dumping grounds to 106 miles offshore from

California to prevent contamination of commercial fishing grounds (Bothner et al., 1994).

A modelling study of the impacts of chronic and acute pollution on a hypothetical commercial fishery indicated negative impacts on biomass, catch and profitability, however recovery occurred after a period of 5 to 10 years (Collins et al., 1998). Economic effects from pollution can occur even when there is no measurable response by the fish population. For example, a coastal fishery might be closed because adjacent shellfish beds have been contaminated by sewage and the demand for all seafood products is generally reduced (Lipton and Strand, 1997).

Aquaculture

Aquaculturists are concerned about health risks to human consumers, economic losses arising from mortality and sub-lethal impacts on their products and the potential benefits of the use of treated sewage for fish production.

The principal problem for human health on a world-wide scale is the existence of pathogenic organisms discharged with sewage to natural waters (GESAMP, 1990). These organisms can affect humans who consume seafood, and molluscs are particularly susceptible to contamination by pathogens from wastewater flows, since the growing sites are often in highly polluted areas near urban centres (GESAMP, 1990). In Iwayama Bay, in the Palau Islands in the Pacific Ocean, where sewage pollution was severe, it was found that crabs, oysters and coral reef fish were contaminated with a high content of faecal coliform bacteria, up to 20,000 times the concentration of the surrounding water.

Sewage may have detrimental impacts on mollusc fisheries (Johnston et al., 1993; ACIL, 1997) and massive mortality of the carpet shell Venerupis decussata was attributed to increasing levels of sewage in the Ria Formosa Lagoon, Portugal (Bebianno, 1995). The commercial cultivation of Sydney rock oysters

Saccostrea commercialis in NSW reached a peak production of 15 million dozen in the 1970s, but had declined by approximately fifty percent by 1996 (ACIL, 1997). This production decline was attributed Table 2.3. Overview of ocean sewage outfall characteristics and extent of monitoring that has been conducted in NSW (derived from MHL 1997 and EPA 1996). ML- Megalitres a day dry flow. NA -

Not applicable, E - Effect reported, I - Indeterminate effect reported, N - No effect reported.

Small Moderate Large Very Large

Size <5 ML 5 to 19 ML 20 to 50 ML >50 ML

Number 16 10 6 3

Type 94% shoreline 80% shoreline 50% shoreline 100%

6% offshore 20% nearshore 17% nearshore deepwater

33% offshore

Treatment 6% primary 30% raw 33% primary 100%

44% secondary 10% primary 33% secondary primary

50% tertiary 30% secondary 33% tertiary

30% tertiary

Monitoring 5(31%) 7 (70%) 6(100%) 3 (100%)

Intertidal 4 (2E, 11, IN) 5(3E, 2N) 2(2E) NA

Subtidal 3(1E, 11, IN) 6(2E, 21, 2N) 3(2E, 11) 3 E

Fish 2(11, IN) 5(4E, IN) 1(E) 3 E

Bioaccum. 4(4N) 4(11, 3N) 4(11, 3N) 3 N

Fishery 0 0 0 0 CHAPTER 2 LITERATURE REVIEW 33 to a number of causes, including pollution effects on water quality, and microbial contamination causing human illness and thus reduced consumer confidence (ACIL, 1997). In some situations, sewage has been treated with chlorine to reduce microbial contamination in aquaculture regions, however this may result in increased mortality, sub-lethal toxicity and ecosystem effects because molluscs are particularly sensitive to chlorine (Abarnou and Miossec, 1992).

In some parts of the world, sewage has been used as a fertiliser to increase fish production (Tsai, 1991).

Fish reared in treated effluent in Suez, Egypt, were suitable for marketing for human consumption and were equal in quality to wild fish (Easa et al., 1995).

CASE STUDY- SEWAGE OUTFALLS IN NSW, AUSTRALIA

This case study describes the sewage disposal system in NSW and some of the common research issues associated with of the monitoring, detection of effects, reporting and management of the environment and human health.

Description. There are 295 licensed sewage outlets in NSW, most of which are small and flow to freshwater rivers. A total average volume of approximately 1200 ML'1 day of sewage is discharged to

NSW waters, of which the major volume of sewage (74%) is discharged to the marine environment, 19% flows to estuaries and rivers, and the remaining 7% is generally discharged to land (EPA, 1997). There are

35 ocean outfalls, most of which discharge less than 5 ML day, and there is a trend for a decreasing number of moderate sized, large and very large outfalls (Table 2.3). The largest outfall in NSW discharges approximately 475 ML day (MHL, 1997). The ocean outfalls have been constructed on the shoreline, nearshore, offshore or deepwater, and most of the small and moderate outfalls discharge secondary or tertiary treated effluent to the shoreline, however 30% of the moderate sized outfalls discharge raw sewage (Table 2.3). As the size of the outfalls increases, there is a trend for locating them further offshore, together with a lesser quality of (primary) treatment (Table 2.3).

Monitoring. In NSW, there has been extensive but generally short-term, scientific research on the very large sewage outfalls off Sydney compared to about 30% of the smaller ocean outfalls (Table 2.3). The CHAPTER 2 LITERATURE REVIEW 34 monitoring that has been undertaken at the sewage outfalls can be grouped as follows: intertidal, subtidal, fish and bioaccumulation. There has been no intertidal research at the very large outfalls because they all discharge to deepwater sites and there has been no fishery monitoring. There has been a greater number of subtidal and bioaccumulation studies (15 each) compared to intertidal and fish studies (11 each) (Table 2.3). The methodology for groups such as fish has varied from underwater visual census, gill nets, trawl nets, plankton nets and longlines, which makes comparisons difficult

(Fagan et al., 1992; Otway, 1995; Scanes, 1996). The design and analysis of some of the above studies may suffer from deficiencies in experimental design (Krogh and Koop, 1996).

Detection of effects. Significant or indeterminate effects of sewage have been detected for intertidal, subtidal and fish communities at approximately half of the small and moderate and all of the large and very large outfalls in NSW (Table 2.3). No significant effects have been detected for bioaccumulation studies, although indeterminate effects were detected at moderate and large outfalls (Table 2.3).

Reporting. There has been a large amount of scientific research on the three very large sewage outfalls

located off Sydney (for example, 19 papers in Marine Pollution Bulletin 1996, Volume 33, nos 7-12) compared to other smaller sewage outfalls (1 paper in the above volume). Several other papers have been published on small outfalls (Smith 1994; Roberts et al., 1998) and there are numerous

unpublished reports. The numerous reports and lack of scientific papers may reflect the limited quality of some research. A comprehensive report of sewage effluent ocean outfall performance in NSW

involved a review of physico-chemical and scientific studies on small, medium, large and very large outfalls and recommended a more systematic collection of existing data (MHL, 1997).

Management of environmental health. Sewage effluent discharge must be licensed in NSW waters

under the Protection of the Environment Operations Act 1998. Approvals and licences issued under

legislation may contain conditions for pollution reduction, monitoring etc. The approvals are generally based on standards such as ‘Australian water quality guidelines for fresh and marine waters’(ANZECC,

1992) and ‘Discharge of wastes to ocean waters’ (EPA, 1993c) which provide values for acceptable CHAPTER 2 LITERATURE REVIEW 35 levels of physio-chemical pollution and acceptable ecological changes to a fish community or to aquatic habitat or fishery.

Managers attempt to improve sewage treatment to ensure safe, beneficial and environmentally acceptable disposal of effluent (SW, 1997), and address the public’s concerns about sewage outfalls: what effect will the outfalls have on the (marine) environment, will the bathing waters be suitable to swim in, and will the fish be safe to eat? (EPA, 1996). Management of sewage outfalls is evaluated in annual reports, parliamentary inquiries and community perceptions. A recent parliamentary inquiry into the management of sewage and sewage by-products in the NSW coastal zone provided direction for strategic planning, effluent reuse, monitoring and consultation (EPA, 1997). The relocation of very large shoreline outfalls to deepwater outfalls in the Sydney region appears to have resolved many of these public concerns (Koop and Hutchings, 1996; SW, 1997), however, there is ongoing community dissatisfaction with the management of sewage effluents in NSW (Beder, 1989; Leadbitter, 1996; EPA,

1997).

RESEARCH ISSUES

Fish are regarded as good bioindicators of sewage pollution because they integrate the effects of many biotic and abiotic variables, are easy to identify, and are of commercial and recreational importance and public interest (Adams, 1990; Harris, 1995). Some aquatic biota such as benthic invertebrates, or aquatic habitats such as seagrasses and corals are often monitored in preference to fish, because of the assumption that if biota at the base of the food chain or habitats are affected then so will the fish

(Stephens et al., 1988; Maurer et al., 1993; Reish and Bellan, 1995; Wright et al., 1995; Schmitt and

Osenberg, 1996; Kelly, 1998). The most comprehensive approach is to undertake multidisciplinary research over a number of years, incorporating field surveys and field experiments to understand the impacts of sewage pollution (Underwood and Peterson, 1988; Gray, 1989b; Bernstein and Dorsey,

1991; Skilleter, 1995). By contrast, most research on sewage has been short-term and limited in scope

(although see Colten (1992) and Hardy et al., (1993) for long-term studies). Table 2.4. Examples of rigorous scientific research which have utilised appropriate experimental design to detect the impacts of sewage on aquatic habitats, fish and fishes. B- Before, A- After, C-

Control, I- Impact, Exp- experiment, E - Effect reported, N - No effect reported, na- not applicable. * - pollution gradient design. The number in brackets indicates the outfalls in the study, although these locations were not directly sampled.

Variate Experimental design Impact Location Reference

B A C I

AQUATIC HABITATS macrobenthos 0 4 2 1 N Australia Underwood and Chapman (1996) macrobenthos 3 3 2 1 E Australia Roberts et al., (1998) algae 1 1 0 3 E UK Hardy et al., (1993) algae 0 1 9* 3 E Australia Fairweather (1990) coral 1 1 12 2 E USA Grigg (1994)

FISH larval fish 3 5 3 3 N Australia Gray (1997) larval fish (Exp) na 1 1 5 E USA Costello and Gamble (1992) benthic fish (Exp) na 1 1 2 E Scotland Waring et al., (1996) rocky reef fish 4 8 2 1 E Australia Smith et al., (1998) rocky reef fish 0 6 3 3 E Australia Krogh and Scanes (1996) rocky reef fish 0 1 24* (3) E Australia Lincoln Smith and Mann (1989) estuarine fish 0 4 2 2 E Australia Growns et al., (1998) estuarine fish 0 5 4 1 E UK Hall et al., (1997)

FISHERIES

Catch per unit effort 2 2 2 2 E USA Chang (1993) CHAPTER 2 LITERATURE REVIEW 36

Physiological measures may be the most sensitive measures of pollution effects (Adams, 1990).

However, measures on communities may provide a better indication of the consequences of that pollution to processes of economic and societal value such as fishing (Underwood and Peterson, 1988;

Keough and Quinn, 1991). A complete assessment of any episode of pollution must include accurate measures of the biological effects of pollution at a number of different scales (Underwood and

Peterson, 1988).

The results of many studies on the impacts of sewage are equivocal (see Chapman et al., 1995; Grigg,

1995; Grigg and Dollar, 1995; Birch, 1996; Roberts, 1996; Krogh and Koop 1996; Underwood and

Chapman, 1996; Gray, 1997; Sindermann, 1997). This may be partially explained by sub-optimal experimental design: there are few Before/After/Control/Impact (BACI) or Beyond BACI studies

(Underwood 1991, 1992, 1994) or pollution gradient studies (Lincoln Smith and Mann 1989b; Keough and Mapstone, 1995) but also because only seldom does sewage pollution take place alone, without other diffuse pollutants or activities also impacting on aquatic ecosystems and confounding the analyses (Grigg, 1994, 1995; Sindermann, 1997). For example, in a study of an infaunal community off California, natural features accounted for over 80% of the variability, while sewage discharge- related effects represented less than 10% of the variability (Diener et al., 1995).

The majority of papers and reports discussed in this review are qualitative, but several quantitative studies that have detected an impact on aquatic habitats, fish and fisheries that could be attributed to sewage pollution have been selected to identify research that has been well designed with adequate controls and replication (Table 2.4). This table also contains several studies that were well designed but did not detect an impact of sewage pollution (Table 2.4). Approximately half of the studies selected as examples of well designed scientific research have used a BACI design and, of these, two studies

(Chang, 1993; Roberts et al., 1998) have a balanced design with the same number of before and after samples. Three studies have a large number (9 to 24) of control locations, to compare the outfalls with both near and far control locations (Fairweather, 1990) or compare the impact of the outfalls at increasing distances (e.g. 0.5, 1.5, 2.5 and 3.5 kms north and south - Lincoln Smith and Mann, 1989b).

Note that this latter study was a pollution gradient sampling design (Keough and Mapstone, 1995) that CHAPTER 2 LITERATURE REVIEW 37 did not sample immediately within the three sewage outfall sites. The number of impact locations or samples varies from 1 to 5, however, most studies have 1 to 3 impact locations and the study that had 5 impacts was a laboratory experiment (Table 2.4).

The lack of standard methods or guidelines for research which could allow comparison between studies of the impacts of sewage has been recognised as a problem (FAO, 1971; Kelly, 1998). It is recommended that the following research on aquatic habitats, fish and fisheries be undertaken as a minimum requirement for future assessment of the impact of proposed sewage outfalls. For detailed information on scoping and environmental assessment of developments which may impact on the aquatic environment see Underwood (1991, 1992, 1993, 1994, 1996), Keough and Mapstone (1995),

Lincoln Smith (1998) and NSW Fisheries (1998).

Recommended research to detect impacts of sewage pollution

Aquatic habitats - A description of physical and biological habitats at the potential impact location(s) and multiple control locations. If sensitive biological habitats such as seagrasses, corals, macroalgae, small rivers or enclosed waterbodies (or protected areas such as Marine Parks) are adjacent to the impact location, then quantitative surveys are recommended at multiple control locations at multiple times (Beyond BACI experimental design) or sampling sites may be located at distances downstream and upstream from the sewage outfall (pollution gradient design).

Fish - (a) A pilot study of a number of survey techniques followed by a quantitative survey of fish abundance and diversity; (b) collection of at least three species of abundant, resident fish and analysis of their physiological condition (minimum n=20) and bioaccumulation (minimum n=5); and (c) experimental manipulation of sentinel biota. All studies must involve the potential impact location and multiple control locations at multiple times (Beyond BACI experimental design) or sampling sites may be located at distances downstream and upstream from the sewage outfall (pollution gradient design).

Fisheries - A survey of existing users of the fisheries resources and their catches or other requirements, including commercial and recreational fishers (catch per unit effort), aquaculturists and non-extractive users such as SCUBA divers. CHAPTER 2 LITERATURE REVIEW 38

MANAGEMENT ISSUES

Conceptually, the logic of management planning is simple. It consists of problem definition - identifying the impacts related to each activity, and problem rectification - incorporating suitable controls and limitations (Kenchington, 1990). However, it is unusual for managers to deal with a simple situation where the issues relate to a single impact and a single impacted group. In most cases that involve sewage pollution the management issues are complex and include issues such as technology, health, economics, politics, legislation, community perceptions, education, research and environment. The environment is naturally dynamic and this makes it difficult to attribute changes in physical, chemical or biotic variables to a sewage source. This complexity has made it difficult for managers to develop legislation, regulations, guidelines or standards that are realistic and enforceable for a variety of criteria. Furthermore, although there has been a large amount of scientific research on the impacts of sewage pollution, some of the research may be poorly designed or contradictory, and therefore of little value for assisting management decisions. For example, sewage disposal may result in significant impacts on selected species or fish but have no immediately identifiable impacts on other species, and in such cases managers must resolve how these apparently contradictory scientific findings relate to current management actions and what is the ecological significance of these changes (Smith et al., 1998).

Society has generally been unable (or unwilling) to reach a consensus on what constitute acceptable ecosystem states and changes (Baird, 1996). In most countries, waterways are used for a variety of conflicting uses, including cheap, rapid removal of waste substances such as sewage. In most countries there are laws relating to sewage discharge, and in the USA these are more restrictive for ocean than land disposal (Connor, 1998) and discharge of sewage sludge or primary treated sewage into waterways is not permitted in some areas of Europe, USA and Australia (Oakley, 1979; ANZECC, 1992; Grigg and Dollar,

1995). Society’s priorities may also change over time, as illustrated by the case of shellfish cultivators in

England who from the 1890s to 1915 lawfully obtained the right to clean water flowing over their property, and sewage flow across a shellfish bed was regarded as a trespass (Parsons, 1996); however, this right has since been lost. CHAPTER 2 LITERATURE REVIEW 39

Fisheries management agencies have a mandate to regulate exploitation rates of fish, yet they generally have little control over aquatic habitats, anthropogenic inputs such as the discharge of sewage or the management or monitoring systems relating to sewage discharge (FAO, 1971; Baird, 1996). However, community and government perceptions appear to be changing (Beder, 1989; Earle, 1995; Leadbitter,

1996; NSW Fisheries, 1998; O’Connor, 1998). In Australia and Canada, for example, fisheries agencies are increasing their role in the management of pollution (NSW Fisheries, 1998).

A precautionary management approach may delineate ‘mixing zones’ which are largely based on bacterial assessment and may apply to zones of 100 to 1000 m surrounding ocean sewage outfalls where humans should not swim or consume seafood (EPA, 1997). In some circumstances these zones are legislatively closed to fishing, and the management action of closing a contaminated area raises additional issues such as: Are small area closures effective given that many fish are mobile? What are the health risks to consumers? Are there any compensation implications for fishers as a result of prohibiting fishing?

It is possible to control and reverse the impacts of sewage pollution. There are well documented studies which have reported rapid improvements in aquatic ecosystems, fish and fisheries following the diversion, reduction, increased treatment or cessation of sewage disposal (Oviatt et al., 1984;

GESAMP, 1990; Hardy et al., 1993; TEL, 1993; Scanes and Philip, 1995; Bokn et al., 1996; Scanes,

1996). For example, in southern California, a decision was made to cease sewage disposal through a deep ocean outfall, and to dispose of it on land, and in New York a nearshore disposal site was phased out in favour of an offshore site (Oviatt et al. 1987; Chang, 1993). After 5 months the biota of a previously grossly polluted system in New York was similar to that at two control sites (Oviatt et al.,

1984). Therefore, managers should strive for mitigation or rehabilitation of the impacts of existing sewage outfalls.

Management decisions to reuse effluent, upgrade treatment of sewage or relocate outfalls are expensive

(and may cost hundreds of millions of dollars). To reduce the impact of sewage pollution, and CHAPTER 2 LITERATURE REVIEW 40 particularly the frequency of algal blooms, on aquaculture, the Korean government plans to spend 4 trillion won for the construction of sewage treatment systems (Lee and Chah, 1996).

CONCLUSIONS

Quantitative information on the number of sewage outfalls, effluent volume, level of treatment etc. may be available in reports (NOAA, 1990; EE, 1995; Zann, 1995, EPA, 1998), but in most countries such information is either not available or is too descriptive to be useful (Lean and Hinrichsen, 1992).

Two examples of good integrated information systems are the National Coastal Pollutant Inventory of estuaries produced by NOAA (1990) and the Rivers of Life produced by EPA (1998). These provide information on pollution sources, fishery resources and the relative susceptibility of areas to pollutants.

Although there are many forms of both point source pollutants (including sediments, thermal, toxic chemicals) and diffuse source pollutants which impact on aquatic habitats, fish and fisheries (Gabric and Bell, 1993; Omari et ai, 1994; Kelso et al., 1995; Dubinsky and Stambler, 1996; NSW Fisheries,

1998), few have stimulated as much social and political interest as sewage effluents. This is probably because humans recognise that increasing population growth and associated sewage disposal has resulted in a significant deterioration of the aquatic environment which will probably continue to deteriorate in the next decade unless strong, coordinated national and international action is taken (GESAMP, 1990; Patin,

1995).

This review has focused on the effects of sewage on aquatic habitats, fish and fisheries, and has described a variety of impacts, depending on the type of sewage effluent, habitat, fish and fishery. The selection of well designed research which detected (or did not detect) impacts that could be attributed to sewage pollution will assist scientists and managers to focus on the best available information. A case study of sewage outfalls in NSW, Australia illustrated the lack of comprehensive research, the need for long-term monitoring and the need for research on the impacts of sewage on fisheries. This paper recommends a multidisciplinary approach to detect the impacts of sewage pollution, encompassing research on aquatic habitats, fish and fisheries, using Beyond BACI or gradient CHAPTER 2 LITERATURE REVIEW 41 experimental designs. It is emphasised that better scientific research will lead to improved management outcomes for both aquatic ecosystems and humans.

REFERENCES

Abarnou, A. and Miossec, L. (1992). Chlorinated waters discharged to the marine environment

chemistry and environmental impact. An overview. The Science of the Total Environment 126,

173-197.

ACIL (1997). Oysters at the crossroads: a strategic plan for the New South Wales oyster industry. A

report to the NSW oyster industry steering committee by ACIL Economics and Policy Pty Ltd.,

Canberra.

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium

No. 8, Bethesda, Maryland.

Adams, M. S, Shepard, K.L., Greeley, M.S., Jimenez, B.D., Ryon, M.G., Shugart, L.R. and McCarthy,

J.F. (1989). The use of bioindicators for assessing the effects of pollutant stress on fish. Mar.

Environ. Res. 28, 459-464.

Anderlini, V.C. and Wear, R.G. (1992). The effect of sewage and natural seasonal disturbances on

benthic macrofaunal communities in Fitzroy Bay, Wellington, New Zealand. Mar. Poll. Bull. 24,

21-26.

Anon (1989). Sewage effects in marine and estuarine environments (Jan 77 - Apr 89). Citations from

the NTIS bibliographic database. National Technical Information Service, Springfield, VA.

Anon (1998). Red tide engulfs HK’s fish stocks. The Australian newspaper. 15 April 1998. pp 7-8.

Anttila, R. (1973). Effects of sewage on the fish fauna in the Helsinki area. Oikos 15, 226-229.

ANZECC (1992). Australian water quality guidelines for fresh and marine waters. Australian and New

Zealand Environment and Conservation Council, Canberra.

Arthington, A.H. and Welcomme, R.L (1995). The condition of large river systems in the world. In

Armantrout, N.B. ed. Condition of the world’s aquatic habitats. Proceedings of the World’s

Fisheries Congress, Theme 1. Science Publishers Inc., Lebanon, USA. pp. 44-75

Avery, E.L., Dunstan, R.H. and Nell, J.A. (1996). The detection of pollutant impact in marine

environments: condition index, oxidative DNA damage, and their associations with metal CHAPTER 2 LITERATURE REVIEW 42

bioaccumulation in the Sydney rock oysiev Saccostrea commercial is. Arch. Environ. Contam.

Toxicol. 31, 192-198.

AWT (1997). Fishing and recreational activity survey. Report No. 97/90 by Australian Water

Technologies (AWT) Ensite and Parametrix Inc for Sydney Water Corporation. Sydney. 17 pp.

Baird, R.C. (1996). Towards new paradigms in coastal resource management: linkages and institutional

effectiveness. Est. 19, 320-335.

Bayly, I.A.E. and Williams, W.D. (1973). Inland waters and their ecology. Longman Australia. 314

pp.

Bebianno, M.J. (1995). Effects of pollutants in the Rio Formosa Lagoon, Portugal. In. Vittozzi, L. ed.

Environmental toxicology: hazards to the environment and man in the Mediterranean region,

Italy, pp 107-115.

Beder, S. (1989). Toxic fish and sewer surfing. Allen & Unwin, Sydney.

Bell, J.D. and Harmelin-Vivien, M.L. (1982). Fish fauna of French Mediterranean Posidonia Oceania

seagrass meadows. 1. Assemblage Structure. Tethys 10, 337 - 347.

Bellgrove, A., Clayton, M.N. and Quinn, G.P. (1997). Effects of secondarily treated sewage effluent on

intertidal macroalgal recruitment processes. Mar. Freshwater Res. 48, 137-46.

Bernstein, B. and Dorsey, J. (1991). Marine monitoring in heterogeneous environments. J. Environ.

Manag. 32, 227-240.

Bickford, G.P. (1996). The effects of sewage organic matter on biogeochemical processes within mid­

shelf sediments offshore Sydney, Australia. Mar. Poll. Bull. 33, 168-181.

Birch, G.F. (1996). Sediment-bound metallic contaminants in Sydney’s estuaries and adjacent offshore,

Australia. Est. Coast. Shelf Sci. 42, 31-44.

Birtwell, I.K., Greer, G.L., Nassichuk, M.D. and Rogers, I.H. (1983). Studies on the impact of

municipal sewage discharged onto an intertidal area within the Fraser River estuary, British

Columbia. Can. Tech. Rep. Fish. Aquat. Sci. 1170. 64 pp.

Boates, W.A.M. and Russell, K.S. (1992). The option to discharge industrial or sewage effluent to the

marine environment. Wat. Sci. Tech. 25, 75-82.

Boddeke, R. and Hagel, P. (1995). Eutrophication, fisheries and productivity of the North Sea

continental zone. In Armantrout, N.B. ed. Condition of the world’s aquatic habitats. Proceedings CHAPTER 2 LITERATURE REVIEW 43

of the World’s Fisheries Congress, Theme 1. Science Publishers Inc., Lebanon, USA. pp. 290-

315.

Bokn, T.L., Frithjof, E. and Walday, M. (1996). Improvement of the shallow water communities

following reductions of industrial outlets and sewage discharge in the Hvaler estuary, Norway.

Hydrobiol. 326/327, 297-304.

Bolotova, N.L., Zuyanova, O.V., Zuyanov, E.A. and Tereshchenko, V.G. (1996). Change in the fish

part of community and in catches under eutrophication of a large northern lake. J. Ichthyol. 36,

492-502.

Bothner, M.H., Takada, H., Knight, I.T., Hill, R.T., Butman, B., Farrington, J.W., Colwell, R.R. and

Grassle, J.F. (1994). Sewage contamination in sediments beneath a deep-ocean dump site off New

York. Mar. Environ. Res. 38, 43-59.

Boynton, W. (1997). Chesapeake Bay eutrophication: current status, historical trends, nutrient

limitation and management actions. Proceedings of the Coastal Nutrients Workshop, Environment

Protection Authority, Sydney, pp. 6-11.

Brown, V.B., Davies, S.A. and Synnot, R.N. (1990). Long-term monitoring of the effects of treated

sewage effluent on the intertidal macroalgal assemblage near Cape Schanck, Victoria, Australia.

Bot. Mar. 33, 85-98.

Caddy, J.F. (1995) Towards a comparative evaluation of human impacts on fishery ecosystems of

enclosed and semi-enclosed seas. In Armantrout, N.B. ed. Condition of the world's aquatic

habitats. Proceedings of the World’s Fisheries Congress, Theme 1. Science Publishers Inc.,

Lebanon, USA. pp. 76-87.

Caine, R.F. (1995). Sea outfalls for the disposal and treatment of wastewater effluents. Wat. Sci. Tech.

7, 79-86.

Cambridge, M.L. and McComb, A.J. (1984). The loss of seagrasses in Cockburn Sound, Western

Australia. I. The time course and magnitude of seagrass decline in relation to industrial

development. Aq. Bot. 20, 229-243.

Chang, S. (1993). Analysis of fishery resources: potential risk from sewage sludge dumping at the

deepwater dumpsite off New Jersey. Fish. Bull. 91, 594-610. CHAPTER 2 LITERATURE REVIEW 44

Chapman, M.G., Underwood, A.J. and Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

J. Exp. Mar. Biol. Ecol. 189, 103-122.

Chapman, P.M., Arthur, A.D., Paine, M.D. and Taylor, L.A. (1993). Sediment studies provide key

information on the need to treat sewage discharged to sea by a major Canadian city. Wat. Sci.

Tech. 28,255-261.

Chisholm, J.R.M., Fernex, F.E., Mathieu, D. and Jaubert, J.M. (1997). Wastewater discharge, seagrass

decline and algal proliferation on the Cote d’Azur. Mar. Poll. Bull. 34, 78-84.

Clark, R.B. (1995). Pollution, people, the press and the Bulletin. Mar. Poll. Bull. 30, 2-3.

Collins, A., Stapleton, M. and Whitmarsh, D. (1998). Fishery-pollution interactions: a modelling

approach to explore the nature and incidence of economic damages. Mar. Poll. Bull. 36, 211-221.

Colton, C.E. (1992). Illinois River Pollution Control, 1900-1970. Dilsaver, L.M. and C.E. Colten eds.

In The American Environment, Interpretations of Past Geographies, Rowman & Littlefied

Publishers, Inc, USA. pp. 193-214.

Connor, T.P. (1998). Comparative criteria: land application of sewage sludge and ocean disposal of

dredged material. Mar. Poll. Bull. 36, 181-184.

Costello, M.J. and Gamble, J.C. (1992). Effects of sewage sludge on marine fish embryos and larvae.

Mar. Environ. Res. 33,49-74.

Costello, M.J. and Read, P. (1994). Toxicity of sewage sludge to marine organisms: a review. Mar.

Environ. Res. 37, 23-46.

Diener, D.R, Fuller, S.C., Lissner, A., Haydock, C.I., Maurer, D., Robertson, G. and Gerlinger, T. (1995).

Spatial and temporal patterns of the infaunal community near a major ocean outfall in southern

California. Mar. Poll. Bull. 30, 861-878.

Dubinsky, Z. and Stambler, N. (1996). Marine pollution and coral reefs. Global Change Biology. 2, 511 -

526.

Earle, S.A. (1995). Sea change. A message of the oceans. G.P. Putnam’s Sons, New York. 354p.

Easa, M., Shereif, M.M., Shaaban, A.I. and Mancy, K.H. (1995). Public health implications of waste

water reuse for fish production. In El Gohary, F. ed. Middle East Wastewater Management, Cairo,

Egypt, pp 145-152 CHAPTER 2 LITERATURE REVIEW 45

EE (1995). Europe’s Environment. The Dobris Assessment. Stanners, D. and Bourdeau, P. eds.

European Environment Agency, Copenhagen.

EPA (1992). The concept of the ‘assimilative capacity’ as a management tool in temperate coastal waters

of Western Australia. Environmental Protection Authority, Technical Series 48, Perth, Western

Australia.

EPA (1993 a). Deformities and associated sublethal effects in fish exposed to sewage-borne

contaminants literature review. Environment Protection Authority, Sydney, Australia, 93/72.

EPA (1993b). The effects of nearshore disposal of construction spoil on epibenthic and rocky reeffish

assemblages at Bondi, Sydney, Australia 1985-91. Environment Protection Authority, Sydney. 64

pp.

EPA (1993c). Discharge of wastes to ocean waters. Heggie, A and MacDonald, R. Environment

Protection Authority, Sydney.

EPA (1996). Sydney Deepwater Outfalls Environmental Monitoring Program. Final Report Series.

Volume 1. Environment Protection Authority, Sydney. 194 pp.

EPA (1997). New South Wales State of the Environment 1997. Environment Protection Authority,

Sydney. 500 pp.

EPA (1998). Rivers of Life, Critical Watersheds for Protecting Freshwater Biodiversity. Environmental

Protection Agency, USA. http://www.consci.tnc.org/library/pubs/rivers/index.htm

Fagan, P., Miskiewicz, A.G. and Tate, P.M. (1992). An approach to monitoring sewage outfalls. A case

study on the Sydney Deepwater Sewage Outfalls. Mar. Pol. Bull. 25, 172-180.

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environ. Monit. Assess. 14, 197-210.

FAO (1971). Pollution. An international problem for fisheries. Food and Agricultural Organisation of

the United Nations, Rome. 85 pp.

Ferraro, S.P., Swartz, R.C., Cole, F.A. and Schults, D.W. (1991). Temporal changes in the benthos

along a pollution gradient: discriminating the effects of natural phenomena from sewage-

industrial wastewater effects. Est. Coast. Shelf Sci. 33, 383-407.

Franklin, F.L. (1983). Laboratory tests as a basis for the control of sewage sludge dumping at sea. Mar.

Poll. Bull. 14,217-223. CHAPTER 2 LITERATURE REVIEW 46

Gabric, A.J. and Bell, R.F. (1993). Review of the effects of non-point nutrient loading on coastal

ecosystems. Aust. J. Mar. Freshwater Res. 44, 261-83.

Gappa, J.L, Tablado, A. and Magaldi, N.H. (1993). Seasonal changes in an intertidal community

affected by sewage pollution. Environ. Poll. 82, 157-165.

GESAMP (1990). The state of the marine environment. IMO/FAO/Unesco/WHO/IAEA/UN/UNEP

Joint Group of Experts on the Scientific Aspects of Marine Pollution. UNEP Regional Seas

Reports and Studies No. 115. UNEP.

Gibbs, P.J. and Miskiewicz, A.G. (1995). Heavy metals in fish near a major primary treatment sewage

plant outfall. Mar. Poll. 30, 667-674.

Gray, C.A. (1996). Intrusions of surface sewage plumes into continental shelf waters: interaction with

larval and presettlement juvenile fishes. Mar. Ecol. Prog. Ser. 139, 31-45.

Gray, C.A., Otway, N.M., Laurenson, F.A., Miskiewicz, A.D. and Pethebridge, R. (1992). Distribution

and abundance of marine fish larvae in relation to effluent plumes from sewage outfall and depth

of water. Mar. Biol. 113, 549-559.

Gray, C.A. (1997). Field assessment of numerical impacts of coastal sewage disposal on fish larvae

relative to natural variability. Environ. Biol. Fish. 50, 415-434.

Gray, J.S. (1989a). Effects of environmental stress on species rich assemblages. Biol. J. Linn. Soc. 37,

19-32.

Gray, J.S. (1989b). Ecological theory and marine pollution monitoring. Mar. Poll. Bull. 16, 224-227.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Mar. Ecol. Prog. Ser. 103, 25-34.

Grigg, R.W. (1995). Coral reefs in an urban embayment in Hawaii: a complex case history controlled

by natural and anthropogenic stress. Coral Reefs 14, 253-266.

Grigg, R.W. and Dollar, S.J. (1995). Environmental protection misapplied: alleged verses documented

impacts of a deep ocean sewage outfall in Hawaii. Ambio 24, 125-128.

Griswold, B.L. (1997). Fisheries and Pollution: a conference overview. Trans. Am. Fish. Soc. 126, 504-

505. CHAPTER 2 LITERATURE REVIEW 47

Growns, I.O., Pollard, D.A. and Gehrke, P.C. (1998). Changes in river fish assemblages associated

with vegetated and degraded banks, upstream of and within nutrient-enriched zones. Fish.

Manag. Eco. 5, 55-69.

Haines, E. (1980). Salt marsh creation: impact of sewage. In Rehabilitation and creation of selected

coastal habitats: proceedings of a workshop. Lewis, J.C. and Bunce, E.W. eds. Biol. Serv.

Program Fish. Wildl. Serv. U.S. pp 148-153.

Hall, J.A., Frid, C.L.J. and Gill, M.E. (1997). The response of estuarine fish and benthos to an

increasing discharge of sewage effluent. Mar. Poll. Bull. 34, 527-535.

Hardy, F.G., Evans, S.M. and Tremayne, M.A. (1993). Long-term changes in the marine macroalgae of

three polluted estuaries in north-east England. J. Exp. Mar. Biol. Ecol. 172, 81-92.

Harmelin-Vivien, M. (1992). Impact des activites humaines sur les peuplements ichtyologues des recifs

coralliens de polynesie francaise. Cybium 16, 279-289.

Harris, J.H. (1995). The use of fish in ecological assessments. Aust. J. Ecol. 20, 65-80.

Hurley, D.A. and Christie, W.J. (1977). Depreciation of a warmwater fish community in the Bay of

Quinte, Lake Ontaria. J. Fish. Res. Can. 34, 1849-i860.

Johnston, P.A., MacGarvin, M., Stringer, R.L., Troendle, S. and Swindlehurst, R.J. (1993). Sewage:

towards realistic environmental protection. Wat. Sci. Tech. 27, 481-491. v Kakuta, I. (1997). Effect of sewage on blood parameters and resistance against bacterial infection of

goldfish, Carassius auratus. Environ. Toxicol. Water Qual. 12, 43-51.

Kakuta, I. and Murachi, S. (1997). Physiological response of carp, Cyprinus carpio, exposed to raw

sewage containing fish processing wastewater. Environ. Toxicol. Water Qual. 12, 1-9.

Kelly, M.G. (1998). Use of community-based indices to monitor eutrophication in European rivers.

Environ. Conserv. 25, 22-29.

Kelso, J.R.M., Steedman, R.J. and Stoddart, S. (1995). Historical causes of change in Great Lakes fish

stocks and the implications for ecosystem rehabilitation. Can. J. Fish. Aquat. Sci. 53, 10-19.

Kenchington, R.A. (1990). Managing Marine Environments. Taylor & Francis, New York.

Keough, M.J. and Mapstone, B.D. (1995). Protocol for designing marine ecological monitoring programs

associated with BEK mills. National Pulp Mills Research Program Technical Report No. 11.,

Australia. 150 pp. CHAPTER 2 LITERATURE REVIEW 48

Keough, M.J. and Quinn, G.P. (1991). Causality and the choice of measurements for detecting human

impacts in marine environments. Aust. J. Mar. Freshwater Res. 42, 539-554.

Kingsford, M.J. and Gray, C.A. (1996). Influence of pollutants and oceanography on abundance and

deformities of wild fish larvae. In Detecting Ecological Impacts: Concepts and Applications in

Coastal habitats. Schmitt, R.J. and Osenberg, C.W. eds. Academic Press, pp. 235-255.

Koop, K. and Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Mar. Poll.

Bull. 33, 121-123.

Krogh, M. and Koop, K. (1996). Design and analysis of the Sydney deepwater ocean outfall

environmental monitoring programme: a critique. Mar. Poll. Bull. 33, 273-280.

Krogh, M. and Robinson, L. (1996). Environmental variables and their association with faecal coliform

and faecal streptococci densities at thirteen Sydney beaches. Mar. Poll. Bull. 33, 239-248.

Krogh, M. and Scanes, P.R. (1996). Organochlorine compound and trace metal contaminants in fish

near Sydney’s ocean outfalls. Mar. Poll. Bull. 33, 213-225.

Leach, J.H., Johnson, M.G., Kelso, J.R.M., Hartmann, J., Numann, W. and Entz, B. (1977). Responses of

percid fishes and their habitats to eutrophication. J. Fish. Res. Can. 34, 1964-1971.

Leadbitter, D. (1992). Management of fishing activities in contaminated areas. Aust. Fish. September. 30-

33.

Leadbitter, D. (1996). Sewage and the community - one view of Sydney’s deepwater outfalls, the

environmental monitoring programme and the future of sewage disposal. Mar. Poll. Bull. 33,

269-272.

Lean, G. and Hinrichsen, D. (1992). Atlas of the Environment. Helicon, Oxford.

Lee, Y. and Chah, E.Y. (1996). Marine pollution as an economic problem - the case of harmful algal

blooming. Ocean Res. 18, 173-181.

Leach, J.H., Johnson, M.G., Kelso, J.R.M., Hartmann, J., Numann, W. and Entz, B. (1977). Responses of

percoid fishes and their habitats to eutrophication. J. Fish. Res. Can. 34, 1964-1971.

Lemly, A. (1996). Wastewater discharges may be most hazardous to fish during winter. Environ. Poll.

93,169-174. CHAPTER 2 LITERATURE REVIEW 49

Lenanton, R.C.J., Potter, I.C., Loneragan, N.R. and Chrystal, P.J. (1984). Age structure and changes in

abundance of three important species of teleost in a eutrophic estuary (Pisces: Teleostei). J. Zool.,

Lond. 203, 311-327.

Lincoln Smith, M.P. and Mann, R.P. (1989). Bioaccumulation in nearshore marine organisms II.

Organochlorine compounds in the red morwong Cheilodactylus fuscus, around Sydney’s three major

sewage ocean outfalls. A report prepared by SPCC and the Ecology Lab Pty Ltd, Sydney.

Lincoln Smith, M.P (1998). Assessment of Aquatic Ecology in EIA. Department of Urban Affairs and

Planning, Sydney.

Lipton, D.W. and Strand, I.E. (1997). Economic effects of pollution in fish habitats. Trans. Am. Fish. Soc.

126,514-518.

Littler, M.M. and Murray, S.M. (1975). Impact of sewage on the distribution, abundance and community

structure of rocky intertidal macro-organisms. Mar. Biol. 30, 277-291.

Love, M.S., Axell, B., Morris, P., Collins, R. and Brooks, A. (1987). Life history and fishery of the

Californian scorpionfish, Scorpaena guttata, within the southern Californian Bight. Fish. Bull. 85,

99-116.

Lye, C.M., Frid, C.L.J., Gill, M.E. and McCormick, D. (1997). Abnormalities in the reproductive

health of flounder Platichthys flesus exposed to effluent from a sewage treatment works. Mar.

Poll. Bull. 34, 34-41.

Madden, C.J. and Kemp, W.M. (1996). Ecosystem model of an estuarine submersed plant community:

calibration and simulation of eutrophication responses. Est. 19, 457-47

MAFF (19934.). Analysis and interpretation of benthic community data at sewage-sludge disposal

sites. Compiled by H.L. Rees, Aquat. Environ. Monit. Rep., MAFF Direct. Fish. Res.,

Lowenstoft, (37), 80 pp.

Marneffe, Y., Comblin, S., Bussers, J. and Thome, J. (1996). Biomonitoring of the water quality of the

River Warche (Belgium): impact of tributaries and sewage effluent. Neth. J. Zool. 46, 337-350.

Maurer, D., Robertson, G. and Gerlinger, T. (1993). San Pedro Shelf California: Testing the Pearson-

Rosenberg Model (PRM). Mar. Environ. Res. 35, 303-321.

May, V. (1985). Observations on algal floras close to two sewerage outlets. Cunninghamia 1, 385-394. CHAPTER 2 LITERATURE REVIEW 50

McFeters, G.A. and Watters, S.K. (1991). Distribution of enteric bacteria in Antarctic seawater

surrounding the sewage outfall at McMurdo station. Antarct. J. U.S. 26, 319-320.

McLean, C., Miskiewcz, A.G. and Roberts, E.A. (1991). Effect of 3 primary treatment sewage outfalls

on metal concentrations in the fish Cheilodactylus fuscus collected along the coast of Sydney,

Australia. Mar. Poll. Bull. 22, 134-140.

Miskiewicz, A.G. (1992). Proceedings of a Bioaccumulation Workshop. Published by the Water Board

and Australian Marine Sciences Association Inc. Sydney. 334 pp.

Mitz, S.V. and Giesy, J.P. (1985). Sewage effluent biomonitoring. 1. Survival, growth and

histopathological effects in channel catfish. Ecotoxicology and Environmental Safety 10, 22-39.

MHL (1997). ‘Review of Sewage Effluent Ocean Outfall Performance in NSW’. Manly Hydraulic

Laboratory Report 769. 65 pp.

Montgomery, M. and Needelman, M. (1997). The welfare effects of toxic contamination in freshwater

fish. Land Economics 73, 211-223.

Moore, M.J., Shea, D., Hillman, R.E. and Stegeman, J.J. (1996). Trends in hepatic tumours and

hydropic vacuolation, fin erosion, organic chemicals and stable isotope ratios in winter flounder

from Massachusetts, USA. Mar. Poll. Bull. 32, 458-470.

Mortimer, M.R. and Connell, D.W. (1995). A model of the environmental fate of chlorohydrocarbon

contaminants associated with Sydney sewage discharges. Chemosphere 30, 2021-2038.

NOAA (1990). Estuaries of the United States. National Oceanic and Atmospheric Administration,

Maryland, USA.

NOAA (1991). The national estuarine eutrophication project: workshop proceedings. National Oceanic

and Atmospheric Administration and the University of Rhode island Graduate School of

Oceanography, USA.

NSW Fisheries.(1998). Policy and guidelines: Aquatic habitat management and fish conservation.

Smith, A.K., Pollard, D.A. eds. NSW Fisheries, Sydney. 76 pp.

Oakley, H.R. (1979). Where should the sewage go? Mar. Poll. Bull. 10, 155-156.

Omori, K., Hirano, T. and Takeoka, H. (1994). The limitations to organic loading on a bottom of a

coastal ecosystem. Mar. Poll. Bull. 28, 73-80. CHAPTER 2 LITERATURE REVIEW 51

Otway, N.M. (1995). Assessing impacts of deepwater sewage disposal: a case study from New South

Wales, Australia. Mar. Poll. Bull. 31, 347-354.

Otway, N.M, Gray, C.A., Craig, J.R., McVea, T.A. and Ling, J.E. (1996a). Assessing the impacts of

deepwater sewage outfalls on spatially and temporally variable marine communities. Mar.

Environ. Res. 41,45-71.

Otway, N.M, Sullings, D.J. and Lenehan, N.W. (1996b). Trophically-based assessment of the impacts

of deepwater sewage disposal on a demersal fish community. Environ. Biol. Fish. 46, 167-183.

Oviatt, C.A., Pilson, M.E.Q., Nixon, S.W., Frithsen, J.B., Rudnick, D.T., Kelly, J.R., Grassle, J.F. and

Grassle, J.P. (1984). Recovery of a polluted estuarine system: a mesocosm experiment. Mar.

Ecol. Prog. Ser. 16, 203 - 217.

Oviatt, C.A., Quinn, J.C., Maughan, J.T., Ellis, J.T., Sullivan, B.K., Gearing, J.N., Gearing, P.J., Hunt,

C.D., Satnpou, P.A. and Latimer, J.S. (1987). Fate and effects of sewage sludge in the coastal

marine environment: a mesocosm experiment. Mar. Ecol. Prog. Ser. 41, 187-203.

Paller, M.H., Heidinger, R.C. and Lewis, W.M. (1988). Impact of nonchlorinated secondary and

tertiary effluents on warm water fish communities. Water Resour. Bull. 24, 65-76.

Parsons, G. (1996). English inshore fisheries 1890-1990: one hundred years of the acceptance of

sewage pollution. In Partnerships in Coastal Zone Management. Taussik, J., Mitchell, J. eds.

Samara Publishing Pty Ltd, Cardigan, UK. pp 327-335.

Pastorok, R. and Bilyard, G.R. (1985). Effects of sewage pollution on coral-reef communities. Mar.

Ecol. Prog. Ser. 21, 175-189.

Patin, S.A. (1995). Global pollution and biological resources of the world ocean. In Condition of the

world’s aquatic habitats. Armantrout, N.B. ed. Proceedings of the World’s Fisheries Congress,

Theme 1. Science Publishers Inc., Lebanon, U.S.A. pp. 81-87.

Pearson, T.H. and Rosenberg, R. (1978). Macrobenthic succession in relation to organic enrichment

and pollution of the marine environment. Oceanogr. Mar. Biol. A. Rev. 16, 229-311.

Pilanowski, R.A. (1992). The effects of ocean disposal of sewage sludge on the relative abundance of

benthic megafauna. Chemistry and Ecology 6, 199-212.

Pritchard, T., Ajani, P., Andrew, D., Calfas, M., Holden, C., Lee, R., Linforth, S. and Rendell, P.

(1997). Relative significance of slope water, estuarine discharges and sewage outfalls for CHAPTER 2 LITERATURE REVIEW 52

nutrients in offshore coastal waters. In Proceedings of the Coastal Nutrients Workshop,

Environment Protection Authority, Sydney, pp. 44-50.

Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, N.C., Tyler, C.R. and Sumpter, J.P. (1994). Estrogenic

effects of effluents from sewage treatment works. Chem. Ecol. 8, 275-285.

Quentin, B. and De Rouville, M. (1986). Submarine sewer outfalls - a design manual. Mar. Poll. Bull.

17, 133-183.

Reish, D.J. and Bellan, G. (1995). The long-term effects of municipal discharges from urban areas on

the marine environment: a review. In Functioning and dynamics of natural and perturbed

ecosystems. Bellan, D., Bonin, G., Emig, C. eds. Lavoisier, France, pp 701-743.

Roberts, D.E. (1996). Patterns in subtidal marine assemblages associated with a deep-water sewage

outfall. Aust. J. Mar. Freshwat. Res. 47, 1-9.

Roberts, D.E., Smith, A., Ajani, P. and Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACE approach. Mar.

Ecol. Prog. Ser. 163, 213-224.

Russo, A..R. (1982). Temporal changes in fish community structure near a sewage ocean outfall,

Mokapu, Oahu, Hawaii. Mar. Environ. Res. 6, 83-98.

Russo, A.R. (1989). Fish community structure on a deep ocean outfall, Barber’s Point, Oahu, Hawaii.

!nt. Revue, ges. Hydrobiol. 74, 499-506.

Scanes, P.R. (1996). ‘Oyster Watch’: monitoring trace metal and organochlorine concentrations in

Sydney’s coastal waters. Mar. Poll. Bull. 33, 226-238.

Scanes, P.R. (1997). Uptake and depuration of organochlorine compounds in Sydney rock oysters

{Saccostrea commercials). Mar. Freshwater Res. 48, 1-6.

Scanes, P.R., Coade, G., Large, D. and Roach, T. (1997). Developing criteria for acceptable loads of

nutrients from catchments. Proceedings of the Coastal Nutrients Workshop, Environment

Protection Authority, Sydney, pp. 89-99.

Scanes, P.R. and Philip, N. (1995). Environmental impact of deepwater discharge of sewage off

Sydney, NSW, Australia. Mar. Poll. Bull. 31, 343-346. CHAPTER 2 LITERATURE REVIEW 53

Scanes, P.R., Scanes, K. and Otway, N.M. (1995). Environmental problems due to disposal of wastes.

In Coastal marine ecology of temperate Australia. Underwood, A.J., Chapman, M.G. eds, UNSW

Press, Sydney, Australia, pp. 297-310.

Schmitt, R.J. and Osenberg, C.W. (1996). Detecting ecological impacts, concepts and applications in

coastal habitats. Academic Press, New York. 198 pp.

Seager, J. and Abrahams, R.G. (1990). The impact of storm sewage discharges on the ecology of a

small urban river. Wat. Sci. Tech. 22, 163-171.

Sherwin, T.J. and Jonas, P.J.C. (1994). The impact of ambient stratification on marine outfall studies in

British waters. Mar. Poll. Bull. 28, 527-533.

Short, F.T. and Wyllie-Echeverria, S. (1996). Natural and human-induced disturbance of seagrasses.

Environmental Conservation 23, 17-27.

Sindermann, C.J. (1997). The search for cause and effect relationships in marine pollution studies.

Mar. Poll. Bull. 34,218-221.

Skilleter, G.A. (1995). Environmental disturbances. In Coastal marine ecology> of temperate Australia.

Underwood, A.J. and Chapman, M.G. eds. UNSW Press, Sydney, Australia, pp. 263-276

Smith, A.K., Ajani, P.A. and Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Marine Environmental Research 47, 1-20.

Smith, A.K., Holliday, J.E. and Pollard, D.A. (1997). Management of seagrass habitats in NSW

estuaries. Wetlands (Australia). 16,48-55.

Smith, S.D. (1994). Impacts of domestic sewage effluent verses natural background variability: an

example from Jervis Bay, New South Wales. Aust. J. Mar. Freshwater Res. 45, 1045-1064.

Steimle, F.W. (1994). Sewage sludge disposal and winter flounder, red hake, and American lobster

feeding in the New York Bight. Mar. Environ. Res. 37, 233-256.

Stephens, J.S, Jr., Hose, J.E. and Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In Marine Organisms as Indicators. D.F. Soule, G.S. Kleppel

eds. Springer-Verlag NY. pp 91-105.

SW (1997). Sydney Water Annual Report 1997, Sydney. CHAPTER 2 LITERATURE REVIEW 54

Tegner, M.J., Dayton, P.K., Edwards, P.B., Riser, K.L., Chadwick, D.B., Dean, T.A. and Deysher, L.

(1995). Effects of a large sewage spill on a kelp forest community: catastrophe or disturbance?

Mar. Environ. Res. 40, 181-224.

TEL (1993). Changes to assemblages of reeffish adjacent to the North Head and Bondi cliffface

outfall following decommissioning. Final Report by The Ecology Lab the Water Board, Sydney.

49 pp.

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final Report by The Ecology Lab for the Water Board, Sydney. 29 pp.

Tsai, C. (1968). Effects of chlorinated sewage effluents on fishes in Upper Patuxent River, Maryland.

Chesapeake Science 9, 83-93.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA. 229 pp.

Tsai, C. (1991). Rise and fall of the Potomac River striped bass stock: a hypothesis of the role of

sewage. Trans. Am. Fish. Soc. 120, 1-22.

Udy, J.W. and Dennison, W.C. (1997). Physiological responses of seagrass used to identify

anthropogenic nutrient inputs. Mar. Freshwat. Res. 48, 605-614.

Underwood, A.J. (1991). Beyond BAC1: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Mar. Freshwat. Res. 42, 569-587.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. J. Exp. Mar. Biol. Ecol. 161, 145-178.

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Aust. J. Ecol. 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15.

Underwood, A.J. (1996). Detection, interpretation, prediction and management of environmental

disturbances: some roles for experimental marine ecology. J. Exp. Mar. Biol. Ecol. 200, 1-27. CHAPTER 2 LITERATURE REVIEW 55

Underwood, A.J. and Chapman, M.G. (1996). Subtidal assemblages on rocky reefs at a cliff-face

sewage outfall (North Head, Sydney, Australia): What happened when the outfall was turned off?

Mar. Poll. Bull. 33, 293-302.

Underwood, A.J. and Peterson, C.H. (1988). Towards an ecological framework for investigating

pollution. Mar. Ecol. Prog. Ser. 46, 227-234.

UNEP (1997). Global Environmental Outlook. United Nations Environment Programme. Global State

of the Environment Report 1997. Executive Summary.

http://www.unep.org/unep/eia/geol/exsum/ex3.htm

Volodin, V.M. (1994). Status of breeding stocks of burbot, Lota lota, in the Sheksninskii reach of

Rybinskoe Reservoir before and after emergency sewage disposal in 1987. J. Ichthyol. 34, 1828.

Vukadin, I. (1992). Impact of nutrient enrichment and its relationship to the algal bloom in the Adriatic

Sea. In Marine Coastal Eutrophication, Vollenweider, R.A., Marchetti, R. and Viviani, R. eds.

Italy, pp 365-370.

Waring, C.P., Stagg, R.M., Fretwell, K., McLay, H.A and Costello, M.J. (1996). The impact of sewage

sludge exposure on the reproduction of the sand goby, Pomatoschistus minutus. Environ. Poll. 93,

17-25.

Weis, P., Weis, J.S. and Greenberg, A. (1990). Treated municipal wastewaters: effects on development

and growth of fishes. Mar. Environ. Res. 28, 1-4.

Weng, H.T. (1990). Fish in shallow areas in Moreton Bay, Queensland and factors affecting their

distribution. Estuar. Coast. Shelf Sci. 30, 569-578.

Williams, B.L (Ed) (1985). Ocean outfall handbook. Water and Soil Miscellaneous Publication No 76,

Wellington, New Zealand, 219 p.

Wright, I.A., Chessman, B.C., Fairweather, P.G. and Benson, L.J. (1995). Measuring the impact of

sewage effluent on the macro in vertebrate community of an upland stream: the effect of different

levels of taxonomic resolution and quantification. Aust. J. Ecol. 20, 142-149.

Zann, L.P. (1995). Our Sea, Our Future. Major Findings of the State of the Marine Environment Report

for Australia. Department of Environment, Sport and Territories, Canberra. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 56

CHAPTER 3

A BASELINE SURVEY OF FISH AND MACROBENTHIC

ASSEMBLAGES AT TWO SEWAGE OUTFALLS IN THE

ILLAWARRA COASTAL ZONE

ABSTRACT

Assemblages of rocky reef fish and macrobenthos adjacent to ocean sewage outfalls and control locations were surveyed on four occasions over a one year period. The subtidal habitats adjacent to the

Bellambi outfall were shallow (2 to 4 m) and were analysed separately from the deeper (12 to 16 m) habitats adjacent to Port Kembla outfall. A total of 58 species of fish and 5,746 individuals were recorded from the Bellambi study and 79 species of fish and 13,106 individuals from the Port Kembla study. The fish fauna was dominated by small, mobile, schooling species. Species richness was relatively small compared to other studies of fish in NSW. The assemblages of macrobenthos were dominated by macroalgae, Ecklonia radiata, Caulerpa filiformis, turfing and encrusting algae and the sea urchin Centrostephanus rodgersii. Multivariate analyses indicated inconsistent differences between assemblages of fish from the Bellambi outfall and control locations, but ther were consistent differences between assemblages of fish from the Port Kembla outfall and control locations. These differences were due to lesser abundances of Trachinops taeniatus (Plesiopidae), Trachurus novaezelandiae (Carangidae) and Chromis hypsilepis (Pomacentridae) at the outfall site. Univariate analyses indicated few differences in fish assemblages from the Bellambi study that could be attributed to sewage pollution. Significantly fewer Trachinops taeniatus, Hypoplectrodes mccullochi

(Serranidae) and species of cryptic fish, and more Crinodus lophodon (Aplodactylidae) and

Cheilodactylus fuscus (Cheilodactylidae) were recorded at the Port Kembla sewage outfall. These differences may be indicative of sewage pollution, but as no ‘before’ data were obtained, other factors such as topography may also contribute to differences between the outfall and control locations. This CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 57 baseline research will be important if the shoreline outfalls are extended offshore, treatment of sewage is upgraded or the volume of sewage discharge is increased.

INTRODUCTION

Options for sewage disposal include discharge of effluent to ocean, rivers and estuaries, dunes and artificial wetlands; local management; minimisation of inputs; upgrading sewage treatment; construction of offshore outfalls to replace the shoreline outfalls; and reuse (GHD 1990a,

1990b, Hall and Rissik 1992, EPA 1995, Koop and Hutchings 1996, EPA 1997). The major volume of sewage (about 74%) is discharged to ocean waters in NSW (EPA 1997). Sewage contains organic (and inorganic) matter which depletes oxygen in the receiving waters in the immediate vicinity of discharge, and for some distance away from the point of release (Carefoot and Simpson 1977). The extent of sewage pollution is dependent on the type of pollutant, its duration and intensity and physical and biological characteristics of the outfall location (Tsai

1975, Pearson and Rosenberg 1978, Scanes et al. 1995).

Ecology of fish and macrobenthos from rocky reefs

Rocky shores comprise about 33% of the NSW coastline (Fairweather 1990). Subtidal rocky reef environments are characterised by great heterogeneity in habitat structure (Foster 1990,

Underwood et al. 1991) and were classified as Fringe, Pyura, Phyllospora, Barrens, Turf,

Ecklonia and Deep Reef in NSW waters by Underwood et al. (1991). The distribution and patchiness of habitats are related to physical properties of the reef, numerical dominance of sessile invertebrates and combinations of depth, wave exposure and a number of biological processes (Jones and Andrew 1990, Underwood and Kennelly 1990, Underwood et al. 1991,

Kennelly and Underwood 1992, Underwood and Chapman 1995). The principal biological assemblages of subtidal rocky reefs are macrobenthos (algae and invertebrates) and fish. The macrobenthic assemblages of coastal rocky reefs comprise algae (red, brown, green), ascidians

(tunicates), poriferans (sponges), cnidarians (corals, hydroids, sea anemones, sea-whips and sea- pens), bryozoans (sea mosses), echinoderms (sea urchins, sea stars), crustaceans (crayfishes, CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 58 crabs) and molluscs (chitons, limpets, bivalves, snails, octopuses, squids, cuttlefishes) (Dakin

1987, Underwood and Chapman 1995).

Qualitative and quantitative descriptions of subtidal marine assemblages in temperate Australia were reviewed by Underwood and Kennedy (1990) and Underwood et al. (1991). The subtidal habitats for the Bellambi study may be described as ‘Fringe’ which extends to depths of 2 to 3 m and is dominated by Ecklonia radiata, fucoids, dictyotalean algae and geniculate corallines.

There is relatively little crustose coralline algae in the fringe habitat although echinoids and gastropods may be locally abundant. The subtidal habitats for the Port Kembla study may be described as ‘Barrens’ habitat which occurs to depths of 23 m, has very little foliose algae, and is dominated by crustose coralline algae. Limpets and Centrostephanus rodgersii are abundant

(Underwood et al 1991).

Fish are a diverse and abundant component of temperate rocky reefs and may be permanent or temporary residents. Permanent residents include fish from the families Plesiopidae,

Pempherididae, Scorpididae, Pomacentridae and Labridae (Lincoln Smith and Jones 1995).

Temporary residents include Carangidae and Sparidae. Many species of rocky reef fish are of value for commercial or recreational fishing (Lincoln Smith et al. 1989, Kingsford et al. 1991).

Spatial scales of variation in assemblages of fish have been reported to be due to variable recruitment (Lincoln Smith et al. 1991, Sale 1991), variation in local habitats (Choat and Ayling

1987, McCormick 1989, Holbrook et al. 1990, 1994), behaviour of individuals (Jones and

Andrew 1990, Spotte et al. 1992) and anthropomorphic effects (Lincoln Smith and Jones 1995).

Temporal scales of variation in assemblages of fish may be due to tide, time of day, lunar phases, seasons and long term environmental effects (Lincoln Smith et al. 1992, Holbrook et al.

1994). CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA

Impacts of sewage on fish and macrobenthos

Fish and macrobenthic assemblages may exhibit a range of responses in a polluted area including death, physiological change or altered patterns of community structure or distribution and/or abundance of individual species (Adams 1990). Benthic flora and fauna are relatively diverse, abundant and sedentary and therefore regarded as good indicators of environmental change. Rocky reef fish have similar characteristics. Several studies have focused on the effect of sewage on marine assemblages (May 1985, Brown et al. 1990, Fairweather 1990, Smith

1994, Chapman et al. 1995, Underwood and Chapman 1996, Roberts et al. 1998). In general, striking effects are very localised (10’s of metres), but detectable effects may be found at distances up to several kilometres (Brown et al. 1990). A few algal genera are consistently reported at sites affected by sewage: Corallina, Codium, Enteromorpha, Viva, Gelidium and

Pterocladia (May 1985). Dominance of the alga Caulerpa filiformis may be due to local increased pollution of the water (May 1976). The rate of recruitment of algae was lower at the

Bellambi outfall compared to control locations (EPA 1995).

There are few studies of the responses of fish assemblages to sewage. Fish assemblages in polluted locations may be characterised by fewer species with large abundances (Bell and

Harmelin-Vivien 1982, Tsai et al. 1991). Enhanced fish populations near outfalls at Hawaii were reported by Grigg (1994). Similar rocky reef fish assemblages were reported at two small outfalls and control locations in the Sydney region (Lincoln Smith 1985). Significantly different fish assemblages have been reported adjacent to larger sewage outfalls in the Sydney region

(TEL 1993). In this latter study the abundance of Trachinops taeniatus, Trachurus novaezelandiae and Parma microlepis were greater at control sites than outfall sites. In contrast,

Acanthopagrus australis (Sparidae) and red morwong Cheilodactylus fuscus were found in greater numbers at the outfall sites.

The physical processes and potential ecological effects of the Bellambi and Port Kembla sewage discharge on the marine assemblage have been categorised by GHD (1990a, 1990b) as: Bulli Point

Collins Rk.

*•' T Bellambi

WOLLONGONG

a Flinders Is. oBass Is. PORT KEMBLA

Port Kembla

SHELLHARBOUR \ Shellharbour

Bass Pt.

□ Atchinson’s Rock

Figure 3.1. Sampling locations for fish and macrobenthos at the Bellambi sewage outfall (•) and control locations (O) and the Port Kembla sewage outfall (■) and controls (□) in the Illawarra region. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 60

(1) Discharge of large volumes of fresh water. It has been postulated that fish (in particular sea

mullet Mugil cephalus (Mugilidae)) may interact with the plug of fresh water from the Port

Kembla outfall, and alter their migration directions from an inshore to an offshore route.

(2) Discharge of suspended solids. It has been postulated from qualitative inspections of the

outfalls that suspended solids would cause an inner and outer impact zone. The small inner zone

(up to 20 m from the outfall at Port Kembla and up to 40 m from the outfall at Bellambi) would

support few if any benthic animals, and the outer zone (of up to 100 m from the outfall) would

support a very diverse fauna that is different from natural benthic assemblages. It is asserted that

rocky reef fish are attracted to the outfalls and use the particles of effluent as food.

(3) Discharge of soluble nutrients. In general, some fast growing species of algae are favoured

by nutrient enriched waters, but this is modified by the limitation of light (via a shading effect of

the plume) GHD (1990b).

The aim of this study was to quantify the abundance and richness of fish and macroinvertebrates

at the sewage outfalls of Bellambi and Port Kembla compared to control locations.

METHODS

Site descriptions

All surveys were undertaken in the Illawarra region (Figure 3.1). There are five ocean outfalls in the

Illawarra region: Bellambi, Coniston Beach, Port Kembla, Barrack Point (Shellharbour) and Bombo

(Kiama). Bellambi and Port Kembla discharge primary treated effluent, and the others have secondary

treatment. Bellambi and Port Kembla outfalls discharge small to medium volumes (49 and 70 Million

Litres Day (MLD) respectively) of effluent compared to large outfalls in the Sydney region (up to 500

MLD, MHL 1997). However, both sewerage treatment plants are presently operating above capacity,

are exceeding license requirements for some pollutants and have high levels of faecal coliform bacteria

(Table 3.1). Table 3.1. Volume, treatment and characteristics of Bellambi outfall and Port Kembla outfall (from

EMU (1991b) and GHD (1990a, 1990b). EP = Equivalent Persons, MLD: Million Litres a day.

Bellambi Port Kembla

Date Commissioned late 1950’s late 1950’s

Design capacity 74,000 EP 45,000 EP

Flow average (MLD) 16.8-23.9 13.1 - 15.1

Maximum flow (MLD) 49.0 70.0 +

Treatment Primary Primary

Type of sewage Domestic Domestic/Industrial

Suspended solids (tonnes) per day 2.2 1.7

Zinc (ug/1) 131 243

Copper (ug/1) 94 89

Cyanide (ug/1) 27 27

Chlordane <1 <1

Heptachlor <0.1 <0.1

Hexachlorobenzene <0.1 <0.1

Mercury <0.5 <0.5

Faecal coliform (counts per 100 ml) 10 x 106 10 x 106 CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 61

The subtidal habitats adjacent to the Bellambi sewage outfall were shallow (2 to 4 m) compared to deeper (12 to 16 m) habitats adjacent to Port Kembla sewage outfall and therefore the studies are described and analysed separately.

Bellambi study

The sewage outfall at Bellambi is located about 50 m offshore, in about 2-4 m water depth

(Figure 3.1). The underwater topography has generally large flat shelves with occasional crevices and boulders. Major habitats include Ecklonia, mixed reef, and sand channels (EMU

1991b). The study sites are 75-100 m from shore and 50 m to the north/east and south/east of the discharge pipe. The control locations at Bellambi and Shellharbour were selected because of similar depth, topography and habitats to the impact location (Figure 3.1).

Port Kembla study

The sewage outfall at Port Kembla is on the eastern end of Red Point (Figure 3.1). The outfall pipe discharges in approximately 3 m of water. The water depth below the pipe is 12-16 m. Port

Kembla has a greater mixture of subtidal habitats and depths than Bellambi (EMU 1991b). The control locations at Flinders Island and Atchinsons Rock were selected because of similar depth, topography and habitats to the impact location (Figure 3.1).

Survey methodology

Fish were surveyed in 2-4 m depths of water for the Bellambi study and 12-16 m depth for the

Port Kembla study. Surveys of fish and macrobenthic assemblages were completed each quarter of the year (winter and spring 1992, summer and autumn 1993), with each survey being completed within a 2 week period. The sewage outfalls were turned off approximately 30 minutes before surveys of the outfalls, by which time the water had usually cleared. A minimum visibility of 5 m and a low swell (less than 1 m) were arbitrarily set as minimum criteria at outfall and control locations to effectively survey fish. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 62

Within each location, sites were sampled along four, 60 metre transects. Two sites were sampled for the Bellambi study (north and south) and one for the Port Keinbla study (boulder habitat).

Kelp habitat was also surveyed at Port Kembla (see TEL 1994) but the results are not reported here for consistency with the studies of fish in the boulder habitats in the Hunter and Sydney regions. Each transect was laid from a boat and orientated parallel to the shoreline. Fishes were counted by a SCUBA diver and classified into two groups; mobile fishes and cryptic fishes, which were surveyed using separate techniques (Lincoln Smith 1988, 1989). Mobile fishes were counted one metre either side of the transect line (total area 120 m^). The diver did not swim at a fixed speed, but swam fast enough to avoid double counting of fish. Cryptic fishes were counted along the same transect after the count of mobile fishes was completed. Cryptic fishes were counted over a smaller area (30 x 2 m) at a swim speed of 3 min per 10 m for Bellambi, and 6 min per 10 m for Port Kembla, determined from a pilot study (TEL 1992). The diver searched crevices, caves, under kelp and around sea urchins. Fish in each transect took approximately 30 minutes to survey and all data were recorded on slates.

Large, conspicuous macrobenthos were selected for survey based on their abundance during preliminary observations and ecological or economic significance. Macrobenthos were counted on the same dates, locations, sites and transects as fish. Within each transect, however, three replicate belt transects of 10 x 1 m were sampled. Quadrats were sampled because of the patchiness of habitats, and the sessile behaviour and high relative abundance of macroinvertebrates.

A SCUBA diver recorded percentage cover of primary and secondary algae, abundance of selected macroinvertebrates and reef topography. The primary cover of algae comprised the canopy-forming species Ecklonia, other macroalgae, and Caulerpa. The secondary cover comprised turfing algae and encrusting algae. Five species of sea urchins (Centrostephanus rodgersii, Heliocidaris erythrogramma, Heliocidaris tuberculata, Phyllacanthus parvispinus and Tripneustes gratilla), four species of gastropods (Cabestana spengleri, Thais orbita and

Turbo torquatus, Australium tentoriforme) and Haliotis ruber, Jasus verreauxii, Octopus spp, CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 63

Sepia spp. were counted. Reef topography was classified as simple, medium or complex. These classifications were scored as: simple=l, medium=2 and complex=3.

Analyses of fish

Data were analysed using non-metric multi-dimensional statistical (MDS) techniques (Clarke 1993) to describe the variation between species abundance and composition at each location, using the PRIMER software package (Plymouth Marine Laboratories, U.K.). The Bray-Curtis similarity measure and untransformed data were used in classification and ordination analyses. Fish assemblages from each location and time (derived by pooling (sites and) transects) were examined. Pooling is often used in this way when the sample size is relatively large to enable easier interpretation of the main factors of interest in two dimensional plots (Clarke 1993). One way Analysis of Similarities (ANOSIM) was used to test the null hypothesis that “fish assemblages are similar between locations”. The SIMPER procedure (Clarke 1993) was used to identify the contribution of individual species to differences between locations.

Univariate techniques, primarily asymmetrical analysis of variance (ANOVA), were used to determine spatial and temporal differences among populations. Prior to analysis, data were tested for homogeneity of variance by Cochran’s test and transformed where necessary

(Underwood, 1981). The standard transformations used in this study were log (x+1) and sqrt

(x+1). A larger number of variates were collected than analysed (TEL 1994). A reduced number of variates was tested to decrease the probability of committing a Type 1 error. The criteria used to choose variates for analysis required the variate to be abundant and ecologically or commercially important.

A 3-way Asymmetrical Analysis of Variance (ANOVA) was used to compare fish assemblages and individual species at locations, sites nested within locations (North and South) and times for the Bellambi study and a 2-way asymmetrical ANOVA for the Port Kembla study (Appendix

1). These designs were based on procedures developed by Underwood (1991, 1992, 1993, 1994,

1997). All factors were considered as random factors. Table 3.2. Numbers and type of fish, percentage cover (%) of alga, reef complexity and numbers of

macrobenthos from two sewage outfalls and four control locations in the Illawarra region. M - mobile,

C - cryptic, BE - Bellambi, BU - Bulli, SH - Shell Harbour, PK - Port Kembla, FI - Flinders Island,

AR - Atchinsons Rock, O - Outfall, C - Control, * - economically important species, T - Tropical

species.

Family Site Type BE BU SH PK FI AR 0 CC O C C Heterodontidae Heterodontus portusjacksoni M 1 0 0 0 0 5 Brachaeluridae Brachaelurus waddi M 1 2 0 0 0 0 Orectolobidae Orectolobus ornatus * C 0 0 1 0 0 0 Rhinobatidae Trygonorhina fasciata M 0 0 2 0 0 0 Dasyatididae Dasyatis brevicaudatus M 1 1 0 0 0 0 Myliobatidae Myliobatis australis M 1 0 0 0 0 0 Urolophidae Urolophus testaceus M 1 0 0 0 0 0 Muraenidae Gymnothorax prasinus C 1 4 6 0 6 5 Aulopidae Aulopus purpurissatus * M 0 0 0 1 4 1 Plotosidae Cnidoglandis macrocephalus C 3 1 0 0 0 0 Batrachoididae Batrachomoeus dubius C 0 1 0 0 0 0 Gobiesocidae Aspasmogaster costatus c 0 1 4 2 5 9 Moridae Lotella rhacinus c 0 0 15 2 0 19 Trachichthyidae Trachichthys australis c 0 0 0 0 1 6 Scorpaenidae Scorpaena cardinalis * c 2 1 2 3 14 4 Triglidae Lepidotrigla papilio * M 0 0 0 0 0 1 Serranidae Acanthistius ocellatus c 1 0 9 0 1 3 Anthias squamipinnis T M 0 0 0 0 8 0 Hypoplectrodes mccullochi C 0 25 0 7 75 81 Hypoplectrodes nigrorubrum c 0 0 0 1 0 0 Trachypoma macracantlius c 0 0 2 0 0 0 Plesiopidae Trachinops taeniatus M 0 188 1302 358 1974 1215 Dinolestidae Dinolestes lewini * M 1 2 1 115 232 0 Carangidae Pseudocaranx dentex * M 0 10 0 100 0 0 Trachurus novaezelandiae * M 0 90 80 64 237 200 Mullidae Parupeneus signatus * M 2 3 0 17 5 5 Upeneichthys lineatus * M 0 0 0 6 3 1 Monodactylidae Schuettea scalaripinnis M 0 1 7 1 17 2 Pempherididae Pempheris affinis C 0 10 0 46 2 0 Pempheris compressus C 0 0 0 164 21 91 Pempheris multiradiata c 0 0 0 15 5 120 Girellidae Girella elevata * M 0 2 0 2 0 0 Girella tricuspidata * M 0 1 7 9 0 0 Kyphosidae Kyphosus sydneyanus M 0 0 0 3 0 0 Scorpididae Atypichthys strigatus M 271 613 1241 373 828 182 Microcanthus strigatus C 0 0 0 1 0 0 Scorpis lineolatus * M 48 19 107 34 14 88 Enoplosidae Enoplosus armatus M 1 0 0 11 1 0 Pomacentridae Chaetodon guntheri T C 0 0 0 0 0 0 Chromis hypsilepis M 0 2 57 53 308 276 Chromis spp. T M 0 0 0 0 1 0 Parma microlepis C 11 55 107 109 93 98 Parma unifasciata C 2 71 133 36 47 7 Parma victoriae c 0 1 0 0 0 0 Pomacentrus coelestis T M 0 11 10 0 0 0 Stegastes gascoynei T c 0 0 3 0 0 0 Chironemidae Chironemus marmoratus c 9 18 42 30 6 2 Aplodactylidae Crinodus lophodon M 85 144 79 139 17 20 Cheilodactylidae Cheilodactylus fuscus * M 26 20 13 47 15 14 (PTO) Table 3.2 contd Cheilodactylus vestitus M 0 0 0 1 0 0 Nemadactylus douglasii * M 0 0 0 1 0 0 Latrididae Latridopsis forsteri * M 0 0 0 0 0 1 Labridae Achoerodus viridis * M • 7 14 21 20 15 17 Coris picta M 0 0 0 0 0 2 Coris sandangeri * M 0 0 0 1 22 1 Eupetrichthys angustipes M 0 0 2 6 8 0 Labroides dimidiatus T M 0 1 0 0 1 0 Notolabrus gymnogenis M 72 105 59 23 26 23 Macropharygodon meleagris T M 0 0 0 0 1 0 Ophthalmolepis lineolatus M 0 2 1 32 70 6 Pictilabrus laticlavius M 25 30 9 11 3 2 Pseudolabrus guntheri M 0 3 1 0 3 0 Labrid spl (white/br stripe) T M 2 0 0 0 1 0 Labrid sp2 (green) T M 2 0 0 0 0 0 Odacidae Odea cyanomelas M 9 48 58 2 3 0 Odea acroptilus M 0 0 0 2 11 4 Blenniidae Petroscirtes fallca C 0 1 0 0 0 0 Pictiblennius tasmaniensis C 1 4 5 1 0 0 Plagiotremis rhinorhynchos T c 0 0 0 0 2 2 Plagiotremus tapeinosoma M 0 0 3 0 1 0 Tripterygiidae Norfolkia clarkei c 136 21 57 13 9 2 Clinidae Cristiceps spp. c 3 3 3 0 0 0 Meter oclinus perspicallatus c 0 1 0 0 0 1 Scombridae Scomber australasicus * M 0 2 0 0 0 0 Acanthuridae Prionurus microlepidotus * M 0 0 2 0 1 0 Zanclus cornutus T M 0 0 0 0 1 0 Monacanthidae Brachaluteres jacksonianus C 0 0 0 1 0 0 Eubalichthys bucephalus * M 0 0 0 7 2 6 Eubalichthys mosaicus * M 0 0 0 0 2 1 Meuschenia flavolineata * M 0 0 0 13 8 2 Meuschenia freycineti * M 0 1 1 1 4 0 Meuschenia trachylepis * M 0 1 4 1 0 0 Monacanthus chinensis * M 0 0 0 1 0 0 Penicipelta vittiger * M 0 0 0 4 26 3 Scobinichthys granulatus * M 0 0 0 0 2 0 Ostraciidae Anoplocapros inermis M 0 0 0 1 1 0 Tetraodontidae Canthigaster callisterna T M 0 0 0 0 0 1 Tetractenos glaber C 30 1 0 0 0 0

MACROBENTHOS Ecklonia radiata % 2.6 14.7 1.7 30.7 33.7 8.0 other macroalgae % 46.8 32.4 34.6 1.3 6.4 0.3 turfing algae % 21.9 35.0 40.0 43.3 36.6 19.6 encrusting algae % 10.2 22.1 45.6 48.2 53.1 56.0 Caulerpa fdiformis % 31.9 20.7 0 0 0 0 Cidaridae Phyllacanthus parvispinus 0 0 0 6 0 7 Centrechinidae Centrostephanus rodgersii 270 2954 6794 926 1840 2646 Echinidae Tripneustes gratilla 12 1 2 0 0 0 Strongylocentrotidae Heliocidaris tuberculata 330 34 115 1 2 0 Heliocidaris erythrogramma 1833 1608 409 1 9 0 Octopodidae Octopus spp. * 1 1 1 3 0 0 Sepiidae Sepia spp. * 0 1 2 1 1 1 Palinuridae Jasus verreaiaii * 1 0 0 0 0 0 Haliotidae Haliotis ruber * 43 49 40 2 9 1 Cassididae Cabestana spengleri 43 58 30 128 41 13 Muricidae Thais orbita 28 68 45 1 5 3 Turbinidae Turbo torquatus 212 269 240 8 5 6 Australium tentoriforme 27 172 1147 195 146 442

COMPLEXITY Reef complexity 1.1 1.8 1.9 1.8 1.9 2.1 Table 3.3. R-statistics (Clarke 1993) from one-way ANOS1M and pairwise comparisons of fish assemblages at Bellambi and Port Kembla. B = Bellambi, BU = Bulli, SH = Shellharbour, PK = Port

Kembla, FI = Flinders Island, AR = Atchinsons Rock. Number of permutations = 5000. ns = not significant, * = significant at a<0.05, **= significant at oc<0.025 - alphas adjusted for multiple comparisons.

Bellambi Port Kembla

All locations R = 0.475 * R = 0.602 *

Pairwise comparisons B v BU ns PK v FI *

B v SH * PK v AR *

BU v SH ns FI v AR ns Figure 3.2. MDS plots for the abundance of Fish at Bellambi sewage outfall and control locations

(stress = 0.06) (top) • - Bellambi, A - Bulli, □ - Shell Harbour, and Port Kembla sewage outfall and control locations (stress = 0.1) (bottom). • - Port Kembla, A - Flinders Island, □ - Atchinsons Rock. Table 3.4. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Bellambi, Bulli and Shellharbour. O vs C, outfall versus controls; nt, no test; *, significant

(P<0.05); **, highly significant (P<0.01).

Variate Mobile Cryptic Atypichthys Cheilodactylus Parma Norflokia

Abundance Abundance strigatus fuscus unifasciata clarkei

df F-ratio m.s. m.s m.s m.s m.s m.s

denominater

Time 3 Tx L 35 0.6 2.2 0.3 2.1 0.4

Location (L) 2 nt 233 1.9 4.9 1.3 10.7 0.4

0 vs C (1) nt 295 2.0 7.6 1.9 16.7 0.04

Among C (1) nt 171 1.4 2.1 0.8 4.7 0.7

Site (S) 3 T x S(L) 16 0.1 0.2 1.9 0.6 0.3

0 vs C (1) S(C) 6.4 0.4 0.6 4.5 0.02 0.9

Among C (2) T x S(C) 21 0.03 0.3 0.6 0.9 0.04

Tx L 6 T x S(L) 15.3 0.2 1.4 1.0 0.5 0.4

O vs C (3) T x L(C) 6.7 0.3 0.9 0.3 0.9 0.7

Among C (3) T x S(C) 23.9 ** 0.1 1.9 1.6 * 0.2 0.1

T x S(L) 9 Residual 1.3 0.2 0.6 0.6 0.4 0.3

0 vs C (3) T x S(C) 2.0 0.4 0.2 1.2 0.02 0.5

Among C (6) Residual 1.0 0.1 ** 0.7 ** 0.3 0.7 * 0.1

Residual 72 4.2 0.05 0.2 0.7 0.2 0.03 CM Mean abundance (120m2) Abundance (60m2) Meai E Figure (b) unifasciata, (c) (a) (b)

abundance

3.3.

Mean (f)

of Norfolkia

cryptic abundance

fish, clarkei.

(±

(c)

1

Trachinops SE) □

-

of Bellambi,

Fishes

taeniatus,

from

♦

-

the Bulli, CM E CM CM £ E

(d) Bellambi

<2> Cheilodactylus

-

(e) Shellharbour.

study

(a)

fuscus,

abundance

(e)

Parma of

mobile

fish,

Table 3.5. Relative mean abundances and cumulative percentage (Cum %) of the five species that were most responsible for differences between fish assemblages in the boulder habitat at the outfall location at Port Kembla and the two control locations at Flinders Island (FI) and Atchinsons Rock (AR).

Outfall FI Cum %. AR Cum %

Trachinops taeniatus 89 493 40.2 303 30.5

Atypichthys strigatus 93 207 52.6 45 40.2

Trachurus novaezelcmdiae 16 59 60.6 69 48.5

Chromis hypsilepis 13 77 67.5 50 55.8

Dinolestes lewini 29 58 73.5 - -

Pempheris compressus 41 - - 23 62.9 CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 64

Analysis of macrobenthos

Graphs of the dominant macrobenthic variates are presented. Regression was used to determine

the relationship between percentage cover of kelp, Caulerpa filiformis, encrusting and turfing

algae and abundance of selected invertebrate species compared to fish abundance and richness.

The Dunn-Bonferroni Procedure was used to control type I error rates (Winer 1991).

RESULTS

Fish assemblages in the Bellambi study

A total of 58 species and 5,746 individuals were recorded from all locations during the study

(Table 3.2). Mobile fish comprised 62 % of species and 86 % of individuals. Fish of economic

value comprised 27 % of species and 4 % of individuals, and fish of tropical origin comprised

9% of species and less than 1% of individuals (Table 3.2). There was no consistent significant

difference between fish assemblages from outfall and control locations, although fish

assemblages at Bellambi were different from Shellharbour (Table 3.3, Figure 3.2). There were

some obvious differences between the dominant species at outfall and control locations. For

example, Trachinops taeniatus and Parma unifasciaia were abundant at the control sites but

were not generally recorded at the outfall locations (Table 3.2, Figure 3.3). The mean

abundance of mobile fish, crypic fish, Atypichthys strigatus, Cheilodactylus fuscus, Parma

microlepis and Norfolkia clarkei differed among control locations inconsistently through time

(Table 3.4, Figure 3.3). N. clarkei were very abundant at one time at the outfall location (Figure

3.3).

Fish assemblages in the Port Kembla study

A total of 79 species and 13,106 individuals were recorded from four surveys at Port Kembla and control locations. Mobile fishes comprised 46 % of species and 86 % of individuals. Fish of economic value comprised 33 % of species and 15% of individuals and tropical species comprised 15 % of species and less than 1 % of individuals (Table 3.2). There were consistent significant differences between fish assemblages from outfall and control locations, (Table 3.3, (a) (d)

cm CM E E C-

•a

I<

(b) (e)

cl CM E E

CM oj C. (D 1 •o I <

(c)

CM E

Figure 3.4. Mean richness or abundance (± 1 SE) of fishes from the Port Kembla study (a) richness of

cryptic fishes, (b) Crinodus lophodon (c) Cheilodactylus fuscus, (d) Hypoplectrodes mccullochi, (e)

Trachinops taeniatus. □ - Port Kembla, ♦ - Controls. Table 3.6. Summaries of analyses comparing spatial and temporal variations in fish assemblages and species at Port Kembla, Flinders Island and Atchinsons Rock. O vs C, outfall versus controls; *, significant (P<0.05); **, highly significant (P<0.01).

Variate Cryptic Crinodus Cheilodactylus Hypoplectrodes Trachinops

richness lophodon fuscus mccullochi taeniatus

df F-ratio m.s. m.s m.s m.s m.s

denominater

Time 3 Tx L 6.0 4.8 0.002 0.2 0.3

Location (L) 2 Tx L 5.0 302 0.6 1.8 6.7

O vs C (1) L (C) 10.0 * 605 * 1.1 ** 3.5 ** 12.9

Among C (1) T x L(C) 0.03 0.3 0.008 0.005 0.5

Tx L 6 Residual 4.0 13.1 0.1 0.06 0.3

O vs C (3) T x L(C) 6.0 23.2 0.01 0.04 0.6 **

Among C (3) Residual(C) 1.9 3.1 0.2 0.07 0.005

Residual 37 2.4 23.8 0.07 0.05 0.3 CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 65

Figure 3.2). These differences were largely due to lesser abundances of Trachinops taeniatus,

Trachurus novaezelandiae and Chromis hypsilepis at the outfall compared to the control locations (Table 3.5). More Atypichthys strigatus were recorded at Flinders Island and less were recorded at Atchinsons Rock compared to the outfall location (Table 3.5). Less Dinolestes lewini and more Pempheris compressus were also recorded at one of the controls compared to the outfall location (Table 3.5).

Significant location effects indicated an impact due to sewage disposal at Port Kembla.

Differences in abundance between “impact” and control locations were found for species richness of cryptic fish, Trachinops taeniatus, Crinodus lophodon, Cheilodactylus fuscus and

Hypoplectrodes mccullochi (Table 3.6, Figure 3.4). Species richness of cryptic fish and abundance of Trachinops taeniatus, and Hypoplectrodes mccullochi were greater at the control locations compared to the Port Kembla outfall location. More Crinodus lophodon and

Cheilodactylus fuscus were recorded at the outfall location compared to the control locations

(Figure 3.4).

Macrobenthos assemblages in the Bellainbi study

A total of 18 variates were recorded during the study (Table 3.2). Caulerpa filiformis was not recorded at Shellharbour (Table 3.2). Reef complexity was smallest at Bellambi (Table 3.2).

Percentage cover of Ecklonia radiata and encrusting algae was small at the outfall location

(Figure 3.5) Abundance of Centrostephanus rodgersii and Australium tentoriforme was small and Heliocidaris tuberculata was larger at Bellambi outfall compared to control locations

(Figure 3.5).

Abundance and richness of fish at the six sites were compared with six macrobenthos variates

(Table 3.7). Seventy comparisons were not significant, two comparisons were significant at alpha < 0.05, and one comparison was also significant at alpha < 0.0086 (Dunn-Bonferroni procedure n=6). A positive association was found between fish abundance and topographic complexity at the Shellharbour south site (R= 0.651, P<0.01) (Table 3.7). Table 3.7. Associations between (a) abundance and (b) richness of fishes, and macrobenthos from six sites from the Bellambi study. Significance is based on regression at the P < 0.05 level and P <

0.0086# (Dunn-Bonferroni Procedure to correct for type I error) . Statistically significant positive

(+ve), negative (-ve) and no significant associations (-) are shown.

(a)abundance

Bellambi Bulli Shellharbour

Macrobenthos North South North South North South

Ecklonia radiata - - - -

Encrusting algae - - - -

Turfing algae - - - -

Centrostephanus rodgersii - - - -

Caulerpa filiform is +ve - - -

Topographic complexity - - - +ve#

(b) species richness

Bellambi Bulli Shellharbour

Macrobenthos North South North South North South

Ecklonia radiata - - - -

Encrusting algae - - - -

Turfing algae - - --

Centrostephanus rodgersii - - - -

Caulerpa filiform is - - - -

Topographic complexity - - - - Table 3.8. Associations between (a) abundance and (b) richness of fishes, and macrobenthos from boulder habitat at three locations from the Port Kembla study. Significance is based on regression at the P < 0.05 level and P < 0.01# (Dunn-Bonferroni Procedure to correct for type I error). Statistically significant positive (+ve), negative (-ve) and no significant associations (-) are shown.

(a) abundance

Port Kembla Flinders Is Atchinson’s Rock

Macrobenthos Boulder Boulder Boulder

Ecklonia radiata -ve +ve# -

Encrusting algae - - -

Turfing algae - - -

Centrostephanus rodgersii +ve# - -

Topographic complexity - - +ve

(b) species richness

Port Kembla Flinders Is Atchinson’s Rock

Macrobenthos Boulder Boulder Boulder

Ecklonia radiata - +ve# -

Encrusting algae -

Turfing algae - - -

Centrostephanus rodgersii - - -

Topographic complexity - - - Percentage cover (30 m2) Percentage cover (30 m2) Per 100 Australium encrusting Figure (c) O)

( -i a

)

Winter

Winter 3.5. Spring

algae, tentoriforme. Mean

(c) abundance Spring Spring

Caulerpa Summer

□

-

Summer Summer of Bellambi,

filiform

macrobenthos

is,

Autumn ♦

Autumn Autumn (d)

-

Bulli, Centrostephanus

from

< j > the

-

Shellharbour.

Bellambi ? 2 m

300 200 100 40 30 20

rodgersii, (f)

(e)

(d) -i - - - - -

study Winter Winter

(a)

(e)

Ecklonia

Heliocidaris Spring Spring

radiata,

tuberculata Summer Summer

(b)

,

Autumn (f) Autumn

(a) (c) 100 i

Winter Spring Summer Autumn Winter Spring Summer Autumn

(b) (d)

300 -i

8 200 -

100 -

Winter Spring Summer Autumn Winter Spring Summer Autumn

Figure 3.6. Mean abundance of macrobenthos from the Port Kembla study (a) Ecklonia radiata, (b)

Centrostephanus rodgersii, (c) Heliocidaris erythrogramma, (d) Australium tentoriforme. □ - Port

Kembla, ♦ - Flinders Island, <£> - Atchinsons Rock. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 66

Macrobenthos assemblages in the Port Kembla study

Percentage cover of Ecklonia radiata was similar at the outfall and Flinders Island and consistently smaller at Atchinson’s Rock (Figure 3.6). Abundances of Centrostephanus rodgersii and Heliocidaris erythrogramma were consistently smaller at Port Kembla (Figure

3.6). Abundances of Australium tentoriforme were consistently larger at Atchinson’s Rock

(Figure 3.6).

Comparison of fish abundance and macrobenthos resulted in twenty-five non-significant comparisons, two comparisons which were significant at alpha < 0.05, and three comparisons were also significant at alpha < 0.01 (Dunn-Bonferroni procedure n=5) (Table 3.8). Cover of

Ecklonia radiata had a positive association with abundance and richness of fish at Flinders

Island. Positive associations were found for Centrostephanus rodgersii and abundance of fish at

Port Kembla (Table 3.8).

DISCUSSION

The reefs at Bellambi and Port Kembla were physically very different and direct comparisons of

the fishes and macrobenthos at these outfalls are limited. Obvious differences between the two

studies were the greater abundance and species richness of fishes at Port Kembla and the

presence of Caulerpa filiformis and larger numbers of echinoderms at Bellambi. These

differences are largely due to the different depths and habitats associated with the two studies.

Species richness of fish from the Bellambi study was similar to Gerroa (MPR 1989), but less than previous studies of fish in the Illawarra region (EPA 1992), and other studies in NSW (TEL

1992, Lincoln Smith et al. 1992, EPA 1993, Smith et al. 1998). This may be due to the shallow depth of water that was surveyed for the Bellambi study, the low topographic complexity, the presence of the outfall or unknown variables. Species richness of fish from the Port Kembla study was similar to previous studies of fish in the Illawarra region (EPA 1992). The small spatial and temporal variation between assemblages from outfall and control locations, and the CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 67 large variability within locations (i.e., at the scale of sites) is similar to findings of Lincoln Smith

(1985), Smith (1994), Chapman et al. (1995) and Underwood and Chapman (1996). These results are contrasted by the large spatial variation in assemblages surveyed at the small outfall at Boulder Bay (Roberts et al. 1998, Smith et al. 1998) and large outfalls at Sydney (TEL 1992).

These differences may be due to the scale of impact.

The surveys in the Illawarra region were done after the outfalls had been turned off and the water had cleared. This procedure does not account for the possibility of rapid immigration or emigration of Fish in response to the clearer water and lack of sewage. The possibility of rapid migration would probably not apply to territorial or small cryptic species such as the pomacentrids and some labrids, but may apply to mobile species such as girellids.

There were significant differences between outfall and control locations for the Port Kembla

region. The richness of cryptic fish was greater at the control locations compared to the Port

Kembla outfall location. More Crinodus lophodon and Cheilodactylus fuscus and fewer

Hypoplectrodes mccullochi and Trachinops taeniatus were generally recorded at the outfall

location compared to control locations. There were no significant differences between fish

assemblages at Bellambi and control locations although there were some obvious differences for

individual species such as absence of Trachinops taeniatus at Bellambi outfall. Some of these

species of fish and macrobenthos may be useful indicator organisms of pollution. The patterns

of increased or decreased abundance of some of these fish species at outfall locations have been

reported previously by TEL (1992) and Smith et al. (1998). It is not possible to unambiguously

attribute such differences to the presence of the outfalls as there are no data on the assemblages

of fish and macrobenthos existing before they were commissioned (sensu: Green 1979,

Underwood 1992). Differences between the outfalls and control locations may be due to many

other factors, such as variation due to aspect, effects of storms, shifting sediments, differences in

topography, and random events such as settlement of small juveniles (Lincoln Smith et al. 1991,

Kennedy and Underwood 1992, Schmitt and Osenberg 1996). Other potential sources of

influence within the study region are industrial pollution from Port Kembla Harbour, other CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 68 outfalls (e.g. at Coniston and Barrack Point), stormwater runoff and commercial and recreational fishing activities.

The assertion that fish (in particular Mugil cephalus) are interacting with the plume of freshwater from the Port Kembla outfall and altering their migration directions from an inshore to an offshore route (GHD 1990b) was not supported or refuted in this study because no Mugil cephalus were recorded. This species is rarely, if ever, encountered in visual surveys of rocky reef fish in NSW waters. General statements such as “rocky reef fish are attracted to the outfalls and utilise the effluent particles as food” (GHD 1990b) are not supported by this research which shows that the abundance of some species did not change, some increased and some decreased adjacent to the outfalls compared to control locations.

There were very few associations between fish and macrobenthos. Associations were found between fish abundance and topographic complexity, Ecklonia radiata and Centrostephanus rodgersii, however, most comparisons were not significantly different. This indicates that the abundance of most fish assemblages were generally not affected in a positive or negative direction by the cover of algae, abundance of urchins or topographic complexity. Holbrook et al.

(1990) also reported few differences in fish species richness and only weak differences in species composition among reefs of different habitat types. However, algae, urchins or topographic complexity plays an important role in abundance of some species of fish (Choat and

Ayling 1987, Jones 1988, Connell and Jones 1991, Patton et al. 1994).

This study provides baseline information and it is recommended that further environmental monitoring should be done prior to and after any changes in sewage treatment, construction of an offshore outfall or if discharge volumes of sewage in the Illawarra region are increased. A balanced experimental design of an additional four surveys of fish and macrobenthos at similar times of the year prior to any changes to the sewage outfalls would provide a measure of intra- and inter-annual variability. Following changes to the outfalls, surveys should be done four CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 69 times per year over two years to provide a balanced comparison with ‘before’ data to enable rigorous assessment of the effects of sewage disposal on fish and macrobenthos.

REFERENCES

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium

No. 8, Bethesda, Maryland.

Bell, J.D., Harmelin-Vivien, M.L. (1982). Fish fauna of French Mediterranean Posidonia oceanica

seagrass meadows. 1. Assemblage Structure. Tethys, 10, 337-347.

Brown, V.B., Davies. S.A., Synnot. R.N (1990). Long-term monitoring of the effects of treated

sewage effluent on the intertidal macroalgal assemblage near Cape Schanck, Victoria, Australia.

Botanic a Marina 33, 85-98.

Carefoot, T., Simpson, R.D. (1977). Seashore Ecology. University of Queensland Press. St Lucia. 278

pp.

Chapman, M.G., Underwood, A.J., Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

J. Exp. Mar. Biol. Ecol. 189, 103-122.

Choat, J.H., Ayling, A.M. (1987). The relationship between habitat structure and fish faunas on New

Zealand reefs. J. Exp. Mar. Biol. Ecol. 110, 257-284.

Clarke, K.R. (1993). Non-parametric multivariate analyses of change in assemblage structure. Aust. J.

Ecol. 18, 1: 117-143.

Connell, S.D., Jones, G.P. (1991). The influence of habitat complexity on postrecruitment processes in

a temperate reef fish population. J. Exp. Mar. Biol. Ecol. 151, 271-294.

Dakin, W.J. (1987). Australian seashores. Angus and Robertson, Sydney.

EMU (1991a). Marine investigations for the Bellambi and Port Kembla sewage treatment plant

upgrading. Volume 1. Physical Oceanography. Environment Management Unit, Water Board,

Sydney.

EMU (1991b). Marine investigations for the Bellambi and Port Kembla sewage treatment plant

upgrading. Volume 2. Biology, sediments, bioaccumulation and water quality. Environment

Management Unit, Water Board, Sydney. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 70

EPA (1992). Port Kembla dredge spoil report. Published by the Environmental Protection Authority,

Sydney 71 pp.

EPA (1995). Effects of sewage outfalls on marine organisms at Port Kembla and Bellambi, Illawarra

region, NSW. Environmental Protection Authority, Chatswood, Sydney.

EPA (1997). Public Inquiry into the Management of Sewage and Sewage By-products in the NSW

Coastal Zone. Final report prepared by the Environment Protection Authority, Sydney. 299 pp.

Fairweather, P.G (1990). Sewage and the biota on seashores: assessment of impact in relation to

natural variability. Environ. Monit. Assess. 14, 197-210.

Foster, M.S. (1990). Organisation of macroalgal assemblages in the northwest Pacific: the assumption

of homogeneity and the illusion of generality. Hydrobiologia 192, 21-34.

GHD (1990a). Environmental assessment of proposed chemically assisted sedimentation facilities at

the Bellambi wastewater treatment plant. Gutteridge Haskins and Davey Pty Ltd, NSW.

GHD (1990b). Environmental assessment of proposed chemically assisted sedimentation facilities at

the Port Kembla wastewater treatment plant. Gutteridge Haskins and Davey Pty Ltd, NSW.

Green, R.H. (1979). Sampling Design and Statistical Methods for Environmental Scientists. Wiley-

Interscience. New York.

Grigg, R.W., (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Mar. Ecol. Prog. Ser. 103, 25-34.

Hall, J., Rissik, D. (1992). Bellambi and Port Kembla sewage treatment plant upgrades feasibility

study: offshore biological component. Environmental Management Unit. Sydney.

Holbrook, S.J., Schmitt, R.J., Ambrose, R.F. (1990). Biogenic habitat structure and characteristics of

temperate reef fish assemblages. Aust. J. Ecol. 15, 489-503.

Holbrook, S.J., Kingsford, M.J., Schmitt, R.J., and Stephens, J.S. (1994). Spatial and temporal patterns

in assemblages of temperate reef fish. Amer. Zool. 34, 463-475.

Jones, G.P. (1988). Ecology of rocky reef fish of north-eastern New Zealand: a review. N.Z. J. Mar.

Freshwat. Res. 22, 445-462.

Jones, G.P., Andrew, N.L. (1990). Herbivory and patch dynamics on rocky reefs in temperate

Australasia: the roles of fish and sea urchins. Aust. J. Ecol. 15, 505-20. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 71

Kennelly, S.J., Underwood, A.J. (1992). Fluctuations in the distribution and abundances of species in

sublittoral kelp forests in New South Wales. Aust. J. Ecol. 17, 367-382.

Koop, K., Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Mar. Poll. Bull.

33, 121-123.

Lincoln Smith, M.P. (1985). The development and application of visual survey procedures for fish

assemblages on shallow rocky reefs. MSc thesis, Univ. of Sydney, Australia. 187 pp.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate

rocky reef fish assemblages. Mar. Ecol. Prog. Ser. 43, 223-231.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Environ. Biol. Fish. 26, 29-37.

Lincoln Smith, M.P., Bell. J.D., Hair, C.A. (1991). Spatial variation in abundance of recently settled

rocky reef fish in southeastern Australia: implications for detecting change. Mar. Ecol. Prog. Ser.

77, 95-103.

Lincoln Smith, M.P., Hair, C.A., Bell, J.D. (1992). Fish associated with natural rocky reefs and with

artificial structures. Jervis Bay Marine Ecological Study. Fisheries Research Institute report for

Department of Defence, Canberra.

Lincoln Smith, M.P., Jones, G.P. (1995). Fishes of shallow coastal habitats, pp. 240-253 in

Underwood, A.J., Chapman, M.G. (Eds) Coastal marine ecology of temperate Australia. UNSW

Press, Sydney.

May, V. (1976). Changing dominance of an algal species (Caulerpa filiformis (Suhr)Hering). Telopea.

1, 136-138.

May, V. (1985). Observations on algal floras close to two sewerage outlets. Cunninghamia 1, 385-

394.

McCormick, M.I. (1989). Spatio-temporal patterns in the abundance and population structure of a large

temperate reef fish. Mar. Ecol. Prog. Ser. 53, 215-225.

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Report MHL769 by

Manly Hydraulics Laboratory and the Ecology Lab Pty Ltd, Sydney. 65 pp. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 72

MPR (1989). Intertidal and subtidal investigations for Gerringong and Gerroa Ocean Outfall

Preliminary Site Investigations. Marine Pollution Research Pty Ltd for Water Board, Illawarra

Region. 28 pp.

Patton, M.L., Valle, C.F., Grove, R.S. (1994). Effects of bottom relief and Fish grazing on the density

of the giant kelp Adacrocystis. Bull. Mar. Sci. 55, 631-644.

Pearson, T.H., Rosenberg, R. (1978). Macrobenthic succession in relation to organic enrichment and

pollution of the marine environment. Oceanogr. Mar. Biol. A. Rev. 16, 229-311.

Roberts, D.E., Smith, A., Ajani, P. and Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACI’ approach. Mar.

Ecol. Prog. Ser. 163,213-224.

Sale, P.F. (1991). (Ed.) The Ecology Of Fishes On Coral Reefs. Academic Press, Inc., New York.

Scanes, P.R., Scanes, K. and Otway, N.M. (1995). Environmental problems due to disposal of wastes.

In Coastal marine ecology of temperate Australia. Underwood, A.J., Chapman, M.G. eds, UNSW

Press, Sydney, Australia, pp. 297-310.

Schmitt, R.J. and Osenberg, C.W. (1996). Detecting ecological impacts, concepts and applications in

coastal habitats. Academic Press, New York. 198 pp.

Smith, A.K., Ajani, P.A., Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Marine Environmental Research 47, 1-20.

Smith, S.D. (1994). Impacts of domestic sewage effluent verses natural background variability: an

example from Jervis Bay, New South Wales. Aust. J. Mar. Freshwater Res. 45, 1045-1064.

Spotte, S., Bubucis, P.M., Adams, G. (1992). Diurnal occupancy of crevices and overhangs by fishes

on the Caicos Bank, Turks and Caicos Islands, British West Indies. Bull. Mar. Sci. 5, 66-82.

TEL (1992). Surveys of fish and epibenthic assemblages in the Illawarra coastal zone. Preliminary

report by The Ecology Lab Pty Ltd for the Water Board Environmental Management Unit.

Sydney.

TEL (1993). Changes to assemblages of reef fish adjacent to the North Head and Bondi cliff-face

sewage outfalls following decommissioning. Final report by The Ecology Lab Pty Ltd for the

Water Board. Sydney. 49 pp. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 73

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final Report by The Ecology Lab for the Water Board, Sydney. 29 pp.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA. 229 pp.

Tsai, C., Wiley, M., Chai, A. (1991). Rise and fall of the Potomac striped bass stock: a hypothesis of

the role of sewage. Trans. Am. Fish. Soc. 120, 1-22.

Underwood, A.J. (1981). Techniques for analysis of variance in experimental marine biology and

ecology. Ann. Rev. Oceanogr. Mar. Biol. 19, 513-605.

Underwood, A.J. (1991). Beyond BACI: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Aust. J. Mar. Freshwat. Res. 42, 569-588.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. J. Exp. Mar. Biol. Ecol., 161, 145-178..

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Aust. J. Ecol. 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15.

Underwood, A.J. (1997). Experiments in Ecology. Their Logical Design and Interpretation using

Analysis of Variance. Cambridge University Press. 504 pp.

Underwood, A.J., Chapman, M.G. (1995). (eds) Coastal marine ecology of temperate Australia.

UNSW Press, Sydney, Australia. 341 pp.

Underwood, A.J., Chapman, M.G. (1996). Subtidal assemblages on rocky reefs at a cliff-face sewage

outfall (North Head, Sydney, Australia): What happened when the outfall was turned off? Mar.

Poll. Bull. 33, 293-302.

Underwood, A.J., Kennelly, S.J. (1990). Ecology of marine algae on rocky shores and subtidal reefs in

temperate Australia. Hydrobiologia 192, 3-20.

Underwood, A.J., Kingsford, M.J., Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Aust. J. Ecol. 6, 231-249. CHAPTER 3 EFFECTS OF SEWAGE ON FISH IN ILLAWARRA 74

West, R.J., Thorogood, C.A., Walford, T.R., Williams, R.J. (1985). An estuarine inventory for New

South Wales, Australia. Fisheries Bulletin 2. Department of Agriculture New South Wales.

Winer, B.J. (1991). Statistical Principles in Experimental Design. Second Edition. McGraw Hill,

Tokyo. 907 pp Sewage treatment plant (top) and one of the sewage diffusers (bottom) at Boulder Bay, Hunter region. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 75

CHAPTER 4

A BEYOND BACI SURVEY OF THE IMPACT OF SEWAGE ON

FISH ASSEMBLAGES AND A SEA URCHIN IN THE HUNTER

COASTAL ZONE

ABSTRACT

Underwater visual surveys involving a Before/After/Control/Impact (BACI) experimental design were used to investigate the abundance and species richness of temperate rocky reef fishes and the abundance of a sea urchin following the discharge of sewage effluent from a deepwater outfall.

Multivariate analyses indicated that fish assemblages were significantly different at the outfall and control locations, and these differences were largely attributed to the declines in relative abundance of several common, resident species of reef fish, such as the eastern hulafish, Trachinops taeniatus and yellowtail Trachurus novaezelandiae and the sea urchin Centrostephanus rodgersii.

Univariate analysis indicated that the abundance and richness of groups of fish and individual species varied over times, periods and locations. The abundance and species richness of fish from the outfall location were generally less than for fish assemblages at the control locations. We detected decreases of 33 percent in the species richness of fish and 50 percent in the abundance of Centrostephanus rodgersii at the outfall location and, conversely, the abundances of some cryptic (small, cave dwelling or camouflaged) fishes increased over time at the outfall. This study indicates that the discharge of sewage effluent resulted in an environmental impact affecting fish assemblages, some fish species and a sea urchin, and it was concluded that the outfall was not complying with current pollution management guidelines. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 76

INTRODUCTION

Proper treatment and disposal of sewage has become increasingly important because of the large numbers of people living in close proximity to waterways, and the recognition that sewage effluent can

be extremely damaging to the aquatic environment and associated industries such as tourism, fishing

and aquaculture (Beder 1989, Fagan et al. 1992, Otway 1995, Scanes et al. 1995).

The aquatic flora and fauna associated with temperate rocky reefs are relatively diverse, abundant and

sedentary, and therefore regarded as suitable indicators of environmental change (Warwick 1993,

Lincoln Smith and Jones 1995). Several studies have focused on the effects of sewage on such marine

assemblages (Bell and Harmelin-Vivien 1982, May 1985, Love et al. 1987, Brown et al. 1990,

Fairweather 1990, TEL 1991, 1992, 1993, 1994, Grigg 1994, Smith 1994, Chapman et al. 1995,

Roberts 1996). In particular, fish may exhibit a range of responses in a polluted area. Sewage can affect

the abundance, mortality, fecundity and size of fish, lead to toxic effects, increase susceptibilities to

infections, and alter behavioural responses (Gray 1989, Adams et al. 1993, EPA 1993a).

The volume, periodicity and level of treatment of sewage and the type of biota that are sampled are

important determinants in the detection and scale of any environmental impact (Keough and Quinn

1991, Boates and Russell 1992, TEL 1994, Smith 1994). In general, striking effects are very localised

(tens of metres), but detectable effects may be found at distances of up to several kilometres (Brown et

al. 1990) from the source of the sewage pollution. A number of studies have detected an impact at a

community or trophic level (TEL 1993, Grigg 1994) and several have reported an impact restricted to

a small number of species (TEL 1992, 1994). However, several studies have reported no significant

impact on aquatic biota as a result of sewage discharge (Lincoln Smith 1985, Smith 1994, Scanes et al.

1995).

Despite the large quantity of literature describing the effects of sewage on fish, there are very few

studies that have been published that have adopted an appropriate Before/After/Control/Impact (BACI)

experimental design (Underwood 1991, 1992, 1993, 1994) with pre-disturbance data and multiple Scale (fcro)

PORT STEPHENS

Point stePhcns

A «* ------'O Boulder Bay r^

AUSTRALIA

152.6°E

Figure 4.1. The study locations in the Hunter region, NSW, Australia. Boulder Bay is the impacted location with the sewage outfall. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 77 control locations. This limitation may be partially explained by the relatively recent advances in experimental design and the impossibility of obtaining ‘before’ data for existing outfalls that may have been discharging sewage for decades. An inadequate design reduces the ability to make predictions on the causal changes associated with fish assemblages due to sewage.

The study of temperate reef fish was undertaken at one impact location, an ocean sewage outfall, and two control locations. I hypothesised that fish assemblages at the three locations would be similar

before commissioning of the ocean outfall, and that fish assemblages at the ‘outfall’ location would not

be significantly different post-commissioning from the outfall location.

MATERIALS AND METHODS

Fish surveys

Surveys of reef fish were conducted at three locations. The outfall location was at Boulder Bay

(32°45.86’S, 152° 09.83’E) and the control locations were at Point Stephens (32°44.48’S, 152°

12.39’E) and Tomaree Head (32°42.87’S, 152° 11.34’E) (Figure 4.1). The outfall was commissioned

on 15 November 1993 and discharges approximately 4.3 ML/day of secondarily treated effluent from

diffuser(s) located in 15 m depth of water. A possible confounding factor in this study was the location

of a small primary shoreline outfall that discharged untreated sewage into Boulder Bay from the late

1970s to 1993. This outfall continued to discharge sewage until the new offshore outfall was

commissioned. The control locations were chosen because they were topographically similar to the

outfall location but a considerable distance away from the outfall location, so that they were not

exposed to the effects of sewage effluent. The substrate at all locations was dominated by ‘white

boulder’ habitat (Underwood et al., 1991), with occasional patches of the kelp Ecklonia radiata.

Fish were counted during three periods (before commissioning, immediately post-commissioning and

one year post-commissioning), with four samples in each period, comprising a total of twelve sampling

times (Table 4.1). The construction of the outfall was completed prior to sampling. At each time, four

replicate 60 m weighted, line transects were randomly laid by boat during daylight hours, at a depth of

12 to 16m, approximately parallel with the shoreline, at each location. The outfall location was within a Table 4.1. Sampling details for fish at three time periods and twelve times.

Period Time Dates

1 (Before) 1 24-25/6/93

1 2 31/7/93

1 3 27-28/8/93

1 4 1/10/93

2 (Immediately after) 5 13/12/93

2 6 22-23/1/94

2 7 5/3/94

2 8 23/4/94

3 (one year after) 9 30/8/94

3 10 15/10/94

3 11 19/11/94

3 12 15/2/95 CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 78

100 m radius of the shoreward diffuser. The abundance of mobile and cryptic (small, cave dwelling or camouflaged) fish species was recorded by a SCUBA diver within 1 m either side of the transects as described by Lincoln Smith (1988, 1989). Individual mobile fish were recorded while swimming steadily along the transect and cryptic fish were counted by searching at slow swimming speed

(Lincoln Smith 1988, 1989). Sea urchins were recorded using the same methodology and at the same time as we recorded cryptic fish. Sea urchins were recorded because they were abundant, easy to survey while counting cryptic fish and a potential indicator of sewage pollution (TEL 1994). Data were recorded on a perspex slate. Horizontal visibility (water clarity) was estimated at the start of each transect.

Multivariate analyses

Data were analysed using non-metric multi-dimensional statistical (MDS) techniques (Clarke 1993) to describe the variation between species abundance and composition at each location and at each period,

using the PRIMER software package (Plymouth Marine Laboratories, U.K.). Abundance and

presence/absence data from each location and period were analysed using the Bray-Curtis similarity

measure, and the results were presented as 2-dimensional plots. If the stress levels from the 2-

dimensional plots were greater than 0.2 the plots were considered difficult to interpret (Clarke 1993).

One-way analysis of similarity (ANOSIM) tests were used to test the hypothesis that fish assemblages

from the three locations and three periods (times were grouped) were similar. The SIMPER procedure

(Plymouth Marine Laboratories, UK) was used to identify the contribution of individual species to

differences between locations and periods.

Univariate analyses

A 3-way nested asymmetrical Analysis of Variance (ANOVA) was used to compare assemblages of

fish (e.g. abundance and richness of total, mobile and cryptic species) and individual species among

periods, times and locations. There were three time periods, with four random times within each period

(Table 4.1). Data were transformed to log(x+l) if variances proved heterogeneous using Cochran’s

Test (Winer et al., 1991). Table 4.2. Mean abundance of fish at outfall (O) and control (C) locations from three periods: PI - Period 1

(before commissioning), P2 - Period 2 (immediately post commissioning), P3 - Period 3 (one year post­ commissioning),. M - mobile, C - cryptic, t - tropical, * - commercial/recreational importance.

0 C 0 C 0 C Family Species Type Total PI PI P2 P2 P3 P3 Heterodontidae Heterodontus portjacksoni C 11 0 1.5 0 0 0 4 Brachaeluridae Brachaelurus waddi C 1 0 0 0 0.5 0 0 Parascyllidae Parascyllium spl (w.spot) C 1 0 0 0 0 0 0.5 Orectolobidae Orectolobus spl * C 25 1 3.5 1 3.5 0 4.5 Hypnidae Hypnos monopterygium c 1 0 0 0 0.5 0 0 Dasyatididae Dasyatis brevicaudatus M 2 1 0 0 0 0 0.5 Myliobatidae Myliobatis australis M 2 1 0 0 0.5 0 0 Muraenidae Gymnothorax prasinus C 22 6 4 0 1.5 0 2.5 Muraenidae Enchelycore ramosa c 1 0 0.5 0 0 0 0 Aulopidae Aulopus purpurissatus * M 15 0 3 2 0 3 2 Plotosidae Cnidoglanis macrocephalus C 1 0 0 1 0 0 0 Moridae Lotella rhacinus C 10 0 0.5 3 1 0 2 Trachichthyidae Trachichthys australis C 5 1 0.5 0 0.5 0 1 Scorpaenidae Scorpaena cardinal is * C 19 6 2.5 2 0.5 1 2 Serranidae Acanthistus ocellatus C 10 4 1 0 1 0 1 Hypoplectrodes mccullochi C 336 32 67 15 35 11 38 Epinephelus undulatostriatus*T c 1 0 0.5 0 0 0 0 Trachypoma macracanthus c 1 0 0.5 0 0 0 0 Anthias squamipinnis c 1 0 0 0 0 0 0.5 Plesiopidae Trachinops taeniatus M 1539 413 182 194 203 34 65 Dinolestidae Dinolestes lewini * M 239 0 58 1 8.5 0 53 Echeneididae Remora remora M 1 1 0 0 0 0 0 Carangidae Pseudocaranx dentex * M 58 0 3 0 1 0 25 Seriola lalandi * M 4 0 0 0 0 0 2 Trachurus novaezelandiae * M 2448 200 483 251 251 16 257 Sparidae Acanthopagrus australis * M 205 0 24 0 59 0 21 Pagrus auratus * M 7 0 1 0 1 1 1 Rhabdosargus sarba * M 15 0 2 0 3.5 0 2 Gerreidae Gerres subfasciatus * M 13 0 0 0 4.5 0 2 Lethrinidae Lethrinus nebulosus * T M 1 0 0 0 0 0 0.5 Sciaenidae Argyrosomus hololepidotus * M 1 0 0 0 0.5 0 0 Mullidae Parupeneus signatus * M 86 21 6 1 17 0 9.5 Upeneichthys lineatus * M 11 3 1 1 1 0 1.5 Monodactylidae Schuettea scalaripinnis M 2176 0 387 0 373 0 329 Pempherididae Pempheris affinis C 527 2 100 15 89 0 67 Pempheris compressus c 2402 64 245 27 493 258 289 Pempheris multiradiata c 24 4 0.5 0 2 14 0.5 Girellidae Girella elevata * M 5 0 0.5 0 0.5 0 1.5 Girella tricuspidata * M 400 0 73 0 5.5 0 122 Kyphosidae Kyphosus sydneyanus M 20 0 0 0 7 0 3 Scorpididae Atypichthys strigatus M 2773 101 549 114 326 85 362 Microcanthus strigatus M 25 0 1 0 9.5 0 2 Scorpis lineolatus * M 160 6 15 16 28 9 22 Chaetodontidae Chaetodon guntheri T C 11 0 1 0 1.5 0 3 Enoplosidae Enoplosus armatus M 33 2 4.5 1 5.5 1 4.5 Pomacentridae Chromis hypsilepis M 454 6 46 9 78 1 96 Heniochus spl T C 2 0 1 0 0 0 0 Mecaenichthys immaculatus C 23 2 3.5 1 2 1 4 Parma microlepis c 436 54 66 43 46 29 43 (PTO) Table 4.2 contd Parma unifasciata c 102 1 15 1 17 1 19 Pomacentrus coelestis T M 2 0 0 0 0 0 1 Pomacentrid spl T C 4 0 0 0 0.5 1 1 Chironemidae Chironemus marmoratus c 107 12 19 3 15 5 9.5 Aplodactylidae Crinodus lophodon M 92 28 10 3 7 15 6 Cheilodactylidae Cheilodactylus fuscus * M 794 52 110 35 91 81 113 Cheilodactylus vestitus M 5 1 0.5 0 1 0 0.5 Nemadactylus douglasii * M 1 0 0 0 0 0 0.5 Latrididae Latridopsis forsteri * M 1 0 0 0 0.5 0 0 Labridae Achoerodus viridis * M 253 16 43 5 33 9 37 Anampses elegans M 2 0 0.5 0 0 0 0.5 Austrolabrus maculatus M 3 0 0.5 0 0 0 1 Coris pi eta M 211 44 19 0 30 11 30 Coris sandageri M 16 13 1 0 0 1 0 Eupetrichthys angustipes M 18 0 0.5 6 1 5 2 Labroides dimidiatus T M 6 0 0 0 1 0 2 Notolabrus gymnogenis M 326 44 41 35 35 25 36 Ophthalmolepis lineolatus M 430 70 45 57 40 60 37 Pictilabrus laticlavius M 37 5 1.5 6 6 9 1 Pseudolabrus fusciola M 1 0 0.5 0 0 0 0 Pseudolabrus guntheri M 26 2 2 5 4 0 3.5 Thalassoma lunare T M 3 0 0 0 0 0 1.5 Suezichthys arquatus T M 1 0 0.5 0 0 0 0 Stethojulis strigiventer M 1 0 0 0 0 0 0.5 Unid.wrasse-brown,yellow nose M 2 0 0 0 0 0 1 Unid.wrasse-horiz. stripes M 3 1 1 0 0 0 0 Unid.wrasse-green M 2 0 0.5 0 0 0 0.5 Unid.wrasse-silver belly M 1 1 0 0 0 0 0 Unid wrasse-scrib. black,white M 1 0 0 0 0.5 0 0 Unid. wrasse- white/red M 2 0 0 0 0 0 1 Unid. wrasse M 3 0 0.5 0 1 0 0 Odacidae Odax acroptilus M 12 29 6.5 44 6.5 32 9 Odax cyanomelas M 149 1 2 0 0.5 3 1.5 Scaridae Unid. scarid M 1 0 0 0 0 0 0.5 Blenniidae Plagiotremis rhinorhynchos T C 2 0 0 0 1 0 0 Plagiotremis tapeinosoma C 11 0 0.5 0 4.5 0 0.5 Acanthuridae Prionurus maculatus * M 139 0 24 0 17 0 30 Prionurus microlepidotus M 5 0 0 0 2.5 0 0 Siganidae Siganus fuscescens * M 12 0 0 0 5 0 1 Scombridae Scomber australasicus * M 11 0 0 0 0 0 5.5 Cybiosarda elegans * M 30 0 0 0 15 0 0 Monacanthidae Eubalichthys bucephalus M 26 4 3 5 2.5 2 2 Eubalichthys mosaicus * M 33 1 3 2 7 1 4.5 Meuschenia flavolineata M 22 3 1.5 3 1.5 1 4.5 Meuschenia freycineti * M 24 2 2.5 2 1 3 5 Meuschenia trachylepis M 47 5 6 4 4.5 4 6.5 Penicipelta vittiger M 31 8 7 0 0.5 3 2.5 Scobinichthys granulatus M 1 0 0.5 0 0 0 0 Ostraciidae Anoplocapros inermis M 12 0 1 0 4 0 1 Tetraodontidae Canthigaster callisterna M 1 0 0.5 0 0 0 0 Diodontidae Dicotylichthys punctulatus M 19 0 2.5 0 5 0 2 Centrostephanus rodgersii C 3254 178 404 149 475 76 548 relationship Figure

4.2.

Horizontal

between VISIBILITY (METRES) 30 time

visibility -] and periods.

at

the

three

locations TIME 6

7

over time. 9

O 101112

-

Controls,

A-.

Outfall.

Refer

to

Table

4.1

for

CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 79

RESULTS

A total of 17,608 fish comprising 100 species and 46 families were recorded (Table 4.2). In addition,

3,254 individuals of the sea urchin Centrostephanus rodgersii were recorded. Seventy percent of the fish were classified as mobile and thirty percent were cryptic species. The most abundant mobile species were the mado Atypichthys strigatus, yellowtail Trachurus novaezelandiae and eastern pomfred

Schuettea scalaripinnis. The most abundant cryptic species were the bullseyes Pempheris compressus and P. affinis and the white ear Parma microlepis. Thirty six percent of the fish species observed were of commercial/recreational fisheries importance, and 22% of fish species were tropical species (Table

4.2).

Visibility (water clarity) varied from 5 to 20 metres over the study period, with a mean of approximately 10 m. At time 3 (pre-commissioning) the visibility at Boulder Bay was greater than at the control locations, and at times 6 and 11 (post-commissioning) the visibility was less than at the control locations (Figure 4.2).

Multivariate analyses of fish communities

Multivariate analyses indicated that fish communities varied among locations and periods (Table 4.3).

Stress levels for the MDS ordination were high. Stress was 0.18 for untransformed abundance data, and

0.21 for presence/absence data. The stress level for presence/absence data was considered to be difficult to interpret (Clarke 1993) and no further analyses of these data were undertaken. ANOSIM analyses of abundance data indicated that fish communities from the two control locations and three time periods were significantly different for only 4 out of 15 comparisons, however, fish from control

locations were significantly different from those at the outfall location for 14 out of 18 comparisons

(Table 4.3). However, one control location (Tomaree Head) was significantly different to the outfall at all periods (Table 4.3). The fish communities at the outfall were significantly different from the control

locations one year after commissioning of the outfall (Table 4.3, Figure 4.3).

The SIMPER procedure ranked, in order of importance, those species that contributed most to the dissimilarity within the outfall location before and after commissioning (Tables 4.4a). The species Table 4.3. Summary of one-way analysis of similarities (ANOSIM) comparison of fish assemblages at each location and time period. TH - Tomaree Head, PS - Point Stephens, BB -

Boulder Bay, ns - not significant, * significant (P<0.05).

Location TH PS BB

Period 1 2 3 1 2 3 1 2 3

1

TH 2 ns -

3 ns ns

1 ns ns ns

PS 2 * ns ns

3 ns ns ns

1 * * ns ns ns

BB

3 □ A A

Figure 4.3. MDS plot for the abundance of species at each location and period (stress = 0.18).

O - precommissioning at outfall, •- post-commissioning at outfall, □ - Tomaree Head, A -Point Stephens. Note: symbol size increases from Period 1 to 2 to 3. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 80 ranked in the top 5 are included, and it is clear that the relative decline of the fishes Trachinops taeniatus and Trachurus novaezelandiae (and also the sea urchin Centrostephanus rodgersii), and the relative increase in abundance of Pempheris compressus were major contributors to the difference in the fish assemblages pre and post-commissioning of the outfall. The relative abundance of four of these species, together with the absence of Schuettea scalaripinnis at the outfall location, contributed to the difference between outfall and control locations (Table 4.4b).

Univariate analyses of fish assemblages and species

The abundance and richness of fish assemblages, and the abundance of individual species, fluctuated at various spatial and temporal scales. For the sake of brevity a subset of assemblages and common species are presented (Tables 4.5a and 4.5b, Figures 4.4 and 4.5). These variables may be placed into three groups: (i) those for which a significant difference in abundance or richness occurred between the outfall and controls (but not linked to commissioning of the outfall), (ii) those that decreased after outfall commissioning, and (iii) those that increased after commissioning. The first group included total abundance, mobile species abundance, Ophthalmolepis lineolatus, Trachinops taeniatus, Atypichthys strigatus, Cheilodactylus fuscus and Hypoplectrodes mccullochi. The total abundance, mobile species abundance, and the abundances of Trachinops taeniatus, Atypichthys strigatus, Cheilodactylus fuscus and Hypoplectrodes mccullochi were found to be greater at the control locations compared to the outfall location at all times (Tables 4.5a and 4.5b, Figures 4.4 and 4.5). Conversely, Ophthalmolepis lineolatus was found in greater abundance at the outfall location.

The second group included total richness and the abundances of Parma microlepis and

Centrostephanus rodgersii. Fish species richness declined from an average of 15 before commissioning to 10 post commissioning, a press disturbance (defined as a sustained, long-term chronic perturbation) of approximately 33 percent. Centrostephanus rodgersii abundance declined from an average of 10 to less than 5 individuals per transect at the outfall location, a press disturbance of approximately 50 percent. Parma microlepis declined at both the outfall and control locations. The third group included abundance of cryptic species and Pempheris compressus. Cryptic species abundance increased from an average of 10 in periods 1 and 2, to 20 in period 3, which may be characterised as a pulse disturbance Table 4.4a. Relative abundances of the five species that were most responsible for differences between fish

assemblages pre and post-commissioning of the sewage outfall.

Outfall commissioning Cumulative

Species Post (P2 +3) Pre (PI) Percent

Trachinops taeniatus 28.5 103.2 26.7

Trachurus novaezelandiae 33.4 50.0 44.0

Pempheris compressus 35.6 16.0 54.2

Atypichthys strigatus 24.9 25.2 64.1

Centrostephanus rodgersii 28.1 44.5 71.0

Table 4.4b. Relative abundances of the five species that were most responsible for differences between fish

assemblages at the control locations and post-commissioning (Periods 2 and 3) of the sewage outfall.

Cumulative

Species Outfall Controls Percent

Centrostephanus rodgersii 28.1 118.8 14.8

Atypichthys strigatus 24.9 103.0 28.4

Trachurus novaezelandiae 33.4 82.5 40.8

Schuettea scalaripinnis 0.0 90.7 52.5

Pempheris compressus 35.6 85.5 62.7 6 >; ca CQ CO CO CO CO vO«'ONf'"(NVOroOOOC C c a <5 33 o — —. —. —'OOOr^ 3 1 O CQ 11 £50 •2 £3 _C .go na-DOcOMM'OOONCOOION •— r-~ O(N0nC5OOOOOO «3 I 5 a> X ju * C/5 C/5 C/5 C/5 0^f(N^(N — — r-U-5—^(N (3 > O LO —' cx co** cocococococo C/5 * e * * * cccccc #o '^■•OlLOLOt^'— (N-'t — C'' <03 I o — r) CO CO ...... C/5 C/5 C/5 C/5 ca LOLO?'OOOC3CT\»03<03CNO (O'00;rfj«)ff;'0f0'0'0i0 s CN CN LO — Tt h d 6 _c _§) co

O n ~ o o n'f\ S: O'~ voa-oo'oo'oal...- I (NcNCNva—< m vo cx t-- on «o> — •c 1c CO*** cocococococo * C * * * CCCCCC 15 'C'O'tM'nooooa-oor' ho^^^aoooavoor~ LO <50 CN ld rd rj- LO — O c> —: X a- E 05 I la ”c OOCNrt'—'VOLOOO^J’COf^Tj-rt & CN(Nt'--CT'N»03OOOOOOO Eg—1 5ca oocN-a-'oocdooooo ✓—' ------V V § co co co ca s ✓—s O O O cx U d O is is CO .§> co C CO C C > O > O O £50 co O U O U U C > V__ ✓ N__ X s__ X N---- x O jJJJJsda 'S X X X X X X cx X~N X~"S /**V X"""N * O'""''-a x -a /*“N X^-N • •—■ X“"\ >>H CX CX CX Cl, CX CX CX ft CX "5 E '—X n__ / s—x s—x S*x 'w' n—X Jb co _ _t(NS(NS°0^'^0 •& g C CO O o s co 05 CO O is C/5 Q U O ca | 2 a X X E ST 1 o 9a E Cj J .2 -2 3 <5J w S> L-, L- C/0 ^ g -O -O y' w 5^ ca ?Cj 5 X O O o -a 6j •.§ <2 o 4/ ca •8 ‘g e 8 •g | | E— o a. h J cx H at. ^o o ce CD *0 CO*„* C/5 C/5 C/5 CO I c *;*■*• * cccc (5 oor'Jc-oicooirot^r^m 33 ~c a — no — — — — — o 3 O §■ := CQ S 5 bb _c r<-5 (N T5 CO fNO'bfl(NOWC5n O ON 1 COIOOOcf(SOOc*5N- n»n > a OO O

_

D- co co co co co co co to C C C C C C a 2 ■§ ONOO

f O p h th a lm lin e o la tuE s 1 IS ca > coco** cocococococo to C C * * * CCCCCC 15 rs *^* ^o^dd^o^^'0^00 CL a 0(N — (Nr^^O—^OOO ■§ I ^ 3 & Cj CO u ^ E 15 § 'ST lo' a S' '8 X-*\ o X“*\ O o CL u ^ U i= C/5 _S) co C CO C > O > O o bfl O U O CJ U _C > N---■■' ^---X ^—x s—X

versus a a a a -5 o aa « a ar CL X X X X X X X ■g •o E aaaaaaaS ST S o § n_^x S__x n,__/ N----/ \----✓ s__X CL# s_x CL# ca F-ratio CchHHhhHhDHh'Oi c/5

co S' S' ca ca 00 o o CJ ^ .2 O •2 O o ^ -S .2 C/D c 'i x x ^CUDi 'H U O 13 TJ

_d variation CT) -fd| > X o O O 1) 1) > o f s c a

■^r S « te fe 'S .§ .6 -5 _a> ■a w -B ° m^aaaHH § o u S o

300 -j 300 -i

200 - 200 -

100 - 100 -

7 8 9 101112 7 8 9 101112 TIME TIME

(B) (D)

100 -

8 9 101112 9 101112 TIME TIME

Figure 4.4. Mean abundance or species richness of derived variates from outfall and control locations: (a) total abundance, (b) total number of species, (c) mobile species abundance and (d) cryptic species abundance from outfall and control locations. O - Controls, A-. Outfall. Refer to Table 4.1 for relationship between time and periods. (A) (E)

GO -i

UJ UJ O < 1 Q 2 2 ID CQ CO < < 2 < Hi UI :£

9 101112 9 101112 TIME TIME

(B) (F)

LU ui O 2 < 1 Q 2 2 CQ < < 2 < LU UJ :>

9 1011 12 9 101112 TIME TIME

(C) (G)

ui UJ

2 < 1 Q 2 2 £ < <

UI Ui 2 5

9 101112 9 101112 TIME TIME

(D) (H)

9 101112 9 101112 TIME TIME

Figure 4.5. Mean abundance of individual species from outfall and control locations, (a) Trachinops taeniatus, (b)

Atypichthys strigatus, (c) Cheilodactylus fuscus, (d) Ophthalmolepis lineolatus, (e) Hypoplectrodes mccullochi, (f)

Parma microlepis, (g) Pempheris compressus, and (h) Centrostephanus rodgersii. O - Controls, ▲ -. Outfall.

Refer to Table 4.1 for relationship between time and periods. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 81

(defined as a sudden, short-term and possibly acute stress) of 50 percent (Tables 4.5a and 4.5b, Figures

4.4 and 4.5). The pattern for cryptic species was largely based on the abundance of Pempheris compressus.

DISCUSSION

This study has utilised a Before/After/Control/Impact design with multivariate and univariate analyses to investigate the effects on fish assemblages following the commissioning of an ocean sewage outfall at Boulder Bay, New South Wales, Australia. This type of experimental design is optimal to isolate the effect of a putative impact such as sewage pollution from natural spatial and temporal variability.

Clearly, the abundance and richness of fish assemblages fluctuated through time and among locations, and some common species such as Atypichthys strigatus and Cheilodactylus fuscus showed large variability. However, we found differences in fish assemblages at the outfall and control locations that can be largely attributed to the declines in relative abundances of several common, resident species of reef fish and a sea urchin. One of these species, the yellowtail Trachurus novaezelandiae, is a schooling species that is commercially and recreationally important for food and bait. The difference among fish assemblages at Boulder Bay, before and after commissioning of the outfall, can also be largely attributed to the decline of the eastern hulafish, Trachinops taeniatus. This species has previously been reported to be a potential indicator of sewage pollution (Smith 1997). The decline in abundance of some species of fish may be a negative response to physical or chemical characteristics of the sewage (e.g. increased siltation, Roberts et al. 1998), or unknown factors such as increased predation.

A possible confounding factor in this study was the location of a small primary shoreline outfall that discharged untreated sewage into Boulder Bay from the late 1970s to 1993. This outfall continued to discharge sewage until the new offshore outfall was commissioned. The shoreline outfall had a discharge volume of approximately 3.5 ML/day and was located on the north side of Boulder Bay, approximately 500 m from the present deepwater outfall. The suggestion that the shoreline outfall may have confounded the results of the deep water outfall may be supported by the lower abundances of CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 82 several derived fish assemblage variates and abundances of individual species, such as total fish abundance and the abundance of Hypoplectrodes mccullochi, prior to commissioning of the deep water outfall at Boulder Bay. However, because of the small size of the outfall and the distance of 500 m from the present study, the potential for impact is considered small.

The type of data transformation used for the multivariate analysis of fish communities may be a limitation of the study. Abundance data, which I presented, may lead to a shallow interpretation of the data in which only the pattern of a few very common species are represented (Clarke 1993, Warwick

1993). I chose not to pursue analysis of the presence/absence data because the stress levels from the

MDS were high. However, I recognise that this type of transformation or other transformations such as log or double square root give emphasis to rare species, and would have provided a different perspective of the fish communities at the outfall and control locations.

Comparison with other studies

Many studies of the impacts of sewage are equivocal (see Chapman et al. 1995, EPA 1996, Otw'ay et al. 1996, Roberts 1996). This may be explained by the lack of‘before data’ due to commissioning of outfalls many years previously, poor experimental design or because the potential impacts of sewage pollution are often confounded by other diffuse pollutants or activities which also impact on aquatic ecosystems (Grigg 1994, 1995, Griswold 1997). However, a number of studies have reported improvements in aquatic ecosystems following the reduction or cessation of sewage disposal (Oviatt et al. 1984, Hardy et al. 1993, Scanes and Philip 1995, Bokn et al. 1996).

Enhanced fish populations near deepwater outfalls off Hawaii were reported by Grigg (1994) and

Russo (1982, 1989). Large numbers of the snapper (Lutjanus kasmira) were reported directly above a diffuser, and the relative number of herbivorous fishes increased adjacent to the outfall (Russo 1989).

Localised increases in the abundance of bait fish were correlated with the disposal of effluent from a sewage outfall in the Mediterranean Sea by Bell and Harmelin-Vivien (1982). Increases in the abundances of some species of demersal fish, including gurnard (Lepidotrigla mulhalli) and flathead

(F. Platycephalidae) at Bondi deepwater ocean outfall, and decreases in the abundances of a range of CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 83 species including Australian snapper (Pagrus auratus), flounder (F. Pleuronectidae), flathead, leatherjacket (F. Monacanthidae) and rays (F. Urolophidae), at North Head and Malabar sewage outfalls were reported by Otway (1995), EPA (1996) and Otway et al. (1996).

This research at Boulder Bay outfall showed that the abundances of cryptic fish species and pempherids increased as a result of sewage effluent, but also detected declines of 30 to 50 percent in the abundances of several common, resident species of reef fish and a sea urchin.

Management Implications

Minimal treatment and nearshore disposal of sewage are not acceptable in today’s increasingly populated, regulated and environmentally concerned society (Beder 1989, Johnston et al. 1993, Clark

1995, Baird 1996). The public concern over waste disposal, and in particular about elevated levels of contaminants in fish (EPA 1995), is likely to have been the trigger for upgrading the existing shoreline sewage outfall at Boulder Bay to an offshore deepwater outfall. It is surprising, however, that an

Environmental Impact Statement (EIS) was not prepared to fully assess the environmental impacts of the proposal. If management decisions are made to change a historical practice, such as the location and/or operation of an existing sewage outfall, then decisions should be made on a precautionary approach and supported by sound scientific data (Peterson 1993). Similar upgrades and EISs or

Environmental Management Programs have been undertaken for the large outfalls in the Sydney region

(Otway 1995, Scanes et al. 1995), and many ocean sewage outfalls have been constructed further offshore and in deeper water in other countries (Chang 1993, Grigg 1994, 1995).

A biological survey of Boulder Bay was undertaken by Laurie, Montgomerie and Pettit (1977) prior to the construction of the shoreline sewage outfall. Although fish assemblages were not studied, the authors suggested that the outfall would have no impact on biological communities. No monitoring to test this hypothesis, however, was undertaken in subsequent years.

This study focused on the impacts of sewage effluent on fish assemblages and it has demonstrated that there are some statistically significant differences between fish assemblages at outfall and control CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 84 locations at the community and individual species levels. The question that managers and the community must now resolve is how these scientific findings relate to current management actions and what is the ecological significance of the changes. There are two major documents which currently apply to management of water quality and ocean outfalls in NSW waters; ANZECC’s (1992)

Australian water quality guidelines for fresh and marine waters, and the EPA’s (1993b) discharge of wastes to ocean waters. A policy and guideline document for aquatic habitat management and fish conservation should also be considered (NSW Fisheries 1998).

ANZECC (1992) recommends consideration of four principal factors for biological indicators: species richness, species composition, primary production and ecosystem function (ANZECC 1992). Any change to any of these factors at an impacted location compared to other local, unimpacted locations should not be permitted according to the ANZECC guidelines. No specific values for “significant” are recommended for biological indicators, however, the value will be dependent on the statistical power of the study and the magnitude of difference that is considered biologically important. As a general rule, a change of 20% in a biological indicator, one year after the impact should be regarded as a major impact and require environmental compensation (NSW Fisheries 1998).

The NSW EPA’s (1993b) guidelines are similar to the ANZECC (1992) guidelines; however, one section in the former document indicates that wastes should not be discharged to ocean waters if they give rise to a statistically significant impact on the biological communities. Therefore, the present outfall may not be complying with the EPA (1993b) and ANZECC (1992) guidelines. Changes in fish and sea urchin abundance that were greater than 20% were attributed to sewage discharge, and according to NSW Fisheries (1998) this level of change should generally be regarded as a major impact.

In regard to the ecological significance of these findings, our study has demonstrated impacts in the immediate vicinity of the outfall (i.e. within 100 metres of the diffuser), and suggested that there may have been a confounding impact from the previous shoreline outfall (i.e. within 500 metres of the diffuser). The collection of data to explain the mechanisms for increased or decreased abundance of CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 85

Fish at the outfall was beyond the scope of this study, however, I suggest that a number of mechanisms, including increased siltation (Roberts et al. 1998), may be responsible because I detected both press

(e.g. total fish assemblage, Parma microlepis, Centrostephanus rodgersii) and pulse (e.g. Cryptic species abundance) disturbances.

The outfall at Boulder Bay may be regarded as a small outfall (approximately 5 ML/day) but it is important to consider the cumulative impacts of similar outfalls. There are approximately 35 ocean outfalls in NSW of which half may be regarded as small, with the remainder having an average dry weather flow (ADWF) of up to 50 ML/day and there are three very large outfalls with an ADWF of greater that 140 ML/day (MF1L 1997). If all or most of these outfalls have a significant impact on rocky reef fish communities it is important that we quantify and reduce the impacts over time.

CONCLUSIONS

Fish have frequently been used as bio-indicators as they are relatively easy to identify and are of particular interest to commercial and recreational fishers and the public (Warwick 1993, Harris 1995,

Lincoln Smith and Jones 1995). There are very few cases in which more than one component of the biota has been examined in pollution impact studies (Warwick 1993). The findings of this study are supported by a parallel study of the effects of sewage on the ecology of encrusting marine assemblages

(Roberts et al. 1998) which reported rapid reductions in the cover of crustose and foliose algae and sponges and increased cover of silt.

The present guideline documents, which currently provide direction for the management of ocean outfalls in NSW waters, need to be modified to include specific values for acceptable changes to biological communities. Future monitoring is recommended at Boulder Bay and the control locations, and if biological impacts continue to result in the long-term, then this may be a strong argument for the use of adaptive management strategies (Walters 1986) where particular factors such as volume, periodicity and treatment of sewage may be changed and monitored to improve performance. The integration of an optimal experimental design and research with specific management guidelines and compliance is essential for a better understanding of anthropogenic impacts such as sewage pollution. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 86

REFERENCES

Adams, S.M., Brown, A.M. & Goede, R.W. (1993). A quantitative health assessment index for rapid

evaluation of fish condition in the field. Transactions of the American Fisheries Society 122, 63-

73.

ANZECC (1992). Australian water quality guidelines for fresh and marine waters. Australian and New

Zealand Environment and Conservation Council, Canberra

Baird, R.C. (1996). Towards new paradigms in coastal resource management: linkages and institutional

effectiveness. Estuaries, 19, 320-335.

Beder, S. (1989). Toxic fish and sewer surfing. Allen & Unwin, Sydney.

Bell, J.D. & Harmelin-Vivien, M.L. (1982). Fish fauna of French Mediterranean Posidonia Oceania

seagrass meadows. 1. Assemblage Structure. Tethys, 10, 337-347.

Boates, W.A.M. & Russell, K.S. (1992). The option to discharge industrial or sewage effluent to the

marine environment. Water Science Technology IS, 75-82.

Bokn, T.L., Frithjof, E.& Walday, M. (1996). Improvement of the shallow water communities

following reductions of industrial outlets and sewage discharge in the Hvaler estuary, Norway.

Hydrobiologica 326/327, 297-304.

Brown, V.B., Davies, S.A. & Synnot, R.N. (1990). Long-term monitoring of the effects of treated

sewage effluent on the intertidal macroalgal assemblage near Cape Schanck, Victoria, Australia.

Botanica Marina 33, 85-98.

Chang, S. (1993). Analysis of fishery resources: potential risk from sewage sludge dumping at the

deepwater dumpsite off New Jersey. Fishery Bulletin 91, 594-610.

Chapman, M.G., Underwood, A.J. & Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

Journal of Experimental Marine Biology and Ecology 189, 103-122.

Clarke, K.R. (1993). Non-parametric multivariate analyses of changes in community structure.

Australian Journal of Ecology 18, 117-143. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 87

Clark, R.B. (1995). Pollution, people, the press and the Bulletin. Marine Pollution Bulletin 30, 2-3.

EPA (1993a). Deformities and associated sublethal effects in fish exposed to sewage-borne

contaminants: literature review. Environment Protection Authority Report 93/72. EPA, Sydney.

EPA (1993b). Discharge of wastes to ocean waters. Prepared by Heggie, A. & MacDonald, R., EPA,

Sydney.

EPA (1995). Hunter Environmental Monitoring Program 1992-1994. Prepared by Ajani, P., EPA,

Chatswood, Sydney. 27 pp.

EPA (1996). Assessment of the Deepwater Outfalls. Final Report Series. Volume 1. Prepared by

Pritchard, T., Rendell, P., Scanes, P. & N. Philip. EPA, Chatswood, Sydney. 194 pp.

Fagan, P., Miskiewicz, A.G. & Tate, P.M. (1992). An approach to monitoring sewage outfalls. A case

study on the Sydney Deepwater Sewage Outfalls. Marine Pollution Bulletin 25, 172-180.

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environmental Monitoring and Assessment 14, 197-210.

Gray, J.S. (1989). Effects of environmental stress on species rich assemblages. Biological Journal of

the Linnean Society 37, 19-32.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Marine Ecology Progress Series 103, 25-34.

Grigg, R.W. (1995). Coral reefs in an urban embayment in Hawaii: a complex case history controlled

by natural and anthropogenic stress. Coral Reefs 14, 253-266.

Griswold, B.L. (1997). Fisheries and pollution: a conference overview. Marine Pollution Bulletin 126,

504-505.

Hardy, F.G., Evans, S.M. & Tremayne, M.A. (1993). Long-term changes in the marine macroalgae of

three polluted estuaries in north-east England. Journal of Experimental Marine Biology and

Ecology 172, 81-92.

Harris, J.H. (1995). The use of fish in ecological assessments. Australian Journal of Ecology 20, 65-80. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 88

Johnston, P.A., MacGarvin, M., Stringer, R.L., Troendle, S. & Swindlehurst, R.J. (1993). Sewage:

towards realistic environmental protection. Water Science Technology 27, 481-491.

Keough, M.J. & Quinn, G.P. (1991). Causality and the choice of measurements for detecting human

impacts in marine environments. Marine and Freshwater Research 42, 539-554.

Laurie, Montgomerie & Pettit (1977). Boulder Bay ocean outfall. Abstract of report on environmental

impact on marine communities. Report to Hunter District Water Board, Newcastle. 42 pp.

Lincoln Smith, M.P. (1985). The development and application of visual survey procedures for fish

assemblages on shallow rocky reefs. MSc thesis, Univ. Sydney, NSW, Australia. 187 pp.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate

rocky reef fish assemblages. Marine Ecology Progress Series 43, 223-231.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Environmental Biology of Fishes 26, 29-37.

Lincoln Smith, M.P. & Jones, G.P. (1995). Fishes of shallow coastal habitats, pp 240-253 in

Underwood, A.J. & Chapman, M.G. (Eds) Coastal marine ecology of temperate Australia.

UNSW Press, Sydney.

Love, M.S., Axell, B., Morris, P., Collins, R. & Brooks, A. (1987). Life history and fishery of the

Californian scorpionfish, Scorpaena guttata, within the southern Californian Bight. Fishery

Bulletin 85, 99-116.

May, V. (1985). Observations on algal floras close to two sewerage outlets. Cunninghamia 1, 385-394.

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Report MHL769 by

Manly Hydraulics Laboratory and the Ecology Lab Pty Ltd. 65 pp.

NSW Fisheries (1998). Policy and Guidelines, Aquatic Habitat Management and Fish Conservation

(Eds. A.K. Smith and D.A. Pollard). Sydney, 76 pp.

Otway, N.M. (1995). Assessing impacts of deepwater sewage disposal: a case study from New South

Wales, Australia. Marine Pollution Bulletin 31, 347-354. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 89

Otway, N.M, Sullings, D.J. & Lenehan, N.W. (1996). Trophically-based assessment of the impacts of

deepwater sewage disposal on a demersal fish community. Environmental Biology of Fishes 46,

167-183

Oviatt, C.A., Pilson, M.E.Q., Nixon, S.W., Frithsen, J.B., Rudnick, D.T., Kelly, J.R., Grassle, J.F. &

Grassle, J.P. (1984). Recovery of a polluted estuarine system: a mesocosm experiment. Marine

Ecology Progress Series 16, 203 - 217.

Peterson, C.H. (1993). Improvement of environmental impact analysis by application of principles

derived from manipulative ecology: lessons from coastal marine case histories. Australian

Journal of Ecology 18, 21-52.

Roberts, D.E. (1996). Patterns in subtidal marine assemblages associated with a deep-water sewage

outfall. Marine and Freshwater Research 47, 1-9.

Roberts, D.E., Smith, A., Ajani, P. & Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACF approach.

Marine Ecology Progress Series 163, 213-224.

Russo, A.R. (1982). Temporal changes in fish community structure near a sewage ocean outfall,

Mokapu, Oahu, Hawaii. Marine Environmental Research 6, 83-98.

Russo, A.R. (1989). Fish community structure on a deep ocean outfall, Barber’s Point, Oahu, Hawaii.

Int. Revue, ges. Hydrobiologica 74, 499-506.

Scanes, P.R. & Philip, N. (1995). Environmental impact of deepwater discharge of sewage off Sydney,

NSW, Australia. Marine Pollution Bulletin 31, 343-346.

Scanes, P.R., Scanes, K. & Otway, N.M. (1995). Environmental problems due to disposal of wastes,

pp 297-310 in Underwood, A.J., Chapman, M.G. (Eds) Coastal marine ecology of temperate

Australia. UNSW Press, Sydney, Australia.

Smith, A. (1997). Is the eastern hula a potential indicator of sewage pollution? Abstract from 23rd

Annual Workshop of Australian Society for Fish Biology and Fish and Aquatic Resource

Management Association of Australia, Darwin, NT, Australia. 41 pp. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 90

Smith, S.D. (1994). Impacts of domestic sewage effluent verses natural background variability: an

example from Jervis Bay, New South Wales. Marine and Freshwater Research 45, 1045-1064.

TEL (1991). Changes to assemblages of reef fish adjacent to the North Head cliff face outfall following

decommissioning. Progress Report by The Ecology Lab for the Water Board Environmental

Management Unit, Sydney. 11 pp.

TEL (1992). Surveys of fish and epibenthic assemblages in the Illawarra coastal zone. Preliminary

Report by The Ecology Lab for the Water Board Environmental Management Unit, Sydney. 11

pp.

TEL (1993). Changes to assemblages of reef fish adjacent to the North Head and Bondi cliff face

outfall following decommissioning. Final Report by The Ecology Lab for the Water Board

Environmental Management Unit, Sydney. 49 pp.

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final Report by The Ecology Lab for the Water Board’s

Environmental Management Unit, Sydney. 29 pp.

Underwood, A.J. (1991). Beyond BAC1: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Marine and Freshwater Research 42, 569-

587.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. Journal of Experimental Marine Biology and Ecology 161, 145-178.

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Australian Journal of Ecology 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecological Applications 4, 3-15.

Underwood, A.J., Kingsford, M.J. & Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Australian Journal of Ecology 6, 231-249.

Walters, C.J. (1986). Adaptive management of renewable resources. MacMillan, New York. 374 pp. CHAPTER 4 EFFECTS OF SEWAGE ON FISH IN THE HUNTER 91

Warwick, R.M. (1993). Environmental impact studies on marine communities: Pragmatical

considerations. Australian Journal of Ecology 18, 63-80.

Winer, B.J., Brown, D.R. & Michels, K.M. (1991). Statistical Principals in Experimental Design.

McGraw-Hill, New York, 1057 pp. Sewage plumes from large, primary treatment, shoreline outfall in the Sydney region. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 92

CHAPTER 5

EFFECTS OF SHUTDOWN OF A NEARSHORE SEWAGE

OUTFALL IN NSW, AUSTRALIA, ON ASSEMBLAGES OF

TEMPERATE REEF FISHES.

ABSTRACT

A very large primary treated sewage outfall which discharged into the nearshore subtidal habitat at

North Head, Sydney was shutdown in 1990. Temperate reef fish assemblages were surveyed at two locations previously adjacent to the outfall (north and south), and two control locations, from the day after the outfall was shutdown over a period of 13 months. The abundance of fish were censused visually on the rocky reef environment at boulder and wall habitats by SCUBA divers using a timed- interval technique. Multivariate analyses indicated that assemblages of fish associated with a boulder habitat were significantly different at the outfall and control locations. The outfall locations were differentiated by relatively large numbers of Acanthopagrus australis (Sparidae), Atypichthys strigatus and Scorpis lineolatus (Scorpididae), and Cheilodactylus fuscus (Cheilodactylidae) and small numbers of Trachurus novaezelandiae (Carangidae) and Chromis hypsilepis (Pomacentridae). The abundance of several other species appeared to be positively or negatively impacted by sewage discharge, for example Hypoplectrodes mccullochi (Serranidae) occurred in larger numbers at control locations and

Parma unifasciata (Pomacentridae) were only found at the control locations. The results are compared with those from other studies on the effects of sewage on fish and benthic assemblages and fisheries.

INTRODUCTION

Management decisions regarding treatment and disposal of sewage are being increasingly considered because of large numbers of people living close to waterbodies, increased environmental awareness and studies documenting the harmful effects of sewage effluent to human health and aquatic biota

(Koop and Hutchings 1996, Leadbitter 1996). It is possible to control and reverse the impacts of CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 93 sewage pollution by methods such as effluent reuse, enhanced treatment and relocation of outfalls.

Several studies have reported rapid improvements in aquatic ecosystems, fish and fisheries following the implementation of such methods (Oviatt et al. 1984, GESAMP 1990, Hardy et al. 1993, Scanes and

Philip 1995, Bokn et al. 1996, Scanes 1996).

Fish assemblages from tropical coral and temperate rocky reefs are relatively diverse, abundant and many species are relatively site-attached or sedentary. Because of this, they have been regarded as good indicators of environmental change (Warwick 1993, Lincoln Smith and Jones 1995). Many studies have focused on the effects of sewage on fish populations (Love et al. 1987; Gray et al. 1992;

Grigg 1994; Kingsford and Gray 1996; Otway et al. 1996a,b; Waring et al. 1996; Lye et al. 1997). Fish may exhibit a range of responses in a polluted area including changes to abundance, mortality, fecundity, size, toxic effects, susceptibility to infections and behavioural responses (Gray 1989, EPA

1993, Otway et al. 1996a,b, Smith et al. 1998).

The volume, duration and level of treatment of sewage, and the type of fish that are exposed are important determinants of the scale of any environmental impact. In general, striking effects are very localised (10’s of metres), but detectable effects may be found at distances of up to several kilometres.

Several studies have reported no significant impact due to sewage discharge (Lincoln Smith 1985,

Scanes et al. 1995), while others have reported community or trophic impacts, (Harmelin-Vivien

1992, Grigg 1994, Otway et al. 1996b, Smith et al. 1998) or have reported an impact restricted to a small number of species (e.g. Otway et al. 1996a,b).

The recovery of aquatic biota after a sewage outfall is shut down would be predicted to be reasonably rapid if movement of fish was a major determinant of fish population structure, because sewage generally causes chemical rather than a catastrophic or habitat altering disturbance. For example, the upgrade of sewage treatment in the Thames, UK and the Potomac River, USA has resulted in the return of salmonids (Arthington and Welcomme 1995). Cessation of sewage sludge dumping in the New

York Bight, USA resulted in increased biomass of the American lobster Homarus americanus (Chang

1993). Other biota such as macroalgae (Hardy et al. 1993) and macroinvertebrates (Oviatt et al. 1984) CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 94 may also recover rapidly (less than one year) if the impact of sewage is mitigated or removed.

Conversely, recovery of aquatic biota may also take several decades after a sewage outfall is shut down or modified. For example, discharge of industrial chemicals was widespread in the early 1960s and

1970s, and although discharge has since been restricted, the stable nature of chemicals such as DDT and PCBs has caused long-term impacts on aquatic biota (CEPA 1991, Bignert et al. 1998).

Qualitative information on fish adjacent to the North Head sewage outfall was obtained by Jones

(1977). He used underwater visual census in depths from 2 to 27 m and recorded 53 species. The most common species were Pempheris compressus, Parma microlepis, Chironemus marmoratus and

Cheilodactylus fuscus.

The previous nearshore sewage outfall at North Head resulted in beach pollution, abundant intertidal

areas of green alga Ulva sp, accumulation of contaminants in biota and the prohibition of fishing with

500m of the outfall (Beder 1989, Fairweather 1990, Philip and Pritchard 1996). These issues contributed to the construction of deepwater ocean outfalls and the closure of the nearshore outfalls in the Sydney region (Fagan et al. 1992). A study of contamination in fish and oysters clearly showed that

shutdown of the nearshore outfall at North Head has resulted in a significant drop in metal and

organochlorine levels in biota (Scanes 1996, Krogh and Scanes 1996). A study of subtidal sessile fauna

after shutdown of the nearshore outfall indicated declines in abundance of ascidians Pyura stolonifera

and P. gibbosa at the outfall, however there were few significant differences in either mean measures

or variances of most fauna at the outfall compared to control locations (Chapman et al. 1995,

Underwood and Chapman 1996).

The aim of the present study was to measure changes in reef fish at a large outfall compared to

reference locations immediately and for over one year following shut-down of the outfall. In this case,

as with many other sewage outfalls, there were no quantitative data available on fish at the outfall and

control locations prior to commissioning of the outfall. This would have allowed an optimal sampling

design conforming to the Beyond BACI approach (Underwood 1991, 1992, 1993, 1994). An

alternative approach in this case was to rely on spatial comparisons at times after shut-down (e.g. oNHN • NHS

SYDNEY HARBOUR

|A Gap

□ DH

Figure 5.1. The study locations in the Sydney region, NSW, Australia. NHN - North Head North and

NHS - North Head South are the sewage outfall locations. The Gap and DH - Dover Heights are the control locations. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 95

Green 1979). This approach has been adopted in other studies of outfalls (e.g. Chapman et a/. 1995) and detecting the effects of other human activities, such as marinas (Glasby 1997).

METHODS

Study designs

The North Head outfall (Sydney, Australia. 33°50’S, 151°15’E) which discharged an average dry weather release of 293 ML (Millions of litres) day"1 of primary treated effluent was shut-down in

December 1990. Reef fish were surveyed at two locations at North Head (denoted north and south), and at two control locations, one each at Dover Heights and The Gap (Figure 5.1). Fish were censused at all locations five times, from 2 days to 59 weeks after the outfall was turned off (Table 5.1).

All locations were at the base of steep cliffs along the open coast and exposed to a wide range of sea conditions. The North Head locations had relatively distinctive subtidal habitats for the region, which limited the number of control locations available for comparison. At each location a vertical wall extended from the intertidal to depths of 10-15 m. At the base of the wall a boulder field sloped away to depths in excess of 25-30 m, very close to the shoreline. This boulder field contained massive boulders rising up to 5 m above the substratum. The wall habitat was dominated by turfing brown and red algae with numerous solitary ascidians. There were also several cracks in the wall (typically <30 cm wide and < 100 cm deep) which sheltered sea urchins and gastropods. The boulder habitat was made up of rock “barrens” (Underwood et al. 1991) dominated by sea urchins (particularly

Centrostephanus rodgersii) and gastropods, but the tops of the higher boulders contained turfing algae, ascidians and some Ecklonia radiata.

Fish surveys

Fish were surveyed on the wall habitat (at 3-8 m depth) and in the boulder habitat (at 15-18 m depth).

Fish were categorised as mobile or cryptic species (Lincoln Smith 1988, 1989) and each category was surveyed separately. Mobile fish which comprised fast, large or conspicuous species, were visually censused by a SCUBA diver using a timed interval technique (Lincoln Smith 1985, Thresher and Gunn

1986, McCormick and Choat 1987, Sale 1994, Gillanders 1997). This technique was chosen because of Table 5.1. Survey number, start and completion of survey, weeks since North Head cliff-face outfall ceased operation and season.

Survey Started Completed Weeks Season

T1 20/12/90 11/01/91 1-4 Summer

T2 24/01/91 15/02/91 6-9 Summer

T3 5/04/91 28/05/91 16-23 Autumn

T4 29/07/91 5/08/91 32-34 Winter

T5 16/01/92 28/01/92 57-59 Summer CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 96 time constraints of SCUBA diving in deep water, and the difficulty of surveying vertical wall habitats with transect techniques. Six replicate counts each of 3 minutes duration were made during daylight hours as a diver swam 1-2 m above the reef, approximately parallel with the shoreline. Fishes were recorded to a distance of approximately 5 m on either side of the diver.

Cryptic fish which comprised small, cave dwelling or camouflaged species were visually censused by a

SCUBA diver using a line transect technique (Lincoln Smith 1988, 1989). Four replicate 25 m line transects were laid over the reef and cryptic fish were counted 1 m either side of the transect.

Statistical analyses

Data for fish assemblages were analysed using non-metric multi-dimensional statistical (MDS) techniques (Clarke 1993) to describe the variation between locations and habitats, using the PRIMER software package (Plymouth Marine Laboratories, UK). The two categories of fishes were analysed separately using the Bray-Curtis similarity measure with untransformed data. One-way analysis of similarity (ANOSIM) tests were used to test the hypothesis that fish assemblages of each category from four locations were similar. Two-way crossed ANOSIM tests were used to test the hypothesis that fish assemblages from outfalls and controls did not vary over time. The similarity percentage (SIMPER) procedure (Plymouth Marine Laboratories, UK) was used to identify the contribution of individual species to differences between locations.

A 2-way Analysis of Variance (ANOVA) was used to compare selected variables (e.g. total abundance, species richness and abundant species) among the 4 locations and 5 times. Only variables from the boulder habitat are presented, because the wall habitat generally contained large variability. All factors were considered as fixed factors. Data were transformed to log(x+l) to stabilise variances (Underwood

1997). CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 97

RESULTS

Abundance and richness of fish

Overall, the assemblage was both diverse (121 species from 49 families) and abundant (-91,000 fish recorded) (Table 5.2). Seventy seven percent of the fish were categorised as mobile and 23% were cryptic. The most abundant mobile species were Trachurus novaezelandiae, Atypichthys strigatus and

Acanthopagrus australis. The most abundant cryptic species were Trachinops taeniatus, Parma microlepis and Chironemus marmoratus. Some species of fish were recorded in large numbers in

boulder habitat (e.g. Pempheris multiradiata), wall habitat (e.g. Chironemus marmoratus) or were only recorded from outfall locations (e.g. Tetractenos glaber) or control locations (e.g. Cor is picta) or at only one location (e.g Argyrosomus japonicus) (Table 5.2).

Multivariate analyses of fish assemblages

Multivariate analyses indicated that assemblages of mobile and cryptic fish varied between habitats and

locations (Table 5.3, Figure 5.2) and times (Figure 5.3). ANOSIM analyses indicated significant differences between locations for mobile fish in boulder habitat and cryptic fish in both habitats, while

no difference was detected for mobile fish on wall habitat. Pairwise comparisons indicated that fish

assemblages at the two outfall locations were similar, fish assemblages at outfall locations were

generally different from The Gap but were similar to Dover Heights (Table 5.3). The only consistent

outfall effect was for mobile fish in boulder habitat where the North Head North location was different

to the two controls (Table 5.3). In this case, however, the two control locations were significantly

different to each other. Thus, the ANOSIM results were ambiguous in terms of detecting effects of the

outfall.

As a result of a consistent outfall effect, further analysis was undertaken on mobile fish in boulder

habitat to investigate temporal patterns. ANOSIM indicated a significant difference between times

(a<0.05), however, no significant pairwise temporal correlations (e.g. Outfalls at Time 1 verses

Outfalls at Time 2) were detected. Although the fish assemblages from the outfall and control locations

changed over time, there was no ‘recovery’ of fish assemblages at the outfall (Figure 5.3). Table 5.2. Total abundance of fish at outfall and control locations in boulder and wall habitats. M -

mobile, C - cryptic, NN - North Head North (outfall), NS - North Head South (outfall), G - The Gap, D -

Dover Heights.

BOULDER WALL Family Species Type NN NS G D NN NS G D Heterodontidae Heterodontus portjacksoni M 1 0 0 0 0 0 2 0 Brachaeluridae Brachaelurus waddi C 0 2 0 0 4 6 1 3 Brachaelurus sp. C 0 0 2 0 1 0 0 0 Orectolobidae Orectolobus sp c 3 0 2 0 2 0 0 0 Dasyatididae Dasyatis brevicaudatus M 1 0 1 0 0 0 0 0 Dasyatis thetidis M 4 0 0 0 0 0 0 0 Myliobatidae Myliobatis australis M 0 0 0 0 0 0 0 1 Muraenidae Gymnothorax prasinus C 20 8 4 5 3 8 18 8 Aulopidae Aulopus purpurissatus M 3 3 3 2 0 0 0 1 Synodontidae Synodus variegatus M 0 0 1 0 0 0 0 0 Gobiesocidae Aspasmogaster costatus C 1 0 1 0 0 0 0 0 Moridae Lotella rhacinus C 26 26 0 13 1 4 3 0 Trachichthyidae Optivus elongatus c 3 0 0 0 0 0 0 0 Trachichthys australis c 14 25 6 32 0 1 0 0 Scorpaenidae Pterois volitans c 1 0 0 0 0 0 0 1 Scorpaenodes scaber c 1 0 0 0 0 0 0 0 Scorpaena cardinalis c 0 6 5 18 1 3 1 3 Serranidae Acanthistus ocellatus c 30 15 10 8 6 7 12 4 Hypoplectrodes annulata c 0 0 0 1 0 0 0 0 Hypoplectrodes mccullochi c 22 24 184 132 6 3 14 16 Trachypoma macracanthus c 0 0 0 1 0 0 0 0 Anthias sp. (orange) c 0 0 0 0 2 1 3 0 Plesiopidae Paraplesiops bleekeri M 0 0 0 3 0 0 0 0 Trachinops taeniatus c 1292 1458 2943 1448 1394 1019 3073 2019 Apogonidae Apogon aureus c 0 0 0 2 0 0 0 0 Apogon cooki c 1 0 0 0 0 0 0 0 Apogon limenus c 0 0 0 4 0 0 0 0 Apogon sp. c 0 0 1 0 0 0 0 0 Dinolestidae Dinolestes lewini M 254 6 187 149 3 4 19 4 Pomatomidae Pomatomus saltator M 0 200 0 0 9 0 19 6 Echeneididae Remora remora M 1 0 0 0 0 0 0 0 Carangidae Pseudocaranx dentex M 113 127 7 196 23 61 0 0 Seriola lalandi M 7 10 93 22 3 0 5 56 Trachurus declivis M 0 0 6 0 0 0 0 0 Trachurus novaezelandiae M 900 3625 20819 4033 100 2002 7812 5097 Sparidae Acanthopagrus australis M 842 783 3 85 267 465 20 147 Pagrus auratus M 2 1 3 24 0 0 0 0 Rhabdosargus sarba M 86 83 0 30 2 9 0 0 Sciaenidae Argyrosomus japonicus M 0 49 0 0 0 0 0 0 Mullidae Parupeneus signatus M 13 15 7 17 6 1 1 12 Upeneichthys lineatus M 1 0 0 6 0 0 0 0 Monodactylidae Monodactylus argenteus M 0 0 0 0 0 0 5 0 Schuettea scalaripinnis M 0 10 133 109 100 67 349 164 Pempherididae Pempheris affinis C 99 5 37 5 0 0 0 0 (PTO) Table 5.2 contd Pempheris anal is C 0 11 0 1 0 0 0 0 Pempheris compressus C 202 261 94 120 57 0 25 30 Pempheris multiradiata c 251 309 74 218 0 0 0 0 Girellidae Girella elevata M 26 45 24 8 8 9 3 1 Girella tricuspidata M 29 21 0 43 669 445 66 119 Kyphosidae Kyphosus sydneyanus M 2 0 20 0 0 0 1 0 Scorpididae Atypichthys strigatus M 815 519 400 230 2950 2049 2576 560 Microcanthus strigatus M 3 11 14 0 0 0 0 0 Scorpis lineolatus M 591 438 175 433 174 73 41 33 Lutjanidae Lutjanus argentimaculatus M 1 0 0 0 0 0 0 0 Paracaesio xanthurus M 0 0 0 0 0 0 0 14 Chaetodontidae Chaetodon guntheri C 3 5 0 1 2 0 0 0 Chaetodon lunula C 0 0 0 0 0 0 1 0 Chelmonops truncatus C 1 1 0 0 0 0 0 0 Heniochus sp. C 1 2 0 0 0 0 0 0 Enoplosidae Enoplosus armatus M 14 11 24 2 0 0 3 0 Pomacentridae Abudefduf vaigiensis M 0 0 0 0 0 0 0 2 Chromis hyps Hep is M 44 53 354 1307 18 71 163 80 Dischistodus prosopotaenia C 0 0 0 0 0 1 1 0 Mecaenichthys immaculatus C 1 3 0 1 0 0 0 0 Parma microlepis C 245 280 281 305 79 79 138 148 Parma unifasciata C 0 0 128 32 46 36 201 51 Pomacentrus coelestis M 3 6 0 2 0 0 0 0 Stegastes gascoynei C 0 0 0 0 1 0 0 0 Pomacentrid sp. C 0 0 0 0 1 0 1 0 Cirrhitidae Cirrhitichthys aprinus c 4 4 1 0 0 0 0 0 Chironemidae Chironemus marmoratus c 49 8 11 2 361 250 308 182 Aplodactylidae Crinodus lophodon M 7 8 7 13 79 153 440 225 Cheilodactylidae Cheilodactylus fuscus M 396 403 81 212 81 58 26 46 Cheilodactylus spectabilis M 0 0 0 0 1 1 0 0 Cheilodactylus vestitus M 0 1 0 0 0 0 0 0 Nemadactylus douglasii M 0 2 0 0 0 1 0 0 Latrididae Latridopsis forsteri M 26 3 0 5 0 0 0 0 Labridae Achoerodus viridis M 264 253 30 72 172 177 173 108 Austrolabrus maculatus M 0 0 1 0 0 0 0 3 Coris dorsomaculata M 0 0 0 2 0 0 0 0 Cor is picta M 0 0 15 39 0 0 0 0 Coris sandageri M 0 0 0 1 0 0 0 0 Eupetrichthys angustipes M 0 0 0 2 0 0 0 0 Labroides dimidiatus M 10 6 2 2 0 0 3 0 Notolabrus gymnogenis M 36 35 33 31 60 31 61 80 Ophthalmolepis lineolatus M 8 22 14 31 0 0 0 0 Pictilabrus laticlavius M 0 0 0 0 3 0 3 6 Pseudolabrus guntheri M 0 0 1 0 1 0 3 4 Stethojulis interrupta M 0 0 0 2 0 0 0 0 Thalassoma lunare M 0 0 6 3 1 0 1 0 Odacidae Odax cyanomelas M 0 0 1 0 0 0 18 0 Blenniidae Aspidontus taeniatus C 0 0 2 0 0 0 0 1 Aspidontus dussummieri C 1 1 0 0 0 0 0 0 Petroscirtes fallax c 2 0 1 0 1 0 2 4 Pictiblennius tasmaniensis c 0 1 0 0 1 0 0 0 Plagiotremis rhinorhynchos c 1 0 3 5 0 0 0 0 Plagiotremis tapeinosoma c 7 8 12 5 1 1 16 16 Tripterygiidae Norfolkia clarkei c 1 1 1 0 0 0 2 0 Clinidae Heteroclinus perspicillatus c 0 0 0 0 0 2 0 0 (PTO) Table 5.2 cont’d Heteroclinus whiteleggii C 0 0 0 0 0 0 0 1 Gobiidae Ptereleotris evides M 0 0 1 0 0 0 0 0 Ptereleotris microlepis M 0 0 0 0 29 0 7 0 Acanthuridae Acanthurus nigrofuscus M 0 0 2 0 0 0 0 0 Prionurus maculatus M 2 3 0 14 0 18 0 0 Prionurus microlepidotus M 1 0 2 1 3 1 2 2 Siganidae Siganus fuscescens M 0 0 0 0 0 0 4 0 Scombridae Sarda australis M 0 16 0 0 0 0 0 0 Monacanthidae Eubalichthys bucephalus M 10 6 9 8 0 0 0 3] Eubalichthys mosaicus M 7 8 5 0 0 0 0 0 Meuschenia flavolineata M 6 7 6 6 0 0 7 0 Meuschenia freycineti M 1 3 2 2 0 1 2 0 Meuschenia trachylepis M 20 17 0 0 6 23 13 12 Penicipelta vittiger M 0 0 0 0 0 0 0 1] Scobinichthys granulatus M 0 0 0 0 0 2 2 1 Monacanthid sp. M 0 0 0 0 0 1 1 6 Ostraciidae Anoplocapros inermis M 3 0 0 0 15 16 0 1 Ostracion cubicus M 0 1 1 1 0 0 0 0 Tetraodontidae Tetractenos glaber M 0 0 0 0 5 51 0 0 Canthigaster callisterna M 0 0 2 3 0 0 0 0 Canthigaster valentini M 1 0 1 0 0 0 0 0 Diodontidae Dicotylichthys punctulatus M 3 2 1 0 0 1 1 0 CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 98

The SIMPER procedure ranked, in order of importance, those species that contributed most to the differences between the outfall location and the controls (Table 5.4). The species ranked in the top five are included and it is clear that the relatively small numbers of Trachurus novaezelandiae and large numbers of Acanthopagrus australis, Atypichthys strigatus and Scorpis lineolatus contributed to the differences between the outfall location and the control locations (Table 5.4). The relatively small numbers of Chromis hypsilepis and large numbers of Cheilodactylus fuscus at the outfall location also contributed to the differences between North Head and Dover Heights and The Gap, respectively

(Table 5.4).

Univariate analyses of fish assemblages and species

Variables examined by ANOVA (Table 5.5) fluctuated significantly among locations (Figure 5.4) and/or times (Figure 5.5). These variables can be differentiated into two groups: (1) those for which a significant increase in mean number occurred at the outfall relative to the controls, and (2) those for which a significant decrease occurred at the outfall relative to the controls. The first group includes species richness of cryptic fish and abundances of Pempheris multiradiata, Cheilodactylus fuscus,

Atypichthys strigatus and Acanthopagrus australis (Figures 5.4 and 5.5). The abundance of

Cheilodactylus fuscus consistently averaged 12 at the outfall compared to 7 and 3 at the controls,

indicative of an impact of 100 to 400 percent (Figure 5.4). The abundance of Pempheris multiradiata

was 300 percent more at the outfall compared to one control but similar to the other control (Figure

5.4). The abundance of Acanthopagrus australis averaged 50 at Times 1 and 2 at the outfall, compared

to less than 5 fish at the control locations (1000 percent increase). However, the abundance of A. australis declined over time, and abundance at one outfall location was similar to controls after 13

months (Figure 5.5).

The second group includes abundance of mobile fish, abundance of cryptic fish, richness of mobile

fish, Trachinops taeniatus, Parma unifasciata, Hypoplectrodes mccullochi and Trachurus novaezelandiae. The abundance of cryptic fish and Trachinops taeniatus indicated a similar pattern

with 100 percent declines at the outfalls compared to one control but similar to the other control

(Figure 5.4). No Parma unifasciata were recorded in the boulder habitat at the outfall locations (Fig. Table 5.3. R-statistics (Clarke 1993) from one-way ANOSIM and pairwise comparisons of mobile and cryptic fish at boulder and wall habitats (4 analyses). NN = North Head North outfall, NS = North Head

South outfall, G = Gap control, DH = Dover Height Control. Number of permutations = 5000. ns = not significant, R (* = significant at a<0.05), Pairwise comparisons (**= significant at a<0.016 - alphas adjusted for multiple comparisons).

Mobile fish Cryptic fish

Boulder Wall Boulder Wall

All locations R = 0.295 * R = 0.139 ns R = 0.263 * R = 0.274 *

Pairwise comparisons

NN vNS ns NN vNS ns NN vNS ns NN vNS ns

NN vG ** NN vG ns NN vG ** NN vG **

NN v DH ** NN v DH ns NN v DH ns NN v DH ns

NS vG ** NS vG ns NS vG ** NS vG **

NS v DH ns NS vDH ns NS vDH ns NS vDH ns

G v DH ** G v DH ns G v DH ns G v DH ns Figure mobile cryptic Heights. CRYPTIC MOBILE

•

5.2. fish fish o

MDS in in

• 0 BOULDER

boulder wall o

o plots

habitat.

□ •

habitat for 1 •

the

0

O □ □

□ □

abundance (b) -

North

mobile

stress stress • >

Head 3

fish of

= A =

mobile A North,

0.06

0.08 in

wall

• and

habitat, -

North cryptic

(c) Head A

O species

cryptic • 8 o

• South, A P a

WALL at

fish

□

boulder n □

A □ □

in -

The □ boulder

and

Gap,

wall

stress habitat, stress

A* □

habitats -

A Dover

A

=

= and

0.16 0.04 □ _

(a) (d)

Table 5.4. Relative abundances and cumulative percentage (Cum %) of the five mobile species that were most responsible for differences between fish assemblages in the boulder habitat at the outfall location at

North Head North and the two control locations at The Gap and Dover Heights (DH).

Outfall The Gap Cum %. DH Cum %

Trachurus novaezelandiae 180 4163 81.4 806 43.4

Chromis hypsilepis 9 - - 261 58.8

Acanthopagrus australis 168 1 85.5 17 68.3

Atypichthys strigatus 163 80 88.4 46 76.4

Scorpis lineolatus 118 35 90.8 86 82.6

Cheilodactylus fuscus 79 16 92.4 - - Figure 5.3. MDS plot for the abundance of mobile species at the boulder habitat over time. O - sewage outfall locations, □ - control locations. The size of the symbols increases over time from Times 1 to 5

(see Table 5.1). CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 99

4), although they did occur in the wall habitat (Table 5.2). The abundance of Hypoplectrodes mccullochi averaged one at the outfall locations compared to 6 and 9 at the control locations, indicative of a negative impact of up to 900 percent (Figure 5.5). The abundance of both mobile fish and

Trachurus novaezelandiae varied over time but was consistently large at one control location and small at one outfall location (Figure 5.5).

DISCUSSION

This discussion focuses on three issues. First, the differences between assemblages of fish and selected species at outfall and control locations. Second, a comparison with other studies on the effects of sewage on subtidal benthic and fish assemblages. Third, the implications of these results for future research and fisheries management are discussed.

Differences between outfall and control locations

Many authors have reported that the effect of outfalls on benthic fish were evident at highly impacted

locations for a few key species (Stephens et al. 1988, Gray et al. 1992, Grigg 1994, Otway et al,

1996a,b, Smith et al. 1998). I found a similar situation, a consistent outfall effect for the mobile fish

assemblages and several individual species in boulder habitat. The outfall effect appeared to be

consistent for a period of at least one year following shutdown. The outfall effect was explained by

relative differences in abundances of several common species of temperate reef fish and particularly

the decreased abundance of Trachurus novaezelandiae and increased abundances of Acanthopagrus

australis and Atypichthys strigatus. The species that occurred in large numbers at outfalls were

generally large, mobile ‘opportunistic’ species. A. strigatus were described as ‘opportunistic’ because

more fish were found in disturbed areas and the feeding rate was greater in disturbed areas (Glasby and

Kingsford 1994). The magnitude of the outfall effect also differed between species and ranged from

1000 percent increases of Acanthopagrus australis to 900 percent decreases of Hypoplectrodes

mccullochi. The temporal effect varied between species with abundance of Acanthopagrus australis

declining within about 10 weeks after the outfall was shut down, however Cheilodactylus fuscus were

found in consistently large numbers at the outfall locations over the 13 month period of study. Table 5.5. Summary of mean square (m.s) and F ratios from analyses of variance of fish in the boulder habitat, ns - not significant, * p<0.05, ** p<0.01.

Variate TIME (T) LOCATION (L) TXL Residual

m.s F m.s F m.s F

Abundance of mobile fish 0.34 2.47 2.38 17.4 0.42 3.08** 0.14

Abundance of cryptic fish 0.09 3.01* 0.20 7.16** 0.04 1.29ns 0.03

Richness of mobile fish 0.02 4.15 0.07 12.3 0.02 3.44** 0.01

Richness of cryptic fish 0.07 10.8** 0.02 2.96* 0.01 0.92ns 0.01

Cheilodactylus fuscus 0.51 4.74** 3.50 32.6** 0.14 1.35ns 0.11

Trachinops taeniatus 0.14 0.74ns 1.02 5.43** 0.19 1.00ns 0.19

Pempheris multiradiata 0.25 1.23ns 1.81 8.74** 0.19 0.95ns 0.21

Parma unifasciata 0.10 1.08ns 1.51 16.1** 0.14 1.48ns 0.09

Hypoplectrodes mccullochi 0.20 5.31 2.89 76.9 0.08 2.20* 0.03

Trachurus novaezelandiae 4.96 5.71 22.1 25.5 1.96 2.26* 0.87

Atypichthys strigatus 1.81 12.2 1.28 8.67 0.37 2.51** 0.15

Acanthopagrus australis 1.54 22.7 12.4 182 0.38 5.66** 0.07

Achoerodus viridis 0.25 . 6.45 3.80 98.2 0.12 3.32** 0.04 species Figure unifasciata,

Mean abundance (+SE) Mean richness (+SE) Mean abundance (+SE

5.4. richness 200 200 150 150 100 50 12 10

0 4 6 8 0 2 Mean

and - - 4877 (c)

(f) of f\f\| NN

abundance

cryptic

Cheilodactylus NS NS

fish,

from

(c)

Gap Gap

outfall fuscus. Trachinops

and DH DH

control

taeniatus,

locations:

(d)

Pempheris

(a)

total

multiradiata, abundance 3793

of

(e) cryptic

Parma

fish,

(b)

Mean abundance (+SE) Mean richness (+SE) Mean abundance (+SE) 1200 1000 1200 1000 800 600 400 200 600 800 200 400 Acanthopagrus species Figure for

-

-

- - - - - * - - - - t - sampling

5.5. richness

Mean periods Time Time

Time

australis of

abundance

mobile

(weeks) (weeks) (weeks)

(f)

fish, Achoerodus

from 15

-r (c)

outfall Trachurus

viridis

and

Time

control novaezelandiae,

and

(weeks)

(g)

locations

Hypoplectrodes 80

nr

(d) of

(a) Atypichthys

total

mccullochi.

abundance Time Time

strigatus,

(weeks) (weeks)

Refer

of

mobile (e) to

Table • -----

fish, a

— 5.1

- (b)

NHN NHS Dover

Hts CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 100

An unresolved question is why increases or decreases of several species occurred adjacent to the outfalls. Possible explanations include (1) food supply, and (2) avoidance of predators in the polluted and turbid water. Sewage contains nutrients which may be directly or indirectly utilised as food by fish and conversely sewage may also result in degraded conditions which limit or prevent feeding

(Stephens et al. 1988). Increased abundances of planktivorous species have been reported adjacent to sewage outfalls (Bell and Harmelin-Vivien 1982, Pastorok and Bilyard 1985). Large numbers of the carnivorous California scorpionfish Scorpaena outtata near a sewage outfall may be due to the presence of high numbers of the ridgeback prawn Sicyonia inoentis which were attracted to the nutrient-rich substrata (Love et al. 1987). Increased species richness of a trophic group of fish that fed predominantly on crustaceans, echinoderms, polychaetes and fish was attributed to increased • abundance of these food organisms at sewage locations (Otway et al. 1996b).

Polluted and turbid water may result in the outfall locations functioning as “marine protected areas” which result in increased abundances of exploited species and large carnivores (Buxton and Smale

1989). The explanation for this second hypothesis is that human exploiters such as fishers and spearfishers were unable to legally fish in the waters adjacent to the sewage outfalls for health reasons.

Natural predators (such as dolphins, cormorants) may also be unable to capture prey in polluted water with poor visibility. This explanation may account for the increased abundances of mobile, exploited species such as Acanthopagrus australis and Cheilodactylus fuscus but it does not explain the decreased abundances of cryptic species such as Hypoplectrodes mccullochi and Trachinops taeniatus at the outfall locations.

Comparison to other studies

The relationship between sewage disposal and the ecology of fish is specific to species and locations

(Tsai 1975, Pilanowski 1992, Kingsford and Gray 1996, Otway et al. 1996a, Smith et al. 1998). For example, large numbers of the carnivorous snapper Lutjanus kasmira were reported directly above a diffuser, and the relative number of herbivorous fishes increased adjacent to deepwater outfalls off

Hawaii (Russo 1982, 1989; Grigg 1994). Increases in the abundances of some species of demersal fish, including flathead (F. Platycephalidae) and decreases in the abundances of a range of species including CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 101 snapper Pagrus auratus, flounder (Pleuronectidae) and leatherjacket (Monacanthidae), were reported at deepwater ocean outfalls off Sydney (Otway et al. 1996a). Decreased abundance of Trachinops taeniatus, Trachurus novaezelandiae and increased abundance of Pempheris compressus were attributed to discharge of a small sewage outfall (Smith et al. 1998).

The survey of rocky reef fish adjacent to the North Head sewage outfall in the 1970s (Jones 1977) contained a very similar list of species to this study, although Jones (1977) recorded fewer species.

Several fish, including Labridae and Monacanthidae, were recorded by Jones (1977) but not observed in this study. Quantitative information was recorded by Jones (1977) for ten species, and of these abundances of Parupeneus signatus, Pempheris compressus, Parma microlepis, Chironemus marmoratus, Cheilodactylus fuscus, Achoerodus viridis appeared to be similar to the present study, while Heterodontus portjacksoni were more abundant and Pseudocaranx dentex and Chromis hypsilepis were less abundant compared to the present study.

Few studies have examined changes in fish assemblages or species following the cessation of sewage discharge. The cessation of disposal of sewage off California resulted in no difference in abundance of the total demersal finfish or the dominant species, little skate Raja erinacea and winter flounder

Pseudopleuronectes americanus (Pilanowski 1992). A number of studies have reported that areas that were very polluted and had no fish were recolonised by fish when the pollution source is removed or mitigated (Tsai 1975, GESAMP 1990, Arthington and Welcomme 1995).

Benthic invertebrate fauna of a polluted system may recover rapidly after pollutant loading is discontinued (Oviatt et al. 1984). A study of shallow subtidal invertebrate and algal assemblages at the

North Head outfall compared to control locations was done over a similar period to this study

(Chapman et al. 1995, Underwood and Chapman 1996). The study indicated that variability among locations differed according to the taxa being examined and the spatial scale of sampling and there were surprisingly few significant differences in either mean measures of variances at outfall compared to control locations (Chapman et al. 1995, Underwood and Chapman 1996). One possible effect was a greater percentage cover and density of Pyura stolonifera and other filter feeding ascidians at the CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 102 outfall (Chapman et al. 1995). Other studies have reported a link or cause/effect relationship between benthic and fish assemblages and the impacts of sewage (Otway et al. 1996a,b, Hall et al. 1997,

Roberts et al. 1998,.Smith et al. 1998).

Future research and fisheries management

Natural environmental factors result in large variations in fish assemblages (Lincoln Smith et al. 1991,

Holbrook et al. 1994, Schmitt and Osenberg 1996, Gray 1997). However, by using an appropriate experimental design it is possible to measure natural fluctuations, and assess the impacts of man-made perturbations such as sewage disposal. In this present study, the outfall and reference locations were examined only after an affect and because I did not have ‘before information’ any conclusions of probable outfall effects may be partially or completely due to unknown or ecological factors. Although underwater visual census is a widely used sampling technique because is provides a rapid, inexpensive, precise measure of fish assemblages (Lincoln Smith and Jones 1995) it is not suitable to survey fish adjacent to large, primary treatment sewage outfalls because of poor water visibility. Several other methods have been used to survey fish is such environments, such as Catch Per Unit Effort of standardised fish gear (e.g. traps, line or nets). These could have been used to survey fish before the outfall was shut down.

This study emphasises common problems with assessing impacts on assemblages and these include selection of appropriate control locations and study period (Bernstein and Zalinsky 1983, Choat 1995,

Schmitt and Osenberg 1996). The assemblages of fish at the control location at The Gap were significantly different to the control location at Dover Heights. Future studies should include more than two control locations. The study period of 13 months has limited potential to determine changes to fish assemblages at the outfall locations because long-lived or resident species may buffer against rapid change (Holbrook et al. 1994). This may be an explanation for why Cheilodactylus fuscus (restricted home range, longevity of over 20 years) had consistently larger abundances at the outfall in comparison to a migratory species such as Acanthopagrus australis whose abundance declined within about 10 weeks of shutdown of the outfall. A study period of at least three years would be recommended to detected changes due to the shutdown of shoreline sewage outfalls in rocky reef fish. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 103

The relative abundance of several commercially and recreationally important species, particularly

Acanthopagrus australis and Cheilodactylus fuscus, at outfall locations would appear to indicate that

sewage outfalls are a positive benefit for fish in NSW waters. However, there are a number of

physiological disadvantages for fish that occur at sewage outfalls, such as an increased incidence of

deformities, reduced reproduction and bioaccumulation of contaminants (Beder 1989, Scanes and

Philip 1995). These physiological effects in fish may also result in increased health risks for human

consumers of fish from such polluted locations (Leadbitter 1992, 1996).

REFERENCES

Arthington, A.H. and Welcomme, R.L (1995). The condition of large river systems in the world. In

Condition of the World’s Aquatic Habitats, ed. N.B. Armantrout, pp. 44-75. Proceedings of the

World’s Fisheries Congress, Theme 1. Science Publishers Inc., Lebanon, U.S.A.

Beder, S. (1989). Toxic Fish and Sewer Surfing. Allen & Unwin, Sydney, Australia.

Bell, J.D. and Harmelin-Vivien, M.L. (1982). Fish fauna of French Mediterranean Posidonia Oceania

seagrass meadows. 1. Assemblage Structure. Tethys 10, 337 - 347.

Bernstein, B.B and Zalinsky, J. (1983). An optimum sampling design and power tests for

environmental biologists. Journal of Environmental Management 16, 35-43.

Bignert, A., Olsson, M., Persson, W., Jensen, S., Zakrisson, S., Litzen, K., Eriksson, U., Haggberg, L.,

Alsberg, T. (1998). Temporal trends of organochlorines in Northern Europe, 1967-1995. Relation

to global fractionation, leakage from sediments and international measures. Environmental

Pollution 99, 177-198.

Bokn, T.L., Frithjof, E. and Walday, M. (1996). Improvement of the shallow water communities

following reductions of industrial outlets and sewage discharge in the Hvaler estuary, Norway.

Hydrobiologica 326/327, 297-304.

Buxton, C.D. and Smale, M.J. (1989). Abundance and distribution patterns of three temperate marine

reef fish, Teleostei: Sparidae, in exploited and unexploited areas of the southern cape coast.

Journal of Applied Ecology 26, 441 -451. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 104

CEPA (1991). A study of chemical contamination of marine fish from southern California. California

Environmental Protection Agency. 12 pp.

Chang, S. (1993). Analysis of fishery resources: potential risk from sewage sludge dumping at the

deepwater dumpsite off New Jersey. Fishery Bulletin 91, 594-610.

Chapman, M.G., Underwood, A.J. and Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

Journal Experimental Marine Biology and Ecology 189, 103-122.

Choat, J.H. (1995). Long-term vs short-term studies. Long-term monitoring programs in sub-tidal

environments; a problem with fishes. Abstract from Third International Temperate Reef

Symposium, University of Sydney, February.

EPA (1993). Deformities and associated sublethal effects in fish exposed to sewage-borne

contaminants - literature review. Environment Protection Authority, Sydney, Australia, Report

93/72, August.

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environmental Monitoring Assessment 14, 197-210

Fagan, P., Miskiewicz, A.G. and Tate, P.M. (1992). An approach to monitoring sewage outfalls. A case

study on the Sydney Deepwater Sewage Outfalls. Marine Pollution Bulletin 25, 172-180.

GESAMP (1990). The state of the marine environment. IMO / FAO / Unesco / WHO / IAEA / UN /

UNEP Joint Group of Experts on the Scientific Aspects of Marine Pollution. UNEP Regional

Seas Reports and Studies No. 115. UNEP.

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 reef. Environmental

Biology of Fishes 49, 153-173.

Glasby, T.M. (1997). Analysing data from post-impact studies using asymmetrical analysis of

variance: a case study of epibiota on marinas. Australian Journal of Ecology 22, 448-459.

Glasby, T.M and Kingsford, M.J. (1994). Atypichthys strigatus (Pisces: Scorpididae): An opportunistic

planktivore that responds to benthic disturbances and cleans other fishes. Australian Journal of

Ecology 19, 385-394. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 105

Gray, J.S. (1989). Effects of environmental stress on species rich assemblages. Biological Journal of

the Linnean Society 37, 19-32.

Gray, C.A. (1997). Field assessment of numerical impacts of coastal sewage disposal on fish larvae

relative to natural variability. Environmental Biology of Fishes 50, 415-434.

Gray, C.A., Otway, N.M., Laurenson, F.A., Miskiewicz, A.D. and Pethebridge, R. (1992). Distribution

and abundance of marine fish larvae in relation to effluent plumes from sewage outfall and depth

of water. Marine Biology 113, 549-559.

Green, R.H. (1979). Sampling Design and Statistical Methods for Environmental Biologists. Wiley,

New York.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Marine Ecology Progress Series 103, 25-34.

Hall, J.A., Frid, C.L.J. and Gill, M.E. (1997). The response of estuarine fish and benthos to an

increasing discharge of sewage effluent. Marine Pollution Bulletin 34, 527-535.

Hardy, F.G., Evans, S.M. and Tremayne, M.A. (1993). Long-term changes in the marine macroalgae of

three polluted estuaries in north-east England. Journal Experimental Marine Biology and Ecology

172,81-92.

Harmelin-Vivien, M. (1992). Impact des activites humaines sur les peuplements ichtyologues des recifs

coralliens de polynesie francaise. Cybium 164, 279-289.

Holbrook, S.J., Kingsford, M.J., Schmitt, R.J., and Stephens, J.S. (1994). Spatial and temporal patterns

in assemblages of temperate reef fish. American Zoologist 34, 463-475.

Jones, A.R. (1977). An ecological survey of nearshore waters east of Sydney, NSW 1973-75. The

Australian Museum, Sydney.

Kingsford, M.J. and Gray, C.A. (1996). Influence of pollutants and oceanography on abundance and

deformities of wild fish larvae. In Detecting Ecological Impacts: Concepts and Applications in

Coastal Habitats, eds R.J. Schmitt and C.W. Osenberg, pp. 235-255. Academic Press, N.Y.

Koop, K. and Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Marine

Pollution Bulletin 33, 121-123.

Krogh, M. and Scanes, P.R. (1996). Organochlorine compound and trace metal contaminants in fish

near Sydney’s ocean outfalls. Marine Pollution Bulletin 33, 213-225. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 106

Leadbitter, D. (1992). Management of fishing activities in contaminated areas. Australian Fisheries.

September, pp. 30-33.

Leadbitter, D. (1996). Sewage and the community - one view of Sydney’s deepwater outfalls, the

environmental monitoring programme and the future of sewage disposal. Marine Pollution

Bulletin 33, 269-272.

Lincoln Smith, M.P. (1985). The development and application of visual survey procedures for fish

communities on shallow rocky reefs. MSc thesis, University of Sydney, Australia.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate rocky

reef fish assemblages. Marine Ecology Progress Series 43, 223-231.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Environmental Biology of Fishes 26, 29-37.

Lincoln Smith, M.P., Bell. J.D. and Hair, C.A. (1991). Spatial variation in abundance of recently

settled rocky reef fish in southeastern Australia: implications for detecting change. Marine

Ecology Progress Series 77, 95-103.

Lincoln Smith, M.P. and Jones, G.P. (1995). Fishes of shallow coastal habitats. In Coastal Marine

Ecology of Temperate Australia, eds. A.J. Underwood and M.G Chapman, pp. 240-253. UNSW

Press, Sydney, Australia.

Love, M.S., Axell, B., Morris, P., Collins, R. and Brooks, A. (1987). Life history and fishery of the

Californian scorpionfish, Scorpaena guttata, within the southern Californian Bight. Fishery

Bulletin 85, 99-116.

Lye, C.M., Frid, C.L.J., Gill, M.E. and McCormick, D. (1997). Abnormalities in the reproductive

health of flounder Platichthys flesus exposed to effluent from a sewage treatment works. Marine

Pollution Bulletin 34, 34-41.

McCormick, M.I. and Choat, J.H. (1987). Estimating total abundance of a large temperate-reef fish

using visual strip-transects. Marine Biology 96, 469-478.

Otway, N.M, Gray, C.A., Craig, J.R., McVea T.A. and Ling, J.E. (1996a). Assessing the impacts of

deepwater sewage outfalls on spatially and temporally variable marine communities. Marine

Environmental Research 41, 45-71. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 107

Otway, N.M., Sullings, D.J. and Lenehan, N.W. (1996b). Trophically-based assessment of the impacts

of deepwater sewage disposal on a demersal fish community. Environmental Biology of Fishes

46, 167-183.

Oviatt, C.A., Pilson, M.E.Q., Nixon, S.W., Frithsen, J.B., Rudnick, D.T., Kelly, J.R., Grassle, J.F. and

Grassle, J.P. (1984). Recovery of a polluted estuarine system: a mesocosm experiment. Marine

Ecology Progress Series 16, 203 - 217.

Pastorok, R. and Bilyard, G.R. (1985). Effects of sewage pollution on coral-reef communities. Marine

Ecology Progress Series 21, 175-189.

Pilanowski, R.A. (1992). The effects of ocean disposal of sewage sludge on the relative abundance of

benthic megafauna. Chemistry and Ecology 6, 199-212.

Philip, N.A. and Pritchard, T.R. (1996). Australia’s first deepwater sewage outfalls: design

considerations and environmental performance monitoring. Marine Pollution Bulletin 33, 140-

146.

Roberts, D.E., Smith, A., Ajani, P. and Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACI’ approach.

Marine Ecology Progress Series 163, 213-224.

Russo, A.R. (1982). Temporal changes in fish community structure near a sewage ocean outfall,

Mokapu, Oahu, Hawaii. Marine Environmental Research 6, 83-98.

Russo, A.R. (1989). Fish community structure on a deep ocean outfall, Barber’s Point, Oahu, Hawaii.

Int. Rev. ges. Hydrobiol. 74, 499-506.

Sale, P. (1994). Ecology of Coral Reef Fish. Academic Press, New York.

Scanes, P.R. (1996). ‘Oyster Watch’: monitoring trace metal and organochlorine concentrations in

Sydney’s coastal waters. Marine Pollution Bulletin 33, 226-238.

Scanes, P.R. and Philip, N. (1995). Environmental impact of deepwater discharge of sewage off

Sydney, NSW, Australia. Marine Pollution Bulletin 31, 343-346.

Scanes, P.R., Scanes, K. and Otway, N.M. (1995). Environmental problems due to disposal of wastes.

In Coastal Marine Ecology of Temperate Australia, eds. A.J. Underwood and M.G. Chapman, pp.

297-310. UNSW Press, Sydney, Australia. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 108

Schmitt, R.J. and Osenberg, C.W. (1996). Detecting ecological impacts, concepts and applications in

coastal habitats. Academic Press, New York.

Smith, A.K., Ajani, P.A. and Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Marine Environmental Research 47, 1-20.

Stephens, J.S, Jr., Hose, J.E. and Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In Marine Organisms as Indicators, eds. D.F. Soule and G.S.

Kleppel, pp. 91-105. Springer-Verlag, NY.

Thresher, R.E. and Gunn, J.S. (1986). Comparative analysis of visual census techniques for highly

mobile, reef associated piscivores, Carangidae. Environmental Biology of Fishes 17(2), 93-116.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA.

Underwood, A.J. (1991). Beyond BACI: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Marine and Freshwater Research 42, 569-

587.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. Journal of Experimental Marine Biology and Ecology 161, 145-178.

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Australian Journal of Ecology 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecological Applications 4, 3-15.

Underwood, A.J. (1997). Experiments in Ecology. Their Logical Design and Interpretation using

Analysis of Variance. Cambridge University Press. 504 pp.

Underwood, A.J. and Chapman, M.G. (1996). Subtidal assemblages on rocky reefs at a cliff-face

sewage outfall (North Head, Sydney, Australia): What happened when the outfall was turned off?

Marine Pollution Bulletin 33, 293-302.

Underwood, A.J., Kingsford, M.J. and Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Australian Journal of Ecology 6, 231-249. CHAPTER 5 EFFECTS OF SEWAGE ON FISH IN SYDNEY REGION 109

Underwood, A.J. and Peterson, C.H. (1988). Towards an ecological framework for investigating

pollution. Marine Ecology Progress Series 46, 227-234.

Warwick, R.M. (1993). Environmental impact studies on marine communities: Pragmatical

considerations. Australian Journal of Ecology 181, 63-80.

Waring, C.P., R.M. Stagg, K. Fretwell, H.A. McLay and Costello, M.J. (1996). The impact of sewage

sludge exposure on the reproduction of the sand goby, Pomatoschistus minutus. Environmental

Pollution 93(1), 17-25. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 110

CHAPTER 6

IMPACTS OF SEWAGE EFFLUENT DISCHARGE ON

ABUNDANCE AND BIOMASS OF TEMPERATE REEF FISH

ASSEMBLAGES

ABSTRACT

Underwater visual surveys of fish assemblages at three regions on the central coast of New South

Wales found lower abundance, but greater estimated biomass, of fish at locations where sewage was

discharged compared to control locations. Multivariate analyses (MDS and ANOSIM) indicated that

fish assemblages were significantly different between outfall and control locations at each of the three

regions studied. Large abundances of small, schooling species such as Trachurus novaezelandiae

(Carangidae), Trachinops taeniatus (Plesiopidae) and Atypichthys strigatus (Scorpididae) at control

locations were mainly responsible for these differences. Alternatively, the greater biomass of large

species such as Achoerodus viridis (Labridae), Acanthopagrus australis (Sparidae), Cheilodactylus fuscus (Cheilodactylidae), Seriola lalandi (Carangidae) and Argyrosomus japonicus (Sciaenidae) at

outfall locations accounted for most differences. Two species, T. novaezelandiae and A. australis may

be useful indicators of sewage pollution as they ranked highly for discriminating fish assemblages from

outfall and control locations using both abundance and biomass.

If a wide spectrum of size categories is represented, the choice of abundance versus biomass data as the

most suitable measure for environmental assessment of fish assemblages provide different

interpretation of which species are most important. Implications of these findings are discussed in the

context of future management options favouring particular species of fish or fishing interest groups. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 111

INTRODUCTION

Changes in the structure of marine assemblages are widely used for the detection and monitoring of anthropogenic impacts. Species abundance data are by far the most commonly used in environmental impact studies at the assemblage level (Lincoln Smith 1988; 1989; Jones 1988; Bianchi and Haisaeter

1992; Warwick 1993; Holbrook et al. 1994; Underwood 1994). Biomass is potentially a more useful measure, as functional importance may vary among species of different sizes (Warwick 1993; Schmitt and Osenberg 1996). Usually only abundance or biomass of the biota is measured to represent community structure. Other community-level measures (species richness, diversity) generally have questionable theoretical justification, have no demonstrable causal links to the impact, and depend on the taxonomic expertise available (Keough and Quinn 1991; Wright et al. 1995). Many studies are a compromise between the scientific ideal and political, financial and logistical constraints (Keough and

Quinn 1991; Warwick 1993).

Disposal of sewage has been identified as the single most important pressure on coastal ecosystems

(Boates and Russell 1992; Earle 1995; Baird 1996; Koop and Hutchings 1996) and often occurs in the vicinity of temperate reefs. In this case, the use of fish surveys for environmental impact assessment has several advantages: fish are relatively easy to identify, they are of particular interest to commercial and recreational fishers, and they may indicate environmental degradation (Warwick 1993; Harris

1995; Barrett 1995; Lincoln Smith and Jones 1995; Smith et al. 1998).

Congregations of fish have been reported around sewage outfalls (Tsai 1975; Pastorok and Bilyard

1985; Russo 1989; Grigg 1994; Growns et al. 1998). Further, large individuals of some fish species are frequently found adjacent to sewage outfalls (Tsai 1975; 1991; Khalaf et al. 1986; Lincoln Smith and

Mann 1989; Grigg 1994; Nowak 1996). The presence and abundance of large fish may be related to

high nutrient levels providing a direct or indirect source of food (Tsai 1975; Stephens et al. 1988;

Lincoln Smith and Mann 1989; Hall et. al. 1997). The indirect source of food occurs where sewage attracts invertebrates and small baitfish (Bell and Harmelin-Vivien 1982) which may be preyed upon

by carnivorous fish (Love et al. 1987; Tsai 1991). Factors such as the depth of water (McCormick and

Choat 1987; Jones 1988), location (Gillanders 1997), food supply (Cole 1994) and intensity of fishing (a) Hunter region (3 surveys)

QTomaree Head

O Point Stephens

• Boulder Bay

(b) Sydney region (3 surveys)

New South Wales North Head

Q The Gap QDover Heights

(c) Illawarra region ( 1 survey)

A Bcllainbi

£ A Hinders Island

’A Port Kembla

.A Shell Harbour

A Atchiusous Rock

Figure 6.1. Study locations for surveys of reef Fish in the (a) Hunter, (b) Sydney and (c) Illawarra regions.

The sewage outfall locations are represented by Filled symbols. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 112

(Bell 1983; Roberts and Polunin 1991; Watson and Ormond 1994; Jennings et al. 1995) may also affect the size of fish occurring in an area, regardless of the presence of a sewage outfall.

This study examined the relative utility of abundance versus biomass measures for assessing the effects of sewage outfalls on rocky reef fish assemblages.

METHODS

Fish surveys

Temperate reef fish assemblages were surveyed at three regions off the central coast of NSW

(Figure 6.1) during summer (Dec-Feb) between 1990 and 1995 by the same observer using

SCUBA. In the Hunter region, three locations were surveyed at three times in the period 1993-

95 (Smith et al. 1998). Control locations were at Tomaree Head and Point Stephens, and the

sewage outfall was located at Boulder Bay (Figure 6.1a), which discharged 4.3 ML/day of

secondary treated sewage (MHL 1997).

In the Sydney region, four locations were surveyed at three times in the period 1991- 92.

Controls locations were at the Gap and Dover Heights, and there were two sewage outfall sites

at North Head, one each immediately north and south of the outfall (Figure 6.1b). This outfall

discharged 293 ML/day of primary treated sewage until it was shutdown in December 1991.

In the lllawarra region, six locations were surveyed at one time in 1993 (TEL 1994). Control

locations were at Bulli, Flinders Island, Shellharbour and Atchinsons Rock (Figure 6.1c).

Outfalls were located at Bellambi and Port Kembla, discharging 20 ML/day and 16.8 ML/day of

primary treated sewage, respectively.

Fish were generally surveyed in the boulder or barrens habitat (Underwood et al. 1991) at

depths of 12 to 18 m, with the exception of Bellambi outfall and controls at Bulli and

Shellharbour, which were surveyed at depths of 2 to 5m. Table 6.1. Estimated average biomass (g) of individal fish for the 107 species surveyed during summer at three regions on the central coast of NSW in 1991-95.

Category Species 1 Aspasmogaster costatus, Aspidontus taeniatus, Labroides dimidiatus, Norfolkia clarkei, Petroscirtes fallax, Pictiblennius tasmaniensis, Plagiotremis rhinorhynchos, Plagiotremis tapeinosoma, Ptereleotris evides, Pomacentrus coelestis, Trachinops taeniatus 10 Atypichthys strigatus, Cristaceps australis, Eupetrichthys angustipes, Hypoplectrodes mccullochi, Ostracion cubicus 25 Apogon aureus, Austrolabrus maculatus, Canthigaster callisterna, Canthigaster valent ini, Chaetodon guntheri, Gerres subfasciatus, Hypoplectrodes nigrorubrum, Mecaenichthys immaculatus, Pempheris affinis, Pictilabrus laticlavius, Stegastes gascoynei, Synodus variegatus, Trachichthys australis, Trachurus novaezelandiae, Unid wrasse 1, Unid wrasse 2, Unid wrasse 3 50 Chelmonops truncatus, Chromis hypsilepis, Parma microlepis, Parma unifasciata, Pempheris compressus, Pempheris multiradiata, Penicipelta vittiger, Schuettea scalaripinnis, Scorpis lineolatus, Thalassoma lunare, Trachypoma macracanthus 100 Coris picta, Enoplosus armatus, Heniochus species, Microcanthus strigatus, Odax acroptilus, Ophthalmolepis lineolatus, Pseudolabrus fusciola, Pseudolabrus guntheri 250 Anoplocapros inermis, Cheilodactylus vestitus, Chironemus marmoratus, Cnidoglandis macrocephalus, Coris sandageri, Dinolestes lewini, Enchelycore ramosa, Gymnothorax prasinus, Lotella rhacinus, Meuschenia flavolineata, Meuschenia freycineti, Meuschenia trachylepis, Monacanthus chinensis, Notolabrus gymnogenis, Paraplesiops bleekeri, Parupeneus signatus, Pseudocar anx dentex, Rhabdosargus sarba, Scobinichthys granulatus, Scorpaena cardinal is, Siganus fuscescens, Upeneichthys lineatus 500 Acanthistus ocellatus, Acanthopagrus australis, Acanthurus nigrofuscus, Aulopus purpurissatus, Cheilodactylus fuscus, Crinodus lophodon, Cybiosarda elegans, Epinephelus undulatostriatus, Eubalichthys bucephalus, Girella tricuspidata, Odax cyanomelas, Pomatomus saltatrix, Prionurus maculatus, Prionurus microlepidotus, Sarda australis 1000 Brachaelurus waddi, Eubalichthys mosaicus, Girella elevata, Urolophus testaceus 2500 Achoerodus viridis, Dicotolichthys punctulatus, Kyphosus sydneyanus, Latridopsis forsteri, Lutjanus argentimaculatus, Nemadactylus douglasii, Trygonorhina fasciata 5000 Argyrosomus japonicus, Heterodontus portjacksoni, Orectolobus species, Seriola lalandi 10000 Dasyatis brevicaudatus, Dasyatis thetidis CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 113

Within the Hunter and Illawarra regions, all fish were sampled along four, 60 metre-long transects at each time. Mobile fish in the Sydney region were surveyed by timed interval counts (which covered a similar area to the transects), and cryptic fish were sampled along four 60 metre-long transects (TEL

1994). Each transect was laid from a boat orientated parallel to the shoreline, and located within 100 m

of the outfall. Fish were sampled during periods of low swell (less than 1 m) and when water visibility exceeded 5 m. Fish were counted one metre either side of the transect line (total area 120 m^) (Lincoln

Smith 1988; 1989). For the multivariate analyses data from the four individual transects at each time

were summed to obtain a measure of fish abundance at each location (total area 480 m^),.

Estimation of biomass

All species recorded were sorted into biomass categories in accordance with the average biomass of

individuals reported in the literature (Henry 1984; Lincoln Smith et al. 1989; Steffe et al. 1996) or

estimated from personal observations (Table 6.1). There is a large body of scientific literature

concerning fish biomass (e.g. Jennings et al. 1995; Jennings and Polunin 1996; Samoilys 1997; Taylor

and Wiilis 1998), however, the procedures for calculation of biomass require estimation of length,

weight or destructive sampling of fish, which were not measured in this study. Here, average biomass

was multiplied by observed fish abundance to obtain an estimate of total biomass for each species.

Sensitivity analyses were undertaken on the biomass data to determine whether changes in the biomass

categories of several of the most abundant species would result in changes to significant differences

between outfall and control locations and the contribution of individual species. The following changes

were made to decrease the contribution of several abundant species: blue groper Achoerodus viridis

(Steindachner) from 2500 g to lOOOg, yellowfin bream Acanthopagrus australis (Owen) from 500 g to

250 g and yellowtail Trachurus novaezelandiae (Richardson) from 25 g to 10 g, and multivariate

analyses were repeated and the rank importance compared with the original analyses. Figure 6.2. Mean (+ SE) for (a) species abundance, (b) biomass and (c) average weight, of rocky reef fish assemblages at sewage outfall (dark) and control locations (striped) from three regions (H = Hunter, S -

Sydney, I = Illawarra, o = outfall, c = control) (all samples combined). CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 114

Multivariate analyses

Non-metric multi-dimensional scaling (MDS) was performed on data using the Bray-Curtis similarity measure (Clarke 1993). Separate analyses were performed for abundance and biomass data and on untransformed and double square root transformed data. The untransformed data give greater emphasis to abundant species and the double square root transformed data give greater emphasis to rare species.

Significant differences in fish assemblages between fish from outfall and control locations and regions were tested for by two-way crossed analysis of similarity (ANOSIM). Similarity of percentages

(SIMPER) was used to identify the contribution of individual species to differences between outfall and control locations. Multivariate analyses were performed using the PRIMER package (Plymouth

Marine Laboratories, UK).

RESULTS

A total of 107 species, 42 905 individuals and an estimated total biomass of 4 965 kg were recorded.

Trachurus novaezelcmdiae was numerically dominant (49%) but contributed only 10% to total biomass. Achoerodus viridis contributed the greatest biomass (19%) but comprised <1% of total abundance. The assemblages of fish from each sample (n=27, area = 480 m2) ranged from 15 to 42 species (mean 31 + 1.5 SE), 144 to 7314 (mean 1589 + 338) individuals and 19 to 708 kg total biomass

(mean 184 + 37).

The mean abundance of fish from the three regions indicated a pattern of greater abundance at control locations than outfalls (Figure 6.2a). The mean biomass of fish was approximately 1 kg m2 at the

Sydney outfall (Figure 6.2b), which was greater than other regions, but there was no consistent pattern in biomass between outfall and control locations. The average weight of fish was greater at the outfalls compared to the control locations in the Sydney and lllawarra regions, but there was no difference in the Hunter region (Figure 6.2c).

Abundance of fish

MDS ordinations using species abundance showed that fish assemblages differed between outfall and control locations for all regions (Figure 6.3). Stress values ranged from 0.12 to 0.15 giving usable, but CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 115 not good representations (Clarke 1993) of the original data in two dimensions. ANOSIM tests showed that the differences between outfall and control locations were statistically significant at all regions

(Table 6.2).

Two interpretations of the data were possible because I used data that were not transformed and data that had been double square root transformed. For the first case, species that contributed most to distinguishing outfall and control locations over all regions are listed in Table 6.3. Abundances of small, schooling species such as Trachurus novaezelandiae, eastern hula Trachinops taeniatus

(Gunther) and mado Atypichthys strigatus (Gunther) at control locations were mainly responsible for differences between assemblages at outfall and control locations at all regions (Table 6.3). Fish species that were more abundant at outfall locations included Acanthopagrus australis, sweep Scorpis lineolatus (Kner) and red morwong Cheilodactylus fuscus (Castelnau). Few one-spot chromis Chromis hypsilepis (Gunther) and eastern pomfred Schuettea scalaripinnis (Steindachner) were recorded at outfall compared to control locations.

The results of the transformed and untransformed data were similar, with seven out of the ten species important in terms of abundance in distinguishing outfall and control locations in both analyses (Table

6.4). The species that were included as important in the transformed results were white ear Parma unifasciata (Steindachner), Dinolestes lewini and tarwhine Rhabdosargus sarba (Forsskal), and the species that were excluded were Atypichthys strigatus, Cheilodactylus fuscus and Schuettea scalaripinnis.

The sensitivity analyses using three species indicated no changes in the species contributing to the ten most responsible for distinguishing outfall and control locations, however, the ranks of some individual species changed. For example, Cheilodactylus fuscus replaced Achoerodus viridis as the most important species, and Trachurus novaezelandiae changed from a rank of 3 to 6 (Table 6.3). CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 116

Biomass of fish

MDS ordinations using biomass data showed that fish assemblages differed between outfall and control locations for all regions (Figure 6.3). Stress values ranged from 0.11 to 0. 16 giving usable, but not good representations (Clarke 1993) of the original data in two dimensions. ANOS1M tests showed that the differences between outfall and control locations were statistically significant in all regions (Table

6.2).

The ten species that contributed the most to distinguishing outfall and control locations over all regions are listed in Tables 6.3 and 6.4. The results from the untransformed data (Table 6.3) indicate that large species such as Achoerodus viridis, Acanthopagrus australis, Cheilodactylus fuscus, Seriola lalandi and mulloway Argyrosomus japonicus (Lacepede) were mainly responsible for differences between fish assemblages at outfall and control locations. A greater biomass of A. viridis, A. australis, C. fuscus occurred at the outfall locations, and A. japonicus and tailor Pomatomus saltatrix (Linnaeus) were recorded only at outfalls. A greater biomass of T. novaezelandiae and S. lalandi occurred at control

locations.

The results of the transformed and untransformed data were similar. Once again, seven out of the ten

species were important in distinguishing outfall and control locations (Table 6.4). The species that

were included as important in the transformed results were Chromis hypsilepis, Rhabdosargus sarba

and rock blackfish Girella elevata (Macleay) and the species that were excluded were Cheilodactylus fuscus, Agyrosomus japonicus and Pomatomus saltatrix.

The sensitivity analyses indicated that there were no changes in the species contributing to the ten most

responsible for distinguishing outfall and control locations, however, the ranks of some individual

species changed slightly by one or two positions (Table 6.4). Table 6.2. R-statistics (Clarke 1993) from two-way crossed ANOSIM for comparisons of outfall and control locations and three regions for untransformed and double square root transformed data at abundance and biomass levels. ***P <0.001.

Double square root

Untransformed transformed

Abundance

Outfalls vs controls 0.567 *** 0.664 ***

Regions 0.920 *** 0.824 ***

Biomass

Outfalls vs controls 0.686 *** 0.586 ***

Regions 0.727 *** 0.789 *** Untransformed Double square root

Stress = 0.12 Stress = 0.15

□ □

O O

Stress = 0.11 Stress = 0.16

Figure 6.3. MDS ordinations of rocky reef fish assemblages at sewage outfall and control locations using untransformed and double square root transformed data at abundance and estimated biomass measures. The

Hunter, Sydney and Illawarra regions are represented by circles, squares and triangles, respectively. The sewage outfall locations are represented by filled symbols. H 2 le 6.3. Summary of results of SIMPER analyses for abundance and biomass data (untransformed data). The species contributing the mosi > GD 3 3 <5 o 3 bfi < o c c o c < < m £ * „ o E

.2 5 U CQ m X3 ~a u « < J x i cd CO CO cd CO C/5 o E E cd CO CO o E o E c E o c 3

£ £ £ .2 00 SP CD C/3 CL cn^ON«/brbvo

oo -2 t" m m i-i — (N iX * •S I# 5* d cn «/n '

o oo 3 vo cn (N —

.1 vo ^ -2 *q § 03

rn cn •3" ON ■3- ■3" >/n o o (N (N

r-~ O *S nf wn rn vo »o O ^ •a o o ST 3 £ § o s n

■3 co cn cn vn ■|,~2 o 1

vq O t"- (N oo VO V) vo O (N 0\ O ■8|! OO (N oo '/•> ■ t-~ ■*-n — o a 2 » a § n 1

CN ^ oo m O •3- vn U 's £ ^ p. n

co -2 < C3

*>: oo oo rn ^ cn (N (N r-. o — CL, II o 5 & < '

oo O d ub od ON o — 2 — v

Tl o t" 00 t >2 I

O >0 O 'cf cn .

n CN

VO OO -3- CN

(N Ch r-~ vo h CN *0 3 "X3 CJ r NO t'~- O (N 0 a R L. n <

CO n ;s 3" ^ "

— o li I o£ O & § R i

1.54 1.40 9 1.9 71.6

on •a 2 3 In E s

9.09 0.00 10 1.7 73.3 o 3 f c

>

60 jjs 3 e r- Tt > 3 C/5 .CD £

u ■ct (N Jc 3 £ (N •—< >—1 •—1 (N

5 X) 3 3 do £ xs jg .3 C/5 E <2 o 0/ O NO DO do (N co rf r? m +3 3 .2 3 3; t"; d *0 do Tt -3 cq O no Tj- T3 Q. d cd vO d (N —' co C/5 C/l CCS 'I cd VOOCCTOMIONOCN 6 E 'OvtfNOMctOOMcovt O .3 C QO •••••••• 0J o . . co no oo —■ co n© oo o IS 60 Q. u £ On ’—1 ’—1 -—1 P-S (N

T3 o

•§* § J '—1 DO NO — DO I •f U-i 03 o o +2 3 1 <0 ■is s: c5 o 5 - E CD *§ I s •2 .5 o ^ 'C Id Q 3 IN& o s § > DO •s g, o 5 C/5 ! | I a -5 -3 S.| s: Jj ■ct ■5 2 ^3 CD £ 3 ^ NO C/5 •a & & 1 60 sli k ~s: -« § _CD CCS .2 -s: -s: | E o o c 'I ^ O CD 3 u 3. pS: CD CD ^3 -b H 3 DO £ £ 3 O Q, O, 3 S L5 O CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 117

DISCUSSION

Abundance verses biomass measures

The question of whether measures of the abundance or biomass should be used to describe fish communities is rarely addressed (Bianchi and Hnisaster 1992). Multivariate analyses of abundance and biomass data in the present study gave remarkably similar results, regardless of the data transformations and despite the fact that different species were responsible for discriminating outfalls and controls locations. Thus measures of abundance and biomass were equally useful for examining differences in temperate reef fish assemblages between outfall and control locations.

Trachurus novaezelandiae were the numerically dominant species and Achoerodus viridis contributed the greatest biomass to the fish assemblages surveyed. Research on temperate rocky reef fish assemblages in New Zealand has shown a numerical dominance of open-water feeders such as bullseye

Pempheris adspersus (Griffin) and biomass dominated by carnivores in general, and individually by the herbivore, luderick Girel/a tricuspidata (Macleay) (Jones 1988). Three species of fish from the

Pempherididae were recorded in this study, and the abundance of two of these species contributed to distinguishing outfall and control locations over all regions. G. tricuspidata was commonly observed during this study, but was not important for establishing differences between outfall and control locations.

Indicators of sewage pollution

Two different types of species are generally used as indicators of pollution (Keough and Quinn 1991).

“Positive indicators” are opportunistic species that may occur in large numbers in polluted areas, either because environmental conditions become favourable for these species and/or detrimental to other species or because superior competitors are removed. Examples of such species are green algae Ulva spp. (Fairweather 1990), the polychaete worm Capitella capitata (Schmitt and Osenberg 1996) and carp Cyprinus carpio (Tsai 1975). “Negative indicators” are species that are common in relatively unpolluted areas and decline in abundance in response to pollution. Examples are the seagrass

Posidonia spp. (Chisholm et al. 1997) and perhaps Chaetodontidae (Hourigan et al. 1988). The study indicates that small, schooling species such as Trachurus novaezelandiae, Trachinops taeniatus, and CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 118

Chromis hypsilepis are negative indicators of sewage pollution as they were more abundant at outfall

than control locations. This may be due to the generally planktivorous feeding behaviour of these

species, or their small size and vulnerability to the larger number of predators at the outfall locations.

Biomass data indicate that large species such as Achoerodus viridis, and perhaps Argyrosomus japonicus and Pomatomus saltatrix, are positive indicators of sewage pollution, with the latter two

species only recorded at outfalls, while yellowtail kingfish Seriola lalandi (Valenciennes), another

large species was a negative indicator. The measures of abundance and biomass both distinguish

polluted and unpolluted locations but give a different interpretation of which species are important in

establishing these differences. Rather than highlighting the numerical importance of Trachurus

novaezelandiae or the biomass importance of Achoerodus viridis, it may be wise to select indicator

species which have been clearly identified as important from analyses of both biomass and abundance.

Two species, Trachurus novaezelandiae and Acanthopagrus australis were important in distinguishing

fish communities between outfall and control locations using either abundance or biomass measures

and untransformed and transformed data. These species support valuable commercial and recreational

fisheries and these may be useful indicator species. Cheilodactylus fuscus was important for both

abundance or biomass measures but only when untransformed data were used. This species is regarded

as a suitable bioindicator of heavy metal and organochlorine pollution in NSW coastal waters (Beder

1989; Lincoln Smith and Mann 1989; Andrijanic 1992; Leadbitter 1992; Gibbs and Miskiewicz 1995;

Nowak 1996).

Ecologists generally believe that some species are more “important” that others in ecological

communities (Schmitt and Osenberg 1996). Underwood and Peterson (1988) argue that indicator

species should be chosen for their direct ecological importance, either as “keystone” species in

organising the community, or because they form an important link in food webs leading to

economically important consumers. Large carnivorous species such as Argyrosomus japonicus ,

Seriola lalandi and Pomatomus saltatrix or small prey species such as Trachurus novaezelandiae

which are preyed upon by these large carnivores, potentially meet some of these criteria for

ecologically important indicator species. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 119

Comparison with other studies

Many studies of the environmental impact of sewage on aquatic biota compare one impact location with several control locations (e.g. Smith 1994; Chapman et al 1995; Underwood and Chapman 1996).

The present study has compared fish assemblages from four outfalls and eight control locations to enable a broad-scale analysis of environmental impact. Although the samples from the three regions

(Hunter, Sydney and Illawarra) were collected in different years between 1991 and 1995, all samples were taken in summer months. No pre-impact data are presented in this paper (but see Roberts et al.

1998; Smith et al. 1998).

Abundance/biomass comparison (e.g. ABC plots) have been widely used to determine the level of disturbance (pollution-induced or otherwise) on benthic macrofauna communities (Ritz et al. 1989;

Warwick 1993; Otway et al. 1996). Under stable conditions of infrequent disturbance the competitive dominants in macrobenthic communities are K-selected, or conservative species with the attributes of a large body size and long-life span: these are rarely dominant numerically but are dominant in terms of biomass. Under moderate pollution, the large competitive dominants are eliminated, and as pollution becomes more severe, benthic communities become increasingly dominated by one of a few very small species.

In contrast, a large biomass of K-selected fish species at the polluted locations and a numerical dominance of small species at the control locations were observed in the present study. A number of other studies have reported that small, demersal fish avoid polluted waters (Kainz and Gollmann 1990;

Stoykov et al. 1994). However, other studies have found no impact of sewage on demersal fish abundance or biomass, because fish are wide ranging and are two or three trophic levels removed from sewage (Musick et al. 1993). Further, Lim and Hong (1994) report the presence of large K-strategists in a polluted bay in Korea. Inconsistent patterns in ABC plots over time were reported for macrofauna associated with a domestic sewage outfall from Jervis Bay, New South Wales (Smith 1994). Thus, responses to pollution may vary according to the species that are investigated, and previously accepted generalisations and models of community behaviour should be interpreted with caution for other species and locations. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 120

Research implications

Underwater visual census is a non-destructive method of sampling which enables the collection of quantitative information on the fish assemblages, condition and behaviour (Lincoln Smith 1989;

Thompson and Mapstone 1997). Although the length of some species of fish have been estimated in situ (Grigg 1994; Jennings et al. 1995; Samoilys 1997), it is very difficult to estimate weight and it is generally calculated from length estimates. The average weight of individual species were obtained from published literature and estimated from personal observation. The estimation may have some errors, because the average weight of individual species may be different at outfall and control locations (Khalaf et al. 1986). However, sensitivity analysis indicated that when the average biomass of the three most important species was shifted down one size class, the interpretation of the species responsible for differences between outfall and control sites were similar. Additional sensitivity analysis (e.g. more species, increased average weights) could have been undertaken to investigate the effects of biomass of fish assemblages from outfalls and controls, however, this was not pursued further.

There are several methods for estimating fish biomass. The most common is to obtain information on fish length and convert to weight or biomass using length-weight relationships (Samoilys 1997). Fish may also be collected and weighed using a destructive sampling technique (Taylor and Willis 1998), although it would be difficult to obtain sufficient samples for a large number of species (over 100) and different sampling methods (e.g. nets, traps, spear, poison) may be biased. The biomass estimates that were obtained from other published studies may also be biased, because a maximum weight is generally provided in fish identification books for large, recreationally important taxa (e.g. Hutchins and Swainston 1986), and surveys of fishers indicate that they target larger, edible species (i.e. above minimum length) (Lincoln Smith et al. 1989; Kingsford et al. 1991).

Abundance and biomass measures both distinguished polluted and unpolluted locations, and future research on temperate reef fish should collect information on fish length to enable routine estimation of CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 121

biomass. The utility of using both measures is that it provides two different interpretations of which

species are important in establishing these differences.

Management implications

It is important to consider all the issues when discussing how to deal with sewage disposal (Koop and

Hutchings 1996). The impact of sewage on assemblages of rocky reef fish may be regarded as

relatively insignificant by some members of the community but a priority for resource users such as

fishers and divers. Managers have a responsibility to prevent or mitigate the impacts of anthropogenic

activities on biological factors such as species composition (ANZECC 1992). Therefore, managers

should consider the options of maintaining the current sewage discharge which is preferred by some

large species of fish, or shutdown the outfalls and facilitate the restoration of some small species of

fish.

Interest groups such as commercial and recreational fishers may support some sewage outfalls if they

were convinced that the waters adjacent to the outfalls contained increased abundance and/or biomass

of prized angling species such as Acanthopagrus australis, Achoerodus viridis and Argyrosomus japonicus. However, these decisions must be balanced by the decreased abundance and/or biomass of

other prized angling species such as Seriola lalandi and the reduced abundance of numerous smaller

bait species. Also, bioaccumulation of contaminants in fish captured adjacent to sewage outfalls and

the risks to human consumers (Beder 1989; McLean et al. 1991; Scanes and Philip 1995; Gibbs and

Miskiewicz 1995; Otway et al. 1996), together with the loss of recreational amenity to some fishers

and other water users such as SCUBA divers must also be considered.

REFERENCES

Andrijanic, S. (1992). Bioaccumulation of organochlorines in red morwong (Cheilodactylus fuscus) in

the Sydney region: management implications. In: Proceedings of a Bioaccumulation Workshop.

(ed. T. Miskiewicz 1992) pp. 55-64. Published by the Water Board and Australian Marine

Sciences Association Inc., Sydney. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 122

ANZECC (1992). Australian Water Quality Guidelines for Fresh and Marine Waters. Australian and

New Zealand Environment and Conservation Council, Canberra

Baird, R.C. (1996). Towards new paradigms in coastal resource management: linkages and institutional

effectiveness. Est. 19, 320-35.

Barrett, N.S. (1995). Short- and long-term movement patterns of six temperate reef fishes (Families

Labridae and Monacanthidae). Mar. Freshwater Res. 46, 853-60.

Beder, S. (1989). Toxic Fish and Sewer Surfing. Allen and Unwin, Sydney, Australia.

Bell, J.D. (1983). Effects of depth and marine reserve fishing restrictions on the structure of a rocky

reef fish assemblage in the north-western Mediterranean Sea. J. Appl. Ecol. 20, 357-69.

Bell, J.D., Harmelin-Vivien, M.L. (1982). Fish fauna of French Mediterranean Posidonia Oceania

seagrass meadows. 1. Assemblage Structure. Tethys 10, 337-47.

Bianchi, G., Hoisaeter, T. (1992). Relative merits of using numbers and biomass in fish community

studies. Mar. Ecol. Prog. Ser. 85, 25-33.

Boates, W.A.M., Russell, K.S. (1992). The option to discharge industrial or sewage effluent to the

marine environment. Wat. Sci. Tech. 25, 75-82.

Chapman, M.G., Underwood, A.J., Skilleter, G.A. (1995). Variability at different spatial scales

between a subtidal assemblage exposed to the discharge of sewage and two control assemblages.

J. Exp. Mar. Biol. Ecol. 189, 103-22.

Chisholm, J.R.M., Fernex, F.E., Mathieu, D., Jaubert, J.M. (1997). Wastewater discharge, seagrass

decline and algal proliferation on the Cote d’Azur. Mar. Poll. Bull. 34, 78-84.

Clarke K.R. (1993) Non-parametric multivariate analyses of change in community structure. Aust. J.

Ecol. 18, 117-43.

Cole, R.G. (1994). Abundance, size structure, and diver-oriented behaviour of three large benthic

carnivorous fishes in a marine reserve in northeastern New Zealand. Biol. Conserv. 70, 93-99.

Earle, S.A. (1995). Sea Change. A Message of the Oceans. G.P. Putnam’s Sons, New York..

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environ. Monit. Assess. 14, 197-210.

Gibbs, P.J., Miskiewicz, A.G. (1995). Heavy metals in fish near a major primary treatment sewage

plant outfall. Mar. Poll. Bull. 30, 667-74. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 123

Gillanders, B.M. (1997). Comparison of growth rates between estuarine and coastal reef populations of

Achoerodus viridis (Pisces: Labridae). Mar. Ecol. Prog. Ser. 146, 283-87.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Mar. Ecol. Prog. Ser. 103, 25-34.

Growns, I.O., Pollard, D.A., Gehrke, P.C. (1998). Changes in river fish assemblages associated with

vegetated and degraded banks, upstream of and within nutrient-enriched zones. Fish. Manag.

Ecol. 5, 55-69.

Hall, J.A., Frid, C.L.J., Gill, M.E. (1997). The response of estuarine fish and benthos to an increasing

discharge of sewage effluent. Mar. Poll. Bull. 34, 527-35.

Harris, J.H. (1995). The use of fish in ecological assessments. Aust. J. Ecol., 20, 65-80.

Henry, G.W. (1984). Commercial and recreational fishing in Sydney estuary. Fisheries Bulletin 1,

NSW Fisheries, Sydney. 47 pp.

Holbrook, S.J., Kingsford, M.J., Schmitt, R.J., Stephens, J.S. (1994). Spatial and temporal patterns in

assemblages of temperate reef fish. Amer. Zool. 34, 463-75.

Hourigan, T.F., Tricas, T.C., Reese, E.S. (1988). Coral reef fishes as indicators of environmental stress

in coral reefs. In: Marine Organisms as Indicators, (eds. D. Soule and G.S. Kleppel) pp. 107-35.

Springer-Verlag, New York.

Hutchins, B., Swainston, R. (1986). Sea Fishes of Southern Australia. Swainston Publishing, Perth.

Jennings, S., Grandcourt, E.M., Polunin, N.V.C. (1995). The effects of fishing on the diversity,

biomass and trophic structure of Seychelles reef fish communities. Coral Reefs 14, 225-35.

Jennings, S., Polunin, N.V.C. (1996). Effects of fishing effort and catch rate upon the structure and

biomass of Fijian reef fish communities. J. App. Ecol. 33, 400-412.

Jones, G.P. (1988). Ecology of reef fish of north-eastern New Zealand: a review. N.Z. J. Mar.

Freshwat. Res. 22, 445-62.

Kainz, E., Gollmann, P. (1990). Contributions to the distribution of some small-sized fish species of

Austrian running waters. Pt 3: gudgeon {Gobio gobio L.; Cyprinidae). Oesterr. Fisch. 43, 80-6.

Kingsford, M.J., Underwood, A.J., Kennedy, S.J. (1991). Humans as predators on rocky reef in New

South Wales, Australia. Mar. Ecol. Prog. Ser. 72, 1-14. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 124

Keough, M.J., Quinn, G.P. (1991). Causality and the choice of measurements for detecting human impacts

in marine environments. Aust. J. Mar. Freshwater Res. 42, 539-54.

Khalaf, A.N., Al-Yamour, K.Y., Allouse, S.B., Al-Jafary, A., Sadek, S.E. (1986). Age, growth, length-

weight relationships, and distribution of khishni Liza abu (Heckel) (Mugilidae) in a polluted

habitat. J. Biol. Sci. Res. 17, 63-81.

Koop, K., Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Mar. Poll. Ball.

33, 121-23.

Leadbitter, D. (1992). Management of fishing activities in contaminated areas. Aust. Fish.. September. 30-

3.

Lim, H., Hong, J.S. (1994). An environmental impact assessment based on benthic macrofauna in

Chinhae Bay, Korea. Bull. Korean Fish. Soc. 27, 659-674.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate

rocky reef fish assemblages. Mar. Ecol. Prog. Ser. 43, 223-31.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Env. Biol. Fish. 26, 29-37.

Lincoln Smith, M.P., Bell. J.D., Pollard, D.A., Russell. B.C. (1989). Catch and effort of competition

spearfishermen in southeastern Australia. Fish. Res. 8, 45-61.

Lincoln Smith, M.P., Jones, G.P. (1995). Fishes of shallow coastal habitats. In: Coastal Marine

Ecology of Temperate Australia, (eds. A.J Underwood and M.G. Chapman) pp. 240-53. UNSW

Press, Sydney.

Lincoln Smith, M.P., Mann, R.P. (1989). Bioaccumulation in nearshore marine organisms. II.

Organochlorine compounds in the red morwong Cheilodactylusfuscus, around Sydney’s three major

sewage ocean outfalls. A report prepared by SPCC and the Ecology Lab Pty Ltd, Sydney.

Love, M.S., Axell, B., Morris, P., Collins, R., Brooks, A. (1987). Life history and fishery of the

Californian scorpionfish, Scorpaena guttata, within the southern Californian Bight. Fish. Bull. 85,

99-116

McCormick, M.I., Choat, J.H. (1987). Estimating total abundance of a large temperate-reef fish using

visual strip-transects. Mar. Biol. 96, 469-78. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 125

McLean, C., Miskiewcz, A.G., Roberts, E.A. (1991). Effect of 3 primary treatment sewage outfalls on

metal concentrations in the fish Cheilodactylus fuscus collected along the coast of Sydney,

Australia. Mar. Poll. Bull. 22, 134-40.

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Manly Hydraulic

Laboratory Report 769. Sydney, 65 pp.

Musick, J.A., Desfosse, J.C., Wilk, S., McMillan, D., Grogan, E. (1993). Historical comparison of the

structure of demersal fish communities near a deep-sea disposal site in the western North Atlantic.

In: Offshore Disposal: Results of the 106 mile Dumpsite Study. Part 3: Biological Fates and

Effects, (ed. A. Robertson), pp. 149-171. Stony Brook, NY.

Nowak, B.F. (1996). Health of red morwong, Cheilodactylus fuscus, and rock cale, Crinodus

lophodon, from Sydney cliff-face sewage outfalls. Mar. Poll. Bull. 33, 281-92.

Otway, N.M, Gray, C.A., Craig, J.R., McVea, T.A., Ling, J.E. (1996). Assessing the impacts of

deepwater sewage outfalls on spatially and temporally variable marine communities. Mar.

Environ. Res. 41,45-71.

Pastorok, R., Bilyard, G.R. (1985). Effects of sewage pollution on coral-reef communities. Mar. Ecol.

Prog. Ser. 21, 175-89.

Ritz, D.A., Lewis, M.E., Shen, M. (1989). Response to organic enrichment of infaunal macrobenthic

communities under salmonid cages. Mar. Biol. 103, 211-14.

Roberts, C.M., Polunin, N.V.C. (1991). Are marine reserves effective in management of reef fisheries?

Rev. Fish Biol. Fisheries 1, 65-91.

Roberts, D.E., Smith, A., Ajani, P., Davis, A.R. (1998). Rapid changes in encrusting marine

assemblages exposed to anthropogenic point-source pollution: a ‘Beyond BACF approach. Mar.

Ecol. Prog. Ser. 163,213-24.

Russo, A.R. (1989). Fish community structure on a deep ocean outfall, Barber’s Point, Oahu, Hawaii.

Int. Revue, ges. Hydrobiol. 74, 499-506.

Samoilys, M. (1997). (Ed). Manual for assessing fish stocks on Pacific coral reefs. Department of

Primary Industries, Queensland. 75 pp.

Scanes, P.R., Philip, N. (1995). Environmental impact of deepwater discharge of sewage off Sydney,

NSW, Australia. Mar. Poll. Bull. 31, 343-46. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 126

Schmitt, R.J., Osenberg, C.W. (1996). Detecting Ecological Impacts, Concepts and Applications in

Coastal Habitats. Academic Press, New York.

Smith, A.K., Ajani, P.A., Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Mar. Environ. Res. 47, 1-20.

Smith, S.D. (1994). Impacts of domestic sewage effluent verses natural background variability: an

example from Jervis Bay, New South Wales. Aust. J. Mar. Freshwater Res. 45, 1045-64.

Steffe, A.S., Murphy, J.J., Chapman, D.J., Tarlinton, B.E., Gordon, G.N.G. & Grinberg, A. (1996). An

assessment of the impact of offshore recreational fishing in New South Wales waters on the

management of commercial fisheries. Final report to The Fisheries Research and Development

Corporation from the Fisheries Research Institute, Sydney. 139 pp.

Stephens, J.S, Jr., Hose, J.E., Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In: Marine Organisms as Indicators, (eds. D.F. Soule and

G.S. Kleppel) pp. 91-105. Springer-Verlag, NY.

Stoykov, S., Kolarov, P., Stanev, T., Murdjeva, D., Atanassova, V., Kolemanova, K. (1994).

Ecological state of the Bourgas Bay biota. Izv. Inst. Ribni-Resur. Varna-Proc Inst. Fish Varna.

22, 5-57.

Taylor, R.B., Willis, T.J. (1998). Relationships amongst length, weight and growth of north-eastern

New Zealand reef fishes. Mar. Freshwater Res. 49, 255-260.

TEL (1994). A Quantitative Survey of the Impact of Sewage on Fish and Macrobenthic Assemblages in

the Illawarra Coastal Zone. Final Report by The Ecology Lab for the Water Board, Sydney.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA.

Tsai, C. (1991). Rise and fall of the Potomac River striped bass stock: a hypothesis of the role of

sewage. Trans. Am. Fisheries. Soc. 120, 1-22.

Thompson, A.A., Mapstone, B.D. (1997). Observer effects and training in underwater visual surveys of

reef fishes. Mar. Environ. Res. 154, 53-63.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15. CHAPTER 6 EFFECTS OF SEWAGE ON ABUNDANCE AND BIOMASS 127

Underwood, A.J., Chapman, M.G. (1996). Subtidal assemblages on rocky reefs at a cliff-face sewage

outfall (North Head, Sydney, Australia): What happened when the outfall was turned off? Mar.

Poll. Bull. 33, 293-302.

Underwood, A.J., Kingsford, M.J., Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Aust. J. Ecol. 6, 231-49.

Underwood, A.J., Peterson, C.H. (1988). Towards an ecological framework for investigating pollution.

Mar. Ecol. Prog. Ser. 46, 227-34.

Warwick R.M. (1993). Environmental impact studies on marine communities: Pragmatical

considerations. Aust. J. Ecol. 18, 63-80.

Watson, M., Ormond, R.F.G (1994). Effect of an artisanal fishery on the fish and urchin populations of

a Kenyan coral reef. Mar. Ecol. Prog. Ser. 109, 115-29.

Wright, I. A., Chessman, B.C., Fairweather, P.G., Benson, L.J. (1995). Measuring the impact of sewage

effluent on the macroinvertebrate community of an upland stream: the effect of different levels of

taxonomic resolution and quantification. Aust. J. Ecol. 20, 142-49. Red Morwong Cheilodactylus fuscus White Ear Parma microlepis CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 128

CHAPTER 7

CONDITION OF TWO SPECIES OF RESIDENT REEF FISH

EXPOSED TO SEWAGE EFFLUENT DISCHARGE

ABSTRACT

The potential impact of sewage on the condition of two species of fish, red morwong Cheilodactylus fuscus (Cheilodactylidae) and white ear Parma microlepis (Pomacentridae), was assessed at three

scales: (1) outfalls versus controls, (2) among sites, and (3) male and female fish within sites. Each

scale gave a different perspective to the interpretation of a sewage impact. The comparison of outfalls

verses controls was confounded by significant differences between sites. Significant differences were

found between male and female fish for some variables and indices at some sites. The Gonadosomatic

Index for female C. fuscus and P. microlepis was significantly greater at sewage sites. Morphological

analysis found smaller male C. fuscus at North Head, and larger organs for female C. fuscus at Potter

Point. “Controf’sites within the Sydney region may also be impacted by diffuse pollution, and

additional distant control sites, and further temporal sampling is needed to validate these results. A

comparison of C. fuscus collected in this study with fish collected during the 1980s, indicated that liver

index increased by approximately 10% at all sites over time, suggesting further environmental stress on

this species.

INTRODUCTION

Exposure to sewage pollution is manifest at lower levels of biological organisation before disturbances

are realized at the population, community or ecosystem levels (Underwood and Peterson, 1988;

Adams, 1990). Studies concerned with the condition of fish have measured body fluids, cells, tissues or

other biotic variables that indicate the presence and magnitude of stress responses (Adams, 1990;

Adams et al., 1989; Adams et al., 1993; Goede and Barton, 1990; Nowak 1996). Stress may induce

structural and functional responses in fish such as reduced condition, reduced growth rate, impaired CHAPTER 7' EFFECTS OF SEWAGE ON FISH CONDITION 129 fecundity and/or increased deformities, neoplasia (cancers), contaminant levels, parasites and diseases

(Brown et al., 1987; Adams, 1990; EPA, 1993; Avery et al, 1996; Moore et al, 1996; Waring et al.,

1996; Kakuta and Murachi, 1997).

Detection of an impact on the condition of fish that can be attributed to pollution may be difficult.

Even fish of the same species, population, age and sex may respond differently to sewage pollution due to differences in their general health status, habitat, diet and season (Gibbs and Miskiewicz, 1995;

Krogh and Scanes, 1996). Small juveniles and larval fish are particularly susceptible to pollution and relatively brief exposures (24 - 96 h) of larvae to sewage plumes off Sydney has resulted in deformities

(Kingsford and Gray, 1996). Conversely, large fish are frequently found adjacent to sewage outfalls because of enhanced feeding opportunities, although a negative physiological impact may also occur

(Tsai, 1975; 1991; Khalaf et al, 1986; Lincoln Smith and Mann, 1989a,b; Nowak, 1996).

The study described in this paper investigates the effects of sewage on the condition of two species of common, resident, temperate reef fish. Red morwong (Cheilodactylus fuscus) and white ear {Parma

microlepis) were chosen for investigation because underwater visual surveys in the Sydney and

lllawarra regions have consistently found greater numbers of C. fuscus and lesser numbers of P.

microlepis at sewage outfalls relative to unpolluted sites (TEL 1993, 1994). This study sought a

physiological explanation for why these species were found in greater and lesser numbers. One of these

species, C. fuscus has been widely used as an indicator species of sewage pollution (Nowak, 1996),

however, previous studies have focused on chemical (Beder, 1989; Lincoln Smith and Mann, 1989b;

McLean et al., 1991; Andrijanic, 1992; Leadbitter, 1992; Miskiewicz, 1992; Gibbs and Miskiewicz,

1995; EPA, 1996; Krogh and Scanes, 1996) or histological analyses (Nowak, 1996). Some biological

characteristics have been measured in these studies, which found the mean weight of C. fuscus, and a

liver weight index, were greater at sites closer to nearshore sewage outfalls in the Sydney region

(Lincoln Smith and Mann, 1989b). Some research has also been conducted on the impacts of

contaminants in sewage pollution on P. microlepis (Tricklebank, 1994). O North Head SYDNEY

■ Rosa Gully

O Gordons Bay

■ Potters Point

Figure 7.1. Location of sampling sites for collection of white ear Parma microlepis and red morwong

Cheilodactylus fuscus at sewage outfall locations (■) and control locations (O) in the Sydney region. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 130

Chemical and histological analyses are useful techniques to investigate impacts of pollution on fish,

however, they cannot be rapidly applied to field studies. This study investigated the applicability of a

rapid and inexpensive autopsy-based assessment of morphometric and organ indices (Goede and

Barton, 1990; Adams et al., 1993) to detect gross changes in fish condition. As this study’s sampling

occurred at only one time, a comparison was made of biological characteristic of C. fuscus collected in

this study with fish collected in the 1980s to investigate any putative long-term impacts.

MATERIALS AND METHODS

Species, collection and measurements of fish

Cheilodacty/us fuscus is a large (65 cm and 3 kg), common, resident species of rocky reefs and is

distributed from Hervey Bay, Queensland to Mallacoota, Victoria (Hutchins and Swainston, 1986). C. fuscus is a microcarnivore (Bell, 1979) which spawns in autumn (Lowry, 1997). It is one of the most

common species captured by spearfishers (Lincoln Smith et al., 1989).

Parma microlepis is a small (20 cm and 200 g), common, resident species of rocky reefs, and is

distributed from Byron Bay, NSW to Port Phillip Bay, Victoria (Hutchins and Swainston, 1986). P.

microlepis is a herbivore (Kuiter, 1993), spawning in early spring to early summer (Tzioumis and

Kingsford, 1995).

A total of 20 C. fuscus and 20 P. microlepis were collected from each of 4 sites in the Sydney region

(n=160) during February 1996. Fish were collected using a spear from two sewage outfall sites (Rosa

Gully, Potter Point) and two control sites (North Head and Gordons Bay, Figure 7.1). The outfall at

Rosa Gully discharged approximately 5.1 ML/day of raw sewage and Potter Point discharged

approximately 46 ML/day of primary treated sewage (MHL, 1997).

All fish were stored on ice and measured and dissected on the same day that they were captured.

Measurements and observations were obtained on total length (mm), fork length, girth, body depth,

head length, eye diameter, caudal peduncle width, total weight (g), gutted weight, external condition,

sex, gonad weight, liver weight, liver colour (red - healthy, coffee-brown - unhealthy (Goede and CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 131

Barton, 1990)) and stomach weight. Liver health was assessed as a binary scale: healt hy = 1, unhealthy = 0. A morphometric index of condition was calculated as 105 weight/total length3 (Goede and Barton, 1990). Three organ indices of condition were calculated: Hepatosomatic index (HSI)

((liver weight/gutted weight)* 100), Gonadosomatic index (GSI) ((gonad weight/gutted weight)* 100) and Stomach weight index (SWI) ((stomach weight/gutted weight)* 100)).

The fork length, weight, liver weight and liver index ((liver weight/total weight)* 100) of C.fuscus from the four sites in this study were compared with fish collected from similar sites i n the late 1980s by Lincoln Smith and Mann (1989b) and Andrijanic (1991). These fish were all collected using the same method but at different times: February 1996 (this study), July 1987 (Lincoln Smith and Mann,

1989b) and from September to November 1989 (Andrijanic, 1991).

Statistical analyses

Organ indices were checked for independence of length by regression analysis. One-way analysis of

variance (ANOVA) techniques (Underwood, 1997) were used to test the null hypotheses that there were no differences in fork length, total weight, condition, liver health, HSI, SWI and GSI between the

fish from four sites. A binary system was used for statistical analysis of liver health with 1 (healthy -

red in colour) and 0 (unhealthy - coffee in colour).

Non-metric multi-dimensional scaling (MDS) was undertaken on data using the Bray-Curtis similarity

measure (Clarke, 1993). Separate morphometric and organ analyses were undertaken for male and

female C.fuscus, and male and female P. microlepis, using standardised data (total of 8 analyses).

Significant differences in morphometric or organ indices between fish from outfall and control sites

were tested for by one-way analysis of similarity (ANOSIM) (Clarke, 1993). The similarity of

percentages (SIMPER) procedure was used to identify the contribution of individual variables to

differences between outfall and control sites (Clarke, 1993). Multivariate analyses were performed

using the PRIMER package (Plymouth Marine Laboratories, UK). Table 7.1 . Mean (±SE) measurements, condition factor, organ indices and sex ratios for white ear

Parma microlepis and red morwong Cheilodactylus fuscus from sewage outfall (n=40) and control locations (n=40) in the Sydney region.

Parma microlepis Cheilodactylus fuscus

Measurement Outfalls Controls Outfalls Controls

Total length (L, mm) 184 (2.1) 163 (2.5) 423 (5.8) 395 (5.7)

Fork length (mm) 160(1.7) 139 (2.3) 377 (5.1) 351 (5.1)

Total weight (W, g) 157(5.2) 113 (5.1) 1021 (45) 871 (42)

Condition factor (105W/L3) 2.48 (0.03) 2.55 (0.02) 1.32 (0.02) 1.37(0.02)

Liver (% normal) 0.05 0.37 0.30 0.37

Hepatosomatic Index 1.64 (0.04) 1.55 (0.04) 1.12(0.03) 1.04 (0.03)

Stomach Weight Index 10.37(0.41) 10.85 (0.51) 5.10(0.17) 5.42 (0.21)

Gonadosomatic Index 0.25 (0.04) 0.29 (0.03) 0.65 (0.09) 0.36 (0.05)

Sex (% female) 0.35 0.57 0.55 0.50 CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 132

RESULTS

Fish characteristics

C.fuscus varied in size from 339 to 530 mm (mean 409.2 + 4.3, n=80) and 551 to 2123 g (mean 946.2

+ 32.0, n=80). P. microlepis varied in size from 125 to 200 mm (mean 173.5 + 2.0) and 50 to 204 g

(mean 135.4 + 4.4, n=80). Generally, fish from outfalls were larger and heavier than fish from control

locations (Table 7.1). However, one way analysis of variance indicated that there were often significant

differences between the four sites (Tables 7.2, 7.3; Figures 7.2, 7.3). ANOVA indicated that C.fuscus

from the control site at North Head, and P. microlepis from the control site at Gordons Bay were

significantly smaller in length and lighter in weight than fish from other sites (Tables 7.2, 7.3). These

patterns were further complicated by significant differences between sexes for some variables at some

sites (Figure 7.3). For example, male C.fuscus were larger than females at two sites (Figure 7.3).

Separate multivariate analyses were undertaken for morphometric indices for male and female C. fuscus, and male and female P. microlepis (4 analyses). Stress values for the analyses of morphometric

data ranged from 0.02 to 0.05 suggesting good representations (Clarke 1993) of the original data in two

dimensions. All ANOS1M tests showed some significant differences for the morphometric variables of

male and female C.fuscus and P. microlepis between sites (Table 7.4). Consistent morphological

differences were detected for male C.fuscus at North Head compared to the other three sites (Table

7.4, Figure 7.4a). The variable that contributed most to discriminating between sites was weight.

Condition index

P. microlepis are a relatively smaller, more thickset fish than C.fuscus and had a greater condition

index (means of approximately 2.50 verses means of 1.35, respectively) (Table 7.1). C.fuscus from the

outfall locations (mean 1.32 + 0.02) had a lower condition index than fish from control locations (mean

1.37 + 0.02) (Table 7.1). Significant differences in condition between sites were evident for C.fuscus

(Table 7.2) but no significant differences were detected for P. microlepis (Table 7.3). Table 7.2. Red morwong Cheilodactylus fuscus. Results of one way analysis (3,76 df) of variance for fork length, weight, Condition, liver health, Hepatosomatic Index, Stomach Weight Index and

Gonadosomatic Index for males (3,33 df) and females (3,37 df)NH = North Head, RG = Rosa Gully outfall, GB = Gordons Bay, PP = Potters Point outfall. * < 0.05, ** <0.01 (untransformed data), df = degrees of freedom.

VARIABLE MS F TUKEYS TEST

Fork length 10125 11.8 ** NH < GB = PP = RG

Weight 536049 8.3 ** NH < RG = PP = GB

Condition 0.07 6.3 * * RG = NH < PP = GB

Liver health 0.246 1.1 ns GB = RG = PP = NH

Hepatosomatic Index 0.076 1.9 ns NH = GB = PP = RG

Stomach Weight Index 13.0 12.1 * * NH = RG = PP < GB

Gonadosomatic Index 1.21 5.8 ** GB = NH = RG < PP

Gonadosomatic Index (M) 0.009 1.5 ns GB = RG = PP = NH

Gonadosomatic Index (F) 0.61 7.1 * * NH = GB < RG < PP Table 7.3. White ear Parma microlepis. Results of one way analysis (3,76 df) of variance table for fork length, weight, Condition, liver health, Hepatosomatic Index, Stomach Weight Index and

Gonadosomatic Index for males (3,38 df) and females (3,32 df)- NH = North Head, RG = Rosa Gully,

GB = Gordons Bay, PP = Potters Point. * < 0.05, ** <0.01 (untransformed data), df = degrees of freedom.

VARIABLE MS F TUKEYSTEST

Fork length 5018 54.7 ** GB < NH = PP < RG

Weight 25581 42.1 ** GB

Condition 0.54 2.4 ns PP = RG = GB = NH

Liver health 0.74 5.1 ** PP < RG = GB = NH

Hepatosomatic Index 0.48 7.4 ** NH = RG < GB = PP

Stomach Weight Index 90.1 17.0 ** NH = RG < PP < GB

Gonadosomatic Index 0.01 0.2 ns PP = RG = NH = GB

Gonadosomatic Index (M) 0.001 2.5 ns NH = RG = PP = GB

Gonadosomatic Index (F) 0.12 3.3 * GB = NH < PP = RG (a) Parma microlepis (b) Cheilodactylus fuscus

£ f I s>

£

o> CJ ■5

! o E s TO 5 NH GB RG PP NH GB RG PP

Controls Outfalls Controls Outfalls

Figure 7.2. Mean (±SE) length, weight, condition, liver health, hepatosomatic index, stomach weight index and gonadosomatic index for (a) white ear Parma microlepis and (b) red morwong

Cheilodactylus fuscus from two control and two outfall sites in the Sydney region. (a) Parma microlepis (b) Cheilodactylus fuse us

I £ CT> jy

"O TO Cn

I s

NH Controls Outfalls Controls Outfalls

Figure 7.3. Mean (±SE) length, weight, condition, liver health, hepatosomatic index, stomach weight index and gonadosomatic index for (a) male and female white ear Parma microlepis and (b) male and female red morwong Cheilodactylus fuscus from two control and two outfall sites in the Sydney region. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 133

Organ indices

A minority of the P. microlepis and C.fuscus from both outfall and control locations had ‘normal’ livers. P. microlepis from outfall locations had significantly fewer normal livers (5%) compared to

37% at control locations (Tables 7.1, 7.3). Female P. microlepis had a greater proportion of normal livers than males (Figure 7.3).

Linear regression of hepatosomatic and stomach weight indices on length showed no significant association for both species. Gonadosomatic indices were independent of length for male and female P. microlepis and male C.fuscus, but were positive for female C.fuscus (y = 1.38 + 6.43e-3x, R2 = 0.27).

This relationship for female C.fuscus was not incorporated into further analyses because of the small size range sampled.

MDS ordinations and ANOS1M showed that the organ indices of male and female C.fuscus and P. microlepis differed between the four sites (Table 7.4). Stress values ranged from 0.04 to 0.07 giving good representations (Clarke 1993) of the original data in two dimensions. ANOSIM tests showed that there were no differences for male organs for P. microlepis at the four sites. All other ANOSIM tests showed some significant differences between sites (Table 7.4). Consistent differences were detected for female C.fuscus at Potter Point compared to the other three sites (Table 7.4, Figure 7.4b). The variables that contributed most to distinguishing these fish at Potter Point from other sites were stomach weight (Gordons Bay and Rosa Gully) and gonad weight (North Head).

Long-term trends

Mean fork length, weight, liver weight and liver index of C.fuscus from three sampling periods (1987,

1989, 1996) and 4 sites were tabulated (Table 7.5) but not analysed because the original raw data were not available. There did not appear to be any consistent long-term patterns in the fork length or weight of C.fuscus from this study when compared to fish collected in the 1980s. However, there did appear to be a consistent pattern for increased liver index over time at the four sites. The liver index appeared to increase by approximately 10 to 30 % at three of the four sites over the period 1987 to 1996. Table 7.4. R-statistics (Clarke 1993) from one-way ANOSIM and pairwise comparisons of four sites for morphometric and organ indices of male and females of two species of fish, red morwong

Cheilodactylus fuscus and white ear Parma microlepis (8 analyses). NH = North Head control, RG =

Rosa Gully outfall, GB = Gordans Bay control, PP = Potter Point outfall. Data was standardised.

Number of permutations = 5000. ns = not significant, R (* = significant at oc<0.05), Pairwise

comparisons (**= significant at oc<0.008 - alphas adjusted for multiple comparisons).

Morphometric Organ

Male Female Male Female

C. fuscus All sites

R = 0.295 * R = 0.297 * R = 0.161 * R = 0.431 *

C. fuscus Pairwise comparisons

NH v RG * * NH v RG ** NH vRG ns NH v RG ns

NH vGB ** NH vGB ns NH vGB ** NH vGB ns

NH vPP ** NH v PP ** NH v PP ns NH v PP * *

RG vGB ns RG vGB ns RG vGB ns RG vGB * *

RG v PP ns RG v PP ns RG v PP ns RG v PP * *

GB v PP ns GB v PP ns GB v PP ns GB v PP * *

P. microlepis All sites

R = 0.347 * R = 0.429 * R = 0.100 * R = 0.217 *

P. microlepis Pairwise comparisons

NH vRG ** NH vRG ns NH vRG ns NH vRG ns

NH vGB ** NH vGB ** NH vGB ns NH vGB **

NH v PP ns NH vPP ns NH v PP ns NH v PP ns

RG vGB ** RG vGB ** RG vGB ns RG vGB * *

RG v PP ns RG v PP ns RG vPP ns RG vPP ns

GB vPP ** GB v PP ** GB v PP ns GB v PP ns Figure 7.4. MDS ordinations of (a) morphometric variables for male red morwong Cheilodactylus fuscus (stress = 0.02), and (b) organ indices for female C.fuscus (stress =0.06 ), at four sites in the

Sydney region. The North Head and Gordons Bay control sites are represented by small and large circles respectively. The Rosa Gulley and Potter Point sewage outfall sites are represented by small and large squares respectively. Significant differences are represented by filled symbols. Table 7.5. Mean (± SE) (± SD for 1989 data) fork length, weight, liver weight and liver index of red morwong Cheilodactylus fuscus from this study (1996) compared to fish captured at the same sites during 1989 (Andrijanic 1991) and 1987 (Lincoln Smith and Mann 1989b). na- not applicable as fish not sampled.

Year Fork length Weight Liver Weight Liver Index

North Head 1996 332 (4.8) 701 (26) 6.69 (0.36) 0.95 (0.03)

1989 236(11.3) 485 (88) 4.64 (1.73) 0.88 (0.27)

1987 359 (7.8) 860 (44) 11.43 (1.64) 0.85 (0.12)

Rosa Gully 1996 383 (6.0) 1015 (48) 11.05 (0.68) 1.08 (0.04)

1989 363 (4.5) 997 (33) 11.94 (4.01) 1.05 (0.43)

1987 378 (10.7) 1081 (96) 12.09(1.46) 0.98(0.12)

Gordons Bay 1996 370 (6.6) 1041 (61) 10.21 (0.70) 0.98 (0.04)

1989 244 (2.1) 275 (94) 2.58 (1.77) 0.80 (0.46)

1987 337 (11.1) 867 (82.6) 12.44(1.29) 0.71 (0.03)

Potter Point 1996 372 (8.3) 1028 (78) 10.35 (0.91) 1.00 (0.04)

1989 352 (3.9) 889 (338) 8.47 (4.4) 0.93 (0.21)

1987 na na na na CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 134

It should be noted that the liver index measured by Lincoln Smith and Mann (1989a; 1989b) and

Andrijanic (1991) is probably not as accurate a measure of stress as HSI, because the liver index is a

ratio of body weight which may vary according to gonad stage and stomach contents, and HSI is a ratio

of gutted weight.

DISCUSSION

Three main issues will be discussed. Firstly, the results of the analyses of morphometric indices are

compared with other studies. Secondly, the results of analyses of organ indices are compared with

other studies, including long-term trends for increasing liver index of C.fuscus. Finally, the

implications of sample sizes and selection of control sites are discussed and recommendations are made

for future research.

Use of morphometric indices in detecting sewage pollution effects

There was a general trend for greater length and weight of both species at outfall locations compared to

control locations. However, when the data were analysed at finer scales (sites and sex of fish) there

were significant differences between sites which could not be attributed solely to a sewage pollution

effect. The only consistent morphometric effect was that male C.fuscus from the North Head site were

smaller. Several interpretations of this data are possible. Male C.fuscus from North Flead may be more

stressed than fish from the other three sites as a result of point and non-point source pollution

discharged to Sydney Harbour (see Krogh and Scanes, 1996). Secondly, overall length and weight

were not useful in discrimination of the two species of fish from outfall and control locations at one

time.

In a natural population of fish, condition and growth are sensitive measures of stress, and toxic effects

on them and their environment (Mitz and Giesy, 1985; Goede and Barton, 1991). Congregations of

large and small sized fish around sewage outfalls have been reported in separate studies (Tsai, 1975;

Pastorok and Bilyard, 1985; Russo, 1989; Grigg, 1994; Growns et al., 1998). The presence and

abundance of large species may be attributed to high nutrient levels providing a direct or indirect

source of food (Tsai, 1975; Stephens et al., 1988; Lincoln Smith and Mann, 1989b; Hall et. al. 1997), CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 135 although small, demersal species may avoid polluted waters (Kainzand Gollmann, 1990; Stoykov et al., 1994). In Santa Monica Bay, California, it was reported that spotted turbot Pleuronichthys ritteri

from the vicinity of a sewage outfall were of a significantly lighter weight than those caught at control

sites (Tsai, 1975). Lincoln Smith and Mann (1989b) sampled C.fuscus from 24 sites in the Sydney

region, and reported that fish captured within 500 m of three large cliff face outfalls (now

decommissioned) were significantly heavier than fish caught further away from the outfalls.

Use of organ indices in detecting sewage pollution effects

Three organ indices were investigated in this study. Two of these, Hepatosomatic Index (HSI) and

Gonadosomatic Index (GSI) have been widely used. The third, Stomach Weight Indices (SWI), has

rarely been used for detection of condition, although Goede and Barton (1991) used both a hind gut

index and a mesenteric fat index.

Most of the fish in this study had ‘unhealthy’ livers which were light or coffee-cream colour, indicative

of high fat deposition as compared to ‘normal’ red livers. A fatty liver is usually a pathological state

attributable to excessive accumulation of lipids as a result of exposure to PCBs and other organic

compounds and enhanced detoxification activities (Adams et al., 1993). A significant impact was

detected for one species, P. microlepis, where only 5% of fish from outfall locations contained healthy

livers compared to 37% at control locations. There was some difference between locations and species

for the HSI and SWI, but no impact could be specifically attributed to sewage pollution. Other studies

which have reported an increased liver weight of flounder Platichthys flesus exposed to sewage in the

Tyne estuary, England (Lye et al., 1997) and a negative correlation of liver weight of plaice

Pleuronectes platessa with distance from a sewage sludge dumpsite in the Firth of Clyde, Scotland

(Secombes et al., 1995).

Previous studies of C.fuscus by Lincoln Smith and Mann (1989a; 1989b) reported that fish from

Malabar sewage outfall contained a 46% larger liver than fish from two control sites, and in a more

comprehensive study of fish in the Sydney region the liver index varied between sites and significantly

greater liver indices were reported from one site near North Head cliff face outfall. Liver indices of C. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 136 fuscus collected in the 1980s by Lincoln Smith and Mann (1989b) and Andrijanic (1991) were

compared with the fish collected in this study and the liver index appears to have increased at each of

the four sites by approximately 10 to 30% over a period of 10 years. This may be an indicator of

increasing pollution in the Sydney region resulting in increased stress on fish. This interpretation may

be limited because there were only three times and no statistical analyses were undertaken. One

possible confounding effect may be that fish were collected in February in this study and fish were

collected in July by Lincoln Smith and Mann (1989b) and September-November by Andrijanic (1991)

and there may be seasonal changes in liver weight.

The GSI of both C. fuscus and P. microlepis indicated a sewage impact, attributed to a greater mean

gonad weight for female fish. The multivariate analyses supported an effect on organs of female C. fuscus at the Potter Point site. No impact was detected for male fish. An increase in GSI is contrary to

most other studies which have reported a decrease in GSI or egg size when fish are exposed to sewage

pollution. For example, sand gobies Pomatoschistus minutus exposed to 0.1 % sewage sludge produced

a decreased number of eggs (Waring et al., 1996) and the reproduction potential of burbot Lota lota in

the Rybinskoe Reservoir, Russia, has been reduced by approximately 50% due to sewage disposal

(Volodin, 1994). An unresolved question is whether increased GSI of C. fuscus and P. microlepis may

be related to increased reproductive success. It is probable that other factors are also important as a

study of the hulafish Trachinops taeniatus reported no difference in GSI, but an increased number of

eggs and decreased egg size for fish adjacent to sewage outfalls (Smith and Suthers unpublished).

Research and management implications

The number of samples required in a study is often determined by a the power required to detect

differences (Andrew and Mapstone, 1987; Otway, 1992; Underwood, 1994), pilot studies, catchability

and cost. Similar studies have sampled 30 fish (Adams et al., 1993), 20 fish (Goede and Barton, 1990),

10 fish (Nowak, 1996) and 8 fish (Lincoln Smith and Mann, 1989b; Andrijanic, 1991) per site. Other

studies have been limited due to insufficient samples or an unbalanced design (Krogh and Koop,

1996). The collection of 20 fish per site for this study, is at the upper range in comparison to other . CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 137 studies. However, the male and female fish of each species from each site were also analysed separately, and sample sizes ranged from 6 to 14.

The choice of control sites has a great effect on the potential to detect effects. Sites were selected where spearfishing is prohibited (North Harbour is an Aquatic Reserve and Gordons Bay is closed to spearfishing) so that if an impact was detected it could not be confounded due to the impact of fishing activities. However, recent research indicates that C.fuscus throughout the Sydney region may be contaminated with organochlorines and/or heavy metals (Krogh and Scanes, 1996), and thus fish from the ‘control’ sites at North Head and Gordons Bay may be effected by other point and diffuse pollution sources. Sampling at additional outfall and control sites, and control sites that are further away from the outfalls could help increase the ability to detect the impacts of sewage pollution. Alternatively, fish could be placed in cages at replicate control and outfall sites to investigate biological impacts. Caging experiments of invertebrates such as oysters are commonly used to detect the impact of pollution on condition (Avery et al., 1996; Scanes, 1996). Caging of fish would require exposure for long periods to detect gross biological changes (Gossett et al., 1984; Mitz and Giesy, 1985) and this is difficult logistically.

It is recommended that future research on the condition of both species should only measure Fork length, total weight, HSI, liver index, % normal livers and GSI for male and female fish, separately.

Additional sampling of C.fuscus at North Head should be undertaken to determine if these fish are impacted by pollution from Sydney Harbour or other factors such as habitat or age are responsible for the relatively smaller size of fish at this site. Research should also be conducted on the livers of P. microlepis which were frequently ‘unhealthy’ at the outfall locations.

The rapid autopsy-based assessment of morphometric and organ indices developed by Goede and

Barton (1990) and Adams et al., (1993) provided a useful tool for investigating the impacts of sewage pollution of fish. Several morphological and organ indices of temperate reef fish, particularly fish length, liver health and gonad tissue indicated significant impacts that could be attributed to sewage pollution, however, additional temporal and spatial sampling is recommended to validate these CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 138 patterns. The trend for an increase in liver index of C.fuscus over the past decade may indicate increased pollution in the Sydney region. Managers should consider that C.fuscus is the most common species captured by spearfishers in south-eastern Australia (Lincoln Smith et al., 1989), and if deleterious impacts of sewage on fish condition are occurring they may render the species unsuitable for human consumption and reduce the viability of stocks. It is recommended that a long-term monitoring study of the condition of C.fuscus and P. microlepis should be initiated in the Sydney region and adjacent to other sewage outfalls in NSW waters.

REFERENCES

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium 8,

Bethesda, Maryland.

Adams, M. S, Brown, A.M. and Goede, R.W. (1993). A quantitative health assessment index for rapid

evaluation of fish condition in the field. Transactions of the American Fisheries Society. 122, 63-

73.

Adams, M. S, Shepard, K.L., Greeley, M.S., Jimenez, B.D., Ryon, M.G., Shugart, L.R. and McCarthy,

J.F. (1989). The use of bioindicators for assessing the effects of pollutant stress on fish. Marine

Environmental Research 28, 459-464.

Andrew, N.L. and Mapstone, B.D. (1987). Sampling and the description of spatial patterns in marine

ecology. Oceanography Marine Biology’ Annual Review 25, 39-90.

Andrijanic, S. (1991). Organochlorine residues in fish: bioaccumulation in red inorwong

{Cheilodactylus fuscus) in the Sydney region. Progress Report. Sydney Water. 15 pp.

Andrijanic, S. (1992). Bioaccumulation of organochlorines in red morwong (Cheilodactylus fuscus) in

the Sydney region: management implications. In Proceedings of a bioaccumulation workshop, ed.

A. Miskiewicz, pp. 55-64. Water Board and Australian Marine Sciences Association Inc. Sydney.

Avery, E.L., Dunstan, R.H. and Nell, J.A. (1996). The detection of pollutant impact in marine

environments: condition index, oxidative DNA damage, and their associations with metal

bioaccumulation in the Sydney rock oyster Saccostrea commercialis. Archives Environmental

Contamination Toxicology 31, 192-198.

Beder, S. (1989). Toxic fish arid sewer surfing. Allen & Unwin, Sydney. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 139

Bell, J. (1979). Observations on the diet of red morwong, Cheilodactylus fuscus Castelnau (Pisces:

Cheilodactylidae). Australian Journal of Marine and Freshwater Research 30, 129-133.

Brown, D.A., Bay, S.M., Greenstein, D.J., Szalay, P., Hershelman, G.P., Ward, C.F., Westcott, A.M.

and Cross, J.N. (1987). Municipal wastewater contamination in the southern California Bight:

part 2 - cytosolic distribution of contaminants and biochemical effects in fish livers. Marine

Environmental Research 21, 135-161.

Clarke K.R. (1993) Non-parametric multivariate analyses of change in community structure.

Australian Journal of Ecology 18, 117-143.

EPA (1993). Deformities and associated sublethal effects in fish exposed to sewage-borne

contaminants literature review. Environment Protection Authority, 93/72. August. Sydney.

EPA (1996). Sydney Deepwater Outfalls Environmental Monitoring Program. Final Report Series.

Volume 1. Environment Protection Authority, Bankstown, Sydney.

Gibbs, P.J. and Miskiewicz, A.G. (1995). Heavy metals in fish near a major primary treatment sewage

plant outfall. Marine Pollution Bulletin 30, 667-674.

Goede, R W. and Barton, B.A. (1990) Organismic indices and an autopsy-based assessment as

indicators of health and condition of fish. In Biological Indicators of Stress in Fish. ed. M. S.

Adams, pp. 93-108. American Fisheries Symposium 8, Bethesda, Maryland.

Gossett, R.W., McHugh, S.R., Szalay, P., Rosenthal, K.D. and Brown, D.A. (1984). Xenobiotic

organics and biological effects in scorpionfish caged near a major southern California municipal

outfall. Marine Environmental Research 14, 449-450.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Marine Ecology Progress Series 103, 25-34.

Growns, I.O., Pollard, D.A. and Gehrke, P.C. (1998). Changes in river fish assemblages associated

with vegetated and degraded banks, upstream of and within nutrient-enriched zones. Fisheries.

Management and Ecology 5, 55-69.

Hall, J.A., Frid, C.L.J. and Gill, M.E. (1997). The response of estuarine fish and benthos to an

increasing discharge of sewage effluent. Marine Pollution Bulletin. 34, 527-35.

Hutchins, B. and Swainston, R. (1986). Sea Fishes of Southern Australia. Swainston Publishing, Perth. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 140

Kakuta, I. and Murachi, S. (1997). Physiological response of carp, Cyprinus carpio, exposed to raw

sewage containing fish processing wastewater. Environmental Toxicology Water Quality 12, 1-9.

Kainz, E. and Gollmann, P. (1990). Contributions to the distribution of some small-sized fish species

of Austrian running waters. Pt 3: gudgeon (Gobio gobio L.; Cyprinidae). Oesterriche Fische 43,

80-6.

Khalaf, A.N., Al-Yamour, K.Y., Allouse, S.B., Al-Jafary, A. and Sadek, S.E. (1986). Age, growth,

length-weight relationships, and distribution of khishni Liza abu (Heckel) (Mugilidae) in a

polluted habitat. Journal Biological Science Research 17, 63-81.

Kingsford, M.J. and Gray, C.A. (1996). Influence of pollutants and oceanography on abundance and

deformities of wild fish larvae. In Detecting Ecological Impacts: Concepts and Applications in

Coastal habitats ed. R.J. Schmitt and C.W. Osenberg, pp. 235-255. Academic Press, San Diego.

Krogh, M. and Koop, K. (1996). Design and analysis of the Sydney deepwater ocean outfall

environmental monitoring programme: a critique. Marine Pollution Bulletin 33, 273-280.

Krogh, M. and Scanes, P. (1996). Organochlorine compound and trace metal contaminants in fish near

Sydney’s ocean outfalls. Marine Pollution Bulletin 33, 213-225.

Kuiter, R.H. (1993). Coastal Fishes of South-Eastern Australia. Crawford House Press, Bathurst.

Leadbitter, D. (1992). Management of fishing activities in contaminated areas. Australian Fisheries.

September. 30-33.

Lincoln Smith, M.P., Bell. J.D., Pollard, D.A. and Russell. B.C. (1989). Catch and effort of

competition spearfishermen in southeastern Australia. Fisheries Research 8, 45-61.

Lincoln Smith, M.P. and Mann, R.P. (1989a). Bioaccumulation in nearshore marine organisms 1.

Organochlorine compounds and trace metals in rocky reef animals near the Malabar ocean outfall.

State Pollution Control Commission and the Ecology Lab Pty Ltd, Sydney.

Lincoln Smith, M.P. and Mann, R.P. (1989b). Bioaccumulation in nearshore marine organisms 11.

Organochlorine compounds in the red morwong Cheilodactylus fuscus, around Sydney’s three major

sewage ocean outfalls. State Pollution Control Commission and the Ecology Lab Pty Ltd, Sydney.

Lowry, M. (1997). Population structure of the red morwong Cheilodactylus fuscus. PhD thesis.

University of New South Wales. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 141

Lye, C.M., Frid, C.L.J., Gill, M.E. and McCormick, D. (1997). Abnormalities in the reproductive

% health of flounder Platichthys flesus exposed to effluent from a sewage treatment works. Marine

Pollution Bulletin 34, 34-41.

McLean, C., Miskiewcz, A.G. and Roberts, E.A. (1991). Effect of 3 primary treatment sewage outfalls

on metal concentrations in the fish Cheilodactylus fuscus collected along the coast of Sydney,

Australia. Marine Pollution Bulletin 22, 134-140.

Miskiewicz, A.G. (1992). Proceedings of a Bioaccumulation Workshop. Water Board and Australian

Marine Sciences Association Inc. Sydney.

Mitz, S.V., and Giesy, J.P. (1985). Sewage effluent biomonitoring: 1. Survival, growth, and

histopathological effects in channel catfish. Ecotoxicology 10, 22-39.

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Manly Hydraulic

Laboratory Report 769. June. 65 pp.

Moore, M.J., Shea, D., Hillman, R.E. and Stegeman, J.J. (1996). Trends in hepatic tumours and

hydropic vacuolation, fin erosion, organic chemicals and stable isotope ratios in winter flounder

from Massachusetts, USA. Marine Pollution Bulletin 32, 458-470.

Nowak, B.F. (1996). Health of red morwong, Cheilodactylus fuscus, and rock cale, Crinodus

lophodon, from Sydney cliff-face sewage outfalls. Marine Pollution Bulletin 33, 281-292.

Otway, N.M. (1992). Bioaccumulation studies on fish: choice of species, sampling designs, problems

and implications for environmental management. In Proceedings of a bioaccumulation workshop.

ed. A. Miskiewicz, pp 103-113. Water Board and Australian Marine Sciences Association Inc.

Sydney.

Pastorok, R. and Bilyard, G.R. (1985). Effects of sewage pollution on coral-reef communities. Marine

Ecology Progress Series 21, 175-89.

Russo, A.R. (1989). Fish community structure on a deep ocean outfall, Barber’s Point, Oahu, Hawaii.

Int. Revue ges. Hydrobiology 74, 499-506.

Scanes, P.R. (1996). ‘Oyster Watch’: monitoring trace metal and organochlorine concentrations in

Sydney’s coastal waters. Marine Pollution Bulletin 33, 226-238. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 142

Secombes, C.J., White, A., Fletcher, T.C., Stagg, R.. and Houlihan, D.F. (1995). Immune parameters of

plaice, Pleuronectes platessa, L. along a sewage sludge gradient in the Firth of Clyde, Scotland.

Ecotoxicology 4, 329-340.

Stephens, J.S, Jr., Hose, J.E. and Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In: Marine Organisms as Indicators, eds. D.F. Soule & G.S.

Kleppel, pp. 91-105. Springer-Verlag, NY.

Stoykov, S., Kolarov, P., Stanev, T., Murdjeva, D., Atanassova, V. and Kolemanova, K. (1994).

Ecological state of the Bourgas Bay biota. Izv. Inst. Ribni-Resur. Varna-Proc. Inst. Fish. Varna.

22, 5-57.

TEL (1993). Changes to assemblages of reef fish adjacent to the North Head and Bondi cliff face

outfall following decommissioning. Water Board, Sydney.

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Water Board, Sydney.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of the literature. CRC

Publication No. 36. Centre for Environmental and Estuarine Studies, University of Maryland,

USA.

Tsai, C. (1991). Rise and fall of the Potomac River striped bass stock: a hypothesis of the role of

sewage. Transactions of the American Fisheries Society 120, 1-22.

Tricklebank, K.A. (1994). The distribution and effects of aldrin in the damselfish Parma microlepis: an

experimental field study. International symposium on aquatic animal health: program and

abstracts, University of California, School of veterinary medicine.

Tzioumis, V. and Kingsford, M.J. (1995). Periodicity of spawning of two temperate damselfishes:

Parma microlepis and Chromis dispulus. Bulletin of Marine Science 57, 596-609.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecological Applications 4, 3-15.

Underwood, A.J. (1997). Experiments in Ecology. Their Logical Design and Interpretation using

Analysis of Variance. Cambridge University Press. 504 pp.

Underwood, A.J. and Peterson, C.H. (1988). Towards an ecological framework for investigating

pollution. Marine Ecology Progress Series 46, 227-234. CHAPTER 7 EFFECTS OF SEWAGE ON FISH CONDITION 143

Volodin, V.M. (1994). Status of breeding stocks of burbot, Lota lota, in the Sheksninskii reach of

Rybinskoe Reservoir before and after emergency sewage disposal in 1987. Journal Ichthyology

34, 1828.

Waring, C.P., Stagg, R.M., Fretwell, K., McLay, H.A and Costello, M.J. (1996). The impact of sewage

sludge exposure on the reproduction of the sand goby, Pomatoschistus minutus. Environmental

Pollution 93, 17-25. Eastern Hula Trachinops taeniatus Design of the cages (float, cage, weight) which were placed at sewage outfall and control locations to investigate mortality of eastern hulafish Trachinops taeniatus. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 144

CHAPTER 8

EFFECTS OF SEWAGE EFFLUENT DISCHARGE ON THE

ABUNDANCE, CONDITION AND MORTALITY OF HULAFISH,

Trachinops taeniatus (PLESIOPIDAE)

ABSTRACT

The effects of sewage effluent on the abundance, condition (length and reproduction) and mortality of hulafish Trachinops taeniatus were investigated at multiple outfall and control locations on the central coast of New South Wales, Australia. Underwater visual surveys found consistently fewer T. taeniatus at locations where sewage was discharged compared to control locations. The condition of T. taeniatus was investigated by comparing mean length and reproductive indices of fish from two outfall and two control locations. Fish from the largest outfall location were significantly smaller in comparison with fish from control locations. Gravid female fish from outfall locations had similar gonadosomatic indices but a significantly greater number of eggs and smaller size of eggs in comparison with fish from control locations. Mortality of T. taeniatus was investigated during two week, in situ, caging experiments at multiple locations and times and 80% of fish survived, although mortalities of up to

73% per cage were recorded at one sewage outfall. T. taeniatus may be a suitable environmental indicator species of sewage pollution.

INTRODUCTION

Environmental pollution and its effects on the state of our aquatic resources continues to be a problem of growing, world-wide concern (Earle, 1995; Baird, 1996; Koop & Hutchings, 1996). The most common cause of point source pollution to the world’s waterways is the disposal of sewage (Dubinsky

& Stambler, 1996). The early detection or forecasting of human impacts on the distribution and abundance of organisms will enable effective management of coastal environments (Jones & Kaly,

1996) and ensure the survival of aquatic ecosystems and associated activities, such as fisheries and CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 145 coastal tourism. However, a principal challenge posed in assessments of environmental impacts is to isolate the effect of interest from spatial and temporal variability (Underwood & Peterson, 1988;

Schmitt & Osenberg, 1996).

A range of ecological responses in fish have been attributed to sewage pollution including: increased mortality, increased or decreased abundance or diversity, and changes to size, reproduction, contaminant levels, parasites, infections or behaviour (Tsai, 1975; Love et al., 1987;Grigg, 1994,

1995; Siddall et al., 1994; Lemly, 1996). Many authors have investigated the impacts of sewage on fish abundance and/or condition (Adams, 1990; Gray et al, 1992; Otway, 1995; George et al., 1995;

Svanberg, 1996), but few have conducted field experiments. The field experiments have generally placed freshwater fish, such as salmonids and cyprinids, in cages adjacent to sewage outfalls and these fish generally suffered high mortalities (Tsai, 1975; Mitz & Giesy, 1985; Kakuta & Murachi, 1997).

The idea that a single taxon may be used to indicate broader impacts on the community is now widely accepted (Stephens et al., 1988; Jones & Kaly 1996). Two different types of species are generally used as indicators of pollution (Keough & Quinn, 1991): Positive indicators are opportunistic species that may occur in large numbers in polluted areas, either because environmental conditions become favourable for these species and/or detrimental to other species or because superior competitors are removed. Examples of such species include green algae, particularly Ulva spp (Fairweather, 1990;

Hardy et al., 1993; Bokn et al., 1996), the polychaete worm Capitella capitata (Schmitt & Osenberg,

1996) and carp Cyprinus carpio (Tsai, 1975). Negative indicators are species that are common in relatively unpolluted areas and decline in abundance in response to pollution. Examples are the seagrass Posidonia spp (Chisholm et al., 1997) and perhaps Chaetodontidae (Hourigan et al. 1988).

The eastern hulafish Trachinops taeniatus, Gunther 1861 (Plesiopidae), is a common rocky reef fish endemic to Australian waters from Noosa (Queensland) to Cape Conran (Victoria). It is a schooling, planktivorous species which attains a maximum total length of round 10 cm (Hutchins & Swainston,

1986). T. taeniatus has no direct value as a commercial or recreational food or game fish, although it is sought by collectors of aquarium fish. HUNTER

lOTomaree Head O Point Stephens

• Boulder Bay

SYDNEY

0 A North Head 1 Gap A Rosa Gully

New South Wales ;Y'0 A Gordons Bay

• A Potter Point

10 ILLAWARRA

*.*A O Flinders Island

• Port Kembla

O Atchinsons Rock

Figure 8.1. Study locations for surveys of T. taeniatus in the Hunter, Sydney and lllawarra regions. The locations in the Sydney region marked with A were also used for the condition and mortality studies. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 146

To investigate the effects of sewage discharge on T. taeniatus, I used a multidisciplinary approach that involved estimates of population abundance, measures of condition of individual fish, and a field experiment to measure mortality.

METHODS

Abundance of T. taeniatus

T. taeniatus were surveyed at three regions off the central coast of NSW (Figure 8.1) in similar subtidal habitats. All regions were surveyed using the same methodology and personnel. The time of surveys (between July 1992 and February 1998), and size and treatment of the outfalls, differed (TEL, 1994; Smith et ai, 1998). I therefore present results from the three regions as separate studies.

(1) Hunter region. Three locations were surveyed: A small outfall at Boulder Bay (Figure 8.1), discharging 4.3 ML/day of secondary treated sewage (MHL, 1997) and two controls sites at

Tomaree Head and Point Stephens,

(2) Sydney region. Four locations were surveyed: Two outfalls at Rosa Gully discharging 5.1

ML/day (raw sewage), and Potter Point discharging 46 ML/day (primary sewage) (Figure 8.1), and two controls at North Head and Gordons Bay.

(3) Illawarra region: An outfall at Port Kembla, discharging 16.8 ML/day (primary sewage) (Figure

8.1), and two controls at Flinders Island and Atchinsons Rock.

Fish were surveyed by SCUBA divers in the boulder or barrens habitat (Underwood et al.,

1991) at a depth range of 12 to 15 m. Within each location, fish were surveyed along four 60 metre-long transects. Each transect was laid parallel to the shoreline, and located within 100 m of the outfall. Fish were sampled during daylight hours during periods of low swell (less than 1 m) and when water visibility exceeded 5 m. T. taeniatus were counted one metre either side of the transect line (total area 120 m^) (Lincoln Smith, 1988; 1989). Schools of T. taeniatus CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 147 appeared to have a restricted home range and were not visibly attracted or repelled by a SCUBA diver (pers. obs.).

Data were tested for homogeneity of variance by Cochrans test, transformed if necessary, and analysed using Analysis of Variance.

Condition of T. taeniatus

For this component of the study, fish were collected in the Sydney region at two control locations:

North Head and Gordons Bay and two outfalls: Rosa Gully and Potter Point (Figure 8.1). Fish were not collected in the Hunter and Illawarra regions. Fish were collected by a SCUBA diver with two small

(30 cm diameter) hand-held dip nets on five occasions between February 1996 and February 1998. At least 100 fish were captured from each site at each time. Total lengths of all fish were measured in the laboratory on the same day they were collected using callipers (0.01 mm). For all fish collected in 1996 and December 1997 (n=5162) 1 recorded the number of gravid females and measured their gonad weights. Using a binocular microscope I counted the total numbers of eggs and measured the diameters of a subsample of twenty eggs from each female. Gonadosomatic Index (GSI) was calculated as gonad weight/total weight* 100. A small proportion of gravid female fish did not have distinct eggs. The sex of juveniles was not determined and male gonads were not weighed as the largest were less than 0.0 lg.

Data for total length, weight, GSI, egg number and egg diameter were tested for homogeneity of variance by Cochrans test, transformed to log(x+l) if necessary, and analysed using Analysis of

Variance. Significance of means were tested using Ryan’s test.

Mortality of T. taeniatus

Mortality was investigated by placing fish in cages at control and outfall locations in the Sydney region

(Figure 8.1). Cages were constructed according to a lantern cage design. A pilot study determined that this cage design appeared robust in coastal swells less than 2 m. Cages were supported by a 15 cm foam float and anchored by a 25 kg concrete block. The cages were constructed of 1 m deep cylinders CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 148 of 6 mm raschel netting which were supported by two stainless steel rings of 40 cm diameter. The netting was fastened at the top and bottom by a drawstring.

Two cages were deployed at each of two control and two sewage outfall locations (Figure 8.1) for approximately two weeks during both March 1996 and March 1997 (total of sixteen samples). Cages containing 15 Fish captured from the same location (Gap) were deployed in the rocky reef habitat in

water depths of approximately 12-15 m. Cages at the sewage outfall locations were deployed within 50

to 100 m of the outlet pipe and within the sewage plume. All cages were inspected on the day

following deployment and all fish were alive. Cages were retrieved after two weeks and fish were

counted.

Data on survivorship were tested for homogeneity of variance by Cochran’s test, transformed to

log(x+l) if necessary, and analysed using Analysis of Variance.

RESULTS

Abundance of T. taeniatus

A total of 1539 T. taeniatus was counted in the Hunter region. ANOVA indicated that there was a

significant location x period effect (Table 8.1). The abundance of T. taeniatus clearly declined by at

least 50% after commissioning of the sewage outfall at Boulder Bay (Figure 8.2a).

A total of 1220 T. taeniatus was counted in the Sydney region, and the total abundances of T. taeniatus

at the control locations were significantly greater by 300-600% than at the outfall location at Potter

Point (Table 8.2, Figure 8.2b). The abundance of T. taeniatus at the outfall location at Rosa Gully was

similar to the control location at Gordons Bay but significant less than the control location at North

Head (Figure 8.2b).

A total of 3547 T. taeniatus was counted in the Illawarra region. Greater abundances of fish were

recorded from the control locations by 300-500% compared to the outfall locations (Table 8.3, Figure

8.2c). Table 8.1. Summary of Beyond BACI analysis of variance of the abundance of Trachinops taeniatus from the Hunter region at 1 outfall and 2 control locations, 3 periods (before, immediately.after, 1 year after - 4 times in each period) and with 4 replicates, ns = p> 0.05; *=p< 0.05; **=p< 0.01; ***=p<

0.001.

Source of variation df MS F P

Period = P 2 2.78 11.31 * * * *

Time(Period) = T(P) 9 0.54 0.96 ns

Location = L 2 5.76 23.46 * * *

Outfall vs Control Locations = O v C 1 2.26 0.24 ns

Between Control Locations = C 1 9.26 38.92 *

PxL 4 0.63 2.55 *

PxOvC 2 0.74 1.44 ns

P x Between C 2 0.51 2.16 ns

T(P) x L 18 0.51 2.08 **

T(P) x O v C 9 0.57 1.23 ns

T(P) x Between C 9 0.46 1.87 ns

Residual 108 0.25 Table 8.2. Summary of analysis of variance of the abundance of Trachinops taeniatus from the Sydney region at 2 outfall and 2 control locations, 2 times and with 4 replicates, ns = p> 0.05; *=p< 0.05;

**=p<0.01; ***=p<0.001.

Source of variation df MS F P

Time 1 0.01 0.04 ns

Location 3 1.57 7.10 * * *

Time x Location 3 0.18 0.81 ns

Residual 24 0.22

Table 8.3. Summary of analysis of variance of the abundance of Trachinops taeniatus from the lllawarra region at 1 outfall and 2 control locations, 4 times and with 4 replicates, ns = p> 0.05; *=p<

0.05; **=p< 0.01; ***=p< 0.001.

Source of variation df MS F P

Time 3 2853 1.55 ns

Location 2 2687 1.46 ns

Outfall vs Control Locations = O v C 1 5148 22.9 ns

Between Control Locations = C 1 225 0.70 ns

Time x Location 6 1838 2.26 ns

TxOvC 3 3356 10.4 *

T x Between C 3 321 0.39 ns

Residual 37 813 (3) E3 Pre 40 - CD Post

Tomaree Point Stephens Boulder Bay

300 -

200 -

100 -

------,------1------1------1---- North Head Gordons Bay Rosa Gully Potter Point

150 -

100 -

Flinders Island Atchinsons Rock Port Kembla

Figure 8.2. Mean abundance (± SE) of T. taeniatus (per 120 m2) at outfall and control sites in the (a)

Hunter, (b) Sydney, and (c) Illawarra regions. Sewage outfalls are indicated by solid bars. Pre - pre- commissioning, post - post-commissioning of the sewage outfall in the Hunter region. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 149

Condition of T. taenia/us

From a total of 6657 T. taeniatus, 1 measured at least 100 fish from each of four locations and five times (Figure 8.3). The total lengths of all fish ranged from 13-98 mm, with a mean length of 29.6 mm

(+0.15 SE). One way ANOVA indicated that significantly larger fish were captured at the control location at Gordons Bay and significantly smaller fish were captured at the sewage outfall at Potter

Point (Table 8.4, Figure 8.4).

Relatively small sample sizes were obtained to investigate the impacts of sewage on the reproductive condition of fish. Only 25 ripe female T. taeniatus were collected from a total of 1495 fish (1.7 %) in

February-March 1996, and 58 ripe females were collected from a total of 1636 fish (3.5 %) in

December 1997. I therefore pooled the two temporal samples into outfalls and controls locations prior to analysis. There was no difference in the length, weight or GSI of gravid female T. taeniatus from outfall and control locations (Table 8.4). I detected significant differences in the total number of eggs and the egg diameter (Table 8.4). The total number of eggs per female was greater at outfall locations by approximately 15 % and the mean diameter of eggs was approximately 10% less at the outfall locations (Figure 8.5).

Mortality of T. taeniatus

Of the sixteen cages deployed, fourteen were retrieved and two were lost. The overall mortality of T. taeniatus was 19%; however this ranged from 0 - 73% mortality (15 to 4 fish) for individual cages

(Table 8.5). The greatest mortality was recorded from a cage at the Potter Point outfall. ANOVAs were conducted separately for the mortality in 1996 and 1997, and the combined years and these showed no significant effects, although p = 0.07 was near significant (Table 8.6). Also, the power of the analysis was low (0.36 for year and 0.19 for the interactions). Table 8.4. Summary of analysis of variance of condition (length and reproduction) indices for

Trachinops taeniatus at 2 outfall and 2 control locations in the Sydney region, ns = p> 0.05; *=p<

0.05; **=p< 0.01; ***=p< 0.001.

Source of variation df MS F P

LENGTH

Location 3 65537 675 * * *

Time 4 5453 56 * * *

Location x Time 12 2678 27 * * *

Residual 6637 97

REPRODUCTION

Length 1 135 3.55 ns

Residual 78 38

Weight 1 0.168 0.912 ns

Residual 78 0.184

Gonadosomatic Index 1 48.7 1.48 ns

Residual 78 33.0

Number of eggs 1 177197 4.05 *

Residual 59 43764

Diameter of eggs 1 0.819 38.6 ***

Residual 578 0.021 NORTH 111. AIJ U )KI)()NS BAY TO l l l.K I'UINT FEB 96 30.2 (1.1.91 ii = 1:7

MAR 1996 27.1 (0.61 n = 277 n=l 36

MAR 1997 2S.5(0.5) 29.4 (0.4) 34.7 (0 3) 22.8 (0.4) n=??8 n=704 n=511 11=432

DEC 1 997 21.3 (0.6) 30.6 (0.9) 41.9 (0.8) 19.3(0.1) n=311 n=269 11=227 n=829

FEB 98 46.0 (0.6) 25.4 (0.6) 11=273 n=486

10 21 31 41 SI 61 71 81 91 10 21 31 41 SI 61 71 81 91 10 21 31 41 SI 61 71 81 91

Figure 8.3. Length frequency distribution (total length - mm) of T. taeniatus from four locations in the

Sydney region and five times (5 mm increments). Note the scale of the y-axis varies. Figure five times.

8.4. JU •fE £ o re W) Average Mem 20 25 30 40 35 45 50 15 total

------

North length A q t a

Head

(±

SE) Rosa

of •

I q fi 6 Gully T.

taeniatus

Gordons

$ o $ from

Bay

four

Potter A o s o □

locations A 6 o Nov97 Mar97 6 Feb98 Mar96 Fcb96

Point

in

the

Sydney

region

and 14

-a c

c3 E 12- so ■§ c o 11- a § dj 2 10-

0.85-

0.83-

0.80-

0.78-

Controls Outfalls

Figure 8.5. Reproductive indices (± SE) of female T. taeniatus from control and outfall locations in the

Sydney region (a) mean Gonadosomatic index, (b) mean abundance of eggs, and (c) mean diameter of eggs. Table 8.5. Number of Trachinops taeniatus (out of a total of 15 per cage) which survived experimental caging at control and outfall locations in the Sydney region (x = missing cage).

Year 1996 1997

Cage number 1 2 1 2

Control North Head 11 15 13 11

Gordons Bay 9 15 14 X

Outfall Rosa Gully 13 12 15 14

Potter Point 4 X 11 13

Table 8.6. Analysis of variance of the survivorship of Trachinops taeniatus in cages. O - Outfall sites

(Potter Point and Rosa Gully), C - Control sites (North Head and Gordons Bay).

Source df MS F P

Year 1 24.2 4.68 0.07

OC 1 24.5 4.74 0.07

Year * OC 1 16.2 3.14 0.12

Site (OC) 2 16.4 3.19 0.11

Year * Site * OC 2 9.0 1.74 0.25

Residual 6 5.2 CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 150

DISCUSSION

Abundance of fish impacted by sewage

I found decreased abundances of T. taeniatus at the outfall sites compared to the control sites for the three regions. The scale of impact varied from approximately 50 to 600%, with a smaller impact at the small outfall in the Hunter region and larger impacts at the larger outfalls in the Illawarra and Sydney region. There was a large degree of variability in the abundance of T. taeniatus among control locations, particularly for the Hunter region, however, we used a beyond BACI experimental design

(Underwood, 1994) which allowed the detection of significant differences.

The increased or decreased abundances of fish adjacent to sewage outfalls may be site or species specific (Pilanowski, 1992; Grigg, 1994; Otway, 1995; Smith et al, 1998). Other studies have reported that sewage impacts on fish populations may result in a replacement of valuable species (gadoids, flatfish and lobsters in marine areas, and salmonids and sturgeon in freshwater/estuarine habitats), by lower-value small pelagic fish and perciforms respectively (Caddy, 1995). Decreased abundances of

Australian snapper Pagrus auratus and increased abundances of gurnard Lepidotrigla mulhalli were consistent patterns attributed to deepwater ocean outfalls, while the abundance of mosaic leatherjacket

Eubalichthys mosaicus increased at one outfall and decreased at one outfall off Sydney (Otway, 1995).

Condition of fish impacted by sewage

We found consistent patterns of smaller fish from the larger coastal outfall at Potter Point and significantly larger fish at a control location at Gordons Bay. Possible explanations for the smaller mean size of fish observed at the sewage outfall at Potter Point may be increased mortality, dispersal of larger fish away from the sewage, increase in the predation on larger T. taeniatus, or preferential recruitment. These explanations appear to be supported by the length frequency distributions of T. taeniatus (Figure 8.3), which show a large abundance of recently recruited T. taeniatus (less than 20 mm) at Potter Point at all times, and very low abundances of large fish.

Although only a small proportion of T. taeniatus were gravid females, the fish from the outfall locations had a greater mean egg number and smaller egg diameter. Other studies have reported a CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 151 range of impacts of sewage on fish reproductive condition (Donaldson, 1990; Waring et al., 1996; Lye et al., 1997). For example, a substantial proportion of the male flounders Platichthys flesus exposed to sewage effluent had malformed testes (Lye et al., 1997). Sand gobies Pomatoschistus minutus were exposec to sewage sludge and no significant effect was found on GSI, however there was a major reducticn in the number of eggs and larvae produced in the sludge-exposed population, which reflected a failure of some females to spawn (Waring et al., 1996).

Mortality of fish impacted by sewage

Most T. taeniatus survived the experimental manipulation of caging, however, a large proportion

(73%) died in one of the cages at the sewage outfall at Potter Point in 1996. Unfortunately, the loss of several cages reduced the power of the statistical tests, and no significant impact could be attributed to sewage pollution. The experimental design could have been improved if we had additional replicate cages to increase the power of the analysis.

The relatively short experimental periods of two weeks was selected because previous researchers had demons'.rated rapid mortality of fish exposed to sewage. For example, mortality of juvenile chinook salmon Oncorhynchus tshawytscha occurred at all sites within 4.4 km of a 1530 ML/d sewage treatment plant outfall which discharged to the Fraser River, British Columbia (Birtwell et al., 1983).

Mortality was rapid, and fish placed 2.2 km from the outfall died within 9 minutes as a result of low dissolved oxygen (Birtwell et al., 1983). Experimental caging studies also demonstrated that acute mortality of channel catfish Ictalurus punctatus occurred at sites 300 and 500 m downstream from a 76

ML/d sewage treatment plant outfall on the Flint River, Michigan (Mitz & Giesy, 1985). However, three species of rock fish were placed in cages in a sewage effluent zone in San Francisco Bay and no significant mortality could be demonstrated on fish after 96 hours of exposure to 1% effluent (Tsai,

1975). CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 152

Future research and management

T. taeniatus were considered to be a good organism to study because they are a resident rocky reef species which are abundant, easy to capture and their small size and planktivorous feeding were important for achieving the practical limitations of maintaining fish in cages.

Three avenues of research deserve attention in the light of these findings. The scale of impact on abundance of T. taeniatus varied according to the different regions and locations with declines from approximately 50 to 600% in comparison to control locations. Surveys of the abundance of this species have been undertaken at four ocean outfalls, and there are over 40 outfalls in

NSW waters. The cumulative impact of sewage discharge in NSW waters on the abundance of this species should be investigated.

Secondly, differences in condition (mean length and reproductive indices) of T. taeniatus at outfalls and control locations and the generality of these pattern compared to other, more distant and pristine locations should be investigated. The samples for the condition study were only collected in the Sydney region and the selection of control sites may be partially confounded by the impact of diffuse sources of pollution, such as stormwater. Diffuse sources of pollution were not measured but were likely to be considerable for the site at North Head at the entrance to

Sydney Harbour, and this may explain why the length of fish from this control location were significantly smaller than the control location at Gordons Bay. T. taeniatus had more eggs and smaller eggs at outfall locations and laboratory experiments could be initiated to determine spawning success.

Thirdly, the use of bioindicators or sentinel animals in cages to monitor pollution has given rise to the establishment of national and international programmes (Phillips & Segar, 1986; Avery et al., 1996;

Jones & Kaly, 1996). To date these programs have focused on using oysters and mussels in Australian waters (Scanes et al., 1995; Avery et al., 1996), although fish have been used in overseas studies (Tsai

1975; Mitz and Giesy, 1985, Stephens et al., 1988). Although the T. taeniatus experiments did not detect a significant impact on mortality, I suggest that this may be partially due to the low power of the CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 153

experimental design and the loss of replicate cages. It is logical to monitor a species that is very

abundant in the coastal waters of NSW, in preference, or in addition to the existing monitoring

programs which transplant an estuarine species, the Sydney Rock oyster Saccostrea commercialis into

a coastal environment.

In conclusion, this research detected different scales of impact of sewage on T. taeniatus with

approximately 50 - 600% declines in abundance, increases and decreases of 10 - 50% in

condition indices (length, reproduction) although no detectable impact on mortality. While other

studies have focused on the effects of sewage on fish abundance (Gray et al., 1992; Otway et

al., 1996), condition (Waring et al., 1996; Kakuta & Murachi, 1997; Lye et al., 1997) or

mortality (Lemly, 1996), this is one of the few studies to have used a multidisciplinary approach

to investigate the impacts of sewage pollution (but see Liu & Morton, 1998). Similar

multidisciplinary studies should be undertaken on other species of fish and biota that are

exposed to sewage pollution. The identification of the effects of sewage on T. taeniatus

increases the choice of possible bioindicators in Australian waters.

REFERENCES

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium 8,

Bethesda, Maryland. 191 pp.

Avery, E.L., Dunstan, R.H. & Nell, J.A. (1996). The detection of pollutant impact in marine

environments: condition index, oxidative DNA damage, and their associations with metal

bioaccumulation in the Sydney rock oyster Saccostrea commercialis. Arch. Environ. Contain.

Toxicol. 31, 192-198.

Baird, R.C. (1996). Towards new paradigms in coastal resource management: linkages and institutional

effectiveness. Est. 19, 320-335.

Birtwell, I.K., Greer, G.L., Nassichuk, M.D. & Rogers, I.H. (1983). Studies on the impact of municipal

sewage discharged onto an intertidal area within the Fraser River estuary, British Columbia. Can.

Tech. Rep. Fish. Aquat. Sci. 1170. 64 pp. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 154

Bokn, T.L., Frithjof, E. & Walday, M. (1996).' Improvement of the shallow water communities

following reductions of industrial outlets and sewage discharge in the Hvaler estuary, Norway.

Hydrobiologia 326/327, 297-304.

Caddy, J.F. (1995) Towards a comparative evaluation of human impacts on fishery ecosystems of

enclosed and semi-enclosed seas. In Condition of the world’s aquatic habitats, ed. N.B.

Armantrout, Proceedings of the World’s Fisheries Congress, Theme 1. Science Publishers Inc.,

Lebanon, USA. pp. 76-87.

Chisholm, J.R.M., Fernex, F.E., Mathieu, D. & Jaubert, J.M. (1997). Wastewater discharge, seagrass

decline and algal proliferation on the Cote d’Azur. Mar. Poll. Bull. 34, 78-84.

Donaldson, E.M. (1990). Reproductive indices as measures of the effects of environmental stressors in

fish. In Biological Indicators of Stress in Fish. ed. S.M. Adams. American Fisheries Symposium

8, Bethesda, Maryland, pp. 109-122.

Dubinsky, Z. & Stambler, N. (1996). Marine pollution and coral reefs. Global Change Biology 2, 511 -

526.

Earle, S.A. (1995). Sea change. A message of the oceam. G.P. Putnam’s Sons, New York. 354 pp.

Fairweather, P.G. (1990). Sewage and the biota of seashores: assessment of impact in relation to

natural variability. Environ. Monit. Assess. 14, 197-210.

George, S.G., Wright, J. & Conroy, J. (1995). Temporal studies of the impact of the Braer oilspill on

inshore feral fish from Shetland, Scotland. Arch. Environ. Contam. Toxicol., 29, 530-534.

Gray, C.A., Otway, N.M., Laurenson, F.A., Miskiewicz, A.G. & Pethebridge, R.L. (1992). Distribution

and abundance of marine fish larvae in relation to effluent plumes from sewage outfalls and depth

of water. Mar. Biol. 113, 549-559.

Grigg, R.W. (1994). Effects of sewage discharge, fishing pressure and habitat complexity on coral

ecosystems and reef fishes in Hawaii. Mar. Ecol. Prog. Ser. 103, 25-34.

Grigg, R.W. (1995). Coral reefs in an urban embayment in Hawaii: a complex case history controlled

by natural and anthropogenic stress. Coral Reefs 14, 253-266.

Hardy, F.G., Evans, S.M. & Tremayne, M.A. (1993). Long-term changes in the marine macroalgae of

three polluted estuaries in north-east England. J. Exp. Mar. Biol. Ecol. 172, 81-92. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 155

Hourigan, T.F., Tricas, T.C. & Reese, E.S. (1988). Coral reef fishes as indicators of environmental

stress in coral reefs. In Marine Organisms as Indicators, eds. D. Soule & G.S. Kleppel. Springer-

Verlag, New York. pp. 107-35.

Hutchins, B. & Swainston, R. (1986). Sea fishes of Southern Australia. Swainston Publishing, Perth.

180 pp.

Jones, G.P. & Kaly, U.L (1996). Criteria for selecting marine organisms in biomonitoring studies. In

Detecting Ecological Impacts: Concepts and Applications in Coastal Habitats, eds. R.J. Schmitt

& C.W. Osenberg. Academic Press, San Diego, pp. 29-48.

Kakuta, I. & Murachi, S. (1997). Physiological response of carp, Cyprinus carpio, exposed to raw sewage

containing fish processing wastewater. Environ. Toxicol. Wat. Qual. 12, 1-9.

Keough, M.J. & Quinn, G.P. (1991). Causality and the choice of measurements for detecting human

impacts in marine environments. Aust. J. Mar. Freshwater Res. 42, 539-554.

Koop, K. & Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution. Mar. Poll.

Bull. 33, 121-123.

Lerr.ly, A.D. (1996). Wastewater discharge may be most hazardous to fish during winter. Environ.

Poll. 93, 169-174.

Lincoln Smith, M.P. (1988). Effects of observer swimming speed on sample counts of temperate rocky

reef fish assemblages. Mar. Ecol. Prog. Ser. 43, 223-231.

Lincoln Smith, M.P. (1989). Improving multispecies rocky reef fish censuses by counting different

groups of species using different procedures. Environ. Biol. Fish. 26, 29-37.

Liu, J.H. & Morton, B. (1998). The impacts of pollution on the growth, reproduction and population

structure of Hong Kong limpets. Mar. Poll. Bull. 36, 152-158.

Love, M.S., Axell, B., Morris, P., Collins, R. & Brooks, A. (1987). Life history and fishery of the

Californian scorpionfish, Scorpaena guttata, within the southern Californian Bight. Fish. Bull. 85,

99-116.

Lye, C.M., Frid, C.L.J., Gill, M.E. & McCormick, D. (1997). Abnormalities in the reproductive health

of flounder Platichthys flesus exposed to effluent from a sewage treatment works. Mar. Poll. Bull.

34, 34-41. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 156

Mitz, S.V. & Giesy, J.P. (1985). Sewage effluent biomonitoring: 1. Survival, growth, and

histopathological effects in channel catfish. Ecotoxicol. 10, 22-39..

MHL (1997). Review of Sewage Effluent Ocean Outfall Performance in NSW. Manly Hydraulic

Laboratory Report 769, Sydney. 65 pp.

Otway, N.M. (1995). Assessing impacts of deepwater sewage disposal: a case study from New South

Wales, Australia. Mar. Poll. Bull. 31, 347-354.

Otway, N.M., Gray, C.A., Craig, J.R., McVea, T.A. & Ling, J.E. (1996). Assessing the impacts of

deepwater sewage outfalls on spatially- and temporally -variable marine communities. Mar.

Environ. Res. 41,45-71.

Pearson, T.H. & Rosenberg, R. (1978). Macrobenthic succession in relation to organic enrichment and

pollution of the marine environment. Oceanog. Mar. Biol. Ann. Rev. 16, 229-311.

Phillips, D.J.H. & Segar, D.A. (1986). Use of bio-indicators in monitoring conservative contaminants:

program design imperatives. Mar. Poll. Bull. 17, 10-17.

Pilanowski, R.A. (1992). The effects of ocean disposal of sewage sludge on the relative abundance of

benthic megafauna. Chem. Ecol 6, 199-212.

Schmitt, R.J. & Osenberg, C.W. (1996). Detecting Ecological Impacts: Concepts and Applications in

Coastal Habitats. Academic Press, San Diego.

Siddall, R., Pike, A.W. & McVicar, A.H. (1994). Parasites of flatfish in relation to sewage sludge

dumping. J. Fish Biol. 45, 193-209.

Smith, A.K. (1997). Is the eastern hula a potential indicator of sewage pollution? Abstract from 23rd

Annual Workshop of Australian Society for Fish Biology and Fish and Aquatic Resource

Management Association of Australia, Darwin, NT, Australia, p 41.

Smith, A.K., Ajani, P.A. & Roberts, D.E. (1998). Spatial and temporal variation in fish assemblages

exposed to sewage and implications for management. Mar. Environ. Res. 47, 1-20.

Stephens, J.S, Jr., Hose, J.E. & Love, M.S. (1988). Fish assemblages as indicators of environmental

change in nearshore environments. In Marine Organisms as Indicators eds. D.F. Soule & G.S.

Kleppel. Springer-Verlag, New York. pp. 91-105.

Svanberg, O. (1996). Monitoring of biological effects. Resources, Conservation and Recycling 16,

351-360. CHAPTER 8 EFFECTS OF SEWAGE ON HULAFISH 157

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final Report by the Ecology Lab for the Water Board’s Environmental

Management Unit, Sydney. 29 pp.

Tsai, C. (1975). Effects of sewage treatment plant effluents on fish: a review of literature. CRC

Publication No. 36, University of Maryland, America. 229 pp.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15.

Underwood, A.J., Kingsford, M.J. & Andrew, N.L. (1991). Patterns in shallow subtidal marine

assemblages along the coast of New South Wales. Aust. J. Ecol. 6, 231-249.

Underwood, A.J. & Peterson, C.H. (1988). Towards an ecological framework for investigating

pollution. Mar. Ecol. Prog. Ser. 46, 227-234.

Waring, C.P., Stagg, R.M. & Fretwell, K., McLay, H. and Costello, M.J. (1996). The impact of sewage

sludge exposure on the reproduction of the sand goby, Pomatoschistus minutus. Environ. Poll. 93,

17-25. CHAPTER 9 GENERAL CONCLUSIONS 158

CHAPTER 9

GENERAL CONCLUSIONS

The general conclusions in regard to the null hypotheses (Chapter 1) that were tested in this thesis were:

Study 1: There is no difference in the abundance of fish assemblages at sewage outfall and control

locations (Chapters 3, 4 and 5).

General conclusions: The null hypothesis was rejected. Quantitative underwater surveys of fish

assemblages at sewage outfalls and control locations indicated significant effects of sewage on fish

assemblages and individual species at three different regions (Illawarra, Hunter and Sydney). These

surveys also identified potential ‘indicator’ species.

Study 2: There is no difference in the biomass of fish assemblages at sewage outfall and control

locations (Chapter 6).

General conclusions: The null hypothesis was rejected. Biomass measures differentiated between

outfall and control locations. If biomass is used instead of abundance measures, large species

accounted for most differences between outfall and control locations.

Study 3: There is no difference in the condition of Cheilodactylus fuscus, Parma microlepis and

Trachinops taeniatus collected at sewage outfall and control locations (Chapters 7 and 8).

General conclusions: The null hypothesis was accepted for all species for some condition indices and

rejected for others. Changes in condition included increased or decreased body size, liver health,

Gonadosomatic index, egg number and egg size. GSI of C. fuscus and P. microlepis was greater at

outfall locations. T. taeniatus collected at outfalls contained more eggs and smaller diameter eggs than

fish from control locations. However, some significant effects were found at only one location. T.

taeniatus collected from one outfall location were consistently smaller than fish from other locations. CHAPTER 9 GENERAL CONCLUSIONS 159

The study of condition of fish in the Sydney region may be confounded by other pollutants at ‘control’ sites.

Study 4: There is no difference in the mortality of Trachinops taeniatus deployed in cages at sewage outfall and control locations (Chapter 8).

General conclusions: The null hypothesis was accepted. No statistically significant mortality from sewage pollution was detected. However, experimental manipulation and caging of T. taeniatus was logistically difficult in the dynamic coastal environment and there were some statistical constraints.

Further information on the conclusions as well as a proposed model of the impacts of sewage on temperate reef fish and recommendations for future research and management are provided below:

SURVEYS OF THE ABUNDANCE OF FISH ASSEMBLAGES

Baseline information on the impact of sewage on temperate rocky reef fish assemblages was collected in three regions; Hunter, Sydney and Illawarra along the central coast of NSW. The outfalls in the lllawarra region contained moderate sized outfalls which discharged primary treated effluent, the outfall in the Hunter region was small and discharged secondary treated effluent and the large outfall in the Sydney region discharged primary treated effluent until it was shutdown. The first study was in the

Sydney region (1991-92), followed by the Illawarra region (1992-1993) and the Hunter region (1993-

95). The timing of these three surveys depended on a number of factors including; fieldwork logistics, construction of the deepwater outfalls and shutdown of the shoreline outfalls, and construction of the outfall at Boulder Bay.

Over 130 species of fish and 200,000 individuals were recorded by underwater visual census. Fish assemblages were generally significantly different at outfalls compared to control locations. Although fish were the main focus of this study, selected macrobenthic organisms such as kelp Ecklonia radiata and the green alga Caulerpa filiformis were also surveyed in the Illawarra region and the sea urchin

Centrostephanus rodgersii was surveyed in the Hunter region. There was no ‘before’ information available at two of the regions, Sydney and Illawarra. However, it was during the period of my CHAPTER 9 GENERAL CONCLUSIONS 160 research that major advances were being made in designing studies to detect environmental impacts and a BACI design (Before After Control Impact) (Underwood 1991, 1992, 1993, 1994, 1997) was

used for the final study of fish assemblages in the Hunter region. The results indicated a significant

‘press disturbance’ with the turning-on of the outfall, and no such changes were observed at the control

locations.

Another significant question is what happens to fish assemblages when a sewage outfall is shutdown.

Fish assemblages sampled on the day that the sewage outfall at North Head was shutdown and during

the following year indicated that very large numbers of some species of fish occurred at the outfall

(Chapter 5). The abundances of yellowfin bream Acanthopagrus australis were up to 100 times more

abundant at the outfalls immediately after decommissioning but declined rapidly over a period of

weeks. The fish assemblages at the previous sewage outfall at North Head were still significantly

different to control locations after one year, which in the absence of before information may be

interpreted as site specific differences or alternatively longer-term impacts may have occurred.

Qualitative observations of fish at North Head during 1992-1997 indicated that large kelp beds

recolonised and fish assemblages were similar to adjacent areas, supporting the theory of a recovery

period of greater than 1 year.

Broad generalisations on the impact of sewage on the abundance and biomass of fish assemblages in

NSW were investigated using multivariate analyses of a subset of data from the three regions (Chapter

6). This synthesis of information from a number or regions is potentially the most important part of the

thesis because some general patterns such as sewage outfalls contained fewer fish but a greater biomass

were found

The abundance and biomass measures gave a similar result of significant differences between outfall

and control locations, but a different interpretation of which ‘indicator’ species are most important in

generating these differences. Large abundances of small, schooling species such as yellowtail

Trachurus novaezelandiae, hulafish Trachinops taeniatus and mado Atypichthys strigatus were mainly

responsible for these differences. If biomass is used instead of abundance measures, large species such CHAPTER 9 GENERAL CONCLUSIONS 161 as blue groper Achoerodus viridis, yellowfin bream Acanthopagrus australis, red morwong

Cheilodactylus fuscus, yellowtail kingfish Seriola lalandi and mulloway Argyrosomus japonicus accounted for most differences between outfall and control locations. Analyses designed to give greater emphasis to abundant species (raw data), or rare species (double square root transformation), indicated that Cheilodactylus fuscus, Acanthopagrus australis, Trachurus novaezelandiae and Trachinops taeniatus are potential indicator species. Only one species of fish, C. fuscus has previously been described as a potential ‘indicator’ (Lincoln Smith and Mann 1989) of sewage pollution in NSW waters. Instead, scientists have used other taxa such as green algae, polychaetes and the Sydney rock oyster. A number of constraints were apparent including the different years for sampling, the different physical characteristic of the sewage outfalls in the three regions (habitat, volumes and levels of treatment of sewage effluent) and the estimation of fish biomass.

CONDITION OF THREE SPECIES OF FISH

A suitable indicator species must be affected by sewage effluent, common, resident and easy to sample.

Using these criteria four species were initially selected for further investigation; red morwong

Cheilodactylus fuscus, half-banded seaperch Hypoplectrodes mccullochi, white ear Parma microlepis and hulafish Trachinops taeniatus. Hypoplectrodes mccullochi is difficult to capture in large numbers and this species was not considered further.

Cheilodactylus fuscus, Parma microlepis (Chapter 7) and Trachinops taeniatus (Chapter 8) are large, moderate and small sized fish, respectively. Different sampling strategies were used and the first two species were collected at multiple outfall and control locations in the Sydney region but at only one time (20 per location). T. taeniatus were easily collected in large numbers (over 100 per location) at five separate times. The variables used by Adams et al. (1993) and Goede and Barton (1990) for rapidly assessing general fish health in field situations were adapted for this study. A range of measurements and indices such as length, weight, girth, Gonadosomatic index (GSI) and

Hepatosomatic index were made on C. fuscus and P. microlepis. The same range of measurements and

GSI were made on the small fish (Trachinops taeniatus), however, the weights of small organs such as liver and stomach were not measured. CHAPTER 9 GENERAL CONCLUSIONS 162

A number of international and Australian studies have reported a strong relationship between chemical contamination (from sewage effluent) and decreased body size, increased prevalences of liver disease

(Collier et al. 1998), increased liver size (Lincoln Smith and Mann 1987, Lye et al. 1997), and a wide range of reproductive impairment in fish (Waring et al. 1996, Collier et al. 1998). Similar ‘negative’ relationships were found for one or more of the species of fish in this study which suggests an impact from sewage effluent, however, ‘positive’ or no significant relationships were also found and these were inconsistent between the three species of fish. Trachinops taeniatus were consistently smaller and

Cheilodactylus fuscus and Parma microlepis were larger at sewage outfalls. P. microlepis had a greater proportion of‘unhealthy’ liver tissue but this pattern was not significant for C. fuscus. The GS1 of female C. fuscus and P. microlepis was larger at sewage outfalls. Female T. taeniatus did not have a significantly different GS1 at outfall locations, however, fish contained a larger number of eggs which were a smaller diameter than fish from control locations.

Biological information on Cheilodactylus fuscus collected in the late 1980s was compared with the findings in this study. The liver index of Cheilodactylus fuscus appears to have increased by 10% since the last decade, possibly due to long-term declines in water quality in the Sydney region. This trend requires further data for confirmation and may be confounded by time or location.

One of the possible limitations of this study was the selection of control locations within the Sydney region. Previous authors (Krogh and Scanes 1996, Nowak 1996) have suggested that Cheilodactylus fuscus in the Sydney region may be widely contaminated by pollution, which appears to be supported by the ecological findings in this study.

EXPERIMENTAL MANIPULATION OF ONE SPECIES OF FISH

There are often complex sets of physical and biological processes acting together to determine the abundance of any local population. Community change may result from unknown factors, such as specific contaminants (Austen et al. 1994). One approach that is commonly used for detecting the impact of a particular process is to use manipulative experiments such as the use of caging. The use of CHAPTER 9 GENERAL CONCLUSIONS 163 sentinel animals in cages to monitor pollution impacts has been successfully used for oysters and mussels (Scanes et al. 1995; Avery et al. 1996). However, there has been a lack of manipulative studies on rocky reef fish due to the difficulty in sampling fish because of the harsh physical environment and large temporal variability and mobility of fish (Choat 1995) and the large scale of manipulative field experiments (Carr 1989, 1991).

An experimental approach designed to investigate the potential mechanism for the decreased abundances of Trachinops taeniatus at sewage outfalls involved transplanting fish into cages and placing them at outfall and control locations (Chapter 8). The small size, schooling behaviour and planktivorous feeding of the Trachinops taeniatus were a good basis to conduct pilot studies using cages. Although 80% of the fish survived the experiment, there were some constraints including the loss of a number of cages, short time of the experiment and a low statistical power due to few and lost replicates. Mortalities of up to 73% of the Trachinops taeniatus per cage were recorded at one outfall location but there was no overall significant mortality due to sewage.

A MODEL OF THE IMPACTS OF SEWAGE ON TEMPERATE REEF FISH

A number of models designed to explain the impacts of sewage on aquatic biota have been proposed over the past 60 years (see Chapter 2). Most of these models have been based on zones of distribution of several species of freshwater fish and have limited quantitative value. A model proposed by Pearson and Rosenberg (1978) (PRM) is more theoretical and is based on changes in species, abundance and biomass of macrofauna compared to increasing polluted conditions. The PRM has been criticised for its limited applicability (Maurer et al. 1993). Some data on reduced abundance of fish species at different distances from sewage outfalls (500 m and 2 km) may be consistent with the model (EPA

1996). Therefore the PRM was compared to data from this study (Table 9.1).

Species, abundance and biomass of rocky reef fish from ‘control’ and outfall locations from three study regions are graphed in the same format as the PRM (Figure 9.1). A number of assumptions were made: Table 9.1. Percentages change in species, abundance and biomass (kg) of fish assemblages from three regions and three categories of pollution. ‘Unpolluted’ refers to control locations. ‘Slightly polluted’ refers to outfall locations in the lllawarra and Hunter regions and control locations in the Sydney region.

‘Polluted’ refers to outfall locations in the Sydney region. S - Species, A- Abundance, B - Biomass. All times are in summer; time 12 is one year after the Hunter outfall was turned on, time 1 is immediately after the Sydney outfall was shutdown. * All numbers are means except for the single outfall location in the Hunter region.

POLLUTION CATEGORY

Unpolluted Slightly Polluted Grossly

REGION Polluted Polluted

HUNTER S 31 S 22 * (71%)

(time 12) A 337 A 144* (43%)

B 87 B 26 * (31%)

ILLAWARRA S 32 S 34 (106%)

(time 3) A 1484 A 621 (42%)

B 90 B 70 (78%)

SYDNEY S 51 S62 (121%)

(time 1) A 15832 A 11392 (72%)

B 1539 B 2592 (168%) Percentage change

igure NORMAL

abundance NORMAL Total Number I

of 9.1.

ILLAWARRA HUNTER pollution. Abundance

Biomass

Models (from

of (A

Species O) Pearson

of Model

and types and Increased TRANSITION TRANSITION Increasing

biomass and at

of

top

impacts Rosenberg

of Abundance I

(B

page

A)

I on Pollution

of is

biological 1978).

a fish

conceptual SYDNEY

assemblages Pollution

Model POLLUTED POLLUTED

communities Species

at model Indicator

bottom

from I

of

species, with three of

page GROSSLY

increasing

regions

abundance graphs No

Macrofauna

(see

POLLUTED data distance

Table

and

on

species

biomass

from 9.1

for

a

(S

of

CHAPTER 9 GENERAL CONCLUSIONS 164

(1) Pollution categories. Normal is generally a category for control locations. Control locations in the

Hunter and lllawarra regions'were placed in this category. However, control locations in the Sydney region were impacted by other sources of pollution (Krogh and Scanes 1996, Nowak 1996) and were therefore placed in the transition category. The Hunter outfall is small and has secondary treatment and was therefore categorised as transition. The lllawarra outfalls were relatively small compared to

Sydney, had few significant impacts and therefore were also regarded as transition. The Sydney outfall was larger and received primary treatment and was categorised as polluted. No outfalls in this study were categorised as grossly polluted (Table 9.1). A limitation of the model is the pollution categories are relatively subjective are not related to distances from a single pollution source.

(2) Data from a number of studies have been used and they are confounded by time and location. To minimise seasonal bias, data were chosen from summer samples. The Hunter outfall was a new outfall and time 12 (the last sampling period) was probably the most impacted as the outfall had been operating for the longest time. The Sydney outfall was shutdown, how'ever, time 1 was probably the most impacted from sewage pollution

A comparison of the PRM with the data from this study indicates that there are few similarities (Figure

9.1). The PRM indicates large increases and decreases in species, abundance and biomass as a result of pollution. The data on fish suggests that the number of species does not change much from sewage pollution (+ 25%), abundance declines consistently (30 to 60%), and biomass may be the most variable

(+ 70%).

A comparison of individual variables reveals further information:

Species richness in the PRM is low at normal conditions, then increases to a maximum in transition before declining. The species of fish is similar for normal, transition and polluted conditions, although there is a decline of about 25% at one transition outfall and an increase of 20% at a polluted outfall. CHAPTER 9 GENERAL CONCLUSIONS 165

Abundance in the PRM is very low at normal conditions, increases slightly in transition and increases considerably in polluted conditions. The percentage of fish abundance declines consistently at the outfall compared to control locations. The values for the Hunter and Illawarra studies are very similar.

Biomass of species in the PRM parallels the pattern for species in the PRM with an initial increase

followed by a rapid decrease. The biomass of fish appears to be the opposite pattern with an initial

decrease followed by a rapid increase.

In summary, Figure 9.1 suggests the following: the number of species of fish is the most resilient

variable to pollution, while abundance is likely to decline with increasing pollution, and biomass may

decline initially but may then increase (as large individuals dominate). The PRM reflects a dominance

of a few benthic species, which have very large abundances and a small average biomass in polluted

conditions. These characteristics of the PRM indicate very different responses of macrofauna and fish

to sewage pollution and the next logical approach would be to test additional data on fish species to see

if these trends apply at other sewage outfalls.

FUTURE RESEARCH AND MANAGEMENT

The literature review included research from throughout the world, although there was a focus on fish

and sewage in NSW coastal waters. The review indicated that research on sewage is voluminous (over

one hundred papers per year), however, it is often qualitative, specific to an area or issue or suffers

from poor experimental design. Well designed research which has detected (or did not detect) impacts

that could be attributed to sewage was identified which will be of great benefit to other researchers and

managers who are interested in the effects of sewage on aquatic ecosystems.

Fish are regarded as good bioindicators of sewage pollution because they integrate the effects of many

biotic and abiotic variables, are easy to identify, and are of commercial, recreational and public interest

(Adams 1990, Harris 1995). The mobility of some species of fish may limit the degree to which the

spatial distribution of pollutant effects can be resolved (TEL 1993, 1994). Most research on sewage has

been short-term and limited in scope. CHAPTER 9 GENERAL CONCLUSIONS 166

The previous research on the effects of sewage on fish in NSW waters has focused on bioaccumulation and/or the effects of the deepwater outfalls off Sydney. There has been limited research on the effect of sewage on fish in NSW waters (11 studies from 35 outfalls, MHL (1997)). The level and type of research is probably reasonable when compared to other states and countries. Significant effects have been detected at most (9) of these studies in NSW. Recommendations are made for future standardised research to detect the impacts of sewage on aquatic habitats, fish and fisheries to facilitate comparisons between studies (see Chapter 2).

It is strongly recommended that research on fish should be undertaken at all ocean outfalls in NSW waters. In addition, the baseline surveys for the three regions, Hunter, Sydney and Illawarra should be repeated every 5-10 years to enable investigation of long-term impacts.

Studies on the effect of sewage on fish assemblages in NSW waters should present separate analyses for the following potential indicator species:

Common name Scientific name hulafish Trachinops taeniatus yellowtail Trachurus novaezelandiae half-banded sea perch Hypoplectrodes mccullochi red morwong Cheilodactylus fuscus blue groper Achoerodus viridis bream Acanthopagrus australis sea urchin Centrostephanus rodgersii

The limitation of research on biomass was that data were collected on fish abundance without detailed length information (compare to Samoilys 1997) and therefore I estimated average biomass for temperate reef species. Some published data were available on the average biomass of commercially and recreationally important species and this was comparable to my biomass estimates. CHAPTER 9 GENERAL CONCLUSIONS 167

It is recommended that future underwater visual surveys should record size categories of fish and that these categories should be based on estimated length. The discrimination of categories of small juvenile, juvenile and adult fish has been recorded for several previous surveys in NSW waters (Lincoln Smith et al. 1991, 1992). For example, surveys of Trachinops taeniatus should discriminate small juveniles if less than 20 mm (which are transparent), juveniles if between 21 to 40 mm, and adults if greater than 41 mm. In some cases, the length of fish can be estimated to

+ 1 cm (Jennings and Polunin 1996, Samoilys 1997). The collection of size category information would allow a more accurate estimation of recruitment and biomass.

Studies on the condition of three species of fish, Cheilodactylus fuscus, Parma microlepis and

Trachinops taeniatus should also be extended to other sewage outfalls in NSW waters. Fish should be collected in summer to allow comparison with the data collected in this thesis. Because there were differences in the condition of male and female fish an improved sampling design would collect 20 male and 20 female fish for each species. An alternative would be to collect only female fish, given that the GSI of male fish was not affected by sewage pollution. The sex of all three species can be readily determined during collection: male Cheilodactylus fuscus have larger cranial horns, female

Parma microlepis are brown wheras the males are black, and ripe female Trachinops taeniatus have large orange gonads that can be observed through the white skin.

Future research should be conducted on Cheilodactylus fuscus at North Head to determine if these fish are impacted by pollution from Sydney Harbour or whether other factors such as habitat or age are responsible for the relatively smaller size of fish at this site. Liver health of Parma microlepis was very low at the outfall locations and should also be investigated. The liver index (Lincoln Smith and Mann

1989, Andrijanic 1991) is probably not as accurate a measure as HSI, because the liver index is a ratio of body weight which may vary according to gonad stage and stomach contents, and HSI it is a ratio of gutted weight. However, the long-term data on the liver index of Cheilodactylus fuscus indicate an increase over time and therefore this index should continue to be measured. CHAPTER 9 GENERAL CONCLUSIONS 168

The use of sentinel fish in cages in the dynamic coastal environment was difficult and is not recommended for future research unless the technique deploys caged fish in sheltered coastal bays or estuaries. The existing ‘oyster watch’ program uses the Sydney rock oyster Crassostrea commercialis and has demonstrated that this is a robust species for monitoring sewage pollution in the coastal environment. Future research will probably continue to rely on oysters rather than fish as sentinel organisms.

The management of sewage pollution is of prime importance for the health of coastal cities (Koop and

Hutchings 1996). The scientific studies in this thesis suggest that the sewage outfalls in the Hunter,

Sydney and Illawarra regions are resulting in a significant environmental impact on fish assemblages, condition of selected species and perhaps increased mortalities. Therefore it is disappointing that the

NSW Governments’ previous stance of‘no more ocean outfalls’ has been diluted to ‘future sewage outfalls will be assessed on a case by case basis’ (EPA 1997). I would argue that a more prescriptive approach advocating that no sewage disposal to enclosed waterbodies, marine parks, aquatic reserves, recognised fishing grounds or aquaculture leases or the habitat of endangered fish species or seagrasses should be considered. The interests of human aquatic resource users such as fishers, divers, and most importantly the aquatic biota, are often negatively impacted by sewage and this may result in ecosystem change, enormous economic cost, human illness and death.

REFERENCES

Adams, M. S. (1990). (Ed). Biological Indicators of Stress in Fish. American Fisheries Symposium

No. 8, Bethesda, Maryland.

Adams, M. S, Brown, A.M., Goede, R.W. (1993). A quantitative health assessment index for rapid

evaluation of fish condition in the field. Trans. Am. Fish. Soc. 122, 63-73.

Andrijanic, S. (1991). Organochlorine residues in fish: bioaccumulation in red morwong

(Cheilodactylus fuscus) in the Sydney region. Progress Report. Sydney Water. 15 pp.

Avery, E.L., Dunstan, R.H., Nell, J.A. (1996). The detection of pollutant impact in marine

environments: condition index, oxidative DNA damage, and their associations with metal CHAPTER 9 GENERAL CONCLUSIONS 169

bioaccumulation in the Sydney rock oyster Saccostrea commercialis. Arch. Environ. Contam.

Toxicol. 31, 192-198.

Austin, M.C., McEvoy, A.J., Warwick, R.M. (1994). The specificity of meiobenthic community

responses to different pollutants: results from microcosm experiments. Mar. Poll. Bull. 28, 557-

563.

Carr, M.H. (1989). Effects of macroalgal assemblages on the recruitment of temperate zone reef fishes.

J. Exp. Mar. Biol. Ecol. 126, 59-76.

Carr, M.H. (1991). Habitat selection and recruitment of an assemblage of temperate zone reef fishes. J.

Exp. Mar. Biol. Ecol. 146, 113-137.

Choat, J.H. (1995). Long-term vs short-term studies. Long-term monitoring programs in sub-tidal

environments; a problem with fishes. Abstract from Third International Temperate Reef

Symposium. University of Sydney February.

Collier, T.K., Johnson, L.L., Stehr, C.M., Myers, M.S., Stein, J.E. (1998). A comprehensive assessment

of the impacts of contaminants on fish from an urban waterway. Mar. Environ. Res. 46, 243-247.

EPA (1996). Sydney Deepwater Outfalls Environmental Monitoring Program. Final Report Series.

Volume 1. Environment Protection Authority, Sydney. 194 pp.

EPA (1997b). Public Inquiry into the Management of Sewage and Sewage By-products in the NSW

Coastal Zone. Final report prepared by the Environment Protection Authority, Sydney. 299 pp.

Goede, R.W., Barton, B.A. (1990) Organismic indices and an autopsy-based assessment as indicators

of health and condition of fish. In Biological Indicators of Stress in Fish. ed. M. S. Adams, pp.

93-108. American Fisheries Symposium 8, Bethesda, Maryland.

Harris, J.H. (1995). The use of fish in ecological assessments. Aust. J. Ecol. 20, 65-80.

Jennings, S., Polunin, N.V.C. (1995). Biased underwater visual census biomass estimates for target-

species in tropical reef fisheries. J. Fish. Biol. 47, 733-736.

Koop, K., Hutchings, P. (1996). Disposal of sewage to the ocean - a sustainable solution? Mar. Poll. Bull.

33, 121-123.

Krogh, M., Scanes, P.R. (1996). Organochlorine compound and trace metal contaminants in fish near

Sydney’s ocean outfalls. Mar. Poll. Bull. 33, 213-225. CHAPTER 9 GENERAL CONCLUSIONS 170

Lincoln Smith, M.P., Bell, J.D., Hair, C.A. (1991). Spatial variation in abundance of recently settled

rocky reef fish in southeastern Australia: implications for detecting change. Mar. Ecol. Prog. Ser.

77,95-103.'

Lincoln Smith, M.P., Hair, C.A., Bell, J.D. (1992). Jervis Bay marine ecological study. Project 4: fish

associated with natural rocky reefs and with artificial breakwaters, final report. Fisheries Research

Institute, Cronulla.

Lincoln Smith, M.P., Mann, R.P. (1989). Bioaccumulation in nearshore marine organisms II.

Organochlorine compounds in the red morwong Cheilodactylusfuscus, around Sydney’s three major

sewage ocean outfalls. A report prepared by SPCC and the Ecology Lab Pty Ltd, Sydney.

Lye, C.M., Frid, C.L.J., Gill, M.E. and McCormick, D. (1997). Abnormalities in the reproductive

health of flounder Platichthys flesus exposed to effluent from a sewage treatment works. Mar.

Poll. Bull. 34, 34-41.

Maurer, D., Robertson, G., Gerlinger, T. (1993). San Pedro Shelf California: Testing the Pearson-

Rosenberg Model (PRM). Mar. Environ. Res. 35, 303-321.

Nowak, B.F. (1996). Health of red morwong, Cheilodactylus fuscus, and rock cale, Crinodus

lophodon, from Sydney cliff-face sewage outfalls. Mar. Poll. Bull. 33, 281-292.

Pearson, T.H., Rosenberg, R. (1978). Macrobenthic succession in relation to organic enrichment and

pollution of the marine environment. Oceanogr. Mar. Biol. A. Rev. 16, 229-311.

Samoilys, M. (Ed.) (1997). Manual for assessing fish stocks on Pacific coral reefs. Department of

Primary Industries, Queensland. 75 pp.

Scanes, P.R., Scanes, K., Otway, N.M. (1995). Environmental problems due to disposal of wastes. In

Coastal marine ecology of temperate Australia. Underwood, A.J., Chapman, M.G. eds, UNSW

Press, Sydney, Australia, pp. 297-310

TEL (1993). Changes to assemblages of reef fish adjacent to the North Head and Bondi cliff face

outfall following decommissioning. Final report by The Ecology Lab for the Water Board,

Sydney. 49 pp.

TEL (1994). A quantitative survey of the impact of sewage on fish and macrobenthic assemblages in

the Illawarra coastal zone. Final report by The Ecology Lab for the Water Board, Sydney. 29 pp. CHAPTER 9 GENERAL CONCLUSIONS 171

Underwood, A.J. (1991). Beyond BACI: experimental designs for detecting human environmental

impacts on temporal variations in natural populations. Mar. Freshwat. Res. 42, 569-587.

Underwood, A.J. (1992). Beyond BACI: the detection of environmental impacts on populations in the

real, but variable, world. J. Exp. Mar. Biol. Ecol. 161, 145-178.

Underwood, A.J. (1993). The mechanics of spatially replicated sampling programmes to detect

environmental impacts in a variable world. Aust. J. Ecol. 18, 99-116.

Underwood, A.J. (1994). On beyond BACI: sampling designs that might reliably detect environmental

disturbances. Ecol. App. 4, 3-15.

Underwood, A.J. (1997). Experiments in Ecology. Their Logical Design and Interpretation using

Analysis of Variance. Cambridge University Press. 504 pp.

Waring, C.P., Stagg, R.M., Fretwell, K., McLay, H.A and Costello, M.J. (1996). The impact of sewage

sludge exposure on the reproduction of the sand goby, Pomatoschistus minutus. Environ. Poll. 93,

17-25. Appendix

Chapter 3. Asymmetrical analysis of variance model and F-ratio for Bellambi study * - Term of interest for effect of sewage

Source of variation df F-ratio Time 3 Tx L Location 2 No Test Impact vs Controls = I 1 No Test Among Controls = C 1 No Test Site (Location) 3 Tx S S (I) 1 S(C) * S(C) 2 T x S (C) Time x Location 6 Tx S TxL(I) 3 T x L (C) * T x L (C) 3 T x S (C) Time x Site (Location) 9 Residual TxS(I) 3 T x S (C) * T x S (C) 6 Residual Residual 72 Total 95

Chapter 3. Asymmetrical analysis of variance model and F-ratio for Port Kembla study * - Term of interest for effect of sewage

Source of variation df F-ratio Time 3 Tx L Location 2 Tx L Impact vs Controls = I 1 C * Among Controls = C 1 T x L (C) Time x Location 6 Residual TxL(I) 3 T x L (C) * T x L (C) 3 Residual Residual 37 Total 48 Appendix

Chapter 4. Asymmetrical analysis of variance model and F-ratio for Hunter study * - Term of interest for effect of sewage

Source of variation df F-ratio Period 2 Residual Time (Period) = T (P) 9 T (P) x I Location 2 T (P) x L Impact vs Controls = I 1 L (C) * Among Controls = C 1 T (P) x L (C) Period x Location 4 T (P) x L P(I) 2 P x L (C) * P(C) 2 T (P) x L (C) Time (Period) x Location 18 Residual TxL(I) 9 T x L (C) * T x L (C) 9 T x S (C) Residual 108 Total 143