Environmental Management of Aquaculture Effluent:

Development of Biological Indicators and Biological Filters

Adrian B. Jones

Environmental Management of Aquaculture Effluent: Development of Biological Indicators and Biological Filters

A Thesis submitted by

Adrian B. Jones B.Sc. (Hons) The University of Queensland, Australia

to the

Department of Botany The University of Queensland AUSTRALIA

in fulfilment of the requirements for the degree of Doctor of Philosophy within The University of Queensland

July 1999

STATEMENT

The work presented in this thesis is, to the best of my knowledge and belief, original, except as acknowledged in the text, and the material has not been submitted, either in whole or in part, for a degree at this or any other University.

Signed................................................

ACKNOWLEDGMENTS

This thesis was initiated from a Fisheries Research Development Corporation (FRDC) grant to Dr Nigel Preston (CSIRO) and Moreton Bay Prawn Farm. Additional funding was provided through grants from the Australian Research Council (ARC) to Dr William C. Dennison, the CRC for Aquaculture and a University of Queensland Postgraduate Research Scholarship.

Its hard to believe its finally finished. From the early days at CSIRO and Moreton Bay Prawn Farm laying concrete slabs and besser brick raceways to cruising Moreton Bay in thunder storms, the all-nighters at Straddie and then the endless days and nights in front of the computer. None of it would have been possible without the support and friendship of many people, mostly from the Marine Botany Group at UQ.

Dr. Bill Dennison, who developed my interest in marine research through his amazing enthusiasm in the field, for his advice and suggestions, and for his ability to make you realise that its not all as bad as it seems when youre in the depths of confusion.

To everyone else in Marine Botany who provided help with field work, help with interpretation and presentation of results, proof reading of manuscripts, and general friendship and support. In particular, thanks to Cindy Heil, Michele Burford, Mark ODonohue and Joelle Prange who reviewed various sections of the thesis. Also a special mention to Ros Murrell who has tirelessly helped me to track down a certain person when times were desperate. You were my link to the outside world during those last six months.

ACKNOWLEDGMENTS

viii

Dr. Nigel Preston, CSIRO Cleveland Marine Laboratories, for his constant encouragement regarding my writing abilities, his invaluable help regarding the initial planning of the thesis topic, his timely pep talks and last minute editing.

Theresa Mitchell, for her help, efficiency, support and advice regarding anything and everything to do with forms, policies, scholarships and finally the thesis submission procedures. Life at Botany just isnt the same without you! Jan Stewart, for conducting the isotopic analyses, and Gordon Moss for analysing the amino acid samples.

Sabine Roberts who read early drafts of all the chapters, picking my grammar to pieces and helping get my thoughts on the right track. Thankyou so much for all your help, encouragement and friendship throughout my PhD.

My parents who put up with me over the last 4 years constantly complaining about how this didnt work and that didnt work, and no, I dont know when I will finish it all. Thankyou for being there to listen to my complaining, and for your support and most importantly, for never pressuring me.

Finally to Tracey, who despite my constant rebuttal continued to maintain that somehow I would finish it before our flight left. Thankyou for your support during my constant state of stress and panic, especially when I was having serious doubts as to how I would manage to finish it on time. Thankyou for putting up with no end of complaints and irrational

ACKNOWLEDGMENTS

ix

behaviour, and for always being there and for accepting my stream of unfulfilled promises. Thankyou for your marathon proof reading / printing / collation efforts at the end, especially amongst the continual printer jams and photocopying nightmares and my accompanying frustration and rampage. Thankyou for somehow managing to keep me together! Most importantly, thankyou for your love, companionship, support and understanding.

ABSTRACT

Rapid global expansion of the aquaculture industry has prompted the need for development of techniques for effective environmental management. In intensively farmed regions, aquaculture effluent has resulted in environmental degradation of receiving waters. The issues to be addressed include analysis of effluent water quality, determination of the ecological impact of effluent on the ecosystem, and development of remediation strategies to reduce these impacts. Physical and chemical water quality analyses can identify elevated concentrations of suspended solids, chlorophyll a, water column nutrients and other components of aquaculture effluent, however, additional biological sampling is required to provide meaningful information about the ecological impacts of effluent discharge on receiving waters. Analyses of the amino acid composition, tissue nitrogen content and stable isotope ratio of nitrogen ( 15N) in seagrasses, mangroves and macroalgae were developed as biological indicators to determine the influence of shrimp farm effluent on a coastal ecosystem. Different responses in these biological parameters revealed that the impacts of aquaculture effluent on receiving waters were qualitatively different to the impacts of sewage effluent. The impacts were also spatially more extensive than identified by water quality analyses, which revealed no elevation in the concentration of water column nutrients, chlorophyll a concentration or total suspended solids further than 400 m from the mouths of the creeks receiving the sewage and aquaculture effluent. The maximum 15N of the mangroves, seagrass and macroalgae associated with the treated sewage discharge was 19.6, which was significantly higher than the influence of the shrimp effluent (7.6). A 15N value of 4.5, which is elevated relative to unimpacted sites, indicated that the impacts extended up to 4 km from the mouths of the creeks. Differences in the concentrations of the amino acids proline, serine, glutamine and alanine in the seagrass and macroalgae were suggested to reflect the source (aquaculture or sewage) of the nutrients taken up by the plants.

To reduce the environmental impacts, effluent treatment techniques using biological filters were investigated. Filtration by oysters ( Saccostrea commercialis) significantly reduced the concentrations of chlorophyll a (phytoplankton), bacteria, total nitrogen, total phosphorus and total suspended solids to 5%, 32%, 67%, 63% and 11% of the initial concentrations, respectively. However, oyster excretion increased the concentrations of the dissolved nutrients, ammonium (from 18 to 51 M), nitrate / nitrite (from 1.0 to 13 M), and phosphate (from 0.5 to 3.3 M), however macroalgal ( Gracilaria edulis) absorption significantly reduced these concentrations to 2.3%, 2.2% and 4.8%, respectively. The ratio of ammonium to nitrate / nitrite in the effluent was also significantly reduced, which has positive implications for recycling of wastewater back into shrimp production ponds, and reducing impacts on receiving waters. The efficiency and condition of the oysters and macroalgae was reduced by fouling from the high concentration of suspended particulates in the effluent. Several novel techniques such as dissolved free amino acid composition, pigment concentrations, PAM fluorescence, tissue nitrogen and 15N were used to assess the condition of the macroalgae. It was observed that an intermediate reduction in the concentration of suspended

ABSTRACT

xii

particulates resulted in the best growth and condition of the biofilters. The concentration of particulates in this treatment (11 nephelometric turbidity units) provided sufficient particulates for oysters to filter, and a source for regeneration of nutrients for macroalgal uptake, as well as reducing the effects of photoinhibition which can occur in Gracilaria spp. at relatively low light intensities. The problems associated with fouling were successfully mitigated by incorporating natural sedimentation prior to oyster filtration, and subsequent macroalgal absorption. This combined system of treatment proved effective at optimising the performance of the biological filters to improve the water quality of the effluent. Using this combination of polyculture, it was estimated that up to 18 kg N ha-1 d-1 and 15 kg P ha-1 d-1 could be removed from commercial shrimp ponds. The water quality of aquaculture effluent and its impact on the receiving waters will vary due with differing environmental conditions, as well as the type of aquaculture being conducted. Regardless, this thesis has demonstrated that filtration / absorption by various marine organisms can be effective tools for monitoring and reducing the environmental impacts of aquaculture effluent.

TABLE OF CONTENTS

Statement Acknowledgments Abstract Table of Contents List of Tables List of Figures List of Plates

v vii xi xiii xviii xx xxiii

CHAPTER 1. INTRODUCTION 1.1 Aquaculture 1.2 Environmental Impacts 1.3 Biological Indicators 1.4 Biological Treatment Options 1.4.1 Oysters 1.4.2 Macroalgae 1.5 Polyculture / Integrated Aquaculture 1.6 Thesis Aims 1.7 Thesis Overview 1.7.1 Chapter Outline 1.9 Publication Status of Thesis Chapters

1 2 4 5 6 9 12 13 14 16 18

CHAPTER 2 ASSESSING ECOLOGICAL IMPACTS OF SHRIMP AND SEWAGE EFFLUENT: BIOLOGICAL INDICATORS WITH STANDARD WATER QUALITY ANALYSES Abstract 2.1 Introduction 2.2 Materials and Methods 19

19 20 25

TABLE OF CONTENTS

xiv

2.2.1 Study Region 2.2.2 Experimental Design 2.2.3 Collection 2.2.4 Analytical Procedures 2.2.5 Statistical Analysis 2.3 Results 2.3.1 Physical and Chemical Water Quality Analyses 2.3.1.1 Salinity 2.3.1.2 Nutrients 2.3.1.3 Phytoplankton 2.3.1.4 Suspended Solids and Secchi Depth 2.3.1.5 Sediment Organic Content 2.3.2 Bioindicators 2.3.2.1 Tissue Nitrogen Content 2.3.2.2 15N Stable Isotope Ratio of Nitrogen 2.3.2.3 Free Amino Acid Composition 2.4 Discussion 2.4.1Water Quality Parameters 2.4.1.1 Effluent Composition 2.4.1.2 Phytoplankton Biomass and Productivity 2.4.2 Biological Indicators 2.4.2.1 Tissue N Content 2.4.2.2 15N Isotopic Signature 2.4.2.3 Amino Acid Composition 2.4.3 Comparison of Impacts 2.4.4 Conclusion 2.4.5 Application for other types of Aquaculture 2.4.6 Remediation Options

25 26 27 27 30 31 31 31 31 32 34 34 35 35 36 40 45 45 45 46 47 47 48 51 55 57 58 59

TABLE OF CONTENTS

xv

CHAPTER 3 OYSTER FILTRATION OF SHRIMP FARM EFFLUENT, THE EFFECTS ON WATER QUALITY Abstract 3.1 Introduction 3.2 Materials and Methods 3.2.1 Experimental Design 3.2.2 Analytical Procedures 3.3 Results 3.3.1 Suspended Solids 3.3.2 Organic content 3.3.3 Chlorophyll a 3.3.4 Bacteria 3.3.5 Total Nutrients 3.4 Discussion 3.4.1 Scaling Up Calculations 3.4.2 Summary 63

63 64 67 67 68 70 70 70 72 72 72 73 75 75

CHAPTER 4 THE EFFICIENCY AND CONDITION OF OYSTERS AND MACROALGAE USED AS BIOLOGICAL FILTERS OF SHRIMP POND EFFLUENT Abstract 4.1 Introduction 4.2 Materials and Methods 4.2.1 Experimental Design 4.2.1.1 Filtration Efficiency Experiments 4.2.1.2 Biofilter Condition Experiments 4.2.2 Analytical Procedures 4.3 Results 4.3.1 Filtration Efficiency Experiments 4.3.1.1 Continual Flow 4.3.1.2 Recirculating Experiments 77

77 78 81 81 81 83 86 90 90 90 92

TABLE OF CONTENTS

xvi

4.3.2 Biofilter Condition Experiments 4.4 Discussion 4.4.1 Efficiency of Biofilters 4.4.2 Condition of Biofilter Organisms 4.4.3 Conclusion

97 103 103 110 115

CHAPTER 5 IMPROVEMENTS IN WATER QUALITY OF AQUACULTURE EFFLUENT AFTER TREATMENT BY SEDIMENTATION, OYSTER FILTRATION AND MACROALGAL ABSORPTION Abstract 5.1 Introduction 5.2 Materials and Methods 5.2.1 Experimental Design 5.2.2 Analytical Procedures 5.3 Results 5.3.1 Suspended Solids 5.3.2 Organic content 5.3.3 Chlorophyll a 5.3.4 Bacteria 5.3.5 Dissolved Oxygen 5.3.6 Total Nitrogen 5.3.7 Total Phosphorus 5.3.8 Ammonium 5.3.9 Nitrate / Nitrite 5.3.10 Phosphate 5.3.11 Nutrient Uptake Rates and Ratios 5.4 Discussion 5.4.1 Sedimentation 5.4.2 Oyster Filtration 5.4.3 Macroalgal Absorption 5.4.4 Nutrient Regeneration 5.4.5 Conclusions 117

117 118 121 121 123 126 126 126 128 129 131 132 133 134 136 137 137 140 140 141 144 147 149

TABLE OF CONTENTS

xvii

CHAPTER 6 CONCLUSION 6. 1 Downstream Impacts 6.2 Efficiency of Biological Filters 6.3 Condition of Biofilters 6.4 Scaling up for Commercial Treatment 6.5 Other Potential Biofilters 6.6 Management Implications and Potential Problems with Biofiltration / Polyculture 6.7 Benefits of Polyculture or Integrated Aquaculture 6.8 Future Research 6.9 Summary BIBLIOGRAPHY

151

151 152 153 155 156

157 158 159 160

163

APPENDIX 1 FACTORS LIMITING PHYTOPLANKTON BIOMASS IN THE BRISBANE RIVER AND MORETON BAY 192

APPENDIX 2 PHOTOSYNTHETIC CAPACITY IN CORAL REEF SYSTEMS: INVESTIGATIONS INTO ECOLOGICAL APPLICATIONS FOR THE UNDERWATER PAM FLUOROMETER 203

LIST OF TABLES Table 2.1. Results of traditional water quality monitoring for the creek with shrimp farm effluent and sewage treatment effluent. DIN = Dissolved Inorganic Nitrogen; DIP = Dissolved Inorganic Phosphorus; Chl a = Chlorophyll a; Phyto Prod = phytoplankton productivity; TSS = total suspended solids; VSS = volatile suspended solids; Secchi = secchi disc depth. Only one replicate measurement was recorded for Secchi disk depth and salinity (Practical Salinity Scale). Table 2.2. Correlations (r2) between the concentration of phytoplankton (chlorophyll a) and phytoplankton productivity (14C uptake) and various water quality parameters. DIN = Dissolved Inorganic Nitrogen; DIP = Dissolved Inorganic Phosphorus; Phyto Prod = phytoplankton productivity (mg C m-3 h-1); Chl a = Chlorophyll a (g L-1); TSS = total suspended solids (mg L-1); VSS = volatile suspended solids (mg L-1); ISS = inorganic suspended solids (mg L-1); Secchi = secchi disc depth (m). Numbers in bold type indicate significant correlations (r2 0.6). Table 2.3. Results of bioindicator monitoring for the creek with shrimp farm effluent and sewage treatment effluent. 15N = Nitrogen stable isotope ratio; %N = Tissue N content; nd = no data (no plants were present). 38 Table 2.4. Results of bioindicator monitoring for the shrimp and sewage creeks. % refers to percentage of total free amino acid pool. SER = serine; ALA = alanine; GLN = glutamine; PRO = proline; Total = total concentration of free amino acids (mol g wet-1); nd = no data (no plants were present). Table 3.1 Combinations of live and dead oysters (Saccostrea commercialis) used in experiments to determine the effects of oyster density on the water quality of shrimp pond effluent. Table 3.2 Concentration of various water quality parameters before and after filtration by oysters at 3 different densities (see Table 3.1). Values for control (no oysters) and shells (dead shells only) are also given. Values in brackets are concentrations expressed as a percentage of the inflow value. Values in italics are standard errors. Table 4.1 Water quality parameters after filtration by oysters under flow through conditions in raceways. Total N = total Kjeldahl nitrogen; Total P = total phosphorus. 92 71 68 42 35 33

LIST OF TABLES

xix

Table 4.2 Water quality parameters after filtration by oysters after the first circuit during recirculating flow in raceways. TSS = total suspended solids; Organic = organic component of TSS (loss on ignition); Inorganic = inorganic component of TSS. Table 4.3 Changes in the free amino concentration and composition of macroalgae for various treatments in laboratory settling experiments. % refers to percentage of total free amino acid pool. CIT = citrulline; GLU = glutamate; ALA = alanine; GLN = glutamine; PHE = phenylalanine; SER = serine; Total = total concentration of free amino acids (mol g wet-1). Table 5.1 Percentage of original concentrations of various water quality parameters after settling, filtration by oysters and filtration by macroalgae.
*** *

94

101 p 0.05; ** p 0.01;

p 0.001. Percentage of highest concentration represents the final concentration as a

percentage of the highest recorded concentration after sedimentation and oyster filtration. The percent of initial concentration represents the final concentration as a percentage of the initial concentration in the untreated effluent. The only differences between the two values are for the dissolved inorganic nutrients (NH4+, NO3-, & PO43-). 128 Table 5.2 Nutrient uptake and release rates for sedimentation, oyster filtration and macroalgal absorption. Negative symbols represent nutrient uptake, and positive represent nutrient release. The top value for each treatment is the gross value, the middle value is the control and the bottom value (in bold type) is the net value after correction for nutrient changes in the control tanks. The last row of results represent the rates of macroalgal nutrient uptake over the first hour, when nutrient concentrations were still saturating uptake kinetics. 136

LIST OF FIGURES Figure 2.1 Map of study sites in Moreton Bay, including the location of shrimp and sewage effluent discharges. Figure 2.2. Map showing the values of %N in seagrass (Zostera capricorni), macroalgae (Catenella nipae), and mangroves (Avicennia marina) at the study sites (see Fig. 2.1 for site references). Figure 2.3. Map showing the values of 15N in seagrass (Zostera capricorni), macroalgae (Catenella nipae), and mangroves (Avicennia marina) at the study sites (see Fig. 2.1 for site references). 40 Figure 2.4. Map showing the amino acid composition of seagrass (Zostera capricorni) at the study sites (see Fig. 2.1 for site references). 43 Figure 2.5. Map showing the amino acid composition of macroalgae (Catenella nipae) at the study sites. Pie graphs have been reduced to quarters for layout purposes. The remaining three quarters of the pie graphs not represented are a continuation of the other amino acid category (not serine or alanine) (see Fig. 2.1 for site references). Figure 2.6. Conceptual model of the two creeks and the range and type of impacts from the different effluent sources. Figure 3.1. Location map of Moreton Bay Prawn Farm near Brisbane, Australia. Figure 3.2. Schematic representation of tank and waterflow layout. Figure 4.1 Diagrammatic representation of experimental setup, a) single raceway with baffles and oyster trays, and b) laboratory settling experiment. NTU = nephelometric turbidity units. The oysters used in the experiments were Sydney Rock oysters, Saccostrea commercialis and the macroalgae was Gracilaria edulis. Effluent was from an intensive Penaeus japonicus shrimp farm. Figure 4.2 Impacts of effluent on biofilters: a) Oyster mortality (%) from upper, middle and lower trays after 2 weeks at low, medium and high oyster stocking densities in raceways supplied with unsettled shrimp effluent, and b) change in dissolved nutrient concentrations after passing effluent through low, medium and high macroalgal stocking densities in raceways supplied with unsettled shrimp effluent. Positive change represents an increase, negative change represents a decrease. 91 84 61 66 67 44 39 27

LIST OF F IGURES

xxi

Figure 4.3 Particle size distribution, a) before and after control and oyster treatment raceways during single continuous flow, b) before and after consecutive circuits through oyster treatment raceways (linear scale), and c) before and after consecutive circuits through oyster treatment raceways (log scale). 93 Figure 4.4 Concentrations of water quality components before and after consecutive circuits through oyster treatment raceways, a) bacterial numbers, b) chlorophyll a concentration, and c) total suspended solids (TSS). Figure 4.5 Growth of oysters and macroalgae after 8 weeks in tanks supplied with shrimp effluent pre-settled for 0, 1, 6 & 24 h. a) change in oyster growth rate expressed as changes in oyster volume (cm3 oyster -1), and b) macroalgal biomass. n.d. = no data. 98 Figure 4.6 Response of macroalgae to 8 weeks in tanks supplied with shrimp effluent presettled for 0, 1, 6 & 24 h. a) macroalgal growth expressed as number of news shoots per tank, and b) concentration of the photosynthetic pigments, chlorophyll a (CHL) and phycoerythrin (PE). shrimp effluent of different settlement times, a) %N, and b) 15N. Figure 4.8 The response of electron transport rate (ETR) versus photosynthetically active radiation (PAR) in macroalgae incubated in seawater (control) or shrimp effluent (settled 24 h plus oyster filtered for 12 h). Figure 5.1 Design of integrated treatment system stocked with oysters (40 g Saccostrea commercialis), and macroalgae (Gracilaria edulis) . 124 Figure 5.2 Changes in total suspended solids (A) and phytoplankton biomass (chlorophyll a) (B) from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. Figure 5.3 Concentration of particles settled per litre from sedimentation and oyster filtration. Figure 5.4 Changes in the organic content of the a) total suspended solids (TSS) and b) settled particles in the effluent water from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. 130 Figure 5.5 Changes in bacterial numbers from sedimentation, oyster filtration and macroalgal absorption. 131 129 127 102 99 Figure 4.7 Macroalgal nitrogen content (a) and 15N (b) after 8 weeks in tanks supplied with 100 96

LIST OF F IGURES

xxii

Figure 5.6 Changes in water column dissolved oxygen concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. Figure 5.7 Changes in water column total N (A) and P (B) concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. Figure 5.8 Changes in water column NH4+, NO3-, PO43- concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. Figure 5.9 Changes in water column total N: P ratio (A) and DIN: DIP ratio (B) from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible. 139 Figure 6.1 Diagrammatic design of water flow for typical untreated shrimp farms (left) and a design to incorporate physical (sedimentation) and biological (oyster filtration and macroalgal absorption) treatment (right). 162 135 133 132

LIST OF PLATES Plate 1.1 Ponds at Moreton Bay Prawn Farm, an intensive shrimp farm (Penaeus japonicus) near Moreton Bay, Queensland, Australia. Plate 1.2 Shrimp Farm plume discharging into Moreton Bay, Queensland, Australia. Plate 1.3 High phytoplankton concentration in plume from shrimp farm discharging into Moreton Bay, Queensland, Australia. Plate 1.4 Penaeus japonicus from ponds at Moreton Bay Prawn Farm, Queensland, Australia. Plate 1.5 Sydney Rock Oysters (Saccostrea commercialis) cultured in Moreton Bay, Queensland, Australia. Plate 1.6 Gracilaria edulis collected from Moreton Bay, Queensland, Australia. Plate 4.1 Raceways constructed at Moreton Bay Prawn Farm, Queensland, Australia. 7 10 82 5 3 2 3

Plate 4.2 Control raceway on the left with no oysters and treatment stocked at low density 55 g oysters. Demonstrates changes in water clarity (reduction in suspended solids) with the oyster tray clearly visible in the raceway stocked with oysters, but not in the control raceway. Plate 4.3 First chamber (foreground) and second chamber (background) of an oyster treatment raceway showing the improvement in water clarity (reduction in suspended solids) within the raceway. Plate 4.4 Fouling of oysters by settling particulates in raceways. Plate 5.1 Experimental setup with sedimentation drum (background) and control, oyster, macroalgal filtration tanks (foreground). Plate 5.2 Water samples collected: a) before sedimentation; b) after sedimentation; and c) after biofiltration. 150 124 105 108 104

CHAPTER 1 INTRODUCTION

1.1 Aquaculture The UN Food and Agricultural Organisation has estimated that by 2020 more than 50% of fisheries production will need to come from aquaculture due to human population growth, continuing demand for seafood, and static or declining natural fish harvests. However, in many countries aquaculture practices have already resulted in the destruction of coastal vegetation, salinisation of land, pollution of waterways and massive crop losses (Phillips et al., 1993). Further expansion using current technologies is simply not justifiable or sustainable. If the level of demand for seafood is to be met the only alternative is to develop new technologies that require less space and have minimal adverse environmental impacts.

Penaeid prawn (shrimp) farming has been one of the most economically successful of all intensive aquaculture industries. In the early days of shrimp farming and other forms of aquaculture, the perception was that they were completely clean industries (Weston, 1991). Recent reviews of intensive shrimp aquaculture have emphasised the need for more effective controls on the quality of effluent water discharged into the environment (Phillips et al., 1993; Primavera, 1994).

Shrimp farming can be separated into extensive, semi intensive and intensive culture systems (Macintosh & Phillips, 1992). Extensive culture systems have large pond sizes (>5 ha), relatively low stocking densities (<10 per m2), no aeration, and natural food sources (through fertilisation). Intensive farming consists of smaller ponds (1 ha), very high stocking densities (>20 per m2), aeration, and formulated high protein feed pellets (Plate 1.1). Intensive farming

CHAPTER 1

is becoming more prominent, increasing the potential for environment impact from shrimp farming (Phillips et al., 1993).

Plate 1.1 Ponds at Moreton Bay Prawn Farm, an intensive shrimp farm ( Penaeus japonicus) near Moreton Bay, Queensland, Australia.

1.2 Environmental Impacts Intensive shrimp aquaculture systems rely on high protein feed pellets to produce high rates of growth, but a large proportion of the pellets are not assimilated by the shrimps (Primavera, 1994). Approximately 10% of the feed is dissolved and 15% remains uneaten. The remaining 75% is ingested, but 50% is excreted as metabolic waste, producing large amounts of gaseous, dissolved and particulate waste (Lin et al., 1993). Subsequently, the effluent contains elevated concentrations of dissolved nutrients (primarily ammonia), plankton and other suspended solids (Ziemann et al., 1992). The dissolved nutrients and organic material in shrimp ponds stimulate rapid growth of bacteria, phytoplankton, and zooplankton (Lin et al., 1993). These untreated wastes are usually discharged directly into the environment, where they may enhance eutrophication, organic enrichment and turbidity of the local waterways (Plates 1.2 & 1.3) (Eng et al., 1989; O' Connor et al., 1989; Prakash, 1989).

INTRODUCTION

Plate 1.2 Shrimp Farm plume discharging into Moreton Bay, Queensland, Australia.

Plate 1.3 High phytoplankton concentration in plume from shrimp farm discharging into Moreton Bay, Queensland, Australia.

CHAPTER 1

Australia has a small but expanding coastal aquaculture industry. From 1984 to 1998, the shrimp farming sector rose from 15 to 2 000 t. The industry is well placed to take advantage of developments in integrated aquaculture systems, such as the use of natural biofilters and recirculating systems. In southeast Queensland, there are two shrimp species farmed, Penaeus monodon Fabricius and P. japonicus Bate (Plate 1.4), with stocking of post larvae in October and harvest the following April to June. With increasing development of the industry, concerns have risen about the impact of effluent from the farms. The effluent from shrimp farms discharging into Moreton Bay, Queensland often has concentrations of dissolved nitrogen and phosphorus which are 60 fold higher than receiving waters, chlorophyll a concentrations 200 fold higher, and total suspended solids (TSS) 20 fold higher (Jones et al., in prep a; Chapter 2). Australian waters are relatively low in nutrients compared with other coastal waters (State of the Environment Council, 1996), and therefore impacts may be potentially more acute. In Moreton Bay, background concentrations of water quality parameters are: NH4+ < 2 M; NO3- / NO2- ~ 0.1 M; PO43- ~ 0.2 M; chlorophyll a < 1 g L-1; TSS < 20 mg L-1.

1.3 Biological Indicators Due to the close proximity of shrimp farm discharges to several other point and non-point nutrient sources (ie. sewage effluent, agricultural runoff), it can be difficult to determine the impacts of aquaculture on the environment (Grant et al., 1995). Techniques are needed to distinguish the effect of each source and its range of impact so that appropriate discharge limits can be applied.

INTRODUCTION

Plate 1.4 Penaeus japonicus from ponds at Moreton Bay Prawn Farm, Queensland, Australia.

Traditional water quality analyses provide little information as to the impact of nutrients on the biota in the ecosystem (Lyngby, 1990). As a result there is a lack of data on the ecological impact of aquaculture effluent (Gowen et al., 1990). The use of biological indicators can provide information as to the nutrient source, the bioavailability of the nutrients, and the integration of short lived nutrient pulses (Lyngby, 1990; Horrocks et al., 1995; Jones et al., 1996; Udy & Dennison, 1997b; Jones et al., 1998; Appendix 1).

1.4 Biological Treatment Options Concerns about the possible adverse impacts of aquaculture discharge have become a risk factor for the industry (Braaten, 1991). This has prompted efforts to develop cost-effective methods of effluent treatment. In addition to prohibitive costs, because of the large volume of effluent, sewage treatment practices have proved inefficient due to the high suspended solid load (Tetzlaff & Heidinger, 1990) and the high salinity of aquaculture effluent. There are a number of commercially available bacterial systems to promote nitrification and subsequent denitrification to remove nitrogen from the effluent. However, the effectiveness of such systems for treating shrimp pond effluent have yet to be examined rigorously.

CHAPTER 1

The use of filter feeding bivalves such as oysters to consume phytoplankton, zooplankton, and bacteria (Lin et al., 1993), and macroalgae to assimilate the remaining dissolved nutrients (Haines, 1975) may prove to be an efficient and economically viable alternative for improving the water quality of shrimp farm discharges (Hopkins et al., 1993a; Lin et al., 1993). In addition to filtering organic food particles, oysters can also improve the quality of pond effluent by reducing the concentration of inorganic suspended solids.

1.4.1 Oysters The oyster industry in Moreton Bay, Queensland is based on the Sydney Rock Oyster Saccostrea commercialis (Iredale & Roughley) (Plate 1.5), an estuarine species native to Port Stevens, N.S.W., Australia and found from Victoria, Australia to Thailand (Angell, 1986). This species is considered marketable between 29 and 40 g (bottle oysters) and 40 to 67 g (plate oysters) whole weight (Witney et al., 1988).

Culture of oysters on traditional leases from spat to marketable size takes two to three years (Witney et al., 1988). Local availability of oysters is seasonal, with oysters being fat and ready for sale in summer, while in winter when phytoplankton concentrations are lower (Dennison et al., 1993), they are lean and growth is substantially slower. The use of land based aquaculture systems has been trialed to improve productivity and year-round marketability, but most attempts have been relatively unsuccessful, primarily due to the high cost and unreliability of mass algal culture. In an attempt to find alternative sources of microalgae as food for enhanced oyster production, shrimp farm effluent has been trialed (Wang & Jakob, 1991; Hopkins et al., 1993a; Jakob et al., 1993; Lin et al., 1993). Very fast

INTRODUCTION

rates of growth has been observed for oysters grown under controlled conditions with shrimp pond effluent (Jakob et al., 1993).

Plate 1.5 Sydney Rock Oysters ( Saccostrea commercialis) cultured in Moreton Bay, Queensland, Australia.

Oysters are suspension feeders and use their gills to filter phytoplankton, zooplankton, bacteria and other microscopic particles. Bivalves can remove phytoplankton from the water with high efficiency (Jrgensen, 1966), but their filtering ability is affected by a number of factors including the water flow rate (Walne, 1972), temperature (Loosanoff & Tommers, 1948), salinity (Djangmah, 1979), reproductive effort, and silt concentration (Loosanoff & Tommers, 1948; Angell, 1986). A temperature of approximately 30C (Angell, 1986) and salinity of 35 (Nell & Gibbs, 1986) is optimal, although the survival range for S. commercialis is 15 50.

Shrimp pond effluent water typically has elevated concentrations of total suspended solids, a large fraction being small inorganic clay minerals (Smith, 1996). In waters with high concentrations of silt, oyster pumping and therefore feeding can be greatly inhibited (Loosanoff & Tommers, 1948), or they may even cease pumping entirely.

CHAPTER 1

To overcome the problems of oyster fouling, sedimentation ponds could be used to remove the larger settleable particles prior to oyster filtration (Wang, 1990). The remaining particles are either motile, or are small particles (<5 m) with a specific gravity similar to seawater and therefore do not easily settle out of suspension (Rubel & Hager Inc., 1979). Oysters can then be used to filter the organic food particles and convert them to meat and faeces. Oysters can also combine small inorganic clay particles (which otherwise would not settle out) with mucous into larger particles and egest them as pseudofaeces (Tenore & Dunstan, 1973), which readily settle out of suspension.

Phytoplankton can be ingested, excreted as pseudofaeces, or simply settle out of suspension without being available as food (Scura et al., 1979). High concentrations of plankton can also inhibit feeding in bivalves (Ali, 1970; Schulte, 1975). At high food concentrations oysters produce large amounts of pseudofaeces (ie. food filtered but not ingested) and therefore have high deposition rates, demonstrating an inefficient use of filtered food (Tenore & Dunstan, 1973).

Although oyster stocking density does not appear to affect survival, high densities may decrease individual weight due to competition for food (Holliday et al., 1991). Given the high concentrations of phytoplankton in shrimp ponds and pond effluent, it is more likely to be the high concentrations of fine inorganic particles that control filtering rates.

A significant portion of the research to date regarding growing oysters in the effluent from shrimp ponds has focussed on the advantages for growing the oysters. The highest ever reported sustained growth rates for the American oyster (Crassostrea virginica), were when oysters were supplied with effluent from a shrimp pond. Growth rates were reported to be

INTRODUCTION

0.5 g d-1, from seed size (0.04 g) to market size (55.0 g), in just 4 months (Jakob et al., 1993). The authors state that it was clearly shown that undiluted, semi-intensive, marine shrimp pond water provides all the requirements for the very rapid growth of the American oyster C. virginica from 0.05 g spat through 78 g adults (Jakob et al., 1993). Evidence that the quality of oysters remains high in aquaculture was observed by Lam & Wang (1989) who used shrimp pond water to produce excellent quality half-shell oysters, grown from 0.1g to 54.2 g in 198 days with 96% survival.

The use of oysters as biofilters can improve the quality of water leaving aquaculture ponds, and potentially provide a secondary cash crop. After filtration by oysters most of the nutrients (those bound up in phytoplankton and other suspended solids), are deposited as faeces and pseudofaeces, while the rest are incorporated into oyster tissue. However, oysters can also contribute significant amounts of ammonia to the effluent through excretion (Srna & Baggaley, 1976). Ammonia toxicity to shrimp is one of the primary reasons farmers undertake water exchange (Kou & Chen, 1991), and therefore it must be removed before effluent water can be recycled back into production ponds. Consequently, removal of these deposited nutrients from the system entirely will require either physical removal of the settled sediment, denitrification, or assimilation of the dissolved nutrients (from the remineralisation of faeces and pseudofaeces) by macroalgae such as Gracilaria spp. (Funge-Smith & Briggs, 1998).

1.4.2 Macroalgae Macroalgae can absorb significant quantities of dissolved inorganic and organic nutrients, usually with a preference for NH4+ (D'Elia & DeBoer, 1978; Haines & Wheeler, 1978; Hanisak & Harlin, 1978; Harlin, 1978). The ability of macroalgae to rapidly take up nutrients

CHAPTER 1

10

for growth, and store luxury reserves in the form of amino acids and pigments makes them ideal for stripping nutrients from aquaculture effluent (Haines, 1975). Additionally, macroalgae are known to absorb and store heavy metals (Burdin & Bird, 1994), which may be a potential pollutant in shrimp pond effluent.

Removal of nutrients by macroalgae is also efficient as harvesting is relatively simple, and provides an additional cash crop (Hopkins et al., 1995b). Macroalgae can also assimilate metabolic wastes from mariculture animals, which is beneficial to shrimp production ponds if the wastewater is to be recycled (Qian et al., 1996). Macroalgae can be used to ensure complete removal of inorganic nitrogenous excreta from the bivalves (Mann & Ryther, 1977). In particular, commercial red seaweeds such as species from the genera Chondrus, Gracilaria, Agardhiella and Hypnea are candidates as a final polishing step to leave the effluent virtually free of inorganic nitrogen (Ryther et al., 1975) (Plate 1.6).

Plate 1.6 Gracilaria edulis collected from Moreton Bay, Queensland, Australia.

Certain species of red macroalgae (Rhodophyta), in particular those from the genera Gracilaria, Gelidium, and Hypnea are harvested commercially. These species contain

INTRODUCTION

11

sulfated galactan agar and carrageenin which are widely used in the pharmaceutical, cosmetic and food industries (Raven et al., 1987). Nutrients are generally the limiting factor to macroalgal growth in natural systems, and attempts have been made to culture them in land based aquaculture systems. The wastewater from aquaculture effluent contains sufficient nutrients to sustain the high growth rates required without fertilisation, but the high concentrations of suspended solids can foul the macroalgae and reduce light availability (Briggs & Funge-Smith, 1993).

There are potentially considerable economic benefits to be gained from growing macroalgae in shrimp pond effluent. The growth of Hypnea musciformis in the effluent from a tropical mariculture system has been estimated as producing a gross harvest value of $107 250 ha -1 annually (Roels et al., 1976). H. musciformis cultured in aquaculture effluent grew at 64.5 g wet wt d-1, compared to deep water growth of 12.1 g wet wt d-1 (Haines, 1975). Percent carrageenin yields however were lower, ie., 16% dry wt versus 29% for the deep water, however the total production of carrageenin is approximately 3 times greater from the aquaculture effluent (Haines, 1975).

The use of bivalves and / or macroalgae to treat the effluent from shrimp farms has been investigated in a number of studies (Wang & Jakob, 1991; Hopkins et al., 1993a; Jakob et al., 1993; Shpigel et al., 1993b; Jones & Preston, 1999; Chapter 3). Using oysters (to filter phytoplankton, bacteria and other suspended solids), and macroalgae (to take up dissolved nutrients) can potentially improve the quality of shrimp pond effluent. In addition to the environmental benefits for receiving waters, there are also economic gains resulting from the conversion of high cost uneaten and dissolved feed pellets into two additional marketable crops (Wang, 1990).

CHAPTER 1

12

1.5 Polyculture / Integrated Aquaculture Polyculture is defined as the culture of several different organisms in the one culture unit. In contrast, integrated aquaculture is the co-culture of different organisms, but in discrete culture units (Chien & Tsai, 1985). These techniques are regarded as being more ecologically sound methods of aquaculture (Mackay & Lodge, 1983), with a more efficient use of resources, and a higher resilience against environmental fluctuation (Chien & Liao, 1995).

Despite the advantages of these types of combined aquaculture, there may be some problems associated with management of several organisms all with differing culture requirements. Management can be more complex with respect to stocking densities, culture techniques and associated infrastructure, harvesting procedures, and effluent flow management (Chien & Liao, 1995). Specific problems for intensive shrimp farming relate to fouling effects from the high concentrations of suspended solids on secondary crop species (and potential biofilters) such as oysters and macroalgae (Ziemann et al., 1992; Funge-Smith & Briggs, 1998). Although the use of these and other biological treatment techniques for facilitating water recycling are ecologically sound, much research is needed to improve the efficiency of these systems (Lin, 1995).

Effective management of aquaculture effluent can be separated into identification of downstream impacts, and effective farm management to reduce these impacts. Identification of impacts to receiving waters may be accomplished with a combination of water and sediment water analyses with biological indicators to elucidate ecological impacts. Effective on-farm management of effluent can probably be accomplished by a combination of physical and biological treatment techniques.

INTRODUCTION

13

1.6 Thesis Aims Characterise the components of shrimp pond effluent, and their concentrations relative to the receiving waters,

Develop the use of various marine plants as bioindicators to determine the effects of prawn farm effluent on receiving waters,

Determine the viability of oysters and macroalgae as biological treatment organisms for shrimp pond effluent,

Determine the differences in biological filter performance with changes in density, size, and water flow regimes,

Identify problems associated with maintaining oysters and macroalgae in the high suspended solids environment and optimise techniques to minimise the impact on their growth, condition, and effectiveness as biofilters,

Design an integrated system to produce the greatest improvements in water quality, while maintaining the condition of the biological filter organisms.

CHAPTER 1

14

1.7 Thesis Overview Despite several reported cases of large scale environmental degradation linked to aquaculture effluent, there has been no successful determination of the ecological impacts, and certainly no techniques to distinguish these downstream impacts in relation to other nutrient inputs. Techniques to improve effluent discharge water quality, including the use of bivalves to filter aquaculture effluent have been undertaken on a small scale by the industry in other regions of the world. However, there has been a distinct lack of quantitative data to determine the most efficient use of these techniques and the ecophysiological responses of the biofilter organisms. This thesis has addressed these shortcomings.

Bioindicator techniques were developed to investigate the ecological impacts of aquaculture effluent and biofilter organisms were employed, not only to mitigate these impacts, but also to provide an efficient use of resources by producing secondary crops from aquaculture farm effluent. The research is this thesis has been conducted using techniques to look at ecophysiological responses of organisms and ecological changes in the system. This contrasts much of the published material in this area, which has been conducted purely at an applied level.

Evaluation of ecological impacts from shrimp farming was conducted using biological indicator techniques (tissue nitrogen content, 15N isotopic signatures, amino acid composition, phytoplankton productivity) in conjunction with more traditional water quality parameters (nutrient concentrations, suspended solids, chlorophyll a) to determine ecophysiological changes in the biota in the receiving waters. Rates of isotopic fractionation of nitrogen in the effluent and the changes in the dissolved free amino acid composition of the macroalgae incubated in shrimp effluent under controlled laboratory conditions provided

INTRODUCTION

15

some of the background responses used for determining the spatial range of impacts of shrimp effluent in receiving waters. These biological or ecological health indicators provided direct measures of the influence of aquaculture discharge.

The effects of different sizes and stocking densities of oysters and different densities of macroalgae on the water quality (total and dissolved nutrients, chlorophyll a, bacteria, total suspended solids, organic versus inorganic particulates, and particle size distribution) of shrimp effluent was determined for a variety of effluent flow regimes and during different stages of the shrimp growout season. In particular, analysis of the particle size distribution of the effluent provided information into the mechanisms by which oysters remove particulates from the water column, especially the small inorganic clay particles that are difficult to remove by sedimentation or mechanical filtration.

The effects of different concentrations of suspended solids from shrimp effluent on oyster and macroalgal condition was determined by physiological responses in the organisms. This information facilitated estimates of the optimum concentration of suspended particulates for efficient filtration performance by oysters and macroalgae, while minimising sedimentation time and / or mechanical filtration costs. A variety of novel techniques such as dissolved free amino acid composition, pigment concentrations, PAM fluorescence, tissue nitrogen and 15N were used to assess the condition of the macroalgae. These techniques also provide information about the bioavailability of the nutrient profile from shrimp effluent, i.e. whether it is suited for uptake by biofilter organisms (e.g. oysters and macroalgae).

Higher effluent flow rates are likely to improve biofilter condition, but may reduce the filtration performance. In an attempt to improve the condition and performance of the

CHAPTER 1

16

biofilters, experiments were conducted to recirculate the effluent though the biofilter organisms several times to test the possibility of increasing the effluent flow rate, without sacrificing improvements in water quality.

The combined efficiency of sedimentation, followed by oyster filtration of particulates and macroalgal absorption of dissolved nutrients proved to be an effective technique for improving shrimp pond effluent water quality. After treatment in this polyculture system, the effluent proved suitable for reuse in shrimp production ponds. The rates of nutrient regeneration from settled particulates, oyster excretion rates, nutrient uptake rates (bacteria, phytoplankton and macroalgae) and loss of N to the atmosphere via volatilisation and denitrification were determined directly, or inferred by difference.

1.7.1 Chapter Outline Chapter 2 Investigations were conducted to determine the impact of effluent from a local shrimp farm on the biota and integrity of the receiving waters in Moreton Bay. Results were compared with data from other unimpacted regions in Moreton Bay and with a nearby sewage treatment plant. Several bioindicator techniques were utilised to characterise the impacts of the effluent.

Chapter 3 Experiments were conducted to determine if oysters would be successful at improving the water quality of shrimp pond effluent, and to assess the optimal stocking density of the oysters to produce the greatest improvements in water quality.

INTRODUCTION

17

Chapter 4 The filtering efficiency of macroalgae and different sized oysters in raceways with flow through effluent supply, and recirculating supply were conducted. Issues regarding fouling of oysters and macroalgae were investigated to determine the maximum concentration of suspended solids that the oysters and macroalgae could tolerate without adversely impacting their health, survival and filtration efficiency.

Chapter 5 The overall efficiency of a polyculture treatment system was tested using sedimentation followed by oyster filtration and macroalgal absorption.

Chapter 6 The conclusions of the study and areas of potential future research and comparisons with the results of other studies are discussed.

CHAPTER 1

18

1.9 Publication Status of Thesis Chapters

Chapter 2 Jones, A.B., O'Donohue, M.J., Udy, J. & Dennison, W.C. (2001) Assessing ecological impacts of shrimp and sewage effluent: Biological indicators with standard water quality analyses. Estuarine, Coastal and Shelf Science 52, 91109. Presented at the Australian Marine Science Association annual conference, Adelaide, Australia, July 1998.

Chapter 3 Jones, A.B. & N.P. Preston (1999) Oyster filtration of shrimp farm effluent, the effects on water quality. Aquaculture Research 30, 51-57.

Chapter 4 Jones, A.B., N.P. Preston & W.C. Dennison (in review) The efficiency and condition of oysters and macroalgae used as biological filters of shrimp pond effluent. Aquaculture Research.

Chapter 5 Jones, A.B., Dennison, W.C. & Preston, N.P. (2001) Integrated treatment of shrimp effluent by sedimentation, oyster filtration and macroalgal absorption: a laboratory scale study. Aquaculture 193 (1-2), 155-178. Presented at the World Aquaculture Society Meeting, Sydney, Australia, May 1999.

CHAPTER 2 ASSESSING ECOLOGICAL IMPACTS OF SHRIMP AND SEWAGE EFFLUENT: BIOLOGICAL INDICATORS WITH STANDARD WATER QUALITY ANALYSES

Abstract
Despite evidence linking shrimp farming to several cases of environmental degradation, there remains a lack of ecologically meaningful information about the impacts of effluent on receiving waters. The aim of this study was to determine the biological impact of shrimp farm effluent, and to compare and distinguish its impacts from a nearby treated sewage discharge. Assessment of impacts was conducted using both water quality / sediment analyses and biological indicators. Water quality and sediment parameters measured included chlorophyll a, total suspended solids, volatile suspended solids, dissolved nutrients, salinity, and sediment organic content. Biological indicator monitoring consisted of analysis of amino acid composition, tissue nitrogen (N) content and stable isotope ratio of nitrogen ( 15N) in seagrasses, mangroves and macroalgae. The study area consisted of two tidal creeks, one receiving effluent from a sewage treatment plant (sewage creek) and the other receiving effluent from an intensive shrimp farm (shrimp creek). The creeks discharged into Moreton Bay, a sub tropical coastal embayment on the east coast of Australia. Water quality in both creeks was significantly modified, but changes were indistinguishable from unimpacted eastern Moreton Bay levels further than 750 m from the c reek mouths. Biological indicators, however, detected significant impacts up to 4 km beyond the creek mouths. The shrimp creek was more turbid due to clay minerals with a relatively high dissolved NH4+ (3.8 M) concentration, whereas the sewage creek had a higher percentage of organic material (35%) and dissolved nutrient concentrations were higher, particularly NO3- / NO2- (65 M) and PO43- (31 M). The sewage creek did not support high phytoplankton productivity (18-20 mg C m-3 h-1), in spite of high nutrient concentrations. Mangroves and macroalgae in the sewage creek were highly enriched with sewage nitrogen (indicated by high 15N), as was seagrass at the creek mouth. The 15N of seagrasses, mangroves and macroalgae ranged from 10.4-19.6 at the site of sewage discharge to 2.9-4.5 at the reference site, 4 km from the creek mouths. The 15N values of seagrass (4.5) and mangroves (3.4) at the reference site were higher than values reported for oligotrophic areas of Moreton Bay, but the 15N of macroalgae (2.9) was close to unimpacted eastern Moreton Bay values. Macroalgae derive nutrients from the water column, whereas seagrass and mangroves take up nutrients from the sediment. Therefore, deposition of effluent derived N into the sediments is implicated in the elevated 15N values of the seagrass and mangroves at the reference site. The free amino acid concentration and composition of seagrass and macroalgae was used to distinguish uptake of sewage and shrimp derived N. Proline (seagrass) and serine (macroalgae) were high in sewage impacted plants and glutamine (seagrass) and alanine (macroalgae) were high in plants impacted by shrimp effluent. The 15N and amino acid composition indicated sewage N extended further from the creek mouths than shrimp N. This analysis of physical / chemical and biological indicators was able to distinguish the composition and subsequent impacts of aquaculture on the receiving waters.

CHAPTER 2

20

2.1 Introduction Aquaculture is a rapidly expanding industry, and its effluent can be a major source of pollution in marine ecosystems (Chua et al., 1989; Twilley, 1989; Gowen et al., 1990; Braaten, 1991; Holmer, 1991; Phillips et al., 1991; Macintosh & Phillips, 1992; Pruder, 1992; Raa & Liltved, 1992; Wu et al., 1994; Samocha & Lawrence, 1997; Hargreaves, 1998). Environmental studies into the effects of shrimp aquaculture are limited and have mostly focussed on in-pond water quality, with little research conducted into the ecological impacts of wastewater on receiving waters (Pillay, 1992).

The monitoring of traditional water quality parameters has identified that downstream impacts of shrimp effluent, and other forms of aquaculture, are only measurable in close proximity to the discharge point. Hensey (1991) observed that environmental monitoring of aquaculture effluent using water quality sampling techniques showed no impacts. Samocha & Lawrence (1997) observed large diurnal fluctuations in water quality parameters measured downstream of shrimp farm discharge points, and that no increase in total suspended solids (TSS) or nutrient concentrations could be measured further than 400 m from the farms discharge. It is possible, however, that sediment impacts such as increased organic matter and anoxia, may extend further (up to 1 km) than water column impacts (Wu et al., 1994).

With the current projected expansion of shrimp farming in most coastal areas of the world, large scale increases in nutrients and suspended solids in the receiving waters are likely. Elevated loadings of particulate material to receiving waters have immediate effects on the receiving environment such as reduced light penetration and smothering of benthic fauna and flora (Abal et al., 1994). In addition, particle loading may also contribute to longer term

ASSESSING E COLOGICAL IMPACTS

21

changes through initial downstream settling, with resuspension into the water column at a later time.

Increases in the concentration of NH4+ in receiving waters from shrimp farms have been observed by many researchers, and as a result nutrient enrichment of poorly flushed embayments may occur (Gowen & Rosenthal, 1993). In some instances the level of impact has been sufficient to result in feedback which affects the aquaculture operation itself (Gowen et al., 1990). Evidence suggests that serious shrimp farm production losses resulting from the outbreak of disease in Asia and Latin America, are due to the environmental impacts of shrimp culture (Phillips et al., 1993). In addition to impacts on the aquaculture operation itself, shrimp farming has been linked to several cases of environmental degradation, however, despite this type of evidence, there is still a lack of quantitative data on the ecological impacts to receiving waters (Phillips et al., 1993).

The need for data on the ecological impact of aquaculture effluent has been identified (Gowen et al., 1990), and it has been shown that physical and chemical water quality monitoring techniques cannot provide this information. Bioindicators have long been used to determine ecological impacts of point source discharges (Worf, 1980; Kramer, 1994). For example, marine macrophytes can be used to provide insights into the ecological impacts of nutrients and suspended particulates by measuring changes in plant distributions, morphology, pigment concentrations and total tissue N (Lyngby, 1990; Alamoudi, 1994; Horrocks et al., 1995; Abal & Dennison, 1996; Udy & Dennison, 1997b). The morphology of seagrasses can change with reduced light availability, as a consequence of elevated concentrations of suspended solids in the water column (Abal et al., 1994). Seagrass

CHAPTER 2

22

distribution and depth penetration are also reduced as a consequence of reduced light availability in shallow estuarine systems (Dennison et al., 1993; Abal & Dennison, 1996).

Recently, marine plants have been used to detect and integrate the long term effects of small and /or pulsed nutrient inputs in well flushed oceanic systems (Costanzo, 1996), and elucidate the possible sources of the nutrient inputs (Jones et al., 1996). Macrophyte amino acid concentrations and composition have been shown to change with various N sources in both controlled laboratory experiments (Nasr et al., 1968; Di Martino Rigano et al., 1992; Jones et al., 1996) and field surveys (Udy & Dennison, 1997b). In particular, accumulation of the amino acids alanine, glutamine, proline and serine in plants (both terrestrial and marine) has been associated with N uptake, with different amino acids responding to different N sources (Steward and Pollard, 1962; Silveira et al., 1985; Lawlor et al., 1987; Kiladze et al., 1989; Di Martino Rigano et al., 1992; Vona et al., 1992; Heuer & Feigin, 1993). Stable isotope ratios of nitrogen (15N) have been used widely in marine systems as tracers of discharged nitrogen from point and diffuse sources, including sewage effluent (Rau et al., 1981; Wada et al., 1987; Van Dover et al., 1992; Macko & Ostrom, 1994; Cifuentes et al., 1996; McClelland & Valiela, 1998). Elevated 15N signatures in seagrass, mangroves and macroalgae have been attributed to plant assimilation of N from treated sewage effluent (Wada et al., 1987; Grice et al., 1996; Udy & Dennison, 1997b; Abal et al., 1998). This study, however, appears to be the first to use all these techniques to study the extent of impacts from aquaculture effluent.

In coastal marine systems a variety of point source inputs from aquaculture ponds, sewage treatment plants, fertiliser plants, agriculture and urban runoff can make it difficult to determine responsibility for ecological impacts (Grant et al., 1995). With increasing conflict

ASSESSING E COLOGICAL IMPACTS

23

between users of coastal resources, it has become essential to determine the specific influence of each source (Teichert-Coddington, 1995).

To assess the potential impact of aquaculture effluent, comparisons between the water volumes and water quality parameters of aquaculture effluent and treated sewage effluent have been conducted (Bergheim & Selmer-Olsen, 1982; Muir, 1982; Solbe, 1982; Macintosh & Phillips, 1992; Paez Osuna et al., 1997). However, there are very few studies comparing the impacts on the receiving waters (Pearson & Rosenberg, 1978; Cifuentes et al., 1996). Despite the differences in the two forms of waste, Pearson & Rosenberg (1978) hypothesised that the impacts of these two sources on receiving sediments would be similar.

Shrimp pond effluent has higher concentrations of suspended solids and phytoplankton (Ziemann et al., 1992), but lower concentrations of nutrients than sewage effluent (Muir, 1982). Dissolved nutrients in shrimp effluent are predominantly NH4+, whereas sewage effluent is proportionally higher in NO3-, and PO43- (Macintosh & Phillips, 1992). Shrimp effluent is typically produced in large volumes (Macintosh & Phillips, 1992), which can equate to up to 40% of the total inputs of N and P in some localised areas (Bergheim & Selmer-Olsen, 1982). Sewage is freshwater, whereas the salinity of shrimp effluent is typically 35-36 on the practical salinity scale. These differences may have a considerable impact on the fate of organisms in the receiving waters when effluent is released into shallow tidal estuaries. Both sewage and aquaculture effluent can be discharged intermittently, resulting in large diel fluctuations in water quality. Difficulties in monitoring these variable discharges can be overcome by the use of biological indicators, which integrate the impacts of these effluents over time (Costanzo, 1996). Unlike traditional chemical analyses of water

CHAPTER 2

24

column nutrients, these biological indicators reflect the availability of biologically available nutrients (Lyngby, 1990) which provides more ecologically meaningful information.

The aims of this study were to assess the influences on the receiving environment of wastewater discharges to a shallow estuarine system. Changes in receiving water and sediment quality analyses were compared with biological impacts measured as a consequence of shrimp farm and sewage effluent discharges. The region of influence of these two pollutant sources is defined, and mechanisms are suggested which may aid in discerning the relative impacts of these two discharges on a common receiving environment.

ASSESSING E COLOGICAL IMPACTS

25

2.2 Materials and Methods 2.2.1 Study Region Moreton Bay is a shallow coastal embayment on the east coast of Australia. The western side of the bay receives a variety point and non point source inputs including agricultural runoff, sewage and aquaculture effluent. The eastern bay is well flushed and influenced by oceanic waters. In eastern Moreton Bay, background concentrations of water quality parameters are: NH4+ < 2 M; NO3- ~ 0.1 M; PO43- ~ 0.2 M; chlorophyll a < 1 g L-1; TSS < 20 mg L-1, and typical 15N values for mangroves, seagrass and macroalgae in the eastern bay are 2-3 (Abal et al., 1998).

Two tidal creeks in close proximity (1.5 km apart) were studied in Moreton Bay, Australia (Fig. 2.1). One creek received discharge (18000 m3 d-1 containing 2.0 mg N L-1 and 0.2 mg P L-1, which equates to 36 kg N d-1 and 3.6 kg P d-1) from a shrimp farm (Jones, unpub. data). The creek was 2-3 m deep, approximately 1 km in length, and the shrimp farm discharge was 500 m from the mouth of the creek. The intensive shrimp farm (6 ha of ponds) was stocked with Penaeus japonicus (35 animals m-2). Ponds were routinely flushed (~20% per day) and water discharged into the creek on low tide. Except during the effluent discharge, the creek runs dry at low tide. The other creek (Eprapah Creek) received discharge (2400 m3 d-1 containing 4.5 mg N L-1 and 8.0 mg P L-1, which equates to 10.8 kg N d-1 and 19.2 kg P d-1) from a sewage treatment plant (Redland Shire Council, pers. comm.). The creek was approximately 2-5 m deep, 15 km in length, has a standing body of water at low tide, and the sewage discharge point was 2 km from the mouth. The sewage treatment plant serviced approximately 14 000 people and utilised secondary (activated sludge) treatment techniques. Both creeks were tidally flushed, and had virtually no freshwater flow during the study period (Autumn, 1997).

CHAPTER 2

26

2.2.2 Experimental Design Three sites were chosen in each creek, the first at the nutrient source (discharge site), the second approximately mid way between the nutrient source and the mouth (middle site), and the final at the mouth of the creek (mouth site) (Fig. 2.1). A site was positioned midway between both creeks and in close proximity to the shore (midway site). Several more sites were selected in Moreton Bay in a radiating pattern out from the creek mouths (Oyster Point, Sewage Plume and Cox Bank), including a reference site located approximately 4 km from the creek mouths. The creek banks at low tide extended the mouth of the sewage creek as far as the sewage plume site. At eight of the sites, traditional water quality parameters (dissolved N & P, total suspended solids, volatile suspended solids, sediment organic content, secchi depth, chlorophyll a, and physico-chemical parameters) were determined.
Oyster Point
Oyster Point Site
Brisbane 0 0.5 kilometres 1.0

Moreton Bay

Shrimp Farm

Mouth Site

le Discharge Midd Site Site

Cox Bank Site

Midway Sewage Site Plume Site Mouth Site Reference Site

Point Halloran
Middle Site Discharge Site

Coochiemudlo Island

Eprapah Ck

Sewage Treatment Plant

Victoria Point

Nutrient Source Sampling Site

ASSESSING E COLOGICAL IMPACTS

27

Figure 2.1 Map of study sites in Moreton Bay, including the location of shrimp and sewage effluent discharges.

Bioindicators were utilised at eleven sites with macroalgae and phytoplankton at all sites, mangroves at the creek sites, and seagrass at the bay sites. Amino acid composition was determined for macroalgae and seagrass, and the 15N signature and total tissue N was determined for all bioindicator species.

2.2.3 Collection Samples of seagrass (Zostera capricorni), mangrove (Avicennia marina), and macroalgae (Catenella nipae) were collected, placed on ice and returned to the laboratory and prepared for analysis of %N, 15N and amino acids. In the case of the seagrass and mangroves, the second youngest leaves were chosen, and for the macroalgae a single mangrove pneumatophore covered in macroalgae was collected for each replicate. Three replicates for each plant type were collected at each site.

2.2.4 Analytical Procedures Salinity was measured with a Horiba U-10 water quality meter (California, U.S.A.) and expressed on the Practical Salinity Scale.

Chlorophyll a concentration was determined by filtering a known volume of water sample through Whatman GF/F filters, which were immediately frozen. Acetone extraction and calculation of chlorophyll a concentration was performed using the methods of Clesceri et al. (1989), and Parsons et al. (1984).

Light-saturated phytoplankton productivity (potential productivity in mg C m-3 h-1) was determined in the laboratory using the 14C-bicarbonate incorporation method

CHAPTER 2

28

(Parsons et al., 1984). One hundred millilitres of water from each site was dispensed to three 120 ml polycarbonate bottles. A common dark control was established for each site by combining 33 ml of sample from each replicate into a fourth bottle wrapped in foil. Aqueous
14

C sodium bicarbonate (4 Ci) was added and bottles were incubated at a light intensity of 1100

to 1200 E m-2 s-1. A recirculating water bath and perspex heat shields maintained temperatures at ambient levels. After approximately two hours, water samples were filtered through 0.4 m polycarbonate filters (Poretics). The filters were placed into 5 ml scintillation vials and two drops of 5N HCl were added to each vial to drive off any remaining 14CO2. Four millilitres of scintillation fluid was added to each vial, and radioactivity as disintegrations per minute (DPM) determined using a scintillation counter (Packard Tricarb 1600TR, Meriden, Connecticut, U.S.A.). Total CO2 concentration in samples was determined from carbonate alkalinity using the method of Parsons et al. (1984).

Total suspended solids concentrations were determined using the methods of Clesceri et al. (1989). A known volume of water was filtered onto a pre-weighed and predried (110 C; 24 h) Whatman GF/C glass fibre filter. The filter was then oven dried at 60 C for 24 h and total suspended solids calculated by comparing the initial and final weights. Volatile suspended solids were determined as loss on ignition by combusting samples in a muffle furnace for 12 h at 525 C (Clesceri et al., 1989). The organic content of the sediment was determined from 10 cm deep core samples collected using 50 mL cut-off syringes. The sediment sample was combusted in a muffle furnace at 525 C for 12 h and the proportion of organic material determined by loss on ignition (Clesceri et al., 1989).

Dissolved inorganic nutrients (NH4+, NO3-/NO2-, and PO43-) were determined by filtering water samples through Whatman GF/F glass fibre filters and freezing them immediately on

ASSESSING E COLOGICAL IMPACTS

29

dry ice. Samples were analysed within two weeks by the NATA accredited Queensland Health Analytical Services Laboratory in accordance with the methods of Clesceri et al. (1989) using a Skalar autoanalyser (Norcross, Georgia, U.S.A.).

For analysis of plant total tissue N, 15N and amino acids, tissue was rinsed in distilled water to remove nutrients and sediment from the thallus surface, and then prepared for analysis. For calculation of total tissue N content and the 15N isotopic signature, samples were oven dried to constant weight (24 h at 60 C), ground and three sub-samples were oxidised in a Roboprep CN Biological Sample Converter (Europa Tracermass, Crewe, U.K.). The resultant N2 was analysed by a continuous flow isotope ratio mass spectrometer (Europa Tracermass, Crewe, U.K.). Total %N of the sample was determined, and the ratio of 15N to
14

N was expressed as the relative difference between the sample and a standard (N2 in air)

using the following equation (Peterson & Fry, 1987): 15N = (15N/14N (sample) / 15N/14N (standard) 1) x 1000 ()

For amino acid analysis, approximately 1.0 g wet weight of plant tissue was weighed and placed in 5 mL of 100% methanol (analytical reagent grade) for 24 h to extract amino acids. The methanol extract was filtered through Millipore Millex - HV13 (0.45 m) filters and injected into a post column derivatisation HPLC amino acid analyser (Beckman System 6300, Fullerton, California, U.S.A.), for detection of ninhydrin positive free amino acid groups at 570 nm. Results were calculated and expressed as mol g-1 wet weight. As well as detecting free amino acids, this technique also measures the concentration of free NH4+ in plant tissue. Changes in amino acid composition were used to infer nutrient source (either shrimp or sewage effluent). This technique was based on responses observed under ambient field, as well as controlled laboratory conditions using artificial nutrient additions (Jones et al., 1996; Udy & Dennison, 1997a).

CHAPTER 2

30

2.2.5 Statistical Analysis For all sampling techniques, three replicates were analysed and means and standard errors were calculated. Differences between treatments were tested for significance using one way analysis of variance (ANOVA) and Tukey's Test for multiple comparison of means at a significance level of 0.05 using Minitab 12.1 software (State College, Pennsylvania, U.S.A.).

ASSESSING E COLOGICAL IMPACTS

31

2.3 Results 2.3.1 Physical and Chemical Water Quality Analyses The concentrations of dissolved nutrients, chlorophyll a, phytoplankton productivity, total suspended solids, volatile suspended solids, and sediment organic content were different between the sewage and shrimp creek discharge sites. However, these parameters failed to detect an impact at the midway site (750 m beyond the mouths of the creeks), with values not significantly higher than at the reference site (4 km from the creek mouths) (Table 2.1).

2.3.1.1 Salinity All sites in the shrimp creek and in the bay were close to full salinity seawater (35-36). At the sewage creek discharge, the salinity was 29 as a consequence of freshwater inputs from the sewage effluent. Salinity increased downstream to 35 at the mouth site.

2.3.1.2 Nutrients The concentrations of dissolved nutrients (NH4+, NO3- / NO2- and PO43-) for each creek were highest at the discharge sites, and declined rapidly towards the mouth site. In particular, NH4+ concentration decreased to near eastern Moreton Bay concentrations at both creek mouth sites. The concentration of dissolved nutrients at the midway site (750 m from the creek mouths) was not significantly different (p > 0.05) from the reference site (~4 km from the creek mouths) (Table 2.1). The dissolved nutrient ratios were significantly different (p < 0.05) between the two discharge sites. At the sewage creek discharge site, NH4+ : NO3- / NO2- was 0.45, compared to 3.8 for the shrimp creek discharge site. The ratio of DIN (NH4+ + NO3- / NO2-) to DIP (PO43-) at the sewage creek discharge site was 3.0, compared to 24 at the shrimp creek discharge site. The concentrations of dissolved nutrients at the midway and

CHAPTER 2

32

reference sites were at eastern Moreton Bay levels, and the relative ratios of the dissolved nutrients were similar to the shrimp creek.

2.3.1.3 Phytoplankton Chlorophyll a concentration was not significantly different (p > 0.05) between the discharge site in the shrimp creek (10.8 g L-1) and the discharge site in the sewage creek (11.1 g L-1). However, at the creek mouth sites the concentration in the sewage creek (5.2 g L-1) was significantly lower (p < 0.001) than in the shrimp creek (17.9 g L-1). The concentration of chlorophyll a at the midway site (2.5 g L-1) was not significantly higher (p > 0.05) than the reference site (1.8 g L-1) (Table 2.1).

Despite the relatively low NH4+ concentration at the shrimp creek discharge site, the phytoplankton productivity (212 mg C m-3 h-1) was significantly higher (p < 0.001) than at the sewage creek discharge site (20 mg C m-3 h-1). However, the high productivity at the shrimp creek discharge site did not result in a significantly higher (p > 0.05) chlorophyll a concentration. The concentration of chlorophyll a in the shrimp creek increased from 10.8 g L-1 at the discharge site to 17.9 g L-1 at the mouth site, probably due to increased light availability resulting from the reduction in the concentration of inorganic and other suspended solids. Phytoplankton productivity at the midway site (9 mg C m-3 h-1) was not significantly different (p > 0.05) to the reference site (11 mg C m-3 h-1) (Table 2.1).

Table 2.1. Results of traditional water quality monitoring for the creek with shrimp farm effluent and sewage treatment effluent. DIN = Dissolved Inorganic Nitrogen; DIP = Dissolved Inorganic Phosphorus; Chl a = Chlorophyll a; Phyto Prod = phytoplankton productivity; TSS = total suspended solids; VSS = volatile suspended solids; Secchi = secchi disc depth. Only one replicate measurement was recorded for Secchi disk depth and salinity (Practical Salinity Scale).

Sampling Site Shrimp Discharge Shrimp Middle Shrimp Mouth Midway Sewage Discharge Sewage Middle Sewage Mouth Reference F Value

Salinity

NH4+ (M)

NO3-/NO2-(M) 1a 0.4a 0.3a 0.2a 65b 8a 2.9a 0.5a 15***

PO43(M) 0.2a 0.3a 0.2a 0.4a 31b 5.5a 2.1a 0.3a 34***

DIN: DIP Ratio 24c 7ab 13bc 3a 3a 2a 3a 6ab 13***

Chl a (g L-1) 10.8a 11.9ab 17.9b 2.5c 11.1ab 9.2ac 5.2ac 1.8c 15***

Phyto Prod (mg C m-3 h-1) 212d 87c 157b 9a 20a 20a 18a 11a 88***

TSS (mg L-1) 63.5a 51.5ab 40.2b 18.2d 44.3b 32.9bc 32.5bc 20.2d 35***

VSS (% of TSS) 19a 21a 26a 25a 35b 33b 28ab 28ab 14***

Secchi (m) 0.5 0.7 0.5 1.0+ 1.1 1.0 1.1 1.9

Sediment %Organic 6.0a 6.1a 8.3ab 7.1a 13.5c 7.5ab 5.6a 11bc 15***

35 35.5 36 36 29 33 35 36

3.8a 1.6a 2.3a 0.8a 29b 5.4a 2.4a 1.2a 40***

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

abc

Means with different letters are significantly different at p < 0.05.

CHAPTER 2

34

2.3.1.4 Suspended Solids and Secchi Depth The concentration of total suspended solids (TSS) at the shrimp creek discharge site (63 mg L-1) was significantly higher than the sewage creek discharge site (44 mg L-1). The concentrations at the midway (18 mg L-1) and reference sites (22 mg L-1) were not significantly different (p > 0.05) from each other, but were significantly (p < 0.05) lower than both creek mouth sites indicating significant sedimentation or dilution (Table 2.1).

The organic fraction of the suspended solids (volatile suspended solids) was significantly higher (p < 0.001) at the sewage discharge site (35%) compared to the shrimp discharge site (19%). In the shrimp creek the concentration of organic particles increased towards the mouth in proportion with the increasing chlorophyll a concentration (r2 = 0.60). In comparison, the concentration of organic particles in the sewage creek decreased in proportion with the concentration of chlorophyll a (r2 = 0.8) (Table 2.2).

Secchi disk depths did not vary along the length of either creek, from discharge site to mouth site. The mean secchi depth in the sewage creek (~1.0 m) was approximately double the depth in the shrimp creek (~0.6 m), but only half that of the reference site (1.9 m). The secchi depth at the midway site was greater than 1 m, but water depth was too shallow to obtain a measurement (Table 2.1).

2.3.1.5 Sediment Organic Content The organic content of the sediment (loss on ignition) in the shrimp creek increased from 6.0% at the discharge site to 8.3% at the mouth site, probably due to sedimentation. In the sewage creek, the organic content declined significantly (p < 0.001) from the discharge site

ASSESSING E COLOGICAL IMPACTS

35

(13.5%) to the mouth site (5.6%), probably due to senescence and subsequent sedimentation of phytoplankton and other organic particulates near the discharge site (Table 2.1).

Table 2.2. Correlations (r2) between the concentration of phytoplankton (chlorophyll a) and phytoplankton productivity ( 14C uptake) and various water quality parameters. DIN = Dissolved Inorganic Nitrogen; DIP = Dissolved Inorganic Phosphorus; Phyto Prod = phytoplankton productivity (mg C m-3 h-1); Chl a = Chlorophyll a (g L-1); TSS = total suspended solids (mg L-1); VSS = volatile suspended solids (mg L-1); ISS = inorganic suspended solids (mg L-1); Secchi = secchi disc depth (m). Numbers in bold type indicate significant correlations (r2 0.6).

TSS (mg L-1) Shrimp Creek Chl a Phyto Prod Sewage Creek Chl a Phyto Prod 0.60 0.27 - 0.85 0.21

VSS (mg L-1)

ISS (mg L-1)

NH4+ (M)

NO3-/NO2(M)

PO43(M)

DIN: Salinity DIP

- 0.60 0.48

- 0.87 0.19

0.12 0.93

0.51 0.56

0.14 - 0.81

0.09 0.94

0.86 0.19

0.80 0.51

0.37 0.11

0.66 0.34

0.63 0.32

0.66 0.35

0.04 0.25

- 0.85 - 0.60

2.3.2 Bioindicators In contrast to the water quality parameters, the responses of the bioindicator parameters at sites beyond the creek mouths were elevated compared to the reference site. For some of the parameters, the reference site appeared to be influenced by nutrients from the discharges.

2.3.2.1 Tissue Nitrogen Content The %N of the macroalgae was responsive to the nutrient sources, with the highest value at the sewage creek discharge site (3.1%), which was significantly higher (p < 0.05) than the shrimp discharge site (1.9%) (Table 2.3; Fig. 2.2). There was no decrease in the %N of the macroalgae with distance from the discharge site in the shrimp creek. However, in the sewage creek the macroalgae at the mouth site had a %N of 1.6%, which was not

CHAPTER 2

36

significantly elevated (p > 0.05) above the reference site (1.5%). The %N of the macroalgae at the midway site (2.3%) was significantly (p < 0.05) elevated above values in the macroalgae at both of the creek mouth sites (Table 2.3; Fig. 2.2).

The %N of seagrass leaves was significantly higher (p < 0.05) at the sewage creek mouth site (2.7%) compared with the shrimp creek mouth site (2.3%). The next three sites distant from the creek mouths (midway, Oyster Point, and Sewage Plume) were not significantly lower than the shrimp creek mouth site (2.3%). The seagrass %N at the next most distant site (Cox Bank) was 2.0%, which was not significantly higher (p < 0.05) than at the reference site (1.7%) (Table 2.3; Fig. 2.2).

The %N of the mangrove leaves appears less sensitive to nutrient inputs, with none of the mangroves at the other sites being significantly higher (p < 0.05) than the reference site (1.7%) (Table 2.3; Fig. 2.2).

2.3.2.2 15N Stable Isotope Ratio of Nitrogen The 15N isotopic signatures of the seagrass, macroalgae and mangroves were significantly different (p < 0.001) between sites (Table 2.3; Fig. 2.3). The highest 15N was in the macroalgae at the sewage creek discharge site (19.6), and the lowest in the macroalgae at the reference site (2.9). 15N in the macroalgae in the sewage creek decreased with distance away from the source. The value at the discharge site in the shrimp creek was 7.1, with no significant difference (p > 0.05) along the length of the shrimp creek to the mouth (7.9). The 15N at the midway site (6.4) was not significantly lower (p > 0.05) than the

ASSESSING E COLOGICAL IMPACTS

37

creek mouth sites, indicating influence of nutrients from the discharges. The 15N at the reference site (2.9) was significantly lower (p < 0.001) than all other sites.

The 15N of seagrass leaves at the sewage plume site (8.0) was significantly higher (p < 0.05) than all other sites. The 15N was not significantly different (p > 0.05) between the two creek mouths (7.1 and 6.8), but both were significantly higher (p < 0.05) than the midway (5.8), Oyster Point (4.7) and Cox Bank (5.5) sites, and the reference site (4.5) (Table 2.3; Fig. 2.3).

The highest 15N of mangrove leaves was 10.4 at the sewage discharge site, compared with 7.7 at the shrimp discharge site. Despite the significant differences (p < 0.05) at the source, the 15N values at the creek mouths were not significantly different (p > 0.05) from each other (4.9 and 4.6). The midway site is a small intertidal sand bank, and as such is a site of sediment deposition. The 15N of the mangroves (7.7) at the midway site was significantly higher (p < 0.05) than at both of the creek mouth sites, possibly due to deposition of 15N enriched particulates from the creeks. The 15N of the mangroves at the reference site (3.4) was significantly lower (p < 0.05) than all other sites (Table 2.3; Fig. 2.3).

CHAPTER 2

38

Table 2.3. Results of bioindicator monitoring for the creek with shrimp farm effluent and sewage treatment effluent. 15N = Nitrogen stable isotope ratio; %N = Tissue N content; nd = no data (no plants were present).

Sampling Site

Mangrove (Avicennia marina) 15N () %N 1.4b 1.7a 1.6a 1.6a 1.7a 1.4b 1.3b nd nd nd 1.7a 12.7***

Macroalgae (Catenella nipae) 15N () 7.1b 8.6b 7.9b 6.4b 19.6d 16.3c 6.4b nd nd nd 2.9a 70.5*** %N 1.9ab 1.8ab 1.9b 2.3c 3.1d 1.7ab 1.6ab nd nd nd 1.5a 49***

Seagrass (Zostera capricorni) 15N () nd nd 7.1de 5.8c nd nd 6.8d 8.0e 5.5bc 4.7ab 4.5a 33.3*** %N nd nd 2.3c 2.3c nd nd 2.7d 2.1bc 2.0ab 2.2bc 1.7a 19.6***

Shrimp Discharge Shrimp Middle Shrimp Mouth Midway Sewage Discharge Sewage Middle Sewage Mouth Sewage Plume Cox Bank Oyster Point Reference F Value

7.7d 6.2c 4.6ab 7.7d 10.4e 9.4e 4.9bc nd nd nd 3.4a 73.5***

p < 0.05; **p < 0.01; ***p < 0.001. abc Means with different letters are significantly different at p < 0.05.

ASSESSING E COLOGICAL IMPACTS

39

Oyster Point 2.2

0.5 kilometres

1.0

Moreton Bay

n Sampling Site Nutrient Source

Shr imp Farm 1.9 1.4 1.8 1.7 2.3 1.9 1.6

Brisbane

2.3 2.3 1.6 2.1 2.7 1.6 1.3

2.0

Point Halloran

1.7 1.5 1.7

1.7 1.4 3.1 1.7 Victoria Point

Coochiemudlo Island Seagrass Macroalgae Mangrove

Eprapah Ck Sewage Treatment Plant

Figure 2.2. Map showing the values of %N in seagrass ( Zostera capricorni), macroalgae ( Catenella nipae), and mangroves ( Avicennia marina) at the study sites (see Fig. 2.1 for site references).

CHAPTER 2

40

Oyster Point 4.7

0.5 kilometres

1.0

Moreton Bay

n Sampling Site Nutrient Source

Shrimp Farm 7.1 7.7 8.6 6.2 7.1 7.9 4.6

Brisbane

5.8 6.4 7.7 8.0 6.8 6.4 4.9

5.5

4.5 2.9 3.4 Point Hall oran Coochiemudlo Island

16.3 9.4 19.6 10.4 Victoria Point

Eprapah Ck Sewage Treatment Plant

Seagrass Macroalgae Mangrove

Figure 2.3. Map showing the values of 15N in seagrass ( Zostera capricorni), macroalgae ( Catenella nipae), and mangroves ( Avicennia marina) at the study sites (see Fig. 2.1 for site references).

2.3.2.3 Free Amino Acid Composition In the macroalgae, the total dissolved free amino acid concentration ranged from 1.7 to 6.5 mol g wet-1), with the highest concentrations at the shrimp (3.6 mol g wet-1) and sewage (4.1 mol g wet-1) creek mouth sites and the midway site (6.5 mol g wet-1). However, there was no direct correlation between the total amino acid concentration and dissolved nutrient availability, with the lowest concentration of amino acids being in the macroalgae at the sewage discharge site (1.7 mol g wet-1). The concentration at the reference site

ASSESSING E COLOGICAL IMPACTS

41

(3.7 mol g wet-1) was only significantly lower (p < 0.05) than the midway site (Table 2.4; Fig. 2.5).

The total free amino acid pool of the seagrass leaves proved to be more responsive to nutrient sources than the macroalgae, ranging from 5.3 to 16.7 mol g wet-1 (Table 2.4; Fig. 2.4). The total free amino acid concentration in the seagrass at the shrimp (16.7 mol g wet-1) and sewage (10.2 mol g wet-1) creek mouth sites and the midway site (10.4 mol g wet-1) were significantly higher (p < 0.05) than at the reference site (5.3 mol g wet-1).

The amino acid composition in the plants close to the nutrient sources was used to infer the source of nutrients being taken up by plants at more distant sites. The percentages of particular amino acids in the total free amino acid pool of the seagrass (glutamine and proline) and macroalgae (alanine and serine) were correlated to either the shrimp or sewage discharge source (Table 2.4; Fig. 2.4; Fig. 2.5). No other amino acids were significantly different (p > 0.05) between the creek discharge sites for the macroalgae and the creek mouths sites for the seagrass.

The %alanine (4.5%) in the macroalgae at the shrimp creek discharge site was significantly higher (p < 0.05) than the macroalgae at the sewage creek discharge site (2.7%). At the sewage discharge site, the %serine (10.3%) in the macroalgae was significantly higher (p < 0.05) than at the shrimp discharge site (6.6%). The %serine at the midway site (9.0%) was not significantly lower (p > 0.05) than the sewage discharge or sewage mouth sites. However, %serine at the reference site (6.8%) was significantly lower (p < 0.05) than at the sewage discharge site. In contrast, the %alanine at all sites, except at sites influenced directly by sewage (the sewage discharge and mouth sites) was not significantly lower (p > 0.05) than

CHAPTER 2

42

the shrimp discharge site, indicating that macroalgae at the reference site may have been influenced by the shrimp effluent (Table 2.4; Fig. 2.5).

In the seagrass at the shrimp creek mouth site, the %glutamine (30%) was significantly higher (p < 0.05) than at the sewage creek mouth (15%). On the other hand, the %proline was significantly higher (p < 0.05) in the seagrass at the sewage creek mouth site (59%) compared to the shrimp creek mouth site (44%). The %glutamine at the midway site (23%) was not significantly lower (p > 0.05) than at the shrimp mouth site, however all other sites were lower, with a minimum of 14% at the reference site. The %proline at all other sites, except the oyster point site (37%) were not significantly lower (p > 0.05) than at the sewage mouth site.

Table 2.4. Results of bioindicator monitoring for the shrimp and sewage creeks. % refers to percentage of total free amino acid pool. SER = serine; ALA = alanine; GLN = glutamine; PRO = proline; Total = total concentration of free amino acids (mol g wet -1); nd = no data (no plants were present).

Sampling Site

Macroalgae (Catenella nipae) %SER 6.6b 7.9ab 9.0ab 10.3a 6.5

ab

Seagrass (Zostera capricorni) %GLN %PRO Total mol g wet-1 nd 30a 23ab nd 15
b

%ALA 4.5a 3.0ab 3.8ab 2.7b 2.2 nd nd nd 2.9ab 5.9**

b

Total mol g wet-1 3.5bc 3.6bc 6.5a 1.7c 4.1

b

Shrimp Discharge Shrimp Mouth Midway Sewage Discharge Sewage Mouth Sewage Plume Cox Bank Oyster Point Reference F Value
*

nd 44b 50ab nd 59
a

nd 16.7a 10.4b nd 10.2b 13.5ab 7.7bc 6.2bc 5.3c 16.4***

nd nd nd 6.8b 6.5*

nd nd nd 3.7bc 9.5**

14b 18b 17b 14b 7.9***

61a 51ab 37b 46ab 5.0**

p < 0.05; **p < 0.01; ***p < 0.001. abc Means with different letters are significantly different at p < 0.05.

ASSESSING E COLOGICAL IMPACTS

43

Oyste r P oint

0.5 kilometres

1.0

Moreton Bay

n Sampling Site Nutrient Source

S hrim p Farm

Brisbane

P oint H allo ran Co och iem u dlo Island

Epra pah C k S ewag e Trea tm ent Plant

15000 10000

V ictoria Point

Proline Glutamine Other

5000

nmol

g -1

wet wt

Figure 2.4. Map showing the amino acid composition of seagrass ( Zostera capricorni) at the study sites (see Fig. 2.1 for site references).

CHAPTER 2

44

Oyster Point

0.5 kilometres

1.0

Moreton Bay

n Sampling Site Nutrient Source

Shr imp Farm

Brisbane

Point Halloran Coochiemudlo Island

5000

Eprapah Ck Sewage Treatment Plant

3000

Victoria Point

1000

Serine Alanine Other

nmol g-1 wet wt

Figure 2.5. Map showing the amino acid composition of macroalgae (Catenella nipae) at the study sites. Pie graphs have been reduced to quarters for layout purposes. The remaining three quarters of the pie graphs not represented are a continuation of the other amino acid category (not serine or alanine) (see Fig. 2.1 for site references).

ASSESSING E COLOGICAL IMPACTS

45

2.4 Discussion Impacts from the shrimp and sewage effluent discharged into the two creeks were significantly different, both within the creeks and in the receiving waters. In addition to differences in the nature of the effluent, the creeks were physically different, with the sewage creek being longer, wider and deeper. The sewage creek had a longer residence time, allowing for greater assimilation of nutrients and deposition of sediments within the creek itself before being released into Moreton Bay. The potential reduction in nutrients and suspended solids within the creek may have a marked effect on the impacts from the sewage creek compared with the shrimp creek which has a low residence time and high suspended solid load.

2.4.1Water Quality Parameters 2.4.1.1 Effluent Composition Although shrimp farm and treated sewage effluents have similar components, the relative concentrations and proportions can be significantly different (Macintosh & Phillips, 1992). The sewage discharge site had significantly higher concentrations of nutrients, but a lower N: P ratio than at the shrimp discharge site. The ratios of NH4+ to NO3- / NO2- and dissolved inorganic nitrogen (DIN) to dissolved inorganic phosphorus (DIP) at the sites in the shrimp creek were higher than in the sewage creek. These differences in nutrient composition may have implications regarding the phytoplankton community structure in receiving waters. Cyanobacteria and green flagellates in shrimp ponds have been observed to dominate over diatoms in conditions of lower light, higher NH4+ concentrations, and higher DIN: DIP (Burford, 1997).

CHAPTER 2

46

2.4.1.2 Phytoplankton Biomass and Productivity Ammonium is generally considered the most biologically available form of N for marine phytoplankton (McCarthy et al., 1977). This is consistent with the positive correlation of phytoplankton productivity to NH4+ concentration (r2 = 0.93) but not to NO3- / NO2(r2 = 0.56) observed in the shrimp creek (Table 2.2).

In contrast, phytoplankton productivity in the sewage creek was low (not significantly higher (p > 0.05) than at the reference site), and not significantly correlated (r2 = 0.34) to NH4+ concentration. The low phytoplankton productivity and diminishing chlorophyll a concentration (from 11.1 g L-1 to 5.2 g L-1) in the sewage creek was probably not a result of nutrient or light limitation, but rather due to senescence of freshwater phytoplankton with increasing salinity downstream (Ahel et al., 1996). This is evidenced by the strong negative correlation between chlorophyll a and salinity (r2 = 0.85).

Partitioning of nutrients within the creeks may be influenced by chlorophyll a concentrations, TSS and salinity gradients. The concentration of chlorophyll a in the sewage creek was positively correlated to TSS (r2 = 0.60), suggesting that the phytoplankton make up a considerable portion of the suspended particulates (Table 2.2). In comparison, 81% of the TSS at the shrimp discharge site was inorganic. Phytoplankton can take up significant amounts of NH4+ from the water column, whereas inorganic particulates usually have a high affinity for sorbing P (Pomeroy et al., 1965).

ASSESSING E COLOGICAL IMPACTS

47

2.4.2 Biological Indicators 2.4.2.1 Tissue N Content The tissue N content (%N) of marine plants is a potential indicator of biologically available nutrient concentrations (Gerloff & Krombholz, 1966; Duarte, 1990), especially in macroalgae (Horrocks et al., 1995) which have the ability to store large reserves of luxury nitrogen for metabolism during times of nutrient stress. All the %N values of the seagrass and macroalgae were shown to be higher at sites in close proximity to either the sewage or shrimp discharges, and at many of the sites within the bay the %N was elevated above the reference site.

In the shrimp creek, there was no decrease in %N of the macroalgae down the creek away from the discharge site. In contrast, the %N of the macroalgae at the sewage mouth site was significantly lower than at the discharge site, indicating significant assimilation of nutrients by the system within the length of the creek. These differences are probably due to the greater length of the sewage creek allowing for much greater assimilation and deposition of nutrients (Cifuentes et al., 1996).

The %N content in the mangrove leaves had no correlation with nutrient availability, and were relatively unchanged between all sites, despite significant (p < 0.05) changes in 15N. This is consistent with another study which found no significant difference between the %N of mangroves throughout the Moreton Bay, thereby making them less sensitive as indicators of nutrient inputs than seagrasses and macroalgae (Rogers, 1998).

CHAPTER 2

48

2.4.2.2 15N Isotopic Signature The 15N of raw sewage and treated sewage particulates discharging into Moreton Bay is around 5.1 and 9.2 respectively (Loneragan et al., in prep.). The elevated 15N signature subsequent to treatment of the sewage effluent is a result of isotopic fractionation during ammonia volatilisation, nitrification and denitrification (McClelland & Valiela, 1998). The 15N of particulates in the ponds of the shrimp farm discharging into the shrimp creek is around 6 (Preston, unpub. data). This is considerably higher than the value of 0.9 reported for suspended particulate matter in shrimp effluent in Ecuador (Cifuentes et al., 1996). The farm in the present study uses artificial pelleted feeds made predominantly of shrimp meal, compared with the farm studied by Cifuentes et al. (1996) in Ecuador which relies predominantly on natural feed (phytoplankton). The markedly greater 15N of the effluent from the farm in the present study may be a result of the animal derived N in the pelleted feeds versus the plant derived N at the farm in Ecuador.

The 15N of seagrass (Udy & Dennison, 1997b), macroalgae (Abal et al., 1998) and mangroves (Rogers, 1998) in Moreton Bay has been shown to be positively correlated to proximity to nutrient sources. The 15N of seagrass at sewage impacted sites is approximately 10 (Udy & Dennison, 1997b), which closely reflects the 15N of treated sewage effluent (9.2) discharged into Moreton Bay (Loneragan et al., in review). In comparison, the 15N of seagrasses at unimpacted eastern bay sites ranges from 2 to 3 (Udy & Dennison, 1997b). In the present study, the 15N stable isotope signature of seagrasses, mangroves and macroalgae ranged from 2.9 to 19.6.

ASSESSING E COLOGICAL IMPACTS

49

Variations in 15N may be more greatly influenced by isotopic fractionation than by changes in the relative contributions of the N sources (Fogel & Cifuentes, 1993). The enrichment of the 15N of NH4+ in estuaries is mediated predominantly by nitrification of NH4+ (Mariotti et al., 1984; Cifuentes et al., 1989; Fogel & Cifuentes, 1993). The 15N values recorded for the macroalgae at the sewage creek discharge site (19.6) are among the highest reported in the literature (Owens, 1987). Given the high concentrations of NO3- / NO2- in the sewage creek, nitrification probably accounts for the high 15N observed in the mangroves and macroalgae, which would predominantly take up the isotopically heavy NH4+, in preference to NO3- / NO2- (Hanisak, 1983). The longer residence times of the sewage creek (compared with the shorter shrimp creek) may also increase the 15N of the organic N, NH4+ and NO3- / NO2- due to further isotopic segregation. The plants may also be taking up NO3- / NO2- or NH4+ or organic N, including dissolved free amino acids (Hanisak, 1983). Uptake of organic forms of N such as urea require more metabolic energy than uptake of NH4+ (Wheeler, 1983), however, direct uptake of DFAAs may be possible and more energy efficient (Jrgensen, 1982).

The lower 15N of the mangroves and macroalgae in the shrimp creek reflects the lower initial 15N of the shrimp effluent (Preston, unpub. data). The apparent lack of isotopic fractionation by bacteria within the shrimp creek may be due to less NH4+ being available for nitrification because the NH4+ in shrimp ponds is typically taken up by phytoplankton and bacteria, rather than oxidised by nitrifying bacteria (Hargreaves, 1998). Therefore, there is usually less enrichment of the 15N isotopic signature in shrimp effluent than in sewage effluent, even though the initial values may be similar. The sewage effluent also contained a much greater proportion of organic material than the shrimp farm effluent. This organic material has the potential to become more greatly enriched in 15N than NH4+ or NO3- due to

CHAPTER 2

50

isotopic fractionation at each step in the microbial conversion of organic N to NH4+, subsequently to NO3- and finally to N2 gas via denitrification (Handley & Raven, 1992). The high concentrations of NO3- / NO2- resulting from nitrification in the sewage effluent may also act as the substrate for denitrification, thereby elevating the 15N of the remaining NO3- / NO2-.

Strong isotopic enrichment was observed in the macroalgae compared to the mangroves at the sewage discharge site. The relatively low concentrations of suspended particulates in the sewage creek may have reduced the transfer of nutrients to the sediments via sedimentation of material with incorporated or adsorbed nutrients. Mangroves obtain their nutrients from the sediment, and as such may have sufficient nutrients available so that they can preferentially take up isotopically light 14N, whereas the macroalgae may be N limited and will therefore take up the both 14N and the heavier 15N isotope (Wada, 1980). Uptake of nutrients from nitrogen fixation associated with decomposing leaves may also provide high concentrations of N with a much less enriched signature (Hicks & Silvester, 1985).

In contrast to the sewage creek discharge site, the mangrove and macroalgal 15N at the shrimp creek discharge site were not significantly different. This may be due to sedimentation of high concentrations of N rich particulates from the shrimp effluent, thereby making these nutrients available to the mangroves. The 15N of the mangroves and macroalgae in the shrimp creek are similar at the discharge site, but diverge downstream with the mangroves having a lower 15N, while the macroalgal signature remains unchanged. This rapid drop in the 15N of the mangroves downstream is probably a result of significant initial concentrated sedimentation of N rich particulates near the discharge. The next major

ASSESSING E COLOGICAL IMPACTS

51

depositional zone appears to be at the sewage plume site (the mouth of the creek at low tide) with a seagrass 15N of 8.0.

The values for 15N in the mangroves (3.4) and seagrass (4.5) at the reference site were higher than those reported for sites in the eastern bay and other western bay sites which are not impacted by point source nutrient discharges (2-3) (Udy & Dennison, 1997b; Abal et al., 1998; Rogers, 1998). These elevated values indicate that the seagrass (and possibly the mangroves) at the reference site are receiving nutrients from point source discharges. Seagrass and mangroves obtain most of their nutrients from the sediment (Iizumi & Hattori, 1982; Short & McRoy, 1984; Ziemann et al., 1984), suggesting that biologically available nutrients from the effluent sources are being transported to the sediment. However, the values of 15N in the seagrass at the two creek mouth sites (6.8 and 7.1) were similar, and so 15N values alone cannot be used to distinguish between shrimp and sewage effluent sources.

2.4.2.3 Amino Acid Composition The seagrass at the shrimp creek mouth site had a significantly higher total free amino acid pool than at the sewage creek mouth site. Increases in the free amino acid pool of both terrestrial and marine plants have been linked to elevated nutrient availability (Nasholm et al., 1994; Ohlson et al., 1995; Jones et al., 1996; Udy & Dennison, 1997b; Udy & Dennison, 1997a). However, there is evidence to suggest that increases in amino acids may indicate growth limitation by factors such as light (Vergara & Niell, 1995) or nutrients (Di Martino Rigano et al., 1992). Limitation of growth may thereby prevent metabolism of free amino acids into proteins. Seagrass at the shrimp creek mouth site have lower growth rates and biomass than several other sites in Moreton Bay, including those close and distant from

CHAPTER 2

52

nutrient sources (Udy & Dennison, 1997b). The comparatively shallow secchi depth at the shrimp creek mouth (50% of the sewage creek mouth site) indicates that the reduced seagrass growth rate may be due to light limitation, thereby resulting in the storage of excess nutrients as amino acids (Udy & Dennison, 1997a).

The macroalgae at the sewage discharge site had the lowest amino acid concentration, but the highest %N content. It has been shown in macroalgae that uptake of NO3- can represent a temporary N storage pool which contributes significantly more to the cellular N content than the dissolved free amino acid pool (Naldi & Wheeler, 1999). The high NO3- / NO2- to NH4+ ratio of dissolved inorganic N in the sewage creek may have stimulated uptake of NO3- by the biota. Although most species usually prefer NH4+ because it does not need to be reduced (D'Elia and DeBoer, 1978), macroalgae can take up NO3- and NH4+ simultaneously (Thomas et al., 1987).

The specific free amino acid composition in the seagrass leaves and macroalgae near to the discharges was shown to reflect the different effluent sources. Previous studies have linked the source of N (NH4+, NO3- / NO2- or organic N) with the amino acid composition of various plants, including macroalgae (Nasr et al., 1968; Di Martino Rigano et al., 1992; Sarimento, 1992; Jones et al., 1996; Flynn et al., 1997). In particular, glutamine concentration is known to be associated with NH4+ uptake and can inhibit uptake of NO3- (Nasr et al., 1968). The presence of elevated proportions of glutamine in the seagrass at the shrimp creek mouth site was reflective of the high NH4+ (relative to the concentration of NO3- / NO2-). The presence of alanine and glutamine in plant tissue is associated with N metabolism, and increases have been observed in response to an increase in N availability (Steward and Pollard, 1962; Vona et al., 1992), particularly with NH4+ and urea (Nasr et al., 1968).

ASSESSING E COLOGICAL IMPACTS

53

The concentration of dissolved free amino acids in marine plants has also been linked directly with the concentration of amino acids in the water column, due to direct uptake from the water column (Jrgensen, 1982; Hanisak, 1983). It was also determined that the concentrations of these amino acids was independent of the water column NH4+ or NO3- / NO2- concentrations. In particular, direct uptake of glutamine, alanine and serine was observed (Jrgensen, 1982).

Under controlled laboratory conditions, macroalgae incubated in shrimp farm effluent has been shown to store both alanine and glutamine (Jones et al., in prep b; Chapter 4). The relative proportions (as a percentage of the total dissolved free amino pool) correlated directly (r2 = 0.66 for alanine; r2 = 0.60 for glutamine) with the concentration of particulates and nutrients in the effluent (Jones et al., in prep b; Chapter 4). Alanine is used by shrimp for osmoregulation and is found in high concentrations in shrimp heads (Teerasuntonwat & Raksakulthai, 1995) which is one of the major components in pelleted shrimp feeds. Up to 25% of the feed pellets applied to the ponds are uneaten and become dissolved in the effluent (Lin et al., 1993), thereby enriching the effluent in dissolved free amino acids such as alanine. Glutamine (which occurred in elevated concentrations in seagrass at the shrimp creek mouth site) is one of the main amino acids in shrimp muscle tissue (Lee et al., 1989). It is also stored in large concentrations in the haemolymph to overcome ammonia toxicity, which can often occur in the pond environment (Chen et al., 1994). Alanine and glutamine are two of the dominant amino acids in shrimp cuticles (Horst, 1989). After moulting the cuticles are broken down, thereby increasing the concentrations of these amino acids in the effluent.

CHAPTER 2

54

The relationships described above provide support for the use of alanine and glutamine as indicators of uptake of shrimp effluent derived nutrients by biota in the receiving waters. The elevated proportions of alanine (%alanine) in the macroalgae at the shrimp creek discharge site may be reflecting the relatively high availability of NH4+, or may be related to the presence of elevated concentrations of alanine in the water column. The magnitude of the responses in the %alanine observed between the shrimp and sewage discharge sites are similar to the responses to NH4+ versus NO3- supply under controlled laboratory conditions (Nasr et al., 1968).

The %serine in the macroalgae was correlated with the proximity to the sewage discharge site, which contained predominantly NO3- / NO2- as the main form dissolved inorganic N. The marine alga, Cyanidium caldarium has been shown to accumulate more serine grown in NO3- than in NH4+ (Di Martino Rigano et al., 1992). Serine has been shown to accumulate in some terrestrial plants in response to increasing NO3- (Silveira et al., 1985; Lawlor et al., 1987; Kiladze et al., 1989), but not with NH4+ supply. In addition to synthesis of serine, some marine plants can take it up directly from the water column (Jrgensen, 1982; Alamoudi, 1988). These relationships suggest that the elevated concentrations of serine in the macroalgae near the sewage discharge were related to the high NO3- concentrations, or perhaps the presence of high concentrations of dissolved serine in the effluent.

The %proline of the seagrass in close proximity to the sewage creek was significantly higher relative to other sites. Proline accumulation in plants as a response to heavy metal toxicity is widespread (Sharma et al., 1998). Plants incubated in sewage effluent can have elevated proline concentrations in response to high cadmium and lead levels found in sewage effluent (Mohan & Hosetti, 1998). In terrestrial plants, proline has been observed to increase in

ASSESSING E COLOGICAL IMPACTS

55

response to NO3- uptake (Heuer & Feigin, 1993), and proline has been shown to stimulate nitrate reductase activity in aquatic plants (Salonen & Simola, 1989), probably by facilitating iron transport which is essential for nitrate reductase (Wilkinson, 1994). Accumulation of proline in response to either heavy metals or nitrate in the sewage effluent may indicate its potential as an indicator in seagrasses.

At the reference site the %proline in the seagrass was elevated above concentrations measured in at unimpacted eastern Moreton Bay sites (Udy & Dennison, 1997b), and was indicative of the response in the seagrass at the sewage creek mouth site. This is consistent with the high 15N value in the seagrass relative to the macroalgae at this site. This reflects the availability of effluent derived N in the sediment, but not in the water column.

Tracing changes in the amino acid composition of seagrass and macroalgae at various distances from the discharges indicated that the N from both sewage and shrimp discharges may be influencing these macrophytes.

2.4.3 Comparison of Impacts Sewage effluent is freshwater, relatively low in suspended solids, and high in NO3- and PO43-. In contrast, shrimp effluent is saltwater, contains a high concentration of highly productive phytoplankton, total suspended solids, NH4+ and organic nitrogen (both of which are very biologically available forms of nitrogen) (Macintosh & Phillips, 1992; Ziemann et al., 1992). The high concentrations of biologically available nutrients in the effluent stimulate high rates of primary productivity, resulting in production of phytoplankton, thereby reducing light availability for seagrasses (Abal & Dennison, 1996) and subsequently increasing sedimentation (Frid & Mercer, 1989).

CHAPTER 2

56

Both adsorbed and interstitial nitrogen are higher at the sewage creek mouth than the shrimp creek (Udy & Dennison, 1997b), and are elevated in comparison to the rest of Moreton Bay (Abal et al., 1998). Because N is typically the limiting nutrient in Moreton Bay (Horrocks et al., 1995; O'Donohue & Dennison, 1997; Udy & Dennison, 1997a; Jones et al., 1998; Appendix 1), and NH4+ is the preferred N source for seagrass (Iizumi & Hattori, 1982), the high concentrations of NH4+ in the sediments at the sewage creek mouth site resulted in higher seagrass shoot density and growth rate (Udy & Dennison, 1997b). Low sediment NH4+ concentrations at the shrimp creek mouth may be a result of the high concentrations of phytoplankton limiting the movement of NH4+ to the sediment (Hargreaves, 1998). NH4+ can also be sorbed to clay particles, but is more likely to be associated with organic particles in the sediment (Rosenfeld, 1979). The high concentration of particulates at the shrimp creek mouth resulted in reduced light availability. The seagrass canopy height at the shrimp creek mouth is higher than at the sewage creek mouth (Udy & Dennison, 1997b), which is consistent with responses observed for light limitation (Abal et al., 1994). Despite the results of some studies regarding the similarity of impacts on sediment (Pearson & Rosenberg, 1978), the results from the present study reveal that there may be quite different impacts from shrimp versus sewage waste.

Results from the seagrass amino acid composition (%proline) and the 15N signature suggest that impacts of the sewage nutrients may be experienced further from the discharge than shrimp impacts. The greater response by the seagrass at the reference site is postulated as being due to transport of nutrients to the sediments. This correlates to observations that impacts from waste discharges on the sediments can extend further than water column impacts (Wu, 1995). See Figure 2.6 for a conceptual representation of these interpretations.

ASSESSING E COLOGICAL IMPACTS

57

2.4.4 Conclusion Physical and chemical water quality analyses alone would have concluded that there was no influence by either the shrimp or sewage effluent at the midway site, which is approximately 750 m from each creek mouth. This is consistent with the findings of Samocha & Lawrence (1997) who found that shrimp effluent could not be detected at a distance of 400 m from the source, and Hensey (1991) who observed no impacts from fish farming using water quality sampling techniques. This indicates that the nutrients are being taken up by the system (sediments and biota), reduced by dilution, or being lost to the atmosphere through volatilisation and denitrification.

Several workers have suggested that aquaculture and other wastewater inputs should be restricted so that their discharge is within the assimilative ability of the ecosystem (Omori et al., 1994). In the present study, the concentrations of dissolved nutrients, chlorophyll a, and total suspended solids were at eastern Moreton Bay concentrations 750 m from the creek mouths, and the biota had increased %N, 15N and free amino pool. This suggests that system was assimilating the nutrients being discharged. However, it can be argued that the assimilation of nutrients by the system does not indicate an absence of ecological impact (Welch, 1992). This is particularly important given recent interest in using mangrove forests as filters for shrimp effluent (Robertson & Phillips, 1995).

Although amino acid composition, 15N signature and other bioindicator parameters do not directly elucidate environmental impact, these physiological changes may help to identify incipient environmental degradation before large scale changes occur in water quality and community structure, which are the usual signs of environmental degradation.

CHAPTER 2

58

Application of bioindicators was successful at identifying the separate impacts from shrimp and sewage effluent, and in contrast to chemical and physical water quality analyses, bioindicators identified spatially broader impacts on ecosystem biota. The 15N isotopic signature and amino acid composition of the macroalgae appeared to be most sensitive to dissolved nutrients. The same parameters in the seagrass appeared to be sensitive to sediment nutrients, thereby elucidating depositional zones of point source derived nutrients.

This study represents one of the first attempts at detecting the influence of shrimp effluent on receiving waters using marine plants as bioindicators. It also attempted to distinguish this influence from sewage, another common point source input. To further develop these techniques for broad application will require more controlled laboratory experiments to enable the elimination of confounding factors to isolate the bioindicator responses from undiluted nutrient sources.

2.4.5 Application for other types of Aquaculture The impacts of aquaculture effluent can vary considerably depending primarily on the quality of water supplying the ponds (Boyd & Musig, 1992; Pruder, 1992), the farm management practices (feeding rates, water exchange), pond soil characteristics, proximity to other farms, and the flushing rates of the receiving water body (Gowen et al., 1990).

Shrimp farming in Australia is still a small but burgeoning industry, with production increasing from 15 t to 2,000 t over the last 14 years (Preston, pers. comm.). The relatively large geographical spread of existing farms means that overcrowding is not currently a problem, in contrast to certain regions in Asia and Latin America (Phillips et al., 1993).

ASSESSING E COLOGICAL IMPACTS

59

The two creeks studied are small, and the receiving waters are well flushed and low in nutrients relative to many of the worlds sites of aquaculture discharge where the quality of the intake / receiving waters can be so poor as to impact on the aquaculture operation itself (Chua et al., 1989; Aitken, 1990; Wong et al., 1992). The farm discharging into the shrimp creek consisted of only 6 ha of ponds, and so represents a minor input relative to more densely farmed regions, where the effluent from several farms are discharging into the same water body. In some regions, with a dense distribution of farms, the effluent from one farm can become anothers intake (New, 1990 cited in Macintosh & Phillips, 1992). The inputs of sewage effluent were also relatively small, with the treatment facility servicing only 14 000 people. Monitoring of small scale discharges like these can provide an indication of how small inputs can still have a significant impact on receiving waters.

Studies into the impacts on receiving waters for shrimp farming have been very poorly quantified in comparison to other forms of aquaculture (Phillips et al., 1993). In comparing the impacts between different operations, there are several differences that need to be considered. Intensive culture of shrimp has the highest water volume requirements of all forms of aquaculture (Phillips et al., 1991). Sedimentation which works well for some forms of aquaculture is typically not as effective for shrimp because of clay particles and high concentrations of phytoplankton which do not easily settle out (Macintosh & Phillips, 1992).

2.4.6 Remediation Options To reduce environmental impacts, reductions in nutrient and sediment discharge can be achieved with changes to management practices (reviewed in Allan et al., 1995; Hopkins et al., 1995b). Options include more regular feeding to help reduce wastage (Villalon, 1991

CHAPTER 2

60

cited in Samocha & Lawrence, 1997), proper design of settlement ponds to better promote sedimentation (Samocha & Lawrence, 1997), reduced water exchange (Hopkins et al., 1993b), sludge removal (Hopkins et al., 1994) and biological filtration using oysters and macroalgae (Wang & Jakob, 1991; Hopkins et al., 1993a; Shpigel et al., 1993b; Samocha & Lawrence, 1997; Jones et al., in prep b; Chapter 4; Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5). In particular, polyculture with oysters and macroalgae is regarded as having considerable environmental and economic benefits (Chandrkrachang et al., 1991; Macintosh & Phillips, 1992), but further research and development is needed (Macintosh & Phillips, 1992).

Figure 2.6. Conceptual model of the two creeks and the range and type of impacts from the different effluent sources.

CHAPTER 3 OYSTER FILTRATION OF SHRIMP FARM EFFLUENT, THE EFFECTS ON WATER QUALITY

Abstract
Shrimp pond effluent water can contain higher concentrations of dissolved nutrients and suspended particulates than the influent water. Consequently, there are concerns about adverse environmental impacts on coastal waters due to eutrophication and increased turbidity. One potential method of improving effluent water quality, prior to discharge or recirculation, is to use bivalves to filter the effluent. This study examined effects of the Sydney Rock Oyster, Saccostrea commercialis (Iredale and Roughley) on the water quality of shrimp pond effluent. Effluent from a shrimp farm stocked with Penaeus japonicus (Bate) was pumped directly into tanks (34 L) stocked with different densities of oysters. Combinations of live and dead oysters were used to test the effects of three different densities of live oysters (24, 16 and 8 live oysters per tank). The concentrations of total suspended solids, proportion of organic and inorganic matter, total nitrogen, total phosphorous, chlorophyll a and the total number of bacteria in the pond effluent water were determined before and after filtration by oysters. The oysters significantly reduced the concentration of all the parameters examined, with the highest oyster density having the greatest effect. Shrimp pond effluent contained a higher proportion of inorganic matter (72%) than organic matter (28%). The organic component appeared to be mainly detritus, with chlorophyll a comprising only a minor proportion. Filtration by the high density of oysters reduced the effluent total suspended solids to 49% of the initial level, the bacterial numbers to 58%, total nitrogen to 80% and total P to 67%. The combined effects of settlement and oyster filtration reduced the concentration of chlorophyll a to 8% of the initial effluent value.

CHAPTER 3

64

3.1 Introduction In shrimp farming, the addition of feeds and action of pond aerators can result in increased nutrient and sediment loads in pond effluent compared to the influent water (Phillips et al., 1993; Briggs & Funge-Smith, 1994). Consequently, there are concerns about adverse environmental impacts on coastal waters due to increased turbidity and eutrophication (Ziemann et al., 1992; Hopkins et al., 1993a). There is a need to develop more effective controls of the quality of effluent water prior to discharge into the environment or recirculation (Phillips et al., 1993; Primavera, 1994).

One potential method of reducing adverse environmental impacts, and recapturing otherwise wasted nutrients, is to use bivalves to filter the pond effluent (Lin et al., 1993). Pond effluent contains organic matter including bacteria, phytoplankton, and detritus (Ziemann et al., 1992) that could provide food for bivalves such as oysters (Hopkins et al., 1993a). Pond effluent can also contain a high proportion small inorganic particles (Hopkins et al., 1995b) that could be removed from suspension by oyster filtration and subsequently expelled as larger, more settleable, particles in the form of pseudofaeces (Tenore & Dunstan, 1973).

This study investigated the filtration effects of oysters, Saccostrea commercialis on the water quality of shrimp farm effluent. The research was conducted in Moreton Bay, Australia (Fig. 3.1), a region that supports shrimp farming and traditional oyster cultivation. In Moreton Bay shrimp farming and oyster cultivation are both seasonal activities. In winter (June to August) shrimp ponds are empty and oysters are generally not harvested from the bay due to their poor condition. In spring (September to November) shrimp ponds are fertilised to promote phytoplankton growth and ponds are stocked with shrimp post-larvae. Oyster growth and conditioning occurs during spring and summer when phytoplankton

EFFECTS OF O YSTER F ILTRATION ON WATER Q UALITY

65

densities in the bay increase (Dennison et al., 1993). Oyster growth to marketable size takes two to three years (Witney et al., 1988).

This study used a system of tanks supplied with water pumped from the effluent canal of a commercial shrimp farm stocked with P. japonicus. The farms management practices can be considered intensive (25 animals per m2), with a regime of periodic water exchange and no recirculation. The objective of the study was to determine the effects of oyster filtration on pond effluent water quality.

CHAPTER 3

66

Figure 3.1. Location map of Moreton Bay Prawn Farm near Brisbane, Australia.

EFFECTS OF O YSTER F ILTRATION ON WATER Q UALITY

67

3.2 Materials and Methods 3.2.1 Experimental Design The effects of oyster filtration on shrimp pond effluent were determined by monitoring the inflow and outflow water in each of 15 outdoor plastic tanks (60 cm 40 cm 25 cm) stocked with oysters (Fig. 3.2). Each tank was supplied with 34 L of water pumped directly from the effluent channel of a commercial shrimp farm. The effluent channel received water from 6 x 1 ha ponds with a stocking density of approximately 25 shrimp m2. At the time of the experiments (Summer 1995 / 1996) the mean wet weight of the shrimp was approximately 10 g.

Tank Layout
Water Inflow

Water Outflow

Single Tank with oyster tray

Outflow

25 cm

Inflow
60 cm
Figure 3.2. Schematic representation of tank and waterflow layout.

40 cm

The 15 tanks were stocked with oysters (S. commercialis), the mean wet weight of individual oysters was 55 g. In the experimental design, combinations of live and dead oysters (empty shells) were used to produce 3 different densities of live oysters; low, medium and high plus

CHAPTER 3

68

one treatment of dead oysters (Table 3.1). There were 3 replicates for each density of oysters, and a control treatment of no oysters. The dead oysters were included to elucidate any changes in water circulation and settling characteristics effected by the physical characteristics of the oyster shells.

Table 3.1 Combinations of live and dead oysters ( Saccostrea commercialis) used in experiments to determine the effects of oyster density on the water quality of shrimp pond effluent.

Treatment Control Shells Low Density Medium Density High Density

Description No oysters 24 dead oysters 8 live oysters and 16 dead 16 live oysters and 8 dead 24 live oysters

Live Oyster Density (m2) 0 0 33 67 100

The inflow water samples were collected at the start of the experiment, and the outflow samples, after a period of two hours during which the oysters were filtering the effluent (no water flow to simulate conditions in a shrimp farm treatment pond). Sampling consisted of collecting three replicate one litre containers of water from the inflow pipe and from the outflow point of each of the 15 tanks. The water samples were returned the laboratory where they were filtered for chlorophyll a extraction and TSS calculation. Sub-samples were taken for total N and P, bacterial numbers and determination of the organic/inorganic ratio.

3.2.2 Analytical Procedures Chlorophyll a was determined by filtering a known volume of water sample through Whatman GF/C filters, which were immediately frozen. Acetone extraction and calculation

EFFECTS OF O YSTER F ILTRATION ON WATER Q UALITY

69

of chlorophyll a concentration was performed using the methods of Clesceri et al. (1989), and Parsons et al. (1984).

Total suspended solids concentrations were determined using the methods of Clesceri et al. (1989). A known volume of water was filtered onto a pre-weighed and predried (110 C; 24 h) Whatman GF/C glass fibre filter. The filter was then oven dried at 60 C for 24 h and total suspended solids calculated by comparing the initial and final weights. Volatile suspended solids were determined as loss on ignition by combusting samples in a muffle furnace for 12 h at 525 C (Clesceri et al., 1989).

Unfiltered samples for nutrient analysis (total Kjeldahl nitrogen and total phosphorus) were collected in 120 mL polycarbonate immediately frozen. They were subsequently analysed within two weeks by the NATA accredited Queensland Health Analytical Services Laboratory in accordance with the methods of Clesceri et al. (1989) using a Skalar autoanalyser (Norcross, Georgia, U.S.A.).

Bacteria samples were preserved with 2% formalin and kept at 4C until analysis. A known volume (0.5 mL - 1 mL) of sample was stained with acridine orange, filtered onto a stained (Irgalan Black) 2m poretics filter and mounted on a slide. Bacteria were counted using epifluorescence microscopy (Hobbie, 1977).

CHAPTER 3

70

3.3 Results 3.3.1 Suspended Solids The mean concentrations of TSS (and organic/inorganic ratio), total nitrogen, total phosphorous, chlorophyll a and the total number of bacteria in the pond effluent water before and after filtration by oysters are summarised in Table 3.2.

Oysters were effective in removing total suspended solids (TSS) from pond effluent with significant reductions by the medium and high oyster densities and no significant reduction in the low density oysters and control treatments. The reduction in effluent TSS varied with oyster density. At the high density the concentration of TSS was reduced to 49% of the initial level in the pond effluent, a significantly greater reduction than at the low density treatment (80%). The level of reduction at the medium density (64%) was not significantly different from the high or low densities.

3.3.2 Organic content Combustion of pond effluent filtrate showed that the effluent contained a higher proportion of inorganic matter (72%) than organic matter (28%). Comparison between the inflow and the control treatments showed that there was no significant settling of total organic or total inorganic matter during the experiment. The oysters removed both organic and inorganic material from pond effluent approximately in proportion to the concentrations initially present.

Table 3.2 Concentration of various water quality parameters before and after filtration by oysters at 3 different densities (see Table 3.1). Values for control (no oysters) and shells (dead shells only) are also given. Values in brackets are concentrations expressed as a percentage of the inflow value. Values in italics are standard errors.

Treatment

TSS g L (%)
-1

Organic g L (%)
-1

Inorganic g L (%)
-1

Total N mg L (%)
-1

Total P mg L (%)
-1

Bacteria no. 10 mL (%)

6

Chlorophyll a g L-1 (%)

Inflow

0.13a 0.0052 0.13a (100) 0.0019

a

0.035a 0.0007 0.038a (110) 0.004 0.036 (104) 0.0006 0.033a (95) 0.004 (64) 0.025
ab a

0.091a 0.004 0.089a (97) 0.003 0.091 (99) 0.002 0.068b (75) 0.004 (71) 0.056 0.001 0.043c (47) 0.001 33.8
*** bc a

1.40ab 0.00 1.46a (104) 0.04 1.53 (110) 0.04 1.24ab (89) 0.15 (62) 1.23 0.04 1.12b (80) 0.06 5.1
* ab a

0.15ab 0.00 0.16ab (107) 0.008 0.17 (116) 0.006 0.15ac (101) 0.02 (88) 0.12
bcd a

22.8a 1.35 18.9a (83) 1.01 19.5 (85) 0.73 20.9a (92) 1.09 (83) 19.2 (84) 0.8 13.2b (58) 1.05 7.5
** a a

44.1a 4.1 16.9b (38) 1.03 20.8b (47) 0.51 13.6bc (31) 0.64 8.3cd (19) 0.53 3.6d (8) 0.39 54.4***

Control

Shells

0.13 (100) 0.002 0.10ab (80) 0.007

bc

Low

Medium

0.08

0.002 High 0.06c (49) 0.0009 F-Value

*

0.0006 0.018b (52) 0.001 6.2.

abc

0.004 0.10d (67) 0.006 8.3

**

21.8

***

**

p 0.05; ** p 0.01; *** p 0.001.

means with different letters are significantly different at p < 0.05.

CHAPTER 3

72

3.3.3 Chlorophyll a Organisms containing chlorophyll a were a minor component of the total organic matter in pond effluent. There was significant settlement of chlorophyll a in the controls containing no oysters or only dead oysters. In addition to settlement in the controls, the concentration of chlorophyll a was significantly reduced by the high and medium density oyster treatments but there was no significant difference between the low density treatment and the controls. The combined effects of settlement and oyster filtration reduced the concentration of chlorophyll a to 8% of the initial effluent value in the high density and 19% in the medium density of oysters.

3.3.4 Bacteria The mean number of total bacteria in pond effluent was 22.8 x 106 mL-1. The controls and the medium and low density oyster treatments had no significant effects on bacterial concentration. The high density of oysters significantly reduced the total bacterial numbers to 58% of the initial effluent concentration.

3.3.5 Total Nutrients The medium and low densities of oysters had no significant effects on the concentration of total N in the effluent, but the high density oyster treatment significantly reduced the total N concentration to 80% of the initial level in the pond effluent. The total P concentration in high and medium density oyster treatments was significantly lower than in the influent water, but the low density treatment had no significant effect. The high and medium density oyster treatment reduced the effluent total P concentration to 67% and 83% of the initial level, respectively.

EFFECTS OF O YSTER F ILTRATION ON WATER Q UALITY

73

3.4 Discussion The sediment and nutrient levels in the effluent from the P. japonicus shrimp farm were comparable to those reported in other studies of shrimp farm pond water and effluent (Ziemann et al., 1992; Burford, 1997; Samocha & Lawrence, 1997), but the concentrations of TKN and TP were an order of magnitude lower than those reported by Hopkins et al. (1993a). The level of chlorophyll a in the effluent from the P. japonicus ponds was indicative of phytoplankton productivity in shrimp ponds (Burford, 1997). However, organisms containing chlorophyll a were a minor component of the organic matter (the rest consisting of faecal material, undissolved feed pellets and detritus) and approximately 60% of the chlorophyll a was removed by settlement alone. Filtration by oysters removed a significant proportion of the remaining suspended phytoplankton together with other organic matter, including bacteria, the level of reduction depending on the oyster density. The control treatments of no oysters or only oyster shells showed that most of the organic matter remained in suspension during the two hour experiment. This organic matter probably consisted of fine detritus derived from shrimp feed (Briggs & Funge-Smith, 1994). The high density of oysters effectively halved the total organic load by removing this organic matter from the effluent.

The net effect of oyster filtration on effluent nutrient loads reflects the balance between uptake, excretion and remineralisation of nutrients from settled faeces. In this study, the proportion of dissolved or particulate fractions in the total phosphorous or nitrogen load were not determined. The net reduction of 23% of the total phosphorus at the highest density of oysters was probably due to the removal of phosphorous bound to organic or inorganic particulates. The net reduction of total nitrogen was less effective, possibly due to removal being balanced by nitrogen excreted by the oysters. It is possible that assimilation of

CHAPTER 3

74

dissolved nutrients by macroalgae could be used to counter the excretion by oysters (Shpigel et al., 1993).

The high proportion (72%) of inorganic matter in the effluent from the P. japonicus farm is characteristic of effluent from unlined earthen shrimp ponds (Ziemann et al., 1992). The lack of any significant settlement of the inorganic component during the two hour period of the experiments indicates that most of the inorganic matter was in the form of very fine particles. Filtration by oysters was effective in removing this suspended sediment, with the high density of oysters approximately halving the effluent load. During filtration, oysters sort particles by size and weight; rejected material (pseudofaeces) is accumulated and then expelled through the inhalant opening (Barnes, 1994). Oysters may, therefore, serve a useful role in removing small inorganic particles from effluent. However, high sediment loads can reduce or even arrest oyster filtration (Loosanoff & Tommers, 1948). Further studies are needed to determine the effects of high inorganic sediment loads on oyster growth and survival. There is also a need to determine the effectiveness of reducing silt loads in sedimentation ponds, prior to oyster filtration. High organic loads can also reduce oyster filtration efficiency (Ali, 1970), resulting in the production of large amounts of pseudofaeces (i.e., food filtered but not ingested) and a concomitant increase in deposition rates due to an inefficient use of filtered food (Tenore & Dunstan, 1973). Although oysters are inefficient at ingesting phytoplankton and other food particles at very high concentrations, their ability to convert this material to pseudofaeces still results in a net removal from the system. However, if concentrations are too high, recycling of the effluent through the oysters several times may improve the efficiency of the system (Jones et al., in prep b; Chapter 4).

EFFECTS OF O YSTER F ILTRATION ON WATER Q UALITY

75

3.4.1 Scaling Up Calculations Despite these limitations, it is possible to calculate an estimate of the number of oysters required to obtain the observed improvements in water quality and the pond area that would need to be devoted to them. Based on 20% water exchange per day, a 1 ha pond would need 120,000 oysters. At the same stocking density used in the high density treatment in this study (1 oyster per 0.01 m2), the oysters would occupy 0.12 ha. Therefore 12% of the area of ponds must be set aside for oyster filtration. If greater improvements in water quality were required the number of oysters (and hence pond area) could be increased. These calculations are comparable to those from Wang (1990), who suggested 150, 000 adult sized oysters per hectare of shrimp pond would produce reductions in suspended solids similar to those observed in the present study. If commercial production of oysters is to be achieved efficiently, a full range of oyster sizes must be stocked in treatment ponds to ensure a constant supply of commercial sized oysters. Estimates by Wang (1990) have shown that 360, 000 oysters of varying size would be needed to produce 12, 000 oysters per week from the effluent from a 1 ha shrimp pond.

3.4.2 Summary In summary, the results of this short-term study demonstrated that oysters can significantly improve the water quality of shrimp pond effluent by the filtration and retention of suspended organic and inorganic matter. Longer term studies are needed to determine the effectiveness of oyster filtration at other stages of the of the shrimp growout season and the effects of shrimp effluent on the growth, filtration rates and survival of the oysters.

CHAPTER 4 THE EFFICIENCY AND CONDITION OF OYSTERS AND MACROALGAE USED AS BIOLOGICAL FILTERS OF SHRIMP POND EFFLUENT

Abstract
Current shrimp pond management practices generally result in elevated concentrations of nutrients, suspended solids, bacteria, and phytoplankton compared to the influent water. Concerns about adverse environmental impacts due to discharging pond effluent directly into adjacent waterways have prompted the search for costeffective methods of effluent treatment. One potential method of effluent treatment is the use of ponds or raceways stocked with plants or animals that act as natural biofilters by removing waste nutrients. In addition to improving effluent water quality prior to discharge, the use of natural biofilters provides a method for c apturing otherwise wasted nutrients. This study examined the potential of the native oyster, Saccostrea commercialis (Iredale and Roughley) and macroalgae, Gracilaria edulis (Gmelin) Silva to improve effluent water quality from a commercial Penaeus japonicus (Bate) shrimp farm. A system of raceways was constructed to permit recirculation of the effluent through the oysters to maximise the filtration of bacteria, phytoplankton and total suspended solids. A series of experiments were conducted to test the ability of oysters and macroalgae to improve effluent water quality in a flow-through system compared to a recirculation system. In the flowthrough system oysters reduced the concentration of bacteria to 35% of the initial concentration, chlorophyll a to 39%, total particles (2.28 m - 35.2 m) to 29%, total nitrogen to 66% and total phosphorus to 56%. Under the recirculating flow regime, the ability of the oysters to improve water quality was significantly enhanced. After four circuits, total bacterial numbers were reduced to 12%, chlorophyll a to 4%, and total suspended solids to 16%. Efforts to increase biofiltration by adding additional layers of oyster trays and macroalgal mesh bags resulted in fouling of the lower layers causing the death of oysters and senescence of macroalgae. Supplementary laboratory experiments were designed to examine the effects of high effluent concentrations of suspended particulates on the growth and condition of oysters and macroalgae. The results demonstrated that high concentrations of particulates inhibited growth and reduced the condition of oysters and macroalgae. Allowing the effluent to settle before biofiltration improved growth and reduced signs of stress in the oysters and macroalgae. A settling time of 6 h reduced particulates to a level that prevented fouling of the oysters and macroalgae.

CHAPTER 4

78

4.1 Introduction In many countries, including Australia, effluent from aquaculture ponds is released into receiving waterways without treatment (Eng et al., 1989). Consequently, there are concerns about potential adverse environmental impacts on coastal waters due to eutrophication and increased turbidity (Ziemann et al., 1992; Primavera, 1994). These concerns have become a risk factor for the aquaculture industry (Braaten, 1991). With the expansion of pond aquaculture sites in Australia, both industry and resource managers are becoming increasingly aware that the industry may require stricter environmental controls, such as those currently required for sewage discharge (Samocha & Lawrence, 1997). This has prompted efforts to develop cost-effective methods of effluent treatment. In addition to the environmental aspects, recovery of the nutrients associated with the degraded pelleted feeds could be of considerable economic benefit.

Due to the use of paddle wheel aerators that scour soil from the sides and bottom of earthen farm ponds, suspended inorganic solids are one of the principal components of pond effluent. The removal of these particulates from the pond effluent could probably be promoted by structural changes to effluent channels such as incorporation of baffles. The effluent also contains significant amounts of suspended organic particulates, predominantly phytoplankton, protists and bacteria. Due to the small size of the suspended organic and inorganic particles in pond effluent, removal by mechanical filtration is often very difficult and prohibitively expensive (Hopkins et al., 1995b). The smaller particles have a similar specific gravity to seawater and therefore do not readily settle out of suspension (Rubel & Hager Inc., 1979). Flocculating agents could be used to enhance sedimentation, but they are considered prohibitively expensive for application in aquaculture (Norris 1994 in Samocha & Lawrence, 1997).

EFFICIENCY AND CONDITION OF BIOFILT ERS

79

The use of filter feeding bivalves such as oysters, mussels or clams as natural biofilters could be effective in removing the small particles from the effluent (Wang, 1990; Hopkins et al., 1993a). After the larger particles have been settled out of suspension (to prevent fouling of the oysters), bivalves will feed on the smaller, non-settleable particles such as phytoplankton, bacteria and other organic material. Small inorganic particles are also filtered, coagulated into larger, more settleable particles and egested as pseudofaeces (Tenore & Dunstan, 1973). The organic particles ingested by the oysters are incorporated into tissue, thereby capturing wasted nutrients and converting them into a secondary cash crop.

To ensure success of oysters cultured in shrimp pond effluent, adequate water flow and circulation must be maintained such that sufficient oxygen is supplied, the buildup of soluble metabolites is minimised and the fouling by settling particles reduced (Thielker, 1981). This circulation also ensures an even distribution of food, which is important to prevent differential growth in the oysters (Scura et al., 1979).

Water flow rate can have a significant effect on the ability of oysters to filter particulates (Walne, 1972). If flow rates cannot be optimised for oyster filtration, treatment may require recirculation through the oyster pond several times, or have a larger treatment area. A large treatment area is economically less viable due to loss of production area for production of the primary high value crop (Chien & Liao, 1995), and consequently recirculation may be more viable depending on pumping costs.

Previous studies have shown that oysters can significantly improve the water quality of effluent from shrimp ponds (Wang & Jakob, 1991; Hopkins et al., 1993a; Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5). The aims of the present study were the

CHAPTER 4

80

following: a) characterise changes in the biological and chemical composition of shrimp pond effluent following biofiltration by oysters and macroalgae in raceways; b) determine the differences in water quality improvements under flow-through and recirculating flow; and c) determine the effects of the high suspended solids in shrimp pond effluent on the growth and condition of oysters and macroalgae.

EFFICIENCY AND CONDITION OF BIOFILT ERS

81

4.2 Materials and Methods 4.2.1 Experimental Design A series of field and laboratory experiments were conducted to test the efficiency of biofiltration of shrimp pond effluent, to improve discharge water quality or enable reuse of water in production ponds. Different densities of oysters and macroalgae, and different flow regimes (continuous flow versus recirculating) were tested for the ability to improve the water quality of shrimp farm effluent, and the impacts of the effluent on the growth and condition of the biofilter organisms.

4.2.1.1 Filtration Efficiency Experiments Twelve 1500 L concrete raceways (Plate 4.1), based on the design by Scura et al. (1979), were constructed and supplied with water from the effluent canal of a commercial Penaeus japonicus (Bate) shrimp farm which discharges into Moreton Bay, Queensland, Australia. Each raceway was divided into chambers by baffles to ensure good mixing and circulation of the water through the oysters (Fig. 4.1A). The raceways were stocked with locally collected Sydney Rock oysters, Saccostrea commercialis (Iredale and Roughley) and the red macroalga, Gracilaria edulis (Gmelin) Silva.

A series of experiments were conducted, with both a continuous flow through and recirculating flow regimes investigated. Before all experiments, oysters were maintained in the raceways for a period of two weeks, and macroalgae for two days prior to study initiation.

The first continuous flow experiment was conducted with 55 g oysters stocked at three different densities in plastic oyster trays stacked three layers deep. The low density treatment was 18 oysters per 100 L (270 per raceway), the medium density was 36 oysters per 100 L

CHAPTER 4

82

(540 per raceway) and the high density 72 oysters per 100 L (1080 per raceway). Three replicate raceways for each density treatment were stocked with oysters, and three replicate raceways containing empty oyster trays were maintained as controls. Water quality parameters were sampled at the inflow and outflow point of each raceway every 12 h for 72 h. The data presented is an average of all six sampling times. Three replicate one litre containers of water were collected from the inflow and outflow point of each raceway for determination of particle size, bacteria, phytoplankton (chlorophyll a), total nitrogen (N) and total phosphorus (P).

Plate 4.1 Raceways constructed at Moreton Bay Prawn Farm, Queensland, Australia.

The second continuous flow experiment used three different densities of macroalgae in mesh bags (100 cm 50 cm) constructed using plastic mesh (15 mm mesh size) folded in half and cable tied together. Three replicate one litre containers of water were collected from the inflow and outflow point of each raceway for determination of dissolved nutrients (NH4+, NO3- / NO2-, and PO43-).

EFFICIENCY AND CONDITION OF BIOFILT ERS

83

The recirculating flow experiment was conducted with 35 g oysters stocked at 10.8 per 100 L (162 per raceway). The lower density was chosen to ensure that, based on estimated oyster pumping rates (Wang, 1990), not all the volume of water in raceways could be filtered within the 2 h residence time. Measurements were made after the first circuit and compared to the control raceways. The effluent from all twelve raceways flowed into a holding tank and then pumped back into the raceways. Therefore, after the first circuit all twelve raceways were functioning as a combined treatment unit. Water quality sampling was conducted at two hourly intervals (the time required for full water exchange through the raceways) for four circuits to determine the increasing improvement in water quality with successive filtering by the oysters. Three replicate one litre containers of water were collected from the inflow and outflow point of each raceway for determination of total suspended solids (including organic content), particle size, bacteria, and phytoplankton (chlorophyll a).

4.2.1.2 Biofilter Condition Experiments Controlled laboratory experiments to determine the impacts of high loadings of suspended particulates on the growth and condition of oysters and macroalgae were conducted over a period of eight weeks with effluent replaced weekly (Fig. 4.1B). Effluent from the discharge channel at Moreton Bay Prawn Farm was collected and transported to the laboratory in 30 L plastic drums.

In the laboratory, any settled particles in the sample were resuspended and a suite of physicochemical parameters (temperature, pH, salinity, dissolved oxygen, and turbidity) were determined with a Horiba U-10 water quality meter (California, U.S.A.). Three drums of effluent were then settled for 1 h, 6 h, and 24 h, respectively. A fourth drum was settled for 24 h and then treated by oyster filtration for another 24 h. The fifth drum was stirred just

CHAPTER 4

84

prior to transfer to the oyster and macroalgal tanks. A control treatment utilised sand filtered seawater from Moreton Bay. This procedure was repeated weekly for eight weeks. The pretreated effluent was transferred into four separate tanks (containing oysters or macroalgae) for each sedimentation treatment.

Figure 4.1 Diagrammatic representation of experimental setup, a) single raceway with baffles and oyster trays, and b) laboratory settling experiment. NTU = nephelometric turbidity units. The oysters used in the experiments were Sydney Rock oysters, Saccostrea commercialis and the macroalgae was Gracilaria edulis. Effluent was from an intensive Penaeus japonicus shrimp farm.

EFFICIENCY AND CONDITION OF BIOFILT ERS

85

Oysters and macroalgae were incubated in 11 L tanks and were raised off the bottom by a plastic mesh tray. Three macroalgal tanks (50 g of macroalgae each) and one oyster tank (8 oysters) were maintained for each sedimentation treatment. Tanks were maintained at 20 23C, and exposed to light on a 12:12 h light / dark cycle using daylight fluorescent tubes which provided approximately 250 mol quanta m-2 s-1. Tanks were aerated vigorously to maintain water movement and dissolved oxygen concentrations.

Water was changed weekly, with water quality analyses conducted prior to changing the incubation water. Physico-chemical parameters (turbidity, salinity, dissolved oxygen, temperature, and pH) were measured, as well as oyster biomass and volume, macroalgal biomass (wet wt), pigment content (chlorophyll a and phycoerythrin), %N, 15N and amino acid content.

Photosynthetic capacity of the macroalgae was also measured as an indicator of macroalgal condition. Experiments were conducted using the same experimental tanks described for the previous biofilter condition experiment. The effluent was settled for 24 h and then filtered by oysters for 12 h prior to incubation with macroalgae. The experiment was conducted under solar radiation, with tanks shaded to 50% of incident light. Three replicate treatment and seawater control tanks were maintained. Effluent was replaced every twelve hours for 7 d. Daily measurements of photosynthetic capacity (electron transport rate) and photosynthetically active radiation (PAR) were taken using a pulse amplitude modulated fluorometer, DIVINGPAM (Walz GmbH. Effeltrich, Germany). Three replicate measurements were taken for each tank (a total of nine seawater control and nine shrimp effluent measurements).

CHAPTER 4

86

4.2.2 Analytical Procedures Chlorophyll a was determined by filtering a known volume of water sample through Whatman GF/F filters, which were immediately frozen. Acetone extraction and calculation of chlorophyll a concentration was performed using the methods of Clesceri et al. (1989), and Parsons et al. (1984).

The filtrate collected from filtering for chlorophyll a analysis was collected in 120 mL polycarbonate containers and immediately frozen. NH4+ and NO3-/NO2- and PO43- were determined by the Queensland Health Analytical Services (NATA accredited) using a Skalar autoanalyser.

Unfiltered samples for nutrient analysis (total Kjeldahl nitrogen and total phosphorus) were collected in 120 mL polycarbonate immediately frozen. They were subsequently analysed within two weeks by the NATA accredited Queensland Health Analytical Services Laboratory in accordance with the methods of Clesceri et al. (1989) using a Skalar autoanalyser (Norcross, Georgia, U.S.A.).

Total suspended solids concentrations were determined using the methods of Clesceri et al. (1989). A known volume of water was filtered onto a pre-weighed and predried (110 C; 24 h) Whatman GF/C glass fibre filter. The filter was then oven dried at 60 C for 24 h and total suspended solids calculated by comparing the initial and final weights. Volatile suspended solids were determined as loss on ignition by combusting samples in a muffle furnace for 12 h at 525 C (Clesceri et al., 1989).

EFFICIENCY AND CONDITION OF BIOFILT ERS

87

For analysis of particle size number and distribution, a subsample of 30 mL was stored in a polystyrene container, and fixed with 5% formalin and refrigerated for subsequent analysis. The sample was later placed into a Coulter counter, which counted the number of particles in the range from 2.282 m to 35.2 m inclusive.

Bacteria samples were preserved with 2% formalin and kept at 4C until analysis. A known volume (0.5 mL - 1 mL) of sample was stained with acridine orange, filtered onto a stained (Irgalan Black) 2 m poretics filter and mounted on a slide. Bacteria were counted using epifluorescence microscopy (Hobbie, 1977).

For analysis of plant pigments (phycoerythrin and chlorophyll a), total tissue N, 15N and amino acids, tissue was removed, rinsed in distilled water to remove any nutrients and sediment from the thallus surface, and then prepared for analysis.

For calculation of total tissue N content and the 15N isotopic signature, samples were oven dried to constant weight (24 h at 60 C), ground and three sub-samples were oxidised (Dumas combustion) in a Roboprep CN Biological Sample Converter (Europa Tracermass, Crewe, U.K.). The resultant N2 was analysed by a continuous flow isotope ratio mass spectrophotometer (Europa Tracermass, Crewe, U.K.). Total %N of the sample was determined, and the ratio of 15N to 14N was expressed as the relative difference between the sample and a standard (N2 in air) using the following equation (Peterson & Fry, 1987): 15N = (15N/14N (sample) / 15N/14N (standard) 1) x 1000 ()

For amino acid analysis, approximately 1.0 g wet weight of algal tissue was weighed and placed in 5 mL of 100% methanol (analytical reagent grade) for 24 h to extract amino acids.

CHAPTER 4

88

The methanol extract was filtered through Millipore Millex - HV13 (0.45 m) filters and injected into a Beckman System 6300 post column derivatisation HPLC amino acid analyser, for detection of ninhydrin positive free amino acid groups at 570 nm. Results were calculated and expressed as mol g-1 wet weight. As well as detecting free amino acids, this technique also measures the concentration of free NH4+ in plant tissue.

For pigment analysis, approximately 0.5 g fresh weight of G. edulis tissue was separated, blotted dry and weighed (wet wt). The tissue was ground with a mortar and pestle in a phosphate buffer solution (pH 6.5) to disrupt the cells. The extract was then poured into a glass graduated centrifuge tube, made up to 10 mL, and centrifuged (20 min at 2500 rpm) to produce a supernatant containing phycoerythrin and a pellet with the remaining tissue. The supernatant was transferred to a cuvette and absorption was determined on a spectrophotometer at 565 nm and 710 nm for phycoerythrin and turbidity blank, respectively. The pellet was resuspended in 5 mL of 80% acetone (analytical reagent grade) and disrupted with a tissue homogeniser to extract chlorophyll. Samples were re-centrifuged (20 min at 2500 rpm), and absorbance was determined at 664 nm and 710 nm for chlorophyll a and turbidity blank, respectively. Pigment concentrations as mg g -1 dry wt were calculated with specific formulas for phycoerythrin (Rowan, 1989) and chlorophyll a (Parsons et al., 1984).

Photosynthetic capacity as electron transport rate (ETR) was determined at a range of light intensities by generation of rapid light curves using the pulse amplitude modulated (PAM) fluorometer. The rapid light curves were generated over a 90 sec period with 10 sec periods of actinic irradiance with 1 sec saturating pulses for measurement of quantum yield (White & Critchley, 1999). Data was stored in the diving PAM for subsequent download to computer.

EFFICIENCY AND CONDITION OF BIOFILT ERS

89

Placement of the fibre optic cable was standardised by attaching it to a leaf clip, which was located near the growing tip of the thallus.

For all sampling techniques, three replicates were analysed and means and standard errors were calculated. Differences between treatments were tested for significance using one way analysis of variance (ANOVA) and Tukey's Test for multiple comparison of means.

CHAPTER 4

90

4.3 Results 4.3.1 Filtration Efficiency Experiments 4.3.1.1 Continual Flow On completion of sampling, oyster mortality was determined in each tray (Fig. 4.2A). For each density treatment, the percent mortality increased from the upper tray to the lower tray. Total mortality for all trays was significantly greater in the high density treatment. This mortality reduced the number of live organisms available to filter the effluent, and resulted in the low density treatment being more effective at improving the effluent water quality. Therefore, only the data from the low density treatment is presented.

During the macroalgal filtration experiment, the total concentration of dissolved nutrients decreased in the control raceway, but increased in all three macroalgal density treatments (Fig. 4.2B). NH4+ was the predominant dissolved nutrient and increased significantly (p < 0.001) after flowing through the macroalgal raceways. This increase was probably due to decay resulting from fouling of the thallus surface by settling material.

Oysters were effective at reducing bacterial numbers from the shrimp pond effluent, significantly (p < 0.01) decreasing the concentration to 39% of the initial level in the effluent, with no significant (p > 0.05) change in the control raceways (Table 4.1). In addition to significant (p < 0.05) settling in the control raceways (to 76% of the initial concentration), oyster filtration significantly (p < 0.001) reduced the concentration of chlorophyll a to 30% of the initial concentration. Oyster filtration significantly (p < 0.001) reduced the concentrations of total N and total P to 64% of initial concentration and to 55% of initial concentration, respectively. There was also some natural sedimentation of particles in the control raceways, with the concentration of total N decreasing significantly (p < 0.05) to

EFFICIENCY AND CONDITION OF BIOFILT ERS

91

89%, and total P decreasing significantly (p < 0.05) to 84% of the initial concentrations (Table 4.1).

A
Upper Tray

Middle Tray

Lower Tray

10

20

30

40

50

60

70

80

90

100

Oyster Mortality (% of initial) High Medium Low

B
Change in Dissolved Nutrient Concentration (M)
12 10 8 6 4 2 0 -2

Control
-4

Low
Macroalgal Density
+ -

Medium
3-

High

NH4

N O3

PO 4

Figure 4.2 Impacts of effluent on biofilters: a) Oyster mortality (%) from upper, middle and lower trays after 2 weeks at low, medium and high oyster stocking densities in raceways supplied with unsettled shrimp effluent, and b) change in dissolved nutrient concentrations after passing effluent through low, medium and high macroalgal stocking densities in raceways supplied with unsettled shrimp effluent. Positive change represents an increase, negative change represents a decrease.

CHAPTER 4 Table 4.1 Water quality parameters after filtration by oysters under flow through conditions in raceways. Total N = total Kjeldahl nitrogen; Total P = total phosphorus.

92

Treatment

Bacteria (no. 109 L-1) 12.3a 13.3a 4.8b 11**

Chlorophyll a (g L-1) 35.6a 27.1b 10.7c 93***

abc

Total N (M) 134a 119b 86c 66***

Total P (M) 5.8a 4.9b 3.2c 190***

Inflow Control Outflow Oyster Outflow F Value

* p < 0.05; ** p < 0.01; ***

p < 0.001.

means with different letters are significantly different at p < 0.05.

Ninety five percent of the suspended particles in the effluent were approximately 2-4 m in diameter (Fig. 4.3A). Filtration by the oysters significantly decreased (p < 0.001) the concentration of particulates to 36% of initial concentration. However, the number of particles in the control raceways increased to 119% of the initial concentration, due to breaking up of aggregated particles. This was evidenced by a reduction in the mean particle size.

4.3.1.2 Recirculating Experiments After the first circuit of the effluent through the raceways, comparisons were made between the water quality parameters in treatment and control raceways (Table 4.2). In addition to significant (p < 0.05) reductions in bacteria (77% of the initial concentration) after passing through the control raceways, oyster filtration further reduced (p < 0.001) the bacterial numbers to 45% of the initial concentration. There was no significant settling of phytoplankton (chlorophyll a) in the control raceway, but oyster filtration decreased the concentration significantly (p < 0.01) to 70% of the initial level.

EFFICIENCY AND CONDITION OF BIOFILT ERS

93

35
Detection

A
Inflow

30

Limit

Particle Concentration

25

(no. 10 L )

-1

20 15 10 5 0

Control Outflow Oyster Outflow

10

12

14

Particle Size (m) 25

Detection Limit

Particle Concentration

20

Initial Control

(no. 10 L )

-1

15

First Second

10

Third
5

Fourth

0 0 2 4 6 8 10 12 14

100000
Detection Limit

Particle Size (m)

Initial Control First

Particle Concentration

10000

(no. 10 L )

-1

1000

Second Third

100

Fourth

10

1 0 5 10 15 20 25 30 35 40

Particle Size (m)

Figure 4.3 Particle size distribution, a) before and after control and oyster treatment raceways during single continuous flow, b) before and after consecutive circuits through oyster treatment raceways (linear scale), and c) before and after consecutive circuits through oyster treatment raceways (log scale).

CHAPTER 4

94

Table 4.2 Water quality parameters after filtration by oysters after the first circuit during recirculating flow in raceways. TSS = total suspended solids; Organic = organic component of TSS (loss on ignition); Inorganic = inorganic component of TSS.

Treatment

Bacteria (no. 109 L-1)

Chlorophyll a (g L-1) 26.3a 20.8ab 18.5b 5.4**

abc

TSS (mg L-1) 140a 130ab 110b 4.9**

Organic (mg L-1) 41 41 32 2.9

Inorganic (mg L-1) 99a 89ab 78b 4.7**

Inflow Control Outflow Oyster Outflow F Value

* p < 0.05; ** p < 0.01;

26.5a 20.5b 11.8c 65***

*** p < 0.001.

means with different letters are significantly different at p < 0.05.

Comparison of total suspended solids concentrations in the inflow and outflow of the raceways showed that oyster filtration was effective in reducing the concentration of both organic and inorganic particulates (Table 4.2). In the control raceways, settling reduced the concentration of total suspended solids to 93% of initial, with all the reduction being inorganic particles (88% of the initial concentration). However, in the oyster raceways the concentration of inorganic particles decreased significantly (p < 0.01) to 76% and organic particles to 78% of the initial concentration. The number of particles decreased to 80% of the initial after the first circuit (Fig. 4.3B).

The results from the first circuit of the recirculating flow experiment revealed that the smaller 35 g oysters (Table 4.2) were considerably less effective at improving the effluent water than the larger 55 g oysters (Table 4.1) used in the continuous flow experiment. The net reductions (relative to the control raceway) in bacteria, chlorophyll a and particle concentration by the 55 g oysters were 50%, 130% and 120% greater than by the 35 g oysters, respectively.

EFFICIENCY AND CONDITION OF BIOFILT ERS

95

After four circuits through the raceways, the concentrations of all water quality parameters were significantly (p < 0.001) lower than after the first circuit and after treatment by the larger oysters in the continuous flow experiment (Fig. 4.4). The number of bacteria was reduced to 12% of the initial concentration, chlorophyll a concentration to 20%, and TSS to 19%. The slope of the curve over the consequent circuits through the raceways varied between parameters. Bacteria and chlorophyll a both exhibited curvilinear relationships, asymptotically approaching zero. In contrast, the relationship between total suspended solids and number of circuits was linear. The first circuit through the oysters reduced the concentration of bacteria by 47% of the initial concentration, 23% in the second circuit, but only 11% and 8% in the third and fourth circuits, respectively. A similar trend was observed with reductions in chlorophyll a, with a 32% reduction in the first, 55% in the second, but only 9% in the third circuit, and a small increase in the fourth circuit. In contrast, total suspended solids was reduced by 23% of the initial concentration, 10% in the second circuit, 33% in the third and 14% in the fourth circuit.

The total number of particles was reduced to 13% of the initial concentration after four circuits. Particle concentration expressed on a log scale revealed an increase in the number of the larger particles with successive passes through the raceways, probably due to coagulation of smaller particles by oysters into larger faeces and pseudofaeces (Fig. 4.3C).

CHAPTER 4

96

30 Bacteria (no. x10 L ) 25 20 15 10 5 0 30 Chlorophyll a (g L ) 25 20 15 10 5 0 140 120 TSS (mg L )

-1 -1 -1

100 80 60 40 20 0 Initial First Second Third Fourth

Number of Circuits

Figure 4.4 Concentrations of water quality components before and after consecutive circuits through oyster treatment raceways, a) bacterial numbers, b) chlorophyll a concentration, and c) total suspended solids (TSS).

EFFICIENCY AND CONDITION OF BIOFILT ERS

97

4.3.2 Biofilter Condition Experiments The mean turbidity of the oyster and macroalgal treatment tanks over the eight week experimental period ranged from 61 nephelometric turbidity units (NTU) (raw effluent) to 8 NTU (24 h settled) (Fig. 4.1B). The turbidity of the filtered seawater control was 0 NTU, and the 24 hr settled + oyster filtered treatment was 2 NTU.

The mean growth rate of oysters was greatest in effluent that had been pre-settled for 6 or 24 h (Fig. 4.5A). However, there was a high level of variation in growth rates and no significant difference in relation to the duration of pre-settlement of effluent. Macroalgal biomass was significantly lower in the seawater control and unsettled effluent, otherwise there was no significant difference in relation to the duration of presettlement of effluent (Fig. 4.5B).

Macroalgal growth (as number of new shoots) and pigment content were maximal in the 24 h plus oyster filtration treatment (Fig. 4.6). There were new shoots present in all treatments except the seawater control. The number of new shoots increased with increased settling time.

The responses in new shoot growth and pigment concentration was correlated with the responses in %N and 15N (Fig. 4.7). Tissue N content increased from the initial (0.7%) in all but the seawater control, in which a decrease to 0.28% was observed. The 0 h and 24 h + oyster treatment had the two highest tissue N values (0.8 and 0.9%, respectively). The 15N of the macroalgae increased from the initial (6.9) in all treatments except the seawater control (unchanged), or the 24 h plus oyster treatment which decreased to 5.0. The highest value was recorded in the 0 h treatment (10.9).

CHAPTER 4

98

Oyster Growth Rate (cm 3 oyster-1)

n.d.
0
60

B
Macroalgal Biomass (g tank-1 )

Initial Biomass

40

20

0 (0 h) Seawater Control 0 1 6 24
Effluent Settling Time (h)

24 h + Oyster Filtration

Figure 4.5 Growth of oysters and macroalgae after 8 weeks in tanks supplied with shrimp effluent pre-settled for 0, 1, 6 & 24 h. a) change in oyster growth rate expressed as changes in oyster volume (cm3 oyster -1), and b) macroalgal biomass. n.d. = no data.

EFFICIENCY AND CONDITION OF BIOFILT ERS

99

A
(no. of new shoots tank -1) Macroalgal Growth

40

30

20

10

B
Pigment Concentration

1.6 1.4

(mg g -1 dry wt)

1.2 1 0.8 0.6 0.4 0.2 0

PE

CHL

(0 h) Seawater Control

24

24 h + Oyster Filtration

Effluent Settling Time (h)

Figure 4.6 Response of macroalgae to 8 weeks in tanks supplied with shrimp effluent pre-settled for 0, 1, 6 & 24 h. a) macroalgal growth expressed as number of news shoots per tank, and b) concentration of the photosynthetic pigments, chlorophyll a (CHL) and phycoerythrin (PE).

CHAPTER 4

100

A
Tissue Nitrogen (%N)

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 12

Initial

10

8
Initial

N
6 4 2 0 (0 h) Seawater Control Effluent Settling Time (h) 0 1 6 24 24 h + Oyster Filtration

Figure 4.7 Macroalgal nitrogen content (a) and 15N (b) after 8 weeks in tanks supplied with shrimp effluent of different settlement times, a) %N, and b) 15N.

The total concentration of free amino acids in the macroalgal tissue was highly variable, ranging from 0.65 mol g wet-1 in the seawater control to 4.1 mol g wet-1 in the 24 h settling treatment (Table 4.3). These changes in the total free amino acid pools are probably due to an interplay between light availability (both light limitation and light inhibition) and nutrient availability. Changes in the amino acid composition were also

15

EFFICIENCY AND CONDITION OF BIOFILT ERS

101

observed (Table 4.3). Citrulline was present only in the 6 h and 24 h treatments, but in these treatments citrulline constituted 32% and 48%, respectively. Other amino acids such as glutamine, phenylalanine and serine showed significant increases in the 24 h plus oyster treatment, whereas alanine was highest in the 0 h treatment. The proportion of glutamate ranged from 21% to 45% of the total free amino acid pool, and like alanine, was highest in the 0 h treatment.

Table 4.3 Changes in the free amino concentration and composition of macroalgae for various treatments in laboratory settling experiments. % refers to percentage of total free amino acid pool. CIT = citrulline; GLU = glutamate; ALA = alanine; GLN = glutamine; PHE = phenylalanine; SER = serine; Total = total concentration of free amino acids (mol g wet -1).

Treatment

CIT (%)

GLU (%) 40 45 30 24 21 35

ALA (%) 2.4 8.5 7.7 7.2 2.6 5.2

GLN (%) 7.7 5.1 3.0 2.2 2.4 11

PHE (%) 0 9.9 9.4 5.6 6.3 12

SER (%) 4.4 8.8 7 5.1 3.8 16

Total (mol g wet-1) 0.65 3.3 1.8 3.5 4.1 2.5

(0 h) Seawater Control Effluent settled for 0 h Effluent settled for 1 h Effluent settled for 6 h Effluent settled for 24 h 24 h + Oyster Filtration

0 0 0 32 48 0

The maximum photosynthetic capacity (as electron transport rate) of the macroalgae incubated in the settled and oyster filtered shrimp effluent was 49 mol e- m-2 s-1 versus 28 mol e- m-2 s-1 in the macroalgae incubated in the seawater control (Fig. 4.8). Slight photoinhibition was evident from 1000 E m-2 s-1 in macroalgae incubated in the seawater control, but not in the shrimp effluent treatment. In both treatments the maximum electron transport rate was observed at 300-500 E m-2 s-1.

CHAPTER 4

102

60 50

ETR (mol e - m-2 s -1)

40 30 20 10 0 0 200 400 600 800 1000

-2

1200
-1

1400

1600

1800

PAR (mol m Control

s )

Settled Shrimp Effluent

Figure 4.8 The response of electron transport rate (ETR) versus photosynthetically active radiation (PAR) in macroalgae incubated in seawater (control) or shrimp effluent (settled 24 h plus oyster filtered for 12 h).

EFFICIENCY AND CONDITION OF BIOFILT ERS

103

4.4 Discussion Previous studies of the use of biofilters in aquaculture include investigation of the effectiveness of bivalves (Mann & Ryther, 1977; Shpigel & Fridman, 1990; Shpigel & Blaycock, 1991; Shpigel et al., 1993a) and macroalgae (Harlin, 1978; Harlin et al., 1979; Vandermeulen & Gordin, 1990; Cohen & Neori, 1991; Neori et al., 1991; Haglund & Pedersen, 1993; Subander et al., 1993). Shrimp farms present a particular problem for the application of biofilters due to the high concentrations of phytoplankton and clay minerals in pond effluent (Ziemann et al., 1992) which do not settle out as easily as the particulates in other forms of aquaculture (Macintosh & Phillips, 1992). Application of biofilters to shrimp pond effluent has been trialed (Wang & Jakob, 1991; Hopkins et al., 1993a; Jakob et al., 1993; Lin et al., 1993; Funge-Smith & Briggs, in prep.), but not widely adopted on a commercial scale due primarily to the problems with fouling from the high concentrations of particulates (Huguenin, 1976; Funge-Smith & Briggs, 1998). The present study is one in a series designed to remedy this problem through a better understanding of the effects of the stocking densities of biofilters, water flow regimes and integrating sedimentation prior to biofiltration (Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5).

4.4.1 Efficiency of Biofilters Previous studies have demonstrated that sedimentation, in conjunction with nutrient uptake by phytoplankton and bacteria, can improve the water quality of effluent from shrimp ponds (Cripps, 1994). Sedimentation alone will not remove phytoplankton, bacteria or small inorganic particles from pond effluent (Henderson & Bromage, 1988). This study has confirmed previous observations that these components can be efficiently filtered from the effluent by oysters (Plates 4.2 & 4.3) (Lam & Wang, 1989; Hopkins et al., 1993a; Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5). In addition, the present study

CHAPTER 4

104

has highlighted the need to consider the flow regime adopted, the potential benefits of recirculation, the size of oysters used, the negative impacts of vertical stocking and the benefits of pre-settlement of effluent prior to treatment by oysters and macroalgae.

Plate 4.2 Control raceway on the left with no oysters and treatment stocked at low density 55 g oysters. Demonstrates changes in water clarity (reduction in suspended solids) with the oyster tray clearly visible in the raceway stocked with oysters, but not in the control raceway.

In a previous study, Jones & Preston (1999) (Chapter 3) examined the effectiveness of oysters in filtering effluent in still water (no flow-through). Comparisons with the results of the present study indicate that oysters may be more effective in a flow-through system. The relative efficiency between the two flow regimes can be compared in terms of the percentage reduction of effluent particulates by the oysters, corrected for the differences in oyster stocking density. Under a flow-through system (the present study), oysters were approximately 7, 6, and 1.3 times more effective at removing bacteria, suspended particulates and chlorophyll a respectively, than under the still water system examined in the previous study. The relatively minor difference in removal of chlorophyll a between the two studies may have been due to some settling of phytoplankton in the still water system.

EFFICIENCY AND CONDITION OF BIOFILT ERS

105

High water flow rates can enhance oyster filtration (Loosanoff & Tommers, 1948; Walne, 1972), but low flow rates enable more effective settling of particulates (Piper et al., 1982; Henderson & Bromage, 1988). The current study is the first to examine the potential of effluent recirculation through raceways stocked with biofilters in order to maximise the benefits of both sedimentation and biofiltration.

Plate 4.3 First chamber (foreground) and second chamber (background) of an oyster treatment raceway showing the improvement in water clarity (reduction in suspended solids) within the raceway.

The results showed that, during the first circuit, sedimentation alone in the control raceways did not reduce the concentration of suspended organic particulates. However, there was a reduction in the inorganic particulates, which are not a viable food source for oysters and hence are most likely to foul oysters (Loosanoff & Tommers, 1948). Consistent with lack of settlement of organic particulates recorded for the first circuit of the recirculating experiments, phytoplankton did not settle out of suspension in this

CHAPTER 4

106

experiment, which is in contrast to experiments conducted under still water conditions (Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5). The water flow may be keeping non-motile forms in suspension and hence more available to oysters as a food source. Flowing effluent versus the non-flow regime of other studies (Jones & Preston, 1999; Chapter 3; Jones et al., in review; Chapter 5) may also be responsible for the increase in the number of particles observed in the control raceways in the present study. Based on the reduction in the mean particle size in the control, the increase was due to breaking up of aggregated particles, which is consistent with the observations of Walne (1972).

Biofiltration by oysters resulted in significant reductions in the concentration of both organic and inorganic particulates. The production of faeces and psuedofaeces by oysters was shown to increase the mean particle size in the present study. These particles can more easily settle out of suspension, and therefore do not contribute to the particle load leaving the system (Haven & Morales-Alamo, 1970). The increase in the mean size of suspended particulates in the effluent after oyster filtration has been observed previously (Tenore & Dunstan, 1973).

The dual processes of sedimentation and oyster filtration result in the removal of nutrients through the filtration and settlement of particulates and the addition of nutrients by remineralisation (Blackburn et al., 1988) and oyster excretion (Hammen et al., 1966). In the experimental conditions of this study there was a net reduction in total nutrients. This net reduction occurred despite the increase in dissolved nutrients (particularly NH4+) from oyster excretion (Shpigel et al., 1993b; Jones et al., in review; Chapter 5).

EFFICIENCY AND CONDITION OF BIOFILT ERS

107

Most of the material filtered by oysters is deposited as faeces and pseudofaeces, while the rest is incorporated into the oyster tissue. To remove these nutrients from the system entirely will require harvesting of the oysters, and either removal of the sediment, bacterial denitrification, or uptake of the remineralised nutrients by macroalgae (Funge-Smith & Briggs, in prep.), which is subsequently harvested (Hopkins et al., 1995b).

The macroalgae in the present study were shown to be net contributors of dissolved nutrients due to senescence from fouling by settling particulates. An increase in nutrients from Gracilaria spp. incubated in shrimp farm effluent was also observed by Funge-Smith & Briggs (in prep.). Given proper sedimentation, and oyster filtration to sufficiently reduce the concentration of suspended solids, macroalgae have been shown to rapidly reduce the concentrations of dissolved nutrients from shrimp pond effluent (Jones et al., in review; Chapter 5).

By the end of the raceway experiment there was considerable mortality on the bottom two oyster trays, presumably due to fouling from the increased sedimentation from the oysters on the tray above (Plate 4.4). This is consistent with high mortality (71%) observed in oysters placed directly on the bottom of a commercial shrimp pond (Hopkins et al., 1993a), although this contrasts with the success of Jakob et al. (1993) in growing oysters in seven layers in tanks fed with shrimp pond effluent. However, their design enabled much higher flow rates and the effluent water was from a semi intensive farm (compared with intensive in the present study) with lower concentrations of particulates. Other studies which have had success growing oysters in multi layer systems have used cultured phytoplankton and not shrimp pond effluent (Scura et al., 1979), which eliminates the high inorganic solids loading. A higher flow regime may

CHAPTER 4

108

help to reduce mortality, although if flow is too high it may inhibit the settling of particulates, or simply result in insufficient residence time for the oysters to effectively filter the effluent. This could be overcome by either a) recirculating the effluent through the oysters several times, with a sedimentation regime in between to facilitate settlement of the larger particles produced by the oysters or, b) simply stacking the trays only one layer deep. In commercial ponds without baffles, one layer of oysters may result in a situation of laminar flow where the oysters only have access to the top portion of the water column, with the rest flowing underneath, thereby reducing filtration performance (Scura et al., 1979).

Plate 4.4 Fouling of oysters by settling particulates in raceways.

Kinne et al. (1997) found that reducing flow rates (i.e., longer retention time of effluent in the raceways) increased the efficiency of oysters to improve the quality of the effluent water. An alternative is recirculation of the effluent through the oysters several times. The second option would probably be more efficient, as reducing the flow rate is likely to cause a buildup of metabolites and a decrease in dissolved oxygen (Thielker, 1981). Slowing the flow rate down in a system with such a high particulate load may increase fouling. A slower flow rate may also fail to distribute the food evenly amongst the oysters (Scura et al., 1979), and may

EFFICIENCY AND CONDITION OF BIOFILT ERS

109

slow down the filtration rate of the oysters, which is positively correlated to water flow rate (Walne, 1972). However, Haven & Morales-Alamo (1970) found that slowing the flow rate increased the total removal of particulates, presumably due to giving the oysters more time to filter the particles. A recirculating treatment system can facilitate a high effluent flow rate to enhance the oyster filtration rate, while increasing total particulate removal by maintaining a long total residence time by recirculation of the effluent through the oysters more than once.

Dissolved nutrient uptake by macroalgae is also related to flow rate (Wheeler, 1980) and water exchange (Ugarte & Santelices, 1992), as well as light availability (Hanisak, 1979; Falkowski, 1983). Consequently nutrient removal by macroalgae may be more efficient in a faster flowing recirculating system stocked at lower densities to improve water flow and reduce the boundary layer (Wheeler, 1980).

There will be cost benefit issues to consider with increased pumping costs involved in recirculating flow. This is the first study to use recirculating treatment with oysters as biofilters, and clearly the improvements in water quality were considerably greater compared with continuous flow. However, the increased costs would have to be weighed up against the improvements to water quality and the income from enhanced oyster production. Simply optimising single flow through by optimising flow rates, stocking densities, aeration regimes, and pre settlement with baffles may prove to be more economically viable.

Sedimentation generally removes larger particles with a high specific gravity, leaving smaller and motile particles in suspension. For the American oyster (Crassostrea virginica), a particle size range of 3-4 m was the most efficiently removed, with removal of 1-2 m particles only half as efficient (Haven & Morales-Alamo, 1970).

CHAPTER 4

110

Particles <5 m were shown to be optimal for ingestion by Saccostrea commercialis (Thielker, 1981). Greater than 95% of the particles remaining in suspension in shrimp pond effluent in the present study were in the optimal size range (<5 m).

High numbers of particulates have been shown to reduce feeding efficiency in oysters (Loosanoff & Tommers, 1948; Wisely & Reid, 1978). For S. commercialis, less than 2 mg L-1 was optimal, with no production of pseudofaeces. At 18 mg L-1, 50% of the oysters were producing pseudofaeces, and at 35 mg L-1 all oysters were producing both faeces and pseudofaeces (Wisely & Reid, 1978).

The lower efficiency of smaller 35 g oysters used in the recirculating experiments is expected due to the differences in pumping rates by different sized oysters (Wang, 1990). Estimates based on Wang (1990) suggest that the 35 g oysters would be pumping around 22 L d-1 and the larger 55 g oysters, 64 L d-1. Based on these pumping rates and the oyster stocking densities, the 55 g oysters filtered the entire volume of effluent in the raceway in the 2 h period taken for full water exchange. However, the 35 g oysters would have only filtered 20% of the total effluent in the raceway after the first circuit, and 80% after four circuits. Species differences, changes in water flow rate and the impacts of high concentration of food and other particulates (Loosanoff & Tommers, 1948; Ali, 1970) on the filtration rate of oysters may affect these calculations (Walne, 1972).

4.4.2 Condition of Biofilter Organisms The high concentrations of suspended solids in shrimp pond effluent are due to faecal matter, uneaten food, phytoplankton (Hopkins et al., 1994), and scouring the bottom

EFFICIENCY AND CONDITION OF BIOFILT ERS

111

sediments of earthen ponds by the action of aerators (Boyd, 1992). These suspended particulates can have considerable impacts on the growth and condition of oysters and macroalgae and their ability to reduce the phytoplankton, bacterial and nutrient concentrations of the effluent (Hopkins et al., 1993a; Funge-Smith & Briggs, 1998; Funge-Smith & Briggs, in prep.). To improve the filtration efficiency and the condition and growth of the biofilters, sedimentation could be carried out. The benefits of sedimentation to the improvements in water quality that can be obtained by biofilters has been shown previously (Wang, 1990; Jones et al., in review; Chapter 5).

The results from the present study indicate that there may be an optimal amount of settling with regard to maintaining long term oyster and macroalgal condition and growth. Oyster volume showed the greatest increase in the 6 h settling treatment, which may be a result of increased food availability, or even the presence of medium concentrations of clay particles which have been shown to increase the growth rate of oysters (Huntington & Miller, 1989). Despite a slight increase in volume in the 6 h treatment, there was no observed increase in total wet weight (meat plus shell). A prolonged lack of food can initiate shell growth rather than meat gain (Brown & Hartwick, 1988). Given previous growth rates recorded for oysters in shrimp effluent (0.04 g to 55 g in 4 months) (Jakob et al., 1993), the weekly exchange of effluent in the present study was probably not sufficient to maintain food supply.

The macroalgae also demonstrated a maximum growth response in the 6 h settling treatment, possibly as a result of increased nutrients from remineralisation of the remaining particulates (Blackburn et al., 1988). Settling the effluent for 6 h appeared to be optimal for growth of oysters and macroalgae in the present study. This time is

CHAPTER 4

112

likely to vary considerably as a result of differences in effluent composition (seasonal and overall farm differences), settlement pond design, water flow regime through the biofilters, and biofilter stocking density, and therefore must be investigated fully and adapted for the specific application.

The reduced light availability in the 6 h settling treatment may have also helped to reduce the effects of photoinhibition which were probably present in the 24 h and the 24 h + oyster filtered effluent treatments. The light availability to these treatments (~200 E m-2 s-1) was the same irradiance shown to limit growth of Gracilaria verrucosa (Engledow & Bolton, 1992). A light intensity of 100 E m-2 s-1 was the optimal light intensity for maximal growth (Engledow & Bolton, 1992). Light levels as low as 50 E m-2 s-1 can saturate growth of Gracilaria and the onset of necrosis can start as low as 65 E m-2 s-1 (Bird et al., 1979). Prior to the start of the present study the macroalgae were showing signs of necrosis due to high light. Incident irradiance was reduced and the macroalgal condition improved, indicating possible photoinhibition even at these relatively low irradiances. Gracilaria requires relatively low light, but is inhibited by the settling of particles onto the mucilage on the thalli (Funge-Smith & Briggs, in prep.).

Photosynthetic pigments (chlorophyll a and phycoerythrin) of the macroalgae were elevated in response to high nutrients in the 24 h settled plus oyster filtration treatment, and in response to low light in the 0 h treatment. Increases in pigments in Gracilaria have previously been positively correlated with increased nutrient availability (Jones et al., 1996), and negatively correlated to light availability (Friedlander & Levy, 1995).

EFFICIENCY AND CONDITION OF BIOFILT ERS

113

The photosynthetic response between the seawater control and the settled shrimp effluent plus oyster filtered treatment may relate to an increase in pigment concentration, or an increase in the photosynthetic efficiency of the pigments in the macroalgae. The observed increases in electron transport rate and reduction in photoinhibition is consistent with the responses to controlled nutrient additions observed in other species of macroalgae (Jones & Dennison, 1998; Appendix 2). The use of rapid light curves to determine electron transport rate under a variety of light intensities provides information about the photosynthetic capacity and condition of the plant, rather than realistic determination of light saturation intensity. This technique has been shown to identify changes in plant condition in response to a variety of environmental stresses including light limitation and inhibition, and nutrient availability (Hanelt et al., 1994; Herrmann et al., 1995; Cunningham et al., 1996; Hader et al., 1996; Jones & Dennison, 1998; Appendix 2). The increase in photosynthetic capacity of the macroalgae incubated in settled and oyster filtered shrimp effluent indicates the potential for increased growth rates and nutrient uptake.

Tissue N was elevated in the treatment with the highest turbidity (0 h) and, based on the nutrient excretion observed by Jones et al. (in review) (Chapter 5), presumably the highest nutrient availability (24 h plus oyster filtration). Tissue N content of Gracilaria has been correlated to nutrient availability (Jones et al., 1996) and the concentration of photosynthetic pigments, especially phycoerythrin (Horrocks et al., 1995; Jones et al., 1996). The high 15N in the 0 h treatment may be a response to higher concentrations of bacteria associated with the high particulate load (Hoch et al., 1994). The considerable depression in 15N in the 24 h plus oyster filtration treatment may be a result of higher discrimination for 14N during nutrient uptake, facilitated by higher nutrient availability (McClelland & Valiela, 1998).

CHAPTER 4

114

The total concentration of free amino acids was also related to changes in light and nutrient availability. In addition to being related to increased nutrient availability (Di Martino Rigano et al., 1992; Jones et al., 1996), high concentrations of free amino acids have been correlated with reduced light availability (Vergara & Niell, 1995). Specific changes in the amino acid composition may also be related to increased nutrient availability (Jones et al., 1996), or light limitation. The percentage of citrulline in Gracilaria has been positively correlated with high NH4+ availability (Jones et al., 1996).

The interplay between light and nutrient availability had significant impacts on the physiology of the macroalgae. Some of these changes may be interpreted as indicators of condition and potential for growth. Both oysters and macroalgae are marketable crops which, as well as improving the quality of effluent released from aquaculture facilities, may be additional sources of income which require no further inputs of nutrients or feed to maintain a viable crop. In addition to direct sale as food on Asian markets, there are several other direct uses for macroalgae including use in pelleted feeds for shrimp (Briggs & Funge-Smith, in review), and as fresh feed for abalone, a rapidly expanding aquaculture species (Oakes & Ponte, 1996). The use of high nutrient content macroalgae such as Gracilaria has been shown to improve growth in abalone (Fleming, 1995). Extraction of agar and carrageenin from red macroalgae is probably the most economically viable use at present (Indergaard & Rueness, 1990). Light and nutrients can have considerable impact on agar and carrageenin production. Increased dissolved nutrients has been shown to reduce the content of agar and carrageenin in macroalgae, but increase the gel strength which is important for viability in commercial processing (Bird et al., 1981).

EFFICIENCY AND CONDITION OF BIOFILT ERS

115

4.4.3 Conclusion Polyculture (culture of several organisms in the one culture unit) and integrated culture, which is the culture of several species in discrete culture units (Chien & Tsai, 1985), are perhaps more ecologically sound methods of aquaculture (Mackay & Lodge, 1983), with the most efficient use of resources and with the highest resilience against environmental fluctuation (Chien & Liao, 1995). Management of a polyculture or integrated aquaculture facility can be more complex with respect to stocking densities, culture techniques and associated infrastructure, harvesting procedures, and effluent flow management (Chien & Liao, 1995). Studies like this, however, can provide key information about the interactions between the effluent and the secondary biofilter crops, and between the biofilters themselves. This information can be used to better design commercial scale facilities. In particular, the successful integration of shrimp, oysters and macroalgae in a commercially viable operation will require proper physical treatment of the effluent prior to biological filtration. It has been demonstrated in the present study that pre-settlement of particulates can improve the growth and condition of both oysters and macroalgae incubated in shrimp pond effluent.

Properly designed and managed sedimentation, oyster filtration of particulates (with an optimised flow regime and appropriate oyster stocking densities), and macroalgal absorption of dissolved nutrients has the potential to effectively improve effluent water quality and produce two additional crops from otherwise wasted nutrients.

CHAPTER 5 IMPROVEMENTS IN WATER QUALITY OF AQUACULTURE EFFLUENT AFTER TREATMENT BY SEDIMENTATION, OYSTER FILTRATION AND MACROALGAL ABSORPTION

Abstract
Effluent water from shrimp ponds typically contains elevated concentrations of dissolved nutrients and suspended particulates, compared to influent water. Attempts to improve effluent water quality using filter feeding bivalves and macroalgae to reduce nutrients have previously been hampered by the high concentration of clay particles typically found in untreated pond effluent. These particles inhibit feeding in bivalves and reduce photosynthesis in macroalgae by increasing effluent turbidity. The effectiveness of a three stage effluent treatment system was investigated. In the first stage, reduction in particle concentration occurred through natural sedimentation. In the second stage, filtration by the Sydney rock oyster, Saccostrea commercialis (Iredale and Roughley) further reduced the concentration of suspended particulates, including inorganic particles, phytoplankton, bacteria, and their associated nutrients. In the final stage, the macroalga Gracilaria edulis (Gmelin) Silva absorbed dissolved nutrients. Pond effluent was collected from a commercial shrimp farm, taken to an indoor culture facility and left to settle for 24 h. Subsamples of water were then transferred into laboratory tanks stocked with oysters and maintained for 24 h, and then transferred to tanks containing macroalgae for another 24 h. Total suspended solids, chlorophyll a, total nitrogen (N), total phosphorus (P), NH4+, NO3-, and PO43-, and bacterial numbers were compared before and after each treatment at: 0 h (initial); 24 h (after sedimentation); 48 h (after oyster filtration); 72 h (after macroalgal absorption). The combined effect of the sequential treatments resulted in significant reductions in the concentrations of all parameters measured. High rates of nutrient regeneration were observed in the control tanks, which did not contain oysters or macroalgae. Conversely, significant reductions in nutrients and suspended particulates after sedimentation and biological treatment were observed. Overall improvements in water quality (as a percentage of the initial concentration) were as follows: total suspended solids (12%); total N (28%); total P (14%); NH4+(76%); NO3(30%); PO43- (35%); bacteria (30%) and chlorophyll a (0.7%). These results suggest that sequential treatment of pond effluent by natural sedimentation, oyster filtration and macroalgal nutrient absorption could significantly improve shrimp farm effluent water quality.

CHAPTER 5

118

5.1 Introduction Quantitative comparisons of shrimp farm influent and effluent water have demonstrated that the effluent can contain elevated concentrations of dissolved nutrients, phytoplankton, bacteria and other suspended organic and inorganic solids (Ziemann et al., 1992). The potential adverse environmental impacts from untreated effluent have raised concerns about the sustainability of shrimp farming (Phillips et al., 1993; Primavera, 1994). This has prompted the search for cost effective methods of improving effluent water quality prior to discharge into receiving waters.

Traditional methods of wastewater (sewage) treatment are ineffective and prohibitively expensive for application in treating shrimp pond effluent (Hopkins et al., 1995b). A potentially viable alternative is biological treatment of the effluent using oysters and macroalgae to remove suspended particulates and nutrients (Shpigel et al., 1993b). The organic component of pond effluent can provide a rich source of food for bivalves (Newell & Jordan, 1983). Bivalves, such as oysters, can also facilitate the removal of fine inorganic matter from suspension by coagulating this material into larger, more settleable particles and egesting them as pseudofaeces (Tenore & Dunstan, 1973).

Previous studies have shown that filtration by oysters can significantly reduce the concentrations of bacteria, phytoplankton, total nitrogen (N) and total phosphorus (P), and other suspended particles in shrimp pond effluent (Jones & Preston, 1999; Chapter 3). However, if sediment loads are too high, oyster filtration can be reduced or cease completely (Loosanoff & Tommers, 1948). Other studies have observed the problems associated with a high concentration of suspended solids on the health of oysters (Hopkins et al., 1993a).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

119

These studies indicate the need to reduce the concentration of suspended particles by sedimentation prior to filtration by oysters.

Regular water exchange in shrimp ponds is required to maintain adequate water quality for optimal growth conditions. In particular, the toxic effects of high ammonia concentration on the shrimp (Chien, 1992) can be a critical factor determining water exchange rates (Wajsbrot et al., 1989; Hopkins et al., 1993a). Although oysters can reduce the concentration of some particulate and dissolved nutrients (Manahan et al., 1982; Dame, 1996), they can also increase the concentration of NH4+ through excretion (Hammen et al., 1966). Of the N absorbed by oysters, approximately 10% goes to growth, 10% to gametes, 2% to byssal threads, with 50% lost as biodeposition and 27% as excretion (Dame, 1996).

Various species of macroalgae can rapidly assimilate large quantities of dissolved organic and inorganic nutrients, usually with a preference for NH4+ (D'Elia & DeBoer, 1978; Haines & Wheeler, 1978; Hanisak & Harlin, 1978; Harlin, 1978; Topinka, 1978; Ryther et al., 1981). Rhodophyta (red macroalgae) are particularly efficient at taking up nutrients rapidly and have mechanisms for storing large reserves of nutrients (Vergara et al., 1993). For example, the red macroalga Gracilaria edulis rapidly assimilates NH4+ (Jones et al., 1996), and another red macroalgae Kappaphycus alvarezii has been effectively used to assimilate waste nitrogen from the pearl oysters Pinctada martensi (Qian et al., 1996).

In addition to improving the water quality of shrimp effluent water, macroalgae and oysters can provide an additional source of income for shrimp farmers. For example, trials into tank culture of Gracilaria chilensis supplied with salmon seawater effluent have demonstrated production rates of four times those in wild beds and up to double the agar content

CHAPTER 5

120

(Retamales et al., 1994). In the absence of waste nutrients from salmon cages, shrimp ponds or other high value aquaculture species, the costs of providing sufficient levels of nutrients for intensive culture of relatively low value species such as oysters and macroalgae are a constraint to profitability (Neish, 1979).

This study used controlled laboratory experiments to test the combined efficiency of a three stage pond effluent treatment system. The water quality after sedimentation, filtration of particulates by oysters (Saccostrea commercialis) and absorption of nutrients by macroalgae (Gracilaria edulis) was assessed.

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

121

5.2 Materials and Methods 5.2.1 Experimental Design Shrimp pond effluent was collected from a commercial Penaeus japonicus shrimp farm in Moreton Bay, Australia. The farm had a total of 8 ha of ponds and at the time of sampling, the shrimp biomass in the ponds was approximately 3 t ha -1. A 60 L water sample was collected from the effluent channel and transported to the laboratory in a plastic drum (35 cm diameter 60 cm height). In the laboratory, any settled particles in the sample were resuspended and 3 replicate 1 L samples were collected for analysis of total suspended solids (and percent organic of particulates), chlorophyll a, bacterial concentration, total Kjeldahl nitrogen (TKN) and total phosphorus (TP), and dissolved N (NH4+ & NO3-/NO2-), and dissolved P (PO43-). A suite of physico-chemical parameters (temperature, pH, salinity, dissolved oxygen, and turbidity) were determined with a Horiba U-10 water quality meter (California, U.S.A.). 30 mL samples of water were collected in test tubes, placed in the dark for 30 minutes and subsequently placed into a Turner Designs Fluorometer 10-005 (Sunyvale, California, U.S.A.) for determination of in vivo fluorescence (as a measure of chlorophyll a).

The effluent was left to settle in the dark for 24 hours in the collection drum, with physicochemical parameters, in vivo fluorescence, TKN and TP measured at 0.25 h, 0.5 h, 1 h, 2 h, 3 h, 6 h, 12 h, and 24 h. The remaining parameters, dissolved N (NH4+ & NO3-/NO2-), dissolved P (PO43-), bacterial concentration, chlorophyll a and total suspended solids were analysed after 24 h only. A sediment trap (23 cm 15 cm) was placed on the bottom of the drum to determine the amount of settled material per litre (after 24 h) and the percent organic content of the settled particles.

CHAPTER 5

122

Experimental tanks (Fig. 5.1; Plate 5.1) were maintained at 20 - 23C, and exposed to light on a 12:12 h light / dark cycle using daylight fluorescent tubes which provided approximately 250 mol quanta m-2 s-1. After 24 h, 10 L of the settled sample was drained into each of four, 11 L aerated tanks (26 cm 17 cm 25 cm), one as a control (with dead oyster shells), and three for replicate oyster filtration treatments. The 11 L tanks were each stocked with 16 oysters with a mean wet weight of 40 g. Dead oyster shells were used in the control tanks to negate any effects on water flow and consequent differences in settling rates. Oysters were placed on plastic mesh, which was situated on top of the sediment trap. Physico-chemical parameters, in vivo fluorescence, TKN and TP were measured at 24.25 h, 24.5 h, 25 h, 26 h, 27 h, 30 h, 36 h, and 48 h. The remaining parameters, dissolved N (NH4+ & NO3-/NO2-), dissolved P (PO43-), bacterial concentration, chlorophyll a, total suspended solids and sediment traps were analysed after 48 h only.

After 24 h of oyster filtration, 5 L from each tank was drained into each of four more 11 L aerated tanks, with one as a control (empty), and three as replicate macroalgal absorption treatments (100 g wet weight of macroalgae per 5 L) (Fig. 5.1). The water from the oyster control tank was transferred into the macroalgal control tank, and the water from the replicate oyster treatment tanks was transferred into the corresponding macroalgal treatment tanks (Fig. 5.1). Both the oyster and macroalgal control tanks acted as controls for the combined biological treatment process using both the oysters and macroalgae. Consequently, the effluent in the macroalgal control tank contained higher concentrations of phytoplankton, bacteria and other suspended solids, because it had not been treated by the oysters. Physicochemical parameters, dissolved N (NH4+ & NO3-/NO2-), and dissolved P (PO43-) were measured at 48.25 h, 48.5 h, 49 h, 50 h, 51 h, 54 h, 60 h, and 72 h. The remaining parameters, total suspended solids, TKN, TP, chlorophyll a, bacterial concentration and

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

123

sediment traps were analysed after 72 h only. At the end of each 24 h period, 3 replicate 1 L samples were collected from each tank and filtered for chlorophyll a extraction, total suspended solids calculation and the percent organic. Subsamples were taken for TKN and TP, dissolved N (NH4+ & NO3-/NO2-), dissolved P (PO43-), and bacterial concentration.

At all time periods excepting 0 h, 24 h, 48 h and 72 h, chlorophyll a and TSS were determined as in vivo fluorescence and turbidity (NTU), respectively. Preliminary analysis determined correlations between in vivo fluorescence and chlorophyll a, and between turbidity (NTU) and total suspended solids to be r2 = 0.86 and 0.9, respectively. In vivo fluorescence measurements were converted to chlorophyll a concentration and turbidity measurements were converted to total suspended solids for those periods of additional sampling at 15 mins, 30 mins, 1 h, 2 h, 3 h, 6 h, 12 h.

5.2.2 Analytical Procedures Chlorophyll a was determined by filtering a known volume of water sample through Whatman GF/F filters, which were immediately frozen. Acetone extraction and calculation of chlorophyll a concentration was performed using the methods of Clesceri et al. (1989), and Parsons et al. (1984).

The filtrate collected from filtering for chlorophyll a analysis was collected in 120 mL polycarbonate containers and immediately frozen. NH4+ and NO3-/NO2- and PO43- were determined by the Queensland Health Analytical Services Laboratory (NATA- accredited) in accordance with the methods of Clesceri et al. (1989) using a Skalar autoanalyser (Norcross, Georgia, U.S.A.).

CHAPTER 5

124

Sedimentation (60 L)
(tank not aerated)

Oysters (10 L)
(tanks aerated)

Macroalgae (5 L)
(tanks aerated)

Raw Effluent

Control (16 oyster shells)

Control (no algae) 100g Macroalgae 100g Macroalgae 100g Macroalgae

16 Oysters

16 Oysters

16 Oysters

0
macroalgae ( Gracilaria edulis).

24

Time (h)

48

72

Figure 5.1 Design of integrated treatment system stocked with oysters (40 g Saccostrea commercialis), and

Plate 5.1 Experimental setup with sedimentation drum (background) and control, oyster, macroalgal filtration tanks (foreground).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

125

Unfiltered samples for nutrient analysis (total Kjeldahl nitrogen and total phosphorus) were collected in 120 mL polycarbonate immediately frozen. They were subsequently analysed within two weeks by the NATA accredited Queensland Health Analytical Services Laboratory in accordance with the methods of Clesceri et al. (1989) using a Skalar autoanalyser (Norcross, Georgia, U.S.A.).

Total suspended solids concentrations were determined using the methods of Clesceri et al. (1989). A known volume of water was filtered onto a pre-weighed and predried (110 C; 24 h) Whatman GF/C glass fibre filter. The filter was then oven dried at 60 C for 24 h and total suspended solids calculated by comparing the initial and final weights. Volatile suspended solids were determined as loss on ignition by combusting samples in a muffle furnace for 12 h at 525 C (Clesceri et al., 1989).

Bacteria samples were preserved with 2% formalin and kept at 4C until analysis. A known volume (0.5 mL - 1 mL) of sample was stained with acridine orange, filtered onto a stained (Irgalan Black) 2 m poretics filter and mounted on a slide. Bacteria were counted using epifluorescence microscopy (Hobbie, 1977).

CHAPTER 5

126

5.3 Results 5.3.1 Suspended Solids The initial concentration of total suspended solids was 0.60 g L-1. After sedimentation, the concentration was reduced to 0.17 g L-1. Oyster filtration further decreased the concentration to 0.02 g L-1. There was no significant change in the oyster control, indicating that material with a specific gravity greater than water had already settled out of suspension during the 24 h sedimentation period. There was no significant change in the concentration of total suspended solids in the macroalgal control and treatments (Fig. 5.2A; Table 5.1). During sedimentation, 1.9 g of material was collected in the sediment traps. In the oyster treatment, 0.42 g was collected, versus 0.02 g in the oyster control. This indicates significant biodeposition of faeces and pseudofaeces by the oysters (Fig. 5.3).

5.3.2 Organic content In the first 24 h, sedimentation removed a greater proportion of inorganic particles than organic particles. The relative organic content of the remaining suspended solids increased from 23% to 31%. The organic content of the settled material collected in the sediment traps was 21%, which was lower than the initial water column content of 23%. Sedimentation in the oyster control tanks did not significantly change the organic content. Oyster filtration was more effective at removing organic material (phytoplankton, bacteria and detritus), decreasing the relative organic content of the suspended solids from 31% to 24%. The organic content of the settled sediment at the end of the oyster treatment was 31%, compared to 21% at the end of the sedimentation treatment (Fig. 5.4B). These changes mirrored the changes in the organic content of the suspended particles (Fig. 5.4A). In the macroalgal treatment and control, the organic content of the suspended solids remained unchanged (Fig. 5.4A).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

127

0.7 0.6 0.5

Settling

Oysters

Macroalgae

TSS (g L -1)

0.4 0.3 0.2 0.1 0

B
Chlorophyll a (ug L-1)

300 250 200 150 100 50 0 0 10 20 30 40 50 60 70 80

Time (h) Control (no oysters or macroalgae)

20 5

Oysters

Macroalgae

Figure 5.2 Changes in total suspended solids (A) and phytoplankton biomass (chlorophyll a) (B) from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

CHAPTER 5

128

Table 5.1 Percentage of original concentrations of various water quality parameters after settling, filtration by oysters and filtration by macroalgae.
*

p 0.05; ** p 0.01; *** p 0.001. Percentage of highest concentration

represents the final concentration as a percentage of the highest recorded concentration after sedimentation and oyster filtration. The percent of initial concentration represents the final concentration as a percentage of the initial concentration in the untreated effluent. The only differences between the two values are for the dissolved inorganic nutrients (NH4+, NO3-, & PO43-).

Percent of highest concentration TSS (g L-1 ) TSS % Organic (g L-1) TKN (M) TP (M) NH4+ (M) NO3- (M) PO43- (M) Bacteria (no. per L) Chl a (g L-1) 12 12 28 14 2.3 2.2 4.8 30 0.7

F Value 1338*** 507*** 609*** 4629*** 2987*** 1176*** 162*** 107*** 202***

Percent of initial concentration 12 12 28 14 76 30 35 30 0.7

F Value 1338*** 507*** 609*** 4629*** 0.69 30** 19.6* 107*** 202***

5.3.3 Chlorophyll a Sedimentation reduced the chlorophyll a concentration from 180 g L-1 to 130 g L-1. In the oyster control, the concentration of chlorophyll a was further reduced from 130 g L-1 to 100 g L-1 due to continued settling. The reduction in the chlorophyll a concentration from 130 g L-1 to 11 g L-1 in the oyster treatment was significantly greater than the control. The concentration of chlorophyll a decreased from 100 g L-1 to 51 g L-1 in the macroalgal control (no macroalgae) and from 11 g L-1 to 1.5 g L-1 in the macroalgal treatment. These decreases indicate additional settling of phytoplankton (Fig. 5.2B; Table 5.1).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

129

0.15

Settling

Oysters

Macroalgae

Settled particles (g L -1)

0.1

0.05

0 0 10 20 30 40 50 60 70 80

Time (h)

Control (no oysters or macroalgae)

Oysters

Macroalgae

Figure 5.3 Concentration of particles settled per litre from sedimentation and oyster filtration.

5.3.4 Bacteria During sedimentation, bacterial numbers did not change significantly, whereas the oyster treatment significantly reduced the concentration of bacteria from 19 1010 L-1 to 6 1010 L-1. There was no significant change in bacterial numbers in the macroalgal controls and treatments (Fig. 5.5; Table 5.1).

CHAPTER 5

130

A
TSS (% Organic)

40

Settling

Oysters

Macroalgae

35

30

25

20

15

B
Settled particulates (% Organic)

40 35 30 25 20 15 0 10 20 30 40 50 60 70 80

Time (h) Control (no oysters or macroalgae)

100

2 50

Oysters

Macroalgae

Figure 5.4 Changes in the organic content of the a) total suspended solids (TSS) and b) settled particles in the effluent water from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

131

25
Bacteria (no. 10 10 L -1)

Settling

Oysters

Macroalgae

20 15 10 5 0 0 20 40
Time (h)

60

80

Control (no oysters or macroalgae)

Oysters

Macroalgae

Figure 5.5 Changes in bacterial numbers from sedimentation, oyster filtration and macroalgal absorption.

5.3.5 Dissolved Oxygen Dissolved oxygen was 6.3 mg L-1 immediately following the initial resuspension of any settled particles prior to sedimentation. After 24 h of sedimentation without stirring or aeration, respiration by bacteria and plankton reduced the concentration to 2.6 mg L-1. Within 2 h of aeration being applied, the concentration of dissolved oxygen was up to 6.3 mg L-1. The concentration peaked at between 7 mg L-1 and 8 mg L-1 in the macroalgal treatment and control tanks (Fig. 5.6).

CHAPTER 5

132

9 Dissolved Oxygen (mg L -1 ) 8 7 6 5 4 3 2 1 0 0

Settling

Oysters

Macroalgae

20

40
Time (h)

60

80

Control (no oysters or macroalgae)

Oysters

Macroalgae

Figure 5.6 Changes in water column dissolved oxygen concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

5.3.6 Total Nitrogen The sedimentation treatment reduced the concentration of total Kjeldahl nitrogen (TKN) from 290 M to 205 M. The oyster treatment further reduced the concentration to 138 M, while in the control the concentration increased to 266 M. The macroalgal treatment reduced the concentration from 138 M to 81 M, while in the control the concentration increased from 266 M to 279 M (Fig. 5.7A; Table 5.1).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

133

A
Total Nitrogen (M)

350

Settling
300 250 200 150 100 50 0

Oysters

Macroalgae

B
Total Phosphorus (M)

25

20

15

10

0 0 10 20 30 40 50 60 70 80

Time (h) Control (no oysters or macroalgae) Oysters Macroalgae

Figure 5.7 Changes in water column total N (A) and P (B) concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

5.3.7 Total Phosphorus Total phosphorus (TP) also decreased during sedimentation from 21 M to 9.7 M. The oyster control increased the concentration to 12.5 M, whereas the oyster treatment reduced the concentration to 6.1 M. The macroalgal control reduced the concentration from

CHAPTER 5

134

12.5 M to 11.5 M and the macroalgal treatment reduced the concentration from 6.1 M to 2.9 M (Fig. 5.7B; Table 5.1).

5.3.8 Ammonium The concentration of water column NH4+ increased significantly during sedimentation from 1.7 M to 18 M due to regeneration. Regeneration rates were calculated as mol h-1 based on the effluent volumes used during each treatment period (i.e., 60 L for sedimentation, 10 L for the oyster treatment, and 5 L for the macroalgal treatment). Thus, the observed rate of regeneration during sedimentation was 40.5 mol h-1, although this may have been substantially higher if the loss due to volatilisation and uptake by bacteria and phytoplankton was taken into account. The rate of NH4+ regeneration in the oyster control was considerably lower (5.7 mol h-1) increasing from 18 M to 29 M. This was probably a result of the decrease in the concentration of TSS, thereby reducing the source of remineralisation. The NH4+ concentration in the oyster treatment increased from 18 M to 51 M. During the macroalgal treatment, the NH4+ concentration was reduced from 51 M to 1.3 M (to 1.0 M after 2 h). The uptake rate during the first hour (when N concentrations were still saturating uptake kinetics) was 225 mol h-1 for the 100 g of macroalgae. Despite the rapid removal of N from the effluent, this uptake rate is substantially lower than recorded literature values for other species of Gracilaria (Friedlander & Dawes, 1985; Peckol et al., 1994), suggesting that removal rates could be improved. The concentration of NH4+ in the macroalgal control decreased from 29 M to 1.2 M, but at a much slower rate (only to 26 M after 2 h) than in the macroalgal treatment (Fig. 5.8A; Table 5.1; Table 5.2). The reduction in NH4+ in the macroalgal control suggests volatilisation, nitrification, or uptake by the high concentration of bacteria and phytoplankton.

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

135

A
Ammonium (M)

60 50 40 30 20 10 0

Settling

Oysters

Macroalgae

B
Nitrate / Nitrite (M)

16 14 12 10 8 6 4 2 0 4

C
Phosphate (M)

3.5 3 2.5 2 1.5 1 0.5 0 0 10 20 30 40 50 60 70 80

Time (h) Control (no oysters or macroalgae) Oysters Macroalgae

Figure 5.8 Changes in water column NH4+, NO3-, PO43- concentrations from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

CHAPTER 5 Table 5.2 Nutrient uptake and release rates for sedimentation, oyster filtration and macroalgal absorption.

136

Negative symbols represent nutrient uptake, and positive represent nutrient release. The top value for each treatment is the gross value, the middle value is the control and the bottom value (in bold type) is the net value after correction for nutrient changes in the control tanks. The last row of results represent the rates of macroalgal nutrient uptake over the first hour, when nutrient concentrations were still saturating uptake kinetics.

Treatment

TKN mol h-1

TP mol h-1

NH4+ mol h-1

NO3mol h-1

PO43mol h-1

Sedimentation Gross Oysters Gross Control Net Macroalgae Gross Control Net Macroalgae over 1 h Gross Control Net nd nd nd nd nd nd -225 -7 -218 -10 +1 -11 -3.4 -2.1 -1.3 -12 +2.5 -14.5 -0.65 -0.2 -0.45 -10.4 -5.8 -4.6 -2.8 0.0 -2.8 -0.66 -0.26 -0.4 -28 +26 -54 -1.5 +1.15 -2.65 +14 +5.7 +8.3 +5.0 +0.6 +4.4 +1.2 +0.6 +0.6 -220 -28 +40.5 +0.05 +0.18

5.3.9 Nitrate / Nitrite The concentration of NO3-/ NO2- was unchanged during the sedimentation period, where the effluent was not aerated, but increased from 1.0 M to 1.4 M in the oyster control and from 1.0 M to 13 M in the oyster treatment. There was no significant change in the NO3-/NO2-

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

137

concentration in the macroalgal control, but the macroalgae in the treatment tanks reduced the concentration from 13 M to 0.3 M (Fig. 5.8B; Table 5.1).

5.3.10 Phosphate Phosphate was also unchanged during sedimentation, but increased from 0.5 M to 2 M in the oyster control and from 0.5 M to 3.3 M in the oyster treatment. The concentration in the macroalgal control decreased from 2 M to 0.7 M, and in the macroalgal treatment the concentration of P decreased from 3.3 M to 0.16 M (Fig. 5.8C; Table 5.1).

These reductions in PO43- equate to a removal rate of 0.66 mol h-1 in the macroalgal treatment, while loss in the control was only 0.26 mol h-1. During the first hour of macroalgal filtration, the concentration of PO43- in the control decreased at a greater rate than in the macroalgal treatment, presumably due to phytoplankton and bacterial uptake. This may have been due to the presence of significantly more phytoplankton and bacteria in the control tanks, because the control water had not been treated by the oysters.

5.3.11 Nutrient Uptake Rates and Ratios During the macroalgal treatment, NH4+ was taken up in the first 2 h (hour 48 to hour 50), NO3- in the next 4 h (from hour 50 to hour 54), and PO43- over 10 h (from hour 50 to hour 60). N in both forms was taken up at faster rates than PO43-.

The TKN: TP ratio in the water column increased during sedimentation from 6.4 to 9.6, and then during oyster filtration from 9.6 to 10.2, probably due to adsorption of P to settling particulates. There was no significant change in the oyster control, but the macroalgal control

CHAPTER 5

138

showed some further increase to 10.9, and the macroalgal treatment increased to 12.6 (Fig. 5.9A).

The ratio of DIN: DIP increased from 2.6 to 16.2 during sedimentation, predominantly as a result of NH4+ remineralisation. During the oyster treatment and control, the ratio dropped to 8.7 and 7.5, respectively. The macroalgal control increased the ratio to 18.8 after 6 h and then decreased to 2.3, whereas the macroalgal treatment decreased the ratio to 0.8 after 6 h and then increased it to 4.5. These responses coincided with the rapid drop in the PO43- in the macroalgal control in the first six hours, and the rapid uptake of NH4+ in the macroalgal treatment (Fig. 5.9B).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

139

A
TKN:TP Ratio

13

Settling

Oysters

Macroalgae

11

B
DIN:DIP Ratio

20

15

10

0 0 10 20 30 40 50 60 70 80

Time (h)

20 5

100

Control (no oysters or macroalgae)

Oysters

Macroalgae

Figure 5.9 Changes in water column total N: P ratio (A) and DIN: DIP ratio (B) from sedimentation, oyster filtration and macroalgal absorption. Standard error bars have been plotted, but are too small to be visible.

CHAPTER 5

140

5.4 Discussion

Previous studies on the use of sedimentation, bivalves and macroalgae to improve effluent water quality have generally examined each component separately. For example, Shpigel et al. (1993b) examined three separate regimes for treating fish pond effluent and concluded these could be more closely integrated. The present study advanced the concept to the point of laboratory tests to examine the combined efficiency of sedimentation, biofiltration by oysters and nutrient uptake by macroalgae.

The ability of a biofilter system to improve the quality of shrimp effluent is likely to vary depending on the initial water quality. The water quality of shrimp farm effluent depends on a number of factors, including pond soil type, quality of influent water, stage of growout season and management practices employed (Ziemann et al., 1992). The present study was conducted in November, early in the growout season when the concentration of NH4+ was relatively low (1.7 M), compared to later in the growout season (65 M) (Jones et al., in prep a; Chapter 2). However, there were very high concentrations of total suspended solids and phytoplankton in the effluent at the time of the study. The concentrations of total suspended solids (0.6 g L-1), chlorophyll a (180 g L-1) and bacteria (20 1010 L-1), were respectively 4.5, 5 and 10 times the values measured from the same shrimp farm in the previous growout season (Jones & Preston, 1999; Chapter 3).

5.4.1 Sedimentation Sedimentation of shrimp pond effluent was very effective at reducing the total suspended solid (TSS) load. The concentration was reduced to 12% of the initial concentration with proportionally more inorganic particles being removed from suspension. Additional benefits

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

141

were reductions in total Kjeldahl nitrogen (TKN) to 70%, total phosphorus (TP) to 47%, chlorophyll a to 72%, and bacteria to 95% of the initial concentration. The results indicate that the majority of N in the effluent was associated with phytoplankton and bacteria, rather than being bound to readily settleable inorganic particles, or detritus. In contrast to this, a relatively high proportion of the TP appeared to be associated with the readily settleable particles.

In the present study, the rates of TKN removal by sedimentation and oyster filtration were not significantly different, however the rate of TSS removal was much slower during oyster filtration compared with sedimentation. This may be a result of higher N content in the small (<5 m) unsettleable particles (Cripps, 1992) removed by the oysters. Alternatively, it may simply reflect relatively slow settlement rates of phytoplankton compared to TSS. Despite the reduction in TKN and TP during sedimentation, there was a significant increase in the concentration of NH4+ in the water column, most likely due to mineralisation of particulate organic matter (Briggs & Funge-Smith, 1993).

Treatment for the removal of suspended solids must be adapted specifically for the target effluent. The size and density of the particles, along with the flow rates, surface area, and retention time affects the efficiency of sedimentation (Chien & Liao, 1995). Effluent with high algal concentrations may require less sedimentation time, because it can be filtered more effectively by oysters.

5.4.2 Oyster Filtration Pre-sedimentation of suspended particles is highly beneficial to oyster filtration. In particular, the relative increase in the proportion of organic particles (from 23% to 31%) may

CHAPTER 5

142

enhance the oysters ability to assimilate particulate nutrients, as well as filter and egest the smaller inorganic clay particles as pseudofaeces (Loosanoff, 1949). In the present study, the concentration of total suspended solids was reduced from 0.6 g L-1 to 0.17 g L-1 after sedimentation. However, even at this reduced concentration of suspended solids the oyster pumping rates may have been inhibited. For example, a concentration of 0.1 g L-1 has been observed to reduce the pumping rate of oysters by up to 87% (Loosanoff & Tommers, 1948).

Ninety five percent of the suspended particulates in the effluent from this shrimp farm were 2 - 4 m (Jones et al., in prep b; Chapter 4). During filtration oysters sort particles by size, weight (Yonge, 1926) and chemical composition (Loosanoff, 1949; Menzel, 1955). Oysters preferentially ingest organic material, reject inorganic material, and preferentially ingest N rich over C rich particles (Newell & Jordan, 1983). Rejected material is prevented from ingestion by the gills and palps, where it is accumulated and then expelled through the inhalent opening as pseudofaeces (Barnes, 1994). When food concentration exceeds the digestive capabilities of the gut, pseudofaeces may contain some digestible food.

Despite rejecting small inorganic particles, oysters may facilitate removal of these particles from suspension by coagulating them into larger, more settleable particles before egestion as pseudofaeces. In addition to the fine inorganic particles, oysters filter small organic rich particles, such as bacteria and very small phytoplankton that would otherwise stay in suspension (Loosanoff, 1949; Kautsky & Evans, 1987). Oysters can filter particles as small as 1 m, although their efficiency in removing the smaller particles (1-3 m) is two thirds lower than for larger particles (Haven & Morales-Alamo, 1970).

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

143

The oysters removed large concentrations of phytoplankton, bacteria and other suspended solids from the water column, and produced faeces and pseudofaeces at a rate of 78 mg (g dry wt) -1 d-1. This rate is within the range of 66 to 246 mg (g dry wt) -1 d-1 observed for the Pacific oyster, Crassostrea gigas (Boucher & Boucher-Rodoni, 1988).

Decreases in TP were observed in the oyster treatment due to removal of particulate P (bacteria, phytoplankton and clay particles with adsorbed P) at a rate greater than the inputs of PO43-. Shrimp pond effluent often contains a high proportion of small clay particles (Hopkins et al., 1995b) rich in phosphate (Pomeroy et al., 1965). Bacteria may compete with sediments for phosphate thereby maintaining the phosphate in the water column (Pomeroy et al., 1965). The increase in PO43- and TP observed in the oyster control was probably due to sediment release and the greater increase in PO43- in the oyster treatment due to excretion by oysters (Asmus et al., 1995). Most of the P filtered by oysters is converted to biodeposits, with 8% being released as PO43- (Dame et al., 1989), and only 3% absorbed and converted into biomass (Sornin et al., 1986).

In the present study, oyster filtration reduced the concentration of TKN to 28% and TP to 14% of the initial effluent concentration. These reductions represent an uptake ratio of 20: 1. Based on the Redfield ratio of 16: 1 in phytoplankton, the oysters are removing more N than expected. However, this is in contrast to the results of Dame et al. (1989) who observed an N: P uptake ratio by oysters (Crassostrea gigas) of 2: 1.

During oyster filtration, there was a net decrease in the concentration of particulate nutrients, and a net increase in the concentration of dissolved inorganic nutrients. In addition to particulate matter, oysters can also take up dissolved organic and inorganic nutrients and

CHAPTER 5

144

dissolved organic matter (DOM) such as dissolved free amino acids from the water column (Manahan et al., 1982; Dame, 1996). In particular, oysters can assimilate PO43- directly from the water column for carbohydrate metabolism, energy transfer and shell deposition (Pomeroy & Haskin, 1954).

A net NH4+ excretion rate of 0.9 mol g-1 DW h-1, which is in the range described in the literature (Boucher & Boucher-Rodoni, 1988), was measured in the oyster treatment tanks. This takes into account inputs from regeneration, but does not incorporate the losses to phytoplankton and bacterial uptake of NH4+, volatilisation and nitrification which may significantly enhance these rates (Boucher & Boucher-Rodoni, 1988).

The low concentrations of dissolved oxygen (DO) during the first two hours of oyster filtration are likely to have reduced the filtration efficiency. Ingestion rates of bivalves are reduced under conditions of low DO availability (Sobral & Widdows, 1997). The oysters were maintained in no flow environments, apart from the water motion due to aeration. Flow rates have been positively correlated with oyster filtration rates (Walne, 1972), and so oysters in treatment ponds with flowing water might produce significantly greater improvements in water quality.

5.4.3 Macroalgal Absorption Gracilaria edulis in the present study achieved 87% of its total NH4+ uptake within the first hour. A similar rate (80%) was observed for the red macroalga Kappaphycus alvarezii, incubated in the nutrients excreted by the pearl oyster Pinctada martensi (Qian et al., 1996). The initial rate of NH4+ uptake in the present study was 218 mol h-1 per 100 g wet-1 (Table 5.2), which is very close to the rate observed by Qian et al. (1996) for 200 g wet-1 for

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

145

K. alvarezii, and is within the range observed for Gracilaria (Rivers & Peckol, 1995) and other species of macroalgae (Taylor et al., 1998). The NO3- uptake rates in the present study, however, were considerably lower, with a maximum rate of 15% per hour compared to 66% per hour for the K. alvarezii. Low light and high concentrations of NH4+ (D'Elia & DeBoer, 1978) can inhibit NO3- uptake. The ability to assimilate NO3- can also be species specific. Consistent with the findings for G. edulis, uptake of NO3- by Ulva lactuca was found to be minimal (Krom et al., 1995). Preferential uptake of NH4+ before NO3- is typical for most species of macroalgae and is consistent with the results of Oki & Fushimi (1992) who found that as the concentration of NH4+ decreased, the uptake rate of NO3- increased.

In the present study, the fastest rates of net uptake of NH4+, NO3- and PO43- by macroalgae were 26, 2.5 and 2.2 times higher than the net oyster release rate, respectively. The net rate takes into account losses to phytoplankton and bacterial uptake, volatilisation and nitrification (Table 5.2). The proportion of NH4+ to NO3- taken up by the macroalgae varied, being 20: 1 during the period of fastest uptake of NH4+ uptake (in the first hour), but only 1.6: 1 over the entire 24 hours. This overall rate is comparable to the oyster release rates of NH4+ and NO3-, which were 1.9: 1. The uptake of NH4+ and NO3- by K. alvarezii was in similar proportions to the release rates of the pearl oyster (Qian et al., 1996).

Macroalgae did not significantly change the NH4+: NO3-/NO2- ratio, but the uptake of N markedly increased the DIN: DIP on a short time scale, due to the saturating availability of N, and the ability of Gracilaria edulis to store luxury reserves of N for times of nutrient limitation (Gerloff & Krombholz, 1966; Jones et al., 1996). This reduction in the DIN: DIP ratio may have implications for water recirculation and downstream environmental impacts.

CHAPTER 5

146

Analysis of bioindicators downstream of shrimp farms has indicated responses typical of high NH4+ rich effluent (Jones et al., in prep a; Chapter 2).

The nutrient removal efficiency of macroalgae may be improved with increased light, particularly for nitrate uptake (Hanisak, 1979; Falkowski, 1983), increased water flow to reduce the boundary layer (Wheeler, 1980), and higher stocking densities (Ugarte & Santelices, 1992). The present study was conducted indoors under experimental conditions. Under full solar radiation, photosynthetic rates could be considerably higher, and nutrient uptake rates correspondingly higher. In a biological treatment pond, water flow rates would be determined by the physical volume of water being pumped through the pond, and the water movement due to the action of paddlewheels or other aerators. In the present study, the macroalgae was stocked at 2.3 kg m-2, whereas maximum growth rates of Gracilaria chilensis were achieved by stocking at 4 kg m-2 in winter and 8 kg m-2 in summer (Ugarte & Santelices, 1992).

Given the low concentrations of NH4+ in the effluent from the shrimp farm at the time of the study (early in the shrimp growout season), the total reduction in the concentration of NH4+ was lower than expected. The 24% reduction in NH4+ from 1.7 M to 1.3 M may not be indicative of the real potential for this system to remove NH4+. Due to addition of NH4+ to the system through the decay of epiphytes and macroalgal tissue, a concentration of approximately 1 M may be the equilibrium point between uptake and input from decay. Based on the fast rates of uptake observed in this study, it can be expected that during the late phase of the growout season, when the concentrations of NH4+ are considerably higher (up to 60 M) (Jones et al., in prep a; Chapter 2), that overall removal efficiency will be considerably higher.

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

147

The percent reductions by macroalgal nutrient absorption have been expressed as overall reductions from the initial concentration in the shrimp effluent. However, because remineralisation and oyster excretion significantly increased the concentration of NH4+, NO3-, and PO43-, the efficiency of nutrient removal can be expressed as a percentage of the concentration in the water received by the macroalgae. Based on these calculations, the concentration of NH4+ was reduced to 2.3%, NO3- to 2.2% and PO43- to 4.8% (Table 5.1).

5.4.4 Nutrient Regeneration During sedimentation, the rate of NH4+ regeneration was 1.8 mmol m-2 d-1, although this did not take into account loss of NH4+ by phytoplankton and bacterial uptake, or by volatilisation. The NH4+ regeneration rate declined markedly after transfer to the aerated oyster control tanks. This may have been due to the reduction in particulates, or the result of increased aeration. These results are consistent with those of Kamiyama et al. (1997), who observed increases in NH4+ release during low dissolved oxygen conditions, and NO3- release under high dissolved oxygen conditions. The NH4+ regeneration rates during sedimentation were significantly lower than the sediment NH4+ release rates of between 8.4 mmol m-2 d-1 and 18 mmol m-2 d-1 that have been recorded in fish ponds (Riise & Roos, 1997). In the present study, the NH4+ uptake rate by the macroalgae was 118 mmol m-2 d-1. At this rate, the macroalgae would be able to remove up to 6.5 times the NH4+ released from the fish ponds observed by Riise & Roos (1997).

During sedimentation when the concentration of dissolved oxygen was very low, there was no production of NO3-, despite significant amounts of NH4+ remineralisation. This is typical of anaerobic aquaculture ponds (Blackburn et al., 1988). Once under aerobic conditions in

CHAPTER 5

148

the aerated oyster controls, nitrification was observed, although rates were still very low. In the oyster treatment tanks, there was significant production of NO3-, which is consistent with the nutrient release by the pearl oyster Pinctada martensi (Qian et al., 1996).

Oysters excrete NH4+, amino acids, urea, uric acid (Hammen et al., 1966) and PO43(Pomeroy & Haskin, 1954; Dame et al., 1989; Dame, 1996). The nitrogen release rate by the pearl oyster for NH4+ was 0.52 mol h-1 and for NO3-, 0.44 mol h-1, per oyster (Qian et al., 1996), compared to 0.52 mol h-1 and 0.28 mol h-1, respectively, for S. commercialis in the present study. It has been suggested that the observed NO3-/NO2- release from oysters is due to nitrifying bacteria in the digestive tract of the organism (Saijo & Mitamura, 1971; Boucher & Boucher-Rodoni, 1988). There could also be free living nitrifying bacteria in the water column, although there is only negligible production of NO3- in the oyster and macroalgal control tanks.

The apparent enhancement of nitrification by the oysters resulted in more NO3- being available for denitrification. By increasing the proportion of NO3--N relative to NH4+-N, bivalves such as oysters can enhance nitrification / denitrification coupling, promoting greater removal of N from the effluent. Rates of denitrification in mussel beds have been observed to be 21% greater than bare sediment (Kasper et al., 1985), and combined with the mussel harvest, this represented a 68% greater loss of nitrogen than bare sediment.

Bivalves are known to significantly increase the rates of carbon deposition (Doering & Oviatt, 1986; Doering et al., 1986) and the rates of benthic flux of DIN (Doering et al., 1987). Consequently, the sediment NH4+ and organic N pools under bivalves are significantly higher than those in bare sediments (Kasper et al., 1985). Bivalves enhance

COMBINED TREATMENT USING S EDIMENTATION, O YSTERS AND MACROALGAE

149

movement of organic nitrogen to the sediments, where it decomposes. This decomposition in the aerobic surficial sediments yields NH4+ (remineralisation) and NO3-/NO2- (nitrification) and, in the deeper anaerobic sediments, NO3- is converted into N2 gas (denitrification) (Dame, 1996). Stimulation of nitrification by oysters can result in more NO3- present in the water, and less NH4+ (Boucher & Boucher-Rodoni, 1988).

By stimulating nitrification, the oysters significantly reduced the NH4+: NO3-/NO2- ratio from 13: 1 to 4: 1. This has significant management implications, especially in relation to the possibilities for recycling of treatment pond effluent back into the production ponds. Aeration and uptake by phytoplankton and bacteria in the macroalgal control tanks reduced the NH4+: NO3- ratio to 0.5: 1. Although the ratio of NH4+: NO3- was lower in the control, the total concentration of DIN was significantly lower in the macroalgal treatment.

5.4.5 Conclusions Combining the oysters and macroalgae in the same treatment pond may produce greater improvements in water quality. Rapid uptake of the nutrients resulting from remineralisation of oyster biodeposits (Kelly & Nixon, 1984), may reduce potential stimulation of phytoplankton production. There may be a balance between nutrient enhanced phytoplankton productivity and the removal of phytoplankton biomass by oysters (Zeitzschel, 1980).

For co-culture of macroalgae and oysters to be successful, the growth requirements of each species need to be optimised. In particular, if the temperature, water flow rate, or light availability are not adequate for macroalgal growth there may be more biomass decaying than being produced. This would result in a decline in water quality and consequently lower oyster growth rates (Qian et al., 1996).

CHAPTER 5

150

In order to maximise the volume of effluent that this polyculture system can successfully treat, the maximum final concentration of the various water quality parameters needs to be determined. Reductions in all parameters were observed to be non linear over 24 h, and therefore the amount of treatment time would need to be adjusted to maximise the volume of effluent that can be treated, without sacrificing the required water quality. Using this combination of polyculture, it was estimated that up to 18 kg N ha -1 d-1 and 15 kg P ha-1 d-1 could be removed.

The results from the present study indicate the ability of sedimentation, oysters and macroalgae to substantially improve the quality of shrimp pond effluent being released into the receiving waters (Plate 5.2). It may also make it possible to recirculate the effluent back into the production ponds, thereby reducing the need for water exchange.

Plate 5.2 Water samples collected: a) before sedimentation; b) after sedimentation; and c) after biofiltration.

CHAPTER 6 CONCLUSION

The impacts of shrimp pond effluent on the receiving waters were investigated using bioindicators (Chapter 2), and the potential for biological treatment to improve the water quality of effluent was investigated (Chapters 3, 4, & 5). It was shown that shrimp pond effluent can be effectively treated by a combination of sedimentation, oyster filtration of particulates and macroalgal absorption of dissolved nutrients (Chapter 5).

6. 1 Downstream Impacts Bioindicators proved to be useful markers for identifying the region of influence of shrimp effluent discharged into receiving waters. Results indicated that effluent nutrients may be more widespread than can be detected by traditional water quality sampling techniques (Chapter 2). Water quality analyses could detect no differences beyond 750 m from the mouths of the creeks receiving the discharges, but bioindicator responses detected effluent nutrients up to 4 km away. The amino acid composition, tissue nitrogen content and stable isotope ratio of nitrogen (15N) in seagrasses, mangroves and macroalgae were responsive to nutrient inputs from shrimp and sewage effluent discharged into two adjacent creeks. Different responses in these biological indicators revealed that the impacts of shrimp pond effluent were qualitatively different to impacts of treated sewage effluent, and were spatially more extensive than identified by water quality analyses (Chapter 2).

Effective in-pond management can significantly reduce effluent loadings of sediment and nutrients. Remediation techniques such as settlement, wetlands filtering, biological filtering with bivalves and seaweeds, and bacterial inoculation, all have potential as cost-effective

CHAPTER 6

152

treatment methods for shrimp effluent (Samocha & Lawrence, 1997). Although there are some lessons to be learnt from research into the treatment of sewage, shrimp farm effluent has different characteristics, in particular the comparatively high concentrations of suspended inorganic particulates and high salinity. This thesis presents information on some of the differences between the shrimp pond effluent and sewage effluent obtained from a comparative study in Moreton Bay. Shrimp pond effluent was shown to be high in chlorophyll a and TSS, and high in NH4+, whereas sewage effluent was relatively low in chlorophyll a, and TSS, and high in NO3- (Chapter 2).

6.2 Efficiency of Biological Filters Filtration of effluent by oysters significantly reduced the concentrations of chlorophyll a (phytoplankton), bacteria, total nitrogen, total phosphorus and total suspended solids. In particular, oyster filtration significantly reduced the concentration of small organic and inorganic particles (<5 m) which would otherwise remain in suspension. Despite reducing the concentrations of particulates and total nutrients, oyster excretion increased the concentrations of dissolved nutrients (ammonium, nitrate / nitrite, and phosphate). Macroalgae effectively reduced the concentrations of dissolved nutrients from effluent that had been pre-filtered by oysters. Macroalgae can take up dissolved organic matter such as free amino acids, and nutrients such as urea, both of which may be present in shrimp ponds (Hanisak, 1983). Red macroalgae such as Gracilaria are known to have high uptake rates and capacity for storage of excess nutrients and they contain high concentrations of commercially important substances such as agar and carrageenin (Hanisak, 1983). Macroalgae can also be used as a food supply for animals, such as the abalone Haliotis sp. (Sorgeloos & Sweetman, 1993), or livestock (Fielder et al., 1994).

CONCLUSION

153

6.3 Condition of Biofilters The results of this study demonstrated that, when cultured without prior sedimentation of particulates, the efficiency and condition of oysters and macroalgae was reduced by fouling from the high concentration of suspended particulates in the effluent. When oysters were stacked one tray deep, the highest oyster stocking density was the most effective at improving effluent water quality. However, higher stocking densities with macroalgae and oysters stacked three layers deep resulted in the death of oysters and senescence of macroalgae. This may be mitigated by faster water flow, although this would reduce the residence time and perhaps reduce the removal efficiency of the oysters. Recirculating the effluent through the oyster treatment significantly enhanced effluent water quality by increasing residence time without reducing the flow rate.

Integrated treatment incorporating sedimentation prior to biological filtration with oysters and macroalgae, improved growth and reduced signs of stress in the oysters and macroalgae. The ratio of NH4+ to NO3- / NO2- was also reduced, which has positive implications for recycling of wastewater back into shrimp production ponds, and reducing impacts on receiving waters. Settled effluent without subsequent biological filtration demonstrated very high rates of nutrient regeneration.

Despite significant reductions in total suspended solids by natural sedimentation, the concentration of particulates was probably still inhibiting the filtering efficiency of the oysters (Loosanoff & Tommers, 1948). Flocculating agents could be used to reduce settling times, although their use would probably not be cost effective (Norris 1994 in Samocha & Lawrence, 1997). However, the use of screen filters (Cripps, 1994) or the incorporation of baffles into pond design can improve the efficiency of sedimentation, and are likely to be the

CHAPTER 6

154

most economically viable and low maintenance option. The main factors affecting settling rates are particle size and water flow rate. A maximum flow of 4 m min-1, but preferably 1 m min-1 is needed to achieve optimal sedimentation performance (Henderson & Bromage, 1988).

Rates of biodeposition are related to the concentration of suspended particulates. At high food concentrations oysters produce large amounts of pseudofaeces (i.e., food filtered but not ingested) and therefore have high deposition rates, demonstrating an inefficient use of filtered food (Tenore & Dunstan, 1973). Due to the inefficient use of food and high pseudofaeces production at high food concentrations an integrated aquaculture system may be more effective if it also includes bivalves capable of existing in a high silt environment such as clams, deposit feeders such as polychaete worms and carnivores such as flounder (Tenore et al., 1973).

Oysters can potentially grow at much greater rates in shrimp pond effluent compared with natural leases. Growth from spat to market size has been achieved in as little as 4 months (Jakob et al., 1993). In order to maintain adequate improvements in water quality as well as produce a reliable crop of oysters, the stocking of a full range of oyster sizes is necessary. This thesis showed that there are significantly different filtration rates between different sized oysters and stocking densities (Chapters 3 & 4). Filtration performance was found to be positively correlated with stocking density, however, Holliday et al. (1991) found that oyster growth rates were inversely correlated to stocking density, although increasing water flow rates may help to overcome this. To maintain a continuous supply of mature, marketable oysters, an aquaculture facility would need to stock a full size range. Estimates of how many

CONCLUSION

155

of each size class and the concomitant improvements in water quality have been discussed (Wang, 1990).

6.4 Scaling up for Commercial Treatment There are several factors that may affect the performance of the proposed integrated aquaculture system when scaled up to treat the effluent from an entire shrimp farm:

a) the effects of wind on the settling rates of particles in a sedimentation pond compared to the controlled conditions in this study; b) the effects of water flow on the efficiency of the oysters and macroalgae to take up nutrients and particulates; c) the problems associated with ensuring the water is mixed sufficiently to permit optimal biofiltration; d) the potential long term effects of fouling by suspended solids and epiphytes on the condition and efficiency of the biofilters.

The use of a sedimentation pond with high walls may reduce the effects of wind on stirring the water column. Also a smaller, deeper pond may produce a larger percentage of undisturbed water and provide less fetch for wind disturbance, although the distance that the particles have to settle becomes greater. Current velocities in the settling ponds should not exceed 2-4 cm s-1 to prevent resuspension (Warrer-Hansen, 1982).

In the oyster treatment pond, paddlewheels may be ideal to simulate the high recirculating flow rates provided by bubble aeration in the present study. High flow rates, good mixing and aeration are important to reduce the problems of fouling and the buildup of metabolites

CHAPTER 6

156

(Thielker, 1981). These characteristics are also required to reduce boundary layers to facilitate efficient nutrient uptake by the macroalgae (Wheeler, 1980). Oysters and macroalgae must remain close to the surface to avoid fouling, and sufficient mixing is required to ensure the biofilters are exposed to all of the effluent in the pond.

Incorporation of sedimentation and biofilters in the one pond may be problematic because of differences in the optimal depth for biofilter efficiency versus sedimentation efficiency. Sedimentation could be separated from the biofilters to enable slow, undisturbed (laminar) flow for sedimentation, and faster mixed flow for the biofilters. The culture of oysters and macroalgae in separate ponds (oysters first, and then macroalgae) may help to improve the condition and efficiency of the macroalgae because of the reduction in suspended solids effected by the oysters. However, co-culture in the one pond may be more efficient, as Qian et al. (1996) demonstrated that co-culture of bivalves and macroalgae results in higher growth rates than monospecific cultures. Improvements in bivalve growth could be due to removal of metabolites by the macroalgae, while the higher growth rates of macroalgae may be a result of increased nutrient availability from excretion by the bivalves.

6.5 Other Potential Biofilters Bivalves other than oysters have a potential application as biofilters. Mussels may be more effective than oysters due to their higher filtration efficiency (Jrgensen, 1966; Tenore & Dunstan, 1973), although oysters have a higher assimilative capacity (Tenore et al., 1973). Clams could also be better because of their ability to survive burial in high concentrations of silt, although their filtration rates are the lowest of the three bivalves.

CONCLUSION

157

There may also be potential benefits in using macroalgae other than Gracilaria. Despite the success of using high value species of red macroalga such as Gracilaria sp., there may be certain species of macroalgae which are better able to tolerate the high suspended solid load in shrimp farm effluent. Enteromorpha sp. (a filamentous green macroalga) bloomed prolifically in the experimental raceways during non sampling times, and was observed to have very fast growth rates, with no signs of fouling.

6.6 Management Implications and Potential Problems with Biofiltration / Polyculture Although the treatment of shrimp pond effluent through the integration of natural sedimentation, oyster filtration and macroalgal absorption has been shown to be successful (Chapter 5), there are still potential problems associated with very high suspended solid loads (Chapter 4). The concentration of suspended solids in the pond effluent can be affected by the concentration in the influent water, the pond soil type, pond shape, feed management and aerator design and implementation. Reducing the concentration of suspended solids may be best accomplished by changes to in-pond management practices.

Due to the origins of shrimp feed pellets (ground shellfish, crustaceans, and fish), they often contain relatively high concentrations of heavy metals due to bioaccumulation, resulting in elevated levels in the shrimp farm effluent. There has been some concern that shrimp pond effluent may result in elevated concentrations of heavy metals in oysters and macroalgae used as biofilters, thereby making them unsafe for human consumption. However, oysters cultured in shrimp pond effluent were found to be free of human pathogens and fit for human consumption (Shpigel et al., 1993b). Further studies are needed to examine the effects of shrimp pond effluent on the quality of oysters in relation to national and international food standards.

CHAPTER 6

158

6.7 Benefits of Polyculture or Integrated Aquaculture Intensive shrimp farming without water exchange may be possible if feed management practices are modified so that small portions of feed are delivered at frequent intervals (Hopkins et al., 1995a), and biofilters such as oysters and macroalgae are incorporated to reduce the loading of particulates and dissolved nutrients. These techniques not only reduce the impacts of the effluent on the environment, but they can also have considerable advantages for maintaining in-pond water quality and preventing the transfer of disease from contaminated water pumped in after water exchange.

In addition to improving the effluent water quality from shrimp ponds, the culturing of oysters and macroalgae in this nutrient rich wastewater can lead to rapid growth rates in these commercially valuable species. The culture of oysters and macroalgae are traditionally conducted using extensive management practices, in comparison to shrimp, fish and other intensively farmed aquaculture species. Maintaining food supply and nutrients for intensive culture of both oysters and macroalgae can be costly and, it has been suggested that for commercial success, would require development of polyculture with other crops (Neish, 1979).

Polyculture is a more ecologically sound method of aquaculture (Mackay & Lodge, 1983) with an efficient use of resources, and a high resilience to environmental fluctuation (Chien & Liao, 1995). In addition to meeting environmental standards to reduce effluent loads of nutrients and total suspended solids, shrimp farmers may potentially benefit economically from the production of biofilter crops.

CONCLUSION

159

The concept of polyculture is regarded by many as having potential benefits, environmentally and economically, but the ideas have not been widely accepted (Wang & Jakob, 1991). Sedimentation ponds are ideal as primary treatment, with oysters and macroalgae functioning as secondary and tertiary treatment. Possibly the main factor inhibiting the uptake of this type of polyculture is the problem associated with culture and maintenance of these organisms in the pond environment (Chien & Liao, 1995). However, in addition to significant water quality improvements, high yields of oysters and macroalgae can be achieved (Shpigel et al., 1993b).

6.8 Future Research The results of this study demonstrated that natural abundance of stable isotope ratios of nitrogen were useful indicators for detecting the range of influence of pond effluent nutrients in the discharge environment. Analysis of the natural abundance of 13C of seagrass, mangroves and macroalgae could also be used to determine whether the C was from decomposing mangrove leaves or excreted from phytoplankton (Primavera, 1996). Differences in the 13C have also been correlated to differences in phytoplankton species composition (Fogel et al., 1992), which may make it possible to trace the 13C signature back to a nutrient source with a specific phytoplankton composition. The combined analysis of two elements also improves the ability to resolve sources with similar isotopic signatures (Ziemann et al.1984).

To better identify the region of potential impacts from shrimp farm effluent, enriched stable isotope tracer techniques (e.g., 15N labelled NH4+ released in the farms discharge canal) could be used to quantify effluent nutrient uptake by biota in the receiving waters. Controlled

CHAPTER 6

160

laboratory experiments are also needed in order to discern the relationships between shrimp and sewage effluent and the effects on amino composition, 15N and 13C.

Investigations into the use of other species of bivalves and macroalgae, as well as incorporation of sediment fauna such as polychaetes could be conducted. The optimal combination will probably relate to effluent composition, commercial viability and ease and cost of maintenance. Certain species of bivalves and macroalgae could potentially improve the quality of effluent from freshwater aquaculture. The use of duckweed (Lemnaceae) has already been trialed and found to be successful, with very high growth on effluent from carp and tilapia. It was found to significantly reduce dissolved nutrient concentrations and was also used as feed for the cultured fish (Skillicorn et al., 1998).

Application of the integrated treatment system (Chapter 5) needs to be tested on a pond scale to determine the optimal combination of biofilter density, pond depth, aeration and water flow rate to maximise water quality improvements while maintaining good condition and growth of the biofilters. These are likely to be a compromise and the optimal setup will be dependant on the effluent composition, and the species of bivalves and macroalgae used.

6.9 Summary This thesis has developed techniques for identifying the influence of shrimp pond effluent on receiving waters using seagrass, macroalgae and mangroves as biological indicators. Biological indicator responses indicated that the impacts of aquaculture effluent were qualitatively different to sewage effluent, and were spatially more extensive than identified by water quality analyses. To reduce these impacts, effluent treatment techniques incorporating biological filters were investigated.

CONCLUSION

161

Filtration by oysters significantly reduced the concentrations of particulates, but increased the concentrations of NH4+, NO3-, and PO43-. Nutrient uptake by macroalgal significantly reduced the dissolved nutrient concentrations, in particular the concentration of NH4+. Differences in oyster size, stocking density, and water flow regime had significant impacts on the condition of the oysters and macroalgae as well as their ability to improve the quality of the shrimp pond effluent. The efficiency and condition of the oysters and macroalgae was reduced by high suspended particulate loads in the effluent, however, an integrated system incorporating natural sedimentation prior to biological filtration proved effective at optimising oyster and macroalgal performance.

The integrated system effected significant improvements in the water quality of effluent being released from shrimp ponds. These improvements may be sufficient to enable recirculation of effluent back into the shrimp production ponds creating a more controlled system with minimal environmental impacts (Fig. 6.1). Filtration and absorption by various marine organisms can be effective for monitoring and reducing the environmental impacts of aquaculture effluent.

CHAPTER 6

162

Figure 6.1 Diagrammatic design of water flow for typical untreated shrimp farms (left) and a design to incorporate physical (sedimentation) and biological (oyster filtration and macroalgal absorption) treatment (right).

BIBLIOGRAPHY

Abal, E. G., Loneragan, N., Bowen, P., Perry, C.J., Udy, J.W. & Dennison, W.C. 1994. Physiological and morphological responses of Zostera capricorni Aschers. to light intensity. Journal of Experimental Marine Biology and Ecology 178, 113-129. Abal, E. G. & Dennison, W. C. 1996. Seagrass depth range and water quality in southern Moreton Bay, Queensland, Australia. Marine and Freshwater Research 47, 763-771. Abal, E. G., Holloway, K. M. & Dennison, W. C. (eds.) 1998. Interim Stage 2 Scientific Report. Brisbane River and Moreton Bay Wastewater Management Study, Brisbane. Ahel, M., Barlow, R.G. & Mantoura, R.F.C. 1996. Effect of salinity gradients on the distribution of phytoplankton pigments in a stratified estuary. Marine Ecology Progress Series 143, 289-295. Aitken, D. 1990. Shrimp farming in Ecuador. An aquaculture success story. World Aquaculture 21, 7-16. Alamoudi, O. A. 1988. The growth of Tetraselmis striata Butcher in inorganic nitrogen free ASP medium supplemented with amino acids source. Egyptian Journal of Botany 31, 13-22. Alamoudi, O. A. 1994. Species-differences in nitrogen storage reservoirs in macroalgae. Microbios 77, 239-246. Ali, R. M. 1970. The influence of suspension density and temperature on the filtration rate of Hiatella arctica. Marine Biology 6, 291-302. Allan, G. L., Moriarty, D. J. W. & Maguire, G. B. 1995. Effects of pond preparation and feeding rate on production of Penaeus monodon Fabricus, water quality, bacteria and benthos in model farming ponds. Aquaculture 130, 329-349.

BIBLIOGRAPHY

164

Angell, C. L. 1986. The biology and culture of tropical oysters. ICLARM, Manila. Asmus, H., Asmus, R. M. & Zubillaga, G. F. 1995. Do mussel beds intensify the phosphorus exchange between sediment and tidal waters? Ophelia 26, 1-18. Barnes, R. D. 1994. Invertebrate Zoology. Saunders College Publishing, Philadelphia. Bergheim, A. & Selmer-Olsen, A. R. 1982. Estimated pollution loadings from Norwegian fish farms. I. Investigations 1978-79. Aquaculture 28, 347-361. Bird, K. T., Hanisak, M. D. & Ryther, J. 1981. Chemical quality and production of agars extracted from Gracilaria tikvahiae grown in different nitrogen enrichment conditions. Botanica Marina 24, 441-444. Bird, N. L., Chen, C.-M. & McLachlan, J. 1979. Effects of temperature, light and salinity on growth in culture of Chondrus crispus, Furcellaria lumbricalis, Gracilaria tikvahiae (Gigartinales, Rhodophyta), and Fucus serrratus (Fucales, Phaeophyta). Botanica Marina 22, 521-527. Blackburn, T. H., Lund, B. A. & Krom, M. D. 1988. C- and N-mineralization in the sediments of earthen marine fishponds. Marine Ecology Progress Series 44, 221-227. Boucher, G. & Boucher-Rodoni, R. 1988. In situ measurement of respiratory metabolism and nitrogen fluxes at the interface of oyster beds. Marine Ecology Progress Series 44, 229-238. Boyd, C. E. 1992. Shrimp pond bottom soil and sediment management. In Proceedings of the Special Session on Shrimp Farming (Wyban, J., ed). World Aquaculture Society, Baton Rouge, LA., U.S.A. p. 166-181. Boyd, C. E. & Musig, Y. 1992. Shrimp pond effluents: Observations of the nature of the problem on commercial farms. In Proceedings of the Special Session on Shrimp Farming (Wyban, J. A., ed). World Aquaculture Society, Baton Rouge, LA., U.S.A. p. 195-197.

BIBLIOGRAPHY

165

Braaten, B. 1991. Impact of pollution from aquaculture in six Nordic countries. Release of nutrients, effects, and waste water treatment. In Aquaculture and the Environment. (De Pauw, N. & Joyce, J., eds), European Aquaculture Society Special Publication No. 16, Fishing News Books, Oxford, England. 536 pp. Briggs, M. R. P. & Funge-Smith, S. J. 1993. Macroalgae in aquaculture: an overview and their possible roles in shrimp culture. In Proceedings of the Conference on Marine Biotechnology in the Asia Pacific Region. Bangkok, Thailand, 16-20 November 1993. p. 45-53. Briggs, M. R. P. & Funge-Smith, S. J. 1994. A nutrient budget of some intensive marine shrimp ponds in Thailand. Aquaculture and Fisheries Management 25, 789-811. Briggs, M. R. P. & Funge-Smith, S. J. in review. The potential of Gracilaria spp. meal for supplementation of diets for juvenile Penaeus monodon Fabricus. Aquaculture and Fisheries Management . Brown, J. R. & Hartwick, E. B. 1988. Influences of temperature, salinity and available food upon suspended culture of the Pacific oyster, Crassostrea gigas: I. Absolute and allometric growth. Aquaculture 70, 231-252. Burdin, K. S. & Bird, K. T. 1994. Heavy metal accumulation by carrageenan and agar producing algae. Botanica Marina 37, 467-470. Burford, M. 1997. Phytoplankton dynamics in shrimp ponds. Aquaculture Research 28, 351360. Chandrkrachang, S., Chinadit, U., Chandyot, P. & Supasari, T. 1991. Profitable spin-offs from shrimp-seaweed polyculture. INFOFISH International 6/91, 26-28. Chen, J. C., Cheng, S. Y. & Chen, C. T. 1994. Changes of haemocyanin, protein and free amino acid levels in the haemolymph of Penaeus japonicus exposed to ambient

BIBLIOGRAPHY

166

ammonia. Comparative Biochemistry and Physiology A - Comparative Physiology 109, 339-347. Chien, Y. H. 1992. Water quality requirements and management for marine shrimp culture. In Proceedings of the Special Session on Shrimp Farming (Wyban, J., ed). World Aquaculture Society, Baton Rouge, LA., U.S.A. p. 144-156. Chien, Y. H. & Liao, I. C. 1995. Integrated approach to shrimp growout system design. In Swimming through troubled waters, Proceedings of the special session on shrimp farming, Aquaculture '95. (Browdy, C. L. & Hopkins, J. S., eds). World Aquaculture Society, Baton Rouge, LA, U.S.A. p. 167-182. Chien, Y. H. & Tsai, W. S. 1985. Integrated pond culture: a type of spatially sequential polyculture. Journal of the World Mariculture Society 16, 429-436. Chua, T. E., Paw, J. N. & Guarin, F. Y. 1989. The environmental impact of aquaculture and the effects of pollution on coastal aquaculture development in southeast Asia. Marine Pollution Bulletin 20, 335-343. Cifuentes, L. A., Coffin, R. B., Solorzano, L., Cardenas, W., Espinoza, J. & Twilley, R. R. 1996. Isotopic and elemental variations of carbon and nitrogen in a mangrove estuary. Estuarine, Coastal and Shelf Science 43, 781-800. Cifuentes, L. A., Fogel, M. L., Pennock, J. R. & Sharp, J. H. 1989. Seasonal variations in the stable isotope ratio of ammonium in the Delaware estuary. Geochim Cosmochim Acta 53, 2713-2721. Clesceri, L. S., Greenberg, A. E. & Trussel, R. R. 1989. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington. 1134 pp. Cohen, I. & Neori, A. 1991. Ulva lactuca biofilters for marine fishpond effluent. I. Ammonia uptake kinetics and nitrogen content. Botanica Marina 34, 475-482.

BIBLIOGRAPHY

167

Costanzo, S. 1996. Marine macroalgae (Rhodophyta) as a biological indicator of pulsed nutrients in oligotrophic environments. Honours Dissertation, The University of Queensland, Brisbane. 43 pp. Cripps, S. J. 1992. Characterization of an aquaculture effluent based on water quality and particle size distribution data. In Aquaculture 92 Growing Toward the 21st Century. p. 72. Cripps, S. J. 1994. Minimizing output: Treatment. J. Appl. Ichthyol. Z. Angew. Ichthyol. 10, 284-294. Cunningham, A., Levavasseur, G., Estrada, M., Hanelt, D. & Wilhelm, C. 1996. Technical discussion III: Fluorescence measurements. Scientia Marina 60, 301-302. Dame, R. F. 1996. Ecology of Marine Bivalves: an ecosystem approach. CRC Press, New York. 254 pp. Dame, R. F., Spurrier, J. D. & Wolaver, T. G. 1989. Carbon, nitrogen and phosphorus processing by an oyster reef. Marine Ecology Progress Series 54, 249-256. D'Elia, C. & DeBoer, J. 1978. Nutritional studies of two red algae. II. Kinetics of ammonia and nitrate uptake. Journal of Phycology 14, 266-272. Dennison, W. C., O'Donohue, M. J. & Abal, E. G. 1993. Water Quality in the Logan River and Southern Moreton Bay, Queensland A Report on the Results of a 1 Year Monitoring Program Conducted by the Marine Botany Section, Department of Botany, University of Queensland. 65 pp. Di Martino Rigano, V., DeMartino, C., Vona, V., Esposito, S. & Rigano, C. 1992. Amino acid pools in nutrient limited Cyanidium caldarium and responses to resupply. Phytochemistry 31, 1911-1916.

BIBLIOGRAPHY

168

Djangmah, J.S., Shumway, S.E. & Davenport, J. 1979. Effects of fluctuating salinity on the behaviour of the West African blood clam Anadara senilis and on the osmotic pressure and ionic concentrations of the haemolymph. Marine Biology. 50, 209-213. Doering, P. H., Kelly, J. R., Oviatt, C. A. & Sowers, T. 1987. Effect of the hard clam Mercenaria mercenaria on the benthic fluxes of inorganic nutrients and gases. Marine Biology 94, 377-383. Doering, P. H. & Oviatt, C. A. 1986. Application of filtration rate models to field populations of bivalves: an assessment using experimental mesocosms. Marine Ecology Progress Series 31, 265-275. Doering, P. H., Oviatt, C. A. & Kelly, J. R. 1986. The effects of the filter feeding clam Mercenaria mercenaria on carbon cycling in experimental mesocosms. Journal of Marine Research 44, 839-861. Duarte, C. M. 1990. Seagrass Nutrient Content. Marine Ecology Progress Series 67, 201207. Eng, C. T., Paw, J. N. & Guarin, F. Y. 1989. The environmental impact of aquaculture and the effects of pollution on coastal aquaculture development in southeast Asia. Marine Pollution Bulletin 20, 335-343. Engledow, H. R. & Bolton, J. J. 1992. Environmental tolerances in culture and agar content of Gracilaria verrucosa (Hudson) Papenfuss (Rhodophyta, Gigartinales) from Saldanha Bay. South African Journal of Botany 58, 263-267. Falkowski, P. G. 1983. Enzymology of nitrogen assimilation. In Nitrogen in the Marine Environment (Carpenter, E. J. & Capone, D. G., eds). Academic Press Inc., New York, p. 839-868. Fielder, D., Johnson, I. & Warburton, K. 1994. Innovations in Aquaculture: Aquaculture Practices. Queensland Department of Education, Brisbane. 16 pp.

BIBLIOGRAPHY

169

Fleming, A. E. 1995. Digestive efficiency of the Australian abalone Haliotis rubra in relation to growth and feed preference. Aquaculture 134, 279-293. Flynn, K. J., Fasham, M. J. R. & Hipkin, C. R. 1997. Modelling the interactions between ammonium and nitrate uptake in marine phytoplankton. Philosophical Transactions of the Royal Society of London Series B Biological Sciences 352, 1625-1645. Fogel, M. L. & Cifuentes, L. A. 1993. Isotopic fractionation during primary production. In Organic Geochemistry (Engel, M. & Macko, S. A., eds). Plenum Press, New York, p. 73-98. Fogel, M.L., Cifuentes, L.A., Velinsky, D.J. & Sharp, J.H. 1992. Relationship of carbon availability in estuarine phytoplankton to isotopic composition. Marine Ecology Progress Series 82, 291-300. Frid, C. L. J. & Mercer, T. S. 1989. Environmental monitoring of caged fish farming in macrotidal environments. Marine Pollution Bulletin 20, 379-383. Friedlander, M. & Dawes, C. J. 1985. In situ uptake kinetics of ammonium and phosphate and chemical composition of the red seaweed Gracilaria tikvahiae. Journal of Phycology 21, 448-453. Friedlander, M. & Levy, I. 1995. Cultivation of Gracilaria in outdoor tanks and ponds. Journal of Applied Phycology 7, 315-324. Funge-Smith, S. J. & Briggs, M. R. P. 1998. Nutrient budgets in intensive shrimp ponds: implications for sustainability. Aquaculture 164, 117-133. Funge-Smith, S. J. & Briggs, M. R. P. in prep. Growth and ammonia removal by Gracilaria spp. in brackishwater and intensive shrimp farm effluents. Gerloff, G. C. & Krombholz, P. H. 1966. Tissue analysis as a measure of nutrient availability for the growth of angiosperm aquatic plants. Limnology and Oceanography 11, 529537.

BIBLIOGRAPHY

170

Gowen, R. J. & Rosenthal, H. 1993. The environmental consequences of intensive coastal aquaculture in developed countries: What lessons can be learned. In Environment and aquaculture in developing countries (Pullin, R. S. V., Rosenthal, H. & Maclean, J. L., eds). ICLARM Conference Proceedings, 31. p. 102-115. Gowen, R. J., Rosenthal, H., Maekinen, T. & Ezzi, I. 1990. Environmental impact of aquaculture activities. In Business Joins Science (Depauw, N., ed). Spec. Publ. Eur. Aquacult. Soc. no. 12. p. 257-283. Grant, J., Hatcher, A., Scott, D. B., Pocklington, P., Schafer, C. T. & Winters, G. V. 1995. A multidisciplinary approach to evaluating impacts of shellfish aquaculture on benthic communities. Estuaries 18, 124-144. Grice, A. M., Loneragan, N. R. & Dennison, W. C. 1996. Light intensity and the interactions between physiology, morphology and stable isotope ratios in five species of seagrass. Journal of Experimental Marine Biology and Ecology 195, 91-110. Hader, D. P., Lebert, M., Mercado, J., Aguilera, J., Salles, S., FloresMoya, A., Jimenez, C. & Figueroa, F. L. 1996. Photosynthetic oxygen production and PAM fluorescence in the brown alga Padina pavonica measured in the field under solar radiation. Marine Biology 127, 61-66. Haglund, K. & Pedersen, M. 1993. Outdoor pond cultivation of the subtropical marine red alga Gracilaria tenuistipitata in brackish water in Sweden. Growth, nutrient uptake, co-cultivation with rainbow trout and epiphyte control. Journal of Applied Phycology, 271-284. Haines, K. C. 1975. Growth of the carrageenin-producing tropical red seaweed Hypnea musciformis in surface water, 870 m deep water, effluent from a clam mariculture system, and in deep water enriched with artificial fertilisers or domestic sewage. In

BIBLIOGRAPHY

171

Proc. 10th European Symposium on Marine Biology (Ostend, Belgium) (Persoone, G. & Jaspers, E., eds). Universa Press, Wetteren. p. 207-220. Haines, K. C. & Wheeler, P. A. 1978. Ammonium and nitrate uptake by the marine macrophytes Hypnea musciformis (Rhodophyta) and Macrocystis pyrifera (Phaeophyta). Journal of Phycology 14, 319-324. Hammen, C. S., Miller, H. F. & Geer, W. H. 1966. Nitrogen excretion of Crassostrea virginica. Comp. Biochem. Physiol. 17, 1199-1200. Handley, L.L. & Raven, J.A. 1992. The use of natural abundance of nitrogen isotopes in plant physiology and ecology. Plant, Cell and Environment 15, 965-985. Hanelt, D., Li, J. & Nultsch, W. 1994. Tidal Dependence of Photoinhibition of Photosynthesis in Marine Macrophytes of the South China Sea. Botanica Acta 107, 66-72. Hanisak, M. D. 1979. Growth patterns of Codium fragile sp. tomentosoides in response to temperature, irradiance, salinity and nitrogen source. Marine Biology (Berlin) 50, 319-332. Hanisak, M. D. 1983. The Nitrogen Relationships of Marine Macroalgae. In Nitrogen in the Marine Environment (Carpenter, E. J. & Capone, D. G., eds). Academic Press Inc., New York. 900 pp. Hanisak, M. D. & Harlin, M. M. 1978. Uptake of inorganic nitrogen by Codium fragile subsp. tomentosoides (Chlorophyta). Journal of Phycology 14, 450-454. Hargreaves, J. A. 1998. Nitrogen biochemistry of aquaculture ponds. Aquaculture 166, 181212. Harlin, M. M. 1978. Nitrate uptake by Enteromorpha spp. (Chlorophyceae): Applications to aquaculture systems. Aquaculture 15, 373-376.

BIBLIOGRAPHY

172

Harlin, M. M., Thorne-Miller, B. & Thursby, G. B. 1979. Ammonium uptake by Gracilaria sp. (Florideophyceae) and Ulva lactuca (Chlorophyceae) in closed system fish culture. In Proc. Int. Seaweed Symp. p. 285-292. Haven, D. & Morales-Alamo, R. 1970. Filtration of particles from suspension by the American oyster Crassostrea virginica. Biol. Bull. Mar. Biol. Lab. Woods Hole 139, 248-264. Henderson, J. P. & Bromage, N. R. 1988. Optimising the removal of suspended solids from aquacultural effluents in settlement lakes. Aquacultural Engineering 7, 167-182. Hensey, M. P. 1991. Environmental monitoring for fish farms in Ireland. In Aquaculture and the Environment. (De Pauw, N. & Joyce, J., eds). Special Publication of the European Aquaculture Society No. 16. Oxford, England. 536 pp. Herrmann, H., Ghetti, F., Scheuerlein, R. & Hader, D. P. 1995. Photosynthetic oxygen and fluorescence measurements in Ulva laetevirens affected by solar irradiation. Journal of Plant Physiology 145, 221-227. Heuer, B. & Feigin, A. 1993. Interactive effects of chloride and nitrate on photosynthesis and related growth parameters in tomatoes. Photosynthetica 28, 549-554. Hicks, B.J. & Silvester, W. B. 1985. Nitrogen fixation associated with the New Zealand mangrove (Avicennia marina var. resinifera). Applied and Environmental Microbiology 49, 955-959. Hobbie, J. E. 1977. Use of Nucleopore filters for counting bacteria by fluorescence microscopy. Applied Environmental Microbiology 33, 1225-1228. Hoch, M. P., Fogel, M. L. & Kirchman, D. L. 1994. Isotope fractionation during ammonium uptake by marine microbial assemblages. Geomicrobiology Journal 12, 113-127. Holliday, J. E., Maguire, G. B. & Nell, J. A. 1991. Optimum stocking density for nursery

culture of Sydney rock oysters (Saccostrea commercialis). Aquaculture 96, 7-16.

BIBLIOGRAPHY

173

Holmer, M. 1991. Impacts of aquaculture on surrounding sediments: generation of organicrich sediments. In Aquaculture and the Environment. (De Pauw, N. & Joyce, J., eds). Special Publication of the European Aquaculture Society No. 16. Oxford, England. 536 pp. Hopkins, J. S., Browdy, C. L., Hamilton II, R. D. & Heffernan III, J. A. 1995a. The effect of low-rate sand filtration and modified feed management on effluent quality, pond water quality and production of intensive shrimp ponds. Estuaries 18, 116-123. Hopkins, J. S., Hamilton II, R. D., Sandifer, P. A. & Browdy, C. L. 1993a. The production of bivalve mollusks in intensive shrimp ponds and their effect on shrimp production and water quality. World Aquaculture 24, 74-77. Hopkins, J. S., Hamilton, R. D., II, Sandifer, P. A., Browdy, C. L. & Stokes, A. D. 1993b. Effect of water exchange rate on production, water quality, effluent characteristics and nitrogen budgets of intensive shrimp ponds. Journal of the World Aquaculture Society 24, 304-320. Hopkins, J. S., Sandifer, P. A. & Browdy, C. L. 1994. Sludge management in intensive pond culture of shrimp: Effect of management regime on water quality, sludge characteristics, nitrogen extinction, and shrimp production. Aquacultural Engineering 13, 11-30. Hopkins, J. S., Sandifer, P. A. & Browdy, C. L. 1995b. A review of water management regimes which abate the environmental impacts of shrimp farming. In World Aquaculture Society (Browdy, C. L. & Hopkins, J. S., eds). World Aquaculture Society, Baton Rouge, Louisiana, U.S.A. p. 157-166. Horrocks, J., Stewart, G. R. & Dennison, W. C. 1995. Tissue nutrient content of Gracilaria spp. (Rhodophyta) and water quality along an estuarine gradient. Marine and Freshwater Research 46, 975-983.

BIBLIOGRAPHY

174

Horst, M. N. 1989. Association between chitin synthesis and protein synthesis in the shrimp Penaeus vannamei. Journal of Crustacean Biology 9, 257-265. Huguenin, J. E. 1976. An examination of problems and potentials for future large-scale intensive seaweed culture systems. Aquaculture 9, 313-342. Huntington, K. M. & Miller, D. C. 1989. Effects of suspended sediment, hypoxia, and hyperoxia on larval Mercenaria mercenaria (L.). Journal of Shellfish Research 8, 3742. Iizumi, H. & Hattori, A. 1982. Growth and organic production of eelgrass (Zostera marina L.) in temperate waters of the Pacific coast of Japan. III. The kinetics of nitrogen uptake. Aquatic Botany 12, 245-256. Indergaard, M. & Rueness, J. 1990. Marine macroalgae: Traditional and present uses. Blyttia 48, 53-56. Jakob, G. S., Pruder, G. D. & Wang, J. K. 1993. Growth trial with the American oyster Crassostrea virginica using shrimp pond water as feed. Journal of the World Aquaculture Society 24, 344-351. Jones, A. B. & Dennison, W. C. 1998. Photosynthetic capacity in coral reef systems: Investigations into ecological applications for the underwater PAM fluorometer. In Proceedings of the Australian Coral Reef Society 75th Anniversary Conference (Greenwood, J. G. & Hall, N. J., eds). School of Marine Science, The University of Queensland, Brisbane, Heron Island October 1997. p. 67-79. Jones, A. B., Dennison, W. C. & Preston, N. P. in review. Improvements in water quality of aquaculture effluent after treatment by sedimentation, oyster filtration and macroalgal absorption. Aquaculture. Jones, A. B., Dudley, B. J. & Dennison, W. C. 1998. Factors limiting phytoplankton biomass in the Brisbane River and Moreton Bay. In Moreton Bay and Catchment (Tibbets, I.

BIBLIOGRAPHY

175

R., Hall, N. J. & Dennison, W. C., eds). School of Marine Science, The University of Queensland, Brisbane. p. 179-186. Jones, A. B., O'Donohue, M. J. & Dennison, W. C. in prep a. Assessing ecological impacts of shrimp and sewage effluent: Biological indicators with standard water quality and sediment analyses. Estuarine, Coastal and Shelf Science. Jones, A. B. & Preston, N. P. 1999. Oyster filtration of shrimp farm effluent, the effects on water quality. Aquaculture Research 30, 51-57. Jones, A. B., Preston, N. P. & Dennison, W. C. in prep b. The efficiency and condition of oysters and macroalgae used as biological filters of shrimp pond effluent. Aquaculture Research. Jones, A. B., Stewart, G. R. & Dennison, W. C. 1996. Macroalgal responses to nitrogen source and availability: amino acid metabolic profiling as a bioindicator using Gracilaria edulis (Rhodophyta). Journal of Phycology 32, 757-766. Jrgensen, G. B. 1966. Biology of suspension feeding. Pergammon Press, New York. 357 pp. Jrgensen, N. O. G., 1982. Heterotrophic Assimilation and Occurrence of Dissolved Free Amino Acids in a Shallow Estuary. Marine Ecology Progress Series 8, 145-159. Kamiyama, T., Tamai, K. & Tsujino, M. 1997. Effects of sediment conditions and dissolved oxygen concentration of bottom seawater on nutrient regeneration processes from marine sediments. Bull. Nansei Natl. Fish. Res. Inst. Nanseisuikenho 30, 209-218. Kasper, H. F., Gillespie, P. A., Boyer, I. C. & MacKenzie, A. L. 1985. Effects of mussel aquaculture on the nitrogen cycle and benthic communities in Kenepuru Sound, Marlborough Sounds, New Zealand. Marine Biology 85, 127-136. Kautsky, N. & Evans, S. 1987. Role of biodeposition by Mytilus edulis in the circulation of matter and nutrients in a Baltic coastal ecosystem. Marine Ecology Progress Series 38, 201-212.

BIBLIOGRAPHY

176

Kelly, J. R. & Nixon, S. W. 1984. Experimental studies of the effect of organic deposition on the metabolism of a coastal marine bottom community. Marine Ecology Progress Series 17, 157-169. Kiladze, T. D., Tsanava, V. P., Izmailov, S. F. 1989. Biosynthesis and distribution of nitrogen-containing compounds in tea plants fed through roots on nitrogen-15 nitrate and carbon-14 sucrose. Fiziologiya Rastenii (Moscow) 36, 1155-1163. Kinne, P.N., Browdy, C.L., Samocha, T.M., Jones, E.R. & McKee, D.A. 1997. Preliminary studies on biofiltration of intensive shrimp pond effluent. Book of Abstracts. World Aquaculture '97. Feb 19-23, Seattle, Washington, U.S.A. p. 257. Kou, Y.-Z. & Chen, J.-C. 1991. Acute toxicity of ammonia to Penaeus japonicus Bate juveniles. Aquaculture and Fisheries Management 22, 259-263. Kramer, K. J. M. (ed) 1994. Biomonitoring of coastal waters and estuaries. CRC Press, Boca Raton. 327 pp. Krom, M. D., Ellner, S., Van Rijn, J. & Neori, A. 1995. Nitrogen and phosphorus cycling and transformations in a prototype 'non-polluting' integrated mariculture system, Eilat, Israel. Marine Ecology Progress Series 118, 25-36. Lam, C. Y. & Wang, J. K. 1989. Grow-out trial of Crassostrea virginica in a tank fed shrimp pond water in Hawaii. In 1989 Annual Meeting of the World Aquaculture Society, Los Angeles, CA. p. 34. Lawlor, D. W., Oyle, F. A., Young, A. T., Kendall, A. C. & Keys, A. J. 1987. Nitrate nutrition and temperature effects on wheat: Soluble components of leaves and carbon fluxes to amino acids and sucrose. Journal of Experimental Botany 38, 1091-1103. Lee, D. L., Her, B. Y., Cheng, S. J. & Liao, I. C. 1989. Seasonal variations of extractive components of wild grass prawn, Penaeus monodon, in Tungkang (Taiwan) area. Journal of the Fisheries Society of Taiwan 16, 281-292.

BIBLIOGRAPHY

177

Lin, C. K. 1995. Progression of intensive marine shrimp culture in Thailand. In Swimming through troubled water. Proceedings of the Special Session on Shrimp Farming, Aquaculture '95. (Browdy, C. L. & Hopkins, J. S., eds). World Aquaculture Society, Baton Rouge, Louisiana, U.S.A. p. 13-23. Lin, C. K., Ruamthaveesub, P. & Wanuchsoontorn, P. 1993. Integrated culture of the green mussel (Perna viridis) in wastewaster from an intensive shrimp pond: concept and practice. World Aquaculture 24, 68-73. Loneragan, N.R., O'Donohue, M.J.H. & Dennison, W.C. in prep. 15N stable isotope ratios of seagrasses and mangroves as tracers of sewage nitrogen. Loosanoff, V. L. 1949. On the food selectivity of oysters. Science 110, 122. Loosanoff, V. L. & Tommers, F. D. 1948. Effect of suspended silt and other substances on rate of feeding of oysters. Science 107, 69-70. Lyngby, J. E. 1990. Monitoring of nutrient availability and limitation using the marine macroalgae, Ceramium rubrum (Huds.) C. Ag. Aquatic Botany 38, 153-161. Macintosh, D. J. & Phillips, M. J. 1992. Environmental issues in shrimp farming. In Shrimp '92, Proceedings of the 3rd global conference on the shrimp industry (Saram, H. & Singh, T., eds), Hong Kong, p. 118-145. Mackay, K. T. & Lodge, B. 1983. Ecological aquaculture, new approaches to aquaculture in North America. Journal of the World Mariculture Society 14, 704-713. Macko, S. A. & Ostrom, N. E. 1994. Pollution studies using stable isotopes. In Stable isotopes in ecology and environmental science (Latja, K. & Michener, R. H., eds). Blackwell Scientific Publications, Oxford, p. 45-62. Manahan, D. T., Wright, S. H., Stephens, G. C. & Rice, M. A. 1982. Transport of dissolved amino acids by the mussel, Mytilus edulis: demonstration of net uptake from natural seawater. Science 215, 1253-1255.

BIBLIOGRAPHY

178

Mann, R. & Ryther, J. H. 1977. Growth of six species of bivalve molluscs in a waste recycling-aquaculture system. Aquaculture 11, 231-245. Mariotti, A., Lancelot, C. & Billen, G. 1984. Natural isotopic composition of nitrogen as a tracer of origin for suspended organic matter in the Scheldt estuary. Geochim Cosmochim Acta 48, 549-555. McCarthy, J. J., Taylor, W. R. & Taft, J. L. 1977. Nitrogenous nutrition of the plankton in the Chesapeake Bay. 1. Nutrient availability and phytoplankton preferences. Limnology and Oceanography 22, 996-1011. McClelland, J. W. & Valiela, I. 1998. Linking nitrogen in estuarine producers to land-derived sources. Limnology and Oceanography 43, 577-585. Menzel, R. W. 1955. Some phases of the biology of Ostrea equistris Say and a comparison with Crassostrea virginica (Gmelin). Publ. Inst. Mar. Sci. Univ. Texas 4, 69-153. Mohan, B. S., Hosetti, B. B. 1998. Potential phytotoxicity of lead and cadmium to Lemna minor grown in sewage stabilization ponds. Environmental-Pollution 98, 233-238. Muir, J. F. 1982. Economic aspects of waste water treatment in fish culture. In Report of the EIFAC workshop on fish farming effluent. EIFAC Tech. Pap. (Alabaster, J. S., ed), Silkeborg, Denmark, p. 123-135. Naldi, M. & Wheeler, P. A. 1999. Changes in nitrogen pools in Ulva fenestrata (Chlorophyta) and Gracilaria pacifica (Rhodophyta) under nitrate and ammonium enrichment. Journal of Phycology 35, 70-77. Nasholm, T., Edfast, A. B., Erscsson, A. & Norden, L. G. 1994. Accumulation of amino acids in some boreal forest plants in response to increased nitrogen availability. New Phytologist 126, 137-143.

BIBLIOGRAPHY

179

Nasr, A. H., Bekheet, I. A., and Ibrahim, R. K., 1968. The Effect of Different Nitrogen and Carbon Sources on Amino Acid Synthesis in Ulva, Dictyota, and Pterocladia. Hydrobiologia 31, 7-16. Neish, I. C. 1979. Principles and perspectives of the cultivation of seaweeds in enclosed systems. In Actas del Primer Symposium sobre Algas Marinas Chilenas. (Santelices, B., ed). Subsecretara de Pesca, Ministerio de Economa, Fomento y Reconstruccon, Santiago, p. 59-74. Nell, J. A. & Gibbs, P. J. 1986. Salinity tolerance and absorption of L-methionine by some Australian bivalve molluscs. Australian Journal of Marine and Freshwater Research 37, 721-727. Neori, A., Cohen, I. & Gordin, H. 1991. Ulva lactuca biofilters for marine fishpond effluent. II. Growth rate, yield and C:N ratio. Botanica Marina 34, 483-489. Newell, R. I. E. & Jordan, S. J. 1983. Preferential ingestion of organic material by the American oyster, Crassostrea virginica. Marine Ecology Progress Series 13, 47-53. O' Connor, B. D. S., Costelloe, B. F., Keegan, B. F. & Rhoads, D. C. 1989. The use of REMOTS@ technology in monitoring coastal enrichment resulting from mariculture. Marine Pollution Bulletin 20, 384-390. Oakes, F. R. & Ponte, R. D. 1996. The abalone market: Opportunities for cultured abalone. Aquaculture 140, 187-195. O'Donohue, M. J. & Dennison, W. C. 1997. Phytoplankton productivity response to nutrient concentrations, light availability and temperature along an Australian estuarine gradient. Estuaries 20, 521-533. Ohlson, M., Nordin, A. & Nasholm, T. 1995. Accumulation of amino acids in forest plants in relation to ecological amplitude and nitrogen supply. Functional Ecology 9, 596-605.

BIBLIOGRAPHY

180

Oki, Y. & Fushimi, A. 1992. Nutrient removal potential of several aquatic weeds. In Abstracts of the Aquatic Plant Management Society, Inc. Thirty Second Annual Meeting and International Symposium on the Biology and Management of Aquatic Plants. p. 10-11. Omori, K., Hirano, T. & Takeoka, H. 1994. The limitations to organic loading on the bottom of a coastal ecosystem. Marine Pollution Bulletin 28, 73-80. Owens, N. J. P. 1987. Natural variations in 15N in the marine environment. Advances in Marine Biology 24, 411-451. Paez Osuna, F., Guerrero Galvan, S. R., Ruiz Fernandez, A. C. & Espinoza Angulo, R. 1997. Fluxes and mass balances of nutrients in a semi-intensive shrimp farm in northWestern Mexico. Marine Pollution Bulletin 34, 290-297. Parsons, T. R., Maita, Y. & Lalli, C. M. 1984. A Manual of Chemical and Biological Methods for Seawater Analysis. Pergammon Press, Oxford. 173 pp. Pearson, T. H. & Rosenberg, R. 1978. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanography & Marine Biology Annual Review 16, 229-311. Peckol, P., Demeo-Anderson, B., Rivers, J., Valiela, I., Maldonado, M. & Yates, J.T.I. 1994. Growth, nutrient uptake capacities and tissue constituents of the macroalgae Cladophora vagabunda and Gracilaria tikvahiae related to site-specific nitrogen loading rates. Marine Biology 121, 175-185. Peterson, B.J. & Fry, B. 1987. Stable isotopes in ecosystem studies. Annual Review of Ecological Systematics 18, 293-320. Phillips, M. J., Beveridge, M. C. M. & Clarke, R. M. 1991. Impact of aquaculture on water resources. In Aquaculture and Water Quality. Advances in World Aquaculture (Brune, D. E. & Tomasso, J. R., eds), Baton Rouge, Louisana, U.S.A. p. 568-591.

BIBLIOGRAPHY

181

Phillips, M. J., Lin, C. K. & Beveridge, M. C. M. 1993. Shrimp culture and the environment lessons from the world's most rapidly expanding warmwater aquaculture sector. In ICLARM Conference Proceedings (Pullin, R. S. V., Rosenthal, H. & Maclean, J. L., eds), p. 171-197. Pillay, T. V. R. (ed) 1992. Aquaculture and the Environment. Fishing News Books, Oxford, England. 189 pp. Piper, R. G., McElwain, I. B., Orme, L. E., McCraren, J. P., Fowler, L. G. & Leonard, J. R. 1982. Fish Hatchery Management. U.S. Department of Interior, Fish and Wildlife Service, Washington, D.C. 517 pp. Pomeroy, L. R. & Haskin, H. H. 1954. The uptake and utilisation of phosphate ions from sea water by the American oyster Crassostrea virginica (Gmel.). Biological Bulletin 107, 123-129. Pomeroy, L. R., Smith, E. E. & Grant, C. M. 1965. The exchange of phosphate between estuarine water and sediments. Limnology and Oceanography 10, 167-172. Prakash, A. 1989. Environmental implications of coastal aquaculture. Bulletin of the Aquacultural Association of Canada September, 109-111. Primavera, J. H. 1994. Environmental and socioeconomic effects of shrimp farming: the Philippine experience. Infofish International 1, 44-49. Primavera, J.H. 1996. Stable carbon and nitrogen isotope ratios of penaeid juveniles and primary producers in a riverine mangrove in Guimaras, Philippines. Bulletin of Marine Science 58, 675-683. Pruder, G. D. 1992. Marine shrimp pond effluent: Characterisation and environmental impacts. In Proceedings of the Special Session on Shrimp Farming. (Wyban, J. A., ed). World Aquaculture Society, Baton Rouge, Louisiana, U.S.A. p. 187-194.

BIBLIOGRAPHY

182

Qian, P.-Y., Wu, C. Y., Wu, M. & Xie, Y. K. 1996. Integrated cultivation of the red alga Kappaphycus alvarezii and the pearl oyster Pinctada martensi. Aquaculture 147, 2135. Raa, J. & Liltved, H. 1992. An assessment of the compatibility between fish farming and the Norwegian coastal environment. In Aquaculture and the Environment (Depauw, N., ed), p. 51-59. Rau, G. H., Sweeney, R. E., Kaplan, I. R., Mearns, A. J. & Young, D. R. 1981. Differences in animal 13C, 15N and D abundance between a polluted and unpolluted coastal site. Likely indicators of sewage uptake by a marine food web. Estuarine, Coastal and Shelf Science 13, 711. Raven, P. H., Evert, R. F. & Eichhorn, S. E. 1987. Biology of Plants 4th Ed. Worth Publishers Inc., New York. 775 pp. Retamales, C. A., Martinez, A. & Buschmann, A. H. 1994. Long term productivity and agar yield of Gracilaria chilensis tank culture in southern Chile. Revista de Biologia Marina, Valparafso 29, 251-261. Riise, J. C. & Roos, N. 1997. Benthic metabolism and the effects of bioturbation in a fertilised polyculture fish pond in Northeast Thailand. Aquaculture 150, 1-2. Rivers, J.S. & Peckol, P. 1995. Interactive effects of nitrogen and dissolved inorganic carbon on photosynthesis, growth, and ammonium uptake of the macroalgae Cladophora vagabunda and Gracilaria tikvahiae. Marine Biology. 121, 747-753. Robertson, A. I. & Phillips, M. J. 1995. Mangroves as filters of shrimp pond effluent: predictions and biogeochemical research needs. Hydrobiologia 295, 311-321. Roels, O. A., Haines, K. C. & Sunderlin, J. B. 1976. The potential yield of artificial upwelling mariculture. In Proc. 10th European Symposium on Marine Biology

BIBLIOGRAPHY

183

(Persoone, G. & Jaspers, E., eds). Universa Press, Wetteren, Ostend, Belgium, p. 385394. Rogers, J. 1998. Responses of mangrove forests to natural and experimental nutrient gradients in Moreton Bay, Australia. Honours Dissertation, The University of Queensland, Brisbane. 53 pp. Rosenfeld, J. K. 1979. Ammonium adsorption in nearshore anoxic sediments. Limnology and Oceanography 24, 356-364. Rowan, K. S. 1989. Photosynthetic Pigments of Algae. Cambridge University Press, London. 334 pp. Rubel and Hager Inc. 1979. Rubel and Hager, Inc., 4400 E. Broadway, Tuscon, AR. 15 pp. Ryther, J. H., Corwin, N., DeBusk, T. A. & Williams, L. D. 1981. Nitrogen uptake and storage by the red algae Gracilaria tikvahiae (McLachlan, 1979). Aquaculture 26, 107-115. Ryther, J. H., Goldman, J. C., Gifford, C. E., Huguenin, J. E., Wing, A. S., Clarner, J. P., Williams, L. D. & Lapointe, B. E. 1975. Physical models of integrated waste recycling- marine polyculture systems. Aquaculture 5, 163 - 177. Saijo, Y. & Mitamura, O. 1971. Regeneration of nutrients in the waters of a coastal oyster bed. In The ocean world (Uda, M., ed). Jpn. Soc. Promotion Sci., Tokyo, p. 242-248. Salonen, M. L., Simola, L. K. 1989. Effect of nitrate, ammonium and some amino acids on growth and nitrate reductase activity in suspension cultures of Atropa belladonna. Plant and Cell Physiology 30, 1177-1182. Samocha, T. M. & Lawrence, A. L. 1997. Shrimp farms' effluent waters, environmental impact and potential treatment methods. In Interactions between cultured species and naturally occurring species in the environment (Keller, B. J., ed), p. 33-58.

BIBLIOGRAPHY

184

Sarimento, R., Villegas, A., Mazuelos, C., Garcia, J. L. and Troncoso, A. 1992. Influence of nitrogen source and concentration on N fractions and free amino acid levels of grape vine explants. Plant and Soil 144, 255-258. Schulte, E. H. 1975. Influence of algal concentration and temperature on the filtration rate of Mytilus edulis. Marine Biology 30, 331-341. Scura, E. D., Kuljis, A. M., York, R. H. J. & LeGoff, R. S. 1979. The commercial production of oysters in an intensive raceway system. Proceedings of the World Mariculture Society 10, 624-630. Sharma, S. S; Schat, H., Vooijs, R. 1998. In vitro alleviation of heavy metal-induced enzyme inhibition by proline. Phytochemistry 49, 1531-1535. Short, F. T. & McRoy, C. P. 1984. Nitrogen uptake by leaves and roots of seagrass Zostera marina L. Botanica Marina 27, 547-555. Shpigel, M. & Blaycock, R. A. 1991. The Pacific oyster, Crassostrea gigas, as a biological filter for a marine fish aquaculture pond. Aquaculture 92, 187-197. Shpigel, M. & Fridman, R. 1990. Propagation of the Manila clam (Tapes semidecussatus) in the effluent of fish aquaculture ponds in Eilat, Israel. Aquaculture 90, 113-122. Shpigel, M., Lee, J., Soohoo, B., Fridman, R. & Gordin, H. 1993a. Use of effluent water from fish-ponds as a food source for the pacific oyster, Crassostrea gigas Thunberg. Aquaculture and Fisheries Management 24, 529-543. Shpigel, M., Neori, A., Popper, D. M. & Gordin, H. 1993b. A proposed model for "environmentally clean" land-based culture of fish, bivalves and seaweeds. Aquaculture 117, 115-128. Silveira, J. S. M., Sant'anna, R., Rena, A. B. & Garcia, R. 1985. Nitrogen translocation as a function of the proportion of nitrate and ammonium. Pesquisa Agropecuaria Brasileira 20, 15-24.

BIBLIOGRAPHY

185

Skillicorn, P., Spira, W. & Journey, W. 1998. Duckweed Aquaculture: A New Aquatic Farming System for Developing Countries: The World Bank, Emena Technical Department, Agricultural Division. 76 pp. Smith, P. T. 1996. Physical and chemical characteristics of sediments from prawn farms and mangrove habitats on the Clarence River, Australia. Aquaculture 146, 47-83. Sobral, P. & Widdows, J. 1997. Influence of hypoxia and anoxia on the physiological responses of the clam Ruditapes decussatus from southern Portugal. Marine Biology 127, 455-461. Solbe, J. F. 1982. Fish-farm effluents: A United Kingdom Survey. In Report of the EIFAC Workshop on Fish Farm Effluents, EIFAC Tech. Pap. (Alabaster, J. A., ed). p. 65-71. Sorgeloos, P. & Sweetman, J. 1993. Aquaculture success stories. World Aquculture 24, 4-14. Sornin, J.-M., Feuillet, M., Heral, M. & Fardeau, J.-C. 1986. Influences des cultures d'huitres Crassostrea gigas sur le cycle du phosphore en zone intertidale: role de la biodeposition. Oceanologica Acta 9, 313-322. Srna, R. F. & Baggaley, A. 1976. Rate of excretion of ammonia by the hard clam Mercenaria mercenaria and the American oyster Crassostrea virginica. Marine Biology 36, 251258. State of the Environment Council. 1996. State of the Environment, Australia. CSIRO, Collingwood, Vic. 47 pp. Steward, F. C., and Pollard, J. K., 1962. In Amino Acid Pools: Distribution, Formation and Function of Free Amino Acids (Holden, J. T., ed). Elsevier, Amsterdam. 813 pp. Subander, A., Petrell, R. J. & Harrison, P. J. 1993. Laminaria culture for reduction of dissolved inorganic nitrogen in salmon farm effluent. Journal of Applied Phycology 5, 455-463.

BIBLIOGRAPHY

186

Taylor, R.B., Peek, J.T.A. & Rees, T.A.V. 1998. Scaling of ammonium uptake by seaweeds to surface area: volume ratio: geographical variation and the role of uptake by passive diffusion. Marine Ecology Progress Series 169, 143-148. Teerasuntonwat, P. & Raksakulthai, N. 1995. Production of flavouring agent from shrimp heads. ASEAN Food Journal 10, 131-138. Teichert-Coddington, D. 1995. Estuarine water quality and sustainable shrimp culture in Honduras. In Swimming through troubled water, Proceedings of the special session on shrimp farming, Aquaculture '95 (Browdy, C. L. & Hopkins, J. S., eds), Baton Rouge, Louisiana, USA, p. 144-156. Tenore, K. R. & Dunstan, W. M. 1973. Comparison of feeding and biodeposition of three bivalves at different food levels. Marine Biology 21, 190-195. Tenore, K. R., Goldman, J. C. & Clarner, J. P. 1973. The food chain dynamics of the oyster, clam, and mussel in an aquaculture food chain. Journal of Experimental Marine Biology and Ecology 12, 157 - 165. Tetzlaff, B. L. & Heidinger, R. C. 1990. Basic principles of biofiltration and system design. Illinois Aquaculture Resource/Research Center, Southern Illinois University, Carbondale. 127 pp. Thomas, T. E., Harrison, P. J. & Turpin, D. H. 1987. Adaptations of Gracilaria pacifica (Rhodophyta) to nitrogen procurement at different intertidal locations. Marine Biology 93, 569-580. Thielker, J. L. 1981. Design and test operation of an intensive controlled environment oyster production system. Journal of the World Mariculture Society 12, 79-93. Topinka, J. A. 1978. Nitrogen Uptake by Fucus spiralis (Phaeophyceae). Journal of Phycology 14, 241-247.

BIBLIOGRAPHY

187

Twilley, R. R. 1989. Impacts of shrimp mariculture practices on the ecology of coastal ecosystems in Ecuador. In Establishing a Sustainable Shrimp Mariculture Industry in Ecuador (Olsen, S. & Arriaga, L., eds). The University of Rhode Island Coastal Resource Center, Kingston, Rhode Island. p. 23-34. Udy, J. W. & Dennison, W. C. 1997a. Growth and physiological responses of three seagrass species to elevated sediment nutrients in Moreton Bay, Australia. Journal of Experimental Marine Biology and Ecology 217, 253-277. Udy, J. W. & Dennison, W. C. 1997b. Physiological responses of seagrasses used to identify anthropogenic nutrient inputs. Marine And Freshwater Research 48, 605-614. Ugarte, R. & Santelices, B. 1992. Experimental tank cultivation of Gracilaria chilensis in central Chile. Aquaculture 101, 7-16. Van Dover, C. L., Grassle, J. F., Fry, B., Garritt, R. H. & Starczak, V. R. 1992. Stable isotope evidence for entry of sewage-derived organic material into deep-sea food web. Nature 360, 153-156. Vandermeulen, H. & Gordin, H. 1990. Ammonium uptake using Ulva (Chlorophyta) in intensive fishpond systems: mass culture and treatment of effluent. Journal of Applied Phycology 2, 363-374. Vergara, J. J. & Niell, F. X. 1995. Short-term variation of internal nitrogen compounds as a function of irradiance in Corallina elongata. Botanica Marina 38, 285-290. Vergara, J. J., Niell, F. X. & Torres, M. 1993. Culture of Gelidium sesquipedale (Clem.) Born. et Thur. in a chemostat system: Biomass production and metabolic responses affected by nitrogen flow. Journal of Applied Phycology 5, 405-415. Vona, V., Di Martino Rigano, V., Esposito, S., Di Martino, C., and Rigano, C., 1992. Growth, photosynthesis, respiration, and intracellular free amino acid profiles in the

BIBLIOGRAPHY

188

unicellular alga Cyanidium caldarium: Effect of nutrient limitation and resupply. Physiologia Plantarum 85, 652-658. Wada, E. 1980. Nitrogen isotope fractionation and its significance in biogeochemical processes occurring in marine environments. In Isotope Marine Chemistry (Goldberg, E. D. & Horibe, Y., eds). Uchida-Rokakuho. p. 375-398. Wada, E., Minagawa, M., Mizutani, H., Tsujo, T., Imaizumi, R. & Karasawa, K. 1987. Biogeochemical studies on the transport of organic matter along the Otsuchi River Watershed, Japan. Estuarine and Coastal Shelf Science 25, 321-336. Wajsbrot, N., Krom, M. D., Gasith, A. & Samocha, T. 1989. Ammonia excretion of green tiger prawn Penaeus semisulcatus as a possible limit on the biomass density in shrimp ponds. Israeli Journal of Aquaculture Bamidgeh 41, 159-164. Walne, P. R. 1972. The influence of current speed, body size and water temperature on the filtration rate of five species of bivalves. J. Mar. Biol. Ass. U.K. 52, 345-374. Wang, J. K. 1990. Managing shrimp pond water to reduce discharge problems. Aquaculture Engineering 9, 61-73. Wang, J. K. & Jakob, G. S. 1991. Pond design and water management strategy for an integrated oyster and shrimp production system. Aquaculture systems engineering. In Proceedings of the World Aquaculture Society and the American Society of Agricultural engineers, Jointly Sponsored Session. San Juan, Puerto Rico, p. 70-81. Warrer-Hansen, I. 1982. Methods of treatment of waste water from trout farming. In Report of the EIFAC Workshop on Fish-Farm Effluents, Denmark, May 1981 (Alabaster, J. S., ed). EIFAC Technical Paper, p. 113-121. Welch, E. B. 1992. Ecological Effects of Waste Water: Pollution in Freshwater Systems, 2nd Ed. Chapman and Hall, London. 425 pp.

BIBLIOGRAPHY

189

Weston, D. P. 1991. The effects of aquaculture on indigenous biota. In Aquaculture and Water Quality. Advances in World Aquaculture Vol. 3 (Brune, D. E. & Tomasso, J. R., eds), Baton Rouge, Louisana, p. 534-567. Wheeler, A., 1983. Phytoplankton Nitrogen Metabolism. In Nitrogen in the Marine Environment (Carpenter, E.J. and Capone, D.G., eds). Academic Press Inc., New York. 900 pp. Wheeler, W. N. 1980. Effect of Boundary Layer Transport on the Fixation of Carbon by the Giant Kelp Macrocystis pyrifera. Marine Biology 56, 103-110. White, A. J. & Critchley, C. 1999. Rapid light curves: a new fluorescence method to assess the state of the photosynthetic apparatus. Photosynthesis Research 59, 63-72. Wilkinson, R.E. (ed.). 1994. Plant-Environment Interactions. Marcel Dekker Inc, New York. 599 pp. Wisely, B. & Reid, B. L. 1978. Experimental feeding of Sydney Rock Oysters (Crassostrea commercialis = Saccostrea cucullata) I. Optimum particle sizes and concentrations. Aquaculture 15, 319-331. Witney, E., Beumer, J. & Smith, G. 1988. Oyster culture in Queensland. Queensland Department of Primary Industries, Brisbane. 27 pp. Wong, P. K., Cheng, W. W., Wai, C. H. & Wong, C. N. 1992. Comparative studies of water quality in two brackish ponds for shrimp cultivation in a salt marsh. Environmental Pollution 77, 87-92. Worf, D. L. 1980. Biological monitoring for environmental effects. Lexington Books, Lexington. 227 pp. Wu, R. S. S. 1995. The environmental impact of marine fish culture: Towards a sustainable future. Marine Pollution Bulletin 159-166.

BIBLIOGRAPHY

190

Wu, R. S. S., Lam, K. S., Mackay, D. W., Lau, T. C. & Yam, V. 1994. Impact of marine fish farming on water quality and bottom sediment: a case study of the sub-tropical environment. Marine Environmental Research 38, 115-145. Yonge, C. M. 1926. Structure and physiology of the organs of feeding and digestion in Ostrea edulis. J. Mar. Biol. Assoc. U.K. 14, 295-386. Zeitzschel, B. 1980. Sediment water interactions in nutrient dynamics. In Marine benthic dynamics (Tenor, K. R. & Coull, B. C., eds). University of Southern California Press, Columbia, p. 195-218. Ziemann, D. A., Walsh, W. A., Saphore, E. G. & Fulton-Bennet, K. 1992. A survey of water quality characteristics of effluent from Hawaiian aquaculture facilities. Journal of the World Aquaculture Society 23, 180-191. Ziemann, J.C., Macko, S.A. & Mills, A.L. 1984. Role of seagrass and mangroves in estuarine food webs: temporal and spatial changes in stable isotope composition and amino acid content during decomposition. Bulletin of Marine Science 35, 380-392.

APPENDIX 1 FACTORS LIMITING PHYTOPLANKTON BIOMASS IN THE BRISBANE RIVER AND MORETON BAY

Jones, A.B., Dudley, B.J., & Dennison, W.C. 1998. Factors limiting phytoplankton biomass in the Brisbane River and Moreton Bay. In: Tibbets, I.R., Hall, N.J., & Dennison, W.C. (eds.). Moreton Bay and Catchment. School of Marine Science, The University of Queensland, Brisbane. pp. 179-186.

Factors Limiting Phytoplankton Biomass in the Brisbane River and Moreton Bay
Adrian B. Jones, Bernard J. Dudley and William C. Dennison
Department of Botany, The University of Queensland, Brisbane Qld 4072

Abstract
Increasing eutrophication of coastal marine environments has led to the development of nutrient sampling programs to monitor water quality. Various shortcomings of chemical analyses common in the majority of these sampling programs have identified the need to develop biological indicators (bioindicators) that can be used to detect the source, fate and impact of nutrients within eutrophic systems. Using phytoplankton bioassays, we tested the accepted model that phosphorus (P) limits phytoplankton biomass in freshwater and nitrogen (N) is limiting in coastal marine waters. The study was conducted along a salinity gradient from Lake W ivenhoe (0 ppt), through the nutrient rich, highly turbid Brisbane River to Moreton Bay (35 ppt). Water samples from seven sites along the transect were spiked to make treatments with the following nutrient concentrations: 30 M NH4 + ; 200 M NO3 -; 20 M PO4 3-; 66 M SiO3 2+ ; all nutrients combined and an unspiked control. Chlorophyll a fluorescence was measured daily for 7 d in each treatment. Nutrient limitation was inferred if there was an increase in chlorophyll a fluorescence in nutrient spiked samples relative to the control. Light limitation was inferred from an increase in chlorophyll a fluorescence in controls. Phytoplankton bioassay results indicate that phytoplankton biomass was limited by: N and P in the freshwater sites; N in the upper reaches of the Brisbane River; light in the middle reaches; and N and silica (Si) in the lower reaches. Within Moreton Bay, phytoplankton populations exhibited no response to nutrient addition. These results suggest that the generalised models of nutrient limitation may not be applicable to all regions. In particular, P may not limit phytoplankton at any salinity regime within the Moreton region.

Introduction
The Brisbane River is the largest tidal estuary flowing into Moreton Bay, and is characterised by high turbidity. It receives large inputs of nutrients from both point and non-point sources such as terrestrial runoff, sewage treatment plants and release from resuspended sediments (Moss, 1990). To assess the role of nutrients and suspended solids in eutrophication, traditional water quality analyses involved periodic sampling of parameters such as dissolved inorganic nitrogen (DIN) and phosphorus (DIP), chlorophyll a, and total suspended solids (TSS). These techniques are limited as they only provide an instantaneous measurement at the time of water collection whereas large fluctuations in the concentrations of dissolved nutrients can occur on short time scales in estuaries (Wheeler & Bjrnster, 1992; Valiela, 1995). Additionally, these analyses do not directly assess the impact of eutrophication on marine life in the system (Lyngby, 1990). Bioassays can be used to investigate the nutrient responses of the phytoplankton community, thereby providing information on the history of nutrient availability at a site. Some bioassay studies supply all but one nutrient to each treatment containing an axenic culture of phytoplankton (Smayda, 1974; Hitchcock & Smayda, 1977). If the particular treatment shows lower growth rates than the treatment with all nutrients added, then that nutrient is considered to be limiting. The technique used in the present study has been employed widely (Valiela, 1995) and uses ambient phytoplankton populations spiked with single nutrients. Rapid increases (before the other ambient nutrients have been assimilated) in phytoplankton biomass in such treatments indicate that the nutrient was limiting for the ambient phytoplankton community.
In: Tibbetts, I.R., Hall, N.J. & Dennison, W.C. eds (1998) Moreton Bay and Catchment. School of Marine Science, The University of Queensland, Brisbane. pp. 301-308.

Jones, Dudley & Dennison

In Moreton Bay, phytoplankton bioassay techniques have been used to assess both short term (~15 hr) physiological responses to nutrient enrichment and long term (up to 7 d) responses in biomass to nutrient enrichment. Short term bioassays measure CO2 uptake via 14C after 15 h incubations in artificially increased nutrient concentrations (ODonohue & Dennison, 1997). Long term bioassays examine changes in algal biomass with nutrient additions (ODonohue et al., this volume). Changes are measured as in vivo fluorescence, which can be directly correlated to chlorophyll a concentration. Interpretations of the nutrient(s) limiting to phytoplankton can be made based on how rapidly the population increases and to which nutrient(s) they respond. In some cases the population may be adapted to oligotrophic conditions and may not be capable of rapid assimilation of the nutrients. In a system where nutrients are available, but increases in biomass are limited by the absence of one particular nutrient (most commonly nitrogen), then supplying this nutrient will result in increases in biomass. When the phytoplankton community is not limited by either N or P alone; the addition of a combination of N and P may produce a marked response. Light limitation may be inferred from increases in biomass in the controls (no nutrient), after suspended solids settle out of suspension. This investigation was conducted to identify factors limiting phytoplankton biomass in the Brisbane River estuary and Moreton Bay, and to define more accurately where efforts should be directed to best monitor water quality in the region. Ultimately, this information will benefit decision making on a number of management issues, particularly nutrient removal strategies.

Materials and Methods

Seven sites were selected along a transect from Lake Wivenhoe, along the Brisbane River and into Moreton Bay (Figure 1). The transect spans the full salinity range from freshwater (0), through the tidal reaches of the river, to full salinity (35) seawater. Sites will be referred to throughout the text as distance in kilometres upstream (negative) or distance downstream (positive) from the mouth of the Brisbane River.

Water quality
Chlorophyll a was determined by filtering a known volume of water sample through Whatman GF/F filters, which were immediately frozen. Acetone extraction and calculation of chlorophyll a concentration was performed using the methods of Clesceri et al. (1989) and Parsons et al. (1989). The water collected from filtering for chlorophyll a analysis was transferred into 120 mL polycarbonate containers and immediately frozen. NH 4+ and NO3-/NO22- were determined within two weeks of sampling using the methods of Parsons et al. (1989). The concentration of total suspended solids (TSS) was determined by filtering a known volume of water onto a pre-weighed and pre-combusted (110C; 24 h) Whatman GF/C glass fibre filter. The filter was then oven dried at 60C for 24 h and TSS calculated by comparing the initial and final weights. Secchi depth was determined by lowering a 30 cm diameter secchi disk (black and white alternating quarters) through the water column until it was no longer possible to distinguish between the black and white sections.

302

Moreton Bay and Catchment

Limitation of phytoplankton biomass

+33 km -145 km -82 km

Brisbane City

+8 km

-12 km

-36 km -72 km

Figure 1. Map of study sites in the Brisbane River and Moreton Bay, Queensland, Australia. Distances are relative to the mouth of the river (negative upriver from the mouth and positive into Moreton Bay). Site 1 Lake Wivenhoe (-145 km); Site 2 Karana Downs (-82 km); Site 3 Bremer River Junction (-72 km); Site 4 Long Pocket (-36 km); Site 5 Gateway Bridge (-12 km); Site 6 South of Fishermans Island (+8 km); Site 7 Myora (+33 km).

Bioassays
Phytoplankton bioassays were conducted with ambient phytoplankton assemblages collected from seven sites in the Brisbane River and Moreton Bay (Figure 1). One 30 L drum of water was collected from each site, kept cool and shaded, and returned to an outdoor incubation facility. Four litres of water from each site was filtered through a 200 m mesh (to screen out the larger zooplankton grazers) into sealed transparent 6 L plastic containers and placed in incubation tanks filled with water (2 m diameter, 0.5 m deep). Temperature was maintained at 2C of the ambient found at each site by flowing water through the tanks and light levels were maintained at 50% of incident irradiance with neutral density screening. For each site there were six bioassay containers, each with a different nutrient treatment. Samples were spiked to make the following concentrations: NO3- (200 M); NH4+ (30 M); PO43-(20 M); SiO32+ (66 M); all nutrients at those concentrations (+All); and a control (no nutrient addition). The concentrations were chosen to ensure saturation by each particular nutrient based on the highest ambient concentrations at the study sites. In vivo fluorescence was measured for all treatments daily for 7 d, using a Turner Designs Fluorometer. The potential of light and nutrients to stimulate significant increases in phytoplankton biomass (blooms) in the bioassays was investigated. The nutrient control treatments functioned as light response treatments because sedimentation of suspended solids in the samples increased light availability above ambient levels. Light stimulated bloom potential was calculated as the difference between initial and maximum in vivo fluorescence values in the control water sample over 7 d. Nutrient stimulated bloom potential was calculated as the difference between the +All nutrients treatment and the control.
Moreton Bay and Catchment

303

Jones, Dudley & Dennison

Results
Water quality
The seven study sites occur along a salinity gradient from freshwater at -145 km to full strength seawater in the Bay sites. Water column NH4+ concentration ranged from <1.5 M at +33 km to a peak of 11 M at -12 km. NO3- ranged from <4 M at +33 km to 112 M at -36 km, near the middle of the river. TSS concentration ranged from 4 mg/L at +33 km to 23 mg/L at -36 km and the chlorophyll a concentration from 0.5 g/L at +33 km to 12 g/L at -72 km (Figure 2).

Bioassays
Freshwater sites (-145 km and -82 km) demonstrated responses in phytoplankton biomass in the +All treatments, with little or no difference in the controls and other nutrient additions. At the estuarine sites (-36 km, -12 km, +8 km), phytoplankton biomass increased in all treatments, and the control. At -72 km, the response in the control was not as great as that in the treatments. The response of phytoplankton at +8 km was primarily to nitrogen (NH4+ and NO3-) and then to silica (SiO3). Populations at the oceanic site (+33 km) showed no almost no response (Figure 3).

Discussion
Nutrient (source and concentration) and light availability (from secchi depth measurements) vary considerably along the gradient of the Brisbane River and Moreton Bay (Figure 2). The peak in NH4+ at the -12 km site may be due to a fertiliser plant, located at -7 km (Moss, 1990), and the Luggage Point sewage treatment plant near the river mouth (0 km) (Moss et al., 1992). The relatively high concentrations of NO3- (compared to NH4+) at most of the mid and upper reaches of the river may result from non-point sources, enhanced nitrification and preferential uptake of NH4+ by phytoplankton (Lipschultz et al., 1986). The response of the bioassays (Figure 3) to the +All nutrients treatment at the -145 km and -82 km sites indicates limitation by more than one nutrient. At the -72 km and -36 km sites, there was no increase in biomass above the control, indicating a response to increased light due to sedimentation of particles. At the -12 km site there was also a light response, but N and Si stimulated biomass above the control. The +8 km site had almost no response in the control, but strong responses to N and Si treatments. This indicates that light is no longer a factor controlling biomass in this region of the estuary and is consistent with the higher light availability (secchi disk readings) at this site. The +33 km site showed almost no response except for the +All treatment on the seventh day. The responses in the +All treatments at day 7 may be due to bottle effects, as all the other sites responded within the first 4 days. This response suggests that the phytoplankton community within the Brisbane River is better adapted to respond rapidly to high nutrient availability compared with those in the low nutrient waters of eastern Moreton Bay. There was an inverse relationship (r 2 = 0.93) between nutrient concentration and light availability (as total suspended solids) along the study transect. At some sites phytoplankton biomass was light limited (tidal estuary), and at others nutrient limited (freshwater and marine). Light limitation was observed when phytoplankton biomass from the highly turbid mid-river sites increased in controls after the suspended solids in the sample had settled out. This response indicates that once light limitation was removed, controls had sufficient ambient nutrient concentrations to grow as well as the treatments, to which nutrients had been added.

304

Moreton Bay and Catchment

Limitation of phytoplankton biomass

25 20 15 10 5 0

TSS Secchi

TSS (mg/L) Secchi depth (m)

120

Dissolved Nutrients (M)

100 80 60 40 20 0 14

NH4 NO3

Chlorophyll a (g/L)

12 10 8 6 4 2 0 -150

-100

-50

50

Site (distance from river mouth)

Figure 2. Water quality parameters at the seven sites along the Brisbane River/Moreton Bay study transect from Lake Wivenhoe (-145 km) to Myora in eastern Moreton Bay (+33 km).

Bloom potentials, calculated as the difference in maximum biomass (as in vivo fluorescence) between the treatment and control, represent the relative increase in biomass given saturating light or nutrient conditions. The upriver sites show the greatest nutrient-induced bloom potential due to relatively high light availability coupled with a environment containing relatively high concentrations of nutrients. The highly turbid midriver sites had the highest light-induced bloom potential (Figure 4).
Moreton Bay and Catchment

305

Jones, Dudley & Dennison

120 100 80 60 40 20 0 120

-145 km / 0

-82 km / 0

Fluorescence

100 80 60 40 20 0 120 100 80 60 40 20 0 0 1

-72 km / 0.1

-36 km / 7.6

-12 km / 24

+8 km / 33

+33 km / 35

0 0

1 1

2 2

3 3

4 4

5 5

6 6

7 7

Time (days)

Control Control

NO3 NO3

NH NH4 4

PO4 PO4

SiO3 SiO3

All

Figure 3. Phytoplankton bioassay responses at the seven sites along the Brisbane River/Moreton Bay study transect. Salinity at each site is given in parts per thousand.

Our new model (Figure 5) describes N and P limitation in freshwater sites, N limitation in the upper reaches of the river, light limitation in the middle reaches, and N and Si limitation in the lower reaches and the Bay. In contradiction to the widely accepted nutrient limitation model, we found no P limitation of phytoplankton biomass in samples from the freshwater sites. This trend has been observed elsewhere in southeast Queensland, in the Maroochy and Tweed Rivers. This departure from the accepted worldwide trend may be explained by a number of
306
Moreton Bay and Catchment

Limitation of phytoplankton biomass

140 Light stimulated

Bloom potential (g Chla / L )

120 100 80 60 40 20 0 -150

Nutrient stimulated

-100

-50

50

Site (distance from river mouth) Figure 4. Light stimulated and nutrient stimulated bloom potential along the Brisbane River/Moreton Bay study transect.

Riverine
Freshwater

Tidal Estuary
River Mouth
Salinity Gradient Saline

Marine

Light Limitation OLD MODEL: P Limitation

Primary N, Secondary Si

N Limitation
(GraphicsKaren Holloway)

Figure 5. New and old models of factors limiting phytoplankton biomass in the Brisbane River and Moreton Bay system.
Moreton Bay and Catchment

307

Jones, Dudley & Dennison

factors including large inputs from fertilisation, nutrient content of Australian soils and from the high loading of suspended solids to which P binds. These results demonstrate the need to develop a better understanding of the interactions of light and nutrients as factors limiting phytoplankton biomass in the Brisbane River estuary and Moreton Bay. Phytoplankton bioassays indicate that the established model for limiting factors may not apply to this region, and that such assays, in conjunction with traditional water quality measurements, provide key information not available from traditional water quality monitoring programs.

Acknowledgements
We would like to thank the Marine Botany practical class (BT 320) for sample collection and analysis. Andrew Moss (Queensland Department of Environment) and James McEwan (Brisbane River Moreton Bay Wastewater Management Group) provided assistance with interpretation of results.

References
Clesceri, L.S., Greenberg, A.E. & Trussel, R.R. (1989) Standard methods for the examination of water and wastewater. pp. 253-256. American Public Health Association, New York. Hitchcock, G.L. & Smayda, T.J. (1977) Bioassay of lower Narragansett Bay waters during the 19721973 winter-spring bloom using the diatom Skeletonema costatum. Limnology and Oceanography 22: 132-139. Lipschultz, F., Wofsy, S.C. & Fox, L.E. (1986) Nitrogen metabolism of the eutrophic Delaware River ecosystem (USA). Limnology and Oceanography 31: 701-716. Lyngby, J.E. (1990) Monitoring of nutrient availability and limitation using the marine macroalgae, Ceramium rubrum (Huds.) C. Ag. Aquat. Bot. 38: 153-161. Moss, A.J. (1990) Turbidity and nutrient behaviour in the estuary. In: The Brisbane River. A sourcebook for the future (eds Davie, P., Stock, E. & Low Choy, D.) pp. 307-311. Australian Littoral Society Inc. in assn with Queensland Museum, Brisbane. 427 pp. Moss, A.J., Connell, D.W. & Bycroft, B. (1992) Water quality in Moreton Bay. In: Moreton Bay in the Balance (ed. Crimp, O.N.) pp. 103-114. Australian Littoral Society Inc. and the Australian Marine Science Consortium, Moorooka, Queensland. ODonohue, M.J. & Dennison, W.C. (1997) Phytoplankton productivity response to nutrient concentrations, light availability and temperature along an Australian estuarine gradient. Estuaries 20(3): 521-533. Parsons, T.R., Maita, Y. & Lalli, C.M. (1989) A Manual of Chemical and Biological Methods for Seawater Analysis. Oxford, Pergammon Press. 173 pp. Smayda, T.J. (1974) Bioassay of the growth potential of the surface water of lower Narragansett Bay over an annual cycle using the diatom Thalassiosira pseudonana (oceanic clone, 13-1). Limnol. Oceanog. 19: 889-901. Valiela, I. (1995) Marine Ecological Processes 2nd edn. New York, Springer Verlag. 686 pp. Wheeler, P.A. & Bjrnster, B.R. (1992) Seasonal fluctuations in tissue nitrogen, phosphorus and N:P for five macroalgal species common to the Pacific northwest coast. J. Phycol. 28: 1-6.

308

Moreton Bay and Catchment

APPENDIX 2 PHOTOSYNTHETIC CAPACITY IN CORAL REEF SYSTEMS: INVESTIGATIONS INTO ECOLOGICAL APPLICATIONS FOR THE UNDERWATER PAM FLUOROMETER

Jones, A.B. & Dennison, W.C. 1998. Photosynthetic capacity in coral reef systems: investigations into ecological applications for the underwater PAM fluorometer. In: Greenwood, J.G., & Hall, N.J. (eds.). Proceedings of the Australian Coral Reef Society 75th Anniversary Conference, Heron Island October 1997. School of Marine Science, The University of Queensland, Brisbane. pp. 105-118.

Photosynthetic Capacity in Coral Reef Systems: Investigations into Ecological Applications for the Underwater PAM Fluorometer
Adrian B. Jones and William C. Dennison
Department of Botany, The University of Queensland, Brisbane Qld 4072

ABSTRACT
A submersible pulse amplitude modulated (PAM) fluorometer was used to determine the effects of desiccation, ultraviolet radiation, changes in solar radiation and nutrient availability on the photosynthetic apparatus of a variety of marine plants (zooxanthellae, benthic microalgae and macroalgae) at Heron Island, Great Barrier Reef, Australia. The PAM measures photosynthesis as irradiance-dependent photosystem II electron transport. There were a number of interspecific and intraspecific variations in electron transport rate (ETR) based on physiological and morphological differences, and the plants response to changes in environmental conditions. The highest ETR was found in the zooxanthellae of the clam Tridacna maxima , and the lowest in the calcified green macroalga Halimeda opuntia. Factors such as water velocity, ultraviolet radiation, solar radiation (total irradiance and spectral changes), desiccation, nutrient availability and algal pigment content were hypothesised as influencing intraspecific changes in ETR. A series of experimental manipulations were conducted to test these hypotheses. Reef flat algae was shaded to 50% of incident solar radiation and 0% of ultraviolet radiation. Samples of macroalgae were collected from the reef flat and 15 m depth and allowed to desiccate to determine if different populations of the same species could adapt physiologically to different environmental conditions. Reef flat samples were collected and incubated in seawater enriched in nitrogen and phosphorus to test for nutrient limitation. Significant differences in the ETR of the plants tested highlighted the impacts of various environmental parameters on photosynthetic capacity. Samples from regions with higher water velocities on the reef flat had significantly higher ETRs. Screening of ultraviolet radiation increased the maximum ETR of certain species, while prolonged periods of shading reduced the maximum ETR of some species more quickly than others. Desiccation responses were the same between deep collected and reef flat populations, although increased light and temperature did reduce the maximum ETR of the deep collected samples. Fertilisation responses varied between species. The results indicate that PAM fluorometry can be used as a tool for in situ non destructive assessment of the effects of various ecological parameters on photosynthetic activity in marine plants.

INTRODUCTION
Various environmental factors can influence the temperature, nutrient availability, light regime and cellular water content of marine macroalgae, and symbiotic microalgae. These factors can be a significant influence on photosynthetic capacity, and in turn on the productivity of the entire coral reef system (Dring, 1982; Matta & Chapman, 1995). Photosynthesis in marine plants is traditionally measured as oxygen production. This requires containment of the plant in chambers in the laboratory, or in situ. This method is time-consuming, can be destructive to the plant and creates an artificial environment which may not sufficiently simulate natural conditions (Hanelt et al., 1994; Hader et al., 1996b). In contrast, pulse amplitude modulated (PAM) fluorescence enables rapid measurement of photosynthetic responses non-destructively in situ with minimal interference to the plants immediate environment.

In: Greenwood, J.G. & Hall, N.J., eds (1998) Proceedings of the Australian Coral Reef Society 75th Anniversary Conference, Heron Island October 1997. School of Marine Science, The University of Queensland, Brisbane. pp. 105-118.

Jones & Dennison

The development of portable pulse amplitude modulated (PAM) fluorometers, in particular the submersible DIVINGPAM (Walz GmbH. Effeltrich, Germany), has facilitated many areas of novel research on photosystem physiology, regulation and ecological adaptation in marine macroalgae and other aquatic plants (Cunningham et al., 1996). PAM fluorescence has been used successfully to study photosynthesis in corals (Warner et al., 1996), macroalgae (Hanelt et al., 1994; Herrmann et al., 1995; Hader et al., 1996a; Hader et al., 1996b), and benthic microalgae (Hartig et al., 1998; Kromkamp et al., 1998). These studies have also successfully used this technique to determine the effects of environmental parameters such as temperature, desiccation, photosynthetically active radiation (PAR) and ultraviolet radiation (UVR) on the photosynthetic apparatus of the organism. The PAM fluorometer measures photosynthesis as irradiance-dependent photosystem II electron transport. A saturating pulse of light allows measurement of the electron transport rate (ETR) by saturating the plastiquinone pool between PSII and PSI. The PAM measures chlorophyll a fluorescence before and after the saturating pulse allowing subsequent calculation of the photosynthetic yield. The yield is corrected for the specific leaf absorbance and then halved to account for the two quanta of light required per electron. This final figure, the electron transport rate, has been successfully correlated to traditional measures of photosynthesis, such as changes in oxygen evolution (Xiong et al., 1996) and 14C uptake (Hartig et al., 1998). For a detailed explanation of the mechanisms involved in using PAM fluorescence to measure photosynthesis as electron transport rate, refer to Schreiber et al. (1994). Short term changes in ambient solar radiation may confound measurements of true ETR when using the PAM to make single instantaneous yield calculations (Critchley & Gademann, in prep). Generation of rapid light curves (RLCs) by the PAM avoids some of these problems by providing the specimen with periods of increasing actinic light (Critchley & Gademann, in prep), during which the response to various irradiances can be measured. The result is an ETR versus PAR curve, which can be used to determine the maximum potential rate of photosynthesis and the irradiance at which photoinhibition may occur. The aim of this research was to investigate potential uses of PAM fluorescence to determine the impacts of a range of environmental variables, including light availability (PAR and UVR), desiccation stress, water motion, and nutrient availability on the photosynthetic apparatus in a variety of marine plants.

MATERIALS AND METHODS

A DIVINGPAM (submersible pulse amplitude modulated fluorometer) was used to determine the electron transport rate (ETR) in a variety of photosynthetic organisms and their response to changes in certain environmental parameters at Heron Island, Great Barrier Reef, Australia (Figure 1). The island is characterised by a large reef flat (approximately 2 m deep at high tide) with a wide variety of macroalgae (Rogers, 1997). The presence of beachrock on the windward side of the island has led to the formation of a gutter just offshore from the beach (Coote, 1984; Kan, 1995). This region is characterised by higher water flow rates, an increased density of benthic microalgae, and a reduced density of corals and attached macroalgae (Rogers, 1997; Heil et al., in review). Experiments were conducted across the reef flat and down the reef slope out from the southern side of the island. The ETR was determined at a range of light intensities by generation of
106
ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems

Australia

Great Barrier Reef Brisbane

Heron Island
Cay Sampling Transect

Heron Reef

Heron Island

Wistari Reef Wistari Channel

Study Area

Beach Gutter Reef Flat 400 m Reef Crest Reef Slope 15 m

Figure 1. Location of study site at the southern reef flat of Heron Island, Great Barrier Reef, Australia.

rapid light curves (RLCs) using the PAM. The RLC was generated over a 90 sec period with 10 sec periods of actinic irradiance and 1 sec saturating pulses for measurement of quantum yield (Critchley & Gademann, in prep.). Data was stored in the diving PAM for subsequent download to computer. Measurements were standardised by placing the fibre optic cable in a leaf clip, attached to the plant near the growing tip of the thallus. When used with filamentous algae such as Chlorodesmis fastigiata the filaments were arranged to simulate a flat thallus. When used on corals or clams a specially designed coral clip was used to ensure no movement of the fibre optic cable during the period of the RLC. Three replicate RLCs were conducted for each species on neighbouring individuals. In all cases visibly bleached or otherwise stressed plants were avoided, and only unshaded specimens were sampled. For each experiment, measurements were taken at approximately the same time of the day and same phase of the tide to limit the effects of changes in temperature and light history. An initial survey of ten species from the reef flat was conducted to compare maximum ETR, and to determine an appropriate species for testing the effects of changing environmental conditions. C. fastigiata was chosen because it is ubiquitous across all of Heron reef and therefore is capable of adapting to a variety of habitats. The ETR of C. fastigiata was measured along a transect conducted at 20 m intervals from the beach at Heron Island, through the gutter, across the reef flat to the reef crest (Figure 1). Another transect was conducted down the reef slope, with ETR of C. fastigiata measured at 2, 5, 10 and 15 m depths (Figure 1). Further responses in ETR to changes in light availability were determined by shading a number of macroalgal species in situ on the reef flat to 50% of incident solar radiation and 0% of UVR using 1 m2 quadrats of shade cloth and polycarbonate sheeting respectively. Three replicate shade and three replicate UV screens were constructed and placed approximately 10 cm above the macroalgae. ETR was measured on afternoon low tides everyday for 4 d.
ACRS Proceedings - 75th Anniversary Conference

107

Jones & Dennison

Adaptation between populations of the same species to desiccation, high light and temperature stress was determined by collecting C. fastigiata from 15 m and from the reef flat. Samples were maintained in flow-through aquaria (water is pumped from the reef flat) for 1h prior to desiccation to allow the plants to adapt to the same light, temperature and water flow regime. The samples were then desiccated in the shade for one hour, with RLCs performed at 0 mins, 15 mins, 30 mins and 60 mins. The effect of nutrient enrichment on the maximum ETR of macroalgae (C. fastigiata, Padina tenuis and Colpomenia sinuosa) and coral ( Acropora sp.) was determined. Samples were collected from the reef flat and placed in six flow-through aquaria. Three replicate aquaria were fertilised (88g m-2 N, 22g m-2 P) with slow release Osmocote fertiliser and three unfertilised control treatments were maintained. Three replicate samples of each species were maintained in each aquarium. Samples were allowed to acclimate and respond to nutrient additions for 10 days prior to sampling ETR.

RESULTS
Of all the photosynthetic plants surveyed with the PAM, the zooxanthellae in the clam Tridacna maxima had the highest maximum ETR (230 mol e - m-2 s-1), considerably higher than those in the coral Acropora sp. (70 mol e - m-2 s-1). Chnoospora implexa (150 mol e - m-2 s-1) and Colpomenia sinuosa (125 mol e - m-2 s-1) from the Phaeophyta had the highest rates of the macroalgae, followed by Plocamium microcladioides (95 mol e - m-2 s-1) (Rhodophyta) and Chlorodesmis fastigiata (60 mol e - m-2 s-1), Caulerpa racemosa (40 mol e - m-2 s-1), and Halimeda opuntia (15 mol e- m -2 s -1) (Chlorophyta). The maximum ETR of benthic microalgae was 45 mol e - m-2 s-1. The ETR of C. fastigiata exhibited peaks in the gutter (43 mol e- m-2 s-1) and at the reef crest (45 mol e - m-2 s-1), corresponding to the regions of fastest water flow on the reef flat (Heil et al., in review). The ETR of samples across the rest of the reef flat were fairly consistent, ranging from 31 mol e - m-2 s-1 to 33 mol e - m-2 s-1 (Figure 2). The ETRs of the samples taken from depth were higher than those from the reef flat, increasing to a peak of 70 mol e - m-2 s-1 at 10m depth, and dropping to 56 mol e - m-2 s-1 at 15m (Figure 3).

Maximum ETR (mol e m s )

-2

-1

50 40 30 20 10 0 0

Gutter

Reef Crest

50

100

150

200

250

300

350

400

Distance from beach (m)

Figure 2. Maximum ETR versus PAR for Chlorodesmis fastigiata along a transect across the reef flat. 108
ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems

80 ETR (mol e m s )
-1 -2 -

70 60 50 40 30 20 10 0 0 500
-2

1000
-1

1500

PAR (mol m s ) 2m 5m 10 m 15 m

Figure 3. ETR versus PAR curves for Chlorodesmis fastigiata along a depth transect down the reef slope from 2m to 15m.

The samples of C. fastigiata collected from 15 m for the desiccation experiment were maintained in aquaria prior to desiccation. After one hour in flow-through aquaria the maximum ETR was reduced from 55 mol e - m-2 s-1 to 16 mol e - m-2 s-1. This most likely relates to the higher light environment or increased water temperature in the aquaria. During desiccation, the rate of decline in maximum ETR between the two populations was not significantly different, indicating no adaptation to desiccation by the reef flat algae (Figure 4).
60 50

Aquaria

Desiccation

(mol e - m- 2 s - 1)

Maximum ETR

40 30 20 10 0 0 20 40 60 Time (mins) Reef Flat 15m 80 100 120

Figure 4. Maximum ETR of Chlorodesmis fastigiata collected from the reef flat and from 15m depth. During the first hour samples were maintained in flow-through aquaria. For the second hour they were desiccated in the shade.
ACRS Proceedings - 75th Anniversary Conference

109

Jones & Dennison

Shading to 50% of incident light reduced the maximum ETR in C. fastigiata from 60 mol e - m-2 s-1 to 25 mol e - m-2 s-1 after 1 d. No further reduction in ETR occurred over the next 3 d. Photosynthesis in Chnoospora implexa declined from 160 mol e- m-2 s-1 to 100 mol e - m-2 s-1 after 1 d. Further reductions in maximum ETR over the next 3 d reduced the maximum ETR to 50 mol e - m-2 s-1 (Figure 5).

ETR (mol e m s )

80 60 40 20 0 0

-2

-1

Chlorodesmis fastigiata

500

1000

1500

2000

ETR (mol e m s )

-2

-1

180 120 60 0 0

Chnoospora implexa

500

1000 PAR (mol m

-2

1500 s )
-1

2000

Day 1

Day 2

Day 3

Day 4

Figure 5. The response of ETR versus PAR curves in Chlorodesmis fastigiata and Chnoospora implexa to shading (50% of incident light) over four days.

Total shading of UVR resulted in an increase in the maximum ETR in Padina tenuis from 40 mol e - m-2 s-1 to 80 mol e - m-2 s-1 over a 4 day period. There was no change however in the maximum ETR of C. fastigiata (Figure 6). Fertilisation significantly increased the maximum ETR in P. tenuis from 30 mol e- m -2 s-1 to 45 mol e- m-2 s-1. There was no change in the maximum ETR of C. fastigiata, but the photoinhibitory response was reduced. By contrast, Colpomenia sinuosa showed no significant response to fertilisation and Acropora sp. showed a slight decline (Figure 7).
110
ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems

60 50 Chlorodesmis fastigiata

ETR (mol e m s )

-2 -

-1

40 30 20 10 0 0 100 80 60 40 20 0 0 200 400 600 800

-2 -1

500

1000 Padina tenuis

1500

2000

ETR (mol e m s )

-2

-1

1000

1200

1400

PAR (mol m s ) Control UV Screened

Figure 6. The response of ETR versus PAR of Chlorodesmis fastigiata and Padina tenuis to screening of 100% of incident UVR.

DISCUSSION
Comparisons of the maximum ETR indicate that the photosynthetic activity between species is variable, ranging from 15 mol e - m-2 s-1 for Halimeda opuntia to 230 mol e - m-2 s-1 for Tridacna maxima (Table1). Titlyanov et al. (1994) reported that of twenty species of macroalgae tested, Halimeda sp. had the lowest photosynthetic capacity. It is unclear why the maximum ETR of the zooxanthellae of T. maxima is so much higher than those in Acropora sp. It may be due to the presence of a different species of Symbodinium (Rowan & Powers, 1991) or due to the presence of more autofluorescent chromatophores in T. maxima (Schlichter et al., 1994). Autofluorescent chromatophores transform short wavelength irradiance, which is less suitable for photosynthesis, into longer wavelengths which are photosynthetically more effective.
ACRS Proceedings - 75th Anniversary Conference

111

Jones & Dennison

50 30

ETR (mol e - m -2 s -1)

ETR (mol e - m -2 s -1)

Padina tenuis

Chlorodesmis fastigiata

40 30 20 10 0 0 500 1000 1500 2000

25 20 15 10 5 0 0 200 400 600 800 1000 1200 1400

60

ETR (mol e - m -2 s -1)

50 40 30 20 10 0 0 500 1000
-2

ETR (mol e - m -2 s -1)

Colpomenia sinuosa

100 80 60 40 20 0 1500 s )
-1

Acropora s p .

2000

500 PAR (mol m

1000
-2

1500

PAR (mol m

s )

-1

Unfertilised

Fertilised

Figure 7. The response of ETR versus PAR to fertilisation (88g m-2 N, 22g m-2 P).

Table 1.

The maximum recorded ETR for selected photosynthetic marine organisms on Heron Island reef flat

Group Phaeophyta

Species Chnoospora implexa Colpomenia sinuosa Padina tenuis

Maximum ETR (mol e- m -2 s -1) 150 125 40 95 60 40 15 230 70 45

Rhodophyta Chlorophyta

Plocamium microcladioides Chlorodesmis fastigiata Caulerpa racemosa Halimeda opuntia

Microalgae Benthic microalgae 112

Tridacna maxima (zooxanthellae) Acropora sp. (zooxanthellae)

ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems

Although the ETR of the benthic microalgae was not relatively high (45 mol e- m-2 s-1), its high biomass across the reef flat (Heil et al., in review) makes it potentially a very large contributor to total primary production. Interestingly, the species with the highest ETRs are from the Phaeophyta (Chnoospora implexa and Colpomenia sinuosa) and Rhodophyta ( Plocamium microcladioides). The higher ETRs of the brown and red algae relative to the green algae (Chlorophyta) may be due to the enhanced light capturing ability of their accessory pigments (Rowan, 1989). For calculation of the absolute ETR based on PAM fluorometry, it is necessary to carefully measure the mean specific absorption coefficient of the algae (Hartig et al., 1998). Due to the difficulty in obtaining an accurate value, most workers are currently using the default absorbance of 0.84, which was calculated as an average based on analysis of terrestrial leaves. The use of this value is acceptable for comparing the ETR of individuals within one species. It must be realised, however, that there will be some error associated when comparing species with very different morphologies, pigment complements and chloroplast configurations unless the species specific absorption coefficients are calculated (Cunningham et al., 1996). The transect conducted across the reef to measure the maximum ETR of C. fastigiata shows definite peaks at the gutter just offshore from the beach, and at the reef crest. The gutter is formed due to scouring from the action of waves returning from the beachrock with more energy (Coote, 1984; Kan, 1995). The region is characterised by higher current velocities at the sediment surface than the rest of the reef flat (Heil et al., in review). The increased water motion in these areas reduces the boundary layer allowing faster diffusion of CO2 across the cell membrane (Shashar et al., 1996). This may explain the higher ETR measured in C. fastigiata at these sites. Benthic microalgae within this same gutter region at Heron Island have a much higher ETR than other areas on the reef flat (Heil et al., in review). Another possible explanation for the higher rates in the gutter is the increased depth in this region, which may provide slightly more protection from PAR and UVR. The highest ETR in C. fastigiata along our depth transect was recorded at 10 m. This could be due to the increased concentration of chlorophyll a within plants at depth (Titlyanov et al., 1992; Titlyanov et al., 1994) or may be a reflection of photoinhibition in plants at the surface due to exposure to high solar radiation or high UVR. The subsequent decline in ETR observed at 15 m may be due to reduced light availability from attenuation through the water column, or because of changes in the spectrum of available PAR. As a chlorophyte, C. fastigiata absorbs predominantly red and blue light for photosynthesis (Rowan, 1989) and the red wavelength is the first to be absorbed by water. Other workers have also found optimal photosynthetic rates at intermediate depths. In particular, Hader et al. (1997) demonstrated optimal photosynthesis using oxygen evolution measurements of another chlorophyte (Caulerpa prolifera) at a depth of 5 m. Photoinhibition at mid range depths may be further reduced because UVR is removed at a much greater rate than PAR through the water column (Franklin & Forster, 1997). The impact of desiccation on the rate of decline in the maximum ETR of C. fastigiata was not different between the reef flat and 15 m samples. However, the initial decline in the ETR of the 15 m deep samples from 55 mol e - m-2 s-1 to 16 mol e - m-2 s-1 is probably because of the increased solar radiation or increased temperature in the aquaria. This indicates that adaptations to localised reef environments have occurred between individual populations of C. fastigiata. This is consistent with the work of Porst et al. (1997) and Franklin et al. (1996), who observed a drastic decline in the effective quantum yield
ACRS Proceedings - 75th Anniversary Conference

113

Jones & Dennison

when plants collected from depth were exposed to high irradiance at the surface. Previous studies have demonstrated increases in photosynthesis (measured by oxygen evolution) during initial emersion when the thallus has lost 10-20% of its initial water content (Johnson et al., 1974; Quadir et al., 1979; Dring & Brown, 1982; Beer & Eshel, 1983), although after substantial desiccation photosynthesis declines (Dring & Brown, 1982). The initial increase may be due to a reduction in the boundary layer when the water covering the surface of the thallus has evaporated, thereby allowing for faster rates of diffusion of CO2. In this study, there was no initial increase in ETR during desiccation. This may be related to differences in morphology between C. fastigiata and the species tested by other workers. The samples of C. fastigiata for this study were collected from near the gutter region of the reef flat and so are never exposed at low tide. Perhaps if they had been collected from the reef crest, which is exposed at low tide, there may have been adaptations to resist desiccation. Both PAR and UVR can be important inhibitors of photosynthesis in marine plants (Hader et al., 1996a). Shading to 50% of incident light reduced the maximum ETR of C. fastigiata and C. implexa. There was a gradual reduction in maximum ETR over 4 d by C. implexa compared to the relatively quick decline over 1 d in C. fastigiata. This suggests that C. implexa has more efficient photoadaptive mechanisms to better tolerate changes in light availability. It is also interesting to note the lack of photoinhibition in C. implexa at higher PAR during the rapid light curve, another indication of resistance to changes in light availability. The alpha of the ETR versus PAR curves was also significantly different between the two species, with C. fastigiata being more efficient at lower irradiances, even though it was C. implexa which was better able to cope with the reduced light regime. Padina tenuis had a significantly higher maximum ETR after being shaded from 100% of the UVR, which is consistent with the findings of Hader et al. (1996c) and Porst et al. (1997) who observed that UVR had an overproportional inhibitory effect on the photosynthesis of two species of chlorophyte. However, in this study C. fastigiata showed no significant change even after 4 d. The filaments that make up C. fastigiata are each one cell, which allows the plant to move its chloroplasts to the tip for photosynthesis, or down into the base of the filaments for repair. This cytoplasmic streaming has been postulated as a mechanism by which this species can mitigate the photoinhibitory effects of UVR and high PAR (Franklin & Larkum, 1997). The fertilisation responses in P. tenuis and C. fastigiata are consistent with an increase in pigment concentrations facilitating higher photosynthesis and reduced photoinhibition at higher light (Titlyanov et al., 1994). Fertilisation of Acropora sp. resulted in a slight depression in maximum ETR, although the initial fluorescence was significantly higher. Higher initial fluorescence may indicate that there is greater biomass of zooxanthellae within the coral. Several studies (Hoegh-Guldberg & Smith, 1989; Dubinsky et al., 1990; Stimson & Kinzie, 1991) have reported that supplying nitrogen to coral resulted in increases in the population of zooxanthellae, but reductions in the per cell rates of photosynthesis. They hypothesised that this was due to shading and competition for CO2 because of the greater algal density. The host corals may also play a role in preventing access of the zooxanthellae to the intracellular nutrients within the host, which are speculated as being quite low (Muscatine, 1980; DElia & Cook, 1988; Snidvongs & Kinzie III, 1994). Another possible theory for the reduction in photosynthesis in the N and P fertilised treatment relates to the effects of dissolved inorganic phosphorus on inhibition of calcification by the coral host (Simkiss, 1964). This reduces the availability of CO2 to the symbiotic algae (Snidvongs & Kinzie III, 1994).
114
ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems

In conclusion, our investigations reveal that PAM fluorescence measurements can be used to assess the photosynthetic response of macroalgae, zooxanthellae and benthic microalgae to a variety of environmental parameters such as desiccation, light, and nutrient availability.

ACKNOWLEDGEMENTS
The authors would like to thank the students of the field courses, Coral Reef Biology and Geology (1996) and Terrestrial and Marine Environmental Physiology (1997) for help with sample collection. We would like to thank two anonymous reviewers whose comments led to substantial revision and improvement to the manuscript.

REFERENCES
Beer, S. & Eshel, A. (1983) Photosynthesis of Ulva sp. I. Effects of desiccation when exposed to air. Journal of Experimental Marine Biology and Ecology 70: 91-97. Cook, C.B., Muller-Parker, G. & Orlandini, C.D. (1994) Ammonium enhancement of dark carbon fixation and nitrogen limitation in zooxanthellae symbiotic with reef corals Madracis mirabilis and Monastrea annularis. Marine Biology 118: 157-165. Coote, S.M. (1984) A study of beachrock on Heron Island, Great Barrier Reef, Australia . MscQual Thesis, Department of Earth Sciences. The University of Queensland, Brisbane. Critchley, C. & Gademann, R. (in prep.) Rapid light curves: a new approach to determine the actual state of the photosynthetic apparatus. . Cunningham, A., Levavasseur, G., Estrada, M., Hanelt, D. & Wilhelm, C. (1996) Technical discussion III: Fluorescence measurements. Scientia Marina 60: 301-302. DElia, C.F. & Cook, C.B. (1988) Methylamine uptake by zooxanthellae - invertebrate symbioses: insights into host ammonium environment and nutrition. Limnology and Oceanography 33: 1153-1165. Dring, M.J. & Brown, F.A. (1982) Photosynthesis of intertidal brown algae during and after periods of emersion: a renewed search for physiological causes of zonation. Marine Ecology Progress Series 8: 301-308. Dring, M.T. (1982) The Biology of Marine Plants. London, Edwin Arnold Press 199 pp. Dubinsky, Z., Stambler, N., Ben-zion, M., McCloskey, L.R., Muscatine, L. & Falkowsky, P.G. (1990) The effect of external nutrient resources on the optical properties and photosynthetic efficiency of Stylophora pistillata . Proceedings of the Royal Society (Ser B) 239: 231-246. Franklin, L., Seaton, G., Lovelock, C. & Larkum, A. (1996) Photoinhibition of photosynthesis on a coral reef. Plant Cell Environment 19: 825-836. Franklin, L.A. & Forster, R.M. (1997) The changing irradiance environment: consequences for marine macrophyte physiology, productivity and ecology. European Journal of Phycology 32: 207-232. Franklin, L.A. & Larkum, A.W.D. (1997) Multiple strategies for a high light existence in a tropical marine macroalga. Photosynthesis Research 53: 149-159. Hader, D.P., Herrmann, H. & Santas, R. (1996a) Effects of solar radiation and solar radiation deprived of UV-B and total UV on photosynthetic oxygen production and pulse amplitude modulated fluorescence in the brown alga Padina pavonia . FEMS Microbiology Ecology 19: 53-61. Hader, D.P., Lebert, M., Mercado, J., Aguilera, J., Salles, S., FloresMoya, A., Jimenez, C. & Figueroa, F.L. (1996b) Photosynthetic oxygen production and PAM fluorescence in the brown alga Padina pavonica measured in the field under solar radiation. Marine Biology 127: 61-66.
ACRS Proceedings - 75th Anniversary Conference

115

Jones & Dennison Hader, D.P., Porst, M., Herrmann, H., Schaefer, J. & Santas, R. (1996c) Photoinhibition in the Mediterranean green alga Halimeda tuna Ellis et sol measured in situ . Photochemistry and Photobiology 64: 428434. Hader, D.P., Porst, M., Herrmann, H., Schafer, J. & Santas, R. (1997) Photosynthesis of the mediterranean green alga Caulerpa prolifera measured in the field under solar irradiation. Journal of Photochemistry and Photobiology B Biology 37: 66-73. Hanelt, D., Li, J. & Nultsch, W. (1994) Tidal Dependence of Photoinhibition of Photosynthesis in Marine Macrophytes of the South China Sea. Botanica Acta 107: 66-72. Hartig, P., Wolfstein, K., Lippemeier, S. & Colijn, F. (1998) Photosynthetic activity of natural microphytobenthos populations measured by fluorescence (PAM) and C-14-tracer methods: a comparison. Marine Ecology Progress Series 166: 53-62. Heil, C.A., Chaston, K.A., Jones, A.B., Bird, P., Costanzo, S., Longstaff, B., Scheffers, S. & Dennison, W.C. (in review) Benthic microalgae in coral reef sediments. Coral Reefs. Herrmann, H., Ghetti, F., Scheuerlein, R. & Hader, D.P. (1995) Photosynthetic oxygen and fluorescence measurements in Ulva laetevirens affected by solar irradiation. Journal of Plant Physiology 145: 221-227. Hoegh-Guldberg, O. & Smith, G.J. (1989) Influence of the population density of zooxanthellae and supply of ammonium on the biomass and metabolic characteristics of the reef corals Seriatophora hystrix and Sylophora pistillata . Marine Ecology Progress Series 57: 173-186. Johnson, W.S., Gigon, A., Gulmon, S.L. & Mooney, H.A. (1974) Comparative photosynthetic capacities of intertidal algae under exposed and submerged conditions. Ecology 55: 450-453. Kan, G.J.A. (1995). A study of the sedimentology and beachrock formation on Heron Island cay, southern Great Barrier Reef, Queensland, Australia . BSc (hons) Thesis, Department of Earth Sciences, The University of Queensland, Brisbane. Kromkamp, J., Barranguet, C. & Peene, J. (1998) Determination of microphytobenthos PSII quantum efficiency and photosynthetic activity by means of variable chlorophyll fluorescence. Marine Ecology Progress Series 162: 45-55. Matta, J.L. & Chapman, D.J. (1995) Effects of light, temperature and desiccation on the net emersed productivity of the intertidal macroalga Colpomenia peregrina Sauv. (Hamel). Journal of Experimental Marine Biology and Ecology 189: 13-27. Muscatine, L. (1980) Uptake, retention, and release of dissolved inorganic nutrients by marine algae invertebrate associations. In: Cellular interactions in symbiosis and parasitism (eds Cook, C.B., Pappas, P.W. & Rudolph, E.D.) pp. 229-244. Ohio University Press, Columbus. Porst, M., Herrmann, H., Schafer, J., Santas, R. & Hader, D.P. (1997) Photoinhibition in the Mediterranean green alga Acetabularia mediterranea measured in the field under solar irradiation. Journal of Plant Physiology 151: 25-32. Quadir, A., Harrison, P.J. & Widaver, W. (1979) The effects of emergence and submergence on the photosynthesis and respiration of marine macrophytes. Phycologia 18: 83-88. Rogers, R.W. (1997) Brown algae on Heron Reef flat, Great Barrier Reef, Australia: Spatial, seasonal and secular variation in cover. Botanica Marina 40: 113-117. Rowan, K.S. (1989) Photosynthetic pigments of algae Cambridge University Press, Cambridge 334 pp. Rowan, R. & Powers, D. (1991) Molecular genetic identification of symbiotic dinoflagellates (zooxanthellae). Mar. Ecol. Prog. Ser. 71: 65-73. Schlichter, D., Meier, U. & Fricke, H. (1994) Improvement of photosynthesis in zooxanthellate corals by autofluorescent chromatophores. Oecologia 99: 1-2. 116
ACRS Proceedings - 75th Anniversary Conference

Photosynthesis in coral reef systems Schreiber, U., Bilger, W. & Neubauer, C. (1994) Chlorophyll fluorescence as a nonintrusive indicator for rapid assessment of in vivo photosynthesis. In: Ecophysiology of photosynthesis. Ecological Studies., vol. 100 (eds. Schulz, E.D. & M.M. Caldwell), pp. 49-70. Springer-Verlag, Berlin. Shashar, N., Kinane, S., Jokiel, P.L. & Patterson, M.R. (1996) Hydromechanical boundary layers over a coral reef. Journal Of Experimental Marine Biology And Ecology 199: 17-28. Simkiss, K. (1964) Phosphates as crystal poisons of calcification. Biological Revue 39: 487-505. Snidvongs, A. & Kinzie III, R.A. (1994) Effects of nitrogen and phosphorus enrichment on in vivo symbiotic zooxanthallae of Pocillipora damicornis. Marine Biology 118: 705-711. Stimson, J. & Kinzie, R.A. (1991) The temporal pattern and rate of release of zooxanthellae from the reef coral Pocillipora damicornis (L.) under nitrogen-enrichment and control conditions. Journal of Experimental Marine Biology and Ecology 153: 63-74. Titlyanov, E.A., Bil, K.Y., Kolmakov, P.V., Lapshina, A.A. & Parnik, T.R. (1992) Photosynthesis in common macrophyte species in the intertidal and upper subtidal zones of the Seychelles Islands. Atoll Research Bulletin 0: 1-36. Titlyanov, E.A., Bil, K.Y., Kolmakov, P.V., Lapshina, A.A. & Titlyanova, T.V. (1994) Characteristics of Production and Light Adaptation of Marine Macrophyte Species Common to the Seychelles. Russian Journal of Plant Physiology 41: 225-231. Warner, M.E., Fitt, W.K. & Schmidt, G.W. (1996) The effects of elevated temperature on the photosynthetic efficiency of zooxanthellae in hospite from four different species of reef coral: A novel approach. Plant Cell and Environment. Mar 19: 291-299. Xiong, F.S., Lederer, F., Lukavsky, J. & Nedbal, L. (1996) Screening of freshwater algae (Chlorophyta, Chromophyta) for ultraviolet-B sensitivity of the photosynthetic apparatus. Journal of Plant Physiology 148: 42-48.

ACRS Proceedings - 75th Anniversary Conference

117