EFFECTS OF FUNGICIDES ON AUSTRALIAN AMPHIPODS AND ORGANIC MATTER BREAKDOWN IN AQUATIC ENVIRONMENTS

Submitted by Hung Thi Hong Vu

Submitted in fulfillment of the degree of Doctor of Philosophy

February 2017

School of BioSciences

Faculty of Science

The University of Melbourne

ABSTRACT

Fungicides are used widely in agriculture to control fungal diseases and increase crop yield. After application, fungicides may be transported off site via air, soil and water to ground and surface waters therefore have the potential to contaminate freshwater and marine/estuarine environments. However, relatively little is known about their potential effects on aquatic ecosystems. Amphipods are important in ecosystem service as they help with nutrient recycling through the decomposition of organic matter. The aim of this thesis is to investigate the effects of common fungicides on biological responses in two Australian amphipod species, Allorchestes compressa and Austrochiltonia subtenuis, through a combination of single and mixture laboratory experiments. In addition a field experiment investigated the effects of fungicides on organic matter breakdown.

In laboratory studies, juveniles of the marine amphipod A. compressa and the freshwater amphipod A.subtenuis were chronically exposed to two commonly used fungicides, Filan® (active ingredient boscalid) and Systhane™ (active ingredient myclobutanil) at environmentally relevant concentrations. A wide range of endpoints that encompass different levels of biological organization were measured including survival, growth, reproduction, and energy reserves (lipid, glycogen, and protein content). Long term interaction effects of fungicides Filan® and Systhane™ on mature amphipod A. subtenuis was also investigated to evaluate how the results of mixture studies vary between endpoints and to determine suitable endpoints for mixture toxicity studies.

In the field study, leaves and cotton strips were deployed at 26 sites, 24 study and 2 reference sites, in an intensive agricultural region in south-eastern Australia to investigate the effects of fungicides and other anthropogenic stressors on organic matter breakdown. Leaves and cotton strips were deployed at the sites for a three week period and repeated twice in winter and spring. Breakdown rates of leaf and cotton at studied sites were compared to that of the reference site two which has similar altitude to the study sites to determine the effects of fungicides and other stressors on functional stream health. Pesticide concentration and physico-chemical parameters of sites were monitored during the study. The relationship between organic matter breakdown rates and environmental variables was investigated.

Page 1 of 120 Laboratory results demonstrated that Filan® and Systhane™ caused significantly adverse effects on survival, growth, reproduction, and energy reserves of both species of amphipod at environmentally realistic concentrations. Female amphipods were more sensitive to fungicides than males in terms of growth. Reproduction was the most sensitive endpoint and most affected by fungicide exposure. The effects of fungicide mixtures on A.subtenuis were endpoint-dependent and antagonistic effects were observed only on reproduction. Field data showed that organic matter breakdown rate was significantly correlated with pesticide concentrations and nutrients but leaf breakdown was also strongly impacted by temperature. Leaf and cotton degraded differently but both indicated the same results on functional stream health for majority of the sites.

This thesis provides the first evidence of the effects of common fungicides on survival, growth, reproduction, and energy reserves of two Australian amphipods at environmentally relevant concentrations. The results suggest that fungicide pollution could affect the viability of amphipod populations in the natural environments that consequently could cause cascading effects on the ecosystem. This is also the first study to investigate individual relationships between different pesticide groups with organic matter breakdown in a field environment. The results of this study emphasize the importance of considering the long-term effects of fungicides in risk assessments for aquatic ecosystems and contribute to the literature of fungicide toxicity on aquatic environments.

Page 2 of 120

DECLARATION

This is to certify that: i. The thesis comprises only my original work towards the PhD ii. Due acknowledgement has been made in the text to all other material used iii. The thesis is less than 100,000 words in length, excluding tables, maps, bibliographies and appendices

Hung Thi Hong Vu

February 2017

Page 3 of 120 PREFACE

This thesis comprises one introduction chapter (Chapter 1), three scientific papers (Chapters 2, 3, and 4), one manuscript (Chapter 5), and one discussion chapter (Chapter 6).

Chapter 1 - The literature review

Chapter 2

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2016. Effects of the boscalid fungicide Filan® on the marine amphipod Allorchestes compressa at environmentally relevant concentrations. Environmental Toxicology and Chemistry 35:1130-1137.

The major research of this publication is my own work. Other co-authors provide scientific advice, training in laboratory and data analysis techniques, and reviewing the manuscript before submission.

Chapter 3

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2017. Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry 36:720-726.

The major research of this publication is my own work. Other co-authors provide scientific advice, training in laboratory and data analysis techniques, and reviewing the manuscript before submission.

Chapter 4

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2017. Toxicological effects of fungicide mixtures on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry. DOI: 10.1002/etc.3809.

Page 4 of 120 The major research of this manuscript is my own work. Other co-authors provide scientific advice, training in laboratory and data analysis techniques, and reviewing the manuscript before submission.

Chapter 5

Hung T. Vu, Jackie H. Myers, Simon M. Sharp, Claudette R. Kellar, Sara M. Long, Michael J. Keough, and Vincent J. Pettigrove (in preparation). Can organic matter decomposition indicate the effects of multiple anthropogenic stressors on functional stream health?

The research of this manuscript was incorporated in the Western Port Toxicant Study Stage 3 project and some of the results were also reported in “Myers, JH., Sharley, D., Sharp, S., Vu, H., Long, S., and Pettigrove, V. (2016), Final Report Western Port Toxicant Study Stage 3 – Pesticide Sourcing Study and Aquatic Flora and Fauna Assessment, Centre for Aquatic Pollution Identification and Management, Technical Report No. 63A, University of Melbourne, Victoria, Australia.” In this manuscript, the nature and extent of authors’ contributions to the work were the following:

Name Nature of contribution Extent of contribution (%) for student authors only Hung T. Vu Field work, sample processing, data 60% analysis and interpretation of results, contribution to project design, and preparation of manuscript Jackie H. Myers Project manager, assistance with field N/A work, and review of data and manuscript Simon M. Sharp Pesticides and water chemistry analysis, N/A assistance with field work Claudette R. Kellar Macroinvertebrate identification and N/A

Page 5 of 120 review of manuscript Sara M. Long Scientific advice and review of N/A manuscript Michael J. Keough Scientific advice and review of data and N/A manuscript Vincent J. Pettigrove Scientific advice and review of data and N/A manuscript

Chapter 6 – General discussion

Page 6 of 120 ACKNOWLEDGMENTS

I would like to express my special gratitude to my supervisors Associate Professor

Vincent Pettigrove, Professor Michael Keough, and Dr Sara Long. Thank you for your great support, guidance, and encouragement throughout my PhD candidature. I am grateful for many hours, in and out of working time, of discussion and advice you provided me. It has been a great honor and pleasure to be your student.

I would like to thank all the CAPIM staff and students for helping me during my candidature. In particular, I would like to thank Jackie Myers, Dave Sharley, Katy Jeppe,

Claudette Kellar, Cameron Amos, Steve Marshall, Rebecca Brown, Daniel MacMahon,

Simon Sharp, Pat Bonney, Rhianna Boyle, Jessica French, and Tyler Mehler for their help with field work. I would like to thank Allyson O’Brien for providing comments and helping improve the second chapter.

I would like to thank Peter Symes and Therese Turner, Royal Botanic Gardens

Melbourne who provided support and assistance with collecting Pomaderris aspera leaves using in my laboratory studies.

I would also like to thank my parent, my parent in-law, my two sisters, and my brother and his wife for their support and assistance with taking care of my children while I was busy with my study.

I dedicate this PhD thesis to my husband and our two beloved children. Thank you for your accompanies during this challenging but interesting journey. Thank you for joining me in happy moments and cheering me up when I was feeling down.

Page 7 of 120 The Melbourne International Research Scholarship (MIRS) and the Centre of Aquatic

Pollution Identification and Management (CAPIM) funded this study. Extra funding was gained through Melbourne Water, Holsworth Wildlife Research Endowment, the Faculty of Science Travelling Scholarship, Drummond Travel Award, and SETAC Student

Travel Awards.

Page 8 of 120 CONTENTS

Abstract ...... 1 Declaration...... 3 Preface ...... 4 Acknowledgments ...... 7 Contents ...... 9 Chapter 1: Introduction ...... 11 1.1 Fungicides ...... 11 1.2 Organic matter breakdown in aquatic environments...... 12 1.3 Role of macroinvertebrates in leaf litter decomposition ...... 13 1.4 Effects of fungicides on shredding amphipods and leaf litter decomposition ...... 14 1.5 Key knowledge gaps ...... 14 1.6 Studied subjects ...... 15 1.7 Thesis aims and overview ...... 16 REFERENCES ...... 20 Chapter 2: Effects of the boscalid fungicide Filan® on the marine amphipod Allorchestes compressa at environmentally relevant concentrations...... 26 Chapter 3: Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis ...... 38 Chapter 4: Toxicological effects of fungicide mixtures on the amphipod Austrochiltonia subtenuis ...... 48 Chapter 5: Can organic matter decomposition indicate the effects of multiple anthropogenic stressors on functional stream health? ...... 60 5.1 Introduction ...... 60 5.2 Materials and Methods ...... 63 5.3 Results ...... 67 5.4 Discussion ...... 72 5.5 Conclusion ...... 78 5.6 Figures...... 79 5.7 Tables ...... 85 5.8 Supplemental data ...... 89

Page 9 of 120 REFERENCES ...... 103 Chapter 6: General discussion ...... 110 6.1 Effects of fungicides on amphipods – Laboratory perspective ...... 110 6.2 Effects of fungicides on ecosystem function – Linking laboratory results with field observations ...... 113 6.3 Recommendations for future studies ...... 116 REFERENCES ...... 117

Page 10 of 120 CHAPTER 1: INTRODUCTION

1.1 Fungicides

Fungicides are chemical or biological compounds used in agriculture, industry, and the home to control fungal diseases. The global fungicide market has rapidly increased during the last decade reaching 15.1 billion USD in 2015 (Mordor-Intelligence, 2016) from 9.2 billion USD in 2007 (Grube et al., 2011). Approximately 227 million kilograms (kg) of fungicides are applied worldwide (Reigart and Roberts, 2013) and over 2.7 million kg are used in Australia (Radcliffe, 2002) annually. However, there are no publicly available data on the amount of individual fungicide used in terms of active ingredients or formulated products in Australia and other countries. Fungicide use is strongly dominated by agriculture (Grube et al., 2011) as they are an important component of plant disease management plans for agronomic crops (Wightwick et al., 2010). After application, fungicides may be transported off site via air, soil and water to ground and surface water and so have the potential to contaminate both freshwater and marine/estuarine environments.

Fungicides have been detected in aquatic environments in many countries worldwide, with total concentrations ranging from a few nanograms per litre to several tens of micrograms per litre. In the United States, fungicides were detected in 75% of the fresh surface waters and 58% of the ground wells sampled in intense fungicide use areas across the country (Reilly et al., 2012) and in more than 80% of water samples in a central Californian estuary (Smalling et al., 2013). In Europe, fungicides have frequently been detected in surface water in which the common fungicide, pyrimethanil, was measured at concentrations up to 70 µg/L (Seeland et al., 2013) and some others were detected at concentrations above guideline values (Kreuger et al., 2010). In Australia, fungicides have also been detected regularly in both freshwater and estuarine environments. Fungicides were detected in 63% of spot water samples in a horticultural production catchment (Wightwick et al., 2012) and in 76% of samples collected via passive sampling techniques in an agricultural catchment (Myers et al., 2016) in south-eastern Australia. Fungicides were also detected in sediment in three estuaries in Victoria, Australia at concentrations that may pose a risk to resident fauna and flora (Sharp et al., 2013).

Page 11 of 120 While fungicides are produced to kill or inhibit fungal diseases their modes of action are non-specific to fungi and may be deleterious to nonpathogenic fungi and non-target organisms (Maltby et al., 2009). Currently, there are limited toxicological data on single and combined effects of fungicides on non-target species and their environmental effects are largely unknown (Reilly et al., 2012; Wightwick et al., 2012). There is increased interest in effects of fungicides on aquatic leaf decomposing fungi and leaf shredding invertebrates, especially amphipods (Rasmussen et al., 2012; Zubrod et al., 2011; Zubrod et al., 2010; Zubrod et al., 2015) because not only they are vulnerable non-target organisms but they also play critical roles on leaf litter decomposition, a fundamental process in aquatic environments.

1.2 Organic matter breakdown in aquatic environments

Detritus or dead organic matter, especially leaf litter, has been considered a main energy source for both freshwater (Graca, 2001; Imberger et al., 2008; Petersen and Cummins, 1974) and estuarine/marine ecosystems (Kenworthy and Thayer, 1984; Lastra et al., 2008). Many researchers have confirmed the important role of allochthonous organic material, derived from riparian trees, in the total energy budget of stream communities ranging from 66 to 99% depending on stream locations (Fisher and Likens, 1972; Nelson and Scott, 1962; Teal, 1957). In most estuarine/marine ecosystems, consumption of organic matter is mainly through leaf litter and only a small amount of live plant tissue is consumed in situ (Fenchel, 1970; Parker et al., 2008; Wahbeh and Mahasneh, 1985).

The breakdown of leaf litter in aquatic environments is influenced by biotic and abiotic factors and can be separated into three phases:

a. Loss of soluble substances (leaching).

b. Microorganism colonization (conditioning) which makes the leaves more palatable for the animal because of enhanced nutrient availability.

c. Fragmentation of leaf material by macroinvertebrate feeding activity and physical abrasion (Anderson and Sedell, 1979; Petersen and Cummins, 1974; Webster and Benfield, 1986).

Page 12 of 120 Although this complex process occurs sequentially, some phases can happen simultaneously and affect each other (Silva-Junior and Moulton, 2011). It has been shown that invertebrates prefer to eat leaves conditioned by microorganisms rather than freshly fallen leaves (Barlocher and Kendrick, 1975; Cummins et al., 1973) or part of leaves that are heavily colonized by fungi (Graca et al., 2000). In turn, the feeding activities of macroinvertebrates increase the microbial biomass due to increase total surface area by decreasing the size of particulate organic matter (Fenchel, 1970) or providing nutrient-rich feces on the leaf surface (Graca et al., 2000). Thus, the factors that drive feeding on detritus are pivotal to the flow of energy and nutrients in aquatic ecosystems (Parker et al., 2008).

1.3 Role of macroinvertebrates in leaf litter decomposition

Macroinvertebrate feeding activity has a major influence on leaf litter decomposition (Cummins et al., 1973). The role of macroinvertebrates, particularly shredders, in accelerating leaf litter decomposition has been reported in both freshwater and estuarine/marine ecosystems. More than 50% of pignut hickory (Carya glabra) mass was lost in the presence of the shredder Tipula abdominalis compared to 26% loss in the control after 110 days (Cummins et al., 1973). The presence of the snail Goniobasis clavaeformis in laboratory streams increased the weight loss from leaves six times more than without snails (Mulholland et al., 1985). Hieber and Gessner (2002) found that shredders had the highest contribution to the leaf mass loss, 64% on alder and 51% on willow leaves, compared to fungi and bacteria. Robertson and Lucas (1983) studied the importance of the amphipod Allorchestes compressa in the turnover of the kelp species Ecklonia radiata and reported that the breakdown rate of E. radiata by the amphipod is comparable with those measured for the physical breakdown and microbial decomposition. Mariano et al. (2008) estimated that the population of the talitrid amphipod (Megalorchestia corniculata) could process on average 55% of the brown macroalgae Macrocystis along a Californian beach. Because of their significant role in fragmenting leaf litter and favoring microbial colonization, any adverse effects on shredders will have strong impacts on leaf litter decomposition. Studies on anthropogenic effects on leaf litter decomposition have focused on amphipods as they are typical

Page 13 of 120 shredders (Graca, 2001). Amphipods are widespread throughout a diverse range of freshwater and marine habitats and can be the dominant part of many benthic macroinvertebrate assemblages, in terms of both numbers and/or biomass (MacNeil et al., 1997).

1.4 Effects of fungicides on shredding amphipods and leaf litter decomposition

Fungicides are often considered to have a low toxicity to aquatic animals consequently they have received little attention compared to herbicide and insecticides in terms of effects on aquatic organisms. Currently, there are few studies on the effects of fungicides on amphipod shredders, especially with respect to leaf litter decomposition. The majority of studies have been laboratory based and have shown that fungicides have affected amphipod survival (Zubrod et al., 2014), reproduction function (Jacobson and Sundelin, 2006; Jubeaux et al., 2012), feeding rate and growth (Bundschuh et al., 2011; Dimitrov et al., 2014; Feckler et al., 2016; Zubrod et al., 2010), and energy reserves (Zubrod et al., 2011). Furthermore, their effects increased when amphipods were exposed to fungicides in combination with increasing temperature (Jacobson et al., 2008), other fungicides (Zubrod et al., 2014; Zubrod et al., 2015), or other insecticides (Flores et al., 2014; Rasmussen et al., 2012). Most studies have shown that the reduction in leaf consumption by amphipods was due to the poor nutritional quality of food as a result of altered fungal biomass and community composition following fungicide exposure rather than direct effects on the amphipods themselves (Bundschuh et al., 2011; Feckler et al., 2016; Zubrod et al., 2011). However, amphipods may increase their leaf consumption to compensate for the nutritional deficiency to meet energy requirements for basic physiological processes (Rasmussen et al., 2012).

1.5 Key knowledge gaps

Most of the aforementioned studies (Chapter 1.4) focused on short term exposure and used concentrations that exceeded those detected in natural environments. To my knowledge, there are no available studies on long term effects of single and mixtures of fungicides on shredder amphipods at environmentally relevant concentrations. The effects of fungicides on amphipod reproduction and transgenerational effects after long

Page 14 of 120 term exposure are lacking. Amphipod reproduction success and healthy juveniles are important for population fitness in natural environments. Reduction in numbers of amphipods could have significant impacts on leaf litter breakdown and, therefore, ecosystem function because abundance of consumers is one of the main drivers of organic matter processing rate (Flores et al., 2014). Furthermore, most of the current studies were conducted in Europe using European indigenous species (Fernandez et al., 2015; Zubrod et al., 2014; Zubrod et al., 2011; Zubrod et al., 2010) while studies in other geographical regions are missing. The current world trend is to use native species to assess the effects of priority toxicants as they have adapted to the local environmental conditions and can thus provide much more representative outcomes than those obtained with a foreign species (Giusto and Ferrari, 2014). Finally, fungicides have been shown to affect shredder amphipods and their feeding (Bundschuh et al., 2011; Feckler et al., 2016; Zubrod et al., 2014) which potentially impacts leaf litter breakdown (Flores et al., 2014) however this has never been evaluated in a field environment. There are many other factors (such as water conditions and other toxicants) in the field that could interact and alter fungicide toxicity, which need to be assessed when studying the effects of fungicides in the field.

1.6 Studied subjects 1.6.1 Fungicides

Fungicides used in my thesis were boscalid and myclobutanil. They are used on a variety of crops, fruit trees, vegetables and so on in Australia and other countries (Kreuger et al., 2010; Wightwick et al., 2012). These fungicides have been frequently detected in water and sediments samples in Australia (Myers et al., 2016; Sharp et al., 2013; Wightwick et al., 2012) and other countries (Kreuger et al., 2010; Phillips and Bode, 2004; Reilly et al., 2012; Smalling and Orlando, 2011). Boscalid and myclobutanil are considered chemicals of concern due to their high global use rates, high detection frequency in surface waters, and likely persistence in the environment (Elskus, 2012), therefore understanding the effects of these chemicals is relevant to the global community. Boscalid was applied using the commercially available product Filan® fungicide (Nufarm, Australia), containing 500 g active ingredient (a.i.)/kg. Mycobutanil was

Page 15 of 120 applied using the commercially available product SysthaneTM 400 WP fungicide (Dow AgroSciences, Australia), containing 400 g a.i./kg.

1.6.2 Test species

The chosen amphipods were the freshwater species Austrochiltonia subtenuis (Amphipoda, Ceinidae) and the marine species Allorchestes compressa (Amphipoda, Dogielinotidae). Both amphipods are local species and abundant in aquatic environments. A. subtenuis is widespread in southern Australia (Williams, 1962) and is one of the most abundant species in lowland standing waters in western Victoria (Lim and Williams, 1971; Timms, 1983). A. compressa is abundant on the shores of southeastern (Burridge et al., 1995) and southwestern (Crawley and Hyndes, 2007) Australia.

1.6.3 Leaf species

Hazel pomaderris (Pomaderris aspera) was chosen for freshwater study as it is a representative of the dominant form of litter entering streams in Australia (Boulton and Boon, 1991). Green hazel pomaderris leaves were collected in the Royal Botanic Gardens Melbourne, Victoria, Australia.

The seagrass Zostera muelleri was chosen for the marine study. Z. muelleri occurs throughout Victorian coastal waters and is a common seagrass in Port Philip Bay where the marine amphipods were collected (Warry and Hindell, 2009). Green Z. muelleri leaves were collected at Clifton Springs beach, Victoria. Seagrass has been used as a food source for A. compressa in both acute and chronic tests (Ahsanullah and Williams, 1986; Burridge et al., 1995)

1.7 Thesis aims and overview

This thesis reports on the long term effects of common fungicides on two Australian amphipods in the laboratory at environmentally relevant concentrations and on organic matter breakdown in a field study at locations where fungicides have been detected regularly, which address the gaps in the literature identified in Chapter 1.5. A conceptual model was developed for this study based on relevant theoretical concepts of leaf litter breakdown and key information gaps on effects of fungicides on decomposer-detritivore

Page 16 of 120 systems that were mentioned above. This model demonstrated the potential (direct and indirect) effects of (single and mixture) fungicides on the amphipods and organic matter breakdown.

Fungicides (Single & mixture)

Direct effects Amphipods Organic matter (Survival, growth, breakdown reproduction) Indirect effects (Macroinvertebrate & microorganism activity)

There were three aims to the thesis:

- Investigate long term effects of common fungicides on Australian freshwater and marine amphipods under laboratory conditions at environmentally relevant concentrations. - Investigate long term effects of fungicide mixtures on freshwater amphipods under laboratory conditions at environmentally relevant concentrations and determine suitable endpoints for chronic mixture studies in future. - Assess effects of fungicides on stream ecosystem function in aquatic environments and determine suitable diagnostic tools for fungicide risk assessment in biomonitoring programs.

Page 17 of 120 This thesis is divided into four experimental chapters, an introduction chapter, and a discussion chapter, as outlined below.

Chapter 1: Introduction

Aim: An overview of current knowledge on effects of fungicides on amphipod shredders and leaf litter breakdown and identify the gaps in the literature.

Chapter 2: Effects of the boscalid fungicide Filan® on the marine amphipod Allorchestes compressa at environmentally relevant concentrations

Aim: To investigate long term effects of the most commonly detected fungicide on the marine amphipod A. compressa under laboratory conditions at environmentally relevant concentrations. A wide range of endpoints was measured including biochemical (lipid, protein, glycogen content) and physiological (feeding rate) biomarkers as well as life history traits (growth, reproduction, and survival) to identify the most sensitive endpoint and investigate the relationship between biochemical changes and effects at higher levels of organization. Indirect effects of boscalid on amphipod food quality were also investigated through microbial respiration on seagrass.

Chapter 3: Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis

Aim: To investigate long term effects of two fungicides on the freshwater amphipod A. subtenuis under laboratory conditions at environmentally relevant concentrations. Fungicides with different modes of action could have different effects on non-target organisms. The direct effects of boscalid and myclobutanil on the amphipod at organism level (survival, growth, and reproduction) and sub-organism level (lipid, protein, and glycogen content) as well as indirect effects on amphipod food quality were investigated. This chapter compares the sensitivity of the amphipod A. subtenuis to two fungicides with different modes of action.

Chapter 4: Toxicological effects of fungicide mixtures on the amphipod Austrochiltonia subtenuis

Page 18 of 120 Aim: To investigate the long term interaction effects of fungicides boscalid and myclobutanil on mature A. subtenuis at environmentally realistic concentrations. Multiple endpoints that span different levels of biological organization were used including organism-level responses (survival, reproduction and growth) and sub-organism level (GST) to look at sensitivity in response at low fungicide concentrations. This chapter evaluates how the results of mixture studies vary between endpoints and suggests suitable endpoints for mixture toxicity studies. This chapter also provides more insights on the effects of these fungicides on amphipod reproduction that was determined the most sensitive endpoint in Chapter 3.

Chapter 5: Can organic matter decomposition indicate the effects of multiple anthropogenic stressors on functional stream health?

Aim: To investigate the effects of fungicides and other co-occurring stressors on stream ecosystem function through the breakdown of leaf and cotton in aquatic environments. Relationship between leaf and cotton breakdown rates and environmental stressors were analyzed to determine the most suitable diagnostic tool for fungicide risk assessment in future biomonitoring programs.

Chapter 6: General discussion

Aim: To summarize the main findings from this study and recommend further research directions.

Page 19 of 120 REFERENCES

Ahsanullah, M., Williams, A.R., 1986. Effect of uranium on growth and reproduction of the marine amphipod Allorchestes compressa. Marine Biology 93, 459-464. Anderson, N.H., Sedell, J.R., 1979. Detritus processing by macroinvertebrates in stream ecosystems. Annual Review of Entomology 24, 351-377. Barlocher, F., Kendrick, B., 1975. Leaf conditioning by microorganisms. Oecologia 20, 359-362. Boulton, A.J., Boon, P.I., 1991. A review of methodology used to measure leaf litter decomposition in lotic environments - Time to turn over an old leaf. Australian Journal of Marine and Freshwater Research 42, 1-43. Bundschuh, M., Zubrod, J.P., Kosol, S., Maltby, L., Stang, C., Duester, L., Schulz, R., 2011. Fungal composition on leaves explains pollutant-mediated indirect effects on amphipod feeding. Aquat Toxicol 104, 32-37. Burridge, T.R., Lavery, T., Lam, P.K.S., 1995. Acute toxicity tests using Phyllospora- comosa (Labillardiere) Agardh, C. (Phaeophyta, Fucales) and Allochestes compressa Dana (Crustacea, Amphipoda). Bulletin of Environmental Contamination and Toxicology 55, 621-628. Crawley, K.R., Hyndes, G.A., 2007. The role of different types of detached macrophytes in the food and habitat choice of a surf-zone inhabiting amphipod. Marine Biology 151, 1433-1443. Cummins, K.W., Petersen, R.C., Howard, F.O., Wuycheck, J.C., Holt, V.I., 1973. Utilization of leaf litter by stream detritivores Ecology 54, 336-345. Dimitrov, M.R., Kosol, S., Smidt, H., Buijse, L., Van den Brink, P.J., Van Wijngaarden, R.P., Brock, T.C., Maltby, L., 2014. Assessing effects of the fungicide tebuconazole to heterotrophic microbes in aquatic microcosms. Science of the Total Environment 490, 1002-1011. Elskus, A.A., 2012. Toxicity, sublethal effects, and potential modes of action of select fungicides on freshwater fish and invertebrates. U.S. Geological Survey Open-File Report 2012–1213, p. 44.

Page 20 of 120 Feckler, A., Goedkoop, W., Zubrod, J.P., Schulz, R., Bundschuh, M., 2016. Exposure pathway-dependent effects of the fungicide epoxiconazole on a decomposer-detritivore system. Science of the Total Environment 571, 992-1000. Fenchel, T., 1970. Studies on decomposition of organic detritus derived from turtle grass Thalassia - testudinum. Limnology and Oceanography 15, 14-20. Fernandez, D., Voss, K., Bundschuh, M., Zubrod, J.P., Schaefer, R.B., 2015. Effects of fungicides on decomposer communities and litter decomposition in vineyard streams. Science of the Total Environment 533, 40-48. Fisher, S.G., Likens, G.E., 1972. Stream ecosystem - organic energy budget. Bioscience 22, 33-35. Flores, L., Banjac, Z., Farre, M., Larranaga, A., Mas-Marti, E., Munoz, I., Barcelo, D., Elosegi, A., 2014. Effects of a fungicide (imazalil) and an insecticide (diazinon) on stream fungi and invertebrates associated with litter breakdown. Science of the Total Environment 476, 532-541. Giusto, A., Ferrari, L., 2014. Biochemical responses of ecological importance in males of the austral South America amphipod Hyalella curvispina Shoemaker, 1942 exposed to waterborne cadmium and copper. Ecotoxicology and Environmental Safety 100, 193-200. Graca, M.A., Newell, S.Y., Kneib, R.T., 2000. Grazing rates of organic matter and living fungal biomass of decaying Spartina alterniflora by three species of salt-marsh invertebrates. Marine Biology 136, 281-289. Graca, M.A.S., 2001. The role of invertebrates on leaf litter decomposition in streams - a review. International Review of Hydrobiology 86, 383-393. Grube, A., Donaldson, D., Kiely, T., Wu, L., 2011. Pesticides industry sales and usage. U.S Environmental Protection Agency, Washington, p. 33. Hieber, M., Gessner, M.O., 2002. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology 83, 1026-1038. Imberger, S.J., Walsh, C.J., Grace, M.R., 2008. More microbial activity, not abrasive flow or shredder abundance, accelerates breakdown of labile leaf litter in urban streams. Journal of the North American Benthological Society 27, 549-561. Jacobson, T., Prevodnik, A., Sundelin, B., 2008. Combined effects of temperature and a pesticide on the Baltic amphipod Monoporeia affinis. Aquatic Biology 1, 269-276.

Page 21 of 120 Jacobson, T., Sundelin, B., 2006. Reproductive effects of the endocrine disruptor fenarimol on a baltic amphipod Monoporeia affinis. Environmental Toxicology and Chemistry 25, 1126-1131. Jubeaux, G., Simon, R., Salvador, A., Queau, H., Chaumot, A., Geffard, O., 2012. Vitellogenin-like proteins in the freshwater amphipod Gammarus fossarum (Koch, 1835): Functional characterization throughout reproductive process, potential for use as an indicator of oocyte quality and endocrine disruption biomarker in males. Aquat Toxicol 112, 72-82. Kenworthy, W.J., Thayer, G.W., 1984. Production and decomposition of the roots and rhizomes of seagrasses, Zostera marina and Thalassia testudinum, in temperate and subtropical marine ecosystems. Bulletin of Marine Science 35, 364-379. Kreuger, J., Graaf, S., Patring, J., Adielsson, S., 2010. Pesticides in surface water in areas with open ground and greenhouse horticultural crops in Sweden 2008, Uppsala, Sweden. Lastra, M., Page, H.M., Dugan, J.E., Hubbard, D.M., Rodil, I.F., 2008. Processing of allochthonous macrophyte subsidies by sandy beach consumers: estimates of feeding rates and impacts on food resources. Marine Biology 154, 163-174. Lim, K.H., Williams, W.D., 1971. Ecology of Austrochiltonia Subtenuis (Sayce) (Amphipoda, Hyalellidae). Crustaceana 20, 19-24. Maltby, L., Brock, T.C.M., van den Brink, P.J., 2009. Fungicide risk assessment for aquatic ecosystems: importance of Interspecific variation, toxic mode of action, and exposure regime. Environ Sci Technol 43, 7556-7563. Mordor-Intelligence, 2016. Global fungicides market - growth, trends and forecasts (2016 - 2021). Mulholland, P.J., Elwood, J.W., Newbold, J.D., Ferren, L.A., 1985. Effect of a leaf- shredding invertebrate on organic matter dynamics and phosphorus spiralling in heterotrophic laboratory streams. Oecologia 66, 199-206. Myers, J., Sharley, D., Sharp, S., Vu, H., Long, S., Pettigrove, V., 2016. Western port toxicant study: pesticide sourcing study and aquatic flora and fauna assessment Nelson, D.J., Scott, D.C., 1962. Role of detritus in the productivity of a rock outcrop community in a Piedmont stream. Limnology and Oceanography 7, 396-413.

Page 22 of 120 Parker, J.D., Montoya, J.P., Hay, M.E., 2008. A specialist detritivore links Spartina alterniflora to salt marsh food webs. Marine Ecology Progress Series 364, 87-95. Petersen, R.C., Cummins, K.W., 1974. Leaf processing in a woodland stream. Freshwater Biology 4, 343-368. Phillips, P.J., Bode, R.W., 2004. Pesticides in surface water runoff in south‐eastern New York State, USA: seasonal and stormflow effects on concentrations. Pest Management Science 60, 531-543. Radcliffe, J.C., 2002. Pesticide use in Australia, p. 309. Rasmussen, J.J., Monberg, R.J., Baattrup-Pedersen, A., Cedergreen, N., Wiberg-Larsen, P., Strobel, B., Kronvang, B., 2012. Effects of a triazole fungicide and a pyrethroid insecticide on the decomposition of leaves in the presence or absence of macroinvertebrate shredders. Aquat Toxicol 118, 54-61. Reigart, J.R., Roberts, R.J., 2013. Fungicides, Recognition and management of pesticide poisonings, Sixth ed. U.S. Environmental Protection Agency, pp. 143 - 160. Reilly, T.J., Smalling, K.L., Orlando, J.L., Kuivila, K.M., 2012. Occurrence of boscalid and other selected fungicides in surface water and groundwater in three targeted use areas in the United States. Chemosphere 89, 228-234. Robertson, A.I., Lucas, J.S., 1983. Food choice, feeding rates, and the turnover of macrophyte biomass by a surf-zone inhabiting amphipod. Journal of Experimental Marine Biology and Ecology 72, 99-124. Seeland, A., Albrand, J., Oehlmann, J., Müller, R., 2013. Life stage-specific effects of the fungicide pyrimethanil and temperature on the snail Physella acuta (Draparnaud, 1805) disclose the pitfalls for the aquatic risk assessment under global climate change. Environmental Pollution 174, 1-9. Sharp, S., Myers, J., Pettigrove, V., 2013. An assessment of sediment toxicants in Western Port and major tributaries, Melbourne Water. Silva-Junior, E.F., Moulton, T.P., 2011. Ecosystem functioning and community structure as indicators for assessing environmental impacts: leaf processing and macroinvertebrates in Atlantic forest streams. International Review of Hydrobiology 96, 656-666.

Page 23 of 120 Smalling, K.L., Kuivila, K.M., Orlando, J.L., Phillips, B.M., Anderson, B.S., Siegler, K., Hunt, J.W., Hamilton, M., 2013. Environmental fate of fungicides and other current-use pesticides in a central California estuary. Marine Pollution Bulletin 73, 144-153. Smalling, K.L., Orlando, J.L., 2011. Occurrence of pesticides in water and sediment from three central California coastal watersheds, 2008–2009. U.S. Geological Survey Data Series 600, p. 70. Teal, J.M., 1957. Community metabolism in a temperate cold spring Ecological Monographs 27, 283-302. Timms, B.V., 1983. A study of benthic communities in some shallow saline lakes of Wentern Victoria, Australia. Hydrobiologia 105, 165-177. Wahbeh, M.I., Mahasneh, A.M., 1985. Some aspects of decomposition of leaf litter of the seagrass Halophila stipulacea from the gulf of Aqaba (Jordan). Aquatic Botany 21, 237- 244. Warry, F.Y., Hindell, J.S., 2009. Review of Victorian seagrass research, with emphasis on Port Phillip Bay. Arthur Rylah Institute for Environmental Research. Draft Report. , Department of Sustainability and Environment, Heidelberg, Victoria, p. 33. Webster, J.R., Benfield, E.F., 1986. Vascular plant breakdown in freshwater ecosystems. Johnston, R. F. (Ed.). Annual Review of Ecology and Systematics, Vol. 17. Xi+714p. Annual Reviews Inc.: Palo Alto, California, USA. Illus, 567-594. Wightwick, A., Allinson, G., Menzies, N., Walters, R., Reichman, S., 2010. Environmental risks of fungicides used in horticultural production systems. INTECH Open Access Publisher. Wightwick, A.M., Bui, A.D., Zhang, P., Rose, G., Allinson, M., Myers, J.H., Reichman, S.M., Menzies, N.W., Pettigrove, V., Allinson, G., 2012. Environmental fate of fungicides in surface waters of a horticultural-production catchment in Southeastern Australia. Archives of Environmental Contamination and Toxicology 62, 380-390. Williams, W.D., 1962. The Australian freshwater amphipods. I. The genus Austrochiltonia (Crustacea: Amphipoda: Hyalellidae). Australian Jour Mar and Freshwater Res 13, 198-216.

Page 24 of 120 Zubrod, J., Baudy, P., Schulz, R., Bundschuh, M., 2014. Effects of current-use fungicides and their mixtures on the feeding and survival of the key shredder Gammarus fossarum. Aquat Toxicol 150, 133-143. Zubrod, J.P., Bundschuh, M., Feckler, A., Englert, D., Schulz, R., 2011. Ecotoxicological impact of the fungicide tebuconazole on an aquatic decomposer-detritivore system. Environmental Toxicology and Chemistry 30, 2718-2724. Zubrod, J.P., Bundschuh, M., Schulz, R., 2010. Effects of subchronic fungicide exposure on the energy processing of Gammarus fossarum (Crustacea; Amphipoda). Ecotoxicology and Environmental Safety 73, 1674-1680. Zubrod, J.P., Englert, D., Feckler, A., Koksharova, N., Konschak, M., Bundschuh, R., Schnetzer, N., Englert, K., Schulz, R., Bundschuh, M., 2015. Does the current fungicide risk assessment provide sufficient protection for key drivers in aquatic ecosystem functioning? Environ Sci Technol 49, 1173-1181.

Page 25 of 120 CHAPTER 2: EFFECTS OF THE BOSCALID FUNGICIDE FILAN® ON THE MARINE AMPHIPOD ALLORCHESTES COMPRESSA AT ENVIRONMENTALLY RELEVANT CONCENTRATIONS

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2016. Effects of the boscalid fungicide Filan® on the marine amphipod Allorchestes compressa at environmentally relevant concentrations. Environmental Toxicology and Chemistry 35:1130-1137

Page 26 of 120 Environmental Toxicology and Chemistry, Vol. 35, No. 5, pp. 1130–1137, 2016 # 2015 SETAC Printed in the USA

EFFECTS OF THE BOSCALID FUNGICIDE FILAN1 ON THE MARINE AMPHIPOD ALLORCHESTES COMPRESSA AT ENVIRONMENTALLY RELEVANT CONCENTRATIONS

HUNG T. VU,* MICHAEL J. KEOUGH,SARA M. LONG, and VINCENT J. PETTIGROVE Centre for Aquatic Pollution Identification and Management (CAPIM), School of Biosciences, The University of Melbourne, Victoria, Australia

(Submitted 1 April 2014; Returned for Revision 12 May 2014; Accepted 14 September 2015)

Abstract: Fungicides are widely used in agriculture to control fungal diseases. After application, fungicides can be transported offsite to surface and groundwater and ultimately enter estuarine and marine environments. The presence of fungicides in the marine environment may pose risks to marine organisms, but little is known about fungicide effects on these organisms, especially invertebrates. The present study investigated the effects of the commonly used boscalid fungicide Filan1 on life history traits, feeding rate, and energy reserves (lipid, glycogen, and protein content) of the marine amphipod Allorchestes compressa over 6 wk under laboratory conditions. Amphipods were exposed to 3 concentrations of Filan (1 mg, 10 mg, and 40 mg active ingredient [a.i.]/L), with 5 replicates per treatment. Lipid content and reproduction were the most sensitive measures of effect, with lipid content reduced by 53.8% at the highest concentration. Survival, growth, and other energy reserves of amphipods were also negatively affected by Filan, and the effects were concentration dependent. Antennal deformities were incidentally observed on the amphipods at a concentration of 40 mg a.i./L. The results of the present study indicate comprehensive effects of the boscalid fungicide Filan on A. compressa at environmentally relevant concentrations. The decline or absence of A. compressa in marine ecosystems could impair the ecosystem function because of their important role in trophic transfer and nutrient recycling. The authors’ results suggest that even though the use of fungicides is often regarded as posing only a minor risk to aquatic organisms, the assessment of their long-term effects is critical. Environ Toxicol Chem 2016;35:1130–1137. # 2015 SETAC

Keywords: Fungicide Marine invertebrates Reproduction Energy reserves Filan1

INTRODUCTION concentrations could vary because of environmental conditions Fungicides are used in agriculture and industry or even (e.g., dry season vs wet season) or increase in sediment because domestically to control fungal infections, which are increas- some of them are persistent in aquatic environments. For ingly recognized as presenting a threat to global food instance, the highest concentration of the fungicide azoxystro- security [1]. After application, fungicides can be transported bin detected in a coastal estuary (California, USA) in the dry m off site via air, soil, and water, and therefore, potentially season was 20.2 g/L, but in the storm season it was detected at m contaminate ground waters, surface waters, freshwaters [2,3], concentrations as high as 4550 g/L [4]. In Western Port and marine and estuarine environments [4,5]. Although (Victoria, Australia), boscalid was the most frequently detected m fungicides are designed to kill or inhibit fungal pathogens, fungicide, with the highest concentration of 22 g/kg [7]. their modes of action are not specific to fungi [6]. Therefore, the Fungicides are usually considered to have low toxicity to presence of fungicides in waterways may pose risks to aquatic aquatic animals compared with other pesticides. Limited organisms. knowledge is available about the levels of fungicides in the Fungicides are often used as a prophylactic crop protectant marine environment as well as the chronic effects of fungicides that is applied at higher frequencies but at lower application on marine species. A few studies have shown that fungicides can rates than other types of pesticides [2]. Fungicides, therefore, affect marine invertebrates at relatively low concentra- – are often detected in surface water in areas of intense fungicide tions [8 10]. When exposed to sublethal concentrations – m use at low concentrations but high frequencies [2,3]. Conse- (367 825 g/L) of the fungicide propiconazole, the shrimp quently, aquatic organisms are likely to be chronically and Litopenaeus vannamei showed morphological deformities of repeatedly exposed to fungicides at relatively low concen- the rostrum, pereopods, and uropods [8]. The fungicide trations, especially during the application season. Furthermore, carbendazim altered the malondialdehyde level, glutathione, the migration of fungicides from these areas to marine or and antioxidant activity of the marine bivalve Donax faba at m m estuarine ecosystem via streams and rivers could lead to concentrations ranging from 52.65 g/L to 842.6 g/L [10]. relatively low concentrations as a result of dilution. For However, these 2 fungicides occurring commonly at such high example, Smalling et al. [5] reported that the fungicide boscalid concentrations in the environment is unlikely. In contrast, the was detected in 100% of water samples in in a coastal estuary fungicide 2-methoxyethylmercuric chloride exhibited broader m (California, USA) throughout the year, but maximum concen- toxicity at very low concentrations (1 g/L) across every criti- trations were lower than laboratory-derived aquatic life bench- cal life transition and stage of the broadcast-spawning coral marks for fish and invertebrates. However, the fungicide Acropora millepora [9]. However, interpreting the ecological impacts of fungicides on aquatic ecosystems without examining the effects of a range of fungicides at environmentally relevant This article includes online-only Supplemental Data. concentrations is difficult. * Address correspondence to [email protected] Published online 17 September 2015 in Wiley Online Library Boscalid is a systemic fungicide that is active against a broad (wileyonlinelibrary.com). range of fungal pathogens and has been used in a wide range of DOI: 10.1002/etc.3247 crops [3]. It is resistant to most environmental degradation and

1130 Page 27 of 120 Toxicity of fungicide Filan1 to Allorchestes compressa Environ Toxicol Chem 35, 2016 1131 is expected to be environmentally persistent [11]. Boscalid further solvents. Filan was dissolved in deionized water to make has been commonly detected in both freshwater [2,3,12] a stock solution with a nominal concentration of 50 mg a.i./L. and estuarine environments [4]. In Western Port, Australia, The stock solution was then diluted to achieve nominal Filan boscalid was often detected in water samples, and the highest concentrations of 1 mg a.i./L, 10 mg a.i./L, and 40 mg a.i./L. Both concentration was recorded at 3.3 mg/L (Centre for Aquatic stock and test medium were prepared immediately before use. Pollution Identification and Management, School of Bioscien- Water samples were collected before the experiment and after ces, The University of Melbourne, Victoria, Australia, unpub- 1 wk exposure and sent to Advanced Analytical Australia Pty lished data, 2013). It is also one of the most frequently detected (North Ryde, NSW, Australia) for analysis of boscalid pesticides (in greater than 90% of the samples) in 3 main coastal concentrations. Sample concentrations were determined using estuaries in California, USA, with concentrations as high as liquid chromatography–tandem mass spectrometry (MS/MS). 36 mg/L [4]. Boscalid has been found in estuarine fish and Water samples were diluted in 30/70 water/(methanol þ 0.1% crabs [5], which could absorb the fungicide directly from the formic acid). An aliquot of diluted water samples was injected environment or through eating contaminated prey. Boscalid is onto an Aglient 1260 Infinity-HPLC (ESI positive mode), and expected to be present in the marine environment, and therefore components were separated on a Phenomenex-Gemini C18 organisms in this ecosystem have the potential to be exposed (150 2 3um) column at 35 8C. The detector was a Varian- to it. To our knowledge, no current data are available on the 320 MS/MS, set at a temperature of 300 8C and a scan time of sublethal effects of boscalid on marine organisms. 2 s. The binary mobile phase was 5 mM ammonium formate The amphipod Allorchestes compressa is abundant and (pH 3.5) and acetonitrile to methanol (4:1) þ 0.2% formic widely distributed along the shores of southeast [13] and acid, using an initial gradient of 30%, which increased to 100% southwest [14] Australia. Allorchestes compressa is a semi- in 3 min, with a flow rate of 0.15 mL/min. Boscalid concen- aquatic amphipod because it inhabits detached macrophytes in trations were quantified by multiple reaction monitoring of the intertidal regions of the shores, and at certain times during 343 m/z >307 m/z, 272 m/z, by external standard quantification. the day (e.g., low tide) the amphipod may not be submerged. It is Boscalid reference material (Novachem) was of 98% purity. an important food source for various fish species [15] and plays The limit of reporting for boscalid was 0.1 mg/L. A spike an important role in the trophic transfer and nutrient recycling in recovery was performed with the analytical batch, on sample 22 marine ecosystems along the Australian coast [14]. This species (reported unspiked at less than the limit of reporting), and the has been used in both acute [13,15] and chronic toxicity recovery was 72%. Reported results were not corrected for tests [16], and they are suitable test organisms for studying recovery. The measured boscalid concentrations were within toxicant effects on growth and reproduction under laboratory 10% of nominal experimental concentrations if these were conditions [16]. corrected for the reported recovery. In the present study, we investigated the effects of a 1 commonly used fungicide, boscalid fungicide Filan , on the Test species marine amphipod A. compressa at a range of concentrations Allorchestes compressa and its food, the seagrass Zostera that have been found to occur in some natural estuarine muelleri, were collected from Clifton Springs beach, Victoria, environments [4,7]. Although these concentrations do not Australia, which is considered to be at low risk of pollution [20]. reflect a realistic environmental situation but rather an unusual Amphipods were maintained in ambient seawater in the event (e.g., caused by accidental releases), investigating how laboratory in groups of 500 at experimental conditions: such rare situations could affect a key marine species is temperature (20 1 8C), salinity (34 2‰), and at a 16:8-h necessary. Furthermore, as a semi-aquatic species, A. com- light:dark photoperiod in 20-L glass aquaria under constant pressa could be exposed directly to relatively high toxicant aeration. The seawater was obtained from a circulating seawater concentrations from agricultural or urban runoff. Under system in School of BioSciences, the University of Melbourne. laboratory conditions, we measured sublethal biochemical Animals were given dry seagrass as food and water was changed (lipid, protein, glycogen content) and physiological (feeding weekly. After the acclimatization period (14 d), gravid females rate) biomarkers as well as life history traits (growth, were separated into clean 2-L glass beakers containing ambient reproduction, and survival). Previous studies have shown that seawater. One week later, the resulting juveniles were fungicides had indirect effects on amphipods through changing transferred to fresh 2-L glass beakers and maintained as the microbial composition or biomass on the leaves that the described previously. Amphipods used in the experiment were amphipods consume, thereby reducing the palatability of their less than 6 wk old. food [17,18]. Oxygen consumption correlated with bacterial growth [19], so changes in microbial respiration could reflect Preconditioned seagrass changes to bacterial biomass. Therefore, we measured During the experiment, amphipods were given precondi- microbial respiration in the seagrass used to feed the amphipods tioned seagrass as food. To precondition the seagrass, the as a way to assess the indirect effects of Filan on the microbial freshly collected seagrass was cleaned with tap water and air- community. We are aware that different groups of micro- dried before use. Approximately 25 mg dried seagrass was organisms could contribute differently to the nutrient quality of weighed and placed in nutrient-enriched seawater (5 mg P as conditioned leaves [17], but that determination is beyond the K2HPO4, 20 mg N as [NH4]2SO4 per 1 L seawater) [21] in scope of the present study. 600-mL glass beakers for 1 wk. This process was carried out weekly to provide freshly preconditioned seagrass for the MATERIALS AND METHODS amphipods throughout the present study. Chemicals Experimental setup Boscalid was applied using the commercially available Twenty A. compressa individuals were randomly placed 1 product Filan (Nufarm, Australia, 500 g active ingredient by 1 in 600-mL glass beakers containing 400 mL aerated [a.i.]/kg) instead of pure active ingredient to avoid adding ambient seawater with the respective Filan concentrations, with Page 28 of 120 1132 Environ Toxicol Chem 35, 2016 H.T. Vu et al. ambient seawater as controls, and preconditioned seagrass was Protein content was determined using a modified Lowry added to each beaker. Each treatment had 5 replicates. The assay (Bio-Rad DC method), with bovine serum albumin as the experiment ran for 6 wk, using the same conditions as standard [26]. described previously in the section Test species. To account Microbial respiration on seagrass was measured using for microbial and abiotic seagrass loss during the experiment, changes in oxygen concentration followed the method descried an additional replicate per treatment was included without by Carlisle and Clements [27]. Dissolved oxygen was measured amphipods and treated like the other replicates. Every week, with a water quality meter (smartCHEM-LAB, TPS, QLD, surviving amphipods were gently transferred by plastic pipettes Australia) at the beginning and end of a 24-h incubation period. to freshly made seawater medium and fresh preconditioned seagrass. The remaining seagrass was cleaned with deionized Statistical analysis water, dried at 60 8C for 24 h, and weighed to determine the Treatment effects on survival, growth, reproduction, feeding animal’s feeding rate. rate, microbial respiration, and energy reserves (lipid, glycogen, At the end of the experiment, the number of surviving adults and protein content) were analyzed using one-way analysis of and produced juveniles was recorded. Three healthy nongravid variance followed by Dunnet’s pairwise comparisons. Simple females per replicate were randomly selected and frozen at and multiple linear regressions were performed to determine the –20 8C for lipid, glycogen, and protein analysis. The remaining relationship between energy reserves and female amphipod surviving adults were preserved in 70% ethanol for further survival and growth. Statistical analysis was performed using examination using a Leica MS5 microscope with an ocular SPSS Ver 22 (IBM). micrometer. Specimens preserved in ethanol were sexed and head length was measured (from the rostrum tip to the posterior RESULTS margin of the head) [22] to determine growth based on the final size, with the assumption that the mean size of amphipods per Survival replicate was the same at the beginning of the experiment, Survival in the control treatments after 6 wk was 86 1.9 % because the amphipods were the same age. The number of (mean standard error [SE]). Survival decreased with increas- gravid females and the number of embryos produced per gravid ing Filan concentrations (Figure 1). Significant differences were female were recorded. found between survival of amphipods from the control and Filan A second experiment was also set up without the amphipods treatments (F3,16 ¼ 4.631, p ¼ 0.016). Survival of the control to assess the effects of fungicide Filan on microbial respiration was significantly different from survival at 10 mg a.i./L and on the seagrass used to feed the amphipods. Preconditioned 40 mg a.i./L Filan (p ¼ 0.046 and p ¼ 0.034, respectively), seagrass was exposed to the same nominal fungicide concen- whereas approximately 70% of animals survived. trations as used in the main experiment for 1 wk. Each treatment had 5 replicates. After 1 wk, the microbial respiration of the Growth seagrass was measured. The size of both male and female amphipods was reduced with increasing fungicide concentrations (Figure 2). Signifi- Determination of feeding rate cant effects on females were found at all Filan treatments 1 mg Feeding rate was expressed as milligrams seagrass mass a.i./L, 10 mga.i./L,and40mg a.i./L (p ¼ 0.008, p < 0.001, and consumed per amphipod per day calculated as follows [23]: p < 0.001, respectively), and female head lengths at the highest concentration were reduced by 12.6% compared C ¼ðLb K LaÞ=ðN TÞ with those in the control treatments. For males, significant effects were observed at 10 mga.i./Land40mga.i./L where Lb and La are initial and final dry mass of seagrass, (p ¼ 0.012 and p ¼ 0.025, respectively) but not at 1 mg a.i./L respectively; N is the number of surviving amphipods (the dead (p ¼ 0.496), and the highest Filan concentration only reduced organisms could contribute to the seagrass consumption, but we size by 6.8%. did not account for this in the present study because the time of death was not recorded daily), T is the feeding time in days, and K is the leaf change correction factor given by ÀÁ S LCa K ¼ LCb n where LCb and LCa are the initial and final dry mass of seagrass in the control replicates without amphipods, n is the number of replicates. Determination of lipid, glycogen, protein content, and microbial respiration The lipid, glycogen, and protein assays were carried out using a Synergy 2 microplate reader (Biotek Instruments). Lipid and glycogen content were measured following the method described by Van Handel [24,25], using commercial vegetable oil and glucose as the standards and modified for the use of a microplate reader. The volume of solution in each well was Figure 1. Percentage of survival (mean standard error) of Allorchestes m 1 60 L; absorbance is measured at 490 nm for lipid and 625 nm compressa in control and Filan treatments after 6-wk exposure (n ¼ 5). for glycogen. *Significant difference from control (p 0.05). Page 29 of 120 Toxicity of fungicide Filan1 to Allorchestes compressa Environ Toxicol Chem 35, 2016 1133

Figure 2. Head length (mean standard error) of Allorchestes compressa in control and Filan1 treatments after 6-wk exposure. Males are dark bars and females are light bars (n ¼ 5). *Significant difference from control (p 0.05).

Reproduction Filan exposure had strong adverse effects on A. compressa reproduction. Neither gravid females nor offspring were present in any of the Filan treatments. Gravid females were first observed in the control at the beginning of week 4. The average number of offspring per replicate in the control was 7.6 0.68 (mean SE). The average number of offspring per single female was 0.58 0.04 (mean SE). The average number of gravid females per replicate in the control was 1.4 0.51 (mean SE). The average number of embryos per gravid female in the control was 5.14 0.37 (mean SE). Feeding rate Filan exposure had no significant effect on A. compressa feeding rates throughout the 6-wk exposure period (all p > 0.05). As expected, the feeding rates in the control and treatments increased from week 1 to week 6 (Supplemental Data, Figure S1). Energy reserves Energy reserves of the amphipod A. compressa decreased with increasing Filan concentrations (Figure 3). Filan had significant effects on lipid content at all concentrations of 1 mg a.i./L, 10 mg a.i./L, and 40 mg a.i./L (p ¼ 0.001, p ¼ 0.002, and p < 0.001, respectively), but only at the highest concentration of 40 mg a.i./L for glycogen and protein content (p ¼ 0.013 and p ¼ 0.007, respectively). A simple linear regression showed that all 3 types of energy positively correlated to the female size Figure 3. Concentrations of lipid (A), glycogen (B), protein (C), (mean standard error) of Allorchestes compressa in control and Filan1 (Figure 4), with lipid content having the highest correlation treatments after 6-wk exposure (n ¼ 5). *Significant difference from control (R ¼ 0.705), then glycogen content (R ¼ 0.637) and protein (p 0.05). content (R ¼ 0.560). However, a multiple linear regression performed on all 3 types of energy reserves simultaneously showed that only lipid and protein content significantly fi contributed to the predicted model (Supplemental Data, no signi cant difference was seen in microbial respiration Table S1) and had a significant increase in the coefficient of in the seagrass between the control and Filan treatments 2 (F3,16 ¼ 3.125, p ¼ 0.055). determination (R ¼ 0.724, F3,16 ¼ 13.99, p < 0.001). No sig- nificant relationship was seen between energy reserves and survival (all p > 0.05). Deformities Some deformities were observed in the antennae of Microbial respiration amphipods when head length measurements were conducted. Microbial respiration increased with increasing Filan The antennae were either missing (Figure 5B) or shortened concentrations (Supplemental Data, Figure S2). However, (Figure 5B and C). No deformities were observed in the control. Page 30 of 120 1134 Environ Toxicol Chem 35, 2016 H.T. Vu et al.

Figure 5. Normal antennae of Allorchestes compressa (A) and deformed specimens in Filan1 treatment of 40 mg active ingredient/L (B, C) after 6-wk exposure. Two antennae were missing; 1 antenna was shortened (B); all 4 antennae were shortened (C).

the feeding rate. However, the levels of effect were different among endpoints. Figure 4. Relationship between female head length and lipid (A), glycogen At the organism level, reproduction was the most sensitive (B), and protein (C) content. Lipid content was most positively correlated endpoint. No female reproduction occurred in all Filan to female size, then glycogen and protein content. treatments. This finding is in agreement with the results of previous studies on chronic effects of toxicants on marine and estuarine amphipods that showed that reproduction was Malformations only occurred in the 40 mg a.i./L Filan treatment delayed [28] or significantly reduced [29], and was a much in 13 individuals of 68 examined amphipods. more sensitive metric than survival [30]. The significant effects of Filan on A. compressa reproductive success could be partially DISCUSSION explained by the reduction in growth of female amphipods. Body size is a determining factor for the onset of the Effects of boscalid exposure on A. compressa at different reproductive phase of amphipods [31], because they have to endpoints reach a certain size before reproduction can occur [32]. Filan had effects at environmentally relevant concentrations Therefore, reduced growth can lead to reduced reproduction. on almost all endpoints measured in the present study, except The relationship between female size and reproductive output Page 31 of 120 Toxicity of fungicide Filan1 to Allorchestes compressa Environ Toxicol Chem 35, 2016 1135 has been documented for some amphipod species such as metabolized to meet the energy needs of an organism [41,42], Hyalella azteca [33] and Gammarus minus [34]. The effects of and it can be quickly synthesized when carbohydrate supplies Filan on amphipod reproduction at environmentally relevant are available [39]. Animals exposed to Filan at concentrations concentrations should be considered in fungicide risk assess- of 1 mg a.i./L and 10 mg a.i./L might have a chance to replenish ments, because a delay in reproduction could have strong the glycogen they used. The ability of the animal to quickly negative effects on the viability of the population at an refill the used glycogen is supported by Hervant et al. [43], ecological scale. who reported a significant overshoot of the glycogen content Filan exposure also had a significant effect on the growth of in the amphipods Niphargus rhenorhodanensis and Niphargus A. compressa. Growth is routinely used as a sublethal endpoint virei during the first week of recovery from nutritional stress, in chronic toxicity studies, and it is often affected by reaching 127% and 121% of fed value, respectively, before contaminant exposure [35]. A few studies have shown that returning to the prestarvation levels. The results of biochemical female amphipods were more sensitive than males [36,37]. The biomakers suggest that exposure to 40 mg a.i./L Filan caused results of the present study were consistent with previous serious stress to the amphipods and that lipid content is a studies, because we found significant effects on female growth sensitive biomarker that could be used to assess the effects of in all Filan treatments whereas only at the higher concentrations fungicides on amphipods. for males. The difference in sensitivity of growth in sexually mature male and female amphipods may be partially explained The link between biochemical changes and effects at higher by the increase in energy requirements during oogenesis and levels of organization brooding in females compared with the less energy-demanding Biomarkers have been used increasingly to investigate process of spermatogenesis in males [36]. This will result in less environmental impacts of pollutants because of a number of energy being available for growth and to cope with toxic stress advantages compared with conventional toxicity tests, which in females. generally use mortality as an endpoint [44]. Biochemical As expected, the survival endpoint was less sensitive than parameters are very sensitive to sublethal concentrations of reproduction and growth. Significant effects of boscalid many chemicals [40] and are often considered as initial changes fungicide Filan on amphipod survival occurred at concen- caused by toxicants that ultimately lead to adverse effects trations of 10 mg/L and 40 mg/L. To our knowledge, no chronic at higher levels of biological organization [44]. However, toxicity data are available for boscalid on marine invertebrates. currently limited knowledge is available on the ecological A. compressa seems to be more sensitive to boscalid compared relevance of biomarker signals [38], and this could be assessed with other invertebrates. For example, a 21-d chronic exposure through investigating their relationship with several life history to Daphnia magna recorded no observed adverse effects traits [45]. concentration of 3.06 mg/L (J. Jatzek, Experimental Toxicology Our results showed a strong relationship between energy and Ecology, BASF Aktiengesellschaft, Ludwigshafen, reserves and the growth of female A. compressa. Organisms use Germany, unpublished data). This may be attributable to stored energy for a variety of needs, but most energy is used for exposure time, because the present study was a 42-d test growth, reproduction, and basal metabolism [39,40]. Increased compared with 21-d exposure for the for D. magna study. energy expenditure in basal metabolism to cope with toxic stress Longer-term exposure is known to cause a significant effect on will lead to a reduction in growth and reproduction [38]. Similar survival [35]. observations of reduction of growth with the concurrence of At the biochemical level, lipid content was the most decreased energy reserves were found in D. magna exposed to sensitive of the energy stores measured, although all 3 types of Cd [46] and Gammarus pseudolimnaeus exposed to pentachlo- energy reserves decreased with increasing fungicide concen- rophenol [47]. However, the present study further suggests that trations. This suggests that lipids were the primary source of lipid content is the energy most correlated to the growth of energy to cover for the increased demand incurred from Filan female amphipods because it had the highest standardized exposure. The present results concur with those of Zubrod coefficients for the predicted model (Supplemental Data, et al. [18], who observed that the fungicide tebuconazole Table S1). The relationship of lipid and growth in amphipods significantly reduced the lipid content of the amphipod has previously been reported in the literature [31,48]. Further- Gammarus fossarum but had no effects on leaf consumption. more, the present results also demonstrated that lipid and protein However, the authors also pointed out that the fungicide content are both important in amphipod growth, because they tebuconazole could alter the food quality of the amphipod overall significantly increased the coefficient of determination through the effects on the microbial colonization of the leaf for multiple linear regression analysis (R2 ¼ 0.724) compared material. De Coen et al. [38] also reported that lipid reserves with the simple linear regression with only lipid (R2 ¼ 0.498) or was the most sensitive endpoint among all cellular energy protein (R2 ¼ 0.314). allocation components of D. magna exposed to 6 different A strong connection also was seen between lipid content and toxicants. Lipids are often mobilized to meet the increased amphipod reproduction. Lipids are prominent storage compo- energy demand associated with toxic stress because lipid is nents in most marine invertebrates [39]. Therefore, not only a prominent long-term energy store in most aquatic crusta- are they an important energy source for growth they but also ceans [31], and they provide more than twice as much play a critical role in amphipod reproductive success, because potential metabolic energy per unit mass as proteins or lipids are used in the development of reproductive tissue carbohydrates [39]. and embryos [31,48]. Studies on amphipod reproduction Protein and glycogen contents were less sensitive than lipid have shown that lipid content correlated to egg production [31] content; significant effects were only observed at the highest and increased during the reproductive period [49,50]. In the concentration. Proteins are often used by animals during periods present study, a significant reduction of lipid content and of high energy demand [40], and they are the last energy sources lack of reproductive output in all Filan treatments convincingly to be mobilized in stressed organisms after the metabolization of demonstrated the important role of lipid reserves in the lipid and carbohydrates [41]. In contrast, glycogen is rapidly reproductive success of A. compressa. Page 32 of 120 1136 Environ Toxicol Chem 35, 2016 H.T. Vu et al.

Finally, protein content could be an explanation for the 5. Smalling KL, Kuivila KM, Orlando JL, Phillips BM, Anderson BS, morphological abnormalities of A. compressa. The mechanism Siegler K, Hunt JW, Hamilton M. 2013. Environmental fate of underlying the occurrence of deformities in aquatic inverte- fungicides and other current-use pesticides in a central California estuary. Mar Pollut Bull 73:144–153. brates exposed to contaminants and the consequences of 6. Maltby L, Brock TCM, van den Brink PJ. 2009. Fungicide risk deformities to these organisms remain unclear [51,52]. In assessment for aquatic ecosystems: Importance of interspecific the present study, we observed antennal deformities in the variation, toxic mode of action, and exposure regime. Environ Sci amphipod A. compressa exposed to 40 mg a.i./L of Filan at Technol 43:7556–7563. fi 7. Sharp S, Myers J, Pettigrove V. 2013. An assessment of sediment which a signi cant reduction of protein content also occurred. toxicants in Western Port and major tributaries. CAPIM Technical We propose the possibility of a relationship between protein Report No 27, Melbourne Water. Centre for Aquatic Pollution content and deformities, because a large proportion of an Identification and Management (CAPIM), School of Biosciences, organism’s body is composed of structural proteins [41], and the The University of Melbourne, Victoria, Australia. decrease in protein content might be attributable to a mechanical 8. Betancourt-Lozano M, Baird DJ, Sangha RS, Gonzalez-Farias F. 2006. Induction of morphological deformities and moulting alterations in lipoprotein formation that will be used to repair damaged cells, Litopenaeus vannamei (Boone) juveniles exposed to the triazole- tissues, and organs [40]. This proposal was supported by David derivative fungicide tilt. Arch Environ Contam Toxicol 51:69–78. et al. [53], who reported a significant reduction in total, soluble, 9. Markey KL, Baird AH, Humphrey C, Negri AP. 2007. Insecticides and and structural proteins in deformed tadpoles of Duttaphrynus a fungicide affect multiple coral life stages. Mar Ecol Prog Ser 330:127–137. melanostictus exposed to sublethal concentrations of cyper- 10. JanakiDevi V, Nagarani N, YokeshBabu M, Kumaraguru AK, methrin. On the contrary, Arambourou et al. [54] reported Ramakritinan CM. 2013. A study of proteotoxicity and genotoxicity alterations in energy reserves of C. riparius exposed to lead- induced by the pesticide and fungicide on marine invertebrate (Donax spiked sediment, but no mentum morphological defects were faba). Chemosphere 90:1158–1166. observed. 11. Elskus AA. 2012. Toxicity, sublethal effects, and potential modes of action of select fungicides on freshwater fish and invertebrates. In: US Geological Survey Open-File Report 2012 –2013. US Department of the CONCLUSION Interior, Washington, DC. p 44. fi 12. Schafer RB, Pettigrove V, Rose G, Allinson G, Wightwick A, von der The present study provides the rst evidence of the effects of a Ohe PC, Shimeta J, Kuhne R, Kefford B. 2011. Effects of pesticides fungicide on survival, growth, reproduction, and energy reserves monitored with three sampling methods in 24 sites on macro- on a marine amphipod at environmentally detected concen- invertebrates and microorganisms. Environ Sci Technol 45:1665–1672. trations. Although these effects were observed under laboratory 13. Burridge TR, Lavery T, Lam PKS. 1995. Acute toxicity tests using conditions, the results suggest that fungicides could affect the Phyllospora-comosa (Labillardiere) Agardh, C. (Phaeophyta, Fucales) and Allochestes compressa Dana (Crustacea, Amphipoda). Bull viability of amphipod populations in natural ecosystems. There Environ Contam Toxicol 55:621–628. also could be cascading effects on the ecosystem, because 14. Crawley KR, Hyndes GA. 2007. The role of different types of detached semiaquatic amphipods such as A. compressa are often a main macrophytes in the food and habitat choice of a surf-zone inhabiting food source for many fish species and have major influence on amphipod. Mar Biol 151:1433–1443. 15. Ahsanullah M, Florence TM. 1984. Toxicity of copper to the marine detrital turnover in surf-zone environments. Further research amphipod Allorchestes compressa in the presence of water-and lipid- needs to address the long-term effects of fungicides on aquatic soluble ligands. Mar Biol 84:41–45. ecosystems and to assess the condition of A. compressa 16. Ahsanullah M, Williams AR. 1986. Effect of uranium on growth and populations in intertidal zones polluted with fungicides. reproduction of the marine amphipod Allorchestes compressa. Mar Biol 93:459–464. 17. Bundschuh M, Zubrod JP, Kosol S, Maltby L, Stang C, Duester L, Supplemental Data—The Supplemental Data are available on the Wiley Schulz R. 2011. Fungal composition on leaves explains pollutant- Online Library at DOI: 10.1002/etc.3247. mediated indirect effects on amphipod feeding. Aquat Toxicol 104:32–37. — Acknowledgement We thank J. Myers, R. Reid, and D. McMahon for 18. Zubrod JP, Bundschuh M, Feckler A, Englert D, Schulz R. 2011. fi kindly assisting in the eld collection of amphipods. We also thank Ecotoxicological impact of the fungicide tebuconazole on an aquatic ’ A. O Brien for providing comments and helping improve the manuscript. decomposer-detritivore system. Environ Toxicol Chem 30:2718–2724. H. T. Vu was funded Melbourne International Research Scholarship 19. Greig ME, Hoogerheide JC. 1941. The correlation of bacterial growth during the conduct of the present study. Funding for the present study with oxygen consumption. J Bacteriol 41:549–556. fi was supported by the Centre for Aquatic Pollution Identi cation and 20. Victoria F, Supply AM. 2014. Triennial report for the Clifton Springs Management (CAPIM). Aquaculture Fisheries Reserves, Fisheries Victoria (DPI) and Advance Mussel Supply Pty. Victoria State Government, Melbourne, Victoria, Data availability—Data can be assessed by contacting corresponding author Australia. ([email protected]). 21. Bird GA, Kaushik NK. 1985. Processing of elm and maple leaf discs by collectors and shredders in laboratory feeding studies. Hydrobiologia REFERENCES 126:109–120. 22. Gonzalez MJ, Burkart GA. 2004. Effects of food type, habitat, and 1. Fisher MC, Henk DA, Briggs CJ, Brownstein JS, Madoff LC, McCraw fish predation on the relative abundance of two amphipod species, SL, Gurr SJ. 2012. Emerging fungal threats to animal, plant and Gammarus fasciatus and Echinogammarus ischnus. J Great Lakes Res ecosystem health. Nature 484:186–194. 30:100–113. 2. Reilly TJ, Smalling KL, Orlando JL, Kuivila KM. 2012. Occurrence of 23. Maltby L, Clayton SA, Wood RM, McLoughlin N. 2002. Evaluation of boscalid and other selected fungicides in surface water and groundwater the Gammarus pulex in situ feeding assay as a biomonitor of water in three targeted use areas in the United States. Chemosphere quality: Robustness, responsiveness, and relevance. Environ Toxicol 89:228–234. Chem 21:361–368. 3. Wightwick AM, Bui AD, Zhang P, Rose G, Allinson M, Myers JH, 24. Van Handel E. 1985. Rapid determination of total lipids in mosquitos. Reichman SM, Menzies NW, Pettigrove V, Allinson G. 2012. J Am Mosq Control Assoc 1:302–304. Environmental fate of fungicides in surface waters of a horticultural- 25. Van Handel E. 1985. Rapid determination of glycogen and sugars in production catchment in southeastern Australia. Arch Environ Contam mosquitoes. J Am Mosq Control Assoc 1:299–301. Toxicol 62:380–390. 26. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein 4. Smalling KL, Orlando JL. 2011. Occurrence of pesticides in water and measurement with the folin phenol reagent. J Biol Chem 193:265–275. sediment from three central California coastal watersheds, 2008–2009. 27. Carlisle DM, Clements WH. 2005. Leaf litter breakdown, microbial US Geological Survey Data Series 600. US Department of the Interior, respiration and shredder production in metal-polluted streams. Freshw Washington, DC. p 70. Biol 50:380–390. Page 33 of 120 Toxicity of fungicide Filan1 to Allorchestes compressa Environ Toxicol Chem 35, 2016 1137

28. Ringenary MJ, Molof AH, Tanacredi JT, Schreibman MP, Kostarelos 42. Gismondi E, Beisel J-N, Cossu-Leguille C. 2012. Influence of gender K. 2007. Long-term sediment bioassay of lead toxicity in two and season on reduced glutathione concentration and energy reserves of generations of the marine amphipod Elasmopus laevis, SI Smith, Gammarus roeseli. Environ Res 118:47–52. 1873. Environ Toxicol Chem 26:1700–1710. 43. Hervant F, Mathieu J, Barre H. 1999. Comparative study on the 29. Green A, Moore D, Farrar D. 1999. Chronic toxicity of 2,4,6- metabolic responses of subterranean and surface-dwelling amphipods trinitrotoluene to a marine polychaete and an estuarine amphipod. to long-term starvation and subsequent refeeding. J Exp Biol 202: Environ Toxicol Chem 18:1783–1790. 3587–3595. 30. Van den Heuvel-Greve M, Postina J, Jol J, Kooman H, Dubbeldam M, 44. De Coen WM, Janssen CR, Giesy JP. 2000. Biomarker Applications in Schipper C, Kater B. 2007. A chronic bioassay with the estuarine Ecotoxicology: Bridging the Gap Between Toxicology and Ecology. amphipod Corophium volutator: Test method description and Kluwer Academic, Dordrecht, The Netherlands. confounding factors. Chemosphere 66:1301–1309. 45. Campero M, Ollevier F, Stoks R. 2007. Ecological relevance and 31. Lehtonen KK. 1996. Ecophysiology of the benthic amphipod sensitivity depending on the exposure time for two biomarkers. Environ Monoporeia affinis in an open-sea area of the northern Baltic Sea: Toxicol 22:572–581. Seasonal variations in body composition, with bioenergetic consider- 46. Knowles CO, McKee MJ. 1987. Protein and nucleic acid content in ations. Mar Ecol Prog Ser 143:87–98. Daphnia magna during chronic exposure to cadmium. Ecotoxicol 32. Conradi M, Depledge MH. 1999. Effects of zinc on the life-cycle, Environ Saf 13:290–300. growth and reproduction of the marine amphipod Corophium volutator. 47. Graney RL, Giesy JP. 1986. Effects of long-term exposure to Mar Ecol Prog Ser 176:131–138. pentachlorophenol on the free amino acid pool and energy reserves 33. Kubitz JA, Besser JM, Giesy JP. 1996. A two-step experimental design of the freshwater amphipod Gammarus pseudolimnaeus Bousfield for a sediment bioassay using growth of the amphipod Hyalella azteca (Crustacea, Amphipoda). Ecotoxicol Environ Saf 12:233–251. for the test endpoint. Environ Toxicol Chem 15:1783–1792. 48. Nalepa TF, Hartson DJ, Buchanan J, Cavaletto JF, Lang GA, Lozano 34. Glazier DS. 2000. Is fatter fitter? Body storage and reproduction in ten SJ. 2000. Spatial variation in density, mean size and physiological populations of the freshwater amphipod Gammarus minus. Oecologia condition of the holarctic amphipod Diporeia spp. in Lake Michigan. 122:335–345. Freshw Biol 43:107–119. 35. Prato E, Parlapiano I, Biandolino F. 2013. Sublethal effects of copper 49. Clarke A, Skadsheim A, Holmes LJ. 1985. Lipid biochemistry and on some biological traits of the amphipod Gammarus aequicauda reproductive biology in two species of Gammarida (Crustacea: reared under laboratory conditions. Chemosphere 93:1015–1022. Amphipoda). Mar Biol 88:247–263. 36. McCahon CP, Pascoe D. 1988. Increased sensitivity to cadmium of the 50. Sundelin B, Rosa R, Wiklund AKE. 2008. Reproduction disorders in freshwater amphipod Gammarus pulex (L.) during the reproductive the benthic amphipod Monoporeia affinis: An effect of low food period. Aquat Toxicol 13:183–194. resources. Aquat Biol 2:179–190. 37. Sornom P, Felten V, Medoc V, Sroda S, Rousselle P, Beisel JN. 2010. 51. Beguer M, Pasquaud S, Noel P, Girardin M, Boet P. 2008. First Effect of gender on physiological and behavioural responses of description of heavy skeletal deformations in Palaemon shrimp Gammarus roeseli (Crustacea Amphipoda) to salinity and temperature. populations of European estuaries: The case of the Gironde (France). Environ Pollut 158:1288–1295. Hydrobiologia 607:225–229. 38. De Coen WM, Janssen CR. 2003. The missing biomarker link: 52. Meregalli G, Bettinetti R, Pluymers L, Vermeulen AC, Rossaro B, Relationships between effects on the cellular energy allocation Ollevier F. 2002. Mouthpart deformities and nucleolus activity in field- biomarker of toxicant-stressed Daphnia magna and corresponding collected Chironomus riparius larvae. Arch Environ Contam Toxicol population characteristics. Environ Toxicol Chem 22:1632–1641. 42:405–409. 39. Hadley NE. 1986. The Adaptive Role of Lipids in Biological Systems. 53. David M, Marigoudar SR, Patil VK, Halappa R. 2012. Behavioral, Wiley, New York. morphological deformities and biomarkers of oxidative damage as 40. Sancho E, Villarroel MJ, Andreu E, Ferrando MD. 2009. Disturbances indicators of sublethal cypermethrin intoxication on the tadpoles of in energy metabolism of Daphnia magna after exposure to tebucona- D. melanostictus (Schneider, 1799). Pestic Biochem Physiol 103:127–134. zole. Chemosphere 74:1171–1178. 54. Arambourou H, Gismondi E, Branchu P, Beisel JN. 2013. Biochemical 41. Huggett RJ. 1992. Biomarkers: Biochemical Physiological and and morphological responses in Chironomus riparius (Diptera, Histological Markers of Anthropogenic Stress. Lewis Publishers Chironomidae) larvae exposed to lead-spiked sediment. Environ (CRC Press), Boca Raton, FL, USA. Toxicol Chem 32:2558–2564.

Page 34 of 120 Supporting document

Effects of the boscalid fungicide Filan® on the marine amphipod Allorchestes compressa at environmentally relevant concentrations

Hung T. Vu†* , Michael J. Keough† , Sara M. Long†, Vincent J. Pettigrove †

†Centre for Aquatic Pollution Identification and Management, School of Biosciences, The

University of Melbourne, Victoria, 3010, Australia.

*Corresponding Author: [email protected]

Phone: + 61 3 8344 4331

Fax: +61 3 8344 7909

Page 35 of 120

Table S1. Coefficients of female head length and energy reserves of multi regression analysis

Model Unstandardized coefficients Standardized coefficients t Sig

Constant .389 27.562 .000

Lipid .002 .511 3.413 .004

Glycogen .001 .254 1.598 .130

Protein .035 .361 2.538 .022

Page 36 of 120 0.12

0.1

0.08

Control 0.06 1 µg a.i./L 10 µg a.i./L 0.04 40 µg a.i./L

0.02 Feeding rate (mg seagrass/amphipod/day) 0 1 2 3 4 5 6 Weeks

Figure S1. Feeding rate (mean ± SE) of Allorchestes compressa in control and Filan® treatments during six weeks, (n = 5), (*) denotes significant difference from control (p ≤ 0.05)

0.7 0.6

0.5 0.4 0.3 0.2 (mg O2/g DW/h) 0.1 0.0 Microbial oxygen consumption Control 1 10 40 Filan® concentration (µg a.i./L)

Figure S2. Microbial respiration (mean ± SE) in control and Filan® treatments after one week, (n = 5), (*) denotes significant difference from control (p ≤ 0.05)

Page 37 of 120 CHAPTER 3: EFFECTS OF TWO COMMONLY USED FUNGICIDES ON THE AMPHIPOD AUSTROCHILTONIA SUBTENUIS

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2017. Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry 36:720-726.

Page 38 of 120 Page 39 of 120 Page 40 of 120 Page 41 of 120 Page 42 of 120 Page 43 of 120 Page 44 of 120 Page 45 of 120 SUPPLEMENTAL DATA

Table S1. Nominal and measured concentrations of Filan® and Systhane™ (µg a.i./L)

Filan® Systhane™

Nominal Measured Nominal Measured concentration concentration* concentration concentration*

0 < 0.1 0 < 0.1

1 1.4 0.3 0.5 10 10 3 3.7 40 48 30 39

*Limit of detection (LOD) = 0.1 µg/L, reported values were based on one replicate.

A 110 Control 100 1 µg a.i./L 90 10 µg a.i./L 80 40 µg a.i./L 70

Survival Survival (%) 60 50 40 0 1 2 3 4 5 6 7 8 Time (week)

B 110 Control 100 0.3 µg a.i./L 90 3 µg a.i./L 80 70 30 µg a.i./L

Survival Survival (%) 60 50 40 0 1 2 3 4 5 6 7 8 Time (week)

Page 46 of 120 Figure S1. Mean cumulative survival of Austrochiltonia subtenuis changed over time after exposure to Filan® (A) and Systhane™ (B) for up to 8 wk.

.30 .30

.25 .25

.20 .20

(mg/gDW/h) .15 (mg/gDW/h) .15 Oxygen Oxygen consumption Oxygen consumption .10 .10 Control 1 10 40 Control 0.3 3 30 Filan® concentration (µg a.i./L) Systhane™ concentration (µg a.i./L)

Figure S2. Microbial respiration (mean ± SE) in control, Filan®, and Systhane™ treatments after 7 d exposure, (n = 6), (*) denotes significant difference from control (p ≤ 0.05).

Page 47 of 120 CHAPTER 4: TOXICOLOGICAL EFFECTS OF FUNGICIDE MIXTURES ON THE AMPHIPOD AUSTROCHILTONIA SUBTENUIS

Hung T. Vu, Michael J. Keough, Sara M. Long, and Vincent J. Pettigrove. 2017. Toxicological effects of fungicide mixtures on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry. DOI: 10.1002/etc.3809.

Page 48 of 120 Environmental Toxicology and Chemistry, Vol. 9999, No. 9999, pp. 1–9, 2017 # 2017 SETAC Printed in the USA

Environmental Toxicology

TOXICOLOGICAL EFFECTS OF FUNGICIDE MIXTURES ON THE AMPHIPOD AUSTROCHILTONIA SUBTENUIS

HUNG T. VU,* MICHAEL J. KEOUGH,SARA M. LONG, and VINCENT J. PETTIGROVE Centre for Aquatic Pollution Identification and Management, School of BioSciences, The University of Melbourne, Victoria, Australia

(Submitted 5 January 2017; Returned for Revision 23 February 2017; Accepted 24 March 2017)

Abstract: Approaches to assess the toxicity of mixtures often use predictive models with acute mortality as an endpoint at relatively high concentrations. However, these approaches do not reflect realistic situations where organisms could be exposed to chemical mixtures over long periods at low concentrations at which no significant mortalities occur. The present study investigated chronic effects of 2 common fungicides, Filan1 (active ingredient [a.i]) boscalid) and SysthaneTM (a.i. myclobutanil), on the amphipod Austrochiltonia subtenuis at environmentally relevant concentrations under laboratory conditions. Sexually mature amphipods were exposed singly and in combination to Filan (1, 10, and 40 mg a.i./L) and Systhane (3 mg a.i./L) over 28 d. Survival, growth, a wide range of reproduction endpoints, and glutathione-S-transferase (GST) activity were measured at the end of the experiment. Both fungicides had significant independent effects on male growth, sex ratio, and juvenile size. Filan mainly affected female growth and the number of embryos per gravid female, whereas Systhane mainly affected the time for females to become gravid. The combined effects of these fungicides on numbers of gravid females and juveniles were antagonistic, causing a 61% reduction in the number of gravid females and a 77% reduction in the number of juveniles produced at the highest concentrations (40 mg a.i./L of boscalid and 3 mg a.i./L of myclobutanil) compared with the controls. There were no significant effects on survival or GST activity. The present study demonstrated that the effects of mixtures were endpoint dependent and that using a variety of endpoints should be considered for a comprehensive understanding of mixture effects. Also, chronic studies are more informative than acute studies for environmentally relevant fungicide concentrations. Environ Toxicol Chem 2017;9999:1–9. # 2017 SETAC

Keywords: Mixture toxicology Pesticides Aquatic invertebrates Reproductive toxicity

INTRODUCTION concentrations [5]. However, in reality, organisms are typically Fungicides are a group of pesticides that are widely used in exposed to mixtures of chemicals over long periods [9] at fi agriculture to protect plants from fungal infection. They have relatively low concentrations [5]. The main dif culties in become an important component of plant disease management studying chronic mixture toxicity are the complex nature of plans for agronomic crops, because fungal diseases have the toxic mechanisms, as well as time and resource limitations [6,9]. potential to destroy crops, rendering them unsaleable [1]. Most At present, relatively few studies have observed the effects of modern fungicides have a single-site mode of action because chronic mixtures on aquatic organisms compared with acute – this is associated with lower potential for negative impact on the mixture effects [9 12]. These studies have demonstrated that environment, including nontarget organisms [2]. However, this the toxicity of mixtures varies with duration of exposure, and the can lead to greater fungal resistance because just a single gene chronic mixture effects could not be predicted from acute mutation can alter a target site and reduce the vulnerability of mixture effects or chronic effects of individual chemicals. fungi to the fungicide [3]. To resolve this problem, one common Synergistic effects (toxicity of the mixture is greater than fi and effective strategy used in pesticide resistance management predicted [5]) are a speci c concern in joint toxicity studies programs is to apply mixtures of fungicides with different because of the potential of individual contaminants to increase modes of action [4]. As a result, fungicides with different modes the toxicity in combination [6]. In the literature, many studies of action are often detected simultaneously in agricultural areas. were conducted to assess and predict mixture effects and to The interaction effects of chemical mixtures are of great identify the chemical groups that have a high potential of concern to both the public and regulatory authorities [5], leading causing synergistic effects [5]. The azole fungicides constitute to numerous studies of chemical mixtures over the past few one pesticide group over-represented in the synergistic decades [6]. Typical approaches to assess the toxicity of mixtures [5]. They are known to interfere with a broad range mixtures often use predictive models, such as concentration of cytochrome P450 monooxygenases that are present in almost addition and independent action, with acute mortality as an all living cells and are enzymes responsible for the phase I endpoint [6,7]. These approaches are mathematical rather than biotransformation of lipophilic compounds [5]. Hence the biological in nature [8], and cannot explain observed toxicity of lipophilic compounds is often substantially enhanced interactions nor explain why mixture effects can change in when in the presence of azole fungicides [5]. Azoles have been time and between endpoints [7]. The concentrations used in used extensively in agriculture not only for preventing fungal these approaches often exceed environmentally relevant infection but also for fungal treatment [13]. One main reason for their widespread use in agriculture is their long-lasting stability [14] with half-lives in soil ranging from a month to This article includes online-only Supplemental Data. more than a year [15,16]. Consequently, they have been * Address correspondence to: [email protected] Published online 28 March 2017 in Wiley Online Library frequently detected in natural environments at levels of low (wileyonlinelibrary.com). nanograms to several micrograms per liter [17–19]. A few DOI: 10.1002/etc.3809 studies have investigated the joint toxicity of azoles fungicides

1 Page 49 of 120 2 Environ Toxicol Chem 9999, 2017 H.T. Vu et al. with other pesticides [20–22] and have reported synergistic boscalid and myclobutanil are persistent in aquatic environ- effects on aquatic organisms on different endpoints. Azole ments [15], their concentrations would not substantially change fungicides have been shown to enhance the effect of a during the experiment. Samples from each treatment of the first pyrethroid insecticide, alpha-cypermethrin, toward Daphnia week were sent to the School of Chemistry at The University of magna in the immobilization test up to 12-fold (prochloraz) [20] Melbourne (VIC, Australia) for analysis of boscalid and or 13-fold (propiconazole) [22]. Zubrod et al. [21] observed a myclobutanil concentration by gas chromatrography–mass synergistic effect on the feeding of Gammarus fossarum (35% spectrometry. Measured concentrations were within 30% of deviation between predicted and observed effect) after exposure nominal concentrations (Supplemental Data, Table S1). In to a mixture of 5 fungicides including an azole fungicide, the present study, reported concentrations are nominal tebuconazole, at a total concentration of 160 mg/L for 7 d. concentrations. However, these studies are based on acute experiments or non- environmentally relevant concentrations. To our knowledge, Preconditioned leaves there is a lack of chronic studies that directly measure toxicity of Green hazel (Pomaderris aspera) leaves were picked from azole mixtures at environmentally relevant concentrations. trees in the Royal Botanic Gardens Melbourne, Victoria, Myclobutanil is an azole fungicide widely used in agriculture Australia. Leaves were cut into leaf discs (diameter 1.3 cm) by because of its broad spectrum of antifungal activity that is hole punch, air-dried, and stored at room temperature until use. effective against a wide range of fungal infections in crops, Before the experiment, leaf discs were preconditioned for 2 wk seeds, fruits, and horticultural production systems [14]. Bo- in a 2-L beaker containing 1 L of nutrient-enriched stream water scalid is another commonly used fungicide in agriculture (5 mg P as K2HPO4, 20 mg N as (NH4)2SO4) [30]. This process because it is a systemic fungicide that is active against a broad was carried out weekly to provide freshly preconditioned leaves range of fungal pathogens [23]. Both fungicides are quite stable for the amphipods throughout the duration of the present study. in aquatic environments [15] and have been detected frequently in water and sediment in agricultural watersheds with very high Test species detection rates in different areas of the world [17–19,24–27]. Austrochiltonia subtenuis and water used in the experiment Boscalid and myclobutanil have also often been found to were collected from a nonpolluted stream, Deep Creek, Victoria, co-occur in streams [17–19]. The adverse individual effects of Australia. Amphipods were maintained in the laboratory at boscalid or mycobutanil on aquatic invertebrates have been experimental conditions: temperature of 21 1 8C and a 16:8-h described [23,28], but no available studies have assessed the light:dark photoperiod in 5-L glass aquaria with site water joint effect of these fungicides. under constant aeration. Organisms were fed preconditioned The present study investigated the chronic effects of the hazel leaves ad libitum and 6 mg ground TetraMin fish food/L 3 1 fungicides Filan (active ingredient [a.i.] boscalid) and times/wk. To obtain a known age of amphipods for the SysthaneTM (a.i. myclobutanil), singly and in combination, on experiment, gravid females were separated after 2 wk of a freshwater amphipod Austrochiltonia subtenuis, at environ- acclimatization into clean 2-L glass beakers. One week later, mentally relevant concentrations. The first objective was to the resulting juveniles were transferred to new 2-L glass beakers investigate the long-term interaction effects of Filan and and maintained as described previously. The sex of <7-wk-old Systhane on mature A. subtenuis at environmentally realistic amphipods was determined under the microscope, with male concentrations using a wide range of endpoints that span amphipods distinguished from females by the second enlarged different levels of biological organization. Endpoints assessed gnathopods [31]. Following this, amphipods were maintained were organism-level responses (survival), physiological re- separately for another week to recover from identification stress sponses (reproduction and growth), and suborganism level before being used in the experiment. The collected site water was responses (using glutathione-S-transferase [GST]) to look at kept at 4 8C and brought to room temperature prior to use in the sensitivity in response to low fungicide concentrations. The experiment. second objective was to evaluate how the results of mixture studies vary between endpoints to propose suitable endpoints Experimental setup for mixture toxicity studies. The present study was a 4 (Filan fungicide: 0, 1, 10, and 40 mg a.i./L) by 2 (Systhane fungicide: 0 and 3 mg a.i./L) MATERIALS AND METHODS factorial design. Fourteen A. subtenuis individuals (<8 wk old; 7 males and 7 females) were placed randomly in 600-mL glass Chemicals beakers containing 400 mL of fungicide-dosed aerated stream The commercially available product Filan fungicide water. Each beaker contained 2 leaf discs (diameter 1.3 cm) as a (Nufarm), containing 500 g a.i./kg, was used for boscalid food source and a 5 5-cm presoaked cotton gauze as a exposures, and Systhane 400 WP fungicide (Dow Agro- substrate for the amphipods. Ground TetraMin fish food was Sciences), containing 400 g a.i./kg, was used for mycobutanil added 3 times/wk as 2.5 mg/beaker to provide additional food. exposures. Filan and Systhane were dissolved in deionized Each treatment had 4 replicates. The experiment was run for 28 water to make a stock solution with a nominal concentration of d using the same conditions as described in the Test species 50 mg a.i./L. The stock solution was diluted in stream water section. Every week, surviving amphipods were gently to achieve nominal boscalid concentrations of 1, 10, and transferred by plastic pipette to freshly prepared test medium 40 mg a.i./L and mycobutanil concentration of 3 mg a.i./L. The with fresh preconditioned leaves. Numbers of gravid females concentrations of boscalid and myclobutanil were based on the were recorded. Juveniles were counted, removed, and preserved concentrations detected in the natural environment, which in 70% ethanol for later size analysis based on head length (from ranged from 2.9 to 36 mg/L [24,26,29]. Both stock and test the rostrum tip to the posterior margin of the head) [32]. media were prepared immediately prior to the initiation of At the end of the experiment, one nongravid female and testing and before water changes. Water samples were collected one male were randomly selected from each replicate and frozen before the experiment and at each water change. Because at –80 8C for GST analysis. The remaining surviving adults were Page 50 of 120 Endpoint dependence of chronic mixture effects Environ Toxicol Chem 9999, 2017 3 preserved in 70% ethanol for further examination using a Leica MS5 microscope with an ocular micrometer. Specimens preserved in ethanol were sexed, and head length was measured to determine growth based on the final size, with the assumption that the mean size of amphipods per replicate was the same at the beginning of the experiment, because the amphipods were the same age. The number of gravid females and the number of embryos produced per gravid female were recorded. Embryo development stages were identified and followed Mann et al. [33]. GST analysis The activity of GST was determined using 1-chloro-2, 4-dinitrobenzene (CDNB) as substrate as described by Habig et al. [34] and Long et al. [35] using a Synergy 2 microplate reader (Biotek Instruments). Briefly, individual amphipods (4 males and 4 females/treatment) were homogenized in 60 mL (for females) or 80 mL(formales)of0.1Mphosphate buffer pH 6.5 (containing 1.4 mM 1,4-dithioerythritol and 1 mM ethylenediamine tetraacetic acid and 20% v/v glycerol). The homogenate was centrifuged at 4 8C and 10 000 rpm for 10 min. Activity of GST was determined following the conjuga- tion of reduced glutathione (GSH) and CDNB at 340 nm, using a mM extinction coefficient of 6.72 (adjusted for the path length of the microplate reader), and an increase in absorbance over time was observed. The reaction buffer contained 0.1 M KH2PO4 (pH 6.5), 3 mM GSH, and 1 mM CDNB. The final volume in each well was 200 mLwith5mL of supernatant. Supernatant was used to analyze protein content using a modified Lowry assay (Bio-Rad DC method) with bovine serum albumin as the standard [35]. For Figure 1. Percentage of survival (mean standard error) of Austrochiltonia all assays, each sample was analyzed in triplicate. Results are subtenuis in control and fungicide treatments after 7 (A) and 28 (B)dof exposure, n ¼ 4. Blue bars are Filan-only treatments, and red bars are expressed as nmol GST activity/min/mg protein. mixtures of Filan and Systhane. Statistical analysis Two-way analysis of variance was used to determine amphipods indicates that older life stages may be less sensitive interaction and independent effects of Filan and Systhane. All than younger life stages in terms of survival. Mortality is a data were checked for normality using a Shapiro–Wilk test and typical endpoint in many mixture studies [6,9], but sublethal homogeneity of variance using Levene’s test. If there were endpoints such as reproduction and growth should be included significant interaction effects, pairwise comparisons were in mixture studies because they are more sensitive than performed to determine the simple effects of Systhane at each mortality and are important for assessing the effects of toxicant Filan concentration, and the synergistic effect was analyzed mixtures on population fitness in natural environments [5]. based on the interaction between 2 trends of the means of the Sex-specific survival. The sex ratio (female:male) of mature mixtures and Filan treatments. If there is a synergism, these amphipods at the end of the experiment was altered by the trends will diverge [36]. Statistical analysis was performed fungicide treatments, with relatively fewer females in treat- using SPSS Ver 23 (IBM). ments than in the controls (Figure 2). There was no interaction RESULTS AND DISCUSSION effect between Filan and Systhane on sex ratio, but individually

Survival Total survival. No significant differences in mortality were observed between treatments and the control after 7 or 28 d (Figure 1). Our previous work showed that both Filan and Systhane significantly reduced the survival of A. subtenuis at 10 and 0.3 mg a.i./L after 7 d of exposure, respectively [28]. In the present study, however, no significant effects were observed at even higher concentrations (40 mg a.i./L of Filan and 3 mg a.i./L of Systhane), both singly and in combination. The main reason for this difference is likely differences in animal age. In previous work, we used juvenile amphipods that were <2 wk old; in the present study, we used mature amphipods that were <8 wk old. Studies in the literature have shown that the response of organisms to toxicants can depend on their life stages [37,38], and juvenile organisms are generally more sensitive to Figure 2. Sex ratio (mean standard error) of Austrochiltonia subtenuis in fungicides than adults [38]. Our previous work with <2-wk- control and fungicide treatments after 28 d of exposure, n ¼ 4. Blue bars are old amphipods together with the present study on <8-wk-old Filan-only treatments, and red bars are mixtures of Filan and Systhane. Page 51 of 120 4 Environ Toxicol Chem 9999, 2017 H.T. Vu et al.

Table 1. Individual and interactive effects of Filan (F) and Systhane (S) on the amphipod Austrochiltonia subtenuis after 28 d of exposure tested with a two-way analysis of variancea

Dependent variable Factor df F p

28-d survival percentage F 3 1.514 0.236 S 1 0.086 0.772 F S 24 0.257 0.855 Sex ratio F 3 5.075 0.007* S 1 8.333 0.008* F S 24 1.382 0.272 Male head length F 3 3.521 0.030* S 1 33.338 <0.001* F S 24 2.216 0.112 Female head length F 3 6.321 0.003* S 1 2.278 0.145 F S 21 2.261 0.108 Juvenile head length F 3 14.262 <0.001* S 1 8.663 0.007* F S 24 2.196 0.115 Time to become gravid F 3 2.867 0.058 S 1 16.200 <0.001* F S 24 0.733 0.542 No. of gravid female F 3 7.707 0.001* S 1 9.366 0.005* F S 24 7.707 0.001* Embryos/female F 3 3.048 0.048* S 1 0.962 0.336 F S 24 0.635 0.600 No. of juveniles F 3 47.813 <0.001* S 1 77.625 <0.001* F S 24 12.060 <0.001* GST activity in male F 3 0.748 0.539 S 1 3.848 0.067 F S 16 1.012 0.413 GST activity in female F 3 0.539 0.667 S 1 0.002 0.964 F S 14 0.250 0.860 aThe degrees of freedom, F values, and p values are shown. *Significant at p < 0.05. GST ¼ glutathione-S-transferase. each fungicide had strong main effects on sex ratio (Table 1). The sex ratio decreased by approximately 23% at 3 mg a.i./L of Systhane compared with the control. Filan also reduced the sex ratio, but significant effects were only observed at low Figure 3. Head length (mean standard error) of males (A), females (B), concentrations of 1 and 10 mg a.i./L, at which the sex ratio and juveniles (C)ofAustrochiltonia subtenuis in control and fungicide ¼ was reduced by approximately 39% and 22%, respectively. treatments after 28 d of exposure, n 4. Blue bars are Filan-only treatments, and red bars are mixtures of Filan and Systhane. The sex ratio data in the present study showed that mature female amphipods were more sensitive to fungicides than males. Sensitivity of female amphipods was also observed in the study by Conradi and Depledge [39], who reported that overall age variation when using mature amphipods. However, the survival of the amphipod Corophium volutator was unaffected sensitivity of mature female amphipods should be considered but female amphipod survival was significantly decreased when in toxicity studies because a reduction in mature females mature amphipods were exposed to zinc (0, 0.2, 0.4, 0.6, and could severely impact the population structure in natural 0.8 mg/L) for 45 d. McCahon and Pascoe [40] also reported that environments. the 48-h median lethal concentration value for cadmium for sexually mature male Gammarus pulex was 12.8 times greater Growth than for sexually mature females not carrying eggs or brooding Fungicide treatments significantly reduced head lengths of unfertilized or stage 1 eggs. Reproduction is not without cost, in male, female, and juvenile amphipods (Figure 3). However, the terms of both post reproductive survival and future reproductive effects differed among males, females, and juveniles. For males, potential [39]. Furthermore, energy requirements during there was no interaction effect, but the main effects of both oogenesis and brooding in females may be higher than during fungicides on male head lengths were statistically significant spermatogenesis in males, resulting in less energy available (Table 1). The male head lengths decreased by 4% at the highest to cope with toxic stress in females [23]. Therefore, post concentration of Filan alone (40 mg a.i./L) and by 6% at 3 mg reproductive females are often more susceptible than post a.i./L of Systhane alone compared with controls. Similarly, the reproductive males [41]. Juvenile amphipods are commonly individual effects of Filan and Systhane on juvenile head length used in laboratory toxicity tests because they are more sensitive were significant (Table 1). There was a 9% and 3% reduction in than at mature stages, and it is also more difficult to control the juvenile head length at 40 mg a.i./L concentration of Filan and Page 52 of 120 Endpoint dependence of chronic mixture effects Environ Toxicol Chem 9999, 2017 5

3 mg a.i./L Systhane, respectively. For females, there was only a significant effect of Filan on head lengths at the highest concentration, 40 mg a.i./L (p ¼ 0.008), which reduced head length by 4%. Growth has been extensively used as an endpoint in toxicity tests of individual chemicals, but it has rarely been used in mixture toxicity tests. Studies have shown that growth is a sensitive parameter in chronic mixture toxicity [10] and that results may differ from the survival endpoint [10,12]. For example, Spehar et al. [10] reported no significant effect on survival of fathead minnows after exposure to metal mixtures (As, Cd, Cr, Cu, Hg, Pb) for 32 d but observed a decrease in growth (a 30% reduction in dry wt) compared with the control. In contrast, Bao et al. [12] observed a synergistic lethal effect but only an additive effect on the developmental Figure 4. Cumulative number of gravid females (mean standard error) of time of copepod larvae after exposure to mixtures of Irgarol Austrochiltonia subtenuis in control and fungicide treatments after 28 d of ¼ andCu(940and50mg/L) in an 18-d life cycle test. In the exposure, n 4. Blue bars are Filan-only treatments, and red bars are mixtures of Filan and Systhane. present study, even though we did not observe interaction effects of the 2 tested fungicides, the independent effects still demonstrated that growth was a more sensitive endpoint than survival. To our knowledge, the present study is the first in The results in terms of time to become gravid and the number aquatic toxicology that assesses the effects of a mixture on of gravid females consistently showed that Systhane delayed growth based on sex and life stages. It was clear that the amphipod maturation. Carrying a brood is likely to cost energy. response was sex-specific, in that both fungicides had If the energy status of a female is reduced (e.g., by stress) to the individual effects on male growth but only Filan reduced extent that by incubating a brood she jeopardizes her own female growth. Furthermore, Filan had a significant effect survival, then her overall fitness may be increased by sacrificing on male head length at 1, 10, and 40 mg a.i./L (p ¼ 0.01, the broodings and reproducing at a later date [39]. Current p ¼ 0.019, p ¼ 0.014, respectively), while a significant effect reproduction versus survival is the most prominent life-history was observed only at the highest concentration of 40 mg a.i./L trade-off for the animal, and current reproduction versus for females (p ¼ 0.008). Our previous studies have also shown parental growth is also another possible trade-off [43] that has that male and female amphipods respond differently to been observed in amphipods [44]. Thus, a possible explanation fungicide exposure in terms of growth [28]. The present study for the effects of Systhane on amphipod maturation could be the further indicated that juveniles were more sensitive to the allocation of energy resources from reproduction to body tested fungicides in terms of growth than mature animals. maintenance or development, thereby increasing the likelihood The observed effect of fungicides on juveniles may have of survival and growth by postponing the reproduction of brood. happened during embryonic development as well as in the Our results in terms of female growth and survival strongly post hatch period, because juveniles were collected weekly in support this hypothesis because Systhane had no effects on our study. Embryos and newborn juveniles are often the most female head length (Table 1) and female survival (data not sensitive stages [39,42], and therefore fungicides are likely to shown). These findings are also in agreement with our previous cause more adverse effects on these life stages than mature study showing that juvenile amphipods exposed to Systhane adults. (0.3, 3, and 30 mg a.i./L) reached maturation later than control amphipods [28]. Reproduction Fecundity. Only Filan affected amphipod fecundity (num- Maturation. In the present study, maturation was investi- ber of embryos/gravid female; Table 1). The number of gated by 2 endpoints: time to become gravid and the number of embryos/gravid female tended to decrease in all Filan treat- gravid females. ments compared with the control (Figure 5), but a significant Only Systhane had a significant effect on the time for females effect was observed only at 10 mg a.i./L. There was no evidence to become gravid (Table 1). It took approximately 1 wk longer of an interaction between the fungicides. than controls for females to become gravid at 3 mg a.i./L of Animal size has been shown to be important in reproductive Systhane (Supplemental Data, Figure S1). success in a variety of species including amphipods, because the The number of gravid females was significantly reduced animals have to reach a certain size before reproduction can in all fungicide treatments (Figure 4). There was a significant occur [39,45]. Amphipod reproductive success is closely linked interaction effect between Filan and Systhane (Table 1) to female [39,44] and male body sizes [44]. The present data on and the effect was antagonistic, because there was no amphipod growth is consistent with the result for female significant difference between the number of gravid females fecundity. Both fungicides had individual effects on male size, in the mixture and in the Filan treatment with increasing but only Filan had a significant effect on female size. As a result, Filan concentrations (Figure 4). It seems that Systhane only Filan had a significant effect on the number of embryos/ was the major factor contributing to the reduction in number gravid female. For single toxicants, the relationship between of gravid females, because the effect of Systhane was growth and reproduction has been extensively documented and observed at 0 mg a.i./L of Filan (F(1,24) ¼ 28.68, p < 0.001), discussed in the literature [23,39,44]; in mixture studies, growth at which the reduction was approximately 61% compared and reproduction endpoints are rarely measured, but the with the control. In fact, the presence of Filan did not affect relationship between them has been reported [11,46]. Growth the cumulative number of gravid females in the fungicide has been considered a more useful parameter than reproduction, mixtures. because determinations of the latter are generally subject to Page 53 of 120 6 Environ Toxicol Chem 9999, 2017 H.T. Vu et al.

Figure 5. Number of embryos per gravid female (mean standard error) of Austrochiltonia subtenuis in control and fungicide treatments after 28 d of Figure 6. Cumulative number of juveniles (mean standard error) of ¼ Austrochiltonia subtenuis in control and fungicide treatments after 28 d of exposure, n 4. Blue bars are Filan-only treatments, and red bars are ¼ mixtures of Filan and Systhane. exposure, n 4. Blue bars are Filan-only treatments, and red bars are mixtures of Filan and Systhane. greater variations [46]. Moreover, if a chemical affects growth Systhane and only Filan had a main effect on amphipod of the exposed organism, it thereby automatically alters the fecundity (Table 1), the mean brood size between early and late toxicokinetics of all mixture components and their effects on stages of development could be reduced if female amphipods reproduction [7,46]. However, the relationship between growth are exposed to contaminants while carrying eggs, because and reproduction has been shown to exhibit a number of forms, embryos developing in the maternal brood pouch could be a each dependent on the manner in which energy is allocated very sensitive life stage [42]. In the present study, the embryo between somatic (growth) and gametic (reproduction) tissue development stage in fungicide treatments was mostly at early development. A proportional (linear) relationship between (1 and 2) and middle (3 and 4) stages (77%); it is likely that the growth and reproduction could be seen if the energy allocations survivorship of embryos at the last stages (5 and 6) before to growth and reproduction are affected similarly by tox- hatching could be affected, resulting in a significant decline in icants [47]. For example, a proportional relationship between the number of juveniles produced. Another plausible explana- growth and reproduction was observed in an experiment in tion is mortality of newborn juveniles. In the present study, which D. magna was exposed to mixtures of 10 organic juveniles were collected on a weekly basis, so it is possible that compounds with diverse modes of actions [46]. Conversely, the mortality of newborn juveniles occurred in between collections. proportional relationship may not exist if growth and Overall findings on reproduction showed that reproduction reproduction respond with differential sensitivity [39]. Cleuvers was the most sensitive endpoint and the only endpoint at which et al. [11] reported that mixture effects of 3 nonsteroidal anti- interactions were observed. Even though Systhane is an azole inflammatory drugs on D. magna growth were associated with fungicide that potentially causes synergistic effects, interaction reproduction except for one treatment where the D. magna body effects observed in the present study were antagonistic. The length increased with increasing mixture concentrations but present study used environmentally relevant concentrations, reproduction clearly decreased at the same time. Further work whereas the synergistic effects of azole fungicides observed needs to address the complicated relationship between growth in other studies occurred at lethal concentrations in acute and reproduction in mixture toxicity. tests [20–22]. It has been reported that true synergistic Fertility. In the present study, fertility was defined as interactions are rare and often occur at high concentrations [5]. cumulative number of viable juveniles per replicate, because the For example, Charles et al. [48] investigated the acute toxic juveniles were collected on a weekly basis and the number of effects of Cu and Ni on the amphipod G. pulex, reporting that females was only determined at the end of the experiment. There toxicity was synergistic when both metals were exposed at equal was a significant interaction effect between Filan and Systhane lethal concentrations (LC1–90), but was antagonistic when Ni on the cumulative number of juveniles (Table 1), resulting in a was used at sublethal concentrations. Furthermore, it has been 54% to 77% reduction of juveniles. Exposure to Systhane reported that when one is dealing with compounds that act resulted in a decrease in the number of juveniles with increasing through different mechanisms in mixture studies, an increase in Filan concentrations (Figure 6). However, significant effects the sensitivity of the parameter studied will lead to a decrease in of Systhane on amphipod fertility were observed at 0, 1, and additivity of the joint action of the mixture [46]. Although the 10 mg a.i./L of Filan (F(1,24) ¼ 83.74, p < 0.001; F(1,24) ¼ 23.13, modes of action of Filan and Systhane on aquatic invertebrates p < 0.001; F(1,24) ¼ 4.71, p ¼ 0.04, respectively) but not at 40 mg are unclear, modes of fungicide action could predict analogous a.i./L (F(1,24) ¼ 2.22, p ¼ 0.15). The effect of the fungicide mechanisms of toxicity, target sites, and toxic effects for mixture on cumulative number of juveniles was antagonistic, nonfungal species, because many biochemical pathways and because the difference between cumulative number of juveniles processes are conserved across species [49]. Filan belongs to the in the mixture and in the Filan treatment decreased with succinate dehydrogenase inhibitor group of fungicides, which increasing Filan concentrations (Figure 6). disrupt fungal respiration, whereas Systhane belongs to the Amphipod fertility is affected by all the previously demethylation inhibitor group, which inhibits sterol biosynthe- mentioned reproduction measurements. The decrease in the sis in fungal membranes [15]. This difference in modes of number of juveniles is closely related to the reduction in number fungicide action could result in a decrease in the additivity of the of gravid females and the increasing time to become gravid. interaction effect on reproduction. Nevertheless, the present Even though there were no interaction effects of Filan and reproductive results suggest that fungicides act selectively on Page 54 of 120 Endpoint dependence of chronic mixture effects Environ Toxicol Chem 9999, 2017 7 reproduction variables; reproduction was the most sensitive following fungicide exposure but varied depending on sex, parameter and could provide insight for a better understanding fungicide concentration, type of fungicide, animal age, and test and evaluation of mixture effects in amphipods. species. As expected, survival was less sensitive than We have demonstrated that mixture toxicity of Filan and reproduction or growth. Fungicide exposure caused no Systhane could be underestimated if only total mortality (both significant effects on survival of amphipods when they were acute and chronic) is used, because no significant effects exposed at the mature stage but caused significant effects when were observed on survival of the amphipods, but significant they were exposed as juveniles. Biochemical biomarkers were effects were observed on growth, sex-specific survival, and altered after fungicide exposure but varied depending on test reproduction. species, type of fungicide, and sex.

GST activity CONCLUSIONS Activity of GST was altered in all fungicide treatments In conclusion, both tested fungicides (Filan and Systhane) fi (Supplemental Data, Figure S2), but no signi cant effects were caused adverse effects on the growth and reproduction of the observed (Table 1). Such activity was 60% higher in males than amphipod A. subtenuis singly and in combination at environ- in females in the control (Supplemental Data, Figure S2). For mentally relevant concentrations. Reproduction was the most males, GST activity increased by approximately 60% in sensitive endpoint and the only one for which interaction effects Systhane treatments compared with the treatments without of 2 fungicides were observed. However, these joint effects fi Systhane, but no signi cant main effect was observed. were antagonistic, and Systhane seems to be a major factor in fi fi The GSTs play a signi cant role in detoxi cation across these interaction effects. Our results demonstrated that the multiple kingdoms and phyla [50] and have been widely used effects of fungicide mixtures were endpoint dependent and that as biomarkers to assess pesticide contamination [51], espe- a variety of endpoints should be considered for a comprehensive cially organochlorine compounds [52]. However, GSTs understanding of mixture effects, because different single generally have not shown predictable behavior in inverte- endpoints could lead to a completely different interpretation. brates [51]. In crustaceans, GST activity was found to be Measuring total mortality only (both acute and chronic) may be fi elevated [52], inhibited [53], or not modi ed [54] after insufficient to assess the toxicity effects of mixtures at pesticide exposure. In addition to the natural seasonal and environmentally realistic concentrations. Growth and repro- spatial variation, factors inherent to the endogenous charac- duction are important endpoints in chronic mixture studies, but teristic of test species, such as size, gender, reproduction toxicants may act selectively on reproduction variables, and fl status, and sexual maturity can also in uence biomarker growth alone is not a good indicator for overall population response in many invertebrates [51,55]. In the present study, fitness. Biochemical biomarker endpoints may not always be we observed higher levels of GST in males than in females, in more sensitive than whole-body endpoints, and their use should both the control and fungicide treatments. This is in agreement be evaluated based on the species and toxicants studied. The with the study conducted by Correia et al. [56], who found that present study emphasizes the importance of chronic mixture glutathione-related enzyme activity in the amphipod Gamma- studies and suggests that reproduction-related endpoints could rus locusta was 50% lower in females than males. The provide better insights for understanding and evaluating mixture difference in GST activity between genders may be related to toxicity. size because the male amphipods are generally bigger than the females. In the present study, the head length of control Supplemental Data—The Supplemental Data are available on the Wiley amphipods was 33% longer in males than females. Depending Online Library at DOI: 10.1002/etc.3809. on species, the size of the test organism may have different effects on level of GST activity. For example, Robillard at Acknowledgment—We thank T. Mehler for kindly assisting in the field al. [57] found that GST activity increased with the length of collection of amphipods and water. Funding for the present study was provided by the Centre for Aquatic Pollution Identification and Manage- the mussel Anodonta cygnea. In contrast, Jemec et al. [55] ment (CAPIM). H. Vu was funded by a Melbourne International Research reported that GST activity in D. magna decreased with size Scholarship during the present study. (directly related to age). The observed high variations in GST activity in both genders in the present study could be related to Data Availability—Data, associated metadata, and calculation tools are reproductive status. The present study used sexually mature available from the corresponding author ([email protected]). amphipods, and during the experiment the amphipods may have experienced one or more reproductive cycles [58]. The REFERENCES amphipods that were randomly chosen for GST analysis may 1. Wightwick A, Allinson G, Menzies N, Walters R, Reichman S. 2010. have been at different phases of the reproductive cycle, which Environmental risks of fungicides used in horticultural production may impact GST activity [59]. Our GST results imply that systems. In Carisse O, ed, Fungicides. InTech, DOI: 10.5772/13032. [cited 2016 December 20]. Available from: https://www.intechopen. biomarkers may not always be more sensitive than whole- com/books/fungicides/environmental-risks-of-fungicides-used-in-hort organism responses, and more endogenous characteristics of icultural-production-systems A. subtenuis need to be known before biochemical biomarkers 2. Brent KJ, Hollomon DW. 1995. Fungicide resistance in crop pathogens: are used in this species as diagnostic tools for environmental How can it be managed? Fungicide Resistance Action Committee, contamination assessments. Brussels, Belgium. 3. McGrath MT. 2015. General guidelines for managing fungicide The present study and our 2 previous studies [23,28] report resistance. Department of Plant Pathology, Cornell University, Ithaca, on the wide range of effects that the fungicides Filan and NY, USA. Systhane have on 2 Australian amphipod species, Allorchestes 4. Wedge DE, Smith BJ, Quebedeaux JP, Constantin RJ. 2007. Fungicide compressa and A. subtenuis, at environmentally relevant management strategies for control of strawberry fruit rot diseases in Louisiana and Mississippi. Crop Prot 26:1449–1458. concentrations under laboratory conditions. In summary, 5. Cedergreen N. 2014. Quantifying synergy: A systematic review of reproduction was the most sensitive endpoint for both single mixture toxicity studies within environmental toxicology. PloS One 9: and mixture exposures. Growth was also significantly affected e96580. Page 55 of 120 8 Environ Toxicol Chem 9999, 2017 H.T. Vu et al.

6. Belden JB, Gilliom RJ, Lydy MJ. 2007. How well can we predict the Environmental fate of fungicides in surface waters of a horticultural- toxicity of pesticide mixtures to aquatic life? Integr Environ Assess production catchment in Southeastern Australia. Arch Environ Contam Manag 3:364–372. Toxicol 62:380–390. 7. Jager T, Vandenbrouck T, Baas J, De Coen WM, Kooijman SA. 2010. 30. Bird GA, Kaushik NK. 1985. Processing of elm and maple leaf discs by A biology-based approach for mixture toxicity of multiple endpoints collectors and shredders in laboratory feeding studies. Hydrobiologia over the life cycle. Ecotoxicology 19:351–361. 126. 8. De Zwart D, Posthuma L. 2005. Complex mixture toxicity for single 31. King RA, Leys R. 2011. The Australian freshwater amphipods and multiple species: Proposed methodologies. Environ Toxicol Chem Austrochiltonia australis and Austrochiltonia subtenuis (Amphipoda: 24:2665–2676. Talitroidea: Chiltoniidae) confirmed and two new cryptic Tasmanian 9. Zou X, Lin Z, Deng Z, Yin D, Zhang Y. 2012. The joint effects of species revealed using a combined molecular and morphological sulfonamides and their potentiator on Photobacterium phosphoreum: approach. Invertebr Syst 25:171–196. Differences between the acute and chronic mixture toxicity mecha- 32. Wilhelm FM, Lasenby DC. 1998. Seasonal trends in the head capsule nisms. Chemosphere 86:30–35. length and body length/weight relationships of two amphipod species. 10. Spehar RL, Fiandt JT. 1986. Acute and chronic effects of water quality Crustaceana 71:399–410. criteria-based metal mixtures on three aquatic species. Environ Toxicol 33. Mann R, Hyne R. 2008. Embryological development of the Australian Chem 5:917–931. amphipod, Melita plumulosa Zeidler, 1989 (Amphipoda, Gammaridea, 11. Cleuvers M. 2008. Chronic mixture toxicity of pharmaceuticals to Melitidae). Crustaceana 81:57–66. Daphnia—The example of nonsteroidal anti-inflammatory drugs. In 34. Habig WH, Pabst MJ, Jakoby WB. 1974. Glutathione S-transferases the Kummerer K, ed, Pharmaceuticals in the Environment. Springer, New first enzymatic step in mercapturic acid formation. J Biol Chem York, NY, USA, pp 277–284. 249:7130–7139. 12. Bao VW, Leung KM, Lui GC, Lam MH. 2013. Acute and chronic 35. Long SM, Tull DL, Jeppe KJ, De Souza DP, Dayalan S, Pettigrove VJ, toxicities of Irgarol alone and in combination with copper to the marine McConville MJ, Hoffmann AA. 2015. A multi-platform metabolomics copepod Tigriopus japonicus. Chemosphere 90:1140–1148. approach demonstrates changes in energy metabolism and the 13. Giavini E, Menegola E. 2010. Are azole fungicides a teratogenic risk for transsulfuration pathway in Chironomus tepperi following exposure human conceptus? Toxicol Lett 198:106–111. to zinc. Aquat Toxicol 162:54–65. 14. Hof H. 2001. Critical annotations to the use of azole antifungals for 36. Slinker BK. 1998. The statistics of synergism. J Mol Cell Cardiol plant protection. Antimicrob Agents Chemother 45:2987–2990. 30:723–731. 15. University of Hertfordshire. 2017. Pesticides Properties Database. 37. Vu HT, Klaine SJ. 2014. Testing the individual effective dose Hertfordshire, UK. hypothesis. Environ Toxicol Chem 33:791–797. 16. Bromilow RH, Evans AA, Nicholls PH. 1999. Factors affecting 38. Kulkarni D, Daniels B, Preuss TG. 2013. Life-stage-dependent degradation rates of five triazole fungicides in two soil types: 2. Field sensitivity of the cyclopoid copepod Mesocyclops leuckarti to studies. Pest Manage Sci 55:1135–1142. triphenyltin. Chemosphere 92:1145–1153. 17. Rose G, Allen D, Allinson G, Allinson M, Bui A, Wightwick A, Zhang 39. Conradi M, Depledge MH. 1999. Effects of zinc on the life-cycle, P. 2009. Melbourne water and DPI agrochemicals in Port Phillip growth and reproduction of the marine amphipod Corophium volutator. catchment streams—Project summary report on 2008–09. Department Mar Ecol Prog Ser 176:131–138. of Primary Industries, Melbourne, VIC, Australia. 40. McCahon CP, Pascoe D. 1988. Increased sensitivity to cadmium of the 18. Smalling KL, Kuivila KM, Orlando JL, Phillips BM, Anderson BS, freshwater amphipod Gammarus pulex (L.) during the reproductive Siegler K, Hunt JW, Hamilton M. 2013. Environmental fate of period. Aquat Toxicol 13:183–194. fungicides and other current-use pesticides in a central California 41. Wenner AM. 1972. Sex ratio as a function of size in marine crustacea. estuary. Mar Pollut Bull 73:144–153. Am Nat 106:321–350. 19. Moreno-Gonzalez R, Campillo J, Leon V. 2013. Influence of an 42. Ford AT, Fernandes TF, Rider SA, Read PA, Robinson CD, Davies IM. intensive agricultural drainage basin on the seasonal distribution of 2003. Measuring sublethal impacts of pollution on reproductive output organic pollutants in seawater from a Mediterranean coastal lagoon of marine Crustacea. Mar Ecol Prog Ser 265:303–309. (Mar Menor, SE Spain). Mar Pollut Bull 77:400–411. 43. Stearns SC. 1989. Trade-offs in life-history evolution. Funct Ecol 20. Nørgaard KB, Cedergreen N. 2010. Pesticide cocktails can interact 3:259–268. synergistically on aquatic crustaceans. Environ Sci Pollut Res 17: 44. Robinson B, Doyle R. 1985. Trade-off between male reproduction 957–967. (amplexus) and growth in the amphipod Gammarus lawrencianus. Biol 21. Zubrod J, Baudy P, Schulz R, Bundschuh M. 2014. Effects of current- Bull 168:482–488. use fungicides and their mixtures on the feeding and survival of the key 45. Ward PI. 1983. Advantages and a disadvantage of large size for male shredder Gammarus fossarum. Aquat Toxicol 150:133–143. Gammarus pulex (Crustacea: Amphipoda). Behav Ecol Sociobiol 22. Kretschmann A, Gottardi M, Dalhoff K, Cedergreen N. 2015. The 14:69–76. synergistic potential of the azole fungicides prochloraz and propico- 46. Deneer J, Seinen W, Hermens J. 1988. Growth of Daphnia magna nazole toward a short alpha-cypermethrin pulse increases over time in exposed to mixtures of chemicals with diverse modes of action. Daphnia magna. Aquat Toxicol 162:94–101. Ecotoxicol Environ Saf 15:72–77. 23. Vu HT, Keough MJ, Long SM, Pettigrove VJ. 2016. Effects of the 47. Sibley PK, Benoit DA, Ankley GT. 1997. The significance of growth 1 boscalid fungicide Filan on the marine amphipod Allorchestes in Chironomus tentans sediment toxicity tests: Relationship to compressa at environmentally relevant concentrations. Environ Toxicol reproduction and demographic endpoints. Environ Toxicol Chem Chem 35:1130–1137. 16:336–345. 24. Kreuger J, Graaf S, Patring J, Adielsson S. 2010. Pesticides in surface 48. Charles J, Crini G, Degiorgi F, Sancey B, Morin-Crini N, Badot P-M. water in areas with open ground and greenhouse horticultural crops in 2014. Unexpected toxic interactions in the freshwater amphipod Sweden 2008. Swedish University of Agricultural Sciences, Uppsala, Gammarus pulex (L.) exposed to binary copper and nickel mixtures. Sweden. Environ Sci Pollut Res 21:1099–1111. 25. Battaglin WA, Sandstrom MW, Kuivila KM, Kolpin DW, Meyer MT. 49. Elskus AA. 2012. Toxicity, sublethal effects, and potential modes 2011. Occurrence of azoxystrobin, propiconazole, and selected other of action of select fungicides on freshwater fish and invertebrates. fungicides in US streams, 2005–2006. Water Air Soil Pollut 218: US Geological Survey Open-File Report 2012–1213. Reston, 307–322. VA, USA. 26. Smalling KL, Orlando JL. 2011. Occurrence of pesticides in water and 50. Oakley A. 2011. Glutathione transferases: A structural perspective. sediment from three central California coastal watersheds, 2008–2009. Drug Metab Rev 43:138–151. US Geological Survey Data Series 600. Reston, VA, USA. 51. Domingues I, Agra AR, Monaghan K, Soares AMVM, Nogueira AJA. 27. Reilly TJ, Smalling KL, Orlando JL, Kuivila KM. 2012. Occurrence of 2010. Cholinesterase and glutathione-S-transferase activities in boscalid and other selected fungicides in surface water and groundwater freshwater invertebrates as biomarkers to assess pesticide contamina- in three targeted use areas in the United States. Chemosphere 89: tion. Environ Toxicol Chem 29:5–18. 228–234. 52. McLoughlin N, Yin DQ, Maltby L, Wood RM, Yu HX. 2000. 28. Vu HT, Keough MJ, Long SM, Pettigrove VJ. 2017. Effects of two Evaluation of sensitivity and specificity of two crustacean biochemical commonly used fungicides on the amphipod Austrochiltonia subtenuis. biomarkers. Environ Toxicol Chem 19:2085–2092. Environ Toxicol Chem 36:720–726. 53. Steevens JA, Benson WH. 1999. Toxicological interactions of 29. Wightwick AM, Bui AD, Zhang P, Rose G, Allinson M, Myers JH, chlorpyrifos and methyl mercury in the amphipod, Hyalella azteca. Reichman SM, Menzies NW, Pettigrove V, Allinson G. 2012. Toxicol Sci 52:168–177. Page 56 of 120 Endpoint dependence of chronic mixture effects Environ Toxicol Chem 9999, 2017 9

54. Anguiano OL, Castro C, Venturino A, Ferrari A. 2014. Acute toxicity 57. Robillard S, Beauchamp G, Laulier M. 2003. The role of abiotic factors and biochemical effects of azinphos methyl in the amphipod Hyalella and pesticide levels on enzymatic activity in the freshwater mussel curvispina. Environ Toxicol 29:1043–1053. Anodonta cygnea at three different exposure sites. Comp Biochem 55. Jemec A, Drobne D, Tisler T, Sepcic K. 2010. Biochemical biomarkers Physiol C 135:49–59. in environmental studies—Lessons learnt from enzymes catalase, 58. Mann RM, Hyne RV, Spadaro DA, Simpson SL. 2009. Development glutathione S-transferase and cholinesterase in two crustacean species. and application of a rapid amphipod reproduction test for sediment- Environ Sci Pollut Res 17:571–581. quality assessment. Environ Toxicol Chem 28:1244–1254. 56. Correia AD, Costa MH, Luis OJ, Livingstone DR. 2003. Age-related 59. Regoli F, Nigro M, Chiantore M, Winston G. 2002. Seasonal variations changes in antioxidant enzyme activities, fatty acid composition and of susceptibility to oxidative stress in Adamussium colbecki, a key lipid peroxidation in whole body Gammarus locusta (Crustacea: bioindicator species for the Antarctic marine environment. Sci Total Amphipoda). J Exp Mar Biol Ecol 289:83–101. Environ 289:205–211.

Page 57 of 120 SUPPLEMENTAL DATA

Table S1. Nominal and measured concentrations of Filan® and Systhane™ (µg a.i./L) Nominal concentration Measured concentration* Filan® Systhane™ Filan® Systhane™ 0 0 <0.1 <0.2 1 0 1.1 <0.2 10 0 13 <0.2 40 0 36 <0.2 0 3 <0.1 4.0 1 3 1 4.2 10 3 12 4.0 40 3 35 3.9 *Limit of detection (LOD) = 0.1 µg/L for Filan® and 0.2 µg/L for Systhane™. Reported values were based on one replicate.

Systhane™ (0 µg a.i./L) 11 Systhane™ (3 µg a.i./L)

10

9 (week) 8

Time to become gravid 7 0 1 10 40 Filan® concentration (µg a.i./L)

Figure S1. Time to get gravid (mean ± SE) of Austrochiltonia subtenuis in control and fungicide treatments (single and mixture of Filan® and Systhane™) after 28 day exposure, n = 4. Blue bars are Filan® only treatments, red bars are mixtures of Filan® and Systhane™.

Page 58 of 120 A Systhane™ (0 µg a.i./L) 60 Systhane™ (3 µg a.i./L) 50 40 30 20 10 GST activity malein

(nmol/min/mgprotein) 0 0 1 10 40 Filan® concentration (µg a.i./L)

B Systhane™ (0 µg a.i./L) 20 Systhane™ (3 µg a.i./L)

15

10

5 GST activity GST activity femalein (nmol/min/mgprotein) 0 0 1 10 40 Filan® concentration (µg a.i./L)

Figure S2. Glutathione S-transferase (GST) activity in males (A) and females (B) (mean ± SE) of Austrochiltonia subtenuis in control and fungicide treatments after 28 day exposure, n = 4. Blue bars are Filan® only treatments, red bars are mixtures of Filan® and Systhane™.

Page 59 of 120 CHAPTER 5: CAN ORGANIC MATTER DECOMPOSITION INDICATE THE EFFECTS OF MULTIPLE ANTHROPOGENIC STRESSORS ON FUNCTIONAL STREAM HEALTH?

Abstract

Pesticides constitute a major anthropogenic stress to freshwater ecosystems and can have effects on both structure and function. Studies have shown the effects of pesticides on the dynamics of macroinvertebrate communities in agricultural streams but relatively little is known about their effects on ecosystem function. The present study investigated the effects of pesticides and other stressors on organic matter (leaf and cotton) decomposition at 26 sites, 24 study and 2 reference sites, in an intensive agricultural region in south-eastern Australia. The study also evaluated the potential of cotton as a standard substrate to measure effects of anthropogenic stressors on organic matter decomposition in aquatic environments. Leaves and cotton strips were deployed at the sites for a 3 week period and repeated twice in winter and in spring. Pesticide concentrations were monitored during the study using passive samplers. Of the 113 pesticides analyzed, 43 were detected during the study period, comprising 9 herbicides, 17 fungicides, and 17 insecticides. Fungicide, herbicide, and insecticide concentrations were 76, 97, and 57 %, respectively, higher at the studied than at the reference site two which has similar altitude to the study sites. The ecosystem function of most of the sites with pesticide concentrations higher than the reference site two was classified as impacted, based on leaf and cotton breakdown rate (79 and 92 % of sites, respectively). Organic matter breakdown rate was significantly correlated with pesticide concentrations and nutrients but leaf breakdown was also strongly impacted by temperature. Leaf and cotton degraded differently but both indicated the same effects on functional stream health for majority of the sites. The results demonstrated that cotton has potential as a standard organic matter substrate to measure effects of anthropogenic stressors on ecosystem function.

Keywords: Pesticides, Litter breakdown, Cotton-strip assay, Ecosystem function.

5.1 Introduction

Traditional methods to assess the effects of anthropogenic stresses on stream health have mostly relied on measurements of ecosystem structure (e.g., community composition of aquatic

Page 60 of 120 organisms, water quality) and neglected ecosystem functional processes (e.g., nutrient recycling, primary production, and organic matter decomposition) (Gessner and Chauvet, 2002; Silva- Junior and Moulton, 2011; Tiegs et al., 2013). Even though structure and function describe different aspects of the same entity and are linked, changes in structure do not always equate to changes in functional ecosystem attributes (Gessner and Chauvet, 2002). For example, macroinvertebrate community structure was more sensitive to metal pollution than leaf breakdown rate in high-altitude streams (Nelson, 2000). However, Bunn and Davies (2000) reported that gross primary production and community respiration responded to nitrogen enrichment and increased water turbidity, while macroinvertebrate community structure failed to detect these stresses. Therefore, a focus solely on structural attributes would not provide a comprehensive understanding of the impacts of anthropogenic stresses on the stream health. Functional measures depict a more general picture than the structure of stream biota as they are not critically dependent on the presence of a specific set of species (Gessner and Chauvet, 2002). They provide an integrative measure of ecosystem integrity which may work not only over time but also across organisms at different organizational levels (Niyogi et al., 2013).

There are many ecosystem-level processes that could be used as functional indicators of stream ecosystem health such as primary production, community respiration, nitrification, sediment respiration, invertebrate production, and organic matter breakdown (Gessner and Chauvet, 2002; Young et al., 2008). Among these, organic matter breakdown has been studied in many streams in different geographical areas (Imberger et al., 2010; Niyogi et al., 2013; Schäfer et al., 2012; Tiegs et al., 2007; Tiegs et al., 2013) and has been proposed as a cost-effective, synthetic, and functional indicator of stream health (Niyogi et al., 2013; Young et al., 2008). Organic matter breakdown directly integrates microorganism (bacteria and fungi) assemblages, energy conversion, and animals, thus provides insights into stream processing and energy transfer that structural measures could not provide. This process is influenced by biotic and abiotic factors and can be separated into two phases: (1) an initial rapid loss of soluble substances (leaching) and (2) a slower microbial/physical breakdown phase (Webster and Benfield, 1986). Breakdown of leaves, cotton strips, and wood sticks has been studied to determine the ecosystem function of organic matter breakdown (Imberger et al., 2010; Niyogi et al., 2013; Schäfer et al., 2012; Tiegs et al., 2007; Vyšná et al., 2014).

Page 61 of 120 Measuring leaf breakdown rate using litter bags is the most common approach for quantifying organic matter decomposition in aquatic environments (Tiegs et al., 2013). Leaf breakdown rate has been demonstrated to be sensitive to many impacts of human activities including agriculture (Rasmussen et al., 2012), mining (Gray and Ward, 1983; Maltby and Booth, 1991), urbanization (Imberger et al., 2008; Paul et al., 2006), and sewage discharge/effluent (Spänhoff et al., 2007). These anthropogenic stresses can depress (e.g., heavy metals, acidification, pesticides) or accelerate (e.g., nutrients, thermal pollution) leaf breakdown rates (Gessner and Chauvet, 2002; Young et al., 2008). While leaf breakdown meets many criteria of being a good functional indicator (Gessner and Chauvet, 2002), it is not without limitations. Some of these shortcomings may compromise the applicability of leaf decomposition as a standardized diagnostic tool, such as inconsistent methods among studies (e.g. mesh size, bag shape, leaf mass) or variation in leaf litter quality both between and within species (Tiegs et al., 2013). Cellulose breakdown using cotton strip assays has been used in the terrestrial environment for more than 50 years (Girling et al., 2000) and has recently been tested and developed as a standardized functional indicator of stream ecosystem health (Imberger et al., 2010; Niyogi et al., 2013; Schäfer et al., 2012; Tiegs et al., 2007; Tiegs et al., 2013). The cotton strip assay offers several advantages over the litter bag assay including standard substrate, less incubation time, less labor intensive, less susceptible to handling, and it is sensitive to small environmental changes (Imberger et al., 2010; Tiegs et al., 2007). However, there is limited mechanistic understanding of cellulose decomposition and thus it is difficult to diagnose the causal factors to a change in composition (Imberger et al., 2010). Wood breakdown is another approach used to evaluate organic matter breakdown. Wood breakdown is comparable to leaf litter decomposition in that it is a complex process involving bacteria, fungi, and invertebrates (Spänhoff and Meyer, 2004). However, wood breakdown is generally slower and has been used to a lesser extent than leaf breakdown. It may be useful in providing a measure that integrates varying water quality for a long period of time (Niyogi et al., 2013).

Pesticides constitute a major anthropogenic stress to freshwater ecosystems and can impact all organisms (Schäfer et al., 2012). Many studies have shown the effects of pesticides on the dynamics of macroinvertebrate communities in agricultural streams (Liess and Ohe, 2005; Schafer et al., 2011; Schulz, 2004; Schulz and Liess, 1999). However, relatively few field studies have analyzed the effects of pesticides on ecosystem function using organic matter breakdown.

Page 62 of 120 To date, three studies examining the relationship between organic matter breakdown and estimated site specific toxicity as derived from measured pesticide concentrations have been published (Rasmussen et al., 2012; Schäfer et al., 2012; Schäfer et al., 2007). All studies found a significant reduction in organic matter breakdown due to exposure to pesticides. Nevertheless, effects of individual pesticide groups (insecticides, herbicides, and fungicides) were not separated. It is important to understand the contribution of individual groups of pesticides as they may have different impacts on the breakdown of organic matter, especially on different substrates. There is no available aquatic study on effects of different pesticide groups on organic matter breakdown but terrestrial studies showed that effects of pesticides on organic matter breakdown differed among pesticide groups (Katayama and Kuwatsuka, 1991) and depended on substrate (Knacker et al., 2003).

The present study investigated organic matter (leaf and cotton) decomposition at 26 sites in three catchments in south-eastern Australia in winter and spring of 2015. These seasons were chosen as they represent different periods of pesticides application. Spring is a major growing season for many crops in Australia and thus represents the intensive period of pesticides use (Wightwick et al., 2012). The use of pesticides in farming is often low in winter which is a dry season in Australia (Radcliffe, 2002). Leaf mass loss and cotton tensile strength loss were compared and analyzed with environmental stressors to address the questions:

1. How do leaf and cotton breakdown respond to environmental stressors?

2. Can cotton be used as standard type of organic matter to measure effects of anthropogenic stressors on ecosystem function in aquatic environments?

5.2 Materials and Methods 5.2.1 Study design and sampling schedule

The study was conducted at 26 sites in the Western Port catchments of Watsons Creek (Sites 1- 12), Western Contour Drain (Sites 13-24), and Cardinia Creek (2 reference sites), an intensive agricultural region of eastern Melbourne, Victoria, Australia. The study involved four three-week sampling programs conducted from 30th June to 12th August 2015 (winter) and 27th October to 9th December 2015 (spring). Hereafter these sampling programs are referred to as periods 1 and 2 (winter); periods 3 and 4 (spring). Initial sites were selected based on a desk-top analysis of

Page 63 of 120 sub-catchment drainage into each system and field scoping of the catchments. In spring, the water level at some sites was too low for deployment due to low rainfall. These sites were removed from the study and new sites were added (Table 1). Two reference sites were included during the deployment periods, both located on Cardinia Creek. Reference site one (R1) was located in the headwaters of the Cardinia catchment which is considered to have good water quality due to continuous flow releases from Cardinia Reservoir. The second (R2) was situated at a similar altitude to the Watsons Creek and Western Contour Drain study sites. This site has had little to no pesticides detected in the past (CAPIM, unpublished data). Site R2 was chosen as the main reference site as it has similar altitude to the study sites. Site R1 was used to compare to R2 within the same catchment but different altitudes.

5.2.2 Pesticide monitoring and recording of environmental variables

Pesticide monitoring

Pesticides were measured at all sites during the study using passive samplers. Passive samplers consisted of an EmporeTM SDB-XC disk as the receiving phase held in a Chemcatcher® housing (Kingston et al., 2000), together with a polyethersulfone (PES) diffusion-limiting membrane. EmporeTM disks and PES membranes were conditioned with methanol followed by deionized water, and then each disk was covered with deionized water and a PTFE screw on lid and maintained on ice during transport to field sites.

Samplers were deployed in cages (plastic mesh pockets 15cm x 15cm) cable tied to steel star pickets for a period of 3 weeks. Samplers were deployed in such a position to ensure they would remain underwater for the deployment period. Upon retrieval, samplers were inverted, filled with site water, a teflon screw-on lid attached, sealed in a plastic zip-lock bag and transported back to the laboratory on ice. In the laboratory EmporeTM disks and PES membranes were removed from the Chemcatcher® holder and dried on a hotplate (35°C). Each EmporeTM disk was then packaged, labelled, and sent to National Measurement Institute (NMI, Sydney, NSW, Australia) for pesticide extraction and analysis.

Pesticides on EmporeTM disks were extracted and analyzed using capillary injection followed by high performance gas chromatography coupled with tandem mass spectrometry (GC/MS; USEPA SW 846).

Page 64 of 120 Physico-chemical parameters and nutrients

Physico-chemical measurements were determined in situ using a multi-parameter water quality analyzer HANNA 98291(HANNA instruments®, Keysborough, VIC, Australia) for temperature, dissolved oxygen, electrical conductivity (EC) and turbidity. In the present study a measurement is defined as any value (e.g. pH, fungicide concentration, leaf breakdown rate, functional stream health etc.) measured at a site in a study period.

Surface water samples were collected for nutrient analysis. Samples were analyzed on site using a Dual wave spectrophotometer (Sensafe Exact EcoCheck) at 425 nm for ammonia, filterable reactive phosphorus (FRP), nitrite and nitrate. Calibration checks, duplicates and blanks were analyzed simultaneously. Additional samples (100 – 200 ml) were collected to provide a comparison between the duel wave spectrophotometer and a standard laboratory method. These were stored on ice while being transported to Australian Laboratory Services (ALS, Melbourne, VIC, Australia). Mean recoveries for all nutrients ranged from 70 to 130 %. Physico-chemical measurements were averaged over the time between deployment and retrieval.

5.2.3 Organic matter breakdown

Leaf bag and cotton strip assays

The leaf bag and cotton strip assays followed methods developed and described in Tiegs et al. (2007) and Schäfer et al. (2012), respectively. In short, two types of litter bags were deployed containing the two substrates at each site. Coarse mesh size bags (4 - 5 mm mesh) were used to estimate the total decomposition rates of by macroinvertebrates and microbial organisms. Fine mesh size bags (1 mm mesh) that prevent macroinvertebrate colonization were used to estimate the decomposition rate by microbial organisms only. The two substrates placed in each bag were 1.025 ± 0.001g (mean ± SE) air-dried, green hazel leaves (Pomaderris aspera) and a single cotton strip. The leaves were collected from Shoreham, Victoria and are a species common to Victoria. They were moistened by dipping in deionized water prior to addition to bags to prevent breakage. Cotton strips (4 x 6 cm) were cut from bolts of Fredrix-brand unprimed 12-oz. heavy- weight cotton fabric, Style #548 (Fredrix, Lawrenceville, GA, USA) and autoclaved for 1 h at 120°C before being placed in each bag.

Page 65 of 120 Five coarse and five fine bags were deployed at each site, cable tied to star pickets approximately 10 cm above the stream bed so that they touched the bottom. After 3 weeks, bags were removed from sites, placed in plastic bags and transported to the laboratory. Each bag was emptied into a shallow plastic tray containing tap water. Each side of the cotton strip and leaves was cleaned with a soft-bristled paint brush to remove adhering debris. Cleaned leaves were oven dried (24h at 60oC) and weighed. Cleaned cotton strips were soaked in 70% ethanol, air dried and stored individually in small plastic bags until tensile strength could be determined. As a procedural control and for pre-deployment tensile strength determination, ten randomly selected cotton strips were soaked in 70% ethanol, air dried and stored as per deployed strips before tensile strength being determined.

Tensile strength was determined using an Instron Microtester 9544 tensiometer (Instron Pty Ltd, Bayswater, Victoria, Australia) as described in Tiegs et al (2013) in a controlled room set at 20oC. Strips were pulled at a fixed rate of 2cm/min and maximum tensile strength recorded for each strip.

The decomposition rate (k) was calculated for leaf (mass loss) and cotton strips (tensile strength loss) by fitting data to a simple exponential decay model (Tiegs et al., 2007)

-kt Xt = Xoe

Where Xt is the leaf mass or tensile strength upon removal of the litter bags from the field, Xo is the initial mass or tensile strength, t is deployment time in days.

Determination of functional stream health

The effects of stressors on functional stream health were assessed using the framework developed by Gessner and Chauvet (2002). Briefly, this framework ranks sites based on the decomposition rate at sites relative to that at a reference site. The functional stream health score is calculated for each site based on the ratio of ksite/kR2

Where ksite = decomposition rate at a study site and kR2 = decomposition rate at the reference site R2.

Page 66 of 120 Each site is given a score 0, 1, or 2 based on the ksite/kR2 value and thresholds proposed by Gessner and Chauvet (2002) (Table 2) then the functional stream status is classified.

5.2.4 Biotic communities

Macroinvertebrates were separated from leaves in all 5 replicates of the litter bags, pooled and preserved in 70% ethanol for identification. Macroinvertebrates were identified to the lowest taxonomic resolution possible - genus/species, with the exception of Oligochatea (class), Acarina (order), Turbellaria (family) and most Diptera (family) under a dissecting microscope. Species richness was calculated for each site and the abundance of dominant species was observed.

5.2.5 Statistical analysis

Nested analysis of variance (nested ANOVA) was used to compare the means of leaf and cotton breakdown rates between catchments, seasons, periods in a season, mesh type, and among sites in each catchment. In this analysis, catchment, season, and mesh type are fixed factors; site and period are random factors nested within catchment and season, respectively. Only sites used in all four periods were considered in the analysis. Differences between breakdown rates of study and reference sites in each period were analyzed using one-way ANOVA followed by Dunnet’s pairwise comparisons. Pearson correlation analysis was used to determine the relationship between environmental stressors and breakdown rates of leaves and cotton strips irrespective of site. This correlation was performed using the mean of five replicates of leaf and cotton breakdown rates. Leaf and cotton breakdown rates were analyzed separately. Correlations among environmental variables were analyzed by scatter plots. Statistical analyses were performed using SPSS Version 23 (IBM).

5.3 Results 5.3.1 Site characteristics

Pesticides

Of the 113 pesticides analyzed, 43 were detected in this study using passive samplers (Table 3), comprising 9 herbicides, 17 fungicides, and 17 insecticides. Herbicides, followed by fungicides

Page 67 of 120 were the most frequently detected pesticides, with twenty pesticides detected in greater than 10% of samples.

Pesticide concentrations varied temporally and spatially throughout the study with fungicides detected at the highest concentrations followed by herbicides and insecticides (Figures 1, 2). Fungicide concentrations were higher in winter than spring (Figures 1A, 2A). Herbicides were detected at almost every site with less seasonal variation compared to fungicides (Figures 1B, 2B). Concentrations of insecticides were relatively low throughout the study compared to fungicides and herbicides (Figures 1C, 2C). Pesticide concentrations were generally higher in Western Contour Drain than in Watsons Creek and were comparatively low in Cardinia Creek (Figures 1, 2). In all pesticides detections, 76, 97, and 57 % of sites had fungicide, herbicide, and insecticide concentrations, respectively, higher than that of site R2.

Physico-chemical parameters and nutrients

Physico-chemical parameters and nutrients for each period are summarized in Table S1 - Supplemental data. Water temperature, pH and dissolved oxygen exhibited slight variation (up to 47% SD), while conductivity, turbidity and nutrients show larger variability (up to 251 % SD) (Table S1).

Water temperature ranged from 7.9°C to 26.8°C. Water temperature in spring was significantly higher than in winter. The pH ranged between 6.4 to 8.6 which were generally within trigger values (TV) for south-east Australian slightly disturbed systems (6.5 – 8.0) (ANZECC and ARMCANZ, 2000). Electrical conductivity at 81% of sites was within TVs (125 – 2200 µs/cm) for south-east Australian slightly disturbed systems (ANZECC and ARMCANZ, 2000). Dissolved oxygen varied throughout the study, ranging from 215 % saturation to a low of 31.7% saturation. However, between sampling periods dissolved oxygen levels did not significantly change (Table S1). Turbidity varied significantly between sites in each sample period ranging from 3.9 to 1000 (NTU).

Nutrient concentrations varied substantially both temporally and spatially (Figures 3, 4). There were 89% and 18% of measurements for FRP and ammonia higher than trigger values for south- east Australian slightly disturbed systems of 0.02 mg/L and 0.02 mg/L respectively. There are no available TVs for nitrite and nitrate for south-east Australian slightly disturbed systems. For NOx

Page 68 of 120 (oxides of nitrogen = nitrite + nitrate) 69% of measurements were higher than the TV for south- east Australian slightly disturbed systems of 0.04 mg/L.

Macroinvertebrate assemblage litter bags

Macroinvertebrates were present in the majority of the litter bags retrieved, with the most abundant and common taxa being Oligochaeta spp., Chironomus spp., Austrochiltonia subtenuis and Potamopyrgus antipodarum. Species richness was low at almost every site throughout the study (Reference: 0-7; Western Contour: 0-10; Watsons Creek: 0-7) and did not vary significantly among sites and deployment periods (Figure S1 – Supplemental data). Taxonomic information for each site in during 4 periods was provided in Table S2 – S5 Supplemental data. Only 8 taxa were present in greater than 10% samples and this is a too small number to conduct community composition analyses.

5.3.2 Organic matter breakdown

Leaf breakdown

Leaf breakdown rates in the reference sites were in the range of breakdown rates of P. aspera in south-eastern Australian streams (Campbell et al., 1992) and similar to the breakdown rate of Alnus glutinosa, a common leaf species used in leaf litter breakdown assessments in northern hemisphere, in unimpacted streams in Europe (Gessner and Chauvet, 2002). Significant differences were observed in leaf breakdown among sites in each catchment (F3, 344 = 9.5, p <

0.001), between winter and spring (F1, 3.9 = 19.3, p = 0.012) but no significant differences in leaf breakdown among catchments (F2, 3.3 = 1.4, p = 0.37); between periods in each season (F1, 344 =

0.36, p = 0.55); or between coarse and fine mesh bags (F1, 1.94 = 3.4, p = 0.21). As a result, to demonstrate the variation between sites and seasons only leaf breakdown data in coarse mesh bags in period 1 in winter and period 4 in spring were further analyzed and displayed (Figure 5).

In winter, leaf breakdown rates in the study sites were lower compared to that of R2 (Figure 5A). There were 4 sites in Western Contour Drain and 8 sites in Watsons Creek where the leaf breakdown rates were significantly different to that of R2. There was no significant difference in breakdown rates between the reference sites.

Page 69 of 120 In spring, leaf breakdown rates were higher at all sites compared to winter (Figure 5). Most of the study sites had leaf breakdown rates higher than that of R2, however significant differences were only observed for 3 sites in Western Contour Drain (Figure 5B). There was no significant difference in breakdown rates between the reference sites.

Cotton breakdown

Nested ANOVA analysis showed that there were significant differences in cotton breakdown among sites in each catchment (F3, 324 = 9.9, p <0.001) and among catchments (F2, 4.3 = 10.1, p =

0.024) but not between seasons (F1, 4 = 1.9, p = 0.25), bag mesh size (F1, 1.5 = 2.6, p = 0.29), or period in each season (F1, 324 = 0.10, p = 0.75). As with leaf litter bag, to demonstrate differences among sites and catchments only data on cotton breakdown rate in coarse mesh size bags in period 1 in winter were further analyzed and displayed (Figure 6).

Most of the sites had cotton breakdown rates higher than that of R2 (Figure 6), 8 sites in Western Contour Drain and 6 sites in Watsons Creek were significantly higher. There was no significant difference in breakdown rates between two reference sites.

Functional stream health

The assessments of functional stream health were summarized in Table 4. Throughout the study, there were 29 and 46 measurements of functional stream health by leaf bag and cotton strip assays, respectively, which indicated that functional stream health was heavily impacted (score of 0) (Table 4). Among these, 24 measurements were identified by both methods. There was a greater number of sites that were designated as heavily impacted in winter (period 1, 2) than in spring (period 3, 4) and by using cotton strip than by using leaf bag method in all 4 periods (Table 4). Measurements indicating mildly impacted (score of 1) stream function were higher using the leaf bag in comparison to cotton strip method for all periods except period 3 (Table 4). Only 12 and 5 measurements indicated no effect on functional stream health based on leaf and cotton strip breakdown rates, respectively.

There were 36 and 47 measurements of leaf and cotton strip breakdown rates, respectively, out of 48 overall measurements that resulted in classification of heavy and mild impacts to ecosystem function; which also correlated with measures of elevated fungicide concentrations

Page 70 of 120 compared to the reference site R2 (Table 5). Forty-eight and 56 measurements, among a total of 61, were classified as heavily and mildly impacted ecosystem function based on leaf and cotton strip breakdown rates, respectively, which also correlated with measures of elevated total pesticide concentrations compared to the reference site R2 (Table 5).

5.3.3 Relationship between environmental variables and organic matter breakdown rate

Correlations between organic matter breakdown and environmental variables were reported in Table 6. Leaf breakdown rate was negatively correlated with total pesticide concentrations (r = - 0.244, p = 0.040), fungicide concentrations (r = -0.244, p = 0.040), nitrate (r = -0.259, p = 0.029) and ammonia (r = -0.308, p = 0.009) and positively correlated with insecticides (r = 0.382, p = 0.001) and temperature (r = 0.689, p < 0.001). Cotton breakdown rate was positively correlated with insecticide (r = 0.379, p = 0.001) and herbicide concentrations (r = 0.442, p < 0.001), nitrite (r = 0.284, p = 0.019) and phosphorus (r = 0.569, p < 0.001). Scatter plot analyses showed that environmental variables are not correlated to each other except fungicide and pesticide concentrations. This is not unexpected as fungicides greatly contributed to total pesticide concentrations. While correlation does not imply causality, it could provide some clarification of the complex relationship between organic matter breakdown rate and environmental variables. Therefore, significant correlations between organic matter breakdown rate and environmental variable were further investigated and confirmed through visual observations at specific sites.

The significantly positive effect of temperature on leaf breakdown was confirmed by the observation of higher leaf breakdown rate in spring than in winter at studied as well as reference sites (Figure 5).

The negative effects of fungicide concentration on leaf breakdown were observed at sites 4, 5 and 7 in Western Contour Drain in period 1 where the fungicide concentrations were three orders of magnitude higher (Figure 1A) and leaf breakdown rates were significantly lower (Figure 5A) than those of R2, respectively. This effect was also observed in Watson Creek where majority of sites (15, 16 and 19 - 24) had leaf breakdown rates significantly lower than R2 (Figure 5A) and fungicide concentrations in these sites were one or two orders of magnitude higher than that of R2 (Figure 1A).

Page 71 of 120 The negative effect of nitrate concentration on leaf breakdown rate was observed at sites 22 – 24 in Watson Creek in winter where nitrate concentrations of these sites were 2 to 3 orders of magnitude higher than R2 (Figure 3A) and leaf breakdown rates were significantly reduced (Figure 5A). In period 1, there were 3 sites (4, 5, and 20) that had ammonia concentrations one order of magnitude higher than R2 (Figure 3C) and leaf breakdown rates of these sites were significantly reduced (Figure 5A). However, two of these sites also had much higher fungicide concentration than R2.

The positive effect of phosphorus concentration on cotton breakdown was confirmed by the observation of significant increases in cotton breakdown rates in the majority of sites in period 1 (Figure 6) and the co-occurrence of high phosphorus concentrations (Figure 3D). There were 5 sites in period 1 (1, 13, 14, 16, 23) that had cotton breakdown rates not significantly different to the R2 (Figure 6) and the phosphorus concentration of these sites were also relatively low compared to the other study sites (Figure 3D).

The herbicide concentrations were relatively high at all sites in period 1 (Figure 1B) where cotton breakdown rates significantly increased (Figure 6). However, these sites also had high phosphorus concentrations and phosphorus could be the main factor affecting the breakdown rate.

5.4 Discussion 5.4.1 Predictors of organic matter breakdown

In the current study, temperature was strongly correlated with leaf breakdown, with increased temperature observed in spring accelerating leaf decomposition. This result is not unexpected as temperature is a well-known naturally controlling factor of leaf litter breakdown (Webster and Benfield, 1986; Young et al., 2008). Temperature primarily affects microbial processes of leaf litter decomposition (Webster and Benfield, 1986) by altering metabolic functions and subsequently growth and survival of microorganisms (Krauss et al., 2011). Thus, leaf breakdown is usually faster in warm water than in cold water. However, this may counteract negative effects of other stressors leading to antagonistic effects on leaf decomposition. Other than temperature, all other variables that affected leaf breakdown are anthropogenic such as pesticides and nutrients.

Page 72 of 120 Leaf breakdown rate was negatively correlated with fungicide and total pesticide concentrations (Table 6). Among 29 measurements that showed stream ecosystem function was severely impacted, 24 and 28 measurements had fungicide and total pesticide concentrations higher than that of R2, respectively (Table 4). Moreover, of these, 21 and 25 measurements which had high fungicide and total pesticide concentrations and ecosystem function was heavily impacted were detected in winter when temperature and its effects on leaf breakdown was low. The inhibition of leaf breakdown by pesticides in the current study is in agreement with other field studies by Rasmussen et al. (2012) and Schafer et al. (2012)who reported that the reduction of leaf breakdown was due to exposure to pesticides. To our knowledge, there is no field study which investigates the relationship between fungicides and leaf litter breakdown. However, several laboratory-based studies have demonstrated the inhibition of leaf breakdown by fungicides (Artigas et al., 2012; Vu et al., 2017; Zubrod et al., 2015).

Nutrient enrichment would be expected to increase leaf breakdown (Gessner and Chauvet, 2002). However, there was a negative correlation between leaf breakdown and nitrate and ammonia in the present study. Dead plant tissues are low in nitrogen (N) as they are composed primarily of carbohydrates and aromatic carbon compounds. Therefore, addition of N is expected to accelerate litter decomposition by supplying decomposer microbes with this limiting nutrient (Carreiro et al., 2000). Many leaf litter decomposition studies supported this prediction (Gulis and Suberkropp, 2003; Menéndez et al., 2003; Robinson and Gessner, 2000). However, some studies found no effects of nutrient enrichment on leaf breakdown (Hopkins et al., 2011; Perez et al., 2013; Royer and Minshall, 2001). Other studies reported the deleterious effects of high nutrients on fungi diversity and sporulation (Menéndez et al., 2011; Pascoal et al., 2005b). These differences suggested that effects of nutrients on leaf litter decomposition are strongly influenced by leaf chemistry and local stream characteristics. Carreiro et al., (2000) reported that addition of N depressed the decay rate coefficient of high-lignin red oak (Quercus rubra) litter up to 25% by reducing microbial production of ligninnolytic enzymes. The hazel leaf used in the present study has lignin content of 24% (Campbell et al., 1992), similar to that of red oak leaf (28.7%) in the study by Carreiro et al. The high lignin content in hazel leaf could be a possible explanation of the negative relationship between ammonia and nitrate and leaf breakdown.

Page 73 of 120 The positive correlation between insecticide concentration and breakdown rates was observed both in leaf and cotton in the present study. Previous studies showed stream invertebrates were sensitive to insecticides (Wallace et al., 1996) while leaf decomposer microbes did not appear to be affected by insecticides (Cuffney et al., 1990; Suberkropp and Wallace, 1992). In headwater streams where shredders are usually abundant and play important roles in leaf litter breakdown (Graca, 2001), effects of insecticides on stream invertebrates could lead to negative effects on leaf breakdown rates (Cuffney et al., 1990). However, in high-order and polluted streams where numbers of shredders are limited, fungi are the major decomposers (Pascoal et al., 2005a; Webster and Benfield, 1986). Both fungi and invertebrates consume leaf substrates and they may even compete for nutrients in leaves (Bärlocher, 1980). Furthermore, fungi may also provide shredders with nutrients not present in leaf tissue (Barlocher, 1985) and the feeding on conditioned leaves by shredders could influence the assemblages of aquatic fungi (Graca, 2001). As a result, the absence of shredders in streams due to insecticide contamination could result in positive effects on fungal community. Suberkropp and Wallace (1992) reported higher concentrations of conidia, higher fungal sporulation frequencies, and lower shredder abundance in streams treated with insecticides than in a reference stream containing a high abundance of shredders. In the present study, macroinvertebrate richness was low at all sites thus the effects of insecticides on leaf and cotton breakdown could be via fungal pathway. Further research need to investigate this secondary effect of insecticides on leaf decomposition.

The present data on leaf breakdown and correlation with biotic and abiotic factors suggested an initial survey of streams conditions including physio-chemical characteristics as well as stream community composition before deploying leaf litter bag to streams to determine an appropriate leaf deployment method (e.g., deployment time, and mesh type) and all requirements information on stream characteristics. For example, species richness was low in all streams in the current study could suggest microorganism may play more important role in leaf litter breakdown. Therefore, only one type of mesh bag needs to be used and the analysis of the microorganism community could provide better information than the macroinvertebrate community on the effect of anthropogenic stressors on leaf litter breakdown. Furthermore, if the microorganisms play an important role in leaf litter breakdown in study stream, a longer deployment period may require as it will take a longer time to breakdown the leaf litter mainly by microorganism. Study season is another factor need to be considered as temperature has a very strong influence of leaf litter

Page 74 of 120 breakdown and this effect could counteract with other effects of anthropogenic stressors making the interpretation of data become more difficult. To avoid this situation, a reference site should be in a similar temperature range with study sites and an all year round study could be done if possible because it will provide full understanding of leaf breakdown in a wide temperature range with a variety of weather conditions.

In contrast to leaf breakdown, cotton breakdown was accelerated by all stressors and the breakdown rate was significantly correlated with insecticide and herbicide concentrations, nitrite and phosphorus (Table 6). Among all the measured parameters, phosphorus has the strongest correlation with cotton breakdown rate. Other studies have found cotton breakdown rate was correlated with phosphorus and nitrogen concentrations (Imberger et al., 2010; Niyogi et al., 2013; Schäfer et al., 2012; Young, 2006). The palatability of cotton strips and the extent to which stream invertebrates feed on them are not well known (Tiegs et al., 2007) but cotton is unlikely to be a preferred food source for stream invertebrates (Young, 2006). Thus, the breakdown of cotton is mostly controlled by microbial activity rather than through invertebrate shredding and affected by nutrient levels in aquatic environments.

Like insecticides, herbicides had a positive correlation with the cotton breakdown rate. To our knowledge, there are no available studies on effects of herbicides and insecticides on cotton (cellulose) breakdown in aquatic environments. However, a terrestrial study found that herbicides and insecticides have no inhibitory effects on cellulose degradation (Katayama and Kuwatsuka, 1991). Furthermore, it has been shown that microbial communities were able to use herbicides as an energy source (Pesce et al., 2009) or utilized the organic proteinaceous compounds released from alga cell lysis because of herbicide exposure, to support their growth (Ricart et al., 2009). These could be a plausible explanation for the positive relationship between herbicide concentrations and cotton breakdown rates in the current studies.

5.4.2 Can cotton be used as a standard type of organic matter to measure effects of anthropogenic stressors on ecosystem function in aquatic environments?

Key criteria for using cotton as proxies for natural leaves are:

Page 75 of 120 1. Breakdown rates of the two materials are related 2. The materials decay in a similar way when exposed to the same environmental conditions (Tiegs et al., 2007)

In the present study, leaf and cotton breakdown rates were not correlated. The number of studies comparing leaf and cotton breakdown were limited and reported different results. A few studies found a weak correlation between leaf mass loss and loss of tensile strength (Niyogi et al., 2013; Schäfer et al., 2012; Tiegs et al., 2007). In contrast, a study by Young (2006) found no correlation between leaf mass loss and tensile strength loss. This difference could be the result of different leaf species used in each study and more importantly the stream characteristics. The correlation between cotton breakdown rate and phosphorus concentrations was observed in the current study and other studies (Imberger et al., 2010; Niyogi et al., 2013; Schäfer et al., 2012). Phosphorus concentration in our study and in the study by Young (2006) were in mg/L range which were much higher than that of other studies, in µg/L range (Imberger et al., 2010; Tiegs et al., 2007). Furthermore, phosphorus has been shown to have significant effects on bacteria growth (Miettinen et al., 1997) and organic matter decomposition (Elwood et al., 1981). As a result, cotton decomposed faster at higher phosphorus concentrations.

Not only the breakdown rates were not correlated, but leaf and cotton also decayed differently. The current study was in agreement with the study by Young (2006) as we observed the contrast between leaf mass loss and tensile strength loss; cotton decomposed much faster in the study sites than at the reference site while the leaf degraded normally or slower than at the reference site. Leaf litter breakdown process and controlling factors are well studied in the literature (Melillo et al., 1982; Petersen and Cummins, 1974; Saunders, 1975; Webster and Benfield, 1986) but little is known about cellulose breakdown mechanisms in aquatic environments (Imberger et al., 2010). Many studies demonstrated the substantial contribution of stream invertebrates and fungi to leaf litter breakdown (Gessner and Chauvet, 1994; Graca, 2001; Hieber and Gessner, 2002; Pascoal et al., 2005a). In contrast, the extent to which fungi colonize and invertebrates consume cotton in streams is poorly characterized (Imberger et al., 2010; Young, 2006). However, cotton could be fragmented by invertebrates through other activities rather than shredding. For example, cotton cloth was found to be a particularly favorable substrate into which the chironomid larvae are able to borrow (Friberg and Winterbourn, 1997; Tank and

Page 76 of 120 Winterbourn, 1996). In the present study, we also observed oligochaete and chironomid larvae burrowed in the cotton strips and faster cotton breakdown at sites had a high density of these animals. Other taxon that could contribute to leaf and cotton breakdown is bacteria. While bacteria had a smaller role on leaf litter breakdown than invertebrates and fungi (Das et al., 2007; Hieber and Gessner, 2002), studies showed they were well colonized and played an important role in cellulose breakdown (Ardón and Pringle, 2007; Friberg and Winterbourn, 1997; Young, 2006). The different roles of invertebrates, fungi, and bacteria on leaf and cotton decomposition suggested that cotton breakdown measure a different combination of ecological processes than leaf breakdown (Tiegs et al., 2007; Young, 2006). In addition to biotic factors, material composition is an abiotic factor could contribute to the differences in breakdown mechanism between leaf litter and cotton strips. Cellulose is a major constituent of cotton (~ 95%) and contributes to a large proportion of leaf composition (70%) (Imberger et al., 2010). Leaf litter is also composed of lignin and small leachable components such as lipids and nutrients (Campbell et al., 1992; Fioretto et al., 2005). Therefore, initial breakdown loss (leaching phase) form leaves will be faster than cotton and independent of fungi/invertebrates but overall leaf could lose less mass in a specified time as the full degradation of leaf requires a suite of enzymes (Tiegs et al., 2007).

Nevertheless, the functional stream health assessment using cotton strip assays reflected the effects of pesticides better than using leaf bag assays (Table 5). The cotton strips also had some advantages in term of responses to environmental variables that could be considered as a promising candidate for biomonitoring programs. Firstly, there were predictable relationships between cotton breakdown rate and anthropogenic stressors such as nutrients and pesticides, although controlled experiments to establish causality is required to investigate these relationships further. Secondly, all correlated environmental stressors resulted in positive effects on cotton breakdown making cotton decompose faster at polluted sites. However, the positive effects on leaf breakdown (e.g. insecticides) could be counteracted by negative effects (e.g. fungicides) and make it difficult to distinguish between healthy and unhealthy ecosystem because this antagonistic effect could make leaves at polluted sites have similar breakdown rates to the reference site. Finally, cotton strips are less sensitive to natural variables such as flow rate (Egglishaw, 1972) or temperature as demonstrated in the current study and study by Niyogi et al., (2013) which make anthropogenic impact assessments more reliable. However, cotton strips

Page 77 of 120 assay is not without shortcomings. Vyšná et al. (2014) found a large within – site variability in cotton decomposition rates and suggested that this a serious limitation to the assessment using reference condition approach. There is also no precise/realistic standardized mathematical model to fits cotton decomposition through time (Imberger et al., 2010). The relationship between cotton decomposition rate, temperature, and time could be more complex than exponential decay model which normally use in the literature (Vyšná et al., 2014). Therefore, the effects of biotic and abiotic factors on cellulose decomposition need to be understood before cotton trip assay becomes a useful and standardized ecological indicator.

5.5 Conclusion

Leaf and cotton breakdown demonstrated effects of pesticides and other stressors on one aspect of ecosystem function. However, multiple stressors may interact resulting in either synergistic or antagonistic effects. Cotton decomposition has potential as a functional indicator of anthropogenic stressors in aquatic environments and could be used as a standardized diagnostic tool in stream health monitoring program. Further research investigating the role of different animal groups and factors controlling cellulose breakdown in aquatic environments could better clarify the effects of anthropogenic stressors on cotton decomposition.

Page 78 of 120 5.6 Figures

A

B

C

Figure 1: Fungicide (A), herbicide (B), and insecticide (C) concentrations measured from passive samplers deployed at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) in winter in 2015, blue bars represent period 1, red bars represent period 2.

Page 79 of 120 A

B

C

Figure 2: Fungicide (A), herbicide (B), and insecticide (C) concentrations measured from passive samplers deployed at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) in spring in 2015, purple bars represent period 3, green bars represent period 4.

Page 80 of 120 A B

C D

Figure 3: Nitrite (A), nitrate (B), ammonia (C), phosphorus (D) concentrations measured at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) in winter in 2015, blue bars represent period 1, red bars represent period 2.

Page 81 of 120 A B

C D

Figure 4: Nitrite (A), nitrate (B), ammonia (C), phosphorus (D) concentrations measured at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) in spring in 2015, purple bars represent period 3, green bars represent period 4.

Page 82 of 120 A 0.015

0.012

0.009 * 0.006

0.003 Breakdown rate (1/d)rate Breakdown 0 R1 R2 1 3 4 5 7 8 10 11 12 13 14 15 16 19 20 21 22 23 24 Site

B .05

.04

.03

.02

.01 Breakdownrate(1/d) .00 R1 R2 1 2 3 5 6 9 11 12 16 17 18 19 23 Site

Figure 5: Leaf breakdown rate (mean ± SE) in coarse mesh bags at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) during period 1 in winter (A) and period 4 in spring 2015 (B). (*) indicate significant difference from R2, p < 0.05, n = 5.

Page 83 of 120 0.16 0.14 0.12 0.1 0.08 * 0.06 0.04 Breakdown rate (1/d) rate Breakdown 0.02 0 R1 R2 1 3 4 5 7 8 10 11 12 13 14 15 16 19 20 21 22 23 24 Site

Figure 6: Cotton breakdown rate (mean ± SE) in coarse mesh bags at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) during period 1 in winter 2015. (*) indicate significant difference from R2, p < 0.05, n = 5.

Page 84 of 120 5.7 Tables

Table 1: Site details for deployment schedule for winter and spring.

Catchment Site Lat Long Winter Spring Code Cardinina Creek REF1 -37.984889 145.381934 x x Cardinina Creek REF2 -38.119202 145.401702 x x Western Contour Drain 1 -38.104245 145.334895 x x Western Contour Drain 2 -38.116544 145.343177 x Western Contour Drain 3 -38.121199 145.344932 x x Western Contour Drain 4 -38.121399 145.345599 x Western Contour Drain 5 -38.12126 145.345815 x x Western Contour Drain 6 -38.12145 145.3461 x Western Contour Drain 7 -38.138261 145.364195 x Western Contour Drain 8 -38.137316 145.356004 x Western Contour Drain 9 -38.142499 145.36109 x Western Contour Drain 10 -38.158925 145.36129 x Western Contour Drain 11 -38.167073 145.359399 x x Western Contour Drain 12 -38.230494 145.236436 x x Watsons Creek 13 -38.19962 145.154898 x Watsons Creek 14 -38.206649 145.163888 x Watsons Creek 15 -38.205129 145.176044 x Watsons Creek 16 -38.205144 145.175932 x x Watsons Creek 17 -38.209675 145.188317 x Watsons Creek 18 -38.209961 145.190291 x Watsons Creek 19 -38.218118 145.195344 x x Watsons Creek 20 -38.218258 145.195228 x Watsons Creek 21 -38.224075 145.2037 x Watsons Creek 22 -38.226736 145.216876 x Watsons Creek 23 -38.231117 145.239641 x x Watsons Creek 24 -38.231543 145.240174 x

Table 2: Framework for assessing functional stream health using organic matter breakdown rate. ksite/kR2 Score Functional stream health status

<0.5 or >2.0 0 Severely impaired ecosystem function

0.5–0.75 or 1.33–2.0 1 Mild effect on ecosystem function

0.75–1.33 2 No evidence of an impact on ecosystem function

Page 85 of 120 Table 3: Pesticides detected in Chemcatcher® passive samplers during the study. Pesticide Type: H = Herbicide, F = Fungicide, I – Insecticide. Limit of reporting 0.01 µg/disc, n = 72.

Pesticide Type %Detects Min Max Median Simazine H 93.1 0.016 4.7 0.5 Diuron H 75.0 0.011 0.98 0.04 Iprodione F 70.8 0.011 2.4 0.052 Metolachlor H 68.1 0.011 6.2 0.7 Prometryn H 62.5 0.037 6.7 0.55 Linuron H 55.6 0.011 0.64 0.051 Metalaxyl F 51.4 0.011 2.1 0.15 Atrazine H 45.8 0.01 0.99 0.017 Procymidone F 41.7 0.011 3.6 0.16 Chlorothalonil F 40.3 0.014 110 0.93 Dimethomorph F 36.1 0.024 3.1 0.21 Tebuconazole F 30.6 0.012 0.17 0.027 Diazinon I 25.0 0.011 4.6 0.056 Dimethoate I 22.2 0.01 0.38 0.043 Propiconazole_II F 20.8 0.011 0.03 0.016 Boscalid F 18.1 0.54 15 2.4 Fenamiphos F 16.7 0.011 0.14 0.016 Difenoconazole F 16.7 0.01 0.27 0.032 Propiconazole_I F 16.7 0.01 0.025 0.015 Cyprodinil F 12.5 0.01 0.21 0.018 Carbaryl I 8.3 0.013 1.3 0.185 Pirimicarb I 8.3 0.011 1.9 0.172 Buprofezin I 6.9 0.01 0.015 0.012 Metribuzine H 6.9 0.018 0.047 0.023 Propiconazole_I_II F 6.9 0.014 0.07 0.028 Prochloraz F 5.6 0.011 0.042 0.026 Pendimethalin H 5.6 0.031 0.29 0.052 Methoprene I 4.2 0.011 0.02 0.019 Azinphos_ethyl I 2.8 0.019 0.021 0.02 Phorate I 2.8 0.012 0.02 0.016 Thiometon I 2.8 0.012 0.032 0.022 Triazophos I 2.8 0.012 0.013 0.013 Permethrin I 2.8 0.012 0.013 0.013 Bupirimate F 2.8 0.013 0.026 0.02 Chlorpyrifos I 1.4 0.25 0.25 0.25 Malathion I 1.4 0.018 0.018 0.018 Fenitrothion I 1.4 0.026 0.026 0.026 Azinphos_methyl I 1.4 0.068 0.068 0.068 Fenchlorphos I 1.4 0.05 0.05 0.05 Deltamethrin I 1.4 0.012 0.012 0.012 Diphenylamine F 1.4 0.043 0.043 0.043 Imazalil F 1.4 0.01 0.01 0.01 Hexazinone H 1.4 0.023 0.023 0.023

Page 86 of 120 Table 4: Functional stream health assessments during the study (period 1-2 in winter and period 3-4 in spring) using leaf breakdown rate (L) and cotton breakdown rate (C) and both (B), value indicates number of measurements.

Period Score* L C B 0 12 14 10 1 1 5 3 2 2 2 2 0 0 14 17 12 2 1 2 1 0 2 3 1 0 0 2 6 1 3 1 4 5 2 2 6 1 0 0 1 9 1 4 1 11 3 3 2 1 1 0 0 29 46 24 All 4 period 1 22 12 7 2 12 5 0

(*) Score of 0: severely impaired ecosystem function, score of 1: mild effect on ecosystem function, score of 2: no evidence of an impact on ecosystem function. These scores are relative to the reference R2.

Table 5: Functional stream health assessment in relation to fungicide and total pesticide concentrations for all 4 study periods.

Variable Number of sites with Material Number of sites Number of sites concentrations higher (H) classified as classified as no or lower (L)than R2 impacted a impacted b Leaf 36 12 H (48) Fungicides Cotton 47 1 Leaf 15 0 L (15) Cotton 11 4 Leaf 48 13 H (61) Pesticides Cotton 56 5 Leaf 2 0 L (2) Cotton 2 0

a: A combination of heavily and mildly impacted sites which had score of 0 or 1 b: No impacted sites which had score of 2

Page 87 of 120 Table 6: Correlation coefficients (r) and significance levels (p) for breakdown rate of leaf and cotton with different environmental variables.

Leaf (n= 71) Cotton (n= 68) Variables r p r p Total pesticide (µg/disk) -0.244 0.040 0.211 0.084 Insecticides (µg/disk) 0.382 0.001 0.379 0.001 Herbicides (µg/disk) -0.152 0.207 0.442 <0.001 Fungicides (µg/disk) -0.244 0.040 0.144 0.241 NO2 (mg/L) -0.109 0.365 0.284 0.019 NO3 (mg/L) -0.259 0.029 0.173 0.157 NH4 (mg/L) -0.308 0.009 0.175 0.154 FRP (mg/L) 0.048 0.693 0.569 <0.001 T (oC) 0.689 <0.001 0.224 0.066 pH -0.215 0.071 0.042 0.736 EC (µS/cm) 0.166 0.168 0.065 0.597 Dissolved oxygen (% sat) 0.172 0.150 0.041 0.739 Turbidity (NTU) -0.030 0.803 0.144 0.242 Species richness 0.007 0.951 0.010 0.936

Page 88 of 120 5.8 Supplemental data

Table S1: Physico-chemistry parameters and nutrients during the study.

Period Variable Min Max Median Mean %SD NO2 (mg/L) 0.005 0.230 0.033 0.052 103.846 NO3 (mg/L) 0.028 78.900 2.450 7.563 237.617 NH4 (mg/L) 0.023 0.550 0.048 0.098 123.469 1 FRP (mg/L) 0.006 3.085 0.298 0.559 145.081 (n = 21) T (oC) 7.868 11.843 10.113 9.862 11.955 pH 6.668 8.540 7.778 7.648 5.845 EC (µS/cm) 212.250 5222.250 2160.750 2342.738 61.071 Dissolved oxygen 62.450 99.075 81.750 81.120 12.357 (% sat) Turbidity (NTU) 3.925 902.900 19.925 97.821 220.560 NO2 (mg/L) 0.008 0.460 0.065 0.094 118.085 NO3 (mg/L) 0.005 25.300 4.095 5.994 117.334 NH4 (mg/L) 0.005 0.505 0.008 0.052 219.231 2 FRP (mg/L) 0.011 2.620 0.350 0.709 106.629 (n = 21) T (oC) 8.313 11.698 9.830 9.886 9.468 pH 7.050 8.668 7.833 7.742 5.619 EC (µS/cm) 191.000 3322.750 1240.500 1470.833 62.910 Dissolved oxygen 55.400 103.575 80.500 78.411 15.838 (% sat) Turbidity (NTU) 11.350 987.875 44.100 115.664 203.291 NO2 (mg/L) 0.005 0.178 0.008 0.035 145.714 NO3 (mg/L) 0.005 17.675 0.005 2.245 236.837 NH4 (mg/L) 0.005 0.178 0.005 0.034 164.706 3 FRP (mg/L) 0.005 2.968 0.220 0.925 115.351 (n = 15) T (oC) 12.800 25.000 16.000 16.720 21.328 pH 6.430 8.615 7.475 7.471 9.115 EC (µS/cm) 91.750 10440 1964.500 3218.617 100.033 Dissolved oxygen 42.400 215.050 84.250 89.787 47.363 (% sat) Turbidity (NTU) 6.650 413.500 19.550 51.930 196.126 NO2 (mg/L) 0.005 0.228 0.013 0.053 139.623 NO3 (mg/L) 0.005 14.775 0.005 2.186 228.088 NH4 (mg/L) 0.005 0.180 0.005 0.024 200.000 4 FRP (mg/L) 0.005 2.690 0.178 0.765 124.183 (n = 15) T (oC) 15.330 26.760 19.595 20.220 17.052 pH 4.565 8.275 7.650 6.807 21.272 EC (µS/cm) 84.000 56025 2008.000 5572.267 251.349 Dissolved oxygen 31.700 144.675 81.850 82.235 36.295 (% sat) Turbidity (NTU) 4.750 1000.000 23.600 131.777 196.466 (FRP) Filterable reactive phosphorus

Page 89 of 120 Table S2: Taxonomic information for each site in period 1.

Taxa Site Major Lowest taxonomic R R 1 1 1 1 1 1 1 1 2 2 2 2 group group 1 2 1 3 4 5 7 8 0 1 2 3 4 5 6 9 1 2 3 4 Nematoda Nematoda spp. 1 3 17 Oligochaeta Oligochaeta spp. 2 2 4 8 7 7 5 2 8 1 3 4 2 1 Turberllaria Dugesiidae spp. 1 7 3 2 3 4 Nemertea Prostoma spp. 3 3 Hirudinea Glossiphonidae immatures 1 3 Helobedella papillornata 1 2 1 1 Acarina Acarina spp. Mollusca Gastropoda immatures Potamopyrgus antipodarum 2 9 5 7 1 2 1 1 2 3 2 Physella acuta 2 1 2 Glyptophysa (Glyptophysa) 2 Crustacea Ceinidae immatures Austrochiltonia 3 subtenuis 1 1 0 6 Amarinus lacustris 1 2 1 Heterias spp. Hemiptera Sigara spp. 1

Page 90 of 120 Table S2: Taxonomic information for each site in period 1 (continued)

Taxa Site R R 1 1 1 1 1 1 1 1 2 2 2 2 Major group Lowest taxonomic group 1 2 1 3 4 5 7 8 0 1 2 3 4 5 6 9 1 2 3 4 Coleoptera Georissus spp. Limnoxenus spp. (l) 1 Notriolus quadriplagiatus (l)

Notriolus spp. (l) Antiporus spp. (l) Ephemeroptera Leptophlebiidae immatures Atalophlebia spp. Atalophlebia sp AV5 Trichoptera Symphitoneuria opposita 3 Triplectides spp.

Triplectides truncatus 1

Triplectides ciuskis

ciuskus 6 Hydrobiosiidae immatures Odonata Coenagrionidae immatures 1 Ishnura spp. 1 1

Austroargiolestes

icteromelas/calcaris 3 Hemicorduliidae immatures Hemicordulia tau 1 1

Page 91 of 120 Table S2: Taxonomic information for each site in period 1 (continued)

Taxa Site Major Lowest taxonomic R R 1 1 1 1 1 1 1 1 2 2 2 2 group group 1 2 1 3 4 5 7 8 0 1 2 3 4 5 6 9 1 2 3 4 Diptera Stratiomyidae spp. 1 Ceratopogonidae spp. 1

Tipulidae spp. 1

Psychodidae spp. Cladotanytarus spp. 1 Tanytarsus spp. 1 Paratanytarsus spp. Chironomus spp. 8 5 2 3 2

Polypedilum spp. 1 2 Kiefferulus spp. Dicrotendipes spp. Orthocladiinae immatures Cricotopus spp. Tanypodinae immatures 1 Paralimnophes spp. 1 Procladius spp. 1

Page 92 of 120 Table S3: Taxonomic information for each site in period 2

Taxa Site Major group Lowest taxonomic group R1 R2 1 3 4 5 7 8 10 11 13 16 20 21 22 23 24 Nematoda Nematoda spp. 7 1 Oligochaeta Oligochaeta spp. 8 23 181 93 12 82 7 1 29 2 11 Turberllaria Dugesiidae spp. 1 1 1 2 Nemertea Prostoma spp. Hirudinea Glossiphonidae immatures 1 Helobedella papillornata 1 1 Acarina Acarina spp. 1 Mollusca Gastropoda immatures 2 Potamopyrgus antipodarum 1 2 3 29 7 11 10 4 2

Physella acuta 1 1 1 1

Glyptophysa (Glyptophysa) 1 Crustacea Ceinidae immatures Austrochiltonia subtenuis 2 3 4 1 1

Amarinus lacustris 5 5

Heterias spp. 1 1 Hemiptera Sigara spp.

Page 93 of 120 Table S3: Taxonomic information for each site in period 2 (continued).

Taxa Site R R 1 1 1 1 2 2 2 2 2 Major group Lowest taxonomic group 1 2 1 3 4 5 7 8 0 1 3 6 0 1 2 3 4 Coleoptera Georissus spp. Limnoxenus spp. (l)

Notriolus quadriplagiatus (l)

Notriolus spp. (l) 1 Antiporus spp. (l) Ephemeroptera Leptophlebiidae immatures 4 Atalophlebia spp. 1

Atalophlebia sp AV5 Trichoptera Symphitoneuria opposita 1 Triplectides spp. 1

Triplectides truncatus

Triplectides ciuskis ciuskus 8 Hydrobiosiidae immatures 1 Odonata Coenagrionidae immatures 1 Ishnura spp. 2 Austroargiolestes icteromelas/calcaris 1

Hemicorduliidae immatures Hemicordulia tau

Page 94 of 120 Table S3: Taxonomic information for each site in period 2 (continued).

Taxa Site Major group Lowest taxonomic group R1 R2 1 3 4 5 7 8 10 11 13 16 20 21 22 23 24 Diptera Stratiomyidae spp. Ceratopogonidae spp.

Tipulidae spp. 1 1

Psychodidae spp. 1 Cladotanytarus spp. Tanytarsus spp. 1 Paratanytarsus spp. Chironomus spp. 20 3 14 3 6

Polypedilum spp. 1 1 Kiefferulus spp. 1 Dicrotendipes spp. Orthocladiinae immatures Cricotopus spp.

Tanypodinae immatures Paralimnophes spp. Procladius spp.

Page 95 of 120 Table S4: Taxonomic information for each site in period 3.

Taxa Site Major group Lowest taxonomic group R2 1 2 3 5 6 9 11 12 16 17 18 19 23 Nematoda Nematoda spp. Oligochaeta Oligochaeta spp. 1 3 1 3 Turberllaria Dugesiidae spp. 1 Nemertea Prostoma spp. 1 Hirudinea Glossiphonidae immatures Helobedella papillornata 1 2 1 1 Acarina Acarina spp. Mollusca Gastropoda immatures 1 Potamopyrgus antipodarum 6 1 1 2 4 2 3 3 Physella acuta 2 2 1 1 Glyptophysa (Glyptophysa) Crustacea Ceinidae immatures Austrochiltonia subtenuis 1 7 2 1 5 3 Amarinus lacustris 2 Heterias spp. 3 Hemiptera Sigara spp.

Page 96 of 120 Table S4: Taxonomic information for each site in period 3 (continued).

Taxa Site Major group Lowest taxonomic group R2 1 2 3 5 6 9 11 12 16 17 18 19 23 Coleoptera Georissus spp. 1 Limnoxenus spp. (l) Notriolus quadriplagiatus (l) 1 Notriolus spp. (l) Antiporus spp. (l) 1 Ephemeroptera Leptophlebiidae immatures Atalophlebia spp. 1 Atalophlebia sp AV5 Trichoptera Symphitoneuria opposita Triplectides spp. Triplectides truncatus Triplectides ciuskis ciuskus 3 Hydrobiosiidae immatures Odonata Coenagrionidae immatures Ishnura spp. 1 Austroargiolestes icteromelas/calcaris 1 Hemicorduliidae immatures Hemicordulia tau

Page 97 of 120 Table S4: Taxonomic information for each site in period 3 (continued).

Taxa Site Major group Lowest taxonomic group R2 1 2 3 5 6 9 11 12 16 17 18 19 23 Diptera Stratiomyidae spp. Ceratopogonidae spp. Tipulidae spp. Psychodidae spp. Cladotanytarus spp. Tanytarsus spp. Paratanytarsus spp. 1 Chironomus spp. 1 5 6 9 1 2 6 7 2 2 2 5 Polypedilum spp. 4 1 Kiefferulus spp. Dicrotendipes spp. 3 Orthocladiinae immatures 1 Cricotopus spp. 1 Tanypodinae immatures Paralimnophes spp. Procladius spp.

Page 98 of 120 Table S5: Taxonomic information for each site in period 4.

Taxa Site Major group Lowest taxonomic group R1 R2 1 2 3 5 9 11 12 16 17 18 19 Nematoda Nematoda spp. Oligochaeta Oligochaeta spp. 1 1 1 Turberllaria Dugesiidae spp. 1 Nemertea Prostoma spp. Hirudinea Glossiphonidae immatures Helobedella papillornata 1 4 Acarina Acarina spp. Mollusca Gastropoda immatures Potamopyrgus antipodarum 1 1 3 3 5 1

Physella acuta 1 2 2

Glyptophysa (Glyptophysa) Crustacea Ceinidae immatures 1 Austrochiltonia subtenuis 1 2 2 3 1 2 4

Amarinus lacustris

Heterias spp. Hemiptera Sigara spp.

Page 99 of 120 Table S5: Taxonomic information for each site in period 4 (continued).

Taxa Site Major group Lowest taxonomic group R1 R2 1 2 3 5 9 11 12 16 17 18 19 Coleoptera Georissus spp. Limnoxenus spp. (l)

Notriolus quadriplagiatus (l)

Notriolus spp. (l) Antiporus spp. (l) Ephemeroptera Leptophlebiidae immatures Atalophlebia spp.

Atalophlebia sp AV5 Trichoptera Symphitoneuria opposita 1 9 Triplectides spp.

Triplectides truncatus

Triplectides ciuskis ciuskus 1 Hydrobiosiidae immatures Odonata Coenagrionidae immatures 1 Ishnura spp.

Austroargiolestes icteromelas/calcaris

Hemicorduliidae immatures 2 Hemicordulia tau

Page 100 of 120 Table S5: Taxonomic information for each site in period 4 (continued).

Taxa Site Major group Lowest taxonomic group R1 R2 1 2 3 5 9 11 12 16 17 18 19 Diptera Stratiomyidae spp. Ceratopogonidae spp. Tipulidae spp. Psychodidae spp. Cladotanytarus spp. Tanytarsus spp. Paratanytarsus spp. 1 Chironomus spp. 2 2 2 6 1 5 4 3 6 4 6 1 Polypedilum spp. 1 Kiefferulus spp. Dicrotendipes spp. 1 Orthocladiinae immatures Cricotopus spp. Tanypodinae immatures Paralimnophes spp. Procladius spp.

Page 101 of 120 A Period 1 12 Period 2 10

8

6

4 Species richness Species 2

0 R1 R2 1 3 4 5 7 8 10 11 12 13 14 15 16 19 20 21 22 23 24 Site

B 9 Period 3 8 7 Period 4 6 5 4 3

Species richness Species 2 1 0 R1 R2 1 2 3 5 6 9 11 12 16 17 18 19 23 Site

Figure S1: Number of macroinvertebrate species in litter bags at sites in Western Contour Drain (1-12), Watsons Creek (13 - 24), and Cardinia Creek (R1& R2) in winter (A) and spring (B).

Page 102 of 120 Acknowledgement: We would like to thank K. Jeppe for kindly providing hazel leaves for this study. We would like to thank R. Boyle, G. Sinclair, J. French, H. Eason, P. Bonney, and J. Boewater for assisting with sample processing. Funding for this research was supported by Melbourne Water, Centre for Aquatic Pollution Identification and Management (CAPIM), and Holsworth Wildlife Research Endowment. Hung T. Vu was funded Melbourne International Research Scholarship throughout this study.

REFERENCES

ANZECC, ARMCANZ, 2000. Australian and New Zealand guidelines for fresh and marine water quality. ANZECC & ARMCANA, Canberra. Ardón, M., Pringle, C.M., 2007. The quality of organic matter mediates the response of heterotrophic biofilms to phosphorus enrichment of the water column and substratum. Freshwater Biology 52, 1762-1772. Artigas, J., Majerholc, J., Foulquier, A., Margoum, C., Volat, B., Neyra, M., Pesce, S., 2012. Effects of the fungicide tebuconazole on microbial capacities for litter breakdown in streams. Aquat Toxicol 122, 197-205. Barlocher, F., 1985. The role of fungi in the nutrition of stream invertebrates. Botanical Journal of the Linnean Society 91, 83-94. Bärlocher, F., 1980. Leaf-eating invertebrates as competitors of aquatic hyphomycetes. Oecologia 47, 303-306. Bunn, S.E., Davies, P.M., 2000. Biological processes in running waters and their implications for the assessment of ecological integrity, Assessing the Ecological Integrity of Running Waters. Springer, pp. 61-70. Campbell, I.C., James, K.R., Hart, B.T., Devereaux, A., 1992. Allochthonous coarse particulate organic material in forest and pasture reaches of two south-eastern Australian streams. II. Litter processing. Freshwater Biology 27, 353-365. Carreiro, M., Sinsabaugh, R., Repert, D., Parkhurst, D., 2000. Microbial enzyme shifts explain litter decay responses to simulated nitrogen deposition. Ecology 81, 2359-2365. Cuffney, T.F., Wallace, J.B., Lugthart, G., 1990. Experimental evidence quantifying the role of benthic invertebrates in organic matter dynamics of headwater streams. Freshwater Biology 23, 281-299.

Page 103 of 120 Das, M., Royer, T.V., Leff, L.G., 2007. Diversity of fungi, bacteria, and actinomycetes on leaves decomposing in a stream. Applied and Environmental Microbiology 73, 756- 767. Egglishaw, H., 1972. An experimental study of the breakdown of cellulose in fast- flowing streams. Memorie dell’Istituto Italiano di Idrobiologia 29, 405-428. Elwood, J.W., Newbold, J.D., Trimble, A.F., Stark, R.W., 1981. The limiting role of phosphorus in a woodland stream ecosystem: effects of P enrichment on leaf decomposition and primary producers. Ecology 62, 146-158. Fioretto, A., Di Nardo, C., Papa, S., Fuggi, A., 2005. Lignin and cellulose degradation and nitrogen dynamics during decomposition of three leaf litter species in a Mediterranean ecosystem. Soil Biology and Biochemistry 37, 1083-1091. Friberg, N., Winterbourn, M.J., 1997. Effects of native and exotic forest on benthic stream biota in New Zealand: a colonization study. Marine and Freshwater Research 48, 267-275. Gessner, M.O., Chauvet, E., 1994. Importance of stream microfungi in controlling breakdown rates of leaf litter. Ecology 75, 1807-1817. Gessner, M.O., Chauvet, E., 2002. A case for using litter breakdown to assess functional stream integrity. Ecological Applications 12, 498-510. Girling, A.E., Pascoe, D., Janssen, C.R., Peither, A., Wenzel, A., Schafer, H., Neumeier, B., Mitchell, G.C., Taylor, E.J., Maund, S.J., Lay, J.P., Juttner, I., Crossland, N.O., Stephenson, R.R., Personne, G., 2000. Development of methods for evaluating toxicity to freshwater ecosystems. Ecotoxicology and Environmental Safety 45. Graca, M.A.S., 2001. The role of invertebrates on leaf litter decomposition in streams - a review. International Review of Hydrobiology 86, 383-393. Gray, L., Ward, J., 1983. Leaf litter breakdown in streams receiving treated and untreated metal mine drainage. Environment International 9, 135-138. Gulis, V., Suberkropp, K., 2003. Leaf litter decomposition and microbial activity in nutrient‐enriched and unaltered reaches of a headwater stream. Freshwater Biology 48, 123-134. Hieber, M., Gessner, M.O., 2002. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology 83, 1026-1038.

Page 104 of 120 Hopkins, J.M., Marcarelli, A.M., Bechtold, H.A., 2011. Ecosystem structure and function are complementary measures of water quality in a polluted, spring-influenced river. Water Air and Soil Pollution 214, 409-421. Imberger, S.J., Thompson, R.M., Grace, M.R., 2010. Searching for effective indicators of ecosystem function in urban streams: assessing cellulose decomposition potential. Freshwater Biology 55, 2089-2106. Imberger, S.J., Walsh, C.J., Grace, M.R., 2008. More microbial activity, not abrasive flow or shredder abundance, accelerates breakdown of labile leaf litter in urban streams. Journal of the North American Benthological Society 27, 549-561. Katayama, A., Kuwatsuka, S., 1991. Effect of pesticides on cellulose degradation in soil under upland and flooded conditions. Soil Science and Plant Nutrition 37, 1-6. Kingston, J.K., Greenwood, R., Mills, G.A., Morrison, G.M., Persson, L.B., 2000. Development of a novel passive sampling system for the time-averaged measurement of a range of organic pollutants in aquatic environments. Journal of Environmental Monitoring 2, 487-495. Knacker, T., Förster, B., Römbke, J., Frampton, G.K., 2003. Assessing the effects of plant protection products on organic matter breakdown in arable fields—litter decomposition test systems. Soil Biology and Biochemistry 35, 1269-1287. Krauss, G.J., Sole, M., Krauss, G., Schlosser, D., Wesenberg, D., Barlocher, F., 2011. Fungi in freshwaters: ecology, physiology and biochemical potential. FEMS Microbiology Reviews 35, 620-651. Liess, M., Ohe, P.C.V.D., 2005. Analyzing effects of pesticides on invertebrate communities in streams. Environmental Toxicology and Chemistry 24, 954-965. Maltby, L., Booth, R., 1991. The effect of coal-mine effluent on fungal assemblages and leaf breakdown. Water Research 25, 247-250. Melillo, J.M., Aber, J.D., Muratore, J.F., 1982. Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63, 621-626. Menéndez, M., Descals, E., Riera, T., Moya, O., 2011. Leaf litter breakdown in Mediterranean streams: effect of dissolved inorganic nutrients. Hydrobiologia 669, 143- 155.

Page 105 of 120 Menéndez, M., Hernández, O., Comín, F.A., 2003. Seasonal comparisons of leaf processing rates in two Mediterranean rivers with different nutrient availability. Hydrobiologia 495, 159-169. Miettinen, I.T., Vartiainen, T., Martikainen, P.J., 1997. Phosphorus and bacterial growth in drinking water. Applied and Environmental Microbiology 63, 3242-3245. Nelson, S., 2000. Leaf pack breakdown and macroinvertebrate colonization: bioassessment tools for a high-altitude regulated system? Environmental Pollution 110, 321-329. Niyogi, D.K., Harding, J.S., Simon, K.S., 2013. Organic matter breakdown as a measure of stream health in New Zealand streams affected by acid mine drainage. Ecological Indicators 24, 510-517. Pascoal, C., Cassio, F., Marcotegui, A., Sanz, B., Gomes, P., 2005a. Role of fungi, bacteria, and invertebrates in leaf litter breakdown in a polluted river. Journal of the North American Benthological Society 24, 784-797. Pascoal, C., Cassio, F., Marvanova, L., 2005b. Anthropogenic stress may affect aquatic hyphomycete diversity more than leaf decomposition in a low-order stream. Archiv Fur Hydrobiologie 162, 481-496. Paul, M.J., Meyer, J.L., Couch, C.A., 2006. Leaf breakdown in streams differing in catchment land use. Freshwater Biology 51, 1684-1695. Perez, J., Basaguren, A., Descals, E., Larranaga, A., Pozo, J., 2013. Leaf-litter processing in headwater streams of northern Iberian Peninsula: moderate levels of eutrophication do not explain breakdown rates. Hydrobiologia 718, 41-57. Pesce, S., Martin‐Laurent, F., Rouard, N., Montuelle, B., 2009. Potential for microbial diuron mineralisation in a small wine‐growing watershed: from treated plots to lotic receiver hydrosystem. Pest Management Science 65, 651-657. Petersen, R.C., Cummins, K.W., 1974. Leaf processing in a woodland stream. Freshwater Biology 4, 343-368. Radcliffe, J.C., 2002. Pesticide use in Australia, p. 309. Rasmussen, J.J., Wiberg-Larsen, P., Baattrup-Pedersen, A., Monberg, R.J., Kronvang, B., 2012. Impacts of pesticides and natural stressors on leaf litter decomposition in agricultural streams. Science of the Total Environment 416, 148-155.

Page 106 of 120 Ricart, M., Barceló, D., Geiszinger, A., Guasch, H., de Alda, M.L., Romaní, A.M., Vidal, G., Villagrasa, M., Sabater, S., 2009. Effects of low concentrations of the phenylurea herbicide diuron on biofilm algae and bacteria. Chemosphere 76, 1392-1401. Robinson, C.T., Gessner, M.O., 2000. Nutrient addition accelerates leaf breakdown in an alpine springbrook. Oecologia 122, 258-263. Royer, T.V., Minshall, G.W., 2001. Effects of nutrient enrichment and leaf quality on the breakdown of leaves in a hardwater stream. Freshwater Biology 46, 603-610. Saunders, G.W., 1975. Decomposition in freswater, in: Anderson, J.M., Macfadyen, A. (Eds.), The Role of terrestrial and aquatic organisms in decomposition processes. Blackwell Scientific Publication, Oxford, England, pp. 341 - 373. Schäfer, R.B., Bundschuh, M., Rouch, D.A., Szöcs, E., Peter, C., Pettigrove, V., Schulz, R., Nugegoda, D., Kefford, B.J., 2012. Effects of pesticide toxicity, salinity and other environmental variables on selected ecosystem functions in streams and the relevance for ecosystem services. Science of the Total Environment 415, 69-78. Schäfer, R.B., Caquet, T., Siimes, K., Mueller, R., Lagadic, L., Liess, M., 2007. Effects of pesticides on community structure and ecosystem functions in agricultural streams of three biogeographical regions in Europe. Science of the Total Environment 382, 272-285. Schafer, R.B., Pettigrove, V., Rose, G., Allinson, G., Wightwick, A., von der Ohe, P.C., Shimeta, J., Kuhne, R., Kefford, B., 2011. Effects of pesticides monitored with three sampling methods in 24 sites on macroinvertebrates and microorganisms. Environ Sci Technol 45, 1665-1672. Schulz, R., 2004. Field studies on exposure, effects, and risk mitigation of aquatic nonpoint-source insecticide pollution. J Environ Qual 33, 419-448. Schulz, R., Liess, M., 1999. A field study of the effects of agriculturally derived insecticide input on stream macroinvertebrate dynamics. Aquat Toxicol 46, 155-176. Silva-Junior, E.F., Moulton, T.P., 2011. Ecosystem functioning and community structure as indicators for assessing environmental impacts: leaf processing and macroinvertebrates in Atlantic forest streams. International Review of Hydrobiology 96, 656-666. Spänhoff, B., Bischof, R., Böhme, A., Lorenz, S., Neumeister, K., Nöthlich, A., Küsel, K., 2007. Assessing the impact of effluents from a modern wastewater treatment plant on

Page 107 of 120 breakdown of coarse particulate organic matter and benthic macroinvertebrates in a lowland river. Water, Air, and Soil Pollution 180, 119-129. Spänhoff, B., Meyer, E.I., 2004. Breakdown rates of wood in streams. Journal of the North American Benthological Society 23, 189-197. Suberkropp, K., Wallace, J.B., 1992. Aquatic hyphomycetes in insecticide-treated and untreated streams. Journal of the North American Benthological Society, 165-171. Tank, J.L., Winterbourn, M.J., 1996. Microbial activity and invertebrate colonisation of wood in a New Zealand forest stream. New Zealand Journal of Marine and Freshwater Research 30, 271-280. Tiegs, S., Langhans, S., Tockner, K., Gessner, M., 2007. Cotton strips as a leaf surrogate to measure decomposition in river floodplain habitats. Journal of the North American Benthological Society 26, 70-77. Tiegs, S.D., Clapcott, J.E., Griffiths, N.A., Boulton, A.J., 2013. A standardized cotton- strip assay for measuring organic-matter decomposition in streams. Ecological Indicators 32, 131-139. Vu, H.T., Keough, M.J., Long, S.M., Pettigrove, V.J., 2017. Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry 36, 720 - 726. Vyšná, V., Dyer, F., Maher, W., Norris, R., 2014. Cotton-strip decomposition rate as a river condition indicator–Diel temperature range and deployment season and length also matter. Ecological Indicators 45, 508-521. Wallace, J.B., Grubaugh, J.W., Whiles, M.R., 1996. Biotic indices and stream ecosystem processes: results from an experimental study. Ecological Applications 6, 140-151. Webster, J.R., Benfield, E.F., 1986. Vascular plant breakdown in freshwater ecosystems. Johnston, R. F. (Ed.). Annual Review of Ecology and Systematics, Vol. 17. Xi+714p. Annual Reviews Inc.: Palo Alto, California, USA. Illus, 567-594. Wightwick, A.M., Bui, A.D., Zhang, P., Rose, G., Allinson, M., Myers, J.H., Reichman, S.M., Menzies, N.W., Pettigrove, V., Allinson, G., 2012. Environmental fate of fungicides in surface waters of a horticultural-production catchment in Southeastern Australia. Archives of Environmental Contamination and Toxicology 62, 380-390.

Page 108 of 120 Young, R., 2006. Functional indicators of river ecosystem health—results from regional case studies of leaf decomposition. Prepared for New Zealand Ministry for the Environment, 31. Young, R.G., Matthaei, C.D., Townsend, C.R., 2008. Organic matter breakdown and ecosystem metabolism: functional indicators for assessing river ecosystem health. Journal of the North American Benthological Society 27, 605-625. Zubrod, J.P., Englert, D., Feckler, A., Koksharova, N., Konschak, M., Bundschuh, R., Schnetzer, N., Englert, K., Schulz, R., Bundschuh, M., 2015. Does the current fungicide risk assessment provide sufficient protection for key drivers in aquatic ecosystem functioning? Environ Sci Technol 49, 1173-1181.

Page 109 of 120 CHAPTER 6: GENERAL DISCUSSION

6.1 Effects of fungicides on amphipods – Laboratory perspective

This thesis reports a wide range of effects fungicides have on two Australian amphipod species at environmentally relevant concentrations under laboratory conditions. In general, reproduction was the most sensitive endpoint for both single (Chapter 2, 3) and mixture (Chapter 4) exposure. Growth was also significantly affected following fungicide exposure but varied depending on sex and fungicide concentration (Chapter 2 - 4), type of fungicide (Chapter 3, 4), animal age (Chapter 4), and test species (Chapter 2, 3). As expected, survival was less sensitive than reproduction and growth (Chapter 2, 4). Fungicide exposure caused no significant effects on survival of amphipods when they were exposed at the mature stage (Chapter 4) but caused significant effects when they were exposed at the juvenile stage (Chapter 2, 3). Biochemical biomarkers were altered after fungicide exposure but varied depending on test species, fungicides (Chapter 2, 3) and sex (Chapter 4).

Chapter 2 demonstrated comprehensive effects of the boscalid fungicide Filan® on Allorchestes compressa, an important intertidal marine amphipod species. In this chapter, strong relationships between biochemical changes (lipid, protein, and glycogen content) and effects at higher levels of organization (growth, reproduction) indicated that these biomarkers are sensitive to fungicides and could be used as early indicators of life history effects of animal exposed to fungicides. The sensitivity of these biochemical biomarkers were tested again in Chapter 3 but with the freshwater amphipod A. subtenuis. Even though these biomarkers were altered after fungicide exposure, they were less sensitive than other endpoints (growth, reproduction) and varied depending on tested fungicides (Chapter 3). The effects of fungicides on amphipod energetics emphasize the importance of understanding fungicide modes of action on non-target species. Furthermore, these results suggested either amphipod food quality was decreased as fungicides caused adverse effects on microbial communities or amphipod utilized their energy reserves to deal with toxic stress. The direct effects via water exposure or indirect effects via food consumption were not fully understood in Chapter 2 and Chapter 3. Microbial respiration increased (Chapter 2) and decreased (Chapter 3) with increasing

Page 110 of 120 fungicide concentration suggesting that microbial communities were affected and food quality was altered. However, the feeding rate of the amphipods in both chapters was not significantly affected (data not shown in Chapter 3). Therefore, it was not clear whether fungicides induce toxic effects on amphipods through reduction of food quality.

The colonization of microbial organisms on leaf litter benefits the shredders in two main ways: by producing enzymes that are not available in invertebrate digest system to promote the breakdown of complex structural compounds of the leaves and to transform them to simple compounds edible to invertebrates, and by providing an additional food source for invertebrates, as fungal mycelia colonizing leaves have higher nutrient content than senescent leaves (Vu et al., 2017). Therefore, fungicide effects on microorganisms that have colonized the leaves could reduce the nutritional quality of leaves and affect shredding invertebrates. Rasmussen et al. (2012a) found that microbial biomass on beech leaves (Fagus sylvatica) was lower than the control after only a 3h exposure to the fungicide propiconazole at 50 mg/L and 500mg/L resulted in a significant increase in consumption of the macroinvertebrates to compensate for the reduced nutritional quality of this leaf litter. Zubrod et al. (2015a) also found that the growth of Gammarus fossarum was significantly reduces (~40%) after 24 d of consumption of the black alder (Alnus glutinosa) which was conditioned in fungicide mixture for 12 d. However, Flores et al. (2014) reported that significantly decreased number of fungal species in the alder leaves exposed to fungicide imazalil (ranging from 0.1mg/L to 100 mg/L) caused no significant effects on consumption rate of the amphipod Echinogammarus berilloni compared with the control. The effects of fungicides on amphipod through food consumption could be the result of either a toxic effect of fungicides accumulated in leaf material or a reduction in food quality through fungicide-induced changes in microbial composition (Dimitrov et al., 2014).

The indirect effects of fungicides on amphipod via food consumption were further investigated in Chapter 4. In this chapter, the amphipod A. subtenuis was not only fed with pre-conditioned leaves like in Chapter 3 but was also provided another food source: ground TetraMin™ fish food which is a common food used for amphipod cultures (Borgmann et al., 1989). If the indirect effects largely contribute to fungicide toxicity on

Page 111 of 120 amphipods, we would expect the toxicity to reduce when amphipods were given an extra food source. However, significant effects of fungicides on amphipod growth and reproduction still occurred. No significant effects on amphipod survival observed in this chapter could be related to animal age rather than indirect effects as mature amphipods were used in this experiment while amphipods used in Chapter 3 were juveniles.

Both Chapters 2 and 3 showed that Filan and Systhane had extreme adverse effects on amphipod reproduction. There were no gravid females or juveniles produced in any fungicide treatments (Chapter 2). There was a significant decrease in the number of gravid females, a large decrease in the number of embryos per gravid females, and no production of juveniles in all fungicide treatments (Chapter 3). These significant effects on amphipod reproduction could have negative impacts on amphipod populations in natural environments. However, results from Chapters 2 and 3 could not fully demonstrate effects of fungicides on the next generation of amphipods. In Chapter 4, the mature amphipods were used to further investigate fungicide effects on amphipod reproduction and to address the questions:

1. Do fungicides cause delayed effect or completely inhibit amphipod reproduction? 2. If a delayed effect was observed then did this affect the quality and quantity of juveniles?

Results from Chapter 4 indicated that fungicides delayed amphipod reproduction and affected both quality and quantity of the next amphipod generation.

Chronic mixture effects of fungicides reflect a more realistic exposure scenario as fungicides are often detected in mixtures in aquatic environments. The main objectives of Chapter 4 were to investigate the long term interaction effects of fungicide mixtures on mature A. subtenuis at environmentally realistic concentrations on a wide range of endpoints that span different levels of biological organization and to evaluate how the results of mixture studies vary between endpoints. Results from Chapter 4 demonstrated the complexity of chronic mixture studies and emphasized that effects of mixtures were endpoint-dependent and using a variety of endpoints is necessary for a comprehensive

Page 112 of 120 understanding of mixture effects as individual endpoints could lead to a completely different interpretation.

As a prophylactic crop protectant, fungicides are often applied at higher frequencies but lower application rates than other types of pesticides (Reilly et al., 2012). Furthermore, many types of fungicides degrade slowly or persist in aquatic environments (Hertfordshire, 2017). Consequently, they are frequently detected in aquatic environments at low concentrations. The use of fungicides is often regarded as posing only a minor risk to aquatic organisms. However, results from all three chapters (Chapter 2 - 4) demonstrated that long term exposure to fungicides at environmentally relevant concentrations significantly affected amphipod survival, growth, and reproduction that consequently could have impacts on amphipod populations in natural environments. These results suggested that it is important to assess the long term impacts of fungicides on aquatic organisms using environmentally relevant concentrations. Chapter 4 further emphasizes the significance of using realistic concentrations in mixture studies as the fungicide myclobutanil was shown to have synergistic effects in other studies using high concentrations but caused antagonistic effects in this study when environmentally relevant concentrations were used.

Laboratory studies suggest that fungicides could reduce amphipod population (Chapters 2- 4) and microbial leaf decomposition (Chapter 3) in natural environments. Because both macroinvertebrates and microorganism have crucial roles in organic matter breakdown in streams, these laboratory results suggest a decreasing organic matter breakdown in fungicide polluted streams. However, extrapolation of laboratory results to the field study is complicated and challenging.

6.2 Effects of fungicides on ecosystem function – Linking laboratory results with field observations

Organic matter breakdown is an integrated physical and biological process in which the role of macroinvertebrates and microorganisms on organic matter breakdown may be altered depending on stream characteristics. In low order streams, shredders are abundant and have a critical role in leaf litter breakdown (Graca, 2001). Therefore, fungicides,

Page 113 of 120 other pesticides, and other anthropogenic pollutants that affect stream macroinvertebrates could significantly inhibit leaf litter breakdown (Cuffney et al., 1990). However, in high order or polluted streams where shredders are less abundant, microorganisms have a more important role in organic matter breakdown than macroinvertebrates (Imberger et al., 2008; Pascoal et al., 2005; Webster and Benfield, 1986). Chapter 5 demonstrated the important role of microorganisms on organic matter breakdown as there was no difference in breakdown rate of leaf and cotton between coarse (allowing macroinvertebrates and microorganisms access to organic matter) and fine (allowing only microorganisms access to organic matter) mesh bags, despite the presence of pesticides. Therefore, the significant effects of fungicides on amphipods (Chapters 2- 4) in laboratory studies do not support what occurred in the field (Chapter 5).

In Chapter 5, leaf breakdown rate was negatively correlated with fungicide/pesticide concentrations suggested the inhibition breakdown effects of fungicides. The inhibition of leaf breakdown by fungicides in the current study is in agreement with other field studies by Rasmussen et al. (2012b) and Schafer et al. (2012) who reported that the reduction of leaf breakdown was due to exposure to pesticides. To my knowledge, there is no field study which investigates the relationship between fungicide concentrations and leaf litter breakdown. However, several laboratory-based studies have demonstrated the inhibition of leaf breakdown by fungicides. Artigas et al. (2012) reported that fungicide tebuconazole (33.1 ± 12.4 g/L) caused significant decrease in breakdown rate of Alnus glutinosa and Populus nigra leaf litter due to the reductions in microbial biomass development and shifts in community structure. Zubrod et al. (2015b) reported that exposure to fungicides not only altered microbial leaf decomposition but also impaired leaf palatability to the leaf-shredder Gammarus fossarum that over all could reduce the leaf litter breakdown rate. However, the results of Chapter 3 demonstrated that fungicide Systhan significantly reduce microbial leaf mass loss, up to 50%, but cause no significant effects on amphipod feeding rates (Data not shown). The effects of fungicides on leaf litter breakdown through the effects on microorganism and macroinvertebrate activity is complicated if both processes are considered simultaneously. On the one hand, the reduction of nutritional quality of leaf litter due to alteration of microbial community could reduce the feeding activity of the shredders as they avoid eating low quality food

Page 114 of 120 (Zubrod et al., 2015b). On the other hand, the shredders may increase their leaf consumption to compensate for the low quality of food (Rasmussen et al., 2012a). Therefore, the overall effects of fungicides on leaf litter decomposition could increase or decrease the leaf litter breakdown depends on the role of each organism group.

Within microorganisms, fungi and bacteria have different roles depending on the chemistry of substrate. Cellulose is a major constituent of leaf litter (~ 70%) and cotton (~ 95%) (Imberger et al., 2010). Leaf litter is also composed of lignin (Fioretto et al., 2005) and relatively small amount of lipids and nutrients (Campbell et al., 1992). This difference in chemistry between leaf and cotton could favor different microbial groups. Fungi play a more important role in leaf litter decomposition than bacteria (Das et al., 2007; Hieber and Gessner, 2002) while bacteria play a more important role than fungi in cellulose breakdown (Ardón and Pringle, 2007; Young, 2006). Thus, the effects of fungicides on leaf breakdown in the laboratory study (Chapter 3) could be used to predict effect of fungicides on leaf breakdown but not to predict cellulose (cotton) breakdown in aquatic environments (Chapter 5).

In laboratory studies, experimental conditions are well controlled and the test subject is generally exposed to only one stressor, while multiple anthropogenic stressors often co- occur in the environment. These multiple stressors could interact in concert to cause complementary synergistic effects or counteract each other resulting in antagonistic effects. Chapter 5 demonstrated that leaf litter breakdown was affected by natural (temperature) and anthropogenic (pesticides and nutrients) stressors and these stressors counteracted each other resulting in a decrease, no effects or an increase of leaf breakdown rate. While, all anthropogenic stressors acted in concert causing additive effects on cotton breakdown rate. Furthermore, other toxicants are likely to exist and have impact on field study subject but are not detected in the study. For example, other pesticides possibly present in streams but not detected in passive samplers of this study (Chapter 5) could also have impacts on organic matter breakdown. Rose et al., (2009) reported that about 48% pesticides detected in grab water samples were observed in passive samplers.

Page 115 of 120 Chapter 5 demonstrated for the first time individual relationships of different pesticide groups on organic matter breakdown. A few studies have shown that pesticides reduce leaf (Rasmussen et al., 2012b; Schäfer et al., 2012; Schäfer et al., 2007) and cotton (Schäfer et al., 2012) breakdown. However the impacts varied depending on pesticide groups and substrate. While insecticides had positive correlation with both leaf and cotton breakdown, fungicides and herbicides had negative correlations with leaf breakdown but positive correlation with cotton breakdown (Chapter 5). These differences among different pesticide groups could be attributed to the different roles of fungi and bacteria on leaf and cotton breakdown as discussed above. The result from Chapter 5 not only support the application of cotton strips as a standard organic matter substrate to measure impacts of anthropogenic stressors on organic matter breakdown but also partially reveal cotton breakdown mechanism in aquatic environments. This chapter supports the role of bacteria and burrowing invertebrates on cotton breakdown rather than the role of fungi and shredding invertebrates.

6.3 Recommendations for future studies

The freshwater amphipod A. subtenuis was shown to be sensitive to fungicides at different life stages (Chapters 3 and 4). It has been reported to be more sensitive to heavy metals than other invertebrates (Thorp and Lake, 1974). Furthermore, A.subtenuis has desirable characteristics of a model test species in toxicity studies because it is widespread in different regions in Australia, important in aquatic food webs, easily collected in large numbers, and easily cultured in the laboratory. While freshwater amphipods have been used as standard toxicity test species in North America (e.g. Hyalella azteca) and Europe (e.g. Gammarus pulex), there are no freshwater amphipods being used as model species in toxicology studies in Australia currently. Austrochiltonia subtenuis used in all available toxicological studies (study by Thorp and Lake (1974) and the present study) were collected from the field. Further studies developing the culture method for A.subtenuis including artificial water and food type and standard toxicity bioassays using this amphipod are necessary.

Fungicide pollution could affect amphipod populations in the natural environments (Chapters 2 - 4). Field investigations on the impacts of fungicides on the dynamics and

Page 116 of 120 production of natural amphipod populations are important to validate the laboratory data. Furthermore, amphipods not only play an important role in leaf litter decomposition but also are an important food source for organisms at higher trophic levels such as fish and birds in both freshwater and marine environments (Dauby et al., 2003; Pen and Potter, 1992; Pollard, 1973). Consequently, these effects of fungicides on amphipods could be transferred to animals at higher trophic levels. This aspect of ecological effects of fungicides was beyond the scope of this thesis and is also a gap in the literature.

The findings in Chapter 5 suggest that cotton strips could be used as a diagnostic tool in biomonitoring programs to assess the impacts of pesticides and other anthropogenic stressors on ecosystem function. To support this, laboratory studies investigating the roles of microorganisms (fungi vs. bacteria) and invertebrates (shredding invertebrates vs. burrowing invertebrates) on cellulose breakdown are needed. Additionally, controlled experiments looking at impacts of different toxicants (singly and in mixtures) on cellulose breakdown in the laboratory could help establish the causality of these toxicants on cellulose breakdown and validate the correlations between environmental stressors and cellulose breakdown rates observed in the field study. The results of Chapter 5 could also be verified by applying leaf bag and cotton strip assays in other agricultural areas but leaf and cotton samples should be collected to determine microorganism biomass and diversity using molecular techniques which will help to identify the roles of fungi and bacteria in leaf and cotton breakdown.

REFERENCES. Ardón, M., Pringle, C.M., 2007. The quality of organic matter mediates the response of heterotrophic biofilms to phosphorus enrichment of the water column and substratum. Freshwater Biology 52, 1762-1772. Artigas, J., Majerholc, J., Foulquier, A., Margoum, C., Volat, B., Neyra, M., Pesce, S., 2012. Effects of the fungicide tebuconazole on microbial capacities for litter breakdown in streams. Aquat Toxicol 122, 197-205. Borgmann, U., Ralph, K.M., Norwood, W.P., 1989. Toxicity test procedures for Hyalella azteca, and chronic toxicity of cadmium and pentachlorophenol to H. azteca, Gammarus

Page 117 of 120 fasciatus, and Daphnia magna. Archives of Environmental Contamination and Toxicology 18, 756-764. Campbell, J.C., James, K.R., Hart, B.T., Devereaux, A., 1992. Allochthonous coarse particulate organic material in forest and pasture reaches of two south‐eastern Australian streams. Freshwater Biology 27, 341-352. Cuffney, T.F., Wallace, J.B., Lugthart, G., 1990. Experimental evidence quantifying the role of benthic invertebrates in organic matter dynamics of headwater streams. Freshwater Biology 23, 281-299. Das, M., Royer, T.V., Leff, L.G., 2007. Diversity of fungi, bacteria, and actinomycetes on leaves decomposing in a stream. Applied and Environmental Microbiology 73, 756- 767. Dauby, P., Nyssen, F., De Broyer, C., 2003. Amphipods as food sources for higher trophic levels in the Southern Ocean: a synthesis. Antarctica in a Global Context. Backhuys, Leiden, 129-134. Dimitrov, M.R., Kosol, S., Smidt, H., Buijse, L., Van den Brink, P.J., Van Wijngaarden, R.P., Brock, T.C., Maltby, L., 2014. Assessing effects of the fungicide tebuconazole to heterotrophic microbes in aquatic microcosms. Science of the Total Environment 490, 1002-1011. Fioretto, A., Di Nardo, C., Papa, S., Fuggi, A., 2005. Lignin and cellulose degradation and nitrogen dynamics during decomposition of three leaf litter species in a Mediterranean ecosystem. Soil Biology and Biochemistry 37, 1083-1091. Flores, L., Banjac, Z., Farre, M., Larranaga, A., Mas-Marti, E., Munoz, I., Barcelo, D., Elosegi, A., 2014. Effects of a fungicide (imazalil) and an insecticide (diazinon) on stream fungi and invertebrates associated with litter breakdown. Science of the Total Environment 476, 532-541. Graca, M.A.S., 2001. The role of invertebrates on leaf litter decomposition in streams - a review. International Review of Hydrobiology 86, 383-393. Hertfordshire, U.o., 2017. Pesticides Properties DataBase. Hieber, M., Gessner, M.O., 2002. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology 83, 1026-1038.

Page 118 of 120 Imberger, S.J., Thompson, R.M., Grace, M.R., 2010. Searching for effective indicators of ecosystem function in urban streams: assessing cellulose decomposition potential. Freshwater Biology 55, 2089-2106. Imberger, S.J., Walsh, C.J., Grace, M.R., 2008. More microbial activity, not abrasive flow or shredder abundance, accelerates breakdown of labile leaf litter in urban streams. Journal of the North American Benthological Society 27, 549-561. Pascoal, C., Cassio, F., Marcotegui, A., Sanz, B., Gomes, P., 2005. Role of fungi, bacteria, and invertebrates in leaf litter breakdown in a polluted river. Journal of the North American Benthological Society 24, 784-797. Pen, L., Potter, I., 1992. Seasonal and size‐related changes in the diet of perch, Perca fluviatilis L., in the shallows of an Australian river, and their implications for the conservation of indigenous teleosts. Aquatic Conservation: Marine and Freshwater Ecosystems 2, 243-253. Pollard, D., 1973. The biology of a landlocked form of the normally catadromous salmoniform fish Galaxias maculatus (Jenyns). V. Composition of the diet. Marine and Freshwater Research 24, 281-296. Rasmussen, J.J., Monberg, R.J., Baattrup-Pedersen, A., Cedergreen, N., Wiberg-Larsen, P., Strobel, B., Kronvang, B., 2012a. Effects of a triazole fungicide and a pyrethroid insecticide on the decomposition of leaves in the presence or absence of macroinvertebrate shredders. Aquat Toxicol 118, 54-61. Rasmussen, J.J., Wiberg-Larsen, P., Baattrup-Pedersen, A., Monberg, R.J., Kronvang, B., 2012b. Impacts of pesticides and natural stressors on leaf litter decomposition in agricultural streams. Science of the Total Environment 416, 148-155. Reilly, T.J., Smalling, K.L., Orlando, J.L., Kuivila, K.M., 2012. Occurrence of boscalid and other selected fungicides in surface water and groundwater in three targeted use areas in the United States. Chemosphere 89, 228-234. Rose, G., Allen, D., Allinson, G., Allinson, M., Bui, A., Wightwick, A., Zhang, P., 2009. Melbourne Water and DPI agrochemicals in Port Philip catchment streams-project summary report on 2008-09, Melbourne, Victoria. Schäfer, R.B., Bundschuh, M., Rouch, D.A., Szöcs, E., Peter, C., Pettigrove, V., Schulz, R., Nugegoda, D., Kefford, B.J., 2012. Effects of pesticide toxicity, salinity and other

Page 119 of 120 environmental variables on selected ecosystem functions in streams and the relevance for ecosystem services. Science of the Total Environment 415, 69-78. Schäfer, R.B., Caquet, T., Siimes, K., Mueller, R., Lagadic, L., Liess, M., 2007. Effects of pesticides on community structure and ecosystem functions in agricultural streams of three biogeographical regions in Europe. Science of the Total Environment 382, 272-285. Thorp, V.J., Lake, P.S., 1974. Toxicity bioassays of cadmium on selected freshwater invertebrates and the interaction of cadmium and zinc on the freshwater shrimp, Paratya tasmaniensis Riek. Australian Journal of Marine and Freshwater Research 25, 97-104. Vu, H.T., Keough, M.J., Long, S.M., Pettigrove, V.J., 2017. Effects of two commonly used fungicides on the amphipod Austrochiltonia subtenuis. Environmental Toxicology and Chemistry 36, 720 - 726. Webster, J.R., Benfield, E.F., 1986. Vascular plant breakdown in freshwater ecosystems. Johnston, R. F. (Ed.). Annual Review of Ecology and Systematics, Vol. 17. Xi+714p. Annual Reviews Inc.: Palo Alto, California, USA. Illus, 567-594. Young, R., 2006. Functional indicators of river ecosystem health—results from regional case studies of leaf decomposition. Prepared for New Zealand Ministry for the Environment, 31. Zubrod, J., Englert, D., Wolfram, J., Wallace, D., Schnetzer, N., Baudy, P., Konschak, M., Schulz, R., Bundschuh, M., 2015a. Waterborne toxicity and diet-related effects of fungicides in the key leaf shredder Gammarus fossarum (Crustacea: Amphipoda). Aquat Toxicol 169, 105-112. Zubrod, J.P., Englert, D., Feckler, A., Koksharova, N., Konschak, M., Bundschuh, R., Schnetzer, N., Englert, K., Schulz, R., Bundschuh, M., 2015b. Does the current fungicide risk assessment provide sufficient protection for key drivers in aquatic ecosystem functioning? Environ Sci Technol 49, 1173-1181.

Page 120 of 120

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: Vu, Hung Thi Hong

Title: Effects of fungicides on Australian amphipods and organic matter breakdown in aquatic environments

Date: 2017

Persistent Link: http://hdl.handle.net/11343/161176

File Description: Effects of fungicides on Australian amphipods and organic matter breakdown in aquatic environments

Terms and Conditions: Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the copyright owner. The work may not be altered without permission from the copyright owner. Readers may only download, print and save electronic copies of whole works for their own personal non-commercial use. Any use that exceeds these limits requires permission from the copyright owner. Attribution is essential when quoting or paraphrasing from these works.