Fish Monitoring Across Regional Catchments of the Adelaide and Mount Lofty Ranges Region 2015–17

Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

David W. Schmarr, Rupert Mathwin and David L.M. Cheshire

SARDI Publication No. F2018/000217-1 SARDI Research Report Series No. 990

SARDI Aquatics Sciences PO Box 120 Henley Beach SA 5022

August 2018

Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 Project

David W. Schmarr, Rupert Mathwin and David L.M. Cheshire

SARDI Publication No. F2018/000217-1 SARDI Research Report Series No. 990

August 2018

II Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

This publication may be cited as: Schmarr, D.W., Mathwin, R. and Cheshire, D.L.M. (2018). Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015-17. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2018/000217-

1. SARDI Research Report Series No. 990. 102pp.

South Australian Research and Development Institute SARDI Aquatic Sciences 2 Hamra Avenue West Beach SA 5024

Telephone: (08) 8207 5400 Facsimile: (08) 8207 5415 http://www.pir.sa.gov.au/research

DISCLAIMER The authors warrant that they have taken all reasonable care in producing this report. The report has been through the SARDI internal review process, and has been formally approved for release by the Research Chief, Aquatic Sciences. Although all reasonable efforts have been made to ensure quality, SARDI does not warrant that the information in this report is free from errors or omissions. SARDI and its employees do not warrant or make any representation regarding the use, or results of the use, of the information contained herein as regards to its correctness, accuracy, reliability and currency or otherwise. SARDI and its employees expressly disclaim all liability or responsibility to any person using the information or advice. Use of the information and data contained in this report is at the user’s sole risk. If users rely on the information they are responsible for ensuring by independent verification its accuracy, currency or completeness. The SARDI Report Series is an Administrative Report Series which has not been reviewed outside the department and is not considered peer-reviewed literature. Material presented in these Administrative Reports may later be published in formal peer-reviewed scientific literature.

© 2018 SARDI This work is copyright. Apart from any use as permitted under the Copyright Act 1968 (Cth), no part may be reproduced by any process, electronic or otherwise, without the specific written permission of the copyright owner. Neither may information be stored electronically in any form whatsoever without such permission.

SARDI Publication No. F2018/000217-1 SARDI Research Report Series No. 990

Author(s): David W. Schmarr, Rupert Mathwin and David L.M. Cheshire

Reviewer(s): George Giatas (SARDI) and Kristian Peters (DEW)

Approved by: Assoc Prof. Qifeng Ye Science Leader – Inland Waters & Catchment Ecology

Signed:

Date: 13 August 2018

Distribution: AMLRNRMB, DEW, SAASC Library, Parliamentary Library, State Library and National Library

Circulation: Public Domain

III Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

TABLE OF CONTENTS

LIST OF FIGURES ...... V LIST OF TABLES ...... VI ACRONYMS ...... VII ACKNOWLEDGEMENTS ...... VIII EXECUTIVE SUMMARY ...... 1 1. INTRODUCTION ...... 3 1.1. Background...... 3 1.2. Objectives ...... 5 2. METHODS ...... 7 2.1. Study site selection ...... 7 2.2. Water quality measurements ...... 10 2.3. Fish sampling methodology ...... 10 3. RESULTS ...... 13 3.1. General results ...... 13 3.2. Site Health Assessment ...... 19 4. DISCUSSION ...... 28 4.1. WMLR Fish condition ...... 28 4.2. Myponga, Bungala, Inman, and Hindmarsh Estuaries ...... 38 4.3. Summary of WMLR fish condition and estuary studies ...... 43 4.4. Recommendations ...... 46 5. REFERENCES ...... 48 6. APPENDIX A ...... 51 7. APPENDIX B ...... 57 8. APPENDIX C ...... 59 9. APPENDIX D ...... 64

IV Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

LIST OF FIGURES

Figure 1. Location of fish monitoring sites (in yellow) surveyed across the WMLR between 2015 and 2017. Sites marked blue are estuary monitoring sites...... 8 Figure 2. Location of fish monitoring sites on Fleurieu Peninsula (in yellow) surveyed between 2015 and 2017. Sites surveyed for the estuarine health study indicated in boxes. Sites marked blue are estuary monitoring sites...... 9 Figure 3. Scatter plot comparing BCG and FHI scores at each site and date. R2 = 0.7489...... 23 Figure 4. Composite map showing the distribution of BCG scores across the WMLR in each sampling season from autumn 2015 to spring 2017...... 24 Figure 5. Composite map showing the distribution of FHI scores across the WMLR in each sampling season from autumn 2015 to spring 2017...... 25 Figure 6. Composite map showing the distribution of BCG scores at Estuary sites in each sampling season from autumn 2015 to spring 2017. Myponga not sampled in spring 2017...... 26 Figure 7. Composite map showing the distribution of FHI scores at Estuary sites in each sampling season from autumn 2015 to spring 2017. Myponga not sampled in spring 2017...... 27 Figure 8. Flow rate in ML/day at Yaldara Gauge (A5050502) on North Para River. Heavy rainfall in the middle and end of September 2016 caused widespread flooding in the MLR...... 31 Figure 9. Landslide and damage to First Creek stream as a result of flooding in spring 2016. ..32 Figure 10. Comparison of riparian and pool habitat at Myponga Pumphouse before (left) and after (right) the spring 2016 flood...... 39 Figure 11. Weir on Inman River blocking passage between estuary and freshwater sites...... 42

V Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

LIST OF TABLES

Table 1. Number of fish monitoring sites surveyed per season across the WMLR between 2015 and 2017...... 13 Table 2. Total catch of native (green) and invasive (red) fish species at sites in the WMLR fish condition study. This includes native diversity score (see section 2.3.2)...... 14 Table 3. Total catch of native (green) and invasive (red) fish species at estuarine health study sites in the WMLR fish condition study. This includes native diversity score (see section 2.3.2)...... 17 Table 4. FHI and BCG scores for the sites in the WMLR fish condition study (I = Historically documented, sensitive, long-lived, or regionally endemic taxa, II = Sensitive-Rare Taxa, III = Sensitive Common Taxa, V = Tolerant taxa, VI = Non-native or intentionally introduced taxa, VII = Organism and population condition, IX = Spatial and temporal extent of detrimental effects, X = Ecosystem connectivity)...... 19 Table 5. FHI and BCG scores for the estuary health study sites (I = Historically documented, sensitive, long-lived, or regionally endemic taxa, II = Sensitive-Rare Taxa, III = Sensitive Common Taxa, V = Tolerant taxa, VI = Non-native or intentionally introduced taxa, VII = Organism and population condition, IX = Spatial and temporal extent of detrimental effects, X = Ecosystem connectivity)...... 22 Table 6. Comparison of average 2015-17 BCG and FHI scores with scores in 2011 and 2013. Note: Historical BCG and FHI are not average values...... 28

VI Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

ACRONYMS

AMLR Adelaide and Mount Lofty Ranges AMLRNRMB Adelaide and Mount Lofty Ranges Natural Resource Management Board BCG Biological condition gradient CPUE Catch per unit of effort DEW Department for Environment and Water DEWNR Department of Environment, Water and Natural Resources DO Dissolved oxygen FHI Fish health index GIS Geographic information system GWAP Goyder Water Allocation Planning MDBC Murray Darling Basin Commission MLR Mount Lofty Ranges NRM Natural Resource Management PIRSA Primary Industries and Regions South Australia PWRA Prescribed water resource area SARDI South Australian Research and Development Institute TL Total length VWASP Verification of Water Allocation Science Project WMLR Western Mount Lofty Ranges WAP Water Allocation Planning

VII Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

ACKNOWLEDGEMENTS

This report was funded by the Natural Resources Adelaide and Mount Lofty Ranges (Natural Resources AMLR) under the Western Mount Lofty Ranges Fish Community Monitoring Project. The field component of this study was extensive and took many months of dedicated work. This would not have been possible without the willing assistance of a range of staff and volunteers. Site selection was assisted by Dr Kristian Peters and Dr Douglas Green (AMLRNRMB/DEW). Field Assistance was provided by SARDI staff Leigh Thwaites, David Cheshire, Rupert Mathwin, Kate Frahn, Alex Dobrovolskis, Nat Navong and Zyg Lorenz, and summer scholarship student Emma Matthews.

Our thanks go to all the community members who allowed us access to their land. Our thanks also go to the reviewers Dr Kristian Peters (DEWNR) and George Giatas (SARDI), whose revisions aided in the final product.

VIII Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

EXECUTIVE SUMMARY

This monitoring study provided information on the status and trends in the condition of important aquatic ecological assets (i.e. fish) in the Adelaide and Mount Lofty Ranges (AMLR) Natural Resource Management (NRM) area. The data informs reporting on state and regional NRM targets and provides baseline information to support various environmental and water resource management processes (e.g. water allocation planning and habitat restoration).

The study consisted of a field-based fish monitoring program that comprised a spatial array of fixed and ad hoc sites in catchments and estuary systems across the AMLR. The program aimed to assess biological condition across the monitoring sites using Biological Condition Gradient (BCG) and SARDI Fish Health Index (FHI) methodologies and the compilation of comprehensive site assessment sheets.

The study monitored 57 sites, including 10 long-term monitoring sites and four estuaries that were monitored every autumn and spring from 2015–2017. Over 90,000 fish were captured from a diverse range of estuarine and freshwater species. The diversity and abundance of native and invasive fish species was used to score each site’s BCG and FHI which provided a spatial and temporal representation of fish community health. Patterns in fish distribution, diversity and abundance, and BCG and FHI scores delivered several key findings:

 Coastal freshwater sites were almost all found to be in the healthy category while sites further inland and further north frequently had intermediate or poor health ratings.  Barriers to fish passage, natural or human-made, appear to have shaped the composition of fish communities.  Presence of brown trout (Salmo trutta) was associated with the absence of native species at a site or reach scale and was frequently associated with poor fish community health.  Native and invasive species were resilient to flood disturbance in the short-term.  The fragmented and limited distribution of mountain Galaxias (Galaxias olidus) indicate that it should remain in the vulnerable category of species protection.  The health of fish populations in the four estuaries monitored over three years suggests that fish populations were tolerant to the water quality conditions during this period.

These findings led to recommendations for prioritising future management including:

 Continuing habitat and flow protection and restoration work and aligning monitoring to assess the long-term effectiveness of broad scale on-ground works to provide adaptive management.

1 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

 Conducting desk-top and field surveys of in-stream barriers to fish passage in priority catchments.  Investigating removal of unnecessary barriers to fish passage.  Targeting monitoring programs in trout-stocked rivers to investigate interactions between trout and native species.  Providing scientific advice to fisheries and natural resource managers to assist in public education and risk management regarding the impact of stocking non-native species in MLR catchments.  Continuing monitoring at appropriate locations and temporal scales to verify long-term fish community recovery from flood disturbance and confirm persistence of isolated populations.  Scaling back monitoring in Myponga, Bungala and Hindmarsh estuaries, and focusing on monitoring Inman estuary and fish passage between estuary and freshwater reach.  Consolidating all regional fish distribution and biological data with historical data to provide an updated Biological Review of the Freshwater Fishes of the Mount Lofty Ranges.

Keywords: aquatic ecosystems, biological condition gradient, biological monitoring, condition monitoring, estuary, freshwater.

2 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

1. INTRODUCTION

1.1. Background

Aquatic organisms such as fish are a robust indicator for the general condition of aquatic ecosystems (Karr 1981). Occupying higher trophic levels, healthy freshwater fish populations reflect positively on the condition of habitat, food-web structure and flow regime, all of which are key components of healthy functioning aquatic ecosystems. Accordingly, many of Australia’s largest aquatic monitoring programs, such as the Murray-Darling Basin’s Sustainable Rivers Audit and the Lake Eyre Basin Ministerial Forum’s Lake Eyre Basin Rivers Assessment, have strong fish monitoring components as an indicator of riverine condition across large spatial and temporal scales (MDBC 2008, Kiri-ganai Research 2009). Aquatic ecosystems in good condition support healthy native fish populations and should possess:

 All expected native fish species based on natural range and habitat requirements

 High abundance commensurate with species traits

 Populations with individuals across a range of ages, including juvenile and adult life stages

 Signs of regular recruitment

 Low numbers of exotic species

 Low incidence of disease and parasites

For streams and rivers in the Mount Lofty Ranges (MLR), the majority of this information can be surveyed using well developed rapid assessment methodologies (McNeil and Hammer 2007, McNeil and Cockayne 2011) with findings compared against historic records and known biotic thresholds (McNeil et al. 2011a, Schmarr et al. 2014). Such surveys, especially those that cover a large number of sites across catchments and regions, provide snapshots of aquatic ecosystems that reflect the current condition of those habitats and regions.

These types of data support a range of management prioritisation and assessment activities. They provide detail on where important or threatened species are distributed, where key populations exist or are absent, where exotic competitors and/or predators may have been introduced, or where populations may be supressed from anthropogenic and natural threatening processes. This information can be extremely effective in directing regional natural resource management (NRM) investment and can assist with the development and setting of condition targets and objectives.

3 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Sequential data sampling (collection of data over time at the same locations) can result in highly effective monitoring programs that provide relevant biological information with excellent temporal resolution allowing objective analysis of key factors (Power 2007). NRM frameworks for the MLR have a clear requirement for assessing and reporting on the outcomes of NRM investment programs, largely against targets set a priori in line with investment priorities and available budgets. As a result, ongoing monitoring programs that utilise consistent methodology at the same sites can be an extremely useful tool for capturing and expressing responses or trajectories in condition that can be measured against desired outcomes or target values.

The condition of river and estuary ecosystems in the AMLR region have been altered since European settlement. Land clearing has resulted in greater sediment and nutrient inputs from land-based sources, which are recognised as key factors driving catchment water quality (Adelaide and Mount Lofty Ranges NRM Board 2013). The Adelaide and Mount Lofty Ranges Natural Resources Management Board (AMLRNRMB) has set regional targets for maintaining and improving the extent, condition and function of aquatic ecosystems and preventing further declines in the conservation status of native species (Adelaide and Mount Lofty Ranges NRM Board 2013). Many AMLR fish species are susceptible to declining water quality and changing habitat provision (Hammer et al. 2009). Recent studies have shown declines in native fish biodiversity across AMLR (Hammer 2005, McNeil and Hammer 2007, Hammer et al. 2009, McNeil et al. 2011a, Schmarr et al. 2014).

These projects have identified changes in the distribution and diversity of native species as well as localised depletion or extinction since historical records were published over fifty years ago (Hammer et al. 2009). These studies also identified a number of exotic species and translocated native species within the AMLR region that through competition, present considerable concern for the sustainability of native fish stocks. Habitat and catchment modifications and changes to flow regime have also been linked to declining distribution and abundances of native fishes as well as threats to population connectivity presented by barriers to fish movement and migrations (Schmarr et al. 2011, McNeil et al. 2011a).

To inform the management of water resources and aquatic ecosystems in the Western Mount Lofty Ranges (WMLR) NRM region, there is a requirement for an understanding of the present distribution and health of native fish populations across the WMLR. The status of these populations will provide a baseline for comparison against future water management regimes. It will also provide data for highlighting and prioritising management actions for improving fish population health.

These projects aim to provide the Board with the knowledge required to effectively allocate water in the Mount Lofty Ranges based on area-specific hydrological and ecological conditions and values. The present component of the project will contribute to this process through surveying the distribution

4 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

of fish species across sites in the WMLR and documenting their current ecological condition. This information is also used in conjunction with water quality and macro-invertebrate data to assist the Board in tracking long-term aquatic ecosystem condition over time.

The fish community monitoring project builds on fish monitoring conducted in 2011 (McNeil et al. 2011), as well as the monitoring conducted for the Barossa Prescribed Water Resource Area (PWRA) fish study and Verification of Water Allocation Science Project (VWASP) (Schmarr et al. 2014). These projects have served to provide a comprehensive baseline dataset of fish community health in the WMLR using a consistent sampling methodology.

In addition to the past sampling efforts, a method of assessing ecosystem health has been adopted using the Biological Condition Gradient (BCG) developed by Davies and Jackson (2006) and trialled by SARDI using the previous 2011 monitoring data (Mathwin et al. 2014).

This report presents results from three years of fish monitoring in the AMLR region. It provides data and information on the distribution and abundance of native and non-native species and a summary of the native and introduced (non-native and translocated) fish community structure at each site. This includes information on the distribution, abundance and biodiversity of native fish populations found in the southern Fleurieu Peninsula’s estuaries (Bungala, Hindmarsh, Inman and Myponga), which have limited information and are not well understood. The report uses a classification process for capturing the relative ecological condition of fish communities and populations to provide a platform of ecological condition against which prioritisation of NRM investments and progress towards management targets for aquatic biodiversity and condition can be measured across the AMLR region.

1.2. Objectives

The overarching aims of this project were to:

i) Provide spatial information of freshwater fish biodiversity across 3 years. This includes sampling approximately 50 sites that comprise both static (sequentially sampled) and specific “fit-for-purpose” sites during each calendar year.

ii) Provide information on fish survivorship and recruitment by conducting surveys during autumn and spring in each calendar year.

iii) Provide spatial information and data on the biodiversity and seasonal structure of fish communities (native, translocated and exotic species) at pre-determined sites across catchments and sub-catchments of the AMLR region, and lower reaches of the Bungala, Hindmarsh and Inman river-estuaries of the lower Fleurieu Peninsula.

5 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

iv) Describe local health and condition of fish populations and aquatic ecosystems across the region by building comparative site condition using Biological Condition Gradients (BCG) and Fish Health Indices (FHI).

v) Compare baseline condition states previously obtained across the AMLR region between 2011 and 2014.

vi) Provide information on the bi-annual fish diversity and population structure that can be used to assist and inform long-term catchment VWASP and Water allocation planning (WAP) programs and flow models for the region.

vii) Collect annual data and information on fish diversity, population structure and aquatic ecosystem information that can be used to inform the Monitoring, Evaluation, Reporting and Implementation (MERI) needs of the AMLRNRMB.

6 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

2. METHODS

2.1. Study site selection

Site selection was undertaken in negotiation with DEW (formerly DEWNR) and AMLRNRMB staff to ensure that project objectives were maintained. This included sampling a range of sites across autumn and spring (Figure 1 and 2, Appendix B) following the criteria below:

 10 VWASP sites

 3 Bungala river estuary

 3 Hindmarsh river estuary

 3 Inman river estuary

 8 additional discretionary sites per year to cover areas not previously sampled

 3 additional sites on Balquhidder Station on southern Fleurieu Peninsula in spring 2016 and autumn 2017

Field sampling took place in autumn and spring between March 2015 and December 2017.

7 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 1. Location of fish monitoring sites (in yellow) surveyed across the WMLR between 2015 and 2017. Sites marked blue are estuary monitoring sites.

8 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 2. Location of fish monitoring sites on Fleurieu Peninsula (in yellow) surveyed between 2015 and 2017. Sites surveyed for the estuarine health study indicated in boxes. Sites marked blue are estuary monitoring sites.

9 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

2.2. Water quality measurements

A point of maximum depth was identified within each site where water quality was recorded. Water quality was determined using dissolved oxygen (DO), water temperature (oC), pH and salinity (conductivity, µS/cm), measured on site using a Horiba U-50 Multiparameter Sonde. To obtain water quality profiles at each site, water samples were assessed at the surface and at intervals of 50cm depth until the sediment layer (riverbed) was reached.

2.3. Fish sampling methodology

To sample fish populations, most sites employed two fyke net designs; ‘small fykes’ (3 m leader, 2 m funnel, 3 mm mesh) and ‘double-wing fykes’ (2 x 5 m wings, 3 m funnel, 3 mm mesh). Nets were anchored using heavy gauge chain clipped to the cod and wing ends. Two polystyrene buoys were placed in each net’s cod end to force a pocket of net above the water’s surface. This created a space where by-catch (turtles, birds or water rats) could take refuge until the net was processed. Two double- wing and four small fykes were deployed at each site with nets positioned to strategically sample the range of in microhabitats present at the site. Double-wing fyke nets were deployed together and in opposition with one opening upstream and the second opening downstream. Single fykes were deployed separately within the microhabitats available at the site, e.g. snag, bare bank, submerged vegetation etc. An additional large double-wing fyke net (2 x 10 m wing, 12 mm mesh, 5 m funnels, 1.2 m high) was set in the estuarine sites. Fykes were set before dusk and collected after dawn ensuring that each site was set for a minimum of 14 hours. This time period allowed capture during crepuscular movement and allowed adequate time for nets to perform. A subset of nets were deployed at sites where conditions were inappropriate to deploy the full set of nets.

2.3.1. Fish processing

Fish captured during each survey were taxonomically identified to species. The only exception were carp gudgeon (Hypseleotris spp.), which were identified to genus. Fish species were considered to be native (endemic to the catchment confirmed by historical data/records), translocated (Australian native fish species not considered endemic to the catchment) (McNeil and Hammer 2007) or exotic (taxa not endemic to Australia). For each species, total length (TL) was recorded for the first 50 fish collected. This is considered a representative subset to create reliable length frequency distributions. The only exception was the exotic species Gambusia (Gambusia holbrooki). This species reproduces continuously throughout the year (Milton and Arthington 1983) so length frequency does not provide information relevant to the health of the site. In such instances, only the first 20 Gambusia recovered at each site were measured. All fish captured were also visually assessed for the presence of fungal infection, subcutaneous endoparasites, spawning condition and congenital abnormalities.

10 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

2.3.2. Site Health Assessments

Two techniques were used to characterise health and condition at each site. The Biological Condition Gradient (BCG) uses a ten step approach to score biotic communities and ecological assets in a landscape context (Davies and Jackson, 2006). Full details for the application of BCG scoring in the WMLR may be found in Mathwin et al. (2014).

The species attribute designations, attribute tables and tier descriptions used in this assessment are presented in Appendix A. Each fish species known in the study area was assigned to an Attribute from I to V. Estuarine specialist species were excluded from assessment in the BCG due to the sheer number of extra species that this would add to the analysis coupled with uncertainties about the known distribution and Attributes of each species. Attribute VI (Non-native and introduced taxa) was determined to include Murray Darling translocations. Where records of exotics exist in the catchment but were not identified at the site, this Attribute was scored a Tier 1 to highlight absence. When scoring Attribute VII (Organism condition) the condition of non-native and introduced taxa (Attribute VI) was not considered. Attribute VIII was not scored in this study as sufficient limnology data is unavailable for these catchments at this time. For each site each relevant attribute was allocated a tier value. These values were added and divided by the number of attributes scored to provide an average BCG score for the site.

The second health assessment approach used the Fish Health Index (FHI). This approach scores the size and composition of fish communities to assess the stability of native fish populations. Detailed methods for this approach may be found in McNeil et al. (2011). The fish population data collected was used to create a scorecard for fish health at each site for three key criteria; Native fish biodiversity, Exotic fish diversity and relative Dominance (based on abundance and population structure) of native fish. The Native Fish Diversity score reflects the number of native fish species identified at each site. It was scored between -5 and 9 points. The presence of one native species of fish scored 3 points, the second and third species each add 2 points and subsequent species add a single point to a maximum of 9 points. An absence of any native fish resulted in a score of -5. For the purpose of this assessment translocated native species were considered ‘native’. Opposing this was the Exotic Fish Diversity Score. This generated a negative score between 0 and -9 points which reflected the number of exotic fish species present at each site. The first exotic fish species identified scored -3 points, the second and third species each scored -2 points and subsequent species score a further -1 point to a maximum of -9 points. A third score, Dominance, was scored to reflect the significance and stability of the native fish populations, as observed from our data. Three points were awarded if a complex population structure was observed in native fish. Three points were deducted if a complex population structure was observed in exotic fish species. Three points were awarded if there were high numbers of native fish present and three points were deducted if there were high numbers of exotic fish present.

11 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Three points were deducted if an exotic predator was present at the site. These values were summed to reach a final Dominance score for each site which ranged from 6 to -9. These three scores were compiled to create a single Fish Health Index for each site reflecting the overall condition of the native fish population present.

As both indices provide different metrics, a combination of both were used to provide greater resolution of condition to inform each site. We assessed differences between outputs scores by conducting a regression analysis to assess any differences produced by BCG and FHI.

To simplify both BCG and FHI outputs for each site, scores were standardised by assigning their scoredvalues to three broad condition categories (Good (green), Intermediate (amber), Poor (red)) which encompassed the range of potential values obtained. Under this approach, sites with scores of 1 to 2.9 (BCG) and 5+ (FHI) were classified in “good” condition; 3 to 4.9 (BCG) and -5 to 5 (FHI) “intermediate condition”; and 5 to 6 (BCG) and < -5 (FHI) were assigned to sites in “poor condition”. Final ratings were compared to identify any differences in score between the two methods. Scores were presented spatially and by sampling season to assess the spatial and temporal trends in fish community health. Mean (average) FHI and BCG was also calculated from each site’s seasonal scores to compare condition indices with historical data. This enabled current scores to be extrapolated at the site level where historical sampling may have occurred during spring or autumn.

12 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

3. RESULTS

3.1. General results

A total of 57 sites were sampled throughout the three years (Figure 1). Fish communities were present at all sites during autumn and spring sampling events where water was present except at two sites (Tanunda Creek Gauge spring 2015 and Victoria Reserve autumn 2017). Overall, 50 site surveys were conducted in 2015, 53 in 2016 and 58 in 2017 (Table 1). To simplify reporting of different components of this study, results are presented in two sections: i) sites sampled for the “WMLR fish condition monitoring” and ii) sites sampled for the “Fleurieu estuarine health monitoring”. In total, 90,489 fish (83,945 native and 6,544 exotic) were captured over the three year period. These comprised eleven freshwater native, eighteen estuarine native, three freshwater translocated and six freshwater exotic fish species (Table 2 and 3).

The most abundant native fishes were common Galaxias (Galaxias maculatus, n = 22,698), black bream (Acanthopagrus butcheri, n = 22,479) and smallmouth hardyhead (Atherinosoma microstoma, n = 18,880). All black bream and smallmouth hardyhead, and over 65% of common Galaxias were recorded during sampling of estuary sites. The most frequently recorded species were common Galaxias (85 occasions, 26 sites), Congolli (Pseudaphritis urvillii, 75 occasions, 21 sites) and climbing Galaxias (Galaxias brevipinnis, 42 occasions, 23 sites).

Gambusia was the most abundant and frequently captured exotic species recorded (n = 6,265, 16 sites) followed by redfin perch (Perca fluviatilis, n = 134, 15 sites). The four remaining exotic species (goldfish, brown trout, speckled livebearer and European carp) each contributed 1, 86, 27, and 31 individuals respectively. The most frequently commonly encountered exotic species were redfin perch (36 occasions) and gambusia (33 occasions).

Water quality data are presented in Appendix C.

Table 1. Number of fish monitoring sites surveyed per season across the WMLR between 2015 and 2017.

Year Season Total sites 2015 Autumn 27 Spring 23 2016 Autumn 24

Spring 29 2017 Autumn 33

Spring 25

13 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Table 2. Total catch of native (green) and invasive (red) fish species at sites in the WMLR fish condition study. This includes native diversity score (see section 2.3.2).

caudimaculatus

brevipinnis

Native Diversity Catchment River Site Date

Gawler Jacobs Creek Jacobs Creek Old Gauge 2/05/2017 31 1 North Para River Gomersal Rd Bridge 24/03/2015 273 70 1 4/03/2016 109 106 23 1 2/05/2017 195 19 1 1 Yaldara 24/03/2015 170 10 245 4 2 7/10/2015 49 58 1 54 4 2 4/03/2016 113 168 181 1 2 9/11/2016 10 21 4 2 2/05/2017 31 1 17 97 1 2 12/12/2017 391 17 2 11 1 2 South Para River Mt Crawford 1/05/2017 8 1 Victoria Creek 3/03/2016 26 1 Victoria Reserve 3/03/2016 0 Tanunda Creek Tanunda Ck Gauge 7/10/2015 0 9/11/2016 7 1 13/12/2017 24 1 Torrens First Creek Chinaman's Hut 11/11/2016 97 1 Waterfall Gully 11/11/2016 103 1 Lower Torrens River Fox Creek 1/05/2017 6 326 2 2 Upper Torrens Talunga Park Bridge 24/03/2015 27 0 Patawalonga Brownhill Creek Craigburn Dam 25/03/2015 333 2 0 DS Brownhill Caravan Park 8/10/2015 1 111 2 10/11/2016 58 1 21/04/2017 4 1 12/12/2017 2 32 2 US Caravan Park 2/03/2016 90 1 Patawalonga West Beach Road Bridge 12/12/2017 33 154 1 Sturt Creek DS Sturt Gorge Retention Dam 8/10/2015 28 31 7 1 US Sturt Gorge FloodRetention Dam 8/10/2015 32 5 2 1 Onkaparinga Aldgate Creek Mylor Bridge 25/03/2015 12 265 2 8/10/2015 67 171 2 2/03/2016 2 162 2 10/11/2016 14 106 2

14 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

caudimaculatus

brevipinnis

Native Diversity Catchment River Site Date

Galaxias Galaxias Galaxias maculatus Galaxias olidus spp Hypseleotris australis Nanoperca grandiceps Philypnodon macrostoma Philypnodon olorum Pseudogobius urvillii Pseudaphritis auratus Carassius Cyprinus carpio holbrooki Gambusia fluviatilis Perca Phalloceros trutta Salmo 21/04/2017 109 56 2 2 13/12/2017 135 265 2 Lenswood Ck Lenswood Gauge 9/11/2016 83 1 1 1/05/2017 41 2 1 12/12/2017 82 7 1 Onkaparinga River MtBoldGate4 20/10/2015 1 1 1 Scott Creek Scott Creek Gauge 2/03/2016 707 1 0 Field River Field River Railway Tunnel 26/03/2015 406 2 10 3 20/10/2015 820 3 20 3 10/03/2016 212 31 2 20/04/2017 187 50 2 14/12/2017 634 34 2 Willunga Creek Willunga Creek Ross Roses 20/04/2017 27 0 St. Johns Rd 20/04/2017 0 Myponga Myponga River Pages Flat 18/03/2015 26 1 150 6 2 Carrickalinga Carrickalinga Riverview Drive 2/11/2016 120 139 7 1 3 Rose Cottage 2/11/2016 8 5 1 13/12/2017 9 3 3 2 Bungala Bungala River Stornoway 10/04/2017 41 394 2 Yankalilla Yankalilla River Ingalalla Falls 27/03/2015 43 10 1 Yankalilla Bridge River 26/03/2015 10 116 31 30 4 4/04/2017 224 6 29 3 Parananacooka Parananacooka Old Bridge 8/12/2017 2287 2 2 Yattagolinga Yattagolinga River Croser 4/04/2017 81 775 2 Hindmarsh Hindmarsh River Hindmarsh Gauge 23/03/2016 360 3 28 3 11/04/2017 148 2 23 3 5/12/2017 103 1 1 21 4 Inman BackValley Trib BackValley Gauge 24/04/2015 6 52 63 3 Inman River Glacier Rock 25/04/2015 10 1 16/10/2015 8 1 1 10/03/2016 472 2 0 2/11/2016 12 4 8 2 11/04/2017 10 5 1 7/12/2017 1 3 1 2 Inman Divine Gauge 24/04/2015 4 159 37 1 114 14 4 16/10/2015 8 13 11 3 2 10/03/2016 213 124 463 8 2

15 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

caudimaculatus

brevipinnis

Native Diversity Catchment River Site Date

16 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Table 3. Total catch of native (green) and invasive (red) fish species at estuarine health study sites in the WMLR fish condition study. This includes native diversity score (see section 2.3.2).

urvillii

brevipinnis

River Site Date Acanthopagrus butche Acanthopagrus tamarensis Afurcagobius forsteri Aldrichetta australis Anguilla bifrenatus Arenigobius japonicus Argyrosomus trutta Arripis Atherinosoma microstoma krefftii Bathygobius depressus Callogobius macrocephalus Cnidoglanis tapirina Rhombosolea Galaxias Galaxias maculatus marmoratus Gymnapistes spp. Hypseleotris argentea Liza Macquaria colonorum platynotus Mugilogobius erebi Nematalosa grandiceps Philypnodon macrostomus Philypnodon georgianus Pseudocaranx olorum Pseudogobius Pseudaphritis nigra Stigmatopora glaber Tetractenos Cyprinus carpio holbrooki Gambusia fluviatilis Perca Native Diversity Bungala River Bungala Caravan Park Bridge 17/03/2015 559 283 104 2713 778 1 10 19 8

13/10/2015 833 1 6 632 14 288 2 530 8

24/03/2016 149 163 66 13915 8 47 8 9 8

18/10/2016 33 2 1 1 2 5

3/04/2017 3 54 1 1570 94 3 1 15 8

8/12/2017 1375 11 13 6 8 2 4 3 15 9

Bungala South Rd 17/03/2015 51 27 2

13/10/2015 168 2 20 3

24/03/2016 37 16 2

18/10/2016 114 3 2

3/04/2017 224 25 2

8/12/2017 96 1 51 3

Hay Flat Rd 17/03/2015 368 19 2

13/10/2015 42 6 2

24/03/2016 475 3 1 8 4

18/10/2016 34 1 5 1 3

3/04/2017 393 11 2

8/12/2017 1 120 55 3

Hindmarsh River Hindmarsh Estuary 20/03/2015 284 3 85 1 32 7 2 7

15/10/2015 2 4 24 268 5 36 7 19 8

11/03/2016 298 5 45 27 1 63 287 3 15 9

4/11/2016 40 1 39 33 90 1 9 1 9 9

7/04/2017 58 12 49 4 5 1 7 29 8 6 10

5/12/2017 32 5 214 12 3 11 18 1 4 51 10 25 12

Lamont Rd 20/03/2015 2 15 3 3

15/10/2015 2 58 82 9 7 5

11/03/2016 22 4 214 2 1 5

4/11/2016 1 106 12 7 4

7/04/2017 11 6 71 2 88 5 3 7

5/12/2017 4 5 2 65 11 35 6

Cootamundra Reserve 20/03/2015 4 650 45 1 66 5

15/10/2015 4 530 25 128 4

11/03/2016 8 1142 45 124 4

4/11/2016 272 364 6 7 4

17 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

urvillii

brevipinnis

River Site Date butche Acanthopagrus tamarensis Afurcagobius forsteri Aldrichetta australis Anguilla bifrenatus Arenigobius japonicus Argyrosomus trutta Arripis Atherinosoma microstoma krefftii Bathygobius depressus Callogobius macrocephalus Cnidoglanis tapirina Rhombosolea Galaxias Galaxias maculatus marmoratus Gymnapistes spp. Hypseleotris argentea Liza Macquaria colonorum platynotus Mugilogobius erebi Nematalosa grandiceps Philypnodon macrostomus Philypnodon georgianus Pseudocaranx olorum Pseudogobius Pseudaphritis nigra Stigmatopora glaber Tetractenos Cyprinus carpio holbrooki Gambusia fluviatilis Perca Native Diversity 7/04/2017 63 71 316 65 144 5

5/12/2017 5 194 16 117 1 4

Inman River Inman estuary 19/03/2015 17515 1 7 2 182 13 58 6

14/10/2015 5 8 1084 1 84 55 30 1 2 8

8/03/2016 261 6 8 18 1 2 24 9 464 1 9

3/11/2016 10 104 2 4 1 1 52 5 1 2 2 1 11

5/04/2017 179 85 9 1 1 18 2 1 22 8 17 1 12

6/12/2017 3 17 2 5 1 2 91 188 10 106 10

Armstrong Rd Bridge 19/03/2015 4 318 1839 2

14/10/2015 25 365 11 2

8/03/2016 17 1465 1 371 3

3/11/2016 27 105 1 2

5/04/2017 1 55 291 1 208 4

6/12/2017 2 522 2 57 1 3

Swains Crossing Road 19/03/2015 4 841 1 2

14/10/2015 3 7 128 1 50 1 4

8/03/2016 51 4 4 3 2

3/11/2016 2 5 1 3 3

5/04/2017 34 1 1 1 1 3

6/12/2017 3 1 1

Myponga River Myponga estuary 18/03/2015 15 47 3 2821 813 24 3 56 8

21/10/2015 245 104 1 1 94 682 7 2 12 9

22/03/2016 424 229 10 105 655 61 111 1 155 9

1/11/2016 1 41 43 7 1 8 1 16 8

1/12/2016 6 1 498 1 28 1 1 76 22 3 39 11

10/04/2017 3 62 393 2 6 9 8 13 55 9

Myponga pumphouse 18/03/2015 87 1 5 3

21/10/2015 53 18 2

23/03/2016 41 8 2

1/11/2016 81 791 45 3

1/12/2016 91 776 86 1 3

10/04/2017 11 1564 225 3

18 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

3.2. Site Health Assessment

FHI and BCG scores for each site are reported in Tables 4 and 5. Comparison of the FHI and BCG site scores showed a strong linear relationship (r2 = 0.75) between the two health assessment methods (Figure 3), suggesting output scores and their relative classification were comparable between methods. The distribution of scores across the study area revealed the spatial and temporal components of fish condition across the WMLR over a three year period (Figures 4–7).

Table 4. FHI and BCG scores for the sites in the WMLR fish condition study (I = Historically documented, sensitive, long-lived, or regionally endemic taxa, II = Sensitive-Rare Taxa, III = Sensitive Common Taxa, V = Tolerant taxa, VI = Non-native or intentionally introduced taxa, VII = Organism and population condition, IX = Spatial and temporal extent of detrimental effects, X = Ecosystem connectivity).

BCG Catchment Watercourse Site Date fish Native biodiversity fish Exotic diversity Relative dominance FHI I II III V VI VII IX X Total Jacobs Creek Gawler Jacobs Creek Old Gauge 2/05/2017 3 0 6 9 1 1 2 4 5 2.6 North Para Gomersal Rd River Bridge 24/03/2015 3 -3 3 3 6 6 3 3 2 5 4 4.1 4/03/2016 3 -5 0 -2 6 6 3 4 2 5 4 4.3 2/05/2017 3 -5 0 -2 6 6 3 4 2 5 4 4.3 Yaldara 24/03/2015 3 -5 -3 -5 1 6 3 4 2 5 4 3.6 7/10/2015 5 -6 -3 -4 1 6 3 4 2 5 4 3.6 4/03/2016 5 -5 -3 -3 1 6 3 4 2 5 4 3.6 9/11/2016 5 -3 0 2 1 6 3 2 2 5 4 3.3 2/05/2017 5 -6 -3 -4 2 6 3 5 2 5 4 3.9 12/12/2017 5 -6 0 -1 1 6 3 4 2 5 4 3.6 South Para River Mt Crawford 1/05/2017 3 0 0 3 4 3 2 4 4 4 3.5 Victoria Creek 3/03/2016 3 0 6 9 1 1 1 5 2 2.0 Victoria Reserve 3/03/2016 0 0 0 0 5 1 4 5 5 4.0 Tanunda Ck Tanunda Creek Gauge 7/10/2015 0 0 0 0 6 1 4 5 5 4.2 9/11/2016 3 0 0 3 5 1 4 5 5 4.0 13/12/2017 3 0 0 3 5 1 4 5 5 4.0 Chinaman's Torrens First Creek Hut 11/11/2016 3 0 6 9 1 1 2 1 1 1.2 Waterfall Gully 11/11/2016 3 0 6 9 1 1 2 1 1 1.2 Lower Torrens River Fox Creek 1/05/2017 6 -3 3 6 3 2 2 4 4 4 3.2 Talunga Park Upper Torrens Bridge 24/03/2015 0 -3 -3 -6 6 6 6 5 5 5.6 Patawalonga Brownhill Creek Craigburn Dam 25/03/2015 0 -5 -6 -11 6 6 6 5 5 5.6 DS Brownhill Caravan Park 8/10/2015 5 0 6 11 4 2 1 2 3 4 2.7 10/11/2016 3 0 6 9 5 2 1 2 3 4 2.8 21/04/2017 3 0 3 6 5 3 1 2 3 4 3.0 12/12/2017 5 0 6 11 4 2 1 2 3 4 2.7 US Caravan Park 2/03/2016 3 0 3 6 5 2 1 2 3 4 2.8 West Beach Patawalonga Road Bridge 12/12/2017 3 -3 -3 -3 3 5 3 5 2 5 5 4.0 DS Sturt Gorge Sturt Creek Retention Dam 8/10/2015 3 -5 -3 -5 5 3 4 4 4 4 4.0 US Sturt Gorge FloodRetention Dam 8/10/2015 3 -5 -3 -5 5 3 4 4 4 4 4.0 Onkaparinga Aldgate Creek Mylor Bridge 25/03/2015 5 0 6 11 3 1 1 1 2 3 1.8

19 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Catchment Watercourse Site Date fish Native biodiversity fish Exotic diversity Relative dominance FHI I II III V VI VII IX X Total 8/10/2015 5 0 6 11 2 1 1 1 2 3 1.7 2/03/2016 5 0 6 11 3 1 1 1 2 3 1.8 10/11/2016 5 0 6 11 3 1 1 1 2 3 1.8 21/04/2017 5 -3 6 8 2 2 2 1 2 3 2.0 13/12/2017 5 0 6 11 2 1 1 1 2 3 1.7 Lenswood Lenswood Creek Gauge 9/11/2016 3 -3 3 3 2 2 2 2 3 2.2 1/05/2017 3 -3 3 3 2 2 2 2 3 2.2 12/12/2017 3 -3 3 3 2 2 2 2 3 2.2 Onkaparinga River MtBoldGate4 20/10/2015 3 -3 -3 -3 6 6 3 3 4 4 3 4.1 Scott Creek Scott Creek Gauge 2/03/2016 0 -5 -6 -11 6 5 6 4 4 5.0 Railway Field River Field River Tunnel 26/03/2015 7 0 6 13 2 1 2 1 2 4 2 2.0 20/10/2015 7 0 6 13 2 1 2 1 2 4 2 2.0 10/03/2016 5 0 6 11 2 1 1 2 4 2 2.0 20/04/2017 5 0 6 11 2 1 1 2 4 2 2.0 14/12/2017 5 0 6 11 2 1 1 2 4 2 2.0 Willunga Creek Willunga Creek Ross Roses 20/04/2017 0 -3 0 -3 6 6 5 5 5.5 St. Johns Rd 20/04/2017 0 0 0 0 6 1 5 5 4.3 Myponga Myponga River Pages Flat 18/03/2015 5 -5 -6 -6 2 4 5 4 4 4 3.8 Riverview Carrickalinga Carrickalinga Drive 2/11/2016 7 -3 3 7 2 2 2 2 3 2 2.2 Rose Cottage 2/11/2016 3 -3 -3 -3 4 6 3 4 3 2 3.7 13/12/2017 5 -3 -3 -1 5 4 3 4 3 2 3.5 Bungala Bungala River Stornoway 10/04/2017 5 0 6 11 2 2 1 2 3 4 2.3 Yankalilla Yankalilla River Ingalalla Falls 27/03/2015 3 -3 0 0 2 3 2 3 4 2.8 Yankalilla Bridge River 26/03/2015 9 0 6 15 3 1 3 1 2 3 3 2.3 4/04/2017 7 0 6 13 3 1 3 1 2 3 3 2.3 Parananacooka Parananacooka Old Bridge 8/12/2017 5 0 6 11 4 1 1 2 3 3 2.3 Yattagolinga Yattagolinga River Croser 4/04/2017 5 0 6 11 1 1 1 1 2 2 1.3 Hindmarsh Hindmarsh Hindmarsh River Gauge 23/03/2016 7 0 6 13 1 1 2 1 1 3 2 1.6 11/04/2017 7 0 6 13 1 1 2 1 1 3 2 1.6 5/12/2017 8 0 6 14 1 1 2 1 1 3 2 1.6 BackValley Inman BackValley Trib Gauge 24/04/2015 7 0 6 13 1 4 1 2 3 3 2.3 Inman River Glacier Rock 25/04/2015 3 0 0 3 4 6 1 4 4 3 3.7 16/10/2015 3 -3 0 0 4 6 2 4 4 3 3.8 10/03/2016 0 -5 -6 -11 5 6 5 4 4 3 4.5 2/11/2016 5 -3 0 2 4 3 2 4 4 3 3.3 11/04/2017 3 -3 0 0 5 3 2 4 4 3 3.5 7/12/2017 5 -3 0 2 4 6 2 4 4 3 3.8 Inman Divine Gauge 24/04/2015 8 -5 -3 0 2 3 5 4 4 4 3 3.6 16/10/2015 5 -5 0 0 3 6 4 4 4 4 3 4.0 10/03/2016 5 -5 -3 -3 2 6 5 5 4 4 3 4.1 11/04/2017 8 -7 0 1 4 2 4 5 4 4 3 3.7 7/12/2017 8 -5 -3 0 4 3 4 4 4 4 3 3.7 Deep Creek Deep Creek Deep Creek Crossing 15/12/2017 3 0 3 6 2 1 2 2 2 1.8 Rangers Pump 25/04/2015 0 -3 -6 -9 5 6 2 2 2 3.4 Bollaparudda Bollaparudda Bollaparudda Creek Creek East campsite 2/12/2016 3 0 6 9 2 3 1 2 2 2 2.0 12/04/2017 5 0 6 11 2 2 1 2 2 2 1.8 Bollaparudda east DS 2/12/2016 5 0 6 11 4 2 1 2 2 2 2.2 12/04/2017 3 0 6 9 5 2 1 2 2 2 2.3 Bollaparudda east gorge 2/12/2016 3 0 6 9 5 2 1 2 2 2 2.3 12/04/2017 7 0 6 13 4 2 1 2 2 2 2.2

20 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Catchment Watercourse Site Date fish Native biodiversity fish Exotic diversity Relative dominance FHI I II III V VI VII IX X Total Bollaparudda Bollaparudda West West Dam 2/12/2016 0 0 0 0 6 6 1 5 5 4.6 Callawonga Callawonga Callawonga Gauge 24/04/2015 0 -3 -6 -9 6 6 6 3 3 2 4.3 22/10/2015 0 -3 -6 -9 6 6 6 3 3 2 4.3 9/03/2016 0 -3 -6 -9 6 6 6 3 3 2 4.3 1/12/2016 0 -3 -6 -9 6 6 6 3 3 2 4.3 18/04/2017 0 -3 -6 -9 6 6 6 3 3 2 4.3 7/12/2017 0 -3 -6 -9 6 6 6 3 3 2 4.3 Eric Bonython Tunkalilla conservation Tunkalilla Creek park 15/12/2017 5 0 6 11 2 3 1 2 2 2 2.0 Boat Harbour Boat Harbour Boat Harbour Gauge 25/04/2015 3 0 6 9 2 1 2 2 2 1.8 22/10/2015 3 -3 3 3 4 2 2 2 2 2.4 9/03/2016 3 0 3 6 2 1 2 2 2 1.8 1/12/2016 3 0 6 9 2 1 2 2 2 1.8 18/04/2017 3 0 6 9 2 1 2 2 2 1.8 14/12/2017 3 0 6 9 2 1 2 2 2 1.8 Rarkang creek Rarkang Creek Rarkang Dam 27/03/2015 5 0 6 11 2 2 1 2 2 2 1.8

21 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Table 5. FHI and BCG scores for the estuary health study sites (I = Historically documented, sensitive, long-lived, or regionally endemic taxa, II = Sensitive-Rare Taxa, III = Sensitive Common Taxa, V = Tolerant taxa, VI = Non-native or intentionally introduced taxa, VII = Organism and population condition, IX = Spatial and temporal extent of detrimental effects, X = Ecosystem connectivity).

BCG Catchment Watercourse Site Date fish Native biodiversity score fish Exotic diversity score Relative dominance FHIscore I II III V VI VII IX X Total Bungala Bungala Caravan Bungala River Park Bridge 17/03/2015 5 0 6 11 2 1 3 1 2 2 2 1.9 13/10/2015 7 0 3 10 3 3 3 1 2 2 2 2.3 24/03/2016 8 0 3 11 3 3 3 1 2 2 2 2.3 18/10/2016 7 0 3 10 3 3 3 1 4 2 2 2.6 3/04/2017 8 0 6 14 2 2 3 1 2 2 2 2.0 8/12/2017 8 0 3 11 2 3 3 1 2 2 2 2.1 Bungala South Rd 17/03/2015 5 0 6 11 2 2 3 1 2 2 2 2.0 13/10/2015 7 0 6 13 2 1 2 1 2 2 2 1.7 24/03/2016 5 0 6 11 2 2 3 1 2 2 2 2.0 18/10/2016 5 0 6 11 3 1 3 1 2 2 2 2.0 3/04/2017 5 0 6 11 2 1 3 1 2 2 2 1.9 8/12/2017 7 0 6 13 2 1 3 1 2 2 2 1.9 Hay Flat Rd 17/03/2015 5 0 6 11 2 1 3 1 2 2 2 1.9 13/10/2015 5 0 6 11 3 2 3 1 2 2 2 2.1 24/03/2016 8 0 6 14 3 1 2 1 2 2 2 1.9 18/10/2016 7 -3 6 10 3 2 2 2 2 2 2 2.1 3/04/2017 5 0 6 11 3 1 3 1 2 2 2 2.0 8/12/2017 5 0 6 11 2 1 3 1 2 2 2 1.9 Hindmarsh Hindmarsh Hindmarsh River Estuary 20/03/2015 8 0 3 11 3 6 3 1 4 3 2 3.1 15/10/2015 8 0 6 14 2 1 3 1 2 3 2 2.0 11/03/2016 8 0 6 14 2 2 5 1 2 3 2 2.4 4/11/2016 8 0 6 14 3 1 3 1 2 3 2 2.1 7/04/2017 8 0 3 11 3 3 3 1 3 2 2.5 5/12/2017 7 0 6 13 2 6 3 1 2 3 2 2.7 Lamont Rd 20/03/2015 5 0 3 8 3 6 3 1 4 3 2 3.1 15/10/2015 8 0 6 14 3 2 3 1 2 3 2 2.3 11/03/2016 7 0 3 10 3 6 5 1 4 3 2 3.4 4/11/2016 7 0 3 10 3 2 3 1 2 3 2 2.3 7/04/2017 8 0 6 14 3 2 3 1 2 3 2 2.3 5/12/2017 8 0 6 14 2 3 3 1 2 3 2 2.3 Cootamundra Reserve 20/03/2015 8 0 6 14 2 1 3 1 2 3 2 2.0 15/10/2015 7 0 6 13 1 1 3 1 2 3 2 1.9 11/03/2016 7 0 6 13 1 1 3 1 2 3 2 1.9 4/11/2016 7 0 6 13 3 1 3 1 2 3 2 2.1 7/04/2017 7 0 6 13 1 1 3 1 2 3 2 1.9 5/12/2017 7 -3 6 10 1 1 3 2 2 3 2 2.0 Inman Inman Inman River estuary 19/03/2015 7 -3 3 7 3 2 6 3 2 2 3 3 3.0 14/10/2015 7 -3 6 10 3 2 1 4 2 2 3 3 2.5 8/03/2016 8 -3 6 11 3 1 3 3 2 2 3 3 2.5 3/11/2016 7 -3 3 7 3 3 2 3 2 2 3 3 2.6 5/04/2017 8 0 3 11 3 1 3 3 2 1 3 3 2.4 6/12/2017 7 0 6 13 3 2 2 4 1 2 3 3 2.5 Armstrong Rd Bridge 19/03/2015 5 -3 0 2 2 6 3 3 5 4 3 4 3.8 14/10/2015 5 -3 3 5 2 6 2 3 2 2 3 4 3.0 8/03/2016 7 -3 6 10 2 4 2 3 2 2 3 4 2.8 3/11/2016 5 -3 6 8 2 6 2 3 2 2 3 4 3.0 5/04/2017 8 -3 3 8 1 4 2 3 2 2 3 4 2.6 6/12/2017 7 -6 0 1 2 4 3 3 2 4 3 4 3.1 Swains Crossing Road 19/03/2015 5 -3 3 5 3 6 3 3 2 4 3 4 3.5

22 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Catchment Watercourse Site Date fish Native biodiversity score fish Exotic diversity score Relative dominance FHIscore I II III V VI VII IX X Total 14/10/2015 8 -6 3 5 3 4 3 3 2 4 3 4 3.3 8/03/2016 5 -6 3 2 3 4 6 3 2 4 3 4 3.6 3/11/2016 7 -3 -3 1 3 4 6 3 2 4 3 4 3.6 5/04/2017 7 -6 3 4 3 4 6 3 2 4 3 4 3.6 6/12/2017 3 -3 -3 -3 3 6 6 3 2 4 3 4 3.9 Myponga Myponga Myponga River estuary 18/03/2015 7 0 6 13 2 1 3 1 2 2 2 1.9 21/10/2015 7 0 6 13 3 1 3 1 2 2 2 2.0 22/03/2016 8 0 6 14 1 1 3 1 2 2 2 1.7 1/11/2016 7 0 6 13 2 3 3 1 2 2 2 2.1 1/12/2016 7 0 6 13 2 2 3 1 2 2 2 2.0 10/04/2017 7 0 6 13 2 3 3 1 2 2 2 2.1 Myponga pumphouse 18/03/2015 7 0 6 13 3 2 3 1 2 2 2 2.1 21/10/2015 5 0 6 11 3 2 3 1 2 2 2 2.1 23/03/2016 5 0 6 11 3 2 3 1 2 2 2 2.1 1/11/2016 7 0 6 13 2 1 3 1 2 2 2 1.9 1/12/2016 7 -3 3 7 2 1 3 1 2 2 2 1.9 10/04/2017 7 0 6 13 1 1 3 1 2 2 2 1.7

6.0 R² = 0.7489 5.0

4.0

BCG 3.0

2.0

1.0 -15 -10 -5 0 5 10 15 FHI

Figure 3. Scatter plot comparing BCG and FHI scores at each site and date. R2 = 0.7489.

23 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 4. Composite map showing the distribution of BCG scores across the WMLR in each sampling season from autumn 2015 to spring 2017.

24 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 5. Composite map showing the distribution of FHI scores across the WMLR in each sampling season from autumn 2015 to spring 2017.

25 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 6. Composite map showing the distribution of BCG scores at Estuary sites in each sampling season from autumn 2015 to spring 2017. Myponga not sampled in spring 2017.

26 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 7. Composite map showing the distribution of FHI scores at Estuary sites in each sampling season from autumn 2015 to spring 2017. Myponga not sampled in spring 2017.

27 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

4. DISCUSSION

4.1. WMLR Fish condition

The condition of sites in each of the catchments sampled throughout the three year study period is discussed below. This discussion includes sites selected: i) as long-term monitoring sites; ii) as VWASP sites; iii) to periodically assess vulnerable or isolated fish populations; and iv) as novel fish monitoring locations to expand knowledge of fish distribution in the WMLR. Where possible, sites scored using the BCG and FHI methods in this study were compared with sites scored in previous studies in 2011 and 2013 (McNeil et al. 2011a, Mathwin et al. 2014) (Table 6). Sites in the four estuary health study reaches are discussed in section 4.2.

Table 6. Comparison of average 2015-17 BCG and FHI scores with scores in 2011 and 2013. Note: Historical BCG and FHI are not average values.

2015- 2015- 2017 2017 Average Average 2013 2011 Catchment Watercourse Site_Name BCG FHI BCG 2013 FHI BCG 2011 FHI Jacobs Creek Gawler Jacobs Creek 2.6 9 2.6 9 Old Gauge North Para Gomersal Rd 4.2 -0.3 4.43 4 River Bridge Yaldara 3.6 -2.5 4.14 0 4 -5 South Para Mt Crawford 3.5 3 4 -1 River Victoria Creek 2 9 2 9

Victoria 4 0 Reserve Tanunda Ck Tanunda Creek 4.1 2 Gauge Chinaman's Torrens First Creek 1.2 9 1.4 9 Hut Waterfall Gully 1.2 9 1.6 6

Lower Torrens Fox Creek 3.2 6 River Talunga Park Upper Torrens 5.6 -6 5.17 -14 Bridge Brownhill Patawalonga Craigburn Dam 5.6 -11 Creek DS Brownhill 2.8 9.3 4.53 3 Caravan Park US Caravan 2.8 6 Park West Beach Patawalonga 4 -3 5.14 -3 Road Bridge DS Sturt Gorge Sturt Creek 4 -5 Retention Dam US Sturt Gorge FloodRetention 4 -5 Dam Onkaparinga Aldgate Creek Mylor Bridge 1.8 10.5 2.88 11

Lenswood Lenswood Ck 2.2 3 Gauge

28 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

2015- 2015- 2017 2017 Average Average 2013 2011 Catchment Watercourse Site_Name BCG FHI BCG 2013 FHI BCG 2011 FHI Onkaparinga MtBoldGate4 4.1 -3 River Scott Creek Scott Creek 5 -11 5.6 -8 Gauge Field River Field River Railway Tunnel 2 11.8 3 11 Willunga Creek Willunga Creek Ross Roses 5.5 -3

St. Johns Rd 4.3 0 4.25 -5 Myponga Myponga Myponga River 2 13.2 estuary Myponga 2 11.3 pumphouse Pages Flat 3.8 -6 4 0 Riverview Carrickalinga Carrickalinga 2.2 7 2.33 11 Drive Rose Cottage 3.6 -2 4.5 1 Bungala Bungala Bungala River Caravan Park 2.2 11.2 Bridge Bungala South 1.9 11.7 3 11 Rd Hay Flat Rd 2 11.3 3 8

Stornoway 2.3 11 3 0 Yankalilla Yankalilla River Ingalalla Falls 2.8 0 4 -3

Yankalilla 2.3 14 2.41 11 Bridge River Parananacooka Parananacooka Old Bridge 2.3 11 Yattagolinga Yattagolinga Croser 1.3 11 River Hindmarsh Cootamundra Hindmarsh 2 12.7 2 13 2 15 River Reserve Hindmarsh 2.5 12.8 Estuary Hindmarsh 1.6 13.3 Gauge Lamont Rd 2.6 11.7 BackValley BackValley Inman 2.3 13 4 3 Trib Gauge Armstrong Rd Inman River 3 5.7 Bridge Glacier Rock 3.8 -0.7 4.38 -3 5 -6 Inman Divine 3.8 -0.4 5 -11 Gauge Inman estuary 2.6 9.8

Swains 3.6 2.3 4 2 Crossing Road Deep Creek Deep Creek Deep Creek 1.8 6 3.75 -11 Crossing Rangers Pump 3.4 -9 3.75 -14 Bollaparudda Bollaparudda Boolaparuda 1.9 10 Creek Creek East campsite Bollaparudda 2.3 10 east DS Bollaparudda 2.3 11 east gorge Bollaparudda Bollaparudda 4.6 0 West West Dam 29 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

2015- 2015- 2017 2017 Average Average 2013 2011 Catchment Watercourse Site_Name BCG FHI BCG 2013 FHI BCG 2011 FHI Callawonga Callawonga Callawonga 4.3 -9 5.2 -14 Gauge Eric Bonython Tunkalilla Tunkalilla conservation 2 11 Creek park Boat Harbour Boat Harbour Boat Harbour 1.9 7.5 2.6 6 Gauge Rarkang creek Rarkang Creek Rarkang Dam 1.8 11

4.1.1. Gawler Catchment

Of the seven sites monitored in the Gawler River catchment, six had been monitored previously and two of those are long-term VWASP monitoring sites Yaldara and Tanunda Creek. All sites previously monitored and scored using BCG and FHI scoring methods maintained similar scores between the 2015-2017 average and the previous years (Table 6).

The VWASP site at Yaldara consistently scored in the intermediate amber category of both the BCG and FHI scores. The consistent scores at this site mostly derive from good long-term populations of native fish, contrasted by long-term populations of invasive species. A scouring flood associated with widespread regional rainfall in spring 2016 (Figure 8) removed a large amount of built up sediment and emergent vegetation at this site (and in the river in general) and had an immediate impact on the native and invasive species present at the site. However, both native and invasive species returned to pre-flood levels and an additional invasive species emerged in European carp, which was present elsewhere in the catchment prior to the flood. Besides the additional invasive species, the main concern with the Yaldara site is the long-term absence of common Galaxias, which has not returned despite adequate flows in 2016. This indicates connectivity problems with downstream populations, most likely as a result of barriers to fish passage (such as the North Para flood retention dam) and limited fish passage resulting from decreased flow duration.

30 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 8. Flow rate in ML/day at Yaldara Gauge (A5050502) on North Para River. Heavy rainfall in the middle and end of September 2016 caused widespread flooding in the MLR.

The Tanunda Creek site was monitored every season for the last three years but was dry every autumn. Despite regular seasonal drying, the site had mountain Galaxias on two or the three occasions when there was sufficient water to sample in spring. These fish are likely to come from an as yet unidentified source population upstream. When the site was dry during autumn sampling, a site on the North Para at the Gomersal Rd bridge was sampled as an alternative. Despite being the nearest permanent pool to the Tanunda Creek site, there were no records of mountain Galaxias at Gomersal Rd during 2015-2017 and the fish population there more resembled the fish population elsewhere in the North Para than Tanunda Creek.

The Victoria Creek and Jacobs Creek Old Gauge sites were monitored once during the sampling period due to their small but significant populations of climbing and mountain Galaxias respectively. It was deemed that frequent sampling at these sites may have an adverse effect on the populations. There appears to be little change in the health of these populations and they should only be subject to occasional monitoring in the future. The Victoria Reserve site was less than one kilometre downstream from the Victoria Creek site and a deep permanent pool maintained by a weir, yet there were no fish present at the site. This site demonstrated the limited range of the climbing

31 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Galaxias population upstream and indicates that the population is likely to be highly vulnerable to disturbance due to its patchy distribution.

4.1.2. Torrens Catchment

The Torrens catchment was monitored at three widely disparate sites including one VWASP site upstream at First Creek. As with some other sites in this study the First Creek sites were only monitored once during the sampling period due to their isolated populations of mountain Galaxias. Once again, it was deemed that frequent sampling at these sites may have an adverse effect on the populations. As with many other sites in the WMLR, the large flood event in spring 2016 caused a large scouring event in First creek associated with significant landslides, and movement of sediment and debris (Figure 9). The decision to monitor this site early was made through concern that the mountain Galaxias population may have been affected by the extreme flow event. However, there appears to be little change in the health of this population and it should only be subject to occasional monitoring in the future.

Figure 9. Landslide and damage to First Creek stream as a result of flooding in spring 2016.

Fox Creek remains an important source population for mountain and climbing Galaxias in the Torrens catchment. These species were occasionally present in low numbers in the nearby Cudlee Creek site during monitoring for the environmental flows project (McNeil et al. 2011b). The site was sampled but not scored for fish population health in a previous condition assessment (McNeil et al. 2011a), although the population structure was very similar to the present study. The additional discovery of redfin perch at this site was of concern and decreased the health score for the site. The site should be considered for regular monitoring.

32 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Sites in the upper reaches of the Torrens have consistently shown to be in an extremely degraded state (McNeil et al. 2011, Mathwin et al. 2014, Schmarr et al. 2014) and the site at Talunga Park Bridge reflects this (Table 6). Despite the presence of mountain Galaxias populations in the connected Millers Creek (Schmarr et al. 2014), the presence of high numbers of invasive species as well as degraded habitat and flow conditions have prevented their reestablishment. Apart from occasional monitoring, work in this part of the Torrens should investigate the factors preventing healthy native fish populations from re-establishing in the reach.

4.1.3. Patawalonga Catchment

Six sites were monitored within the Patawalonga catchment with one VWASP site at Brownhill Creek Caravan Park. In general, the sites in the catchment were in highly urbanised areas and the health scores over this study and previous studies reflect this. Remarkably though, the persistent population of mountain Galaxias with occasional climbing Galaxias presence, and the absence of invasive species at the Brownhill Creek site maintain this site at a healthy score on the BCG and FHI scales (Table 6). Whilst other sites within the catchment often had adequate populations of one native species, they were also home to populations of Gambusia and redfin perch. These sites are healthy enough to support fish populations. Ongoing work to restore habitat and flow may help re- establish other native species – especially diadromous species – at sites where fish passage is unimpeded. Samples for upstream and downstream of Sturt Gorge flood retention dam and Craigburn Lake represent novel records for these sites.

4.1.4. Onkaparinga

Of the four sites monitored during this study period, two were long-term VWASP sites: Mylor Bridge and Lenswood Gauge. Both of these sites scored highly on most occasions due to large populations of climbing and mountain Galaxias at Mylor Bridge and mountain Galaxias at Lenswood, although the presence of invasive redfin perch at Lenswood detracted from the health of that site. Monitoring at the Lenswood site was not conducted in 2015 and early 2016, but previous sampling at the site for the GWAP project did not detect any redfin perch at this site (Maxwell et al. 2015). It cannot be ruled out that the extra connectivity enabled by flooding in spring 2016 enabled redfin to colonise this site from populations down or upstream.

The other two sites in the catchment (Mt Bold Gate 4 and Scott Creek Gauge) had intermediate or poor health scores with both sites being downstream of Mount Bold Reservoir but above Clarendon Weir. The Mt Bold Gate 4 site had a very poor native fish population in combination with the presence of brown trout and Scott Creek had no native species and was dominated by invasive

33 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 species. Whilst the habitat and flow values for these sites were better than many other sites with poor overall scores (i.e. riparian cover restored, flow regime intact), the presence of large invasive fish source populations in Mount Bold Reservoir and Clarendon Weir Pool may be contributing to the poor fish health at these sites. This survey was the first survey by SARDI in the reach above Clarendon Weir and follows a previous recommendation to sample in the reach, further demonstrating the benefit of the adaptive approach taken with this monitoring program.

4.1.5. Field River

The Field River VWASP site sustained a healthy fish population throughout the study period and the health score was comparable to the previous score in 2011 (Table 6). This result is surprising given the highly urbanised and altered habitat upstream of the site. The site is maintained by near- permanent baseflow and is in close proximity to the outlet into Hallett Cove. The fish population is dominated by diadromous species so it is likely that this small pocket of habitat is regularly replenished by fish from other catchments. Nonetheless, this site demonstrates the value of providing healthy coastal stream habitat and flow regime.

4.1.6. Willunga Creek

Willunga Creek was sampled opportunistically as a follow up to the speckled livebearer eradication program conducted in 2010. Previous surveys conducted immediately after the eradication program failed to detect speckled livebearer in Willunga Creek. Unfortunately, the species was detected at one site in the present study. The absence of native species, the re-emergence of livebearer and the highly altered nature of the habitat in this stream all contributed to very low fish population health scores. The site at which the species was detected was downstream of what was assumed to be the original source location at St John’s Road. At this point in time the full extent of the livebearer population is not known, but it is recommended that further surveys are undertaken to reveal how much the livebearer population has recovered.

4.1.7. Myponga River

Besides the sites included in the estuary health part of this project, which are discussed below, only one site was sampled in the Myponga River upstream of the Reservoir. The site at Pages Flat had an intermediate BCG score and a poor FHI score which follow on from intermediate scores in 2011 (Table 6). Whilst the site was home to important populations of mountain and climbing Galaxias, it was dominated by invasive Gambusia and redfin perch.

34 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

4.1.8. Carrickalinga River

The two sites monitored in the Carrickalinga River – Riverview Drive and Rose Cottage - were scored as good and intermediate respectively on both the BCG and FHI scales. Riverview Drive is located at the estuary of this river; however, the salinity and fish population of the pool suggests that it is predominantly a freshwater pool with only diadromous species present and none of the estuarine species present at other estuary sites. Rose cottage had lower numbers of fish and the health scores suffered from the presence of redfin perch. The site is notable as the only site monitored by SARDI over the last 10 years to have had all three species of Galaxias present in a single sample, albeit in very low numbers (Schmarr et al. 2014). It may be important to continue monitoring the Carrickalinga catchment as it is one of the demonstration reaches for low-flow bypasses. Any improvement in fish population health achieved by returning low flows to the catchment should be detectable due to the presence of multiple flow dependent species.

4.1.9. Bungala River

Besides the sites included in the estuary health part of this project which are discussed below, only Stornoway was sampled in the Bungala catchment, which was in the upper reaches of an unnamed tributary. This site scored highly on the BCG and FHI due to abundant numbers of common and climbing Galaxias. The score from the previous 2011 survey improved due to the absence of invasive redfin perch from the site. Due to the small pool size and restricted distribution of this population, only occasional future monitoring at this site is recommended to prevent undue stress on the fish population.

4.1.10. Yankalilla River

The two sites in the Yankalilla River represent sites that have been previously sampled (Table 6). Ingalalla Falls was assessed as being in good health despite previously being scored as intermediate. The improved score was due to the persistent climbing Galaxias population, despite the presence of invasive trout. Whilst trout did not appear to be dominating the fish community at this site, there was a high degree of habitat partitioning between the two species. The other site, Yankalilla River Bridge, was consistently scored in the good category due to the diverse population of native species and absence of invasive species.

4.1.11. Parananacooka River

The site at Parananacooka River was an opportunistic site chosen to inform a works program aimed at clearing aquatic vegetation downstream of the historic bridge in Second Valley. The health of the

35 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 site was determined to be good due to the large population of diadromous common Galaxias and congollis (P. urvillii), and the absence of invasive species. The population was restricted to a small pool underneath the bridge, although other habitats may be available further upstream. Works on the bridge and surrounding area may impact the fish population, but as long as the integrity of the pool is maintained and fish passage is not blocked, there remains the opportunity for diadromous species to recolonise from nearby streams. The site is, however, another demonstration of the value of providing healthy coastal stream habitat and flow regime. The opportunistic nature of monitoring this site also highlights the benefit of maintaining a monitoring program with an element of discretionary site selection.

4.1.12. Yattagolinga River

The Croser site in Yattogolinga River was another site selected to expand knowledge of fish distribution in the WMLR. The site was in an unfenced reach of the river with little riparian vegetation and evidence of damage from stock grazing. Despite the poor riparian condition, the fish population was rated as good due to the abundant population of climbing and common Galaxias and the absence of invasive species.

4.1.13. Hindmarsh River

Besides the sites included in the estuary health part of this project, which are discussed below, only one other site was sampled in the Hindmarsh catchment. This site is the VWASP site at Hindmarsh Gauge, which consistently scored in the good category for fish population health due to the diverse fish population and absence of invasive species. It also had a rare South Australian record of dwarf flathead gudgeon in the Hindmarsh.

4.1.14. Inman River

As with the previous site, lower estuary sites within the Inman catchment are included in the estuary health part of this project and discussed below. Three other sites were sampled in the Inman catchment: Back Valley Gauge, the VWASP site at Glacier Rock and Inman Divine Gauge. Early in this project, sampling at Back Valley Gauge confirmed the presence of the rare southern pygmy perch as well as climbing Galaxias and carp gudgeon. This sample resulted in a fish population health rating of good on both BCG and FHI scales. However, since this survey the landowner at the site has refused to allow further surveys on that property. In future, alternative sites in Back Valley might need to be sought to confirm the ongoing presence of pygmy perch in the catchment. Occasional engagement with the landholder should be considered in case access is granted again.

36 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

The two other sites in the Inman Catchment both consistently scored in the intermediate health range (Glacier Rock dipped into poor condition range once). These poorer scores were a result of low native fish diversity and abundance combined with high abundance and diversity of invasive species, which possibly reflect the degraded nature of habitat and flow in the Inman catchment.

4.1.15. Deep Creek

Most sites in Deep Creek were historically listed as poor or intermediate condition (Table 6) (McNeil et al. 2011, Schmarr et al. 2014) due to the dominance of invasive brown trout. This condition score was also reflected when first sampled at Rangers Pump site in 2015 for this study. However, when sampled at the end of 2017, not only had the brown trout dropped out at Deep Creek Crossing, but climbing Galaxias had returned to the site. This provides potential evidence that efforts to restrict the stocking of trout should have benefits to the native fish community. While not included in the fish population health score, this catchment was one of the four where invasive marron (Cherax cainii) were present, but native yabbies (Cherax destructor) were not present. Marron are known to displace native crustaceans including yabbies (Cherax destructor), which were not present at this site during any sampling event.

4.1.16. Bollaparudda

An additional three sites in the Bollaparudda catchment were monitored in spring 2016 and autumn 2017 as part of a program to investigate the ecosystem values on Balquhidder Station. These surveys provided novel data for fish populations in this catchment. The health of fish populations in the Bollaparudda were rated as good due to the presence of abundant native fish populations and the absence of invasive species.

4.1.17. Callawonga

The VWASP site monitored in Callawonga Creek contrasts the neighbouring Bollaparudda Creek in that the fish population health at the site was rated as intermediate (BCG) or poor (FHI) during this study and poor for both BCG and FHI scores in the previous 2013 study (Table 6). The site only had brown trout present during this study. Populations of climbing Galaxias and common Galaxias have been sampled at other sites upstream and downstream of the current Callawonga Creek site (Schmarr et al. 2014), which further underlines the impact that brown trout may be having at this site.

37 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

4.1.18. Tunkalilla

The site in Tunkalilla Creek has been sampled previously in 2011, although not scored for fish population health at the time. In the current project, the site was sampled once and scored in the good range for both the FHI and BCG due to the presence of climbing and common Galaxias and the absence of invasive fish species. While not included in the fish population health score, this catchment was one of the four where invasive marron were present, although yabbies were still present at this site.

4.1.19. Boat Harbour Creek

The VWASP site at Boat Harbour Creek was sampled repeatedly during this study. The fish population health score at the site was always in the good range for the BCG scale and only recorded intermediate once for the FHI due to the presence of one brown trout in spring 2015. The presence of trout on that occasion coincided with the lowest abundance of climbing Galaxias at the site. While not included in the fish population health score, this catchment was one of the four where invasive marron were present. Native yabbies were still present in very low abundance.

4.1.20. Rarkang Creek

According to museum records, the Rarkang Dam site was the first recorded fish survey from Rarkang Creek (Blowhole Creek). The site scored well on both fish population health measures due to the high abundance of climbing and common Galaxias and the absence of invasive species.

4.2. Myponga, Bungala, Inman, and Hindmarsh Estuaries

4.2.1. Myponga River

Sampling in the lower reaches of the Myponga River was not originally intended to be part of the estuary health component of this project. However, after monitoring was conducted at the site in consecutive studies through 2015 and early 2016, it was decided to continue monitoring the site in spring 2016 and autumn 2017 in response to a large scouring event that occurred in the catchment as a result of an emergency release of water from Myponga reservoir. Effects of this large flow event on riparian vegetation were documented by Emma Matthews in a separate report (Attachment D). Monitoring at the site after the flow event enabled observation of the recovery of the fish population in the immediate weeks following the flooding and also several months later. The repeated monitoring at the two sites in the Myponga also provided a reach to compare with the Bungala, Inman and Hindmarsh estuaries which are located in more developed catchments.

38 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Over the course of the study, the fish population health in both the Myponga Estuary and Myponga Pumphouse sites remained in the good category. Whilst there are no previous site health scores for these two sites, the fish populations are identical to previous studies of fish populations in the Myponga River (McNeil et al. 2009). Despite a small alteration to the estuarine fish community in the estuary site after the flood in 2016, the fish community recovered within a few months to the previous composition. While the composition was similar, there was a change in the dominance of several species at the site (Table 3). This may be a result of a change in salinity at the site due to the clearing of sand at the mouth of the estuary, allowing greater tidal inflow from the sea.

The fish community at the Myponga pumphouse site increased in diversity (with the addition of climbing galaxias) and abundance in response to the flood. One redfin perch was detected in December 2016, but not the following autumn survey. The changes in fish population at this site may be due to the flood increasing habitat availability by removing a large volume of sediment and emergent vegetation (mostly Typha domingensis) from the pools (Figure 10).

The fish community response to the large flood event in spring 2016 appears to highlight the resilience of most species to such disturbances. It may also be a demonstration of the importance of occasional large-scale flooding in resetting habitats after years of sediment and vegetation accumulation.

Figure 10. Comparison of riparian and pool habitat at Myponga Pumphouse before (left) and after (right) the spring 2016 flood.

4.2.2. Bungala River

Monitoring at the three sites in the Bungala River consistently provided scores in the good range during this study. The estuary site consistently had the highest abundance of estuarine specialist

39 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 and diadromous fish species in this study, including high numbers of recruits of the recreationally and commercially important black bream (Table 3). Twelve species were captured in the estuary site, two more than reported for previous studied at this site (Gillanders et al. 2008, AECOM 2010). The estuary site was also one of two locations – the other being the Myponga estuary - where endangered estuary perch (Macquaria colonorum) was observed on Fleurieu Peninsula for the first time. This record expands the range of estuary perch from around the Murray Mouth to Gulf St Vincent. It is uncertain whether estuary perch expanded their range by migrating around Fleurieu Peninsula to sites in Gulf St Vincent or through illegal stocking activities conducted by a member of the public. That the species was found in two nearby estuaries and at no other estuaries between their previous range and the Myponga and Bungala estuaries, may indicate that stocking is the more likely cause of the range expansion. Continued monitoring will determine whether the species persists in estuaries in Gulf St Vincent.

The freshwater sites regularly had 2 or 3 native species and up to 6 species during the course of this study. The study also provided the first record of dwarf flathead gudgeon in Bungala catchment (at South Rd). Only one invasive fish (Gambusia) was recorded for the entire three years, which was likely washed down from source populations upstream during the spring 2016 rain event.

Despite concern over water quality and flow regime in the Bungala catchment (AECOM 2010), the fish population at the three sites monitored during this survey appear to be functioning adequately. Objectives 3 (Improve Ecological Health and Biodiversity), 6 (Maintain Environmental Flows) and 7 (Maintain Marine Connectivity) in the Bungala Estuary Action Plan (AECOM 2010) are relevant to this study. Whilst it is unclear which of the specific actions under these objectives have been implemented in the Bungala, it appears over this short time period that the goals of these objectives have been met. Ongoing monitoring will determine whether the objectives are met over a longer timeframe.

4.2.3. Inman River

The estuary site in Inman River had high abundance and diversity of estuarine specialist and diadromous fish species in this study, including occasional high numbers of recruits of the recreationally and commercially important black bream. Twenty species were captured in the estuary site over the course of this study, 15 more than reported for previous studies at this site (Gillanders et al. 2008). The native species diversity increased during the three year period to a peak of 12 species in autumn 2017 and the invasive species diversity and abundance decreased to zero over the same timeframe. Additionally, an estuarine crab (Amarinus laevis) specimen from

40 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 the Inman Estuary was lodged at the museum, which was the first state record in the Atlas of Living Australia.

In contrast to the estuary site (and the Myponga, Bungala and Hindmarsh freshwater sites), the two freshwater sites in the Inman rated mostly in the intermediate range on both the FHI and BCG scales. The reasons for these scores were the low abundance and diversity of native species (especially sensitive rare or common species), dominance of tolerant native species and dominance of invasive species. Another observation of note was the first record of Marron in the Inman River Spring 2016 at Swain’s Crossing.

The health of sites in the Inman provided conflicting results with an estuary site showing high abundance and diversity of both estuarine specialist species and diadromous species, while the freshwater site appeared to lack healthy populations of diadromous species. One of the most important factors in maintaining diadromous fish populations is providing passage between marine and freshwater habitats. There is a large concrete weir between the estuary and freshwater sites in this system (Figure 11), which is likely to be impassable for many fish species under most flow conditions (with the exception of very high flows).

As with the Bungala, there has been concern over water quality and flow regime in the Inman catchment (SKM 2010a, Steed 2016). In the Inman Estuary Action Plan, action number 17 (Enhance In-stream Ecology) notes that the Inman River estuary previously had poor in-stream ecology, but that water quality and fish diversity figures suggested that ecological function had improved. The results from the present study suggest that in the estuarine part of the Inman River, fish diversity has continued to improve. However, sites upstream of the estuary have not shown similar improvement in ecological function. The poor diversity and abundance of diadromous species in this reach points towards in-stream barriers being an impediment to improving ecological function. Whilst ongoing monitoring will determine whether the lack of fish passage continues to impact fish in this reach, it is also recommended that a study should be undertaken to investigate the feasibility of altering or removing the weir in question.

41 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Figure 11. Weir on Inman River blocking passage between estuary and freshwater sites.

4.2.4. Hindmarsh River

Like the Bungala River, monitoring at the three sites in the Hindmarsh River consistently provided scores in the good range during this study. Fish abundance in the estuary was not as high as the Bungala and Inman estuaries, but the diversity was similar to the Inman and showed a similar improvement in diversity throughout the study. Fourteen species were captured in the estuary site over the course of this study, three more than reported for previous studies at this site (Gillanders et al. 2008). Unlike this previous study, a seasonal trend in the diversity and abundance of estuarine species was not detected, indicating that these species utilised the habitat year-round and were not subject to major water quality impacts.

In contrast to the Inman River, the sites immediately upstream from the Hindmarsh estuary site had healthy populations of native fish, were almost free from invasive species, and had a number of estuarine species present. The salinity at Lamont Road was almost identical to the estuary site,

42 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 while further upstream at Cootamundra Reserve it was much lower (Appendix C). At the VWASP site further upstream again, the water was not saline and only diadromous or obligate freshwater species were present. This salinity gradient between estuary and freshwater sites with the associated changes in fish community is expected in this type of river (Martino and Able 2003).

The Hindmarsh River has had similar water quality and flow regime issues as the Inman catchment (SKM 2010b, Steed 2016). In the Hindmarsh Estuary Action Plan, action number 19 (Enhance In- stream Ecology) notes that the Hindmarsh River estuary previously had fish kills as a result of seasonal declines in water quality, but that water quality and fish diversity figures suggested that ecological function had improved. The results from the present study suggest that the ecological function of the river continues to be maintained. Actions that have been implemented to improve water quality and flow in this catchment should be maintained, and regular monitoring should confirm the ongoing improvement in ecological function.

4.3. Summary of WMLR fish condition and estuary studies

4.3.1. Coastal vs inland habitat

Coastal freshwater sites were almost all found to be in the healthy category. This is due to a range of factors including: protection and rehabilitation of riparian habitat in coastal areas, proximity to the marine environment and maintenance of fish passage between marine and freshwater environments. Many of these sites were managed as public reserves or conservation parks which may have prevented major alteration of habitat and, in instances such as the Field River, the habitat has been restored to regain ecological value. Many sites achieved a healthy rating due to the presence of abundant and diverse populations of diadromous fish species. Given suitable habitat, these species are able to recolonise catchments via the marine phase of their life-cycle.

Conversely, sites further inland and further north frequently had intermediate or poor health ratings. This is partly due to the difficulty for diadromous species to recolonise further inland especially in drier catchments, and partly due to the isolated and fragmented nature of obligate freshwater fish populations such as mountain Galaxias and flathead gudgeons. These species are not diadromous so can only recolonise from locations within a catchment. If recolonisation pathways don’t exist due to instream barriers or lack of flow, or if the species is locally extinct, then there is little prospect for recolonisation.

43 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Catchment management should focus on continuing work that has protected and/or restored habitat in lowland reaches. As far as possible, this work should be extended to reaches further upstream with the consent of and cooperation with stakeholders. In conjunction with these works, monitoring efforts should continue to be aligned with on-ground habitat and flow restoration.

4.3.2. Barriers to fish passage

Barriers to fish passage – natural or human-made – appear to have shaped the composition of fish communities. Some barriers acted as partial barriers, limiting the migration and dispersal of fish species to specific – usually very high – flow bands (i.e. when flow overtopped barriers creating alternative flow paths). In the absence of high flows, due to drought or over-extraction of water from the catchment or if the timing of flows is incompatible with fish life-history stages, these barriers may disrupt fish population structure in the long-term (e.g. Inman River). Other barriers, by their physical size alone, appear to permanently limit the movement of species within catchments (e.g. North Para Flood Retention Dam and Yaldara Weir). In addition to known barriers to fish passage, observation of discontinuous fish populations where source populations of native fish exist nearby in the catchment, indicated the presence of potential barriers to fish passage (e.g. upper Torrens). An inventory of in-stream barriers in the MLR should be collated to provide a better understanding of the constraints on native fish movement and distribution. On-ground surveys may be needed to verify presence of in-stream barriers.

4.3.3. Trout

Presence of trout in this study and many others (McDowall 2006) was frequently associated with the absence of native species at a site or reach scale, and in turn was associated with poor fish community health. Even where trout were present in the same site as native species, the two were partitioned into different habitats or pools. Whilst it would be difficult to remove trout from existing streams, management of trout stocking should continue to be limited to the six streams that it is permitted in, and additional resources should be allocated to educating the public about the harm in illegally stocking invasive species. Up to this date, most monitoring programs have focused on monitoring catchments and sites where trout have not been legally stocked. The bias in this monitoring design may have overlooked interactions between trout and native species. Targeted monitoring in trout stocked catchments may be useful in further elucidating the role that trout play in shaping native fish communities.

44 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

4.3.4. Species resilience

The flooding following the large rainfall event in spring 2016 cleared out sediment and vegetation from many sites and, in some instances, was associated with a change in the composition or abundance of fish communities. However, native and invasive species generally recovered immediately after these events. There were some instances of invasive species dispersing or migrating into new sites and there were also examples of invasive species – especially Gambusia – being eliminated or severely reduced at sites. This observation demonstrates the resilience of native species to flood disturbance and the important role that such events play in maintaining healthy fish communities. Continued monitoring of those sites that experienced flood disturbance will confirm the long-term recovery of native fish populations as well as the impact on invasive species.

4.3.5. Mountain Galaxias distribution

The distribution of freshwater obligate species such as mountain Galaxias was frequently limited to one or two sites per catchment and only five catchments in this study. The fragmented and limited distribution of mountain Galaxias led to it being designated as “vulnerable” in the 2009 freshwater fish action plan (Hammer et al. 2009), and it should remain in this category. Further investigations should be conducted to reveal the full spatial range of mountain Galaxias within these catchments. Monitoring of mountain Galaxias populations in extremely isolated and fragmented habitats should continue to be conducted on an occasional basis to avoid inadvertent harm.

4.3.6. Improved estuary health

The health of the fish populations in the four estuaries monitored over the last three years showed equal or higher diversity of estuarine specialist taxa and diadromous freshwater taxa in comparison to previous reports. This could indicate that previous water quality issues in these reaches may have improved, or that current conditions are within the tolerances suitable to support a number of native species. The recent study by Steed (2016) for example, indicated among water quality indicators, that nitrogen, phosphorus, and turbidity (as sediment and organic matter) were higher and dissolved oxygen lower in the Bungala, Hindmarsh and Inman River estuaries than recommended ANZECC water quality thresholds for estuaries sampled at similar sites to the current study between 2014-2016. Water quality measurements observed in the current study also support these results suggesting the water quality condition values may be within the tolerance range/s of the species identified here. If actions have been implemented to improve water quality and flow in these catchments, these should be maintained, and regular monitoring should confirm the ongoing

45 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17 improvement in ecological function. Further monitoring around the estuarine section of Inman River would help identify whether there is a similar gradient of species along this reach as in Hindmarsh estuary. Targeted sampling around the weir above the estuary in the Inman will further inform its impact on fish communities further upstream.

4.3.7. Consolidated dataset

In the process of collating and analysing data collected in the current project, and reviewing previous projects carried out by SARDI and museum/historical records, it is apparent that the extent of knowledge regarding fish distribution, life history, habitat associations, water quality tolerances and environmental water requirements in the Mount Lofty Ranges has expanded greatly since the Biological Review of the Freshwater Fishes of the Mount Lofty Ranges was published in 2007 (McNeil and Hammer 2007). The review noted that there was “a complete lack of knowledge regarding local distribution, status or biology” and has since been used as a guide for subsequent monitoring programs. Whilst monitoring data are still being collected, given that most of the previous data was collected well over 15 years ago, it would be timely to produce an update or second edition of the biological review.

4.4. Recommendations

Based on the knowledge gained from this monitoring project and previous MLR monitoring, we provide the following recommendations:

 Continue habitat and flow protection and restoration work. Align monitoring program with on-ground works to provide adaptive management.  Desk-top and field survey of in-stream barriers to fish passage in priority catchments.  Investigate removal of unnecessary barriers to fish passage.  Targeted monitoring program in trout-stocked rivers to investigate interactions between trout and native species.  Providing scientific advice to fisheries and natural resource managers to assist in public education and risk management regarding the impact of stocking non-native species in MLR catchments.  Continuing monitoring at appropriate locations and temporal scale to verify long-term fish community recovery from flood disturbance and confirm persistence of isolated populations.

46 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

 Scale back monitoring in Myponga, Bungala and Hindmarsh estuaries, and focus on Inman estuary and fish passage between estuary and freshwater reach.  Consolidate all SARDI fish distribution and biological data with historical data to provide an updated Biological Review of the Freshwater Fishes of the Mount Lofty Ranges.

47 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

5. REFERENCES Adelaide and Mount Lofty Ranges NRM Board (2013). Adelaide and Mount Lofty Ranges Natural Resources Management Plan Volume 1 — Part 2 Strategic Plan 2014-15 to 2023-24. Adelaide and Mount Lofty Ranges Natural Resources Management Board, Adelaide.

AECOM Australia Pty Ltd (2010). Bungala Estuary Action Plan. A report prepared by AECOM Australia Pty Ltd for the Adelaide and Mount Lofty Ranges Natural Resources Management Board. Adelaide.

Davies, P.E., Harris, J.H., Hillman, T.J. and Walker, K.F. (2008). SRA Report 1: A Report on the Ecological Health of Rivers in the Murray–Darling Basin, 2004–2007. Prepared by the Independent Sustainable Rivers Audit Group for the Murray–Darling Basin Ministerial Council.

Davies, S.P. and Jackson, S.K. (2006). "The biological condition gradient: A descriptive model for interpreting change in aquatic ecosystems." Ecological Applications 16(4): 1251-1266.

Gillanders, B.M., Elsdon, T.S., and Hammer, M. (2008). Estuaries of Gulf St Vincent, Investigator Strait, and Backstairs Passage. Chapter 14. In: Natural History of Gulf St Vincent., Eds: Shepherd, S., Bryars, B., Kirkgaard, I.R., Harbison, P., Jennings, J.T. Royal Society of South Australia Inc., Adelaide.

Hammer, M.P. (2005). The Adelaide Hills Fish Inventory: Distribution and Conservation of Freshwater Fishes of the Torrens and Patawolonga Catchments, South Australia. Adelaide: Native Fish Australian (SA) Inc.

Hammer, M.P., Wedderburn, S., and van Weenen, J. (2009). Action Plan for South Australian Freshwater Fishes. Native Fish Australia (SA). Adelaide.

Karr, J.R. (1981). Assessment of biotic integrity using fish communities. Fisheries 6 (6), 21-27.

Kiri-ganai Research (2009). Lake Eyre Basin Rivers Assessment Implementation Plan and business governance model 2010-2018. http://www.lakeeyrebasin.gov.au/sitecollectionimages/71d27602-9826-4d4f-9004- fbc30cde225b/files/lebra-implementation-plan-2010-18.pdf

Martino, E.J. and Able, K.W. (2003). Fish assemblages across the marine to low salinity transition zone of a temperate estuary. Estuarine, Coastal and Shelf Science 56, 969–987.

48 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Mathwin, R., McNeil, D.G. and Schmarr, D.W. (2014). A biological condition gradient model approach for fish-based ecological condition monitoring in the Western Mount Lofty Ranges, South Australia. South Australian Research and Development Institute (Aquatic Sciences), Adeladie. SARDI Research Report Series No. 758. SARDI Publication Number F2013/000020-1. 35pp.

Maxwell, S.E., Green, D.G., Nicol, J., Schmarr, D., Peeters, L., Holland, K. and Overton, I.C. (2015). Water Allocation Planning: Environmental Water Requirements. GWAP Project: Task 4. Goyder Institute for Water Research Technical Report Series No. 15/53, Adelaide, South Australia.

McDowall, R. (2006). "Crying wolf, crying foul, or crying shame: Alien salmonids and a biodiversity crisis in the southern cool-temperate galaxioid fishes?" Reviews in Fish Biology and Fisheries 16(3): 233-422.

McNeil, D. and Cockayne, B. J. (2011). Proposed Guidelines for Fish Monitoring Lake Eyre Basin Rivers Assessment (LEBRA). A report to the Department of Sustainability, Environment, Water, Population and Communities (SEWPaC), Canberra, ACT.

McNeil, D.G., Fredberg, J. and Wilson, P.J. (2009). Coastal Fishes and Flows in the Onkaparinga and Myponga Rivers. Report to the Adelaide and Mount Lofty Ranges Natural Resource Management Board. South Australian Research and Development Institute (Aquatic Sciences), Adeladie. SARDI Research Report Series No. 400. SARDI Publication No. F2009/000410-1. 76pp.

McNeil, D.G. and Hammer, M (2007). Biological review of the freshwater fishes of the Mount Lofty Ranges. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Research Report Series No. 188. SARDI Publication number: F2006/000335. 104pp.

McNeil, D.G, Schmarr, D.W and Mathwin, R (2011). Condition of Freshwater Fish Communities in the Adelaide and Mount Lofty Ranges Management Region. Report to the Adelaide and Mount Lofty Ranges Natural Resources Management Board. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Research Report Series No. 590. SARDI Publication No. F2011/000502-1. 65pp.

McNeil, D., Schmarr, D., Wilson, P. and Reid, D. (2011). Fish Community and Flow Ecology in the Western Mount Lofty Ranges Environmental Water Provisions Trial Reaches. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Research Report Series No. 581. SARDI Publication No. F2011/000313-1. 238pp.

49 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Milton D.A., Arthington A.H., (1983). Reproductive biology of Gambusia affinis holbrooki Baird and Girard, Xiphophorus helleri (Gunther) and X. maculatus (Heckel) (Pisces; Poeciliidae) in Queensland, Australia. Journal of Fish Biology, 23(1):23-41.

Schmarr, D.W., Mathwin, R. and McNeil, D.G. (2011). Mapping threats to aquatic biota movement and recovery with the provision of environmental water in selected reaches of the South Para, Torrens and Onkaparinga Rivers - RESTRICTED. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2011/000418-1. SARDI Research Report Series No. 570. 8pp.

Schmarr, D.W., Mathwin, R. and Cheshire D.L.M. (2014). Western Mount Lofty Ranges Fish Condition Report 2012-13. Incorporating the Barossa Valley Prescribed Water Resource Area Fish Community Study, the Verification of Water Allocation Science Project (VWASP) and the Western Mount Lofty Ranges Fish Community Monitoring. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2014/000113-1. SARDI Research Report Series No. 780. 84pp.

Sinclair Knight Merz (2010). Inman River Estuary Action Plan. A report prepared by Sinclair Knight Merz for Adelaide and Mount Lofty Ranges Natural Resources Management Board. 113pp.

Sinclair Knight Merz (2010). Hindmarsh River Estuary Action Plan. A report prepared by Sinclair Knight Merz for Adelaide and Mount Lofty Ranges Natural Resources Management Board. 123pp.

Steed, S. (2016). Bungala, Inman and Hindmarsh Estuary Monitoring. Estuaries Water Quality Condition 2016. Report to the Adelaide and Mount Lofty Ranges Natural Resources Management Board. Care of Our Environment. 2016. 139pp.

50 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

6. APPENDIX A Designation of Attribute groups to known fish species used in the formulation of the Biological Condition Gradient (BCG) for WMLR sites. Exotic and translocated species are listed in red. Where natural and introduced populations exist, both red and black listings are shown. Note estuarine specialist species are not considered.

Attribute Scientific Name Common Name Code Notes (I – VI) Ambassis agassizii Glassfish Amb aga 1 Anecdotally documented Anguilla australis Shortfinned eel Ang aus 1 Diadromous, Rare, historically rare, possible range extension, life history sensitivities Gadopsis marmoratus River blackfish Gad mar 1 Anecdotally documented Galaxias brevipinnis Climbing Galaxias Gal bre 2 Diadromous, Sensitive and rare, cooler reaches, life history sensitivities Galaxias maculatus Common Galaxias Gal mac 3 Diadromous, Tolerant to salinity, common in downstream reaches, rare in upland reaches Galaxias olidus Mountain Galaxias Gal oli 3 Sensitive to WQ and predation, common Geotria australis Pouched lamprey Geo aus 1 Diadromous, Historically common, now rare Hypseleotris spp. Carp gudgeon species Hyp spp/ 6 (*5) Murray Darling translocation, biological Hyp spp pollution, *considered endemic in the Finniss River Melanotaenia fluviatilis Rainbowfish Mel flu/ 6 (*5) Murray Darling translocation, biological Mel flu pollution, *considered endemic in the Hindmarsh River Mogurnda adspersa Purple-spotted gudgeon Mog ads 1 Historically documented, ?regionally extinct Mordacia mordax Short-headed lamprey Mor mor 1 Diadromous, Historically common, now rare Philypnodon grandiceps Flathead gudgeon Phi gra 5 Tolerant to wide range of impacts including low flow, WQ Philypnodon maculatus Dwarf flathead gudgeon Phi mac 5 In the Onkaparinga and Torrens it is unclear if this species is endemic or a River Murray translocation Pseudaphritis urvillii Congolli Pse urv 2 Diadromous, more commonly found in coastal reaches Pseudogobius olorum Western bluespot goby Pse olo 2 Amphidromous, sensitive to disturbance, sensitivities not well documented, reliance on rocky interstices, rare Tandanus tandanus Eel-tailed catfish Tan tan 6 Murray Darling translocation, biological pollution, but protected Perca fluviatilis Redfin perch Per flu 6 Intentionally introduced taxa Gambusia holbrooki Mosquitofish Gam hol 6 Intentionally introduced taxa Phalloceros Speckled livebearer Pha cau 6 Intentionally introduced taxa caudimaculatus Cyprinus carpio European carp Cyp car 6 Intentionally introduced taxa Carassis auratus Goldfish Car aur 6 Intentionally introduced taxa

51 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Attribute Scientific Name Common Name Code Notes (I – VI) Oncorhynchus mykiss Rainbow trout Onc myk 6 Intentionally introduced taxa Salmo trutta Brown trout Sal tru 6 Intentionally introduced taxa Tinca tinca Tench Tin tin 6 Intentionally introduced taxa, not necessarily harmful to native fishes

52 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Descriptions of the 10 Attributes that distinguish the six Tiers of the biological condition gradient (BCG) modified for fish community assessment.

BCG Tier Description

Attribute I Historically documented, sensitive, long-lived, or regionally endemic taxa (Species with unpredictable occurrence are only scored if historical records exist for locality).

Tier 1 Present

Tier 2 Absent but known at the site (within the last 5 sampling events)

Tier 3 Absent but recolonisation possible

Tier 4 Absent due to global extinction

Tier 5 Absent due to regional extirpation

Tier 6 Absent

Attribute II Sensitive-Rare Taxa (Species with unpredictable occurrence are only scored if a historical record exist for locality).

Tier 1 As predicted for natural occurrence, with at most minor changes from natural densities.

Tier 2 Virtually all are maintained with some changes in densities

Tier 3 Some loss, with replacement by functionally equivalent sensitive-common taxa

Tier 4 Present but markedly diminished

Tier 5 Absent from catchment but likely to occur due to the presence of suitable habitats (supported by recent observation, anecdotal or otherwise from the catchment)

Tier 6 Absent

Attribute III Sensitive Common Taxa

Tier 1 As predicted for natural occurrence, with at most minor changes from natural densities

Tier 2 Present with minor changes in expected density

Tier 3 Common in the catchment but much lower density than expected

Tier 4 Present with reproducing populations maintained; some replacement by functionally equivalent taxa

53 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Tier Description

Tier 5 Markedly diminished

Tier 6 Absent

Attribute IV Taxa of intermediate tolerance

(No WMLR Fish species were assigned to Attribute 4, wording preserved from Davies and Jackson, 2006)

Tier 1 As predicted for natural occurrence, with at most minor changes from natural densities

Tier 2 As naturally present with slight increases in abundance

Tier 3 Often evident increases in abundance

Tier 4 Common and often abundant; common abundance may be greater than sensitive-common taxa

Tier 5 Often exhibit excessive dominance

Tier 6 May occur in extremely high or extremely low densities; richness of all taxa is low

Attribute V Tolerant taxa

Tier 1 As predicted for natural occurrence, with at most minor changes from natural densities

Tier 2 Present with a slight increase or decrease in expected abundance.

Tier 3 May be common in catchment, but locally are either absent or do not exhibit significant dominance

Tier 4 May be common in catchment, but do not exhibit significant dominance

Tier 5 Common in catchment and exhibit significant dominance

Tier 6 Comprise the majority of the assemblage with very high densities.

Attribute VI Nonnative or intentionally introduced taxa

Tier 1 Nonnative taxa not present

Tier 2 Nonnative taxa may be present, but occurrence has a non-detrimental effect on native taxa

Tier 3 Sensitive introduced taxa dominate assemblage

Tier 4 Diversity of non-native taxa dominate assemblage

54 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Tier Description

Tier 5 Fish assemblage is dominated by tolerant nonnative fish(es)

Tier 6 Nonnative fishes are the only representatives of the fish taxa

Attribute VII Organism and population condition (disease, congenital abnormalities, reproduction)

Tier 1 Anomalies not observed, multiple age classes apparent

Tier 2 Any anomalies consistent with naturally occurring incidences and characteristics, multiple age classes apparent

Tier 3 Anomalies are infrequent

Tier 4 Biomass reduced or anomalies common

Tier 5 Biomass reduced and anomalies widespread

Tier 6 Long-lived taxa may be absent; biomass reduced; anomalies widespread and serious; minimal reproduction except for extremely tolerant groups.

Attribute VIII Ecosystem functions

This Tier is not utilised in the current approach. It has been retained to allow future integration of limnological data.

Tier 1 All are maintained within a range of natural variability

Tier 2 All are maintained within a range of natural variability

Tier 3 Virtually all are maintained through functionally redundant system Attributes; minimal increase in export except in high storm flows

Tier 4 Virtually all are maintained through functionally redundant system Attributes, although there is evidence of loss of efficiency (eg increased export or increased import)

Tier 5 Apparent loss of some ecosystem functions manifested as increased export or increased import of some resources and, changes in energy exchange rates (eg P/R, decomposition)

Tier 6 Most functions show extensive and persistent interruption

Attribute IX Spatial and temporal extent of detrimental effects

Tier 1 Not applicable. Natural disturbance regime is maintained

55 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

BCG Tier Description

Tier 2 Limited to small pockets and short durations

Tier 3 Limited to reach scale and/or limited to within a season

Tier 4 Mild detrimental effects may be detectable beyond the reach scale and may include more than one season

Tier 5 Detrimental effects extend far beyond reach scale leaving only a few islands of adequate conditions; effect extends across multiple seasons.

Tier 6 Detrimental effects may eliminate all refugia and colonisation sources within the catchment and affect multiple seasons.

Attribute X Ecosystem Connectivity

Tier 1 No alteration to natural connectivity patterns in space or time

Tier 2 System is fully connected at least annually (longitudinally, laterally and vertically)

Tier 3 Slight loss of connectivity but there are adequate local colonisation sources

Tier 4 Some loss of connectivity but colonisation sources exist within the catchment

Tier 5 Significant loss of ecosystem connectivity is evident. Recolonisation sources do not exist for some species

Tier 6 Complete loss of connectivity lowers reproductive success of most groups. Frequent failures in reproduction and recruitment.

56 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

7. APPENDIX B

Sampling locations for sites monitored in the WMLR fish condition study between March 2015 and December 2017.

Catchment Watercourse Site_Name UTM Inman Inman River Armstrong Rd Bridge 54 H 283146 6063206 Inman BackValley Trib BackValley Gauge 54 H 273970.01 6064352.34 Boat Harbour Boat Harbour Boat Harbour Gauge 54 H 254144.22 6056106.81 Bollaparudda Creek Bollaparudda Creek East Bollaparudda campsite 54 H 264215 6055370 Bollaparudda Creek Bollaparudda Creek East Bollaparudda east DS 54 H 264709 6053330 Bollaparudda Creek Bollaparudda Creek East Bollaparudda east gorge 54 H 264949 6053827 Bollaparudda Creek Bollaparudda West Bollaparudda West Dam 54 H 264543 6053212 Bungala Bungala River Bungala Caravan Park Bridge 54 H 255785 6073996 Bungala Bungala River Bungala South Rd 54 H 256512 6073933 Callawonga Callawonga Callawonga Gauge 54 H 262147 6055176 Torrens First Creek Chinaman's Hut 54 H 345823.84 138411.82 Hindmarsh Hindmarsh River Cootamundra Reserve 54 H 284630 6065460 Patawalonga Brownhill Creek Craigburn Dam 54 H 281583.00 6119478.00 Yattagolinga Yattagolinga River Croser 54 H 246129 6063091 Deep Creek Deep Creek Deep Creek Crossing 54 H 249646.69 6054166.07 Patawalonga Brownhill Creek DS Brownhill Caravan Park 54 H 283296 6126115 Patawalonga Sturt Creek DS Sturt Gorge Retention Dam 54 H 280217 6119735 Tunkalilla Tunkalilla Creek Eric Bonython conservation park 54 H 258268 6056924 Torrens Lower Torrens River Fox Creek 54 H 302416 6138471 Inman Inman River Glacier Rock 54 H 274479 6069038 Gawler North Para River Gomersal Rd Bridge 54 H 311446.67 6176662.60 Bungala Bungala River Hay Flat Rd 54 H 256971 6073610 Hindmarsh Hindmarsh River Hindmarsh Estuary 54 H 284829.96 6064037.01 Hindmarsh Hindmarsh River Hindmarsh Gauge 54 H 284397.00 6066670.00 Yankalilla Yankalilla River Ingalalla Falls 54 H 259093 6064811 Inman Inman River Inman Divine Gauge 54 H 280615 6063902 Inman Inman River Inman estuary 54 H 283376.00 6062134.00 Gawler Jacobs Creek Jacobs Creek Old Gauge 54 H 312982.52 6172765.99 Hindmarsh Hindmarsh River Lamont Rd 54 H 284422.00 6064459.00 Onkaparinga Lenswood Ck Lenswood Gauge 54 H 301139.08 6131972.87 Gawler South Para River Mt Crawford 54 H 313033 6158509 Onkaparinga Onkaparinga River MtBoldGate4 54 H 286359.00 6112613.00 Onkaparinga Aldgate Creek Mylor Bridge 54 H 295761 6120278 Myponga Myponga River Myponga estuary 54 H 262875 6082021 Myponga Myponga River Myponga pumphouse 54 H 262586 6081770 Myponga Myponga River Pages Flat 54 H 271331 6082284 Field River Field River Railway Tunnel 54 H 271807.00 6114210.00

57 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Deep Creek Deep Creek Rangers Pump 54 H 250432.20 6056211.04 Rarkang creek Rarkang Creek Rarkang Dam 54 H 242827.00 6051204.00 Carrickalinga Carrickalinga Riverview Drive 54 H 257132.55 6075782.46 Carrickalinga Carrickalinga Rose Cottage 54 H 259671.11 6074596.75 Willunga Creek Willunga Creek Ross Roses 54 H 277443 6093250 Onkaparinga Scott Creek Scott Creek Gauge 54 H 288009.97 6113417.63 Willunga Creek Willunga Creek St. Johns Rd 54 H 277842.10 6092860.58 Bungala Bungala River Stornoway 54 H 261975 6071430 Inman Inman River Swains Crossing Road 54 H 282000 6064240 Torrens Upper Torrens Talunga Park Bridge 54 H 321660 6150487 Gawler Tanunda Creek Tanunda Ck Gauge 54 H 315074.46 6175575.23 Patawalonga Brownhill Creek US Caravan Park 54 H 283322 6126002 US Sturt Gorge FloodRetention Patawalonga Sturt Creek Dam 54 H 280414 6119532 Gawler South Para River Victoria Creek 54 H 306706 6161154 Gawler South Para River Victoria Reserve 55 H 307970 6161152 Torrens First Creek Waterfall Gully 54 H 345814.38 1384053.64 Patawalonga Patawalonga West Beach Road Bridge 54 H 272953.64 6130215.36 Gawler North Para River Yaldara 54 H 305086 6172145 Yankalilla Yankalilla River Yankalilla Bridge River 54 H 254886.55 6071364.81

58 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

8. APPENDIX C

Water Quality values for sites monitored in the WMLR fish condition study between March 2015 and December 2017. Temperature, salinity, pH and turbidity values are for surface measurements. Dissolved Oxygen is the maximum and minimum values recorded in the profile.

Dissolved Dissolved Oxygen Oxygen Temperature Salinity Turbidity Min Max Catchment Watercourse Site Date (oC) (ppt) pH (NTU) (mg/L) (mg/L) Boat Harbour Boat Harbour Boat Harbour Gauge 25/04/2015 ------

22/10/2015 15.11 0.20 10.03 17.60 5.66 5.66 9/03/2016 18.51 0.20 9.72 25.20 0.46 0.46 1/12/2016 ------19/04/2017 14.86 0.20 9.38 22.90 6.53 7.96 15/12/2017 17.10 0.20 7.54 10.10 6.57 6.65 Bollaparudda Bollaparudda Bollaparudda Creek Creek East campsite 2/12/2016 18.58 0.65 8.36 42.80 2.31 3.59

13/04/2017 17.82 0.70 8.75 24.20 7.59 7.59 Bollaparudda east DS 2/12/2016 23.22 1.10 10.08 16.25 4.00 5.74 13/04/2017 18.16 1.20 9.00 44.10 6.58 7.77 Bollaparudda east gorge 2/12/2016 20.23 0.95 9.89 45.95 3.80 4.33 13/04/2017 15.36 1.00 9.67 33.70 5.43 5.68 Bollaparudda Bollaparudda West West Dam 2/12/2016 21.37 0.60 10.15 22.20 5.14 6.82 Bungala Caravan Bungala Bungala River Park Bridge 17/03/2015 19.89 21.53 8.21 27.23 6.29 7.39

13/10/2015 20.78 24.50 9.07 20.90 6.90 7.72 24/03/2016 ------18/10/2016 13.78 10.47 6.98 143.07 8.96 10.99 4/04/2017 22.16 28.72 8.09 19.30 4.54 5.14 8/12/2017 23.22 12.98 8.17 7.10 4.16 5.65 Bungala South Rd 17/03/2015 16.89 3.80 7.61 24.40 9.78 10.66 13/10/2015 14.98 3.50 8.44 33.63 2.07 4.88 24/03/2016 ------18/10/2016 14.34 0.52 7.52 233.40 6.48 10.00 4/04/2017 15.28 3.70 7.99 19.40 4.86 6.20 8/12/2017 14.63 3.85 8.05 5.00 3.53 4.31 Hay Flat Rd 17/03/2015 17.64 4.37 7.67 63.97 10.00 14.13 13/10/2015 14.79 3.60 8.41 24.10 3.76 3.91 24/03/2016 ------18/10/2016 14.22 0.50 7.10 214.67 4.22 4.33 4/04/2017 16.09 4.80 8.55 23.80 5.90 7.05 8/12/2017 14.59 4.30 7.79 10.50 3.09 4.40 Stornoway 11/04/2017 12.93 1.40 8.02 13.70 4.20 4.20

59 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Dissolved Dissolved Oxygen Oxygen Temperature Salinity Turbidity Min Max Catchment Watercourse Site Date (oC) (ppt) pH (NTU) (mg/L) (mg/L) Callawonga Callawonga Callawonga Gauge 24/04/2015 ------

22/10/2015 15.65 0.40 9.90 93.70 5.39 5.60 9/03/2016 18.85 0.60 9.64 16.50 2.54 4.45 1/12/2016 ------19/04/2017 15.90 0.40 9.74 19.30 5.54 7.81 7/12/2017 15.49 0.30 7.01 844.00 7.44 7.44 Riverview Carrickalinga Carrickalinga Drive 2/11/2016 14.74 1.00 11.85 29.43 2.77 3.91 Rose Cottage 2/11/2016 15.46 0.90 11.61 24.03 2.19 2.58 14/12/2017 20.29 1.55 9.03 13.20 5.62 5.91 Deep Creek Deep Creek Deep Creek Crossing 15/12/2017 18.52 0.80 9.43 43.40 5.43 5.45 Rangers Pump 25/04/2015 ------Railway Field River Field River Tunnel 26/03/2015 15.49 1.80 8.84 37.25 6.32 6.34

20/10/2015 18.41 1.50 13.64 29.87 2.26 4.44 10/03/2016 22.53 0.50 9.73 49.10 0.98 7.38 20/04/2017 19.95 0.50 11.05 247.00 6.96 6.96 14/12/2017 19.51 1.50 9.53 20.30 6.89 7.68 Jacobs Creek Old Gawler Jacobs Creek Gauge 3/05/2017 13.37 0.50 9.65 12.90 8.92 8.92 North Para Gomersal Rd River Bridge 24/03/2015 19.98 3.00 6.90 15.20 5.76 7.32 4/03/2016 20.25 2.80 8.01 17.60 1.81 1.81 3/05/2017 15.43 2.02 10.24 14.60 0.93 3.76 Yaldara 24/03/2015 20.39 5.23 7.25 15.37 3.95 6.43 7/10/2015 18.23 1.70 8.09 19.80 4.70 6.15 4/03/2016 23.54 5.00 8.23 0.00 0.83 3.56 9/11/2016 18.63 1.50 11.05 17.60 5.49 7.01 3/05/2017 12.63 2.40 9.94 15.60 6.02 6.88 13/12/2017 21.78 2.53 9.42 0.00 4.38 5.32 Tanunda Ck Tanunda Creek Gauge 7/10/2015 17.84 0.63 7.25 10.23 3.90 14.50 9/11/2016 17.56 0.40 12.44 10.85 5.90 7.68 13/12/2017 19.85 0.80 9.24 0.00 4.21 4.89 South Para River Mt Crawford 2/05/2017 12.59 1.20 11.21 24.40 4.03 4.54 Victoria Creek 3/03/2016 18.91 0.70 8.07 59.40 5.06 5.06 Victoria Reserve 3/03/2016 22.12 1.00 8.04 83.30 1.44 1.69 Hindmarsh Cootamundra Hindmarsh River Reserve 20/03/2015 22.62 7.70 5.72 47.23 0.78 1.72

15/10/2015 18.19 1.40 8.98 10.55 2.96 3.97 11/03/2016 21.58 2.10 9.70 9.80 2.51 3.18 4/11/2016 18.56 0.90 9.22 16.25 3.71 4.42

60 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Dissolved Dissolved Oxygen Oxygen Temperature Salinity Turbidity Min Max Catchment Watercourse Site Date (oC) (ppt) pH (NTU) (mg/L) (mg/L)

7/04/2017 19.44 12.00 8.09 13.50 1.07 3.44 5/12/2017 21.08 0.90 9.22 0.00 6.80 7.10 Hindmarsh Estuary 20/03/2015 22.83 11.40 8.52 11.88 7.21 10.25 15/10/2015 21.49 14.30 8.61 15.48 3.91 6.27 11/03/2016 24.13 7.63 8.39 14.80 2.54 3.99 4/11/2016 16.34 21.85 8.44 69.43 4.33 5.68 7/04/2017 20.74 20.88 7.99 13.70 1.44 3.73 5/12/2017 21.62 9.48 7.21 0.50 14.57 50.00 Lamont Rd 20/03/2015 23.32 11.57 7.74 14.47 3.83 5.78 15/10/2015 20.22 15.48 8.46 12.74 3.78 10.14 11/03/2016 25.13 7.23 9.29 11.90 2.99 5.44 4/11/2016 16.14 15.08 8.45 15.28 2.67 4.79 7/04/2017 22.12 22.02 7.92 11.70 1.13 2.75 5/12/2017 21.07 9.88 7.92 0.00 4.59 6.25 Hindmarsh Gauge 23/03/2016 ------12/04/2017 17.24 2.00 8.45 12.80 1.15 9.11 5/12/2017 19.79 0.80 9.13 1.00 6.90 7.32 BackValley BackValley Inman Trib Gauge 24/04/2015 ------Armstrong Inman River Rd Bridge 19/03/2015 19.43 2.47 6.79 29.40 1.46 6.42 14/10/2015 16.79 1.70 8.99 18.40 2.05 3.67 8/03/2016 20.67 2.35 9.15 36.60 0.78 4.54 3/11/2016 17.97 0.90 10.74 51.74 2.60 3.83 6/04/2017 18.33 2.63 6.82 30.40 1.06 3.01 6/12/2017 18.78 2.25 7.33 9.60 1.97 4.94 Glacier Rock 25/04/2015 ------16/10/2015 18.04 0.90 9.13 35.63 4.62 4.98 10/03/2016 21.15 1.93 9.73 18.10 1.01 3.86 2/11/2016 ------12/04/2017 15.17 1.60 8.88 17.10 4.43 4.52 7/12/2017 17.09 0.70 9.95 6.50 6.31 6.52 Inman Divine Gauge 16/10/2015 20.48 1.60 9.08 145.80 4.46 5.16 10/03/2016 23.52 2.80 8.77 0.00 1.10 2.39 12/04/2017 16.67 2.50 8.50 34.60 4.79 4.92 7/12/2017 17.70 2.13 7.94 9.40 5.78 6.09 Inman estuary 19/03/2015 25.48 21.20 8.12 18.80 10.77 10.77 14/10/2015 20.70 25.25 8.40 24.35 1.55 3.20 8/03/2016 22.91 23.53 7.96 21.90 0.47 2.44 3/11/2016 17.74 13.40 7.34 41.97 3.21 4.88 6/04/2017 20.28 30.54 6.88 17.60 0.40 4.27 61 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Dissolved Dissolved Oxygen Oxygen Temperature Salinity Turbidity Min Max Catchment Watercourse Site Date (oC) (ppt) pH (NTU) (mg/L) (mg/L)

6/12/2017 21.25 28.67 8.12 3.10 3.89 4.14 Swains Crossing Road 19/03/2015 22.33 4.20 7.30 22.65 6.70 8.04 14/10/2015 18.27 1.80 8.99 15.03 2.36 6.11 8/03/2016 21.37 3.03 9.60 12.80 1.73 4.13 3/11/2016 19.60 1.10 10.12 54.78 1.98 3.41 6/04/2017 16.63 2.60 8.24 25.20 3.12 4.22 6/12/2017 19.86 1.90 7.82 25.30 4.91 5.62 Myponga Myponga Myponga River estuary 18/03/2015 23.56 14.55 7.90 30.80 6.98 14.79

21/10/2015 19.95 2.16 10.13 32.10 4.30 6.69 22/03/2016 ------1/11/2016 16.70 25.30 8.24 17.33 6.18 9.67 11/04/2017 21.92 21.88 7.99 12.40 2.53 5.99 Myponga pumphouse 18/03/2015 19.09 1.47 7.56 10.10 7.73 8.63 21/10/2015 17.84 1.60 9.45 36.36 2.66 3.68 23/03/2016 ------1/11/2016 16.21 1.40 8.73 12.60 0.80 5.70 11/04/2017 16.64 1.46 8.50 12.30 1.58 4.15 Pages Flat 18/03/2015 17.53 0.30 6.84 30.93 1.21 3.34 Onkaparinga Aldgate Creek Mylor Bridge 25/03/2015 14.31 0.20 6.98 88.45 0.33 0.36

8/10/2015 15.18 0.20 7.44 20.60 5.70 10.85 2/03/2016 16.31 0.20 5.90 21.20 4.38 5.14 10/11/2016 13.40 0.20 12.42 14.07 1.70 7.05 21/04/2017 16.45 0.20 10.65 42.40 9.08 9.08 14/12/2017 20.47 0.20 7.10 9.60 5.31 5.52 Lenswood Lenswood Ck Gauge 4/03/2016 19.95 0.30 9.86 23.30 1.62 1.62 9/11/2016 13.52 0.33 12.83 15.43 5.37 5.70 2/05/2017 12.20 0.30 9.87 13.40 7.18 10.10 12/12/2017 15.56 0.40 7.43 0.00 6.14 7.50 Onkaparinga River MtBoldGate4 20/10/2015 13.38 0.30 14.00 22.60 1.83 4.79 Scott Creek Scott Creek Gauge 2/03/2016 19.56 1.00 9.49 15.70 2.26 50.00 Parananacooka Parananacooka Old Bridge 8/12/2017 16.58 1.40 8.01 5.20 6.43 7.00 Brownhill Craigburn Patawalonga Creek Dam 25/03/2015 20.91 1.20 9.06 11.20 7.76 9.90 DS Brownhill Caravan Park 8/10/2015 15.07 0.40 7.90 14.65 3.98 9.66 2/03/2016 19.99 0.50 6.85 47.80 6.94 8.32 10/11/2016 15.66 0.50 13.71 16.48 3.79 6.20 21/04/2017 18.10 0.50 10.84 18.70 4.35 7.22 12/12/2017 18.19 0.50 7.47 3.50 6.25 6.45

62 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

Dissolved Dissolved Oxygen Oxygen Temperature Salinity Turbidity Min Max Catchment Watercourse Site Date (oC) (ppt) pH (NTU) (mg/L) (mg/L) DS Sturt Gorge Retention Sturt Creek Dam 8/10/2015 14.50 0.62 8.68 53.60 0.00 7.55 West Beach Patawalonga Road Bridge 12/12/2017 21.45 0.93 9.73 5.90 5.48 6.67 Rarkang Rarkang creek Rarkang Creek Dam 27/03/2015 16.08 2.30 9.14 36.43 8.16 13.51 Chinaman's Torrens First Creek Hut 11/11/2016 15.32 0.10 14.00 12.27 0.09 0.39 Waterfall Gully 11/11/2016 13.61 0.10 11.36 12.15 0.40 0.79 Lower Torrens River Fox Creek 2/05/2017 12.34 0.40 10.32 0.00 0.10 2.53 Talunga Park Upper Torrens Bridge 24/03/2015 16.62 0.40 4.86 36.60 2.37 2.37 Eric Bonython Tunkalilla conservation Tunkalilla Creek park 15/12/2017 16.85 0.10 7.00 22.00 6.34 6.67 Willunga Creek Willunga Creek Ross Roses 20/04/2017 19.34 0.80 10.60 - 9.07 9.07 St. Johns Rd 20/04/2017 19.11 1.30 11.36 121.00 4.47 4.47 Ingalalla Yankalilla Yankalilla River Falls 27/03/2015 14.67 0.50 8.77 10.50 7.59 7.59 Yankalilla Bridge River 26/03/2015 16.02 4.70 8.98 213.00 11.30 11.30 5/04/2017 16.25 1.93 8.03 20.00 6.16 6.62 Yattagolinga Yattagolinga River Croser 5/04/2017 15.71 1.20 8.18 12.60 3.77 4.11

63 Schmarr, D. et al. (2018) Fish monitoring across regional catchments of the Adelaide and Mount Lofty Ranges region 2015–17

9. APPENDIX D Report by Emma Matthews

Characterisation of aquatic and riparian vegetation after large scouring floods in the Myponga River, South Australia.

Characterisation of aquatic and riparian vegetation after large scouring floods in the Myponga River, South Australia

Emma Matthews

Supervisor: David Schmarr

Summer Scholarship Report 2016-17

Characterisation of aquatic and riparian vegetation after large scouring floods in the Myponga River, South DavidAustralia Schmarr 2016

Emma Matthews

December 2016 - January 2017

Summer Scholarship Project

South Australian Research and Development Institute

Abstract

Located on the Fleurieu Peninsula, South Australia, the Myponga River is a system of both economic and environmental importance. The Myponga Reservoir and has drastically altered hydrology of the system, leading to significant biodiversity loss. September 2016 was a period of intense rainfall for South Australia, which had many environmental impacts, including widespread flooding. The Myponga River was subject to not only natural inflows from flood waters, but operational releases of 3-4 GL of water from the almost full Myponga reservoir.

This study aimed to compare cover pre and post-flood, and characterise aquatic and riparian vegetation after the large scouring floods. Vegetation surveys were conducted at three representative 100m sites, each with three transects containing 15 quadrats. Survey results revealed no relationship between hydrology and vegetation both at species and functional group level. Systematic cover analysis using Google Earth satellite imagery of 10

100m sites, including the three field sites, revealed a drastic reduction in vegetation cover. In conjunction, these results suggest the system is in an early successional state, with the flood and reservoir release the main drivers of current plant distribution. It was concluded that presence or absence of species was a result of resistance or intolerance to flooding, and in the case of abundant species, the exploitation of newly cleared areas. It is expected that over time with the replacement of annuals by perennials the vegetation will return to pre-flood condition.

This study will provide a source of reference data for use in future monitoring of the recovery of the Myponga

River system.

Introduction

The Myponga River is located on the Fleurieu Peninsula, about 70km South of Adelaide. The river drains the

Western Mount Lofty Ranges catchments, discharging freshwater flows into Gulf St. Vincent (McNeil et al. 2009).

European settlement has resulted in the heavy modification of water resources for urban, industrial, and agricultural use, with the construction of weirs and dams recognised as the most significant ecological threat to river systems (McNeil et al. 2009; Bunn et al. 2002; King et al. 2015; Maxwell et al. 2015). Literature reviews have estimated between 87% (Lloyd et al. 2005) and 92% (Poff et al. 2009) of existing studies on the effects of modified flow found significant ecological and/or geomorphological effects. Anthropogenic pressures have highly modified the Myponga River system, most notably as a result of the construction of the Myponga Reservoir

3 in 1962, which provides water to approximately 50,000 people (Bradford et al. 2007; SA Water 2016). The reservoir has a capacity of 26 800 ML, and stores water at the township of Myponga, preventing much of the natural freshwater flows from reaching the coast, although this is supplemented by freshwater runoff downstream of the weir, and by flood events (Bradford et al. 2007; McNeil et al. 2009; SA Water 2016). The reservoir has greatly reduced flow and connectivity between systems located up and downstream of the dam wall, and this has had significant impacts on fish populations, vegetation and water quality, which were intensified by the additional selective pressures of the Millennium Drought period between 1997-2009

(Bradford et al. 2007; McNeil et al. 2009).

Flow regime is a crucial component of river ecosystems which affects many aspects of ecological functioning

(Bunn et al. 2002; Lloyd et al. 2005; Poff et al. 2009). Diversity and abundance of aquatic and riparian vegetation which surround water systems is also largely determined by abiotic flow regime factors, while the rejuvenation of flood plains is wholly reliant on flow (Casanova & Brock 2000; Bunn et al. 2002). Studies have shown that many r-selected invasive species benefit from modified flows, an important consideration in weed mitigation

(Grime 1979; Bunn et al. 2002). Vegetation is influenced by depth, duration, rate and frequency of inundation, and studies have shown that post-flood species composition is a result of the ability to tolerate or respond to altered hydrology (Casanova & Brock 2000). The construction of the reservoir has resulted in a highly controlled and stable water regime, with small and medium flows eradicated (Casanova & Brock 2000). Common impacts of altered hydrology include low flows, desiccation and loss of species, although the characteristics of each system determines the extent and importance of these effects. These factors have made the research and maintenance of freshwater flows a conservation and management priority. This is an important area of research as plants are invaluable to aquatic ecosystems, providing habitat and food sources, maintaining water quality, and assisting with nutrient cycling (Jansen et al. 2001; Bradford et al. 2007; Maxwell et al. 2015).

Desiccation as a result of reduced flow and land management practices have furthered the general degradation of these systems (Lloyd et al. 2005; Bradford et al. 2007; McNeil et al. 2009). Clearance and usage of land for dairying and pasture has restricted vegetation to a thin band surrounding the river, and is comprised largely of exotic tree and grass species, with comparatively little remaining native riparian vegetation, although it is worth noting that below the reservoir vegetation is naturally constrained by a gorge (Bradford et al. 2007; McNeil et al.

2011; Linden et al. 2010). Restoration projects have effectively restored small areas of riparian vegetation

4 through the removal of introduced plants and exclusion of grazing animals, in efforts to conserve the habitat requirements for endangered species and improve water quality, although the vegetation of the system remains fragmented (Bradford et al. 2007; Linden et al. 2010).

September 2016 saw unprecedented rain events for South Australia, resulting in large scouring floods which affected several river systems, including the Myponga River (Bureau of Meteorology 2016). Heavy rainfall, below average temperatures and strong winds were frequent throughout the month, with Myponga receiving 171.6 mm of rain, almost 90 mm more than average, and the highest recorded for September since 1992 (Bureau of

Meteorology 2016). In addition to natural inflows entering the catchment during this period, 3-4 GL of water was released over approximately 12 hours from the Myponga Reservoir, which was nearing its full capacity (SA Water

2016). This release is presumably responsible for many of the flood impacts observed in this area. The quantity of water and time frame over which it was released is rare and therefore knowledge of long term effects is limited.

A pulse-scale flow modification, the altered hydrology of the system has seemingly removed large volumes of aquatic vegetation and potentially displaced native fish species (Lloyd et al. 2005). The extent of damage to

Australian ecosystems as a result of scouring floods remains relatively unknown due to the rarity of these events, and complexity of ecological responses among species with differing ecological requirements (Lloyd et al. 2005;

Poff et al. 2009). A variety of flow types are important drivers of river systems, however the reservoir prevents small and medium flows, meaning large scouring flows are the only type to reach downstream of the reservoir.

Altered hydrology as a result of flow modification adds additional complexity to predicting the outcomes of large flows. With the continued unpredictability of weather events as a result of climate change, an understanding of potential impacts and the ways in which these can be mitigated is an important means of preserving and managing our natural resources (Bardsley et al. 2010; Cooper et al. 2013).

Visual observations and anecdotal evidence from landowners supports the theory that the 2016 flood has greatly altered the vegetation, with the thinning and removal of many scouring intolerant species, and influx of both native and invasive opportunistic species which appear to have taken advantage of this major ecological disturbance.

In this study, I aim to provide a characterisation of the aquatic and riparian vegetation in the Myponga River system after the September 2016 large scouring flood and draw possible conclusions about the impacts of

5 unprecedented flows such as this on Australian river systems. This study will act as a source of reference data for future vegetation studies of the region, and allow for the recovery of the system to be monitored using established field sites.

Methods

Site Locations

Sampling was conducted at the Myponga River catchment on the 15th and 16th of December 2016 (Figure 1).

Three sites deemed to be representative of the system were chosen based on both visual appearance and the underlying geological categories of the region, using SARIG’s 100k Geology layer (2016). The sites are displayed on a satellite map (Figure 2), and coordinates are detailed (Table 1). Coordinates of the transects at each site can be found in Appendix 1.

Figure 1: Location of the Myponga Reservoir and catchment area, 70km south of Adelaide, South Australia.

Figure 2. Approximate locations of the start and end points of the three 100m field sites, Myponga River, South Australia.

Table 1. Coordinates of each of the three 100m field sites at Myponga River, South Australia.

Site 0m (Start) Coordinates 100 m (End) Coordinates 1 Latitude: 35°22'49.06"S Latitude: -35.3794699721 Longitude: 138°23'7.63"E Longitude : 138.3859489951

2 Latitude: -35.3906090185 Latitude: -35.3899600077 Longitude : 138.3896989748 Longitude : 138.3889779635

3 Latitude: -35.3965279832 Latitude: -35.3973319754 Longitude : 138.4054550249 Longitude : 138.4047950339

Sampling Methods

Total Cover Comparisons

Freely available Google Earth imagery was downloaded for the three field sites using GPS coordinates, as well as seven additional 100m sites which were spaced evenly along the remaining river length (see Appendix 2 for coordinates). The most current image (14/10/2016), and historic, pre-flood (17/10/2015) image from the same month were selected for each site using the same location, angle, display and altitude settings. Using the layer feature in Adobe Photoshop Elements, a grid (1.0 cm x 1.0 cm) and guidelines specific to each site were overlayed onto the images. Guidelines extended two grid squares from the maximum width of the water visible in current images at any given point. In this way exactly the same area for both current and historic imagery could be sampled, allowing for direct comparison. The dominant cover type for each grid square was manually recorded by inspection as one of the following categories: water/boulders (white/grey colouration), open land or grass

(light green or brown, smooth), or vegetation (darker green, textured) (see Figure 3). Percent cover for each site and a total for each cover type site wide was calculated, as well as the percent increase/decrease of cover types between the years of 2015 and 2016.

Figure 3: Example of Google Earth imagery with grid overlay and characteristic appearance of each cover category, where W = water/boulders, O = open land or grass, and V= vegetation cover.

Figure 4. An example of the difference in vegetation cover visible using Google Earth imagery for Site 3: A) historic, pre-flood 17/10/2015, and B) Current, post-flood 14/10/2016.

Vegetation Surveys

At each site a measuring tape was extended 100m along the length of the river, upstream to downstream, and the GPS coordinates of the 0 (start) and 100m (end) points recorded (Figure 5). Three transects were set up perpendicular to the river and 100m tape, at random positions along the 100m transect determined using a random number generator (Figure 5). Transects were of either consistent or varying length, and were determined by the maximum water width during the flood, known as the high water mark. The 0 m (start) point of each transect was located on the right of the water, looking downstream. Vegetation surveys were conducted at 15 1.0 m x 1.0 m quadrats which were also spaced along each transect using a random number generator, with the quadrat placed against and upstream of the tape. Where random numbers resulted in overlapping quadrats,

9 the next nearest non-overlapping 1 m section was selected. Current width and assumed flood width (high water mark) of the river at each transect was recorded, as well as the water habitat type for quadrats containing water

(including pool, riffle, run, and riffle/run, modified from Nicol, 2013). Permanent stakes were set up at the distal ends of each transect and 100m site as photo points and for potential use in subsequent studies.

Figure 5. A) Plan view of idealised site showing vegetation survey protocol for Myponga River, South Australia. 100m sites were set up along the river, with three randomly spaced perpendicular transects (T1, T2, T3), each with 15 1.0 m x 1.0 m randomly spaced quadrats (represented by black boxes). B) Example of actual site locations including approximate distal ends of each transect, where T#= transect number and L/R indicates left or right side of the river.

At each of the 15 quadrats per transect, a vegetation survey was conducted involving an estimation of percentage cover using the following categories:

 Soil – all soil types including sand  Gravel – includes stone ranging from small rocks and pebbles to boulders  Bedrock – large solid rock protruding from underground  Water – sources including the main water body, offshoots and puddles  Plants – any form of living vegetation (above or on the water’s surface for water quadrats)  Debris – dead plants, leaf litter, and general debris from flooding

Where possible individual plants were identified to genus or species level using Sainty and Jacobs (1994),

Dashorst and Jessop (1998), and Cunningham et al. (1992). Only plants above the water’s surface were recorded due to time constraints, although the presence of submerged filamentous algae was noted. When plants were not able to be identified on site, coded photographs and small labelled samples were taken for later identification and expert assistance. Any non-native grasses unable to be identified on site were grouped as a single taxon and referred to as “Exotic Grasses”, whilst other species unable to be identified by experts were given a reference code, for example UK1 (unknown species 1). Images of the unidentified species can be found in Appendix 4.

Plant cover was further divided into water plant functional groups present (as per Casanova 2011, see Appendix

3), and a photograph of each quadrat was taken to support written cover estimates and for future reference. At each of the three sites a visual estimate of total cover composition types was also recorded.

Statistical Analysis

Vegetation survey data was used to calculate several species indices (both at functional group and at species level) for individual sites and the combined total. The indices used are defined as follows:

 Species richness, S = total number of Species

 Species count / occupancy, Ki = the number of quadrats in which species i is found  Probability of occurrence = calculated by dividing the number count for species i by the total number of quadrats for that site/total 푆  Species diversity (Shannon-Weaver), H = − ∑푖=1 푃푖푙표푔푃푖 (where Pi = proportional abundance)  Pielou’s species evenness (or equitability), J = H/log(S)

The distance of each quadrat from the water was converted to a percentage using the water width and position of water on, and total length of, the transect1. In this way sample size was effectively doubled through using the data from both the left and right sides of the river, and the relationship between distance from water and the presence of functional groups present could be assessed.

1 Except for Site 1, where the water position was not recorded. For this site it was assumed that the river was in the centre of the transect. 12

Figure 6. Sites 1 (A), 2 (B), and 3 (C), and an example of a transect at each: 1 (D), 2 (E), and 3 (F). Images courtesy of David Schmarr, 2016.

Results

Total Cover Comparisons

Cover type compositions differed greatly between pre-flood and post-flood conditions. The pre-flood imagery had a noticeable lack of visible water, which appeared only at site 6, and comprised less than 1% of the sampled cover for that site (Figure 7A). Pre-flood images had an average of 84% vegetation cover, and 16% open land or grass (Figure 7A). The average composition of cover types in the post-flood imagery was 34% water/stone, 22% open land or grass, and 44 % vegetation cover (Figure 7B). This indicates a 39% decline in vegetation cover between October 2015 and 2016, resulting in a 34% increase in visible water/stone, and 6% increase in open land and grasses.

A. Pre-flood 2015 B. Post-flood 2016

Water/Stone Water/Stone Open Land/Grass Open Land/Grass Vegetation Cover Vegetation Cover

Figure 7. Average composition of river and riparian zone cover types based on Google Earth imagery analysis from A) pre-flood (17/10/2015) and B) post-flood conditions (14/10/2016).

Vegetation Surveys 13

A total of 133 quadrats were sampled across the three different sites, each with 3 randomly spaced transects

containing 15 quadrats.2 At total of 55 taxa were recorded (of which four remained unidentified, see Appendix

4), belonging to a total of five functional groups (Table 2). The functional groups with the most quadrat

observations were terrestrial dry (Tdr), members of which were found in 56 quadrats, and amphibious

fluctuation tolerator emergent (ATe), found in 45 quadrats (Table 2). 65% of taxa observed are considered non-

natives or weeds (Table 2). Species found at all three sites included the sedge Cyperus vaginatus, Ehrharta

longiflora and invasive weeds Rubus fruticosus agg. and Sonchus sp. (Table 2).

Table 2. Complete list of observed plant taxa, corresponding functional groups, and sites they were recorded at (X), Myponga River, South Australia. * indicates non-native or weed status taxon

Presence Taxon Functional Group Code Site 1 Site 2 Site 3 Anagallis arvensis* Terrestrial damp Tda X X Asparagus asparagoides* Terrestrial dry Tdr X Aster subulatus Terrestrial dry Tdr X Atriplex vesicaria Terrestrial dry Tdr X Briza minor * Terrestrial dry Tdr X Bromus catharticus* Terrestrial dry Tdr X Bromus sp.* Terrestrial dry Tdr X Callistemon sieberi Terrestrial dry Tdr X Callistemon sp. Terrestrial dry Tdr X Calystegia sepium Amphibious fluctuation tolerator emergent ATe X Chenopodium album * Terrestrial damp Tda X Cyperus vaginatus Amphibious fluctuation tolerator emergent ATe X X X Cytisus scoparius* Terrestrial dry Tdr X Ehrharta longiflora* Terrestrial dry Tdr X X X Euphorbia sp.* Terrestrial dry Tdr X X Exotic grass * Terrestrial dry Tdr X X Fumaria bastardii* Terrestrial damp Tda X Fumaria capreolata* Terrestrial damp Tda X Galium aparine* Terrestrial dry Tdr X Hordeum sp.* Terrestrial dry Tdr X Hordeum vulgare* Terrestrial dry Tdr X Hydrocotyle sp. * Amphibious fluctuation responder plastic AFRp X Isolepis inundata Amphibious fluctuation tolerator emergent SE X Isolepis sp.* Amphibious fluctuation tolerator emergent ATe X X Juncus kraussii Amphibious fluctuation tolerator emergent ATe X Juncus usitatus Amphibious fluctuation tolerator emergent ATe X Leptospermum lanigerum Terrestrial damp Tda X X

2 Except transect 3, site 3, for which only 13 quadrats were sampled due to the short transect length. 14

Leptospermum sp. Terrestrial damp Tda X X Lobelia anceps* Terrestrial damp Tda X Lophochloa cristata Terrestrial dry Tdr X Medicago sp.* Terrestrial dry Tdr X Melilotus sp. * Terrestrial dry Tdr X Phragmites australis Amphibious fluctuation tolerator emergent Ate X X Picris sp.* Terrestrial dry Tdr X Plantago lanceolata* Terrestrial dry Tdr X Polygonum aviculare* Terrestrial dry Tdr X Polypogon monspeliensis* Amphibious fluctuation tolerator emergent ATe X Pteridium esculentum Terrestrial dry Tdr X Pultenaea largiflorens Terrestrial dry Tdr X Rubus fruticosus agg.* Amphibious fluctuation tolerator emergent ATe X X X Senecio cunninghamii* Terrestrial dry Tdr X X Senecio pterophorus * Terrestrial dry Tdr X Solanum nigrum * Terrestrial damp Tda X Solanum sp.* Terrestrial damp Tda X Sonchus oleraceus * Terrestrial dry Tdr X X Sonchus sp.* Terrestrial dry Tdr X X X Taraxacum officinale* Terrestrial dry Tdr X Themeda triandra Terrestrial dry Tdr X Trifolium sp.* Terrestrial dry Tdr X Typha domingensis Submerged emergent SE X X Ulex europaeus* Terrestrial dry Tdr X Unknown Species 1 NA NA X Unknown Species 2 NA NA X Unknown Species 3 NA NA X Unknown Species 4 NA NA X

Site 3 had the highest species richness and diversity, and was the most even, with scores of 31, 3.22 and 0.94

respectively (Table 3). Site 2 had seven fewer species than site 3 and lower species diversity, with similar

evenness, while site 1 had six fewer species than site 2, lower evenness and lower diversity (Table 3).

Table 3. Species Indices for comparing community diversity between the 3 sites, Myponga River, South Australia.

Species Richness Species Diversity Species Evenness Site 1 18 2.13 0.74 Site 2 24 2.85 0.9 Site 3 31 3.22 0.94 Total 55 NA NA

Most taxa had a relatively low probability of occurrence, with a few higher probability dominant taxa found at each of the three sites (Table 4). The species with the highest probability of occurrence at site 1 was a sedge,

Cyperus vaginatus (60%), followed by annual grasses Ehrharta longiflora (27%) and Bromus catharticus (13%)

(Table 4, Figure 8A). The most common species at site 2 was Typha domingensis, with a probability of 31%, followed by a range of taxa with similar probabilities, including grasses Ehrharta longiflora (11%) and

Phragmites australis (9%), and several invasive weed species such as Asparagus asparagoides (15%), Solanum sp. (11%), and Rubus fruticosus agg. (9%) (Table 4, Figure 8B). Site 3 had a much higher probability of larger shrubs, with the most likely taxon Callistemon sp. (25%), followed by weeds Cytisus scoparius (21%) and Ulex europaeus (16%) (Table 4, Figure 8C). Although only surface plants were identified and recorded, the presence of filamentous algae was noted in 83% of water habitat quadrats.

Table 4. Quadrat occupancy and probability of occurrence (%) for observed plant taxa at sites 1, 2, 3, and total, Myponga River, South Australia.

Count/ Occupancy Probability of Occurrence (%) Species Site 1 Site 2 Site 3 Total Site 1 Site 2 Site 3 Total Anagallis arvensis 1 0 2 3 2.22 0.00 4.65 2.26 Asparagus asparagoides 0 7 0 7 0.00 15.56 0.00 5.26 Aster subulatus 1 0 2 3 2.22 0.00 4.65 2.26 Atriplex vesicaria 0 2 0 2 0.00 4.44 0.00 1.50 Briza Minor 0 0 1 1 0.00 0.00 2.33 0.75 Bromus catharticus 6 0 0 6 13.33 0.00 0.00 4.51 Bromus sp. 0 0 1 1 0.00 0.00 2.33 0.75 Callistemon sieberi 0 0 1 1 0.00 0.00 2.33 0.75 Callistemon sp. 0 0 11 11 0.00 0.00 25.58 8.27 Calystegia sepium 0 1 0 1 0.00 2.22 0.00 0.75 Chenopodium album 0 0 1 1 0.00 0.00 2.33 0.75 Cyperus vaginatus 27 3 1 31 60.00 6.67 2.33 23.31 Cytisus scoparius 0 0 9 9 0.00 0.00 20.93 6.77 Ehrharta longiflora 12 5 3 20 26.67 11.11 6.98 15.04 Euphorbia sp. 0 1 2 3 0.00 2.22 4.65 2.26 Exotic grass 1 0 3 4 2.22 0.00 6.98 3.01 Fumaria bastardii 0 0 1 1 0.00 0.00 2.33 0.75 Fumaria capreolata 0 2 0 2 0.00 4.44 0.00 1.50 Galium aparine 0 0 2 2 0.00 0.00 4.65 1.50 Hordeum sp. 1 0 0 1 2.22 0.00 0.00 0.75 Hordeum vulgare 0 0 3 3 0.00 0.00 6.98 2.26 Hydrocotyle sp. 0 0 1 1 0.00 0.00 2.33 0.75 Isolepis inundata 0 0 4 4 0.00 0.00 9.30 3.01 Isolepis sp. 2 0 2 4 4.44 0.00 4.65 3.01 Juncus kraussii 2 0 0 2 4.44 0.00 0.00 1.50 Juncus usitatus 2 0 0 2 4.44 0.00 0.00 1.50 16

Leptospermum lanigerum 0 1 1 2 0.00 2.22 2.33 1.50 Leptospermum sp. 0 2 2 4 0.00 4.44 4.65 3.01 Lobelia anceps 0 0 3 3 0.00 0.00 6.98 2.26 Lophochloa cristata 2 0 0 2 4.44 0.00 0.00 1.50 Medicago sp. 0 1 0 1 0.00 2.22 0.00 0.75 Melilotus sp. 0 1 0 1 0.00 2.22 0.00 0.75 Phragmites australis 3 4 0 7 6.67 8.89 0.00 5.26 Picris sp. 0 1 0 1 0.00 2.22 0.00 0.75 Plantago lanceolata 0 0 1 1 0.00 0.00 2.33 0.75 Polygonum aviculare 0 1 0 1 0.00 2.22 0.00 0.75 Polypogon monspeliensis 4 0 0 4 8.89 0.00 0.00 3.01 Pteridium esculentum 0 2 2 4 0.00 4.44 4.65 3.01 Pultenaea largiflorens 0 0 2 2 0.00 0.00 4.65 1.50 Rubus fruticosus agg. 1 4 4 9 2.22 8.89 9.30 6.77 Senecio cunninghamii 0 1 1 2 0.00 2.22 2.33 1.50 Senecio pterophorus 0 2 0 2 0.00 4.44 0.00 1.50 Solanum nigrum 0 1 3 4 0.00 2.22 6.98 3.01 Solanum sp. 0 5 0 5 0.00 11.11 0.00 3.76 Sonchus oleraceus 0 0 3 3 0.00 0.00 6.98 2.26 Sonchus sp. 1 2 4 7 2.22 4.44 9.30 5.26 Taraxacum officinale 0 0 2 2 0.00 0.00 4.65 1.50 Themeda triandra 1 0 0 1 2.22 0.00 0.00 0.75 Trifolium sp. 0 2 0 2 0.00 4.44 0.00 1.50 Typha domingensis 2 14 0 16 4.44 31.11 0.00 12.03 Ulex europaeus 0 0 7 7 0.00 0.00 16.28 5.26 Unknown Species 1 0 0 1 1 0.00 0.00 2.33 0.75 Unknown Species 2 0 1 0 1 0.00 2.22 0.00 0.75 Unknown Species 3 0 1 0 1 0.00 2.22 0.00 0.75 Unknown Species 4 0 0 1 1 0.00 0.00 2.33 0.75

A Site 1 70.00 60.00 50.00 40.00 30.00 20.00 10.00

0.00 Probability Probability ofoccurence (%)

Taxa

B Site 2 35.00 30.00 25.00 20.00 15.00

10.00 5.00

0.00 Probability Probability ofoccurence (%)

Taxa

Site 3 30.00 25.00 20.00 15.00 10.00 5.00

0.00

Probability Probability ofoccurence (%)

Isolepissp.

Bromussp.

Exoticgrass

Sonchussp.

Briza Briza Minor

Euphorbia sp.

Lobeliaanceps

Callistemonsp.

Leptospermum…

Galiumaparine

Aster subulatus Aster

Ulex europaeus Ulex

Hydrocotylesp.

Solanumnigrum

Isolepisinundata

Cytisusscoparius

Hordeumvulgare

Fumaria bastardii

Anagallisarvensis

Cyperusvaginatus

Sonchusoleraceus

Leptospermumsp.

Ehrhartalongiflora

Callistemonsieberi

UnknownSpecies 1 UnknownSpecies 4

Plantago lanceolata

Taraxacumofficinale

Chenopodiumalbum

Rubusfruticosus agg.

Senecio cunninghamii

Pultenaealargiflorens Pteridiumesculentum Taxa

D Total 25.00

20.00

15.00

10.00

5.00 Probability Probability ofoccurence (%) 0.00

Taxa

Figure 8. Probability of occurrence (%) for observed plant taxa present at sites 1 (A), 2 (B), and 3 (C), and probability of occurrence for the 20 most abundant taxa overall (D), Myponga River, South Australia.

The group with the highest counts and probability of occurrence for site 1 was ATe, site 2 was SE, and site 3 was

Tdr, which also had the highest total probability of occurrence (Table 5). Amphibious fluctuation responder plastic (ARp) is the group least likely to be found, appearing in only one quadrat at site 3 (Table 5). Site one was dominated by two functional groups, with a 69% probability of ATe and 40% probability of Tdr (Table 5). Site 2 showed a more even distribution of probabilities between functional groups, and site 3 was dominated by Tdr species, with a probability of 63% (Table 5).

Table 5. Quadrat occupancy and probability of occurrence (%) at sites 1, 2 and 3, and total for functional groups

(Casanova, 2011), Myponga River, South Australia.

Count/ Occupancy Probability of Occurrence Functional Group Site 1 Site 2 Site 3 Total Site 1 Site 2 Site 3 Total ATe 31 9 10 50 68.89 20.00 23.26 37.59 Tdr 18 16 27 61 40.00 35.56 62.79 45.86 Tda 0 9 9 18 0.00 20.00 20.93 13.53 SE 1 15 0 16 2.22 33.33 0.00 12.03 ARp 0 0 1 1 0.00 0.00 2.33 0.75

80.00 70.00

60.00

50.00 Site 1 40.00 Site 2 30.00 Site 3 20.00 Total Probability Probability ofoccurence (%) 10.00

0.00 ATe Tdr Tda SE ARp Functional Group

Figure 9: Probability of occurrence (%) of each functional group (Casanova 2011) at each field site, Myponga

River, South Australia. See Appendix 3 for functional group codes.

Correlation coefficients showed no statistically significant relationship between probability of occurrence of individual species and the distance from the water (Figure 10). There was also no statistically significant relationship between presence of functional groups and the distance from the water (Table 6). This is confirmed in Figure 10, which shows poor coefficient of determination (r2) values for all functional group relationships, indicating that distance from the water does not have a constant relationship for these groups at this location.

However, the ATe and submerged emergents (SE) had much higher counts in 0% distance quadrats than anywhere else indicating the influence of being in the water itself (Table 6).

Table 6. Total quadrat occupancy for functional groups (Casanova, 2011) organised by distance from water (%, where 0 includes 0 -<5%), Myponga River, South Australia. See Appendix 3 for functional group codes.

Percent ATe Tdr SE Tda ARp Distance Class (%) 0.00 10 6 12 1.00 0 5.00 2 2 0 0.00 0 10.00 1 1 0 0.00 0 15.00 1 2 0 3.00 0 20.00 2 3 0 1.00 0 25.00 2 2 2 2.00 0 30.00 3 8 0 4.00 0 35.00 1 2 0 0.00 0 40.00 4 2 0 1.00 0 45.00 2 4 1 1.00 0 50.00 1 5 0 2.00 0 55.00 1 2 0 1.00 0 60.00 3 0 0 0.00 0 65.00 2 5 0 3.00 0 70.00 4 2 2 0.00 0 75.00 2 1 0 0.00 0 80.00 3 6 0 1.00 1 85.00 2 2 0 0.00 0 90.00 0 1 0 1.00 0 95.00 2 4 0 0.00 0

Amphibious fluctuation tolerator emergent (ATe) 12

8 R² = 0.1513 6

2 ATe ATe quadrattotalOccupancy 0 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 Distance from water (%)

B Terrestrial dry (Tdr) 10 R² = 0.0262 8 6 4 Tdr 2 0 0.00 20.00 40.00 60.00 80.00 100.00

Total Tdr QuadratOccupancy Distance from water (%)

Submerged Emerget (SE) 14 12 10 8 R² = 0.2898 6 4 2

0 SE SE total Occupancyquadrat -20.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 Distance from water (%)

Terrestrial damp (Tda) 5.00 R² = 0.1153 4.00

3.00

2.00

1.00 0.00

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 TdaTotal Quadrat Occupancy Distance from water (%)

Figure 10. Total quadrat occupancy as a function of distance from water (%) for each observed functional group (Casanova, 2011), Myponga River, South Australia (except ARp, for which there was only one observation).

Water habitat type appeared to have some influence over the functional groups present (Figure 11). The majority of pool habitat quadrats had no plants present, at 71% (Figure 11). SE had the highest occupancies for riffle and run, run, and riffle habitats at 75, 57, and 75%, while of the plants that were present in pools, ATe had the highest occupancy at 21% (Figure 11). SE also had the highest occupancy for water habitats overall, and terrestrial damp

(Tda) the least, closely followed by Tdr, at 1.79 and 4.29% respectively (Figure 11).

80.00

70.00

60.00

50.00 ATE 40.00 Tda

30.00 Tdr

PercentOccupancy SE 20.00 None 10.00

0.00 Pool Riffle and Run Riffle Run Water habitat type

Figure 11. Percent occupancy of functional groups (Casanova 2011) for each water habitat type at Myponga River, South Australia. Note ARp is not shown as it was not found in any water quadrats.

The composition of total site cover estimates varied between the three sites (Figure 12). For all sites plants had the highest percentage cover, with gravel and water as the following highest percent cover types (Figure 12).

Site 2 was the only area with significant bedrock and soil cover, and site 3 the only with significant recorded debris (Figure 12).

Site 1 Site 2

0% 0% 0% Soil 5% Soil Gravel 15% Gravel 38% 40% Bedrock Bedrock

57% Water 20% Water Plants Plants 0% Debris 20% Debris 5%

Site 3

5% Soil 30% Gravel Bedrock

45% Water 0% Plants 20% Debris

Figure 12. Visual estimates of total cover composition for field sites 1, 2 and 3, Myponga River, South Australia.

Discussion

Total Cover Comparisons

Cover comparison studies showed a significant difference in cover compositions under pre- and post-flood conditions, as is to be expected under such circumstances (Poff et al. 2009) (Figure 7). The lack of visible water in pre-flood conditions can be explained by both significantly lower rainfall during 2015 (Bureau of Meteorology

2017), and the presence of thick and dense vegetation as the dominant cover type (Bradford et al. 2007). At 84%, vegetation cover can be expected to obscure the view of water in satellite imagery at this level of resolution. The

39% decrease in vegetation cover evident in the post flood imagery is largely a result of September 2016 weather events, including both natural and controlled release flows (Miller et al. 2013)(Figure 7). A decrease in vegetation cover can be expected during scouring flooding as the force of the water uproots and breaks plants, allowing them to be carried in the water column (Poff et al. 2009; Poff et al. 2009); however the magnitude of the observed change here suggests that the controlled release may have increased these impacts. In addition, extended inundation periods can kill species intolerant to prolonged flooding, such as small terrestrial dry species (Casanova 2011; Poff et al. 2009). The increase in open land/grass is likely a result of this vegetation removal, while the 34% increase in visible stone/water is probably a result of both removal of vegetation and increased volumes of water (Poff et al. 2009) (Table 7). Although the impact of other factors on these changes in cover cannot be excluded, there is strong anecdotal and photographic evidence to support this theory. It is

25 possible that the magnitude of the observed effects are larger than is natural due to encroachment of riparian vegetation as a result of reduced flows (Poff et al. 2009).

The method of cover analysis developed for this study is sufficient in providing an overview of large-scale cover changes, and providing some quantitative evidence for what is evident visually. Advantages of this technique include the accessibility of the extensive free imagery available from the Google Earth software, and the simplicity of the cover assessment method using free applications in Adobe Photoshop Elements. This method is limited however by the manually intensive categorising process and the resolution of imagery for this region, which limits both the accuracy and detail of the data collected. Suggested improvements include using images with more temporal resolution such as Nearmap, or the collection of high resolution imagery using drones. Cover analysis computer programs could also potentially be adapted or created for this application.

Vegetation Surveys

The sampled areas of the Myponga River support visual evidence that the site is a heterogeneous area with varied diversity and distribution of plant species. The observation of 54 different taxa across five functional groups

(Table 2) was considered reasonable coverage for a study of this level. The four taxa observed at all three sites each have characteristics explaining their persistence in multiple habitat types (Table 2). Cyperus vaginatus is a common perennial sedge found along rivers, and grows in many conditions including sand, clay and alluvium rock (Florabase 2017). An ATe, site 1 in particular favoured Cyperus vaginatus, as there was a 68% probability of occurrence for this functional group (Table 5), which survives in water, saturated sands and soils and fluctuations in depth, all of which occurred during the September floods (Casanova 2011). Although the above ground biomass may have been damaged or removed during flooding, Cyperus has extensive rhizome networks allowing persistence during floods, and resprouting afterwards. Ehrharta longiflora, a Tdr grass, was likely favoured by the intolerance of many other plants to sandy soils, as large amounts of sand appeared to be shifted into the river system during flooding. This is particularly true of site 1 which had large quantities of sand, where

Ehrharta longiflora had a 27% probability of occurrence (Table 4). Blackberries (Rubus fruticosus agg.) are stress tolerators, while Sonchus sp. is an opportunistic weeds which thrives in disturbed conditions, and both are established pests in this region and in riparian habitats generally (Jansen et al. 2001; Bradford et al. 2007).

Taxa found in fewer sites may have a narrower range of habitat requirements excluding them from one or more 26 sites pre or post-flood, or be a rare species, such as Unknown 3, which an experienced botanist in this field had never observed (Jason Nicol pers. comm.) and is therefore likely scarce in this area (see Appendix 4). Lack of observation of a taxon at a given site may also be a result of random sampling which does not record every species present sitewide. Exotic species are known to thrive in modified rivers (Cooper et al. 2013). The disturbance and clearing caused by the floods favours r –selected species and may explain the high prevalence of exotic species, which comprised 65% of all observed species (Grime 1979; Poff et al. 2009) (Table 2).

The high diversity and evenness of site 3 (Table 3), can be explained by the presence of a high number of species

(31) found at low even frequencies (Figure 8C). This could be partly due to the high probability of occurrence of the Tdr group (Table 5), for which there are many species members, and varied conditions also supporting the presence of the other three observed groups (Figure 9). The lower diversity and similar evenness of site 2 (Table

3) can be explained by the dominance of Typha domingensis, followed by the similar probability of occurrences for several key groups including Ehrharta longiflora, Phragmites australis, Asparagus asparagoides, Solanum sp. and Rubus fruticosus agg. (Figure 8B). Both Typha and Phragmites are rhizomatous perennials which are adapted for the rapid colonisation of shallow waters, preventing competitors from growing, which explains their abundance (Nicol 2013). The lower diversity and evenness (Table 3) of site 1 is due to the dominance of Cyperus vaginatus at 60% probability (Table 4), which sprouts from rhizomes after flooding.

Members of the Tdr functional group were the most abundant overall (Table 2). These are plants which tolerate damp soils and are frequently found in riparian vegetation buffers and wetland areas, but do not require inundation or flooding (Casanova 2011). This explains both the high abundance (Table 2) and the high probability of occurrence (Table 5, Figure 9) of Tdr taxa observed. Both the larger size of riparian zones and experimental design mean that more non-aquatic quadrats are likely to be sampled and therefore favour the observation of terrestrial plant species. The presence of more cleared, dry soil habitats increases the probability of recruitment of terrestrial dry species. Larger terrestrial species such as Callistemon sp. and gorse were likely to have been present before flooding and survived as a result of deep root systems, and often being located at higher elevations on riverbanks, and therefore suffering shorter inundation periods and lower water velocities.

The high proportion of small terrestrial grasses is probably a result of r-selected type species which capitalise on the disturbance and cleared areas caused by flooding (Grime 1979; Casanova 2011). The short life span and

27 observation of primarily juvenile grasses indicates recent recruitment, which is supported by anecdotal evidence about the increased prevalence of exotic and invasive grasses in post flood conditions.

The group with the second highest probability of occurrence, the amphibious fluctuation tolerator emergents

(Figure 9), are those which can survive in saturated soils and low water levels (Casanova 2011). It is therefore reasonable to expect the ATe species to be observed group across all sites due to their ability to survive in very wet conditions. Tdr species will germinate in damp, water logged soils, often in the bare grounds created during disturbance events, and young members of this group appeared to be emerging (Casanova 2011). Abundance of this group is expected to be limited as water cannot be tolerated once in the vegetative state (Casanova 2011).

The one species of ARp, Hydrocotyle sp., was probably a new germinant sprouted from rhizomes and stimulated by fluctuating water levels, as it was found in an unsheltered area and therefore above ground biomass was unlikely to have survived flooding (Casanova 2011). The lack of many functional groups could be a result of many factors; unsuitable habitat conditions, the effects of a highly disturbed community, limited local species, sampling design, or more likely, the early successional state of this system.

The dominant functional group at each site is consistent with the proportions of different habitat types. Site 1 was dominated by ATe species (Figure 9), which is logical as the area included flooded and water logged soils, and minimal water for truly aquatic plants (Figure 12). Site 2 had the largest proportions of SE taxa (Figure 9), likely due to the much larger quantity of water in this location (Figure 12), while site 3 comprised mostly of Tdr species which could be explained by both the high proportion of plants at this site (Table 3), and the favourable conditions for larger terrestrial species such as larger availability of soil and land.

Although there is no baseline data for the abundance of filamentous algae, the high abundance recorded in this study is almost certainly in response to warm, nutrient rich waters, as nutrient levels are enhanced by both runoff and large volumes of leaf litter and debris entering systems during floods, and are also favoured by river modification (Bradford et al. 2007; Linden et al. 2010; Cooper et al. 2013). As algae is often found in stagnant waters, this may explain the lack of plants seen in pond water habitats (Figure 11).

The lack of a statistically significant relationship between the presence of a functional group or species and the distance from water can be expected in a recently disturbed system (Figure 10). The fundamental characteristics of each functional group involve a plant’s response to, and reliance on, water regimes, and are a determining

28 factor of presence in the absence of large scouring flows (Casanova 2011). The lack of relationship between hydrology and plant distribution could be a result of sampling error, or the dynamic nature of the system and improved by an increased sample size, period, or study reach. However, it is more likely that highly disturbed system has resulted in the absence of normally observed patterns, and that flood related factors, rather than water regime preference, are the current determinants of plant distribution. Although large flows are a natural component of river systems, the removal of small and medium floods through regulation mean that augmented effects can be expected. In order to test this theory, continued monitoring of the system could be used to determine if relationships between hydrology and distribution return over time, as the system returns to its normal low-flow state.

Although ATe and SE did not show an overall linear relationship (Figure 10 A & C), quadrats with 0% linear distance from the water (ie. water quadrats), had the highest counts (Table 6). This provides some indication that members of these functional groups are favoured by water habitats, although the additional use of vertical distance would improve accuracy of these assumptions (Casanova 2011).

The dominance of SE species in water quadrats across all habitat types is to be expected due to the water regime preferences of this group, which is usually found partly submerged in water (Casanova 2011) (Figure 11).

Unsurprisingly, terrestrial functional groups had lower occupancy as a result of the presence of water (Figure

11). The lack of plants in pool habitats could be caused by the increased water depth in these areas which prevents the growth of emergents, which must have their reproductive portions above water, the souring of in- stream flora during the flood, or bedrock substrates not conducive to plant growth (Lloyd et al. 2005)(Figure

11).

The variation in cover composition between the three sites demonstrates the heterogeneity of the system (Figure

12). The varied cover could be in part responsible for the differences in plant communities that are observed at each site, although a larger sample size would be required to support this (Figure 12 & Table 2). The presence of large woody debris provides further support for the anecdotal evidence of the strength of the flood waters, and the removal of large amounts of vegetation (Figure 12).

Vegetation Surveys – benefits and limitations of the experimental design

Benefits of the experimental design include the stratified choice of site locations which allows for data representative of the main zones of the system to be collected in lieu of repetitive sampling of similar areas, or whole river data, the collection of which is impractical due to access complications and time constraints. The random transect and quadrat generation reduces bias and allows for a broader range of areas to be calculated efficiently. The use of functional groups allows conclusions to be drawn on individual species based on their functional group classification, which is beneficial as it extends the applicability of the study beyond the species recorded here. For example, if the members of a functional group showed a particular significant response, it could reasonably be predicted that another unobserved member of the group would respond in a similar way.

The use of functional groups in large studies also reduces the labour load associated with the analysis of large data sets.

Although this study can be used to make assumptions about the impacts of the September 2016 flood events on vegetation cover, the absence of quantitative pre-flood data limits the reliability of these statements (King et al.

2015). The single sampling event (15-16/12/16) is also a limitation due to potential lag effects, which refer to the often delayed response of ecosystems to large scale ecological change (Bunn et al. 2002; Lloyd et al. 2005;

King et al. 2015). Unfortunately the position of the water at site 1 was not recorded, and the assumption of centred water may have affected predicted relationships between distance to water and probability of occurrence. Where random numbers resulted in overlapping quadrats3, the position of the next nearest non- overlapping quadrat was used, but the new position was not recorded, which also slightly impacted the accuracy of distance relationships. Although this study likely captured the response of abundant species, several subsequent sampling events are suggested as a means of monitoring lagged responses, for example the recovery of slow growing, long-lived species (Lloyd et al. 2005; King et al. 2015). As per Maxwell et al. 2015, another significant limitation is the inability of this data to predict whole plant communities, as it provides only the probability of occurrence of a given taxon. The sample size also meant that the number of observed species provided insufficient data to reveal true relationships with hydrology (Maxwell et al. 2015). As previously discussed, random sampling was possibly responsible for the apparent dominance of terrestrial dry species, as land areas are more likely to be sampled, while water areas can be missed (Maxwell et al. 2015). In addition, a 0 count for a particular distance interval may indicate the lack of a quadrat in this region, rather than the lack of a

3 Random numbers were pre-calculated and in the form of percentages due to the inability to know transect length before surveying. 30 species. In order to rectify this issue, a stratified sampling technique could be used, a larger number of quadrats, or a survey of the entire transect (King et al. 2015). The classification of functional groups should be used with caution, as the incorrect classification of certain taxa has been highlighted as an issue in previous studies

(Maxwell et al. 2015).

Although distance from the water was used as a proxy for hydrology, in order to be informative the vegetation surveys need to be conducted as surveyed cross sections in areas with good hydrological data, including vertical distances. As there is no existing hydrological data for this reach and none was collected, this is recognised as a key flaw in the experimental design and is a result of limited resources and time constraints.

Conclusion

It was found that the 2016 scouring flood caused a significant change in overall vegetation cover at Myponga

River. Vegetation cover was reduced, while water, stone, and bare soil cover increased. The absence of an observable significant relationship between presence of functional groups and distance from the water suggests the unprecedented scouring floods of September 2016 may have reset the river system, causing an absence of normally observable patterns. Rather than water regime, a plant’s ability to tolerate flooding is thought to be the main determinant of species distribution. The results of this study will act as a set of baseline data characterising the vegetation after a large disturbance. Further studies with increased sample sizes and improved experimental design should provide information about the capability, duration and modes of recovery, as well as the extent of changes caused by the flood. Continued monitoring of established sites, with the additional collection of hydrological data, is highly recommended to gain a better understanding of the impact of natural and controlled release flooding of this system.

Acknowledgements

Thank you to the landowners Jim Stacy and Lyn Zoumis for allowing us to conduct surveys on their properties, and to SA Water for granting us access to the Myponga Reservoir and surrounding catchment area. Thank you to Kate Frahn who aided in plant identification.

A huge thank you to David Schmarr for envisioning this project, assisting field work, and providing support and assistance throughout, and to Jason Nicol for his drafting and valuable feedback.

References

BARSLEY, D.K. & SWEENEY, S.M. 2010. Guiding climate change adaptation within vulnerable natural resource management systems. Environmental Management, vol. 45, pp. 1127-1141.

BRADFORD, G. FINNEY, S. HE, Y. & MANOU, M. 2008. Myponga Watercourse Restoration Project final report 2000-2007. Environment Protection Authority. ISBN: 978-1-921125-71-3.

BUNN, S.E. & ARTHINGTON, A.H. 2002. Basic Principles and Ecological Consequences of Altered Flow Regimes for Aquatic Biodiversity, Environmental Management, vol. 30, pp. 492-507.

BUREAU OF METEROLOGY 2016. Special Climate Statement 58- record September rains continue wet period in much of Australia. Australian Government. URL: http://www.bom.gov.au/climate/current/statements /scs58.pdf

CASANOVA, M.T. & BROCK, M.A. 2000. How do depth, duration and frequency of flooding influence the establishment of wetland plant communities? Plant Ecology, vol. 147, pp. 237 – 250.

CASANOVA, M.T. 2011. Using water plant functional groups to investigate environmental water requirements. Freshwater Biology, vol. 56, pp. 2637-2652.

COOPER, S.D., LAKE, S.P., SABATER, S., MELACK, J.M. & SABO, J.L. 2013. The effects of land use changes on streams and rivers in Mediterranean climates. Hydrobiologia, vol. 719, pp. 383-425.

CUNNINGHAM, G.M., MULHAM, W.E., MITHORPE, P.L. & LEIGH, J.H. 1992. Plants of Western New South Wales. CSIRO Publishing, Victoria, Australia.

FLORABASE 2017. Names, descriptions, photos and maps. Department of Parks and Wildlife, Western Australia, Australia. URL: https://florabase.dpaw.wa.gov.au/.

GRIME, J.P. 1979. Plant strategies and vegetation processes. John Wiley and Sons Ltd: Chichester.

JANSEN, A. & ROBERTSON, A.I. 2001. Relationships between livestock management and the ecological condition of riparian habitats along an Australian floodplain river. Journal of Applied Ecology, vol. 38, pp. 63-75.

JESSOP, JP., & DASHORST, G.R.M. 2006. Plants of the Adelaide Hills and Plains. Botanic Gardens of Adelaide and State Herbarium of South Australia, Australia.

KING, A.J., GAWNE, B., BEESLEY, L., KOEHN, J.D., NIELSEN, D.L. & PRICE, A. 2015. Improving Ecological Response Monitoring of Environmental Flows, Environmental Management, vol. 55, pp. 991-1005.

LINDEN, L.G., LEWIS, D.M., BURCH, M.D. & BROOKES, J.D. 2010. Interannual variability in rainfall and its impact on nutrient load and phytoplankton in Myponga Reservoir, South Australia, International Journal of River Basin Management, vol. 2, pp. 169-179.

MAXWELL, S., GREEN, D., NICOL, J., SCHMARR, D., PEETERS, L., FLEMING, N., HOLLAND, K. & OVERTON, I. 2015. Water allocation planning: environmental water requirements, GWAP Project Task 4. Goyder Institute for Water Research Technical Report Series No. 15/53.

McNEIL, D.G., FREDBERG, J.F. & WILSON, P. 2011. Coastal Fishes and Flows in the Onkaparinga and Myponga Rivers. Report to the Adelaide and Mount Lofty Ranges Natural Resource Management Board. South Australian Research and Development Institute. SARDI publication No. F2009/000410-1. SARDI Research Report Series No. 400.

McNEIL, D.G., SCHMARR, D. & MATHWIN, R. 2011. Condition of the Freshwater Fish Communities in the Adelaide and Mount Lofty Ranges Management Region. South Australian Research and Development Institute. SARDI publication No. F2011/000502-1. SARDI Research Report Series No. 590.

MILLER, K.A., WEBB, J., DE LITTLE, K.M. & STEWARDSON, M.J. 2013. Environmental flows can reduce the encroachment of terrestrial vegetation into river channels: a systematic literature review Environmental Management, vol. 52, pp. 1202-1212.

NICOL, J. 2013. Characterisation of the in stream and riparian plant communities in the Barossa Prescribed Water Resources Area Data and methods report. South Australian Research and Development Institute. SARDI Publication No. F2013/000413-1. SARDI Research Report Series No. 745.

POFF, L.N. & ZIMMERMAN, J.K.H. 2009. Ecological responses to altered flow regime: a literature review to inform the science and management of environmental flows, Freshwater Biology, vol. 55, pp. 194-205.

SAINTY, G.R. & JACOBS, S. 2003. Water plants in Australia: A Field Guide (3rd Edition), Sainty and Associates Pty Ltd, Potts Point, New South Wales, Australia.

SA WATER 2017. Myponga Reservoir. Australian Government. URL: https://www.sawater.com.au/community- and-environment/our-water-and-sewerage-systems/water-sources/reservoir-data/myponga-reservoir.

Appendices

Appendix 1. Coordinates of distal ends of transects within each 100m site, Myponga River.

Site Transect Left end coordinates Right end coordinates

1 1 Latitude: -35.3793269768 Latitude: -35.3796009813 Longitude: 138.3851109724 Longitude: 138.3856690396 1 2 Latitude: -35.3792439960 Latitude: -35.3796090279 Longitude: 138.3852080349 Longitude: 138.3857210074 1 3 Latitude: -35.3790640365 Latitude: -35.3794790246 Longitude: 138.3854579832 Longitude: 138.3858980332 2 1 Latitude: -35.3904390335 Latitude: -35.3903319966 Longitude: 138.3892640378 Longitude: 138.3894419856 2 2 Latitude: -35.3903210163 Latitude: -35.3901849780 Longitude: 138.3891259879 Longitude: 138.3893129881 2 3 Latitude: -35.3901360277 Latitude: -35.3900950402 Longitude: 138.3889130037 Longitude: 138.3890209626 3 1 Latitude:- 35.3967169952 Latitude: -35.3965279832 Lon: 138.4055299591 Lon: 138.4054550249 3 2 Latitude:-35.3968350124 Latitude: -35.3966449946

Longitude: 138.4053360019 Longitude: 138.4051640052

3 3 Latitude: -35.3969390318 Latitude: -35.3968179971 Longitude: 138.4052520152 Longitude: 138.4050530288

Appendix 2. Details of each of the additional 100m sites at Myponga River, used for cover comparisons.

Site 0m Coordinates 100 m Coordinates

4 Latitude: 35°23'37.17"S Latitude: 35°23'37.81"S Longitude: 138°23'58.69"E Longitude: 138°23'54.39"E 5 Latitude: 35°23'2.59"S Latitude: 35°22'59.59"S Longitude: 138°23'11.06"E Longitude: 138°23'12.58"E 6 Latitude: 35°23'46.50"S Latitude: 35°23'43.36"S Longitude: 138°24'5.48"E Longitude: 138°24'4.45"E 7 Latitude: 35°23'35.98"S Latitude: 35°23'34.85"S Longitude: 138°23'34.80"E Longitude: 138°23'30.99"E 8 Latitude: 35°23'13.84"S Latitude: 35°23'10.29"S Longitude: 138°23'18.30"E Longitude: 138°23'18.12"E 9 Latitude: 35°22'54.44"S Latitude: 35°22'52.06"S Longitude: 138°23'11.87"E Longitude: 138°23'9.13"E 10 Latitude: 35°22'43.76"S Latitude: 35°22'43.07"S Longitude: 138°23'14.11"E Longitude: 138°23'18.02"E

Appendix 3. Water Plant Functional Groups, as defined by Casanova, 2011. Table from Maxwell et al. 2015.

Appendix 4.

Images of the four unidentified species, Unknowns 1,2,3 and 4, with mm scale (Matthews 2016).