MURRAY-DARLING BASIN AUTHORITY Assessing the …Much speculation has surrounded the role of flooding in the spawning and recruitment of native fish in the Murray-Darling Basin. This

MURRAY-DARLING BASIN AUTHORITY

Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest

Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest

Prepared by: Alison King, Zeb Tonkin and John Mahoney Department of Sustainability and Environment

MURRAY-DARLING BASIN AUTHORITY

This contract is funded by The Living Murray and Native Fish Strategy

Published by Murray-Darling Basin Authority Postal Address GPO Box 1801, Canberra ACT 2601 Office location Level 4, 51 Allara Street, Canberra City Australian Capital Territory Telephone (02) 6279 0100 international + 61 2 6279 0100 Facsimile (02) 6248 8053 international + 61 2 6248 8053 E-Mail [email protected] Internet http://www.mdba.gov.au For further information contact the Murray-Darling Basin Authority office on (02) 6279 0100 This report may be cited as: King, A.J., Tonkin, Z. And Mahoney, J. (2007). Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest. Report to the Murray-Darling Basin Commission (now Murray-Darling Basin Authority). Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment. MDBA Publication No. 17/09 ISBN 978-1-921257-95-7 © Copyright Murray-Darling Basin Authority (MDBA), on behalf of the Commonwealth of Australia 2009 This work is copyright. With the exception of photographs, any logo or emblem, and any trademarks, the work may be stored, retrieved and reproduced in whole or in part, provided that it is not sold or used in any way for commercial benefit, and that the source and author of any material used is acknowledged. Apart from any use permitted under the Copyright Act 1968 or above, no part of this work may be reproduced by any process without prior written permission from the Commonwealth. Requests and inquiries concerning reproduction and rights should be addressed to the Commonwealth Copyright Administration, Attorney General’s Department, National Circuit, Barton ACT 2600 or posted at http://www.ag.gov.au/cca. This work was originally commissioned and produced for the Murray-Darling Basin Commission (MDBC) and contains references to the MDBC. In December 2008, the MDBC's rights and its functions were transferred to the MDBA in accordance with the Water Act 2008 (Cth). The views, opinions and conclusions expressed by the authors in this publication are not necessarily those of the MDBC, MDBA or the Commonwealth. To the extent permitted by law, the Commonwealth (including the MDBA) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this report (in part or in whole) and any information or material contained within it. Front Cover image: Hut Lake (Barmah Forest) in flood February 2005. All images by Alison King, DSE, unless otherwise stated.

iii

CONTENTS

Acknowledgments ..................................................................................................................................................................................................................................................................................v

Executive Summary ............................................................................................................................................................................................................................................................................vi

Background ................................................................................................................................................................................................................................................................................ vi

Significant Ecological Findings ........................................................................................................................................................................................................................... vi

Methodological Findings ............................................................................................................................................................................................................................................vii

Outputs of the project ....................................................................................................................................................................................................................................................vii

Conclusion ..................................................................................................................................................................................................................................................................................vii

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

River regulation and environmental flows ............................................................................................................................................................................................. 1

Barmah-Millewa Forest and the use of the Barmah-Millewa Environmental Water Allocation ........................................... 2

River regulation, environmental flows and native fish ............................................................................................................................................................ 4

Fish in the Barmah-Millewa Forest region of the Murray River ................................................................................................................................... 5

Aims of the study................................................................................................................................................................................................................................................................... 5

2. Effects of an environmental flow on the spawning and recruitment of four channel dwelling native fish in the Murray River at Barmah-Millewa Forest ..................................................................................................................................................7

Introduction ................................................................................................................................................................................................................................................................................. 7

Methods ........................................................................................................................................................................................................................................................................................... 8

Results ...........................................................................................................................................................................................................................................................................................10

Discussion ..................................................................................................................................................................................................................................................................................16

Summary of Key Findings .......................................................................................................................................................................................................................................19

3. Spatial and temporal patterns of fish spawning and recruitment in the Barmah-Millewa Forest wetland complex.........................................................................................................................................................................................................................................................................20

Introduction ..............................................................................................................................................................................................................................................................................20

Methods ........................................................................................................................................................................................................................................................................................21

Results ...........................................................................................................................................................................................................................................................................................29

Discussion ..................................................................................................................................................................................................................................................................................52


Key Research Gaps..........................................................................................................................................................................................................................................................60

4. Diel and spatial drifting patterns of eggs and larvae ...............................................................................................................................................................61

Introduction ..............................................................................................................................................................................................................................................................................61

Methods ........................................................................................................................................................................................................................................................................................61

Results and Discussion ..............................................................................................................................................................................................................................................62



5. Testing a modification to a standard passive drift net to capture drifting Ichthyofauna ..........................................................67

Introduction ..............................................................................................................................................................................................................................................................................67

Methods ........................................................................................................................................................................................................................................................................................67

Results and Discussion ..............................................................................................................................................................................................................................................68



iv

Murray–Darling Basin Authority

Closing summary .................................................................................................................................................................................................................................................................................72

References.....................................................................................................................................................................................................................................................................................................74

Appendix 1: Scientific publications and presentations .....................................................................................................................................................................81

Appendix 2: Media communications ...........................................................................................................................................................................................................................83

Appendix 3: Species list and developmental stage collected at all sites in Barmah-Millewa Forest during normal sampling .............................................................................................................................................................................................................................................................93

Appendix 4: Species list of fish collected in the Murray River boat electrofishing sampling targeted at juveniles ....................................................................................................................................................................................................................................................................................................97

Appendix 5: Results of REML post-hoc analyses from Chapter 3......................................................................................................................................98

v

ACKNOWLEDGMENTS

We would like to thank the Barmah-Millewa Forum (now the Barmah-Millewa Coordinating Committee and Barmah-Millewa Technical Advisory Committee) and the Native Fish Strategy section of the Murray-Darling Basin Commission for their financial and technical support throughout this project. We would particularly like to acknowledge the role of the Barmah-Millewa Coordinating Committee and Technical Advisory Committee and the Living Murray section of the Murray-Darling Basin Commission in the technical expertise and management of the 2005 Environmental Water Allocation. Invaluable assistance and cooperation with access to areas within the Forests was provided by Mick Caldwell and other staff at DSE Nathalia and State Forests Deniliquin. We would also like to thank Greg and Megan Gilmore from Morning Glory River Resort for being understanding and flexible with our accommodation requirements throughout the duration of the project. Initial project establishment and ongoing support for the project was provided by an expert steering committee: John Koehn, Matthew Jones, Keith Ward, Paul O’Connor, David Leslie, Amy Webb, Paul Humphries and Mark Lintermans. The majority of field and laboratory work was conducted by the authors, but we also thank John Morrongiello, Andrew Pickworth and John Koehn for their involvement with some field work. John Koehn also provided helpful comments on drafts of the report.

This study was conducted under the following permits:

• DSE Animal Care and Ethics approval permit No. AEC 03/002

• NSW DPI Scientific Collection Permit No. P03/0068-3.1

• Vic DSE Flora and Fauna Guarantee Act Permit No. 10003994

• Forests NSW Special Purposes Permit for Research No. CO33916.

vi


EXECUTIVE SUMMARY

BackgroundThe Barmah-Millewa (B-M) Forest is a highly significant wetland on the Murray River, whose natural hydrology and consequently complex ecosystems have been significantly affected by the river’s now highly regulated flow regime. In an attempt to alleviate some of these effects, the Forest’s water regime is highly managed and includes targeted environmental water allocations. Environmental flows are commonly targeted at enhancing native fish populations, through increasing spawning and successful recruitment events. However, our understanding of the relationship between flow regime and its influence on the early life of the Basin’s fishes is poorly known, and therefore predicting the response of particular environmental flow releases is difficult.

While the fish fauna of the B-M Forest region has undoubtedly changed significantly since European settlement, a large diversity and abundance of native fish still utilise the Forest region. As such, and given the existing environmental water allocation for the Forest, the B-M Forest is an obvious site to evaluate the response of fish to various flow management scenarios and to generate critical knowledge of the flow requirements of native fish. This project aimed to assess the impact of water management (particularly environmental flows) in the B-M Forest on fish spawning and recruitment. The current project was conducted from September to February in 2003–04, 2004–05 and 2005–06 seasons. In October to December 2005, 513 GL of the environmental water allocation was used at the Forest to extend the duration and slightly increase the magnitude of a natural spring flood event. The managed hydrograph resulted in fairly continuous flooding of the Forest from mid-August to mid-December.

Significant Ecological FindingsThe B-M Forest contained a high diversity of native fish, and is a significant area for native fish conservation, with a number of species of conservation significance detected. The majority of native species known to exist in the region utilised a variety of Forest habitat types for both residence and recruitment, during flood and non-flood conditions. This suggests that the diversity of the Forest’s aquatic habitat types needs to be maintained to ensure conservation of the region’s fish fauna. Abundant breeding populations of five introduced species were also recorded in the Forest.

Much speculation has surrounded the role of flooding in the spawning and recruitment of native fish in the Murray-Darling Basin. This study demonstrated that flooding can indeed influence the spawning and recruitment success of golden perch, silver perch, Murray cod and trout cod; but that the response and mechanism varies between the four species. Golden perch and silver perch increased their spawning activity in the main river channel during the flood of 2005–06 compared to the previous two seasons. Murray cod and trout cod appeared not to increase their spawning activity in the flood year, but rather increased the abundance of young-of–year resulting from the flood year compared to the previous year.

A variety of native fish species (generally smaller species) did utilise floodplain habitats to spawn and recruit, however this occurred during both flooding and non-flood seasons. Most native fish were found not to require overbank floods to stimulate spawning despite interannual variations in flow and water temperature, but many species did alter the timing and extent of their spawning period in the 2005 flood season. Counter to general predictions about the role of flooding in fish spawning and recruitment, there was no dramatic increase in the total abundance of all species or the abundance of larvae and juveniles of most native species (especially for the smaller species) associated with the 2005 flood event. The flood year did however indirectly increase the abundance of juvenile southern pygmy perch found in the Forest. The introduced species, carp, goldfish and oriental weatherloach also demonstrated an increase in recruitment strength associated with the 2005 flood event.

vii

Methodological FindingsDuring the project we also examined the diel drifting behaviour of riverine fish and investigated a new method for capturing drifting fish in slower water velocities. The intensive diel study suggested that the highest numbers of fish were drifting at night, and that the peak spawning time of silver perch occurs between 21:00 and 01:00h. The passive drift net was shown to be more effective in the capture of eggs and early larvae however the modified funnel net was more effective at the capture of larger individuals.

Outputs of the projectThe results of this project have been disseminated at numerous scientific, management and community forums, reported in to-date five peer-reviewed scientific journal articles and also in numerous regional, state and national media (TV and print).

ConclusionThis study demonstrates that the 2005 flood event (including use of the environmental flow) did achieve some positive benefits for native fish. The most obvious being the increased spawning activity in golden perch and silver perch, and the increased number of young-of-year Murray cod, trout cod and southern pygmy perch. Flooding also played an invaluable role in habitat maintenance and connectivity of floodplain habitats such as wetlands and creeks for a variety of fish residing and recruiting on the floodplain. Indirectly it also probably provided a boost of nutrients and prey items in returning waters to permanent waterbodies such as the main channel and wetlands.

We believe that this study has provided invaluable insights into the impact of water management on fish recruitment at B-M Forest, and is the first study to demonstrate a strong link between fish spawning and recruitment and the provision of an environmental flood at least in Australia and perhaps the world. However, as flow and other environmental conditions can vary substantially across years, longer-term monitoring across a range of environmental conditions and flow regimes, including managed flow events of different types, needs to occur before strong conclusions on the relationship between flooding and fish spawning and recruitment can be confidently established.

1

1. INTRODUCTION

River regulation and environmental flowsThere is little doubt that the alteration of flow regimes has led to substantial effects on the ecological sustainability and integrity of floodplain rivers throughout the world (Sparks 1995, Ward et al. 1999, Bunn and Arthington 2002). There is also an increasing awareness of the role of the natural flow regime in the ecology of floodplain rivers (Junk et al. 1989, Poff et al. 1997, Puckridge et al. 1998, Lytle and Poff 2004). Consequently, an increasing number of scientists and managers have concluded that the recovery of a more natural flow regime could provide an effective restoration strategy, potentially allowing ecosystem recovery to occur through natural recruitment and growth processes (Stanford et al. 1996, Poff et al. 1997, Rood et al. 2003, Arthington et al. 2006). The relatively new field of ‘environmental flows’ has embraced the concept, with many restoration efforts aiming to mimic components of the rivers natural flow variability, including the magnitude, frequency, timing, duration, rate of change and the predictability of flow events (Arthington et al. 2006). However, as the field is relatively new, ecologists still have much to learn about the significance of individual flow events on specific biota, and therefore long-term manipulative experiments are required (Bunn and Arthington 2002, Poff et al. 2003, Arthington et al. 2006).

The Murray River, in south-eastern Australia, is a highly regulated river system managed for multiple uses. The River has large upstream storages capturing the bulk of winter-spring flows to release for downstream consumptive use primarily in spring-summer, resulting in a seasonal reversal in flow regime and reduced flooding (Close 1990). Recently, increasing attention has been given to the ecological condition of the river (Walker and Thoms 1993, Thoms et al. 2000), and part of a suite of restoration activities being undertaken is the provision of environmental flows. (MDBC 2002a, b), (www.thelivingmurray.mdbc.gov.au).

Plate 1: Murray River near Ladgroves Beach, Barmah-Millewa Forest. May 2005.

2


Plate 2: Black Engine Billabong, Barmah Forest. February 2004.

Barmah-Millewa Forest and the use of the Barmah-Millewa Environmental Water AllocationThe Barmah-Millewa (B-M) Forest is located on the Murray River floodplain upstream of the township of Echuca, (Figure 1.1 and 1.2). In this region, the Murray is a lowland river, with a maximum channel width of 100m and has an annual average discharge of 11,250 GL (Close 1990). The Forest, dominated by River Red Gum (Eucalyptus camal dulensis), is a 70,000 ha highly complex wetland system, with a range of aquatic habitats present including rivers, permanent and ephemeral creeks, wetlands, swamps and the floodplain proper when inundated. The Forest is internationally recognised as an important wetland under the RAMSAR convention and has also received iconic status under the Murray-Darling Basin Commission’s ‘Living Murray Initiative’ (www.thelivingmurray.mdbc.gov.au).

As a result of the flow regulation of the Murray River, the Forest now experiences a reduction in the frequency, duration and inundation area of winter-spring floods, altered timing of all floods and low flow periods, increased frequency of smaller summer floods and a reduced variability in flood flows (Bren et al. 1987, Thoms et al. 2000, Chong and Ladson 2003). This massive alteration in flow regime is the major threat to the environmental values of the Forest (Ward 2005) and the Murray River (Walker and Thoms 1993). In an attempt to mitigate some of the effects of the altered flow regime on Forest ecology, the Forest’s water regime is highly managed through a series of offstream regulators and an annual environmental water allocation (EWA) specific to the site of 150 GL per year. The EWA does not have to be used within any one year, and can be accumulated for a number of years and used in larger volumes.

The B-M EWA has been used three times since its inception in 1993. The EWA was first used in 1998 when 97 GL was provided to supplement a minor spring flood. Despite a range of flora and fauna generally responding to the event (no specific monitoring of fish occurred), in general, the event was believed to achieve only some of the desired ecological objectives, as the period and depth of inundation was thought to be insufficient (Maunsell McIntyre Pty Ltd 1999). The second use of the EWA occurred in spring 2000 to January 2001, and used a total of 341 GL to extend the duration of two large spring flood events (Maunsell McIntyre Pty Ltd 2001).

The event supplemented a one in five year flood event for the Forest, and resulted in a significant waterbird breeding event (Leslie and Ward 2002) and vegetation responses. Again, little targeted monitoring of fish populations or spawning responses was undertaken.

3

1. Introduction

Figure 1.1: Location of Barmah-Millewa Forest on the Murray River, Australia.

Figure 1.2: Map of Barmah-Millewa Forest showing sites used in this study (asterix). R = river, Ck = creek, Lk = lake, Sw = swamp and Bb = billabong.

4


The B-M EWA was again used in October to December 2005, when 513 GL was used to extend the duration of floodplain inundation of a spring flood event (Figure 1.3). A team of multidisciplinary members from various Government agencies actively managed the use and shape of the hydrograph of the EWA. The managed hydrograph resulted in fairly continuous flooding of the Forest from mid-August to mid-December. This included several naturally driven flood spikes from upstream tributary inflows, and mirrored the modelled natural decline in river discharge that would have occurred. Despite approximately 50% of the Forest being inundated, the height of the flood peaks was significantly lower than the 2000/01 EWA use. Previous uses of the EWA were principally targeted at maintaining suitable breeding conditions for colonially-nesting waterbirds and vegetation responses (Ward 2005); however, the 2005 EWA was also aimed at enhancing spawning and recruitment events for native fish species. This was achieved by planning to incorporate peaks in the hydrograph during the floodplain inundation (although this was subsequently naturally achieved via upstream tributary inputs) and also maintaining floodplain inundation for 1-2 months. This approach to achieving benefits for native fish spawning was somewhat speculative given our lack of biological knowledge, but was incorporated into initial management of the EWA as an experimental component that could be validated using the current project.

River regulation, environmental flows and native fishOne of the most obvious, and often to the general community, most unacceptable effects of river regulation is the collapse of the riverine fish community. The effects of river regulation have been implicated in the decline in abundance and distribution of native fishes of the Murray-Darling Basin, Australia (Cadwallader 1978, Walker and Thoms 1993, Gehrke et al. 1995, MDBC 2004). Indeed, aspects of the flow regime are linked to critical components of the life history strategies of riverine fishes, including pre-spawning condition and maturation, spawning cues and behaviour, larval and juvenile survival, movements, and subsequent recruitment (Welcomme 1985, Junk et al. 1989, Humphries et al. 1999, Poff et al. 2003, Lytle and Poff 2004). As successful spawning and the survival of the early life history stages of fish often dictates the strength of the subsequent cohorts (Trippel and Chambers 1997), understanding the relationship between the natural flow regime and its influence on the early life of fishes is vitally important to managing fish populations in flow-altered rivers but is generally poorly known (eg. Humphries et al. 1999, Marchetti and Moyle 2001, King et al. 2003, Balcombe et al. 2006). Environmental flows, whilst broadly used to improve the ecological health of the system, may also target specific biota including enhancing native fish populations, through increasing spawning and successful recruitment events. However, given the general lack of understanding about early life history requirements of Murray-Darling fish species (Humphries et al. 1999), predicting the response of these environmental flows is difficult.

Figure 1.3: Actual mean daily discharge (solid line); simulated mean daily discharge without use of Environmental Water Allocation (EWA) (dotted line); simulated natural flows (dashed line) during the 2005 use of the Barmah-Millewa EWA. Flows above 10,000 ML (straight solid line) represent floodplain inundation. All flows measured or simulated downstream of Yarrawonga on Murray River (upstream of Barmah-Millewa). Simulated natural flows refers to results of modelled flows based on tributary inputs and no upstream river regulation. Data supplied courtesy of Damien Green, Murray-Darling Basin Commission.

5

1. Introduction

Fish in the Barmah-Millewa Forest region of the Murray RiverEarly accounts of the B-M region suggest that native fish were both diverse and highly abundant (see King 2005). Fish were an important component of the diet of the local Aboriginals. The explorer Charles Sturt, travelling through the region in 1838, noted a group of Aborigines fishing with “astonishing success” (Sturt 1838 cited in Leslie 1995) and Curr (1965) reported that “…the supply of their favourite food existed in such abundance, and was so easily procured…”.

These bountiful resources that were such an important food source and cultural activity for local Aboriginals were soon exploited by European settlers. In 1859, Joseph Waldo Rice (Rice’s Weir on Broken Creek is named in his honour) established what was probably the first commercial inland fishing enterprise in Australia, largely centred at Moira Lake (Leslie 1995). The Lake Moira Fishing Company or Murray River Fishing Company (as it was later known) supplied fish to markets principally in Bendigo and Melbourne. Leslie (1995) suggested that this catch would have equated to around 160 tonnes of fish per year (or 32,000 individuals weighing an average 5kg) being removed from the lakes and rivers of the region. In 1883, the first official records from Moama show that around 150 tonnes of fish per year were transported to Melbourne markets. Until the 1860s, most fishing was centered on Moira Lake, with catches dominated by Murray cod, but also including “bream, perch and carp1 (Illustrated Australian News, 22 March 1869). However, by the 1890s the total catch taken to market had declined significantly to around only 35 tonnes per year, and was dominated by golden perch (Leslie 1995).

Catches of native fish soon began to decline, and concern over the decline in native fish numbers in the B-M region were reported as early as the mid 1860s (Leslie 1995), and more broadly throughout inland NSW by the 1880s (Rowland 1989). Today, the B-M region of the River Murray no longer supports a native commercial fishery and recreational fishing opportunities have also declined. Ten of the 18 native species recorded (or likely) from the Forest are currently listed under either State or Commonwealth threatened species legislation, and the fish of the region are also included in the listings of endangered ecological communities for both Victoria and NSW. Although the fish fauna has undoubtedly been significantly reduced in diversity and abundance since early European settlement, a large diversity of native species still utilise the Forest (McKinnon 1997, Stuart and Jones 2002, Jones and Stuart 2004, King 2005) (Table 1.1).

Aims of the studyThe B-M Forest is an obvious site to evaluate the response of fish to various flows (particularly environmental flows) and to generate critical knowledge of the flow requirements of native fish. This study aimed to assess the impact of water management (particularly environmental flows) in the B-M region on fish breeding and recruitment.

Specifically, this project aimed to:

• Describe the distribution, timing and abundance of larval fish communities in B-M

• Determine approximate spawning periods and peak spawning times

• Establish the importance of a range of off-channel habitat types as nursery habitats

• Determine if flow conditions influence spawning triggers and/or survival of larvae and juveniles

• improve our understanding of the importance of floodplain inundation and habitats for native fish recruitment.

• If appropriate, aid in the modification of existing environmental watering strategies and management to optimise native fish recruitment

Whilst the project was focussed on B-M Forest, the information gathered should be generally applicable throughout the Murray Valley and other Murray-Darling Basin floodplain wetlands, and will aid in establishing the most appropriate environmental watering strategies for fish in these systems. The following chapters 2–5 represent reporting of separate components of the project which additionally, will or have been published as individual peer-reviewed journal articles.

1 Bream most likely silver perch, perch probably golden perch, and carp most likely goldfish (Carassius auratus) not Cyprinus carpio as the latter was not recorded in the region until the late 1970s (Koehn, J. D., Brumley, A., and Gehrke, P.C. 2000. Managing the Impacts of Carp. Bureau of Rural Sciences, Department of Agriculture, Fisheries and Forestry, Canberra, Australia. 249 pp.).

6


Table 1.1: Status of fish of the Barmah-Millewa Forest region, reproduced from (King 2005). Last record and relative abundance obtained from recent surveys, historical information in text or probable distribution. Conservation status as listed under NSW Fisheries Management Act 1994, Flora and Fauna Guarantee Act and conservation status DSE (2003), and Environment Protection and Biodiversity Conservation Act 1999. DD = data deficient, V= vulnerable, EP = endangered population, E = endangered, CE = critically endangered, RE = regionally extinct.

Common name Scientific nameLastRecord

RelativeAbundance

Conservation status

NSW Vic National

Native species

Murray cod Maccullochella peelii peelii Recent Common E V

trout cod Maccullochella macquariensis Recent Present E CE E

golden perch Macquaria ambigua Recent Common

silver perch Bidyanus bidyanus Recent Common V CE

freshwater catfish Tandanus tandanus Recent Rare E w

bony herring Nematalosa erebi Recent Present

river blackfish Gadopsis marmoratus Recent Rare

short-headed lamprey Mordacia mordax Recent Rare

Macquarie perch Macquaria australasica 1940s Probably locally extinct

V E E

Murray-Darling rainbowfish

Melanotaenia fluviatilis Recent Rare DD

Murray hardyhead Craterocephalus fluviatilis Recent Rare E CE V

unspecked hardyhead Craterocephalus stercusmuscarum fulvus

Recent Common DD

Australian smelt Retropinna semoni Recent Common

carp gudgeons Hypseleotris spp. Recent Common

flat-headed gudgeon Philypnodon grandiceps Recent Common

southern pygmy perch

Nannoperca australis Recent Rare V

southern purple-spotted gudgeon

Mogurnda adspersa No record, likely to occur

Probably locally extinct

EP RE

flat-head galaxia Galaxias rostratus No record, likely to occur

Rare

climbing galaxias # Galaxias brevipinnis Recent Rare

Introduced species

carp Cyprinus carpio Recent Common

goldfish Carasius auratus Recent Common

oriental weatherloach Misgurnus anguillicaudatus Recent Common

redfin perch Perca fluviatilis Recent Common

gambusia Gambusia holbrooki Recent Common

brown trout Salmo trutta Recent Rare

rainbow trout Oncorhynchus mykiss Recent Rare

tench Tinca tinca Recent Rare

# Climbing galaxias were recorded in Barmah Forest in 1991 by McKinnon (1997), and is thought to be a translocated native species from coastal streams, having probably emigrated via water transfers from the Snowy Mountain Hydroelectric scheme (Waters et al. 2002).

7

2. EFFECTS OF AN ENVIRONMENTAL FLOW ON THE SPAWNING AND RECRUITMENT OF FOUR CHANNEL DWELLING NATIVE FISH IN THE MURRAY RIVER AT BARMAH-MILLEWA FOREST

IntroductionRivers throughout the world support complex and highly diverse ecosystems. However, rivers also support human life and activities, and consequently many have been dammed and the water diverted for offstream purposes (Dynesius and Nilsson 1994). There is little doubt that the alteration of flow regimes has led to substantial effects on the ecological sustainability and integrity of floodplain rivers throughout the world (Sparks 1995, Ward et al. 1999, Bunn and Arthington 2002). There is also an increasing awareness of the role of the natural flow regime in the ecology of floodplain rivers (Junk et al. 1989, Poff et al. 1997, Puckridge et al. 1998, Lytle and Poff 2004). Consequently, an increasing number of scientists and managers have concluded that the recovery of a more natural flow regime could provide an effective restoration strategy, potentially allowing ecosystem recovery to occur through natural recruitment and growth processes (Stanford et al. 1996, Poff et al. 1997, Rood et al. 2003, Arthington et al. 2006). The relatively new field of ‘environmental flows’ has embraced the concept, with many restoration efforts aiming to mimic components of the river’s natural flow variability, including the magnitude, frequency, timing, duration, rate of change and the predictability of flow events (Arthington et al. 2006). However, as the field is relatively new, ecologists still have much to learn about the significance of individual flow events on specific biota, and therefore long-term manipulative experiments are required (Bunn and Arthington 2002, Poff et al. 2003, Arthington et al. 2006).

One of the most obvious, and often to the general community, most unacceptable effects of river regulation is the collapse of the riverine fish community. Indeed, aspects of the flow regime are linked to critical components of the life history strategies of riverine fishes, including pre-spawning condition and maturation, spawning cues and behaviour, larval and juvenile survival, movements, and subsequent recruitment (Welcomme 1985, Junk et al. 1989, Humphries et al. 1999, Poff et al. 2003, Lytle and Poff 2004). As successful spawning and the survival of the early life history stages of fish often dictates the strength of the subsequent cohorts (Trippel and Chambers 1997), understanding the relationship between the natural flow regime and its influence on the early life of fishes is vitally important to managing fish populations in flow-altered rivers but is generally poorly known (eg. Humphries et al. 1999, Marchetti and Moyle 2001, King et al. 2003, Balcombe et al. 2006).

The effects of river regulation have been implicated in the decline in abundance and distribution of native fishes of the Murray-Darling Basin, Australia (Cadwallader 1978, Walker and Thoms 1993, Gehrke et al. 1995, MDBC 2004). The Murray River is a highly regulated river system managed for multiple uses. The River has large upstream storages capturing the bulk of winter-spring flows to release for downstream consumptive use primarily in spring-summer, resulting in a seasonal reversal in flow regime and reduced flooding (Close 1990). Recently, increasing attention has been given to the ecological condition of the river (Walker and Thoms 1993, Thoms et al. 2000), and part of a suite of restoration activities being undertaken is the provision of environmental flows (MDBC 2002a, b), (www.thelivingmurray.mdbc.gov.au). These environmental flows whilst generally used to improve the ecological health of the system, may also target specific biota including enhancing native fish populations, through increasing spawning and successful recruitment events. However, given the general lack of understanding about early life history requirements of Murray-Darling fish species (Humphries et al. 1999), predicting the response of these environmental flows is difficult.

This study reports on the results to date from a three year study on the effects of water management on the spawning and recruitment dynamics of four native fish species in the mid-Murray River system. Fortuitously, whilst two of these years were hydrologically similar, with fully regulated in-channel flow conditions, the third year (2005–06) encompassed an extensive period of floodplain inundation, including the use of the largest environmental flow allocation to date in Australia to the Barmah-Millewa (B-M) Forest.

8


Methods

Study area and hydrology

See Chapter 1 for description of study area and general hydrology

Four target species investigated

This study focussed on four native fish species: golden perch (Macquaria ambigua), silver perch (Bidyanus bidyanus), Murray cod (Maccullochella peelii peelii), trout cod (Maccullochella macquariensis); that are known to have important recreational, conservation and cultural significance. This study was part of a broader investigation into the effects of water management on spawning and recruitment of the entire fish community at Barmah-Millewa Forest (see Chapter 3).

The four species are thought to exhibit two totally different recruitment strategies (see Humphries et al. 1999) particularly in response to floods. Following the principles of the flood pulse concept (Junk et al. 1989). Harris and Gehrke (1994) proposed that flooding enhances recruitment in Murray-Darling fish via two mechanisms: either through the initiation of spawning; or by indirectly enhancing larval and juvenile survival through the provision of abundant food and habitat. The first mechanism was attributed to species such as golden perch and silver perch, based on limited evidence of a link between floods and spawning from aquaculture studies (Lake 1967a), and various anecdotal accounts. More recently the widespread applicability of this relationship has been questioned (Humphries et al. 1999, King et al. 2003), with evidence also emerging of golden perch and silver perch recruitment being able to occur during within channel flows (Mallen-Cooper and Stuart 2003), and no field evidence to support flooding to initiate spawning. The second pathway was proposed for species such as Murray cod and trout cod, which do not require floods to spawn (Humphries 2005, Koehn and Harrington 2006), but may benefit from improved environmental conditions to sustain young during floods. Although the data is very limited, there is some evidence from commercial catch records to suggest that strong recruitment can occur in Murray cod populations following years of high flows (Rowland 1998, Ye et al. 2000).

Collection and processing of fish eggs and larvae

Fish eggs and larvae of the four target species were collected from three sites in the Murray River in the B-M Forest region (Ladgroves Beach, 35°51.677, 145°20.773; Barmah Choke, 35°54.947, 144°57.267; Morning Glory, 36°04.765, 144°57.553) (Figure 1.1). Sampling was conducted overnight, once in mid-September, and fortnightly thereafter until the end of February during 2003–04, 2004–05 and 2005–06 breeding seasons. Due to the occurrence of a late February natural flood pulse, two additional fortnightly sampling trips were also conducted in March 2005. Sampling was conducted using passive drift nets, as the eggs and/or larvae of the four target species are known to exhibit a drifting dispersal phase (Humphries and King 2004). All nets were set on dusk and retrieved as early as possible the following morning, generally before 11:00 hours.

Drift nets were 1.5 m long, with a 0.5 m diameter mouth opening and were constructed of 500µm mesh, which tapered to a removable collection jar. A General Oceanics Inc. (Florida, USA) flow meter was fixed in the mouth of each drift net to determine the volume of water filtered, therefore, enabling raw catch data to be adjusted to a standard volume of filtered water (1000 m3). At each site, two drift nets were attached to a pole to sample the surface and bottom 50 cm of the water column (Figure 2.1). In the field, eggs were removed alive from the samples and returned to the laboratory to hatch, to enable correct identification. Remaining samples were preserved in 95% ethanol in the field and returned to the laboratory for processing, where fish were removed from the samples using a dissecting microscope. Identifications were made by experienced staff using available keys (Serafini and Humphries 2004), and by collating a reference collection of successive larval stages. The presence of eggs or larvae was used as an indication of spawning occurrence. Data for eggs and larval catches were adjusted to a standard volume of water filtered (number of eggs/larvae per 1000 m3), and the data was then pooled across both net position (top or bottom) and across the three river sites. Differences between years were tested using Kruskal Wallis analysis log10 (x+1) transformed total raw numbers of eggs and larvae of all species, using data from all sites and time periods where they were present. If significant, Mann-Whitney tests were used to identify pair-wise differences.

9

2. Spawning and recruitment of four channel dwelling native fish at Barmah-Millewa Forest

Figure 2.1: Schematic of top and bottom paired drift net sampling conducted at river sites.

Plate 3: Drift nets set in position in the Murray River. Note only top net visible.

10


Collection of Young-of-Year

Standardised boat electrofishing surveys were conducted in May of 2005 and 2006 at all three sites, after irrigation flows in the main river had dropped to very low managed winter baseflows. Sampling was not able to be conducted during May 2004, due to a very small period of time available when river levels were low and therefore suitable for sampling. The 5.2m electrofishing boats were equipped with on-board 7.5 Kva Smith-Root Model GPP 7.5 H/L electrofishing systems. The electrofisher usually operated at 1000 V DC, 7.5 amps pulsed at 120 Hz and 35% duty cycle. Within each site, sampling was conducted for 1080 seconds (electrofishing time-on) at each of three randomly assigned inner bend, outer bend and straight reaches. Inner bend reaches occurred at large bends in the river and occurred in the mostly slower flowing, depositional half of the river, and were often associated with beach habitats. Outer bend reaches again occurred at large bends, but occurred in the generally faster flowing, actively eroding zone of the river and were often associated with high steep banks and deeper water. Straight reaches were reaches of the river where no bends occurred, although sampling was restricted to one side of the river only for each replicate. Each replicate reach was then divided into three zones, inner (closest to the bank), middle and outer (middle of the river), to allow a standardised even effort within all available habitats within each reach. Sampling was conducted during the day and then repeated the following night at each of the replicate reaches, leaving at least four hours between each sampling run. All fish captured were weighed and measured (standard length) and then released. Individuals were classified as young-of-year based on their length, as <150 mm for Murray cod and trout cod, and <100 mm for golden perch and silver perch; these lengths were chosen based on otolith ageing of a small number of individuals (unpub. data). Data for the young-of-year sampling was pooled across day and night samples and across reaches, and grouped by ‘season’ and ‘site’. Kruskal-Wallis tests were performed on each species with ‘season’ as the factor.

Results

Hydrographs

The hydrographs of the three years of this study were remarkably different each year (Figure 2.2a). The 2003–04 season, was marked by three winter and early spring floods that inundated the Forest, and continued as fairly stable, bank-full conditions for the remainder of the season. A very minor flood event occurred in late December 2003, which inundated very restricted areas of the Forest. The 2004–05 season, was characterised by one spring flood event, and again remained fairly stable, bank-full conditions for the rest of the season. A minor, natural flood event which inundated low-lying areas of the Forest, occurred in late February 2005. This flood caused a minor blackwater event into the Murray River, but no fish kills or adverse environmental effects were observed (King, pers. obs.). The February flood was also associated with a marked decline in temperature. The 2005–06 season hydrograph was obviously distinct from the two previous two seasons, with floodplain inundation occurring for nearly five months from early August to late December. The flood had a number of peaks, which were principally driven by upstream inflows.

Occurrence of eggs and larvae

While golden perch (eggs and larvae), silver perch (eggs and larvae) and Murray cod (larvae only) were collected in all three years; trout cod larvae were only collected in 2005–06 (Table 2.1). The occurrence and total abundance of the four species varied among the three sites sampled.

Total raw abundance and adjusted abundance of golden perch and silver perch eggs and larvae increased by one to two orders of magnitude in 2005–06 compared to the first two sampling seasons (Table 2.1). Whilst there was a significantly greater number of silver perch eggs captured in 2005–06 compared to the previous seasons (2005–06>2004–05: p<0.05; 2005–06>2003–04: p<0.01; 2003–04=2004–05), there was no significant difference in the number of silver perch larvae amongst years (p=0.072). There was no significant difference in the number of golden perch eggs captured amongst years (p=0.110), but there was a significantly greater number of golden perch larvae captured in 2005–06 compared to 2003–04 (p<0.05) and 2004–05 (p<0.05). The larvae of both golden perch and silver perch that were collected were very newly hatched larvae, approximately up to 2 days old (King, unpub. data), and therefore the presence of eggs or larvae of these two species can be used to indicate the spawning time of these species. The timing of silver perch spawning was fairly consistent each year, occurring between early November to mid-February, and samples were consistently collected in water temperatures above 20ºC, ranging from 17.2ºC to 28.4ºC (Figure 2.2b). By far the greatest abundance of silver perch eggs were collected in the early November and late December 2005 sampling events. The first major spawning event in November occurred on a coinciding rise in water temperature and rise in water level

11


Figu

re 2

.2: (

a) M

ean

daily

dis

char

ge (s

olid

line

), w

ater

tem

pera

ture

(das

hed

line)

and

adj

uste

d to

tal a

bund

ance

of (

b) s

ilver

per

ch a

nd (c

) gol

den

perc

h eg

gs p

er

1000

m-3

col

lect

ed in

dri

ft n

ets

in th

e M

urra

y R

iver

. Dat

a po

oled

for

thre

e si

tes

and

net p

ositi

on. D

isch

arge

and

tem

pera

ture

dat

a fr

om T

ocum

wal

gau

ge o

n M

urra

y R

iver

. Not

e: d

otte

d lin

e re

pres

ents

app

roxi

mat

e fl

oodp

lain

inun

datio

n he

ight

at B

arm

ah-M

illew

a Fo

rest

, sol

id tr

iang

les

on x

-axi

s in

dica

tes

sam

plin

g ev

ent.

Lett

ers

on x

-axi

s re

pres

ent c

alen

dar

mon

ths

in e

ach

year

from

Jul

y to

Mar

ch.

12


during the extended flood conditions, while the second major spawning event in late December coincided with a rapid decline in water level, as flood waters were receding from the Forest (Figure 2.2a & b). The spawning time of golden perch was much shorter, and occurred earlier in all three years compared to silver perch (Figure 2.2c). The earliest golden perch spawned was from early October to mid-December, with water temperatures ranging from 16.9ºC to 24.7ºC. However, the spawning period did vary between the three years. A large golden perch spawning event occurred in early November 2003. However, as with silver perch, by far the greatest abundance of golden perch eggs were collected in the 2005–06 season, on the early November 2005 sampling event. This spawning event occurred on a coinciding rise in water level and temperature during the extended flood conditions (Figure 2.2c).

Table 2.1: Total raw abundance of eggs and larvae of golden perch, silver perch, Murray cod, trout cod and unidentified cod larvae collected during 2003–04, 2004–05 and 2005–06 breeding seasons across the three collection sites.

golden perch silver perch Murray cod trout cod Unid. cod arvae

Eggs Larvae Eggs Larvae Larvae Larvae Larvae

2003–04

Ladgroves Beach 20 0 134 29 9 0 0

Barmah Choke 143 1 219 10 1 0 0

Morning Glory 0 0 0 0 0 0 0

Total raw number 163 1 353 39 10 0 0

2004–05

Ladgroves Beach 14 2 469 2 20 0 0

Barmah Choke 69 0 65 0 12 0 0

Morning Glory 26 0 85 0 3 0 0

Total raw number 109 2 619 2 35 0 0

2005–06

Ladgroves Beach 291 89 1682 176 45 1 3

Barmah Choke 656 14 1069 18 24 2 2

Morning Glory 659 7 843 1 12 1 1

Total raw number 1606 110 3594 195 81 4 6

Total raw number 1878 113 4566 236 126 4 6

Plate 4: Golden perch and silver perch eggs collected in drift nets in the Murray River, 4th November 2005.

13


Murray cod larvae were consistently collected from early November to mid-December in all three years, in water temperatures ranging from 18.6ºC to 24.8ºC (Figure 2.3b). The larvae were collected in similar abundances among the three years and over a wide range of flow conditions (Figure 2.3b), with significantly fewer individuals captured in 2003–04 than in both of the following seasons (p<0.05). Trout cod larvae were only collected in the 2005–06 sampling season and therefore it is not appropriate to draw conclusions from these numbers (Figure 2.3c). The few individuals that were collected, were collected at the same time as Murray cod (Figure 2.3c).

Abundance of young-of-year fish

In total, five golden perch, zero silver perch, 30 Murray cod and 71 trout cod young-of-year were collected during the two sampling years (Table 2.2). The low abundance of golden perch and absence of silver perch young-of-year may due be due to a suspected sampling inefficiency in the collection of smaller individuals of these two species, and it is hoped sampling in subsequent years may further elucidate the success, or otherwise, of the 2005–06 cohort of fish compared to other years. Significantly greater numbers of young-of-year Murray cod and trout cod were captured in 2006 compared to 2005 (Murray cod: Figure 2.4a, p<0.05; trout cod: Figure2.4b, p<0.05).

Table 2.2: Total catch per unit effort of young-of-year golden perch (Macquaria ambigua), silver perch (Bidyanus bidyanus), Murray cod (Maccullochella peelii peelii) and trout cod (Maccullochella macquariensis) collected during 2005 and 2006 sampling events across the three collection sites.

golden perch silver perch Murray cod trout cod

2005

Ladgroves Beach 0 0 5 2

Barmah Choke 1 0 0 1

Morning Glory 2 0 0 1

Total CPUE 3 0 5 4

2006

Ladgroves Beach 0 0 15 57

Barmah Choke 2 0 5 9

Morning Glory 0 0 5 1

Total CPUE 2 0 25 67

Total CPUE 5 0 30 71

14


Figu

re 2

.3: (

a) M

ean

daily

dis

char

ge (s

olid

line

), w

ater

tem

pera

ture

(das

hed

line)

and

adj

uste

d to

tal a

bund

ance

of (

b) M

urra

y co

d an

d (c

) tro

ut c

od la

rvae

per

10

00m

-3 c

olle

cted

in d

rift

net

s in

the

Mur

ray

Riv

er in

200

3/04

, 200

4/05

and

200

5/06

. Dat

a po

oled

for

thre

e si

tes

and

net p

ositi

on. D

isch

arge

and

tem

pera

ture

dat

a fr

om T

ocum

wal

gau

ge o

n M

urra

y R

iver

. Not

e: d

otte

d lin

e re

pres

ents

app

roxi

mat

e fl

oodp

lain

inun

datio

n he

ight

at B

arm

ah-M

illew

a Fo

rest

, sol

id tr

iang

les

on x

-axi

s in

dica

tes

sam

plin

g ev

ent.

Lett

ers

on x

-axi

s re

pres

ent c

alen

dar

mon

ths

in e

ach

year

from

Jul

y to

Mar

ch.

15


Figure 2.4: Mean (+SE) CPUE of young-of-year (a) Murray cod and (b) trout cod collected using standardised boat electrofishing sampling in 2005 and 2006.

Plate 5: Young-of –year trout cod (top) and golden perch (bottom) captured May 2006.

16


Discussion

The role of flooding in spawning and recruitment of native fish

Much speculation has surrounded the role of flooding in the spawning and recruitment of native fish in the Murray-Darling Basin (Harris and Gehrke 1994, Humphries et al. 1999, King et al. 2003). Harris and Gehrke (1994), following the tenets of the Flood Pulse Concept (Junk et al. 1989) proposed two possible mechanisms about how floods could enhance recruitment of native fish, suggesting that fish could either spawn in response to floods or that flooding indirectly increases survival of the young by providing suitable food and habitat resources on the inundated floodplain. However, evidence supporting the use of floodplains by Murray-Darling fish and the role of floods in spawning and recruitment has been limited, and the widespread applicability of this relationship has been questioned (Humphries et al. 1999, King et al. 2003, Mallen-Cooper and Stuart 2003). The present study has shown that flooding can indeed influence the spawning and recruitment success of golden perch, silver perch, Murray cod and trout cod; but that the response and mechanism varies between the four species.

Both golden and silver perch are commonly referred to as flood recruitment specialists (Lake 1967a, Harris and Gehrke 1994, Schiller and Harris 2001), however as discussed by various authors the evidence supporting this is sketchy and inconclusive (Humphries et al. 1999, King et al. 2003). Lake (1967a), using results from constructed ponds, suggested that these two species were stimulated to spawn by increasing water levels and inundating dry ground when water temperatures exceeded 23ºC. Although Lake (1967a) is the first attributed to suggest a flood recruitment model for these species, Mallen-Cooper and Stuart (2003) suggest that Lake’s data also supports the proposal that rises within the river channel at specific temperatures, might be enough to stimulate spawning. Mallen-Cooper and Stuart (2003) presented age verified year-class strength data of the two species from three years of sampling fish moving through the Torrumbarry weir fishway. They concluded that golden perch recruitment was high in non-flood years and poor in flood years, and silver perch recruited in all flow years; supporting a non-flood recruitment model for both of these species. However, a significant problem with their conclusions is the unknown location of spawning and nursery site/s of the fish captured, making it difficult to confidently correlate year-class strength and the timing of spawning to flow conditions experienced by the fish. The present study demonstrates that golden perch and silver perch can spawn at low levels during within channel flow conditions (albeit regulated, fairly stable, irrigation flows) during spring and summer; but that a massive increase in spawning intensity occurred during the 2005 flood conditions. The first major spawning event for both silver perch and golden perch appeared to occur during a coinciding rise in water temperature (of 2.5°C in the previous 7 days) and on a rise in water level in early November 2005, similar to that proposed by Lake (1967a). Silver perch also demonstrated a second increase in spawning activity in late December 2005, as flows were slowly declining and the Forest was draining. However, as this was only one flow event in one year, it is difficult to confidently determine exactly what environmental cue triggered the increased spawning event. Although the increase in spawning of both of these species is likely to have resulted in an increase in subsequent recruitment, unfortunately at present this study cannot shed light on the intensity of recruitment during these different flow years, and conclusions about the overall success of the spawning event and subsequent recruitment cannot be made at this time.

Whilst golden perch and silver perch were stimulated to increase spawning activity during flood conditions, Murray cod and trout cod spawning intensity (inferred from catches of drifting larvae) was relatively similar between the flood year and the two previous breeding seasons. This finding supports previous studies that have demonstrated that they can spawn irrespective of flow conditions in both regulated and unregulated rivers, and that spawning is not enhanced or triggered by flooding (Humphries et al. 2002, Gilligan and Schiller 2003, King et al. 2003, Humphries 2005, Koehn and Harrington 2006). Again, in concert with these other studies, Murray cod and trout cod larvae were collected during November and December across all three years of this study, suggesting that the timing of spawning is more associated with day length or some other consistent interannual variable than flow (Humphries 2005, Koehn and Harrington 2006). However, while spawning intensity did not increase as a response to flooding, there was a massive increase in catch per unit effort of young-of-the-year of both species after the 2005–06 flood conditions compared to that in the 2004–05 season. This suggests that the survival and subsequent recruitment of their young was greater during the flood conditions. Strong year classes of Murray cod following breeding seasons that have experienced high flows or floods have been reported from analysing commercial catch records (Rowland 1998, Ye et al. 2000), but these data have not been age validated and have to be treated with some caution.

17


Plate 6: Live golden perch eggs (left) and 2 day old preserved golden perch larvae (top right) and silver perch larvae (bottom right).

Plate 7: Preserved Murray cod larvae.

The Flood Pulse Concept (Junk et al. 1989) suggested that the inundated floodplain environment provides a suitable spawning habitat and/or high densities of food and habitat for increasing the survival of young fish. Indeed, rises in flow have been thought to provide essential cues for upstream movements in these fish species (Reynolds 1983) where the fish then move laterally onto inundated floodplains to spawn or spawn in-channel where their drifting eggs and/or larvae are washed into inundated nursery habitats on the floodplain (Lake 1967a). However, despite large floodplain areas of the Barmah-Millewa Forest wetland complex being accessible to all stages of fish, no golden perch and silver perch eggs or larvae, and low numbers of Murray cod and trout cod larvae were collected during intensive sampling of the Forest habitats over the three years of the study (see Chapter 3). Instead the early life stages of these four species were either solely or mostly captured in the river. This suggests that the food resources that their young relied upon to grow and survive were generated on the inundated floodplain environment and then transported to the main river channel where they could access them. This emphasises the importance of undisrupted lateral connectivity of floodplains and river ecosystems (Junk et al. 1989).

In summary, this study has demonstrated that golden perch and silver perch are quite flexible in their spawning requirements and can spawn under flood and within channel flows; however, their spawning activity is increased during flood conditions. In contrast, Murray cod spawning activity is not influenced at all by flow conditions, but that recruitment of both Murray cod and trout cod can be substantially increased when floodplain inundation coincides with the presence of their larvae and juveniles.

18


Can flooding using environmental flows be used to improve riverine fish populations?

With much of the world’s riverine fish faunas in severe decline, managers are increasingly looking to improve or restore fish populations through a variety of restoration activities. Aspects of the flow regime are known to be linked to critical aspects of the life history of fishes (Welcomme 1985, Junk et al. 1989, Humphries et al. 1999, Poff et al. 2003, Lytle and Poff 2004), and therefore providing more natural flow regimes in regulated rivers is seen as one of these potential restoration measures (Marchetti and Moyle 2001, Arthington et al. 2006). However, examples of its use are limited and those that are available have had varying successes. For example, Travnichek et al. (1995) and Freeman et al. (2001) demonstrated some recovery in fish populations in hydropeaking reaches when stable minimum flows were returned. However, a test flood in the Colorado River downstream of Glen Canyon Dam had little effect on the distribution, abundance or movement of native fish, and only short-term effects on the densities of some non-native species, suggesting that the flood was of insufficient magnitude to result in a response (Valdez et al. 2001). Conversely, the importance of the appropriate timing of rising flows or floods has been shown to be a strong spawning cue for some species (Nesler et al. 1988, King et al. 1998). King et al. (1998) studied the spawning response of an endangered cyprinid, the Clanwilliam yellowfish (Barbus capensis), in the Olifants River, South Africa, to experimental flow releases from a dam. They concluded that spawning success should increase following flow releases from the dam if these releases were delivered at the appropriate time and water temperatures were suitable. These examples all highlight the need to consider various key attributes of the natural flow regime in a river system to manage flows for improved fish populations.

King et al. (2003) suggested a range of environmental conditions would need to occur to allow successful spawning and recruitment of fish during flood conditions, including: (1) a coupling of high flows and temperatures, (2) the flood pulse needs to be predictable for that system and the fishes within it, (3) the rates of rise and fall need to be slow, (4) the duration of the inundation period needs to be in the order of months and (5) that a large proportion of the floodplain needs to be inundated. The 2005 flood at B-M Forest met all of these requirements; and therefore perhaps the successful spawning and recruitment observed in these species and smaller species (see Chapter 3) may not have occurred if the B-M EWA had not been used. Indeed, the use of the EWA not only increased the magnitude of the flood peaks, but also allowed the duration of the effective inundation period to be extended for an additional two months through October and December without disconnection of the floodplain (Figure 1.3). Importantly this use of the environmental water coincided with natural tributary inflows creating the three flow peaks in October and November, and also closely mirroring the shape of the modelled natural hydrograph during these months (Figure 1.3). The importance of providing as close as possible to natural flow conditions in environmental flow allocations is increasingly being recognised (Arthington et al. 2006). Additionally, the flooding and the use of the environmental water occurred at a time when spawning had been recorded in this region for these species in previous years.

Whilst further study is required to elucidate and confirm the mechanism for the successful spawning and recruitment events observed for some native fish during this study, this research has demonstrated that it is possible to optimise and manage flows to improve native fish spawning and recruitment opportunities. We believe that this is the first study to demonstrate a strong link between fish spawning and recruitment and the provision of an environmental flood at least in Australia, but perhaps in the world, and has important implications for managing flows in regulated rivers in the Murray-Darling Basin and wider. Careful monitoring and scientific input into the planning and management of the 2005 B-M environmental water allocation allowed us to input some considerations of fish spawning and recruitment into the management of the event. The research provided valuable real-time and ongoing management input for optimising benefits of environmental water for restoration of fish communities. However, this event has to be viewed as only a single experiment, and as we still have much to learn about the role of various flows on fish and other biota, longer-term manipulative experimental flows utilising a variety of flow scenarios need to be encouraged (Bunn and Arthington 2002, Poff et al. 2003, Arthington et al. 2006). This needs to occur in an adaptive management context whereby the flow manipulations happen when suitable monitoring is underway, and the lessons learnt from each event can be incorporated into future management of the system. As suggested by Poff et al. (2003) this experimental approach needs to be done in full collaboration with scientists, managers and other stakeholders, as was used in the implementation of this environmental water allocation.

19


Summary of Key Findings• Golden perch and silver perch increased their spawning activity during the flood of 2005–06 compared to

the two previous seasons. However, recruitment success of these two species cannot be determined at this stage.

• Murray cod and trout cod spawning activity appeared not to increase during the flood year compared to the previous two seasons; however, there was a significant increase in the number of young-of-year of both species after the 2005–06 flood season, compared to the previous year.

• These results suggest that flooding is an important component of the life history strategies of these species either as a direct spawning cue or by increasing the survival of young and hence recruitment.

• This research has demonstrated that it is possible to optimise and manage flows to improve native fish spawning and recruitment opportunities.

Key Research Gaps

• Determine key components of the flood that triggered increased spawning activity of golden and silver perch.

• Determine mechanism for increasing recruitment strength

• Conduct follow-up monitoring of young-of-year to determine cohort strength and survival into future years.

20


3. SPATIAL AND TEMPORAL PATTERNS OF FISH SPAWNING AND RECRUITMENT IN THE BARMAH-MILLEWA FOREST WETLAND COMPLEX

IntroductionFloodplain rivers are amongst the world’s most dynamic, biologically diverse and valuable habitats (Power et al. 1995, Sparks 1995, Ward et al. 1999). They contain complex ecosystems structured around a diverse range of habitat types from flowing river channels, permanent and ephemeral creeks and floodways, backwaters, large floodplain lakes to wetlands and forested or grassland plains. The diversity and productivity of these floodplains is integrally linked with the adjacent river’s flow regime and with unregulated floodplain rivers that have naturally variable flow regimes containing a complex variety of ecological communities (Welcomme 1985, Junk et al. 1989). The flood pulse concept (FPC) (Junk et al. 1989) states that predictable flooding is the main driving force behind the existence, productivity and interactions of the major biota (including fish) in river-floodplain systems. The FPC also emphasises the role of the floodplain environment in the ecology of fish, suggesting that floodplain inundation provides a spawning cue for some fishes, and an abundant food and habitat source for all life stages, including larvae (Junk et al. 1989, Bayley 1991).

In the Murray-Darling Basin, the successful spawning and recruitment of fishes has often been linked to floodplain inundation (e.g. Geddes and Puckridge 1988, Lloyd et al. 1989, Harris and Gehrke 1994, Schiller and Harris 2001). Harris and Gehrke (1994) proposed the ‘flood recruitment model’ (FRM) for native Murray-Darling fishes and suggested that flooding either directly acts as a spawning cue for some species (particularly golden perch and silver perch), or indirectly enhances larval and juvenile survival by providing abundant food and habitat. This hypothesis was extrapolated from overseas studies in floodplain rivers (Welcomme 1985, Junk et al. 1989), early experimental breeding studies where golden perch were shown to initiate spawning on a water level rise in aquaculture ponds (Lake 1967a), gonad maturation of some golden perch after flooding (Mackay 1973), correlations of fish catches following flooding (Reid et al. 1997) and some observations (Cadwallader 1977). However strong evidence supporting the use and dependence of Australian native fishes on the floodplain for spawning and recruitment is limited and, more recently, the broad applicability of both the FRM and FPC to the Murray-Darling Basin has been questioned (Humphries et al. 1999, King et al. 2003).

Unlike fishes in many tropical river systems (Welcomme 1985), where the FPC was primarily based, to date no native Murray-Darling Basin fish have been shown to be solely dependent on inundated floodplains or flooding for spawning or recruitment. Even for golden perch and silver perch, which have been shown in this study to be capable of massive increases in spawning activity during appropriate flood conditions (see Chapter 2), some limited spawning was observed during stable within channel flows. Mallen-Cooper and Stuart (2003) and Balcombe et al. (2006) also found that spawning of golden perch was not solely dependent on floods. King et al. (2003) studied the use of floodplain habitats for fish spawning and recruitment in a non-flood and a flood year in the unregulated Ovens River, and found that the only species to increase in larval abundance associated with high flow conditions was the introduced species, carp. Additionally, the two native species (Australian smelt and carp gudgeons) and three introduced species (redfin perch, carp and gambusia) that were found in sufficiently high abundance also occurred as larvae during both years, suggesting that spawning and recruitment of these species is not reliant upon floodplain inundation, and can occur in isolated billabongs or anabranches on the floodplain. These species are also known to reside and recruit within the main channel of rivers (Humphries et al. 2002, King 2004). Interestingly, King et al. (2003) also found that carp gudgeons and gambusia only spawned in the floodplain habitats during low flow conditions in both years, suggesting that these two species may have a preference for spawning and recruitment during low flow conditions, similar to that proposed under the low flow recruitment hypothesis (Humphries et al. 1999). However, with such a limited number of studies having examined the dependence of fish on flooding and inundated floodplain habitats for spawning and recruitment in the Murray-Darling Basin, conclusions on the use of floodplains should be treated with caution (Humphries et al. 1999). Indeed, Graham and Harris (2004) cited this as a key research gap in our understanding of native fishes.

Fish at all life stages can utilise the floodplain environment for a variety of reasons including as refuge from catastrophic events (floods or droughts), permanent or temporary habitat, feeding, spawning and recruitment or a combination of all of these. Closs et al. (2006) suggested that the majority of Murray-Darling fish can be

21

3. Spatial and temporal patterns in the Barmah-Millewa Forest wetland complex

considered regular inhabitants of wetland habitats. Indeed, in complex floodplain river systems that contain a variety of permanent and ephemeral habitat types, the diversity of fish species in floodplain areas can be quite high and distinct compared to the main river channel. Meredith et al. (2002) surveyed larval and adult fish in fast flowing creeks, shallow ponds and a weir pool in the lower Murray floodplain at Lindsay-Whalpolla, and found that although the three habitat types had similar species diversity each contained a subtly different fish community. McNeil (2004) also reported distinct fish communities associated with various floodplain wetland types on the Ovens River floodplain. Despite the distinct fish communities and the number of species that are thought to utilise floodplain habitats, there has been relatively little research on the use of floodplain habitats or wetlands by fish (Graham and Harris 2004, Closs et al. 2006). Information such as the importance of connectivity between the river channel, floodplain and floodplain habitats, and the role of flow regimes including flood frequency, duration and magnitude and drying cycles, is required to effectively manage floodplain habitats to maintain and improve the Basin’s fish communities.

A workshop on native fish and wetlands in the Murray-Darling Basin listed the impact of watering trials and environmental flows to wetlands on native fish as a major research priority (Phillips 2006). This study aimed to assess the impact of water management (particularly environmental flows) in the Barmah-Millewa floodplain on native and introduced fish breeding and recruitment. The study reports on the timing of spawning and nursery habitat use during three years of sampling. This study is also the first to undertake detailed surveys of small fish communities on the B-M floodplain, and reports on the adult habitat preferences of a number of smaller native and introduced species.

Methods

Study area and hydrology

For a description of study area and hydrology see Chapter 1.

Collection and processing of fish eggs and larvae

Sampling of the early life stages of fish and adults of some smaller species was conducted monthly from mid-September to the end of February during 2003–04, 2004–05 and 2005–06 breeding seasons using all methods. An additional sampling trip was also conducted in March 2005, due to the occurrence of a late February natural flood pulse. Sampling was conducted at fourteen sites throughout the forest and was grouped into four broad categories: river channel, large lakes, forest creeks and wetlands (see Table 3.1 for site descriptions). Opportunistic sampling of the inundated floodplain proper was also conducted where possible. Additional fortnightly drift sampling of only the river sites also occurred in all three years. Data from these collections is used in showing total numbers captured and the timing of spawning only. A standardised number of replicate samples of a range of methods known to successfully capture small-bodied fish, including drift nets, light traps, sweep net electrofishing and hand trawls, was taken at each sampling site throughout the duration of the study (see Table 3.2). Only three light trap, Sweep Net Electrofishing and hand trawl samples were taken at wetland sites, as these sites were generally smaller in size and contained a less diverse range of meso-habitat types (woody debris, open water and vegetation) compared to the other broad habitats.

Passive drift nets were used to collect eggs and/or larvae of a number of species known to exhibit a drifting dispersal phase (Humphries and King 2004). All nets were set on dusk and retrieved as early as possible the following morning, generally before 1100 hours. Drift nets were 1.5 m long, with a 0.5 m diameter mouth opening and were constructed of 500 µm mesh, which tapered to a removable collection jar. A General Oceanics Inc. (Florida, USA) flow meter was fixed in the mouth of each drift net to determine the volume of water filtered and enabled raw catch data to be adjusted to a standard volume of filtered water (1000 m3). At the three river sites, two drift nets were attached to the top and bottom of a long pole to sample surface and bottom of the water column (overall depths 3–5 m). At the four creek sites, only one drift net was attached to an immovable piece of natural wood, so that the top of the mouth of the net was just under the water surface. Drift sampling at the Tullah Creek site was only conducted when the creek was flowing. In the field, any fish eggs found in the drift samples were removed alive from the samples and returned to the laboratory for hatching to enable correct identification.

22


Table 3.1: Major habitat types with associated sample sites and descriptions during the Barmah-Millewa recruitment project. Does not include additional floodplain sampling. Site descriptions are hydrology/connectivity with river channel, approximate size, mean depth, dominant habitat and water history.

Habitat SiteHydrology / connectivity

Depth (m) Size

Dominant habitat Water history

Large lake Barmah Lake HR *** 0.6 368 ha EM Connected during regulated river flows

Moira Lake HR *** 0.8 764 ha V / EM Regulated

Forest creeks

Budgee Creek HR *** 1.5 30m width WD / EM Regulated at top end but constantly connected to Barmah Lake

Tongalong Creek HR *** 1.5 40m width WD / EM Unregulated anabranch

Gulpa Creek HR *** 1.5 30m width WD Stable flows during Aug-May irrigation season

Tullah Creek HR ** 0.8 12m width WD Low lying creek with over bank flows during minor floods

Murray River

Ladgroves Beach HRna 2 95m width WD Stable flows during Aug-May irrigation season

Morning Glory HRna 1.8 90m width WD / EM Stable flows during Aug-May irrigation season

Barmah Choke HRna 2.5 75m width WD / EM Stable flows during Aug-May irrigation season

Wetlands Bunyip Billabong UR** 1 1 ha EM / WD Situated close to main river channel, fills during minor floods

Black Engine Billabong

UR* 1 6 ha WD Situated close to main river channel, fills during moderate floods

Tarma Swamp R** 0.8 4 ha WD Part of Tullah creek system, connects to creek during low level floods

Hut Lake R** 0.8 16 ha EM / SM Unique shallow grassy lake which fills during minor floods

Flat Swamp UR*** 1 3 ha WD Connects to Tongalong Creek during high level in-channel flows

HR = highly regulated, UR = unregulated; *** high connectivity, ** medium connectivity, * low connectivity; EM = emergent macrophytes, V = Valisneria, WD = woody debris, SM = submerged macrophytes

23


Table 3.2: Location (latitude, longitude), habitat type, collection methods and replication used at each site. Does not include additional floodplain sampling.

Site Name S E Habitat type Methods used No. reps

Barmah Lake 35’56.768 144’57.500 Lake Light trap 5

Hand trawl 5

SNE 5

Moira Lake 35’56.908 144’56.002 Lake Light trap 5

Hand trawl 5

SNE 5

Budgee Creek 35’56.128 144’58.169 Creek Light trap 5

Hand trawl 5

SNE 5

Drift net 1

Tongalong Creek 35’50.701 145’13.591 Creek Light trap 5

Hand trawl 5

SNE 5

Drift net 1

Gulpa Creek 35’42.352 144’55.278 Creek Light trap 5

Hand trawl 5

SNE 5

Drift net 1

Tullah Creek 35’54.730 145’02.512 Creek Light trap 5

Hand trawl 5

SNE 5

Drift net 1

Ladgroves Beach 35’51.677 145’20.773 Murray River Light trap 5

SNE 5

Drift net 2

Morning Glory 36’04.765 144’57.553 Murray River Light trap 5

SNE 5

Drift net 2

Barmah Choke 35’54.947 144’57.267 Murray River Light trap 5

SNE 5

Drift net 2

Bunyip Billabong 35’50.343 145’13.460 Billabong Light trap 3

Hand trawl 3

SNE 3

Black Engine Billabong 35’51.364 145’16.259 Billabong Light trap 3

Hand trawl 3

SNE 3

Tarma Swamp 35’54.377 145’02.344 Small lake Light trap 3

Hand trawl 3

SNE 3

Hut Lake 35’54.751 144’59.772 Small lake Light trap 3

Hand trawl 3

SNE 3

Flat Swamp 35’51.607 145’14.481 Small lake Light trap 3

SNE 3

Hand trawl 3

24


Plate 8: Some of the sites sampled during this study (clockwise from top left), Barmah Lake, Gulpa Creek, Hut Lake and Murray River at Morning Glory (downstream of Forest).

Plate 9: Drift sampling in Tullah Creek, September 2004.

25


Modified quatrefoil light traps (Floyd et al. 1984, Secor et al. 1992) were used to collect fish from all habitat types, and were useful in collecting fish from a range of meso-habitat types and depths. These were constructed from clear Perspex and steel, with a removable 250 µm sieve attached to the base, as used by Humphries et al. (2002). The light traps were set on dusk with a yellow 12 hour light stick (CYALUME®), and were retrieved as early as possible the next day. Active sampling was also conducted in all habitat types using the Sweep Net Electrofishing (SNE) method (King and Crook 2002), which is suitable for sampling small-bodied fish at wadable depths in a variety of habitats. Briefly, the SNE method is a modified standard backpack electrofishing unit (Smith-Root Model 12), with a 15 cm diameter anode ring, and fitted with a moulded plastic rectangular frame (25 x 30 x 2 cm) with an attached 250 µm mesh sampling net that is similar in shape to a standard sweep or dip net. The frame and mesh net are attached to the bottom of the anode pole so that the anode ring is in the centre of the opening of the net. The SNE method has the advantage of stunning small-bodied fish and then immediately capturing them in the fixed net (King and Crook 2002). Sampling involved approaching the selected habitat quietly from a downstream direction, activating the anode and moving at a constant speed in a forward zig-zag motion to cover all the available depths of the habitat. Replicate samples involved a standard 20 seconds of electrofishing time. Electrofishing was conducted by one operator, with another person always present for sample preservation and safety reasons. Since the SNE method is restricted to wadable depths, hand trawl samples were also taken to sample deeper areas of wetlands, large lakes and creek habitats. The hand trawl net was similar to a standard zooplankton sampling net, with a 30 cm diameter opening and 250 µm mesh net tapering to a removable collection jar. The net was thrown the full length of an attached 5 m rope and pulled quickly through the top of the water column. One replicate hand trawl sample consisted of five pooled 5 m trawls. All samples were preserved in 95% ethanol in the field and returned to the laboratory for processing, where fish were removed from the samples using a dissecting microscope.

Plate 10: Light trap with light stick glowing being set in the Murray River.

26


Plate 11: Light trap being retrieved from Budgee Creek inundated floodplain, November 2005.

Plate 12: SNE method in use in Gulpa Creek.

27


Plate 13: Sweep net electrofishing method, showing small anode ring and square sampling frame attached to anode pole.

Plate 14: Hand trawl sampling in Budgee Creek.

28


Plate 15: Hand trawl net.

Identifications were made by experienced staff using available keys (Serafini and Humphries 2004), and by utilising an existing reference collection of successive larval stages. Water quality parameters (temperature, dissolved oxygen, conductivity, turbidity and pH) were measured in three locations at each site on each sampling trip using a Horiba U10 Water Quality Checker (Horiba Ltd, Japan).

Data analysis

A Restricted Maximum Likelihood (REML) analysis was used to provide a robust statistical approach to model the unbalanced nested design structure across the main factors of interest (Corbeil and Searle 1976, Venables and Dichmont 2004). For each developmental stage and species combination, REML analysis was only conducted where:

• greater than 30 individuals across the three sampling seasons were collected,

• using a method that captured greater than 40% of total individuals, and

• only using trips where different developmental stages were present.

For drift data, each model incorporated ‘Habitat’ (2 levels: Creek, River), ‘Season’ (3 levels) and ‘trip’ (6 levels) as fixed effects. Drift data was adjusted to a standard volume of water filtered (number of eggs/larvae per 1000 m3), and the average number of standardised fish captured per habitat type on each trip was used in the analysis, ie. averaging across both sites and replicate nets within a site (river sites only). For SNE and light trap data (analysed separately) each model incorporated ‘Habitat’ (4 levels: Creek, River, Wetland, Lake), ‘Season’ (3 levels) as fixed effects and ‘trip’ (6 levels). For hand trawl data each model incorporated ‘Habitat’ (3 levels: Creek, Wetland, Lake), ‘Season’ (3 levels) as fixed effects and ‘trip’ (6 levels). The factor trip was significant in nearly all analyses and was not reported in results as this is due to normal within season variation. All data was log10 (x+1) transformed before analysis. The significance of each fixed effect was assessed from Wald statistics, which approximate a χ2 distribution for the appropriate degrees of freedom. However Wald statistics can elevate the probability of a Type 1 error, particularly when sample sizes are small, and so trends associated with χ2 p-values < 0.05 are discussed. REML analyses were performed using GenStat for Windows 6.1™.

The patterns in the fish community between years and sites were examined using non-metric multidimensional scaling (MDS) ordination techniques in the statistical package PRIMER 5 (Clarke and Warwick 2001). Data were reduced to presence/absence of particular species or developmental stage across all sampling methods prior to analysis, and used the Bray-Curtis similarity measure. A two-way ANOSIM (Analysis of Similarity) using 999 permutations and a significance level of < 0.05 was used to test the hypothesis that there were differences in the fish assemblages between habitat types (lake, river, creek and wetland) and season (three sampling seasons).

29


Results

Hydrology and environmental variables

The hydrographs differed markedly between each of the three years of this study (Figure 3.1a). The 2003–04 season was characterised by three winter and early spring floods that inundated the Forest, then continued as fairly stable, bank-full flows for the remainder of the season. A very minor flood event occurred in late December 2003 which inundated very restricted areas of the Forest. In 2003–04 all forest sampling sites were inundated during the September flood and all had varying degrees of connection for the remainder of the sampling season (Figure 3.1b). The 2004–05 season was characterised by one spring flood event and again remained fairly stable with bank-full flows for the rest of the season. A minor, natural flood event that inundated low lying areas of the Forest occurred in late February 2005. This flood caused a minor blackwater event into the Murray River but no fish kills or adverse environmental effects were observed (King, pers. obs.). The February flood was also associated with a marked decline in river water temperature. In 2004–05, all floodplain sites except Black Engine Billabong were connected during the September and February flood events (Fig 3.1b). Bunyip Billabong was not connected during the February flood. The 2005–06 season hydrograph was quite distinct from the two previous seasons, with floodplain inundation occurring for nearly five months from early August to late December, and contained a number of peaks. As previously discussed, the use of the B-M Environmental Water Allocation occurred during October to December 2005. All floodplain sampling sites were connected during this flood event, except Black Engine Billabong (Figure 3.1b). Water level in Black Engine Billabong declined steadily throughout the whole study and was at very low levels (maximum depth 0.5m) during 2005–06 season. All forest creek habitats maintained some connectivity (albeit intermittent at times) to the river during the study period, except Tullah Creek which tended to dry into a series of isolated pools by February/March in 2004–05 and 2005–06.

Mean monthly water temperature ranged from 12.0 to 28.6°C across all habitat types sampled (Figure 3.2a). There was no significant difference in water temperature across the three sampling seasons or across habitats (2-way ANOVA, p>0.05). However, there was a difference between summer (December, January, February) water temperatures across the three years, with 2003–04 summer being 1-3°C warmer than the other two seasons (p<0.01). Mean monthly dissolved oxygen levels ranged from 3.50 to 9.97 mg/L across all habitat types (Figure 3.2b), with no significant difference occurring between sampling seasons (2-way ANOVA, p>0.05). Dissolved oxygen levels did significantly differ between habitat types (p<0.01), with river sites generally having significantly higher oxygen concentrations than all three floodplain habitat types (p<0.01). Mean monthly turbidity varied greatly within season (Figure 3.2c), across season (2-way ANOVA, p<0.05) and between habitats types (p<0.001). Turbidity was significantly greater in 2003–04 than 2005–06 (p<0.05), but there was no difference between other years. River sites had significantly lower turbidity than any of the three floodplain habitats throughout the study period (p<0.001). Mean monthly conductivity ranged from 43.1 to 143.2 µS/cm (Figure 3.2d), with no significant difference occurring between sampling seasons (2-way ANOVA, p>0.05). Conductivity did decline significantly between each of the habitat types, from wetland, creek, lake and river habitats (Figure 3.2d, p<0.05). Mean monthly pH across all habitat types ranged from 6.33 to 8.40 (Figure 3.2e), with very little variation between either season or habitat.

Species composition

In total 46,713 individuals from all life history stages were captured throughout the study period, representing ten native and five introduced species (Table 3.3). All species were collected as eggs, larvae or juveniles indicating recent spawning and recruitment. Most species and life stages were captured every breeding season; however, some species and stages (southern pygmy perch, Murray-Darling rainbowfish, trout cod, golden and silver perch) were captured in very low numbers in the first two seasons and were captured in greater numbers in 2005–06 (Table 3.3). In general, the same species were captured as larvae in each of three seasons; however a small number of trout cod and Murray-Darling rainbowfish larvae were collected for the first time in 2005–06. Across the entire study period, the raw abundance of larvae was heavily dominated by Australian smelt (68%) and carp gudgeons (15%), with other species (including introduced species) representing a very small proportion (<1% each) of the remainder of the catch. The raw abundance of juvenile fish across the entire study period, was dominated by Australian smelt (47.8%), gambusia (23.5%) and carp gudgeons (13.3%). The raw abundance of adults was dominated by carp gudgeons 46.3%), gambusia (28.4%) and Australian smelt (20.2%).

30


Figu

re 3

.1: (

a) M

ean

daily

dis

char

ge (s

olid

line

) and

wat

er te

mpe

ratu

re (d

ashe

d lin

e) a

cros

s th

e th

ree

sam

plin

g ye

ars

and

(b) h

ydro

logi

cal s

tatu

s of

eac

h fo

rest

floo

dpla

in s

ampl

ing

site

ass

esse

d ea

ch s

ampl

ing

trip

. Sol

id b

lack

bar

indi

cate

s in

full

floo

d, s

olid

gre

y ba

r in

dica

tes

low

floo

d, s

olid

whi

te b

ar

indi

cate

s is

olat

ed h

abita

t and

str

iped

bar

indi

cate

s co

nnec

ted

to o

ther

hab

itats

. Dis

char

ge a

nd te

mpe

ratu

re d

ata

from

Toc

umw

al g

auge

on

Mur

ray

Riv

er.

Not

e: d

otte

d lin

e re

pres

ents

app

roxi

mat

e fl

oodp

lain

inun

datio

n he

ight

, sol

id tr

iang

les

on x

-axi

s in

dica

tes

sam

plin

g ev

ent,

lett

ers

on x

-axi

s re

pres

ent

cale

ndar

mon

ths

in e

ach

year

from

Jul

y to

Mar

ch/A

pril.

31


Figure 3.2: Mean monthly (a) water temperature, (b) dissolved oxygen, (c) turbidity (d) conductivity and (e) pH for each habitat type in each sampling season. Blue squares indicate creek habitats, green diamonds indicate lake habitats, light blue circles indicate river habitats and red triangles indicate wetland habitats. Note: Water quality variables were measured as spot measurements in each sampling site each month using a water quality checker.

32


Tabl

e 3.

3: R

aw to

tal a

bund

ance

of l

arva

e (L

), ju

veni

les

(J) a

nd a

dult

s (A

) of a

ll m

etho

ds c

ombi

ned

of e

ach

spec

ies

acro

ss th

e th

ree

sam

plin

g se

ason

s. T

otal

abu

ndan

ce

of g

olde

n pe

rch

and

silv

er p

erch

egg

s sh

own

in b

rack

ets

for

each

sea

son.

# In

clud

es 6

uni

dent

ified

cod

larv

ae.

Com

mon

nam

eSc

ient

ific

nam

e20

03–0

420

04–0

520

05–0

6

LJ

ATo

tal

LJ

ATo

tal

LJ

ATo

tal

Nat

ive

Aust

ralia

n sm

elt

Ret

ropi

nna

sem

oni

7191

3508

649

1134

822

9384

179

739

3159

6821

2364

087

31

carp

gud

geon

sH

ypse

leot

ris

spp.

971

453

1126

2550

1228

654

2171

4053

1169

698

1485

3352

flat-

head

ed g

udge

onP

hily

pnod

on

gran

dice

ps73

201

9414

262

921

385

577

149

unsp

ecke

d ha

rdyh

ead

Cra

tero

ceph

alus

st

ercu

smus

caru

m97

130

9532

212

528

291

498

164

120

9437

8

Mur

ray

cod

Mac

cullo

chel

la p

eelii

pe

elii

290

029

542

056

974

0#

107

trou

t cod

Mac

cullo

chel

la

mac

quar

iens

is0

00

00

10

14

00

4

gold

en p

erch

Mac

quar

ia a

mbi

gua

(290

) 10

01

(109

) 20

02

(190

6)

110

00

110

silv

er p

erch

Bid

yanu

s bi

dyan

us(4

39)

400

040

(619

) 20

02

(358

5)

195

00

195

sout

hern

pyg

my

perc

hN

anno

perc

a au

stra

lis0

01

10

89

170

2921

50

Mur

ray-

Dar

ling

rain

bow

fish

Mel

anot

aeni

a flu

viat

ilis

03

36

00

11

45

211

Intr

oduc

ed

carp

Cyp

rinu

s ca

rpio

196

731

270

1216

303

015

1958

651

210

98

gold

fish

Car

assi

us a

urat

us5

288

4124

139

1617

918

100

412

2

redfi

n pe

rch

Perc

a flu

viat

ilis

2024

145

9463

015

712

620

74

gam

busi

aG

ambu

sia

holb

rook

i94

496

529

1119

234

1239

1498

2971

160

1446

906

2512

orie

ntal

wea

ther

loac

hM

isgu

rnus

an

guill

icau

datu

s1

366

7011

583

9911

4517

73

Uni

dent

ified

130

013

571

159

370

340

Tota

l raw

abu

ndan

ce87

3147

3824

8015

949

5482

3600

4676

1375

886

2052

0131

7917

006

33


This study used four different methods to capture fish, due to the known behavioural and habitat differences among early life stages. The percentage contribution of each species (by developmental stage) differed considerably with sampling method (Table 3.4). For example, while Australian smelt larvae were captured by all four methods, 45% of individuals were captured using the SNE. Other species such as flat-headed gudgeon (larvae), Murray cod (larvae), silver perch (eggs and larvae) and golden perch (eggs and larvae) were almost exclusively captured in drift sampling. The majority of fish species and stages were captured in either drift nets, light traps or SNE sampling. Carp gudgeon larvae were the only species to be mostly captured in hand trawl samples.

Table 3.4: Percentage of each species (by developmental stage) captured in each method.

Common name StagePercentage captured in each method

Total no.D HT LT SNENativeAustralian smelt L 27.50 12.00 15.53 44.96 15450

J 0.23 28.23 59.15 12.39 6472A 0.58 6.24 86.95 6.24 2085

carp gudgeons L 3.26 45.85 20.72 30.17 3315J 6.09 5.81 47.43 40.67 1790A 0.42 0.25 87.69 11.64 4768

Flat-headed gudgeon L 90.67 0.00 4.33 5.00 300J 23.02 0.00 25.18 51.80 139A 6.67 0.00 53.33 40.00 15

unspecked hardyhead L 0.52 3.11 41.19 55.18 386J 0.00 1.88 90.79 7.33 532A 0.36 0.00 95.71 3.93 280

Murray cod L 90.91 0.00 7.95 1.14 176J 0.00 0.00 0.00 100.00 6

trout cod L 100.00 0.00 0.00 0.00 5J 0.00 0.00 0.00 0.00 0

golden perch E 100.00 0.00 0.00 0.00 2005L 100.00 0.00 0.00 0.00 113

silver perch E 99.98 0.00 0.02 0.00 4643L 100.00 0.00 0.00 0.00 237

southern pygmy perch L 0.00 0.00 0.00 0.00 0J 0.00 0.00 75.68 24.32 37A 3.23 0.00 77.42 19.35 31

Murray-Darling rainbowfish L 0.00 25.00 0.00 75.00 4J 0.00 25.00 37.50 37.50 8A 0.00 0.00 83.33 16.67 6

Introducedcarp L 73.19 1.19 8.01 17.61 1936

J 7.46 0.45 44.52 47.57 885A 0.00 0.00 0.00 100.00 1

goldfish L 17.02 10.64 14.89 57.45 47J 1.32 2.63 27.63 68.42 228A 0.00 0.00 0.00 100.00 28

redfin perch L 1.59 0.79 92.06 5.56 126J 0.67 0.00 75.84 23.49 149A 0.00 0.00 0.00 100.00 1

gambusia L 1.43 39.14 17.01 42.42 488J 0.00 26.31 48.79 24.90 3181A 0.07 13.77 75.55 10.60 2933

oriental weatherloach L 6.67 0.00 33.33 60.00 15J 0.00 0.00 11.32 88.68 53A 0.78 0.00 46.09 53.13 128

34


Timing of spawning, duration and abundance of larvae and juveniles

The presence of eggs or newly hatched larvae (protolarvae) was used as an indication of recent spawning (see Figure 3.3). The duration of spawning for most species was fairly consistent each season (ie. short or protracted); however the timing of the onset and end of the spawning period did differ slightly between the seasons for each species. Spawning of Australian smelt, carp, goldfish, redfin and golden perch first occurred early in the season in most years (either September or October), whereas species such as unspecked hardyhead, carp gudgeons and gambusia first spawned much later in each season. Estimated spawning times for Australian smelt and Murray cod were exactly the same between each of the three seasons (September-December and November – January, respectively).

A number of species did seem to alter the pattern of their spawning time as a result of water level changes across the three years (Figure 3.3). Carp, goldfish, golden perch, silver perch, flat-headed gudgeon and unspecked hardyhead were all detected spawning earlier in 2005–06 during the extended flood event compared to the previous seasons. Interestingly, in 2005–06 gambusia delayed their major spawning event (or bearing of young) until slightly later in the season after the flood had passed. Carp demonstrated flexibility in their spawning time, with a second major spawning event occurring on the mid-February 2005 flood event, after they had already spawned successfully earlier in the season.

Whilst there were changes in spawning times of some species associated with flood conditions across the three seasons, there were also some changes in the timing of the peak abundance of larvae and juveniles (Figure 3.4–3.7), which could be attributed to flood conditions. The duration and peak abundance of Australian smelt larvae did not differ between the three sampling seasons. The peak abundance of juveniles captured, however, was greater and occurred a month later (December) in 2005–06 than compared to the two previous seasons (Figure 3.4a). In 2003–04 and 2004–05, the proportion of carp gudgeon larvae captured peaked in December or February; however, in 2005–06 a fairly constant proportion of larvae were collected from October to February with no obvious peaks in larval abundance (Figure 3.4b). In all three seasons the proportion of carp gudgeon juveniles captured steadily increased throughout the season. Although no southern pygmy perch larvae and only a few juveniles were collected throughout the duration of the study, the peak in the occurrence of juveniles occurred in December in both years they were captured (Figure 3.4c). The peak abundance of unspecked hardyhead larvae was fairly even across their occurrence in the first two years, however in 2005–06 the peak abundance occurred in December (Figure 3.5a). The majority of juvenile hardyhead were always collected in February or March in each season. The number of flat-headed gudgeon larvae collected in SNE, hand trawl and light traps across the three years was fairly low compared to the higher numbers collected in drift nets. The proportion of Flat-headed gudgeon larvae and juveniles in the first two sampling seasons was fairly similar across the months, however larval occurrence was one month earlier (November) and juveniles peaked earlier in December in 2005–06 than compared to the previous seasons (Figure 3.5b). Low numbers of Murray Darling rainbowfish were collected in all three sampling seasons, with larvae collected only in December and January 2005–06 and juveniles collected in January and February in 2003–04 and 2005–06 (Figure 3.5c).

The peak abundance and pattern of occurrence of carp larvae differed between each sampling season (Figure 3.6a). The peak abundance of carp larvae occurred in October in the first two sampling seasons after water levels had declined and water temperatures had increased. A second major peak in the abundance of carp larvae also occurred coinciding with the mid-February flood in 2004–05. In 2005–06 two peaks in the abundance of carp larvae occurred, one in September after the first major flood, and another in November during the second flood. The abundance of juvenile carp was fairly consistent between months and seasons, but was slightly greater in December especially in 2005–06 (Figure 3.6a). Goldfish larvae were captured in low numbers using hand trawls, light traps and the SNE method, but were recorded at a similar time in all three seasons, except in 2004–05 where some larvae were captured after the February flood event (Figure 3.6b). The proportional abundance of redfin perch larvae was always at its greatest in October across the three sampling seasons, however the peak abundance of juveniles occurred one month earlier in October in 2005–06 (Figure 3.6c). The pattern of proportional abundance of gambusia larvae varied between the seasons (Figure 3.7a), with a definite peak occurring in February 2003–04, and January 2005–06 seasons. In 2004–05 there was no clear peak in abundance, however, larvae occurred for a greater period of time. Very few individual oriental weatherloach larvae or juveniles were collected for the three seasons, however, the timing of the occurrence of larvae did change significantly between the two seasons they were captured (Figure 3.7b).

35


Figu

re 3

.3: (

a) M

ean

daily

dis

char

ge (s

olid

line

) and

wat

er te

mpe

ratu

re (d

ashe

d lin

e) a

cros

s th

e th

ree

sam

plin

g ye

ars

and

(b) e

stim

ated

spa

wni

ng p

erio

d (e

stim

ated

from

col

lect

ion

of e

ggs

or n

ewly

hat

ched

larv

ae) f

or a

ll o

f the

maj

or s

peci

es c

aptu

red

acro

ss th

e th

ree

sam

plin

g se

ason

s. D

isch

arge

and

te

mpe

ratu

re d

ata

from

Toc

umw

al g

auge

on

Mur

ray

Riv

er. N

ote:

dot

ted

line

repr

esen

ts a

ppro

xim

ate

floo

dpla

in in

unda

tion

heig

ht, s

olid

tria

ngle

s on

x-a

xis

indi

cate

s sa

mpl

ing

even

t, le

tter

s on

x-a

xis

repr

esen

t cal

enda

r m

onth

s in

eac

h ye

ar fr

om J

uly

to M

arch

/Apr

il.

36


Figu

re 3

.4: P

erce

ntag

e of

tota

l col

lect

ed (r

aw n

umbe

r) in

a s

easo

n of

larv

ae a

nd ju

veni

les

capt

ured

per

mon

thly

sam

plin

g tr

ip fo

r (a

) Aus

tral

ian

smel

t,

(b) c

arp

gudg

eon

and

(c) s

outh

ern

pygm

y pe

rch

usin

g al

l met

hods

exc

ept d

rift

s. B

lack

bar

s in

dica

te la

rvae

, whi

te b

ars

indi

cate

juve

nile

s.

37


Figu

re 3

.5: P

erce

ntag

e of

tota

l col

lect

ed (r

aw n

umbe

r) in

a s

easo

n of

larv

ae a

nd ju

veni

les

capt

ured

per

mon

thly

sam

plin

g tr

ip fo

r (a

) uns

peck

ed h

ardy

head

, (b

) flat

-hea

ded

gudg

eon

and

(c) M

urra

y D

arlin

g ra

inbo

wfis

h us

ing

all m

etho

ds e

xcep

t dri

ft n

ets.

Bla

ck b

ars

indi

cate

larv

ae, w

hite

bar

s in

dica

te ju

veni

les.

38


Figu

re 3

.6: P

erce

ntag

e of

tota

l col

lect

ed (r

aw n

umbe

r) in

a s

easo

n of

larv

ae a

nd ju

veni

les

capt

ured

per

mon

thly

sam

plin

g tr

ip fo

r (a

) car

p, (b

) gol

dfish

an

d (c

) red

fin p

erch

usi

ng a

ll m

etho

ds e

xcep

t dri

ft n

ets.

Bla

ck b

ars

indi

cate

larv

ae, w

hite

bar

s in

dica

te ju

veni

les.

39


Figure 3.7: Percentage of total collected (raw number) in a season of larvae and juveniles captured per monthly sampling trip for (a) gambusia and (b) oriental weatherloach using all methods except drift nets. Black bars indicate larvae, white bars indicate juveniles.

Temporal and spatial variation in occurrence and abundance of each species

The total abundance of all larvae captured in both SNE and light trap methods did not vary significantly among habitat types, but did vary among years (Table 3.5, Figure 3.8a), with a greater abundance of larvae captured in 2003–04. The total abundance of all juveniles captured in both SNE and light trap methods did not vary among seasons (Table 3.55, Figure 3.8b). However, the abundance of juveniles captured in the SNE method did significantly vary among habitats, with a significantly greater number of juveniles occurring in wetland habitats than either lake or river habitats (p<0.01). There was no difference between creek and the other habitats.

In general, there were some differences in the mean abundance of species/stages between the three sampling seasons, but most of the majority of variation occurred between the habitats sampled (Table 3.5, Figures 3.9–3.12, see also Appendix 5 for post hoc results). The abundance of Australian smelt larvae varied significantly among both seasons and habitat types (Table 3.5, Figure 3.9a). A significantly greater abundance of larvae were captured in 2003–04 than in 2004–05 (p<0.001), and a significantly greater abundance of larvae occurred in river habitats compared to the other floodplain habitats (p<0.01), while the lowest catches using the SNE method were found in lakes and wetlands. With all methods and years combined, more than 60% of all Australian smelt larvae were captured in creek habitats (Figure 3.10a). The abundance of juvenile Australian smelt varied significantly among season with most captured in 2003–04 (Table 5; 2003–04> 2005–06>2004–05, p<0.01), but there was no significant difference among habitat types. There was no significant difference in the abundance of adult Australian smelt either among seasons or habitats (Table 5, Figure 10a). The abundance of carp gudgeon larvae, juveniles and adults did not significantly differ among seasons, but did among habitat types (Table 3.5, Figure 3.9b), with a significantly greater abundance of all stages captured in wetlands compared to the other

40


habitats (p<0.05). The strong association for wetland habitats for carp gudgeons as both a nursery and adult habitat was also evident when all methods and seasons were combined (Figure 3.10). The abundance of flat-headed gudgeon larvae varied significantly among seasons, but not habitat types (Table 3.5, Figure 3.9d), with significantly fewer larvae captured in 2005–06 than both of the previous seasons (p<0.05). Flat-headed gudgeon larvae were only captured in the flowing habitats of creeks and rivers, and never in wetlands or lake habitats (Figure 3.10). The abundance of juvenile flat-headed gudgeons did not differ significantly among seasons, but did among habitat types (Table 3.5, Figure 3.9d), with a greater abundance of juveniles being captured in river habitats. Across all methods and seasons, juvenile and adult flat-headed gudgeons were also mostly captured in creek or river habitats (Figure 3.10). The abundance of unspecked hardyhead larvae, juveniles and adults did not differ significantly among years, but did among habitats (Table 3.5, Figure 3.9e & f), with a significantly greater abundance of all life stages captured in the river compared to the other habitat types (p<0.01). Across all methods and seasons, all stages of unspecked hardyhead were also mostly captured in river habitats (Figure 3.10).

Table 3.5: Wald statistics and significance level for full fixed REML models of season, habitat and trip (not shown see methods) for abundant species and stages. Estimates of significance are based on a X2 distribution. Bold values indicate significant variables. See methods for more information on statistics.

Species StageSeason Habitat Used in analysis

Wald P value Wald P value Trips MethodAustralian smelt Total larvae 16.46 <0.001 18.18 <0.001 S,O,N,D SNE

Juveniles 34.03 <0.001 4.32 0.229 O,N,D,J,F LTAdults 1.33 0.514 1.46 0.691 All trips LT

carp gudgeons Total larvae 1.55 0.46 48.98 <0.001 O,N,D,J,F HTJuveniles 3.68 0.159 31.66 <0.001 O,N,D,J,F SNEJuveniles 6.34 0.042 15.24 0.002 N,D,J,F LTAdults 3.42 0.181 58.29 <0.001 All trips LT

flat-headed gudgeon Larvae 9.58 0.008 0.02 0.895 S,O,N DriftJuveniles 2.48 0.29 39.08 <0.001 D,J,F SNE

unspecked hardyhead Total larvae 0.65 0.722 56.18 <0.001 D,J,F SNETotal larvae 4.39 0.112 29.85 <0.001 D,J,F LTJuveniles 1.02 0.599 18.8 <0.001 D,J,F LTAdults 0.53 0.766 51.1 <0.001 All trips LT

Murray cod Total larvae 2.94 0.23 0 0.969 N,D Driftgolden perch Eggs 3.04 0.219 4.85 0.028 O,N,D Driftsilver perch Eggs 1.75 0.417 28.48 <0.001 N,D,J,F Driftsouthern pygmy perch Juveniles 5.84 0.050 10.01 0.018 N,D,J LT

Adults 4.55 0.103 7.18 0.066 O,N,D,J,F LTcarp Total larvae 2.42 0.299 0 0.973 All trips Drift

Juveniles 19.05 <0.001 7.23 0.065 O,N,D,J,F SNEJuveniles 2.36 0.308 12.28 0.006 O,N,D,J LT

goldfish Total larvae 2.42 0.298 8.8 0.032 O,N,D SNEJuveniles 9.07 0.011 7.99 0.046 O,N,D,J,F SNE

redfin perch Total larvae 1.64 0.44 5.75 0.124 O,N LTJuveniles 0.74 0.689 5.95 0.114 O,N,D,J LT

gambusia Total larvae 0.72 0.699 19.25 <0.001 D,J,F SNEJuveniles 2.26 0.322 48.94 <0.001 All trips LTAdults 1.6 0.448 43.63 <0.001 All trips LT

oriental weatherloach Juveniles 7.06 0.029 5.36 0.148 O,N,D,J,F SNEAdults 4.1 0.129 7.08 0.069 O,N,D,J,F SNEAdults 6.31 0.043 2.12 0.548 O,N,D,J LT

Total larvae (All species) 6.35 0.042 7.57 0.056 All trips SNE13.52 0.001 3.01 0.39 All trips LT

Total juveniles (All species) 3.28 0.194 14.67 0.002 All trips SNE4.58 0.101 5.1 0.165 All trips LT

41


Figure 3.8: Mean (+ 1 standard error) log10 (x+1) abundance per habitat type for each sampling season for all developmental stages captured for (a) total larvae and (b) total juveniles captured in SNE and light trap methods. Black solid bars indicate creek habitats, grey bars indicate lake habitats, striped bars indicate river habitats and white bars indicate wetland habitats. For each developmental stage and species combination, only methods that captured greater than 40% of the total were analysed, and where greater than 30 individuals in total across the three sampling seasons.

42


Figure 3.9: Mean (+ 1 standard error) log10 (x+1) abundance per habitat type for each sampling season for all developmental stages captured for (a) Australian smelt, (b & c) carp gudgeons, (d) flat-headed gudgeons and (e &f) unspecked hardyhead. Black solid bars indicate creek habitats, grey bars indicate lake habitats, striped bars indicate river habitats and white bars indicate wetland habitats. For each developmental stage and species combination, only methods that captured greater than 40% of the total were analysed, and where greater than 30 individuals in total across the three sampling seasons.

43


Figure 3.10: Percentage composition by habitat type for (a) larvae, (b) juveniles and (c) adults for all species using all methods pooled across the three years. Black solid bars indicate creek habitats, grey bars indicate lake habitats, striped bars indicate river habitats and white bars indicate wetland habitats. Numbers on the top of columns represent the total number of individuals.

44


There was no significant difference in the mean adjusted abundance of Murray cod larvae captured among years or between river and creek habitats (Table 3.5, Figure 3.11a), however overall a greater proportion of individuals were captured in the river compared to creek habitats (Figure 3.10). No Murray cod or trout cod larvae or juveniles were ever captured in wetlands or the two lakes, and while Murray cod were captured in Tongalong Creek in all years, they were rarely captured in the other three creek sites (see Table 3.6). The mean adjusted abundance of golden perch and silver perch eggs collected in only the monthly sampling trips did not differ among years (but see chapter 2), but did differ significantly between river and creek habitats (Table 3.5, Figure 3.11b & c). The eggs and larvae of golden and silver perch were only ever collected at river sites (Figures 3.10, 3.11b & c, Table 3.6). The mean abundance of southern pygmy perch juveniles varied significantly among season (Table 3.5, Figure 3.11d), with a significantly greater abundance captured in the 2005–06 season than 2003–04 (p<0.05), but there was no significant effect of season on adult abundance. Juvenile and adult southern pygmy perch also increased their distribution throughout the Forest in 2005–06 than compared to the other seasons, being detected at more sites (Juveniles: 2005–06 = 5 sites, other seasons = 0 or 1; Adults 2005–06 = 4 sites, other seasons 1 or 2). In the first two years, they were found only in Tarma swamp, while in 2005–06, higher numbers were collected in Tarma swamp, Flat swamp and Tongalong Creek, and lower numbers in Hut Lake, Bunyip Billabong and Tullah Creek. The abundance of juvenile southern pygmy perch varied significantly among habitats,

Figure 3.11: Mean (+ 1 standard error) log10 (x+1) abundance per habitat type for each sampling season for all developmental stages captured for (a) Murray cod, (b) golden perch, (c) silver perch and (d) southern pygmy perch. Black solid bars indicate creek habitats, grey bars indicate lake habitats, striped bars indicate river habitats and white bars indicate wetland habitats. For each developmental stage and species combination, only methods that captured greater than 40% of the total were analysed, and where greater than 30 individuals in total across the three sampling seasons. Note: No golden perch eggs were captured in monthly sampling in 2003/04 season, but they were captured in fortnightly samples (see Chapter 2).

45


Tabl

e 3.

6: R

aw to

tal a

bund

ance

of fi

sh c

aptu

red

in d

rift

net

s in

riv

er a

nd c

reek

sam

plin

g si

tes

over

the

thre

e sa

mpl

ing

seas

ons.

Not

e: o

nly

spec

ies/

stag

es w

ith

>10%

tota

l cat

ch c

aptu

red

in d

rift

net

s (s

ee T

able

3.4

) sho

wn

in th

is T

able

.

Ladg

rove

s B

each

(M

urra

y R

iver

)B

arm

ah C

hoke

(Mur

ray

Riv

er)

Mor

ning

Glo

ry (M

urra

y R

iver

)B

udge

e C

reek

Gul

pa C

reek

Tong

alon

g C

reek

Tulla

h C

reek

03/0

404

/05

05/0

603

/04

04/0

505

/06

03/0

404

/05

05/0

603

/04

04/0

505

/06

03/0

404

/05

05/0

603

/04

04/0

505

/06

03/0

404

/05

05/0

6

gold

en p

erch

(e

gg)

2014

291

143

6965

612

726

659

00

00

00

00

00

00

gold

en p

erch

(p

roto

larv

ae)

02

891

014

00

70

00

00

00

00

00

0

silv

er p

erch

(e

gg)

134

469

1682

219

6510

6986

8584

30

00

00

00

00

00

0

silv

er p

erch

(p

roto

larv

ae)

292

176

100

181

01

00

00

00

00

00

00

Mur

ray

cod

(tota

l lar

vae)

920

451

1224

03

120

20

03

014

13

00

0

trou

t cod

(to

tal l

arva

e)0

01

00

20

01

00

00

00

01

00

00

Aust

ralia

n sm

elt (

tota

l la

rvae

)

22

81

81

150

232

16

00

03

63

2058

1818

06

flat-

head

ed

gudg

eon

(tota

l lar

vae)

964

490

102

26

170

00

00

046

280

1025

0

flat-

head

ed

gudg

eon

(juve

nile

s)

55

10

01

01

140

00

00

04

10

00

0

carp

(tot

al

larv

ae)

94

40

10

6583

839

60

26

10

012

12

711

3

carp

(ju

veni

les)

00

00

00

00

560

00

00

02

21

03

0

gold

fish

(tota

l la

rvae

)0

00

00

01

12

00

10

00

01

00

00

46


with a greater abundance of juveniles found in creek and wetland habitats than compared to lake or river habitats, but there was no significant effect of habitat type on adult abundance. Across all methods and seasons, juvenile and adult southern pygmy perch were only collected in wetland and creek habitats (Figure 3.10). As very low numbers of all stages of Murray-Darling rainbowfish were collected during this study (see Table 3.3), the results discussed here should be treated with caution. All stages of Murray-Darling rainbowfish collected were only collected from one creek (Gulpa Creek) and the river habitats (Figure 3.10). A greater number of individuals of all stages of Murray-Darling rainbowfish were captured in 2005–06 than the previous seasons, and were captured at one additional site.

There was no significant difference in the mean adjusted abundance of carp larvae captured among years or habitat types (Table 3.5, Figure 3.12a). Whilst carp larvae were collected in all method types and all habitats sampled in all years, by far the majority of carp larvae were collected in drift nets in the river site (Morning Glory) downstream of the Forest (see Table 3.6). Similar proportions of the total number of juvenile carp were collected in SNE and light traps, and therefore both methods were analysed, however they displayed quite different effects of season and habitat. There was a significant effect of season for juvenile carp collected using the SNE method, with a significantly greater number of juveniles captured in 2005–06 than 2003–04 (p<0.001), but not 2004–05 (Table 3.5, Figure 3.12a). There was no effect of habitat. In contrast, there was no significant effect of season on the abundance of juvenile carp collected in light traps, but there was among habitat types (Table 3.5). A significantly greater number of juvenile carp were captured in lakes and wetlands compared to river and creek habitats (p<0.05). There was no significant difference between the abundance of goldfish larvae among seasons, but the abundance did differ among habitats (Table 3.5, Figure 3.12b), with a significantly greater abundance captured in wetland habitats than creek or lake habitats (p<0.05). The abundance of juvenile goldfish varied among both season and habitat type, with a significantly lower abundance of juveniles occurring in 2003–04 than the other two seasons (p<0.05); and a significantly lower abundance occurring in river habitats compared to the other floodplain habitat types (p<0.05; Figures 3.10, 3.12b). There were no significant effects of either season or habitat on the abundance of larvae or juvenile redfin perch (Table 3.5, Figure 3.12c). However, across all methods and years, a greater proportion of both larvae and juveniles were collected in creek habitats (Figure 3.10). The abundance of all stages of gambusia did not differ significantly among seasons, but did among habitat types (Table 3.5, Figure 3.12d), with a significantly greater abundance of all stages occurring in wetlands compared to the other habitat types (p<0.05). The abundance of juvenile oriental weatherloach varied significantly among seasons, with a greater abundance of juveniles occurring in 2005–06 than the previous two years (p<0.05, Table 3.5, Figure 3.12e). The abundance of adult oriental weatherloach captured in light traps (but not SNE) also varied among seasons with significantly greater number of adults occurring in 2003–04 than compared to 2005–06 (p<0.05). Whilst there was no significant effect of habitat for either juvenile or adult oriental weatherloach, across all methods and years a greater proportion were captured in creek habitats (Figure 3.10).

Use of the inundated floodplain was assessed on an ad hoc basis in each year depending on the availability of the habitat type at various sites throughout the Forest. Over the three years, very little overall effort was undertaken in sampling these temporary floodplain habitats and therefore the results presented should be treated with caution. In total, only five native and four introduced species utilised the inundated floodplain habitat (Table 3.7). Of most significance was the capture of 11 Murray cod larvae drifting in fast flowing flood waters from Tongalong Creek into Flat Swamp during the 2005–06 flood event. A greater diversity of both native and introduced species were collected during the 2005–06 flood than in previous years. Sampling effort on the floodplain however, was slightly greater during this year.

47


Figure 3.12: Mean (+ 1 standard error) log10 (x+1) abundance per habitat type for each sampling season for all developmental stages captured for (a) carp, (b) goldfish, (c) redfin perch, (d) gambusia and (e &f) oriental weatherloach. Black solid bars indicate creek habitats, grey bars indicate lake habitats, striped bars indicate river habitats and white bars indicate wetland habitats. For each developmental stage and species combination, only methods that captured greater than 40% of the total were analysed, and where greater than 30 individuals in total across the three sampling seasons.

48


Table 3.7: Raw abundance of species collected on inundated floodplain habitats throughout the study. Note: sampling was on an ad hoc basis, and was conducted at different sites, used different methods and sampling times, depending on availability of the habitat type.

2003–04 2004–05 2005–06

Cucumber Gully

Tongalong Creek

Budgee Creek

Barmah Lake

Flat Swamp

Steamer Plain

Method used SNE 3 3 3

LT 3 & 3 3 3

D 1

Oct & Dec Oct Feb Nov Nov Nov

Australian smelt Larvae 4 38 0 0 0 2

Juveniles 14 0 0 0 1 1

Adults 5 3 0 0 0 0

carp gudgeons. Larvae 9 0 0 0 0 0


Adults 143 0 0 0 0 0

flat-headed gudgeon 0 0 0 0 1 0

Murray cod 0 0 0 0 11 0

southern pygmy perch


carp Larvae 0 0 0 0 0 3


goldfish Larvae 0 0 0 0 0 1


gambusia Larvae 0 0 1 0 0 0

Juveniles 29 0 13 1 0 0

Adults 19 0 0 0 0 0

oriental weatherloach

Larvae 0 0 0 0 0 4

Spatial and temporal and variation in fish community

There were significant differences in the fish communities among the habitats using the presence/absence of all species and stages (Figure 3.13, R = 0.299, p<0.05) and larvae only (Figure 3.14, R = 352, p<0.01). River habitats contained a significantly different fish community than the other habitats (all species/stages: p<0.01; larvae only: p<0.05), as they were the only habitats where unspecked hardyhead, silver perch and golden perch occurred. Using the presence/absence of larvae only, there was also a significant difference between the fish communities in wetlands and creeks (p<0.05), with wetlands containing larval gambusia and carp gudgeons more frequently that in creeks (see also Figure 3.10).

There was no significant change in the fish community composition among seasons using the presence/absence of all species and stages or larvae only (all species and stages: p=0.67, R = -0.035; larvae only: p=0.378, R = 0.015; Figures 3.13 & 3.14). In general, the larval fish composition among years for each method type was fairly consistent (Figure 3.15). Australian smelt generally dominated the larval fish community across all methods and seasons, however, carp larvae dominated the catch in drift nets in 2004–05.

49


Figure 3.13: Two-dimensional solution for NMDS of the presence/absence of all species and developmental stages from all methods. The ordination is presented to show groupings of (a) season and (b) habitat. 50 random starts, maximum of 200 iterations, minimum stress = 0.17.

50


Figure 3.14: Two-dimensional solution for NMDS of the presence/absence of larvae only of all species from all methods. The ordination is presented to show groupings of (a) season and (b) habitat. 50 random starts, maximum of 200 iterations, minimum stress = 0.17.

51


Figure 3.15: Percentage composition by season of larvae collected in (a) drift nets, (b) hand trawls, (c) light traps and (d) SNE methods from all habitats. Numbers on the top of columns are the total abundance of larvae collected for that method in that season.

52


Discussion

Composition of the BM forest fish community and the importance of floodplain habitat diversity

Whilst much management attention has previously been focused on some of the B-M Forest’s other key ecological features, such as its importance as a waterbird breeding site and significant vegetation community, the importance of the Forest for native fish has been generally overlooked (see www.livingmurray.mdbc.gov.au). Indeed, the recent focus of fish management and research in the Forest has largely been on the use of the Forest by introduced fish species (particularly carp) (Stuart and Jones 2002, Brown et al. 2005, Stuart and Jones 2006a, b). Whilst native fish have undoubtedly undergone major declines since early European settlement of the region (see King 2005), they are still a highly diverse and important component of the Forest’s and River’s ecology. This study and McKinnon (1997) demonstrate that the native fish fauna of the region is diverse and of conservation significance. This study recorded a total of 15 species, ten native and five introduced, occurring throughout all of the sampled habitats. Stuart and Jones (2002), during surveys targeted at carp in the Forest, collected eight native and five introduced species; while extensive surveys conducted in the early 1990s collected 11 native and seven introduced species (McKinnon 1997). All species collected during this study were collected as eggs, larvae or juveniles indicating recent spawning and recruitment, and demonstrating their use of the Forest as a spawning and/or nursery area. Several species recorded in the Forest in recent surveys including bony herring, river blackfish, climbing galaxias, short-headed lamprey, brown trout and rainbow trout. (albeit in low numbers) were not found during this study. This suggests that these species are either in such low numbers that detecting early life stages is very difficult or that they do not use this region for spawning and recruitment. This study also recorded significant species that have either been absent or rarely recorded in the Forest. Unspecked hardyhead (Craterocephalus stercusmuscarum fulvus) were not recorded by either Stuart and Jones (2002) or McKinnon (1997), but have been collected by Jones and Stuart (2004) and were consistently recorded in large numbers in river habitats in this study. McKinnon (1997) and Jones and Stuart (2004) also reported only one individual of flat-headed gudgeon in each of their studies, while this study has recorded them in high numbers throughout a range of habitat types. Southern pygmy perch and Murray-Darling rainbowfish were not recorded in the Forest by McKinnon (1997) or Jones and Stuart (2004). In this study, southern pygmy perch were collected in all three years particularly in two wetland sites (Tarma and Flat Swamps), while Murray-Darling rainbowfish were collected in very low numbers in all three years. Determining the conservation status of rarer, and particularly rarer small-bodied fish species, requires additional targeted intensive surveys. These surveys should also target other threatened or significant species that have not been recorded in the Forest for some time (see Table 1.1), such as freshwater catfish, river blackfish and Macquarie perch.

Like previous surveys in the B-M Forest, this study found abundant populations of five introduced species utilising the Forest, all of which were able to successfully spawn and recruit during the three years of this study. Of particular concern is the apparent recent increase in the abundance of oriental weatherloach. McKinnon (1997) only recorded two oriental weatherloach during surveys in the early 1990s, while today they appear to be very common and distributed across a range of habitat types (Stuart and Jones 2002, Jones and Stuart 2004, this study). Unlike the native species, introduced species were found to both reside as adults and successfully spawn and recruit in all of the habitat types present within the Forest (Table 3.8). However, despite this apparent flexibility in spawning location and adult habitat use, all of the introduced species demonstrated a preference for spawning in floodplain habitats rather than in the main river channel. For example, gambusia postlarvae and juveniles were only recorded in their highest abundances in wetlands and in low abundances in river habitats. The combined impact of these introduced species on native fish is unknown; however, the limited research that does exist suggests that they have the capacity to significantly alter environmental conditions and trophic interactions, particularly within wetland habitats (King et al. 1997, Robertson et al. 1997, Shirley 2002, Stoffels and Humphries 2003, McNeil 2004).

53


Table 3.8: A summary of habitat use of each species and stage across the four habitat types. *** = Strong use (statistical difference), ** = some use, * = occurred but in very low numbers, blank cell = never recorded in that habitat. E/L = eggs/larvae, L = larvae, J/A = juvenile/adult. † indicates very low numbers captured and habitat use should be treated with caution.

Species Stage River Creek Lake Wetland

Australian smelt L ** *** * *

J/A * * * *

carp gudgeons L * * * ***

J/A * * * ***

flat-headed gudgeon L ** **

J/A ** ** * *

unspecked hardyhead L *** * * *

J/A *** * * *

Murray cod L *** **

J† ** *

golden perch E/L ***

silver perch E/L ***

southern pygmy perch J/A† * ** **

Murray-Darling rainbowfish J/A† ** **

carp L ** ** ** **

J * * ** **

goldfish L * * * **

J * ** ** **

redfin perch L * ** * *

J * ** * *

gambusia L * * * ***

J/A * * * ***

oriental weatherloach L * ** * *

J/A * ** * *

The fish fauna in floodplain rivers throughout the world is generally thought to be reliant on regular connectivity with the main river channel (Welcomme 1985, Copp and Penaz 1988, Junk et al. 1989, Sparks 1995). In Australia, whilst it is widely believed that floodplain habitats provide an important habitat for many fish species, there has been relatively little research on the use of floodplain habitats or wetlands by fish (Closs et al. 2006). The B-M Forest wetland is a complex floodplain river system which contains a variety of permanent and ephemeral aquatic habitat types, and therefore the diversity of fish present should also be quite high and distinct compared to the main river channel. This study found that in general, while a lower diversity of fish species was utilising floodplain habitats than the main river channel, a number of native and introduced species were highly reliant on floodplain habitats such as wetlands and the internal Forest creek systems, as both a nursery and/or adult habitat. Indeed, the whole fish community (across all life stages) and just the larval fish community was quite distinct between the wetland and river habitats (see Table 3.8). Meredith et al. (2002), McNeil (2004) and Arthington et al. (2005) also found a high use of floodplain habitats by fish in their studies. McNeil (2004) suggested that the distinct fish communities occurring throughout the Ovens River floodplain wetland types related to varying aspects of the flow regime, such as frequency and timing of inundation and connectivity with the main channel. Indeed, this also seems important for the B-M system, where wetlands or creeks that received more frequent connection to the river channel (such as Tongalong Creek and Flat Swamp) contained a different native fish assemblage than sites with less connectivity (such as Tarma Swamp and Tullah Creek). Further, inundation of the floodplain provides the only mechanism where individuals from populations isolated on the floodplain can disperse and colonise new waterbodies. This was particularly seen for southern

54


pygmy perch, a rare species within the Forest, which increased its distribution in the Forest during the 2005–06 flood event. The effects of river regulation on the Forest’s natural inundation regime are also likely to have already altered the diversity of habitat types available to fish (King 2005). As recommended for arid zone floodplain river fishes (Arthington et al. 2005), the variability between floodplain habitat elements suggests that the natural water level variability, magnitude, timing, frequency and duration of the connectivity between floodplain elements and the main channel (the natural flow regime Poff et al. 1997) needs to be maintained to sustain the distinctive and diverse aquatic habitats and to preserve and enhance the fish assemblages present within the Forest (Copp 1989, Grift et al. 2003, Arthington et al. 2005). These results also caution against imposing a single flow regime type and thereby conserving only for single habitat features, and also impeding free movement of fishes between the floodplain and the main channel (e.g. impassable regulators on Forest floodways) and between floodplain habitat features (e.g. levee banks).

Temporal variability in spawning time, duration and abundance of larvae and juveniles

Few studies have attempted to examine the spawning times for Murray-Darling fishes, and therefore any work that expands on this knowledge is worthwhile for biological understanding of the species and management. The species can be generally grouped into predominantly spring spawners, such as Australian smelt, carp, goldfish, and redfin perch or summer spawners, such as unspecked hardyhead and gambusia; with some species that can spawn across both periods such as silver perch and carp gudgeons (King et al. 2003). The fish fauna also exhibited both protracted (silver perch, carp gudgeons) and shorter (golden perch, Murray cod) duration spawning periods.

The duration of spawning time for each species was largely consistent across years, although the onset and end of the spawning periods did differ slightly for some species across the years. Murray cod spawning time was relatively similar between all three years, supporting previous studies that have demonstrated that they can spawn irrespective of flow conditions and that spawning is not enhanced or triggered by flooding (Humphries et al. 2002, Gilligan and Schiller 2003, King et al. 2003, Humphries 2005, Koehn and Harrington 2006). In this study, Australian smelt also showed a consistent spawning period from September to December, similar to that observed in the Ovens River (King et al. 2003), but much shorter than that observed by Humphries et al. (2002). Other species such as carp gudgeons, gambusia and carp significantly altered the timing and duration of their spawning across the three years of the study, suggesting that factors such as water temperature and water level fluctuations may play a greater role as a spawning cue for these species. Gambusia for example, seemed to delay the majority of their spawning in 2005–06 until after the flood levels had receded and the wetlands had returned to disconnected isolated waterbodies. However, carp provided the most extreme example of this, as although they were able to spawn in all three years, in a range of habitats and under differing environmental conditions, they were the only species that spawned on the mid-February 2005 flood event, having already spawned earlier in that season. The peak abundance of carp larvae in all years was also associated with flooding. Further, during the 2005–06 extended flood period, two peaks in larval abundance occurred coinciding with the major peaks in water levels. Although carp can spawn under a range of environmental conditions, carp are well known to enhance their spawning activity and recruitment during periods of floodplain inundation (Swee and McCrimmon 1966, Crivelli 1981, King et al. 2003, Stuart and Jones 2006a).

Whilst it is difficult to be certain of the relative role of various environmental factors in initiating spawning using only three years of data and given the variability in the environmental factors of the different habitat types, carp, goldfish, golden perch, silver perch, flat-headed gudgeon and unspecked hardyhead all spawned earlier during the 2005–06 extended flood period than in previous years. Additionally, the peak abundance of larvae or juveniles of some species (Australian smelt, hardyhead and flat-headed gudgeons) also seemed to alter during the 2005–06 season, with the peaks in that season tending to be occurring earlier and during the flood period.

55


Plate 16: unspecked hardyhead larvae.

Plate 17: carp gudgeon larvae.

Plate 18: carp larvae.

Occurrence of larvae and recruitment on the floodplain and the effect of the 2005–06 flood

Flooding is thought to provide a wide range of benefits to native fish including as a potential cue for spawning and adult migration, increasing the amount of food production available for all stages of fish, an increased ability to disperse to new habitats and as an important nursery area for young fish (Finger and Stewart 1987, Junk et al. 1989, Bayley 1991, Harris and Gehrke 1994, Turner et al. 1994, Humphries et al. 1999, King et al. 2003). This study supports recent suggestions (Humphries et al. 1999, King et al. 2003) that most native Murray-Darling fish do not require floodplain inundation to stimulate spawning and are able to spawn and potentially recruit each year, despite interannual variations in flow and daily water temperature. Even for species such as silver perch and golden perch, which were found to increase their spawning activity associated with the 2005–06 flood event, eggs and larvae of both species were also collected during relatively stable, regulated, within channel flow conditions in the two previous years (see Chapter 2, King et al. 2005). This is contrary to some

56


previous models and suggestions for spawning cues for some of the Basin’s fish fauna (see Humphries et al. 1999), and suggests that along with the flexibility in timing of spawning and peak abundance of larvae, the fish fauna may have some flexibility in their spawning requirements, and that they can take advantage of suitable conditions if they are available. Indeed, the only species found to preferentially spawn as a result of flooding was the introduced species carp, which was even able to spawn a second time in the 2004–05 breeding season on a warm February flood. Junk et al. (1989) in their discussion of the FPC highlighted that in temperate systems, the timing of spawning is more likely to be controlled by temperature and light regimes than by flooding cycles.

The FPC (Junk et al. 1989) and the ‘flood recruitment model’ (Harris and Gehrke 1994) also proposed that flooding indirectly enhances larval and juvenile survival by providing abundant food and habitat. Throughout the world, inundated floodplains are widely considered as important nursery areas for young fish (Finger and Stewart 1987, Junk et al. 1989, Bayley 1991, Harris and Gerhke 1994, Turner et al. 1994). Compared to the main channel, inundated floodplains contain a large diversity of habitat types with a distinctive array of environmental characteristics that may be suitable for larvae and juvenile fish. These characteristics would include: lower velocities (Holland 1986), high productivity of suitable food and abundant cover and protection from predators (Welcomme 1985, Junk et al. 1989). However, strong evidence supporting the use and dependence of Murray-Darling fishes on the floodplain for spawning and recruitment is limited (Humphries et al. 1999, King et al. 2003). This study found that a variety of fish species did utilise floodplain habitats to spawn and recruit, however, this occurred during both flooding and non-flood seasons. This was also found in the Ovens River floodplain (King et al. 2003), and suggests that the successful spawning and recruitment of these species is not reliant upon flooding. King et al. (2003) also observed that whilst carp gudgeons and gambusia spawned during both the high and low flow year, they only spawned when water had retreated from the floodplain proper, even in the low flow year. In B-M, carp gudgeons were able to spawn during both high flow and low periods, while the abundance of gambusia larvae was always greatest after flooding had ceased.

Most species that utilised the floodplain (whether under inundated or not) for spawning and recruitment either demonstrated a strong association to particular floodplain habitats as adults or were able to reside in a range of habitat types. However, species that normally reside in the main river channel, such as unspecked hardyhead, golden perch, silver perch and Murray cod, were either rarely or never found as eggs or larvae in floodplain habitats, even during periods of inundation. King et al. (2003) and Koehn and Harrington (2005) also rarely or never recorded Murray cod larvae in floodplain habitats in the Ovens River. This is contrary to Anderson (1919) who described the collection of large numbers of small Murray cod trapped in floodplain waterholes after floodwaters had receded. (Gehrke 1990a, b, 1991) suggested that golden and silver perch larvae may actively avoid floodplains due to low dissolved oxygen and high concentrations of toxic tannins and lignins. This may explain why golden perch and silver perch larvae were not recorded on the floodplain. However, given the abundances of eggs of these species found drifting in the main river channel, it suggests that spawning of these species may preferentially occur in the main river channel only. Whether or not there would be more use of the floodplain by all of these species as a spawning or a nursery area given a greater magnitude flood is uncertain.

The five introduced species recorded during this study were found to spawn and recruit in all years of the study and in all habitat types, but particularly in floodplain habitats. Whilst there was very little difference detected in the abundance of larvae of all of the species across the years, abundance of juvenile carp, goldfish and oriental weatherloach was significantly greater in 2005–06 flood year than compared to the previous two seasons, suggesting that floodplain inundation improved larval survival and recruitment for these species. The inundated floodplain environment is known to be an important spawning and recruitment habitat for carp in Australian lowland rivers (King et al. 2003, Stuart and Jones 2006), with the B-M Forest known to be a major recruitment area for carp (Crook and Gillanders 2006, Macdonald and Crook 2006, Stuart and Jones 2006a). This was also supported in this study with the greatest number of carp larvae captured drifting downstream of the Forest in 2005–06. Stuart and Jones (2006a) proposed a generic model for the movement and recruitment of carp in B-M, whereby spawning occurs in still, off channel waterbodies, larvae move to inundated floodplains as nursery areas and young-of-year actively leave the floodplain and drift downstream into the Murray River. However, our results show that carp can spawn and recruit in a range of floodplain habitat types throughout the Forest in both flood and non-flood periods, and that larvae can either remain on the floodplain or drift downstream from the Forest as larvae or juveniles. Whilst carp were able to spawn along with many native species during the spring flooding events that occurred during this study, they were also able to spawn a second-time on a natural, late summer flood in February 2005. This demonstrates the flexibility in spawning strategy for carp, and suggests that whilst management does need to acknowledge that floodplain inundation will improve the chances of successful carp recruitment, the benefits of flooding for native fish and other ecosystem components also

57


needs to be considered. Whilst the relationship between increased carp spawning and floodplain inundation at B-M Forest has received much attention, it seems that floodplain inundation may also play a vital role in improving recruitment of oriental weatherloach. While the impacts of this species are still uncertain, numbers have increased dramatically in the last decade through the Murray River system (Karolak 2006), and discussion of suitable control options and research into its breeding biology and impacts should be urgently undertaken.

Plate 19: Tullah Creek during summer low water conditions, February 2004.

Plate 20: Tullah Creek during flood conditions, September 2005. Note: tree to the left of photo is same tree in foreground of Plate 19.

58


Although a number of native species were found to utilise the floodplain for spawning and recruitment during a range of flow conditions, there was no dramatic increase in the total abundance of all species or the abundance of most of the native species as larvae or juveniles during the 2005–06 extended flood compared to the previous two years. The exceptions to this were the larger-bodied species residing in the main channel which did demonstrate increased spawning activity (golden and silver perch) or recruitment (Murray cod and trout cod) in 2005–06 than previous years (see Chapter 2), and southern pigmy perch whose abundance increased during the flood year but whose numbers were too low to draw a firm conclusion. The lack of a dramatic increase in the abundance of larvae or juveniles for the smaller native fish in this study associated with the extended duration flood seems to be contrary to a number of elements including: other overseas studies in temperate systems (Finger and Stewart 1987, Killgore and Baker 1996), the predictions of the flood pulse concept (Junk et al. 1989), the flood recruitment model for Murray-Darling fishes (Harris and Gehrke 1994), and to early observations of juvenile fish in Murray-Darling floodplains (Anderson 1919, Cadwallader 1977). There are a range of possible explanations why this may have occurred, and these are discussed below:

• recruitment of these native Murray-Darling fish is not enhanced by flood periods,

• recruitment of these native Murray-Darling fish is enhanced by flood periods but cumulative changes to the region’s fish fauna, Forest and the river’s flow regime since European settlement have restricted the use of the floodplain for spawning and recruitment,

• the optimal environmental conditions for enhanced recruitment and use of the floodplain did not occur in 2005–06 event,

• conditions during the previous two seasons did not have a dramatic enough contrast in environmental conditions as 2005–06 to cause a substantially different effect on spawning or recruitment success,

• the spawning and recruitment of native fish was enhanced during the 2005–06 event, but that given the greater volumes of water present, fish were essentially more diluted across the floodplain and our sampling intensity during the flood periods was not sufficient to detect an effect, or that,

• examination of the success of any one year’s recruitment is better assessed by examining the abundance of adults at the beginning of the following breeding season (not within season as was conducted during this analysis).

The B-M region has also undergone substantial alterations to the natural ecosystem since European settlement. There has been commercial overfishing, removal of instream woody debris, changes to the natural flooding and drying cycles, construction of small regulators and levee banks, cattle grazing and changes in the vegetation structure and dominance (Chesterfield 1986, Bren et al. 1987, Leslie 2001, King 2005, Stuart and Jones 2006a). These factors combined may have contributed to a reduction in the use of inundated floodplains in the B-M Forest for spawning and recruitment. This is also supported by anecdotal evidence from the native commercial fishery in the 1880s, where large numbers of Murray cod and golden perch in spawning condition were annually harvested, mainly from Moira Lake (Leslie 1995). Some of these factors, such as flow regime and vegetation changes could be addressed by improved water and land management of the Forest, and may therefore also improve the use of the inundated floodplain for fish recruitment.

King et al. (2003) proposed a new model suggesting the optimum environmental conditions for use of the floodplain for fish recruitment. The model suggested that the optimum conditions for use of the floodplain would occur when floods and warm temperatures were coupled, the flood pulse was predictable, the rates of rise and fall of the flood were slow, the duration of the inundation period was weeks to months in length, and that a large area of the floodplain was inundated (related to magnitude of the flood). The B-M 2005–06 flood could be viewed as having met the first four of these conditions, however the amount of floodplain inundation may be relevant. The 2005–06 flood event inundated around only 50% of the Forest, and the magnitude of the flood peak was significantly lower than can naturally occur. It is difficult to know what the effect of a larger flood would have been on the recruitment of these species, and this requires further monitoring in future years. Whilst the overall shape of the hydrographs in each of the three years was substantially different, the overall magnitude of the flooding was about the same. Therefore it may be possible that 2005–06 event was not different enough from the previous years to statistically demonstrate an effect for many species. Further monitoring of extreme years such as droughts and floods of varying magnitudes should enable these factors to be separated.

59


It is also plausible that the spawning and recruitment of native fish was enhanced during the 2005–06 flood event, but since there was a larger volume of water present and we maintained the same sampling effort throughout all conditions, fish may have been less concentrated and therefore more difficult to capture, and our sampling intensity was not sufficient to detect an increase. One solution to this is to not only examine the abundance of larvae and juveniles within the season of interest, but also examine the surviving recruits and adults present at the start of the following season, especially for short-lived species such as Australian smelt, hardyhead etc. Brown and Ford (2002) found that the best model that explained the relationship between flow regime and native and introduced fish was one that incorporated the flow regime from the previous year, as this affected reproductive success and the fish community sampling in the following winter/spring. Unfortunately, this type of analysis is not possible at this stage, although sampling has occurred in 2006–07 (and hopefully will be maintained in future years).

Conclusion

This study has demonstrated the significance of the B-M Forest region for native fish, and has emphasised the importance of a diversity of floodplain habitat types for the maintenance of a diverse native fish community (Closs et al. 2006, Phillips 2006). However, in contrast to the current paradigm of the importance of floodplains for fish recruitment, we found that the 2005–06 flood did not demonstrate a significant increase in the abundance of larvae and juveniles of all the native species, especially the smaller native species, compared to the previous two years. However, while the 2005–06 flooding did not achieve all of the expected outcomes, it did achieve some substantial benefits for native fish and was certainly not detrimental to them. The flooding triggered an increase in the spawning activity in golden perch and silver perch, and also indirectly increased the number of young-of-year Murray cod and trout cod (see Chapter 2). The flooding also played an invaluable role in habitat maintenance and connectivity of floodplain habitats such as wetlands and creeks for a variety of fish residing and recruiting on the floodplain. Indirectly it also probably provided a boost of nutrients and prey items in returning waters to permanent waterbodies such as the main channel and wetlands. We believe that this study has provided invaluable insights into the impact of water management on fish recruitment at B-M Forest. However, as flow and other environmental conditions can vary substantially across years, longer-term monitoring across a range of environmental conditions and flow regimes, including managed flow events of different types, needs to occur before strong conclusions on the relationship between flooding and fish recruitment can be established.

Summary of Key Findings• B-M Forest region was found to contain a high diversity of native fish and is a significant area for native fish

conservation.

• This study recorded breeding and adult residence of a number of significant species that have either not been recorded in the Forest for some time or have been recorded in low numbers, such as unspecked hardyhead, Murray-Darling rainbowfish, southern pygmy perch and trout cod.

• Native fish utilised a wide range of habitat types within the river channel and the floodplain, with some species showing strong preference for particular habitat types. The maintenance of the diversity of the Forest’s aquatic habitat types is seen as the key to ensuring the conservation of native species within the Forest.

• Abundant populations of five introduced species were found in the Forest, all of which were able to successfully spawn and recruit in each year of the study. All of the introduced species were found to reside as adults and successfully spawn and recruit in all of the habitat types within the Forest. However all species demonstrated a preference for spawning in floodplain habitats rather than in the main river channel. Of particular concern is the rapidly increasing abundance and distribution of oriental weatherloach throughout the Forest.

• Carp were found to spawn as a direct association with flooding, although they were also able to spawn under low flow conditions. Additionally, whilst there was no significant increase in larvae carp abundance, the high flow year did result in significantly greater numbers of juveniles.

• A variety of native fish species did utilise floodplain habitats to spawn and recruit, however, this occurred during both flood and non-flood seasons. Most native fish were found to not require over-bank floods to stimulate spawning despite interannual variations in flow and water temperature; however, many species did alter the timing and extent of their spawning period in the 2005 flood season.

60


• Counter to general predictions about the role of flooding in fish spawning and recruitment, there was no dramatic increase in the total abundance of all species combined or the abundance of most native species (especially for the smaller species) as larvae and juveniles associated with the 2005 flood event. The flood year did, however, indirectly increase the number of juvenile southern pygmy perch in the Forest. The introduced species, carp, goldfish and oriental weatherloach also demonstrated an increase in recruitment strength associated with the 2005 flood event.

• This study demonstrates that 2005 flood event (including use of the environmental flow) did achieve some positive benefits for native fish. The most obvious were the increased spawning activity in golden perch and silver perch and the increased number of young-of-year Murray cod, trout cod (see Chapter 2) and southern pygmy perch. Flooding also played an invaluable role in habitat maintenance and connectivity of floodplain habitats such as wetlands and creeks for a variety of fish residing and recruiting on the floodplain. Indirectly it also probably provided a boost of nutrients and prey items in returning waters to permanent waterbodies such as the main channel and wetlands.

Key Research Gaps• Conduct targeted surveys of rarer significant species within the Forest to determine their status and

distribution.

• Continue monitoring the effects of various flow regimes (whether managed or natural events) on fish spawning and recruitment, to better determine key hydrological and other environmental features which are required for enhancing native fish spawning and recruitment.

• Research specifically the ecology of oriental weatherloach in the Forest particularly its effect on native fish communities.

• Assess indirect effects of flooding such as nutrient loads and densities of prey items in returning waters to permanent waterbodies such as the main channel and wetlands.

61

4. DIEL AND SPATIAL DRIFTING PATTERNS OF EGGS AND LARVAE

Most of the data presented in this chapter has been published as:

Tonkin, Z., A. King, J. Mahoney & J. Morrongiello (2007). Diel and spatial drifting patterns of silver perch Bidyanus bidyanus eggs in an Australian lowland river. Journal of Fish Biology. 69, 1-5. doi: 1111/j.1095-8649.2006.01281.x (see Appendix 1).

IntroductionDownstream drift of eggs, larvae and juveniles is a common dispersal strategy employed by many riverine fish species (Penaz et al. 1992). There is increasing awareness that numbers of drifting eggs and larvae can show considerable variation in both time and space. Faulkner and Copp (2001) found that the smallest changes in bed sampling position result in significant effects on drift collection. Many studies have reported drifting eggs and larvae having distinct diel patterns (e.g.Gadomski and Barfoot 1998, Copp et al. 2002, Reichard et al. 2002, Zitek et al. 2004, Humphries 2005). These patterns are often both taxon and developmental stage specific (e.g. Brown and Armstrong 1985, Gadomski and Barfoot 1998), with studies such as Auer and Baker (2002) and Zitek et al. (2004) illustrating the need for such specific knowledge to accurately estimate drift abundances of fish embryos and larvae. With the aim of gaining specific knowledge on spatial and temporal drifting patterns, the present study assessed densities of drifting eggs and larvae over three consecutive 24h periods.

MethodsCollections were made using 1.5 m long passive drift nets with a 0.5 m diameter mouth opening, constructed of 500 µm mesh, tapered to a removable collection jar. A General Oceanics Inc. (Florida, USA) flow meter was fixed in the mouth of each drift net to determine the volume of water filtered, thus enabling raw catch data to be standardised among all nets to the number of eggs per 1000 m3 of water filtered.

Drift samples were collected at one site in the Barmah-Millewa Forest region of the mid-Murray River (Ladgroves Beach, 35º51.677, 145°20.773), over three consecutive 24 hr periods in November 2004. Samples were taken at three positions across the main river channel (108 m channel width; daily discharge during sampling = 10200 ML): 1. Near-shore right bank (distance from R. bank 21 m; depth 3.3 m); 2. Mid-channel (distance from R. bank 54 m; depth 3.9 m) and; 3. Near-shore left bank (distance from R. bank 85 m; depth 3.0 m). At each position, two drift nets were attached to a pole (see Figure 2.1) to sample the surface and bottom 50 cm of the water column, giving a total of six nets. The nets were set at 21:00 on the 22nd November 2004, emptied every 4 hr, and removed on the final sample collection at 21:00 on the 25th November 2004. This gave a total of three replicates for each time period and top/bottom net setting (three top/bottom net settings over three 24 hr time periods).

At the time of collection, any eggs were removed alive from the samples and subsequently hatched, enabling a positive identification of either silver or golden perch. Samples were then preserved in 95% ethanol and returned to the laboratory where any remaining eggs and larvae were removed, identified and counted under a dissecting microscope.

At the commencement of sampling, measures of dissolved oxygen (4.49–7.88 ppm), electrical conductivity (54.7–68.8uS/cm), pH (7.05-8.09) and water temperature (20.3–23.9°C) were recorded every hour throughout the experiment.

Due to large variations in numbers of eggs and larvae collected, resulting in a non-normal distribution of data, non-parametric measures were used in analysis. Overall differences in egg numbers between the six time periods as well as between the three positions were tested using the Kruskal-Wallis test, and then if significant, Mann-Whitney tests were used to identify pair-wise differences. Differences between top and bottom nets were analysed as the number of eggs captured per 24 hr using Mann-Whitney tests.

62


Results and DiscussionA total of 239 larvae and 3744 eggs were collected during the three-day period (Table 1). By far the most abundant species/stage collected was silver perch eggs, representing 92% of eggs collected. As such, results and discussion on drifting eggs in this chapter will primarily focus on this species.

Despite large variations in standardised egg numbers among nets, all showed a distinct diel pattern (Figure 4.1). There were significant differences in drifting silver perch egg densities among different time periods for both top and bottom nets (Kruskal-Wallis, P<0.001; P=0.016 respectively), with a significantly greater number of eggs collected in the period 21:00-01:00 (Figure 4.1; Mann-Whitney, all P<0.05).

Few studies have investigated diel patterns for drifting eggs, with diel patterns recorded in the present study likely to be a result of specific adult spawning times. The diel pattern of silver perch egg drift recorded in the present study is in accordance with early observations of spawning behaviour in earthern ponds for the species by Lake (1967a). He observed courtship from 16:00 h until sunset, with eggs collected 3-5 h after sunset. Given this observation, and the short hatching time of the species, egg abundances in the present study suggest a peak spawning period during 21:00-01:00 h on each of the three days sampled (sunset time: 20:15). Reasons for spawning at such specific times are speculative, but could be related to favourable hatch times, or, like many coral reef species that have pelagic eggs and larvae, a possible mechanism by adults to make their eggs less subject to predation by visual planktivores (e.g. Helfman 1993).

Table 4.1: Species composition and raw numbers of eggs and larvae collected over the sampling period.

Common name Species Total

Australian smelt Retropinna semoni Eggs 156

Larvae 179

silver perch Bidyanus bidyanus Eggs 3459

Larvae 9

golden perch Macquaria ambigua Eggs 46

carp gudgeons Hypseleotris spp. Larvae 4

flat-headed gudgeon Philypnodon grandiceps Larvae 37

Murray cod Maccullochella peelii peelii Larvae 10

Total larvae 239

Total eggs + larvae 3744

Plate 21: Sets of drift nets at position 2 (mid-channel) and 3 (near-shore left bank) in the Murray River.

63

4. Diel and spatial drifting patterns of eggs and larvae

Drifting silver perch egg concentrations were significantly higher in bottom nets when compared to top nets at all three positions (Figure 4.2a; Mann-Whitney, all P<0.05). This was particularly evident for positions 1 and 3, which were located closer to the bank and filtered lesser amounts of water. This may be due to a possible near-bank spawning preference of the species, or due to entrainment of the eggs in lower water velocities at the edge of the channel. Higher water velocities in surface nets relative to corresponding bottom nets has been reported in a similar study by Oesmann (2003) (2003), while both Nezdolily (1984) and Copp et al. (2002) found concentrations of drifting eggs to be highest in areas of lower water velocities. This is likely to be due to eggs being by nature, entirely passive particles (Copp et al. 2002), and is a possible explanation for the results presented in this study for two reasons. Firstly, observations on silver perch eggs by Lake (1967b) reported that the fine mat-like chorion of the eggs, readily collects small clay particles, causing eggs to have increased negative buoyancy and causing settling to the bottom in slow and still water. Araujo-Lima and Oliveira (1998) also proposed increased negative buoyancy in the eggs of Characiformes species can be caused by adhesion of fine sediments. Secondly, the amount of organic matter (other negatively buoyant passive particles) observed in the bottom nets was always higher than that found in corresponding top nets. Slower water velocities may also explain the significant difference in standardised silver perch egg numbers between bottom nets across sites (Kruskal-Wallis, P=0.012). The bottom net at position 3 showed a significant difference to the bottom net at position 2 (Mann-Whitney, P<0.001) with the former located in much slower velocities.

Silver perch egg densities also gradually increased throughout the three-day study period (Figure 4.3), perhaps due to an increase in spawning activity with a gradual increase in temperature. This increase in temperature may have also been the trigger for golden perch spawning, whose eggs were only present at lower levels, and restricted to time periods on the final day of the study.

Figure 4.1: Mean (± 1 SE) standardised density (number per 1000 m3) of silver perch eggs captured during each consecutive 4hr-time period over 24 hours. Numbers relate to net position. 1. Near-shore right bank; 2. Mid-channel and; 3. Near-shore left bank. T = top net; B = bottom net. Horizontal bars beneath the plot represent corresponding night (black) and daylight (white). Note varying scale of the y-axis among panels.

64


Figure 4.2: Mean (+ 1 SE) standardised density (number per 1000 m3) of (a) silver perch eggs and (b) total larvae captured in a 24hr period for top (clear bars) and bottom nets (filled bars) at each position across the river channel (1. Near-shore right bank; 2. Mid-channel and; 3. Near-shore left bank).

Figure 4.3: Water temperature and standardised density (number per 1000 m3) of silver perch eggs captured during each consecutive 4hr-time period over 24 hours at position 1 for top and bottom nets.

65

4. Diel and spatial drifting patterns of eggs and larvae

The highest densities of drifting larvae were collected during the night (Figure 4.4; Kruskal-Wallis, P=0.026). Raw numbers of larvae collected were dominated by Australian smelt (75%) and flat-headed gudgeon (15%), with only small numbers of Murray cod, silver perch and carp gudgeons present (Table 4.1). All individuals of Murray cod larvae were collected during the first 24hr period of the study, with the highest densities occurring in the first 4-hr period after sunset (Figure 4.4). Despite this observation, low numbers and high variability meant this result was insignificant (Kruskal-Wallis, P=0.121). Despite this, our observations are in accordance with results from Humphries (2005) who also found densities of drifting Murray cod larvae to be highest between sunset and sunrise. Peak numbers of Australian smelt larvae occurred around the hours of sunrise (Figure 4.4). Drifting larvae of various species have been shown to exhibit strong diel peaks, with many authors suggesting that such diel patterns in dispersal strategies may be to minimise mortality by predation (e.g. Corbett and Powles 1986, Gadomski and Barfoot 1998, Zitek et al. 2004).

Figure 4.4: Mean standardised density (number per 1000 m3) of larvae captured during each consecutive 4hr-time period over 24 hours including mean species composition for each. Horizontal bars beneath the plot represent corresponding night (black) and daylight (white).

Like the spatial drifting patterns of silver perch eggs, densities of the total drifting larvae were highest at positions 1 and 3 when compared to position 2 (Figure 4.2b; Mann-Whitney, both P<0.05) although there was no significant difference between top and bottom nets (Mann-Whitney, P=0.054). This was particularly evident for the small number of silver perch larvae collected, all of which had only recently hatched. As discussed, the highest egg numbers for this species were found in these slower flowing areas, therefore one would assume the highest abundances of newly hatched larvae to also be in the same area.

The diel component of this study has shown that the distribution of larvae and eggs was non-uniform through both time and space. Results suggest a peak spawning time of silver perch between 21:00-01:00h, and a propensity to drift in higher densities near shore and at the bottom of the water column. This study did highlight that greater densities of larvae drift at night even though during the study the catches of drifting larvae were not very high, Such non-uniformity in spatial and temporal distribution needs to be considered when estimating drift abundances of both eggs and larvae.

66


Summary of Key Findings• Peak spawning time of silver perch occurs between 21:00-01:00 h.

• Eggs have a propensity to drift in higher densities near shore and at the bottom of the water column.

• Highest numbers of drifting larvae were captured at night.

Key Research Gaps• Gaining further knowledge of species specific drifting pattern for other species.

• Repeating the experiment in another season would strengthen results.

67

5. TESTING A MODIFICATION TO A STANDARD PASSIVE DRIFT NET TO CAPTURE DRIFTING ICHTHYOFAUNA

IntroductionThere is a large diversity of gear types currently available for the collection of larval and juvenile fish in freshwater habitats (King and Crook 2002), with drift and trawl nets being the most common methods employed for sampling drifting egg, larval and juvenile stages. In recent times there has been much documentation on the downstream drifting dispersal phase of eggs and larvae for a number of species in Australia’s Murray-Darling basin (e.g. Humphries 2005, King et al. 2005, Koehn and Harrington 2005, Tonkin et al. 2007). There have been two main techniques employed to sample such drifting stages. Koehn and Harrington (2005) and later Stuart and Jones (2006) employed Elton glass eel nets set in reverse. Whilst having good capture rates, particularly of later stage larvae and juveniles due to the funnel insert at their mouth, quantitative assessment of numbers was difficult and the large sample volume means processing, particularly when sorting for small eggs and larvae in the laboratory, is difficult. The vast majority of later studies (e.g. King et al. 2005; Tonkin et al. 2007; Humphries 2005) have employed a standard 1.5 m long passive drift net, with a 50cm mouth, tapering to a collection jar. These 500µm mesh nets have a flow meter situated in their mouth, enabling quantitative assessment of the catch, as well as providing manageable sample sizes. This, along with their simplicity to deploy means these nets are able to be set at various depths and at higher numbers of replicates.

Whilst the standard drift nets used in these studies have proved very effective in the capture of drifting eggs and early larval stages of riverine fish, their efficiency in slow flowing water and at detecting downstream dispersal of larger individuals is somewhat limited. The open mouth of the passive drift nets, may allow fish with good swimming abilities, particularly in slow flowing areas, to swim out of the net. Adjustments to standard nets which to accommodate such situations, whilst having minimal effects on the efficiency and advantages of the original nets previously mentioned would therefore be highly desirable.

In an attempt to increase the capture rates of larvae in slower flowing areas as well as larger, stronger swimming juvenile stages, a standard passive drift net containing a conical insert at its mouth was deployed with catches compared to those of the standard net.

MethodsStandard passive drift nets such as those used by King et al. (2005) are 1.5 m long with a 0.5 m diameter mouth opening, constructed of 500 µm mesh and taper to a removable collection jar (Figure 5.1a). This standard net was modified at its mouth with an additional 1 mm rigid mesh conical insert which was attached by a Velcro fastening system. The insert formed a funnel, which extended 30 cm into the net and tapered to a 20 cm opening (Figure 5.1b). It was hypothesised that this modified net combined the advantages of the standard drift net with a funnel insert such as those in the Elton glass eel nets. The Velcro fastening system of the net design has the additional advantage in that it allows a quick conversion to either a passive or new net type.

Once a month from October until February in 2004/2005, drifting stages of fish were sampled in the Barmah-Milewa forest region of the mid-Murray river, Australia, at four creek sites (Tullah Creek, 35°54.730, 145°02.512; Gulpa Creek, 35°42.352, 144°55.278; Tongalong Creek, 35°50.701, 145°13.591; and Budgee Creek, 35°56.128, 144°58.169) (see Figure 1.2). In 2005/2006, sampling was conducted at one creek site (Budgee Creek) and one site on the Murray River (Morning Glory, 36°04.765, 144°57.553) to expose the net type to higher water velocities. At each site the modified ‘funnel net’ was placed alongside the standard passive drift net in order to assess its efficiency over a range of conditions and species. Both nets had a General Oceanics Inc. (Florida, USA) flow meter fixed in the mouth to determine the volume of water filtered thus enabling raw catch data from each net to also be standardised to catch per 1000m3 of water filtered. Comparisons between the two net types were analysed for both numbers and lengths. Both raw and adjusted egg, larvae and juvenile/adult numbers were log transformed and analysed using a two-way analysis of variance (ANOVA), using net type and site as the factors for 2004/2005 and 2005/2006 data. Length comparisons between the two net types for all fish as well as for cyprinids only (carp and goldfish) were made using the Kolmogorov-Smirnov two sample test on the two years of data combined.

68


Figure 5.1: Side view illustrations of (a) the standard passive drift net and (b) the funnel net, highlighting the mesh conical insert.

Results and DiscussionDespite the two net types being placed alongside one another, identical flow rates at the mouth of each net was not always possible. Despite this, for all sites, on average the standard net filtered far more water than its corresponding funnel net (Mean ± 1SE 11.7 ± 5.3 times the volume). This is not an unexpected outcome given that the mesh insert at its mouth increases the rate of clogging, particularly during times of high suspended particulate matter such as during a flow rise.

A combined total of 2021 eggs, larvae and juvenile/adults from 10 species were collected over the two years of sampling (see Table 5.1). There were significantly higher raw numbers of eggs captured in the passive drift net than the corresponding funnel net in 2005/2006 (F=1, 11.09, p<0.01; Figure 5.2) but not in 2004/2005 (F=1, 0.00, p>0.05). This difference between years is due to high egg numbers in the river site as shown by the significant effect of site in the 2005/2006 data (F=1, 10.07, p<0.01). The differences in egg numbers between the two nets at this site is likely to be exaggerated due to a poor set of the funnel net during the time of a large golden perch and silver perch spawning event (see Table 5.1). There was no significant difference in egg numbers between nets in either year for adjusted data (2004/2005, F=1, 1.46; 2005/2006, F=1, 0.52; p>0.05 respectively). Despite this, our raw data result is expected given that the net filtering higher volumes of water (ie. the standard net) would be expected to capture higher numbers of eggs, which drift in a passive manner (Copp et al. 2002).

There was no significant difference in both raw (2004/2005, F=1, 0.01; 2005/2006 F=1, 0.20; p>0.05) and adjusted (2004/2005, F=1, 2.00; 2005/2006 F=1, 0.00; p>0.05) numbers of larvae captured between the net types although raw numbers were slightly higher in the standard net than those in the funnel net Adjusted data highlighted that this was due to a greater amount of water filtered (Figure 5.2).

There was no significant difference in raw numbers of juveniles and adults captured between the two nets for either year (2004/2005, F=1, 2.81; 2005/2006, F=1, 0.05; p>0.05 respectively) but numbers were slightly higher in the funnel net Figure 5.2). Examination of adjusted numbers sampled in 2004/2005 showed the funnel net captured a significantly greater number of juvenile/adults than its corresponding standard net (F=1, 6.32, p<0.05). Fish captured in the funnel net were significantly larger than those in the standard drift net (Figure 5.3; Chi-square 2, 22.57, p<0.001). The funnel net was able to catch fish in excess of 100 mm SL, while catches of fish in excess of 10 mm in the standard passive drift net declined rapidly with increasing length (Figure 5.3). This trend was well represented in catches of cyprinids with fish from the funnel net being significantly larger than

69

5. Testing a modification to a standard passive drift net to capture drifting Ichthyofauna

those in the standard drift net (Figure 5.4; Chi-square 22.57, 2 d.f, p<0.001). In a similar trend, adult oriental weatherloach, which is rarely recorded in the standard drift nets, were also captured in far greater numbers in the funnel net (Table 5.1). The funnel insert on the modified net is therefore successfully stopping or at least reducing the escape of fish with stronger swimming abilities.

Table 5.1: Raw numbers of all species captured from all sites by the two net types during the two years of sampling.

Common name

Standard drift net Funnel drift net

Eggs Larvae Juv / Adult Eggs Larvae Juv / Adult

Australian smelt 12 31 10 15

carp gudgeon 64 80 53 122

flat-headed gudgeon 69 1

Murray cod 17 8

golden perch 441 4 1*

silver perch 517 3*

carp 309 5 150 13

goldfish 3 6 39

redfin perch 1

oriental weatherloach 1 8 38

* Bad set of Funnel net during golden and silver perch spawning event

Figure 5.2: Numbers of eggs, larvae and juvenile/adults captured per trip in 2004/2005 and 2005/2006 by the standard passive drift net (grey bars) and funnel net (white bars). Catches are expressed as (a) log transformed mean (+1SE) raw catch and (b) log transformed mean (+1SE) standardised catch expressed as catch 1000m-3.

70


Figure 5.3: Length frequency histogram for all fish captured by the standard passive drift net (grey bars) and funnel net (white bars). SL = standard length.

Figure 5.4: Length frequency histogram of all cyprinids (carp and goldfish) captured by the standard passive drift net (grey bars) and funnel net (white bars). SL = Standard length.

Overall, despite no significant difference in egg and larval collection between net types, the funnel net does not appear to be as efficient in the capture of eggs and larvae as the standard passive net due to the increased rate of clogging which leads to a reduction in water volume sampled. Factors such as fishing time and the amount of suspended particulate matter present would have to be considered prior to deployment. The funnel net is however, much more effective at capturing greater numbers of larger individuals. There are sampling situations which will be better suited to the new modified drift net. For example, researchers who wish to specifically target spawning events of riverine fish which are detected by the collection of eggs and early larvae in a fast flowing river channel (e.g. King et al. 2005), are best to employ the standard passive nets purely due to their capacity to sample greater volumes. Alternatively, if the research is aimed at targeting a wider range of sizes of fish whose drifting stages can extend from early larvae to juveniles, or the reaearch is to sample very slow flowing areas, then the new funnel net or a combination of both may be appropriate. Researchers must therefore consider both what species and stages they wish to target, as well as the likely flow and particulate matter conditions occurring, and choose their net type accordingly.

71

5. Testing a modification to a standard passive drift net to capture drifting Ichthyofauna

Summary of Key Findings• Standard passive drift net slightly more effective in the capture of eggs.

• Modified funnel net more effective at the capture of larger individuals.

• Net design makes the modification simple to apply or remove depending on the situation.

Key Research Gaps• A comparison of differences in volumes sampled between the net types under a range of different conditions

would be desirable.

72


CLOSING SUMMARY

Predicting the response of the Murray-Darling Basin’s fish populations to environmental flow releases has been difficult as our understanding of the relationship between flow regime and its influence on the early life stages of the fishes is poorly known (Humphries and Lake 2000). The B-M Forest was an ideal location to evaluate the response of fish spawning and recruitment to various flow management scenarios and generate critical knowledge of the flow requirements of native fish. The study reported here was conducted over three spawning seasons, and fortuitously included a 513 GL environmental water release in 2005. As such it has provided invaluable insights into the effects of water management on fish spawning and recruitment not only in the B-M Forest, but should be generally applicable for many areas of the Murray Darling Basin.

This study confirmed that the B-M Forest is an important region for native fish. Native fish were found to utilise a wide range of habitat types within the main river channel and the floodplain, for both residence and recruitment, and during both flood and non-flood conditions. This finding supports the need for the diversity of the Forest’s aquatic habitat types to be maintained to ensure the conservation of the region’s native fish fauna. The B-M forest contains a high diversity of native fish including a number of significant species that have either not been recorded in the Forest for some time or have been recorded in low numbers (e.g. unspecked hardyhead, Murray-Darling rainbowfish, southern pygmy perch and trout cod). Such findings have highlighted the need to conduct targeted but comprehensive surveys for rarer significant species to determine their status and distribution. Most native fish (predominantly the smaller bodied species), were found not to require overbank floods to stimulate spawning and were able to spawn and recruit at similar levels each year despite interannual variations in flow and water temperature. Overall, the 2005–06 flood did not demonstrate a significant increase in the abundance of larvae and juveniles of all the native species (especially the smaller species) compared to the previous two years. However the 2005–06 environmental flooding did achieve some substantial benefits for native fish.

Careful monitoring and scientific knowledge was included in the planning and management of the 2005 B-M environmental water allocation which allowed fish spawning and recruitment to be considered in the management of the event. As a result, this study demonstrated that flooding can indeed influence the spawning and recruitment success of golden perch, silver perch, Murray cod and trout cod. Golden perch and silver perch increased their spawning activity in the main river channel during the flood of 2005–06 compared to the previous two seasons. Murray cod and trout cod appeared not to increase their spawning activity in 2005–06, but rather increased the abundance of young-of–year fish resulting from the flood year. Our results, then, suggest that flooding may be an important component of the life history strategies of these species either as a direct spawning cue or by increasing the survival of young and hence recruitment. Along with the increased spawning and/or recruitment of the four species mentioned, the flood year also increased the abundance of juvenile southern pygmy perch. Furthermore, other species such as carp, goldfish, golden perch, silver perch, flat-headed gudgeon and unspecked hardyhead all appeared to alter the timing and duration of their spawning period to align with the flood conditions. Flooding also played an invaluable role in habitat maintenance and connectivity of floodplain habitats such as wetlands and creeks for a variety of fish residing and recruiting in the Forest. Indirectly it probably also provided a boost of nutrients and prey items in returning waters to permanent waterbodies such as the main channel and wetlands. This project did not assess these potential indirect mechanisms, and hence this is a worthwhile area of future research. This research has demonstrated that it is possible to optimise and manage flows to improve native fish spawning and recruitment opportunities. It is still unclear, however, as to which key components of the 2005–06 flood triggered the increased spawning activity of golden and silver perch. Results from this study must be viewed as a single event in a three year period and, as flow and other environmental conditions can vary substantially across years, longer-term monitoring across a range of environmental conditions and flow regimes, including both managed and natural flow events of different types, needs to occur before strong conclusions on the relationship between flooding and fish recruitment can be confidently established. Furthermore, in order to determine the subsequent cohort strength and survival from such spawning responses, follow-up monitoring of young-of-year is essential.

Abundant populations of five introduced species were also found in the Forest; all of which were able to successfully spawn and recruit in each year of the study. Of particular concern is the rapidly increasing abundance and distribution of oriental weatherloach throughout the Forest. In this iconic wetland, with such a current population explosion of this invasive species and its rapid range expansion downstream (pers. obs), further research on the effects of this species on native fish communities, its basic biology and potential control

73

Closing Summary

options is urgently recommended. This study also found that all of the introduced species demonstrated a preference for spawning in off-channel habitats such as wetlands, creeks and the inundated floodplain and subsequently carp, goldfish and oriental weatherloach all showed an increase in recruitment strength associated with the 2005–06 flood event.

As additional components of the project, we also investigated the diel drifting behaviour of riverine fish over three days and evaluated a new method for capturing drifting fish in slower water velocities. The intensive diel study suggested that the highest numbers of fish eggs and larvae were drifting at night, and that the peak spawning time of silver perch occurred between 21:00 and 01:00h. Such non-uniformity in spatial and temporal distribution needs to be considered when estimating drift abundances of both eggs and larvae (Tonkin et al. 2007). Repeating the experiment in another season should strengthen these results. A second methological study within this project involved testing a new drift net design for use in lower water velocities. We found that the standard drift net was more effective in the capture of eggs and early larvae; however, the new modified funnel net was more effective at the capture of larger individuals. The results from both of these methodological studies will prove valuable in future sampling designs, along with improving our estimates of fish fauna drift abundances.

This study has demonstrated the significance of the B-M forest region for native fish, and has emphasised the importance of a diversity of floodplain habitat types for the maintenance of a diverse native fish community (Closs et al. 2006, Phillips 2006). Furthermore, this is the first study to demonstrate a strong link between fish spawning and recruitment and the provision of an environmental flood at least in Australia and perhaps the world. The study has important implications for managing flows in regulated rivers in the Murray-Darling Basin and wider. Along with the information provided to water management, the results of this project have also been disseminated at numerous scientific, management and community forums, peer-reviewed scientific journal articles and also in numerous regional, State and National media (see Appendices 1 and 2).

74


REFERENCES

Anderson, H. K. (1919). Rescue operation on the Murrumbidgee River. Australian Zoologist 1, 157-160.

Araujo-Lima, C. A. R. M. and E. C. Oliveira (1998). Transport of larval fish the Amazon. Journal of Fish Biology 53, 297-306.

Arthington, A. H., S. R. Balcombe, G. A. Wilson, M. C. Thoms and J. Marshall (2005). Spatial and temporal variation in fish-assemblage structure in isolated waterholes during the 2001 dry season of an arid-zone floodplain river, Cooper Creek, Australia. Marine and Freshwater Research 56, 25-35.

Arthington, A. H., S. E. Bunn, L. N. Poff and R. J. Naiman (2006). The challenge of providing environmental flow rules to sustain river ecosystems. Ecological Applications 16, 1311-1318.

Auer, N. A. and E. A. Baker (2002). Duration and drift of larval lake sturgeon in the Sturgeon River, Michigan. Journal of Applied Icthyology 18, 557-564.

Balcombe, S. R., A. H. Arthington, N. D. Foster, M. C. Thoms, G. G. Wilson and S. E. Bunn (2006). Fish assemblages of an Australian dryland river: abundance, assemblage structure and recruitment patterns in the Warrego River, Murray-Darling Basin. Marine and Freshwater Research 57, 619-633.

Bayley, P. B. (1991). The flood pulse advantage and the restoration of river-floodplain systems. Regulated Rivers: Research and Management 6, 75-86.

Bren, L. J., I. O’Neill and N. L. Gibbs (1987). Flooding in Barmah Forest and its relation to flow in the Murray-Edward River System. Australian Forest Research 16, 357-370.

Brown, A. V. and M. L. Armstrong (1985). Propensity to drift downstream among various species of fish. Journal of Freshwater Ecology 3, 3-17.

Brown, L. R. and T. Ford (2002). Effects of flow on the fish communities of a regualted California river: implications for managing native fishes. River Research and Applications 18, 331-342.

Brown, P., K. P. Sivakumaran, D. Stoessel and A. Giles (2005). Population biology of carp (Cyprinus carpio L.) in the mid-Murray River and Barmah Forest wetlands, Australia. Marine and Freshwater Research 56, 1151-1164.

Bunn, S. E. and A. H. Arthington (2002). Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30, 492-507.

Cadwallader, P. L. (1977). J.O. Langtry’s 1949-50 Murray River Investigations. Ministry for Conservation Fisheries and Wildlife Division, Melbourne, Australia. 70 pp.

Cadwallader, P. L. (1978). Some causes of the decline in range and abundance of native fish in the Murray-Darling river system. Proceedings of the Royal Society of Victoria 90, 211-224.

Chesterfield, E. A. (1986). Changes in the vegetation of the river red gum forests at Barmah, Victoria. Australian Forest Research 49, 16-27.

Chong, J. and A. R. Ladson (2003). Analysis and management of unseasonal flooding of the Barmah-Millewa Forest. River Research and Applications 19, 191-180.

Clarke, K. and R. Warwick (2001). ‘Change in marine communities: an approach to statistical analysis and interpretation. PRIMER-E Ltd.’ PRIMER-E, Plymouth.

Close, A. (1990). The impact of man on the natural flow regime. In ‘The Murray’. (Eds N. Mackay and D. Eastburn) pp 61-76. (Murray-Darling Basin Commission: Canberra, Australia)

Closs, G. P., S. R. Balcombe, P. Driver, D. G. McNeil and M. J. Shirley (2006). The importance of floodplain wetlands to Murray-Darling fish: What’s where? What do we know? What do we need to know? In ‘Native fish and wetlands in the Murray-Darling Basin: Action plan, knowledge gaps and supporting papers. Proceedings of a workshop held in Canberra ACT, 7-8 June 2005.’ (Ed. B. Phillips) pp. 14-28. (Murray-Darling Basin Commission: Canberra)

75

References

Copp, G. H. (1989). The habitat diversity and fish reproductive function of floodplain ecosystems. Environmental Biology of Fishes 26, 1-27.

Copp, G. H., H. Faulkner, S. Dowerty, M. S. Watkins and J. Majecki (2002). Diel drift behavious of fish eggs and larvae, in particular barbel Barbus Barbus (L.), in an English chalk stream. Fisheries Management and Ecology 9, 95-103.

Copp, G. H. and M. Penaz (1988). Ecology of fish spawning and nursery zones in the floodplain, using a new sampling approach. Hydrobiologia 169, 209-224.

Corbeil, R. R. and S. R. Searle (1976). Restricted likelihood (REML) estimation of variance components in the mixed model. Technometrics 18, 31-38.

Corbett, B. W. and P. M. Powles (1986). Spawning and larval drift of sympatric walleyes and white suckers in an Ontario stream. Transactions of the American Fisheries Society 115, 41-46.

Crivelli, A. J. (1981). The biology of the common carp, Cyprinus carpio L. in the Camargue, southern France. Journal of Fish Biology 18, 271-290.

Crook, D. A. and B. M. Gillanders (2006). Use of otolith chemical signatures to estimate carp recruitment sources in the mid-Murray River, Australia. River Research and Applications 22, 871-879.

Curr, E. M. (1965). Recollections of squatting in Victoria: then called the Port Phillip District from 1841 to 1851. Melbourne University Press, Melbourne.

Dynesius, M. and C. Nilsson (1994). Fragmentation and flow regulation of river systems in the northern third of the world. Science 266, 753-762.

Faragher, R. A. H. and Harris, J. H. (1994). The historical and current status of freshwater fish in New South Wales. Australian Zoologist 29, 166-176.

Faulkner, H. and G. H. Copp (2001). A model for accurate drift estimation in streams. Freshwater Biology 46, 723-733.

Finger, T. R. and E. M. Stewart (1987). Response of fishes to flooding regime in lowland hardwood wetlands. In ‘Community and Evolutionary Ecology of North American Stream Fishes’. (Ed. W. J. a. H. Matthews, D.C.) pp 86-92. (University of Oklahoma Press: Norman,USA)

Floyd, K. B., R. B. Hoyt and S. Timbrook (1984). Chronology of appearance and habitat partitioning by stream larval fishes. Transactions of the American Fisheries Society 113, 217-223.

Freeman, M. C., Z. H. Bowen, K. D. Bovee and E. R. Irwin (2001). Flow and habitat effects on juvenile fish abundance in natural and altered flow regimes. Ecological Applications 11, 179-190.

Gadomski, D. M. and C. A. Barfoot (1998). Diel and distributional abundance patterns of fish embryos and larvae in the lower Columbia and Deschutes rivers. Environmental Biology of Fishes 51, 353-368.

Geddes, M. C. and J. T. Puckridge (1988). Survival and growth of larval and juvenile native fish. The importance of the floodplain. In ‘Proceedings of the Native Fish Management Workshop’. (Ed. B. Lawrence) pp. 101-115. (Murray Darling Basin Commission: Canberra, Australia)

Gehrke, P. C. (1990a). Clinotactic responses of larval silver perch (Bidyanus bidyanus) and golden perch (Macquaria ambigua) to simulated environmental gradients. Australian Journal of Marine and Freshwater Research 41, 523-528.

Gehrke, P. C. (1990b). Spatial and temporal dispersion patterns of golden perch, Macquaria ambigua, larvae in an artificial floodplain environment. Journal of Fish Biology 37, 225-236.

Gehrke, P. C. (1991). Avoidance of inundated floodplain habitat by larvae of golden perch (Macquaria ambigua Richardson): Influence of water quality or food distribution? Australian Journal of Marine and Freshwater Research 42, 707-719.

Gehrke, P. C., P. Brown, C. B. Schiller, D. B. Moffatt and A. M. Bruce (1995). River regulation and fish communities in the Murray-Darling River system, Australia. Regulated Rivers: Research and Management 11, 363-375.

76


Gilligan, D. and C. B. Schiller (2003). ‘Downstream transport of larval and juvenile fish in the Murray River. NSW Fisheries Report Series No. 50.’ NSW Fisheries, Sydney, Australia.

Graham, R. and J. H. Harris (2004). ‘Floodplain inundation and fish dynamics in the Murray-Darling Basin. Current concepts and future research: a scoping study.’ Cooperative Centre for Freshwater Ecology. Canberra. 52 pp.

Grift, R. E., A. D. Buijse, W. L. T. Van Denson, M. A. M. Machiels, J. Kranenbarg, J. G. P. Klein Breteler and J. J. G. M. Backz (2003). Suitable habitats for o-group fish in rehabilitated floodplains along the Lower River Rhine. River Research and Applications 19, 353-374.

Harris, J. H. and P. C. Gehrke (1994). Modelling the relationship between streamflow and population recruitment to manage freshwater fisheries. Agricultural Systems and Information Technology 6, 28-30.

Helfman, G. S. (1993). Fish behaviour by day, night and twilight. In ‘Behaviour of Teleost Fishes’. (Ed. T. J. Pitcher) pp. 479-512. (Chapman and Hall: London)

Holland, L. E. (1986). Distribution of early life history stages of fishes in selected pools of the Upper Mississippi River. Hydrobiologia 136, 121-130.

Humphries, P. (2005). Spawning time and early life history of Murray cod, Maccullochella peelii peelii (Mitchell) in an Australian river. Environmental Biology of Fishes 72, 393-407.

Humphries, P. and A. J. King (2004). Drifting fish larvae in Murray-Darling Basin rivers: composition, spatial and temporal patterns and distance drifted. In ‘Downstream movement of fish in the Murray-Darling Basin. Statement, recommendations and supporting papers from a workshop held in Canberra 3-4 June 2003’. (Eds M. Lintermans and B. Phillips) pp. 51-58. (Murray-Darling Basin Commission: Canberra, Australia.)

Humphries, P., A. J. King and J. D. Koehn (1999). Fish, flows and flood plains: links between freshwater fishes and their environment in the Murray-Darling River system, Australia. Environmental Biology of Fishes 56, 129-151.

Humphries, P. and P.S. Lake. (2000). Fish larvae and the management of regulated rivers. Regulated Rivers: Research and Management 16,421–432.

Humphries, P., L. Serafini and A. J. King (2002). River regulation and fish larvae: variations through space and time. Freshwater Biology 47, 1307-1331.

Jones, M. and I. G. Stuart (2004). ‘Impact of flow regulation structures on fish in the Barmah-Millewa forest.’ Arthur Rylah Institute for Environmental Research, Heidelberg. 69 pp.

Junk, W. J., P. B. Bayley and R. E. Sparks (1989). The flood pulse concept in river-floodplain systems. Canadian Special Publication of Fisheries and Aquatic Sciences 106, 110-127.

Karolak, S. (2006). Alien fish in the Murray-Darling Basin. Murray-Darling Basin Commission, Canberra. 83 pp.

Killgore, K. J. and J. A. Baker (1996). Patterns of larval fish abundance in a bottomland hardwood wetland. Wetlands 16, 288-295.

King, A. J. (2004). Ontogenetic patterns of habitat use by fishes within the main channel of an Australian floodplain river. Journal of Fish Biology 65, 1582-1603.

King, A. J. (2005). Fish and the Barmah-Millewa Forest: history, status and management challenges. Proceedings of the Royal Society of Victoria 117, 117-125.

King, A. J. and D. A. Crook (2002). Evaluation of a sweep net electrofishing method for the collection of small fish and shrimp in lotic freshwater environments. Hydrobiologia 472, 223-233.

King, A. J., D. A. Crook, W. M. Koster, J. Mahoney and Z. Tonkin (2005). Comparison of larval fish drift in the Lower Goulburn and mid-Murray Rivers. Ecological Management and Restoration 6, 136-138.

King, A. J., P. Humphries and P. S. Lake (2003). Fish recruitment on floodplains: the roles of patterns of flooding and life history characteristics. Canadian Journal of Fisheries and Aquatic Sciences 60, 773-786.

77

References

King, A. J., A. I. Robertson and M. R. Healey (1997). Experimental manipulations of the biomass of introduced carp (Cyprinus carpio) in billabongs. I. Impacts on water-column properties. Marine and Freshwater Research 48, 435-443.

King, J., J. A. Cambray and N. D. Impson (1998). Linked effects of dam-released floods and water temperature on spawning of the Clanwilliam yellowfish Barbus capensis. Hydrobiologia 384, 245-265.

Koehn, J. D., Brumley, A., and Gehrke, P.C. (2000). Managing the Impacts of Carp. Bureau of Rural Sciences, Department of Agriculture, Fisheries and Forestry, Canberra, Australia. 249 pp.

Koehn, J. D. and D. J. Harrington (2005). Collection and distribution of the early life stages of Murray cod (Maccullochella peelii peelii) in a regulated river. Australian Journal of Zoology 53, 137-144.

Koehn, J. D. and D. J. Harrington (2006). Environmental conditions and timing for the spawning of Murray cod (Maccullochella peelii peelii) and the endangered trout cod (M. macquariensis) on southeastern Australian rivers. River Research and Applications 22, 327-342.

Lake, J. S. (1967a). Rearing experiments with five species of Australian freshwater fishes. I. Inducement to spawning. Australian Journal of Marine and Freshwater Research 18, 137-153.

Lake, J. S. (1967b). Rearing experiments with five species of Australian freshwater fishes. II. Morphogenesis and ontogeny. Australian Journal of Marine and Freshwater Research 18, 155-173.

Leslie, D. J. 1995. Moira Lake: a case study of the deterioration of a River Murray natural resource. MSc thesis. University of Melbourne. 188 pp.

Leslie, D. J. (2001). Effect of river management on colonially-nesting waterbirds in the Barmah-Millewa forest, south-eastern Australia. Regulated Rivers: Research and Management 17, 21-36.

Leslie, D. J. and K. A. Ward (2002). River Murray environmental flows 2000-2001. Ecological Management and Restoration 3, 221-223.

Lloyd, L., J. Puckridge and K. Walker (1989). The significance of fish populations in the Murray-Darling system and their requirements for survival. In ‘Conservation in Management of the River Murray System - Making Conservation Count. Proceedings of the Third Fenner Conference on the Environment, Canberra. September 1989’. (Ed. T. A. C. Dendy, M) pp 86-99. (Australian Academy of Science, South Australian Department of Environment and Planning: Adelaide, Australia)

Lytle, D. A. and L. N. Poff (2004). Adaptation to natural flow regimes. Trends in Ecology and Evolution 19, 94-100.

Macdonald, J. I. and D. A. Crook (2006). ‘Contribution of the Barmah-Millewa floodplain lakes to carp populations in the mid-Murray River, 200506.’ Arthur Rylah Institute for Environmental Research, Heidelberg. 32 pp.

Mackay, N. J. (1973). Histological changes in the ovaries of the golden perch, Plectroplites ambiguus, associated with the reproductive cycle. Australian Journal of Marine and Freshwater Research 24, 95-101.

Mallen-Cooper, M. and I. G. Stuart (2003). Age, growth and non-flood recruitment of two potamodromous fishes in a large semi-arid/temperate river system. River Research and Applications 19, 697-719.

Marchetti, M. P. and P. B. Moyle (2001). Effects of flow refime on fish assemblages in a regulated California stream. Ecological Applications 11, 530-539.

Maunsell McIntyre Pty Ltd (1999). ‘Report on Barmah-Millewa flood of October 1998 and the first use of Barmah-Millewa allocation. A report prepared for the Barmah-Millewa Forum and Murray-Darling Basin Commission.’ Maunsell McIntyre Pty Ltd, Canberra, Australia. 121 pp.

Maunsell McIntyre Pty Ltd (2001). ‘Report on Barmah-Millewa Forest flood of spring 2000 and the second use of Barmah-Millewa Forest environmental water allocation, spring-summer 2000-01. A report prepared for the Barmah-Millewa Forum and Murray-Darling Basin Commission.’ Maunsell McIntyre Pty Ltd,, Canberra, Australia. 127 pp.

McKinnon, L. J. (1997). Monitoring of Fish Aspects of the Flooding of Barmah Forest. Marine and Freshwater Resources Institute, Queenscliff, Australia. 184 pp.

78


McNeil, D. G. (2004). Ecophysiology amd behavious of Ovens River floodplain fish: hypoxia tolerance and the role of the physicochemical environment in structuring Australian billabong fish communities. PhD thesis. LaTrobe, Bundoora.

MDBC. (2002a). The Living Murray Initiative Proposal for investment - an implementation program of structural works and measures (2003/04 to 2005/06). Murray-Darling Basin Commission, Canberra, Australia. 38 pp.

MDBC. (2002b). The Living Murray. A discussion paper on restoring the health of the River Murray. MDBC, Canberra, Australia. 51 pp.

MDBC. (2004). Native Fish Strategy for the Murray-Darling Basin 2003-2013. Murray-Darling Basin Commission, Canberra, Australia. 50 pp.

Meredith, S., B. Gawne, C. Sharpe, N. Whiterod, A. Conallin and S. Zukowski (2002). ‘Dryland floodplain ecosystems: influence of flow pattern on fish production.’ Technical Report 1/2002, Murray-Darling Freshwater Research Centre, Mildura, Australia. 38 pp.

Nesler, T. P., R. T. Muth and A. F. Wasowicz (1988). Evidence for baseline flow spikes as spawning cues for Colorado squawfish in the Yamba River, Colorado. American Fisheries Society Symposium 5, 68-79.

Nezdolily, V. K. (1984). Downstream migration of young fishes during the initial period of flow regulation of the Ili River. Journal of Ichthyology 24, 34-46.

Oesmann, S. (2003). Vertical, lateral and diurnal drift patterns of fish larvae in a large lowland river, the Elbe. Journal of Applied Ichthyology 19, 284-293.

Penaz, M., A. L. Roux, P. Juradja and J. M. Oliver (1992). Drift of larval and juvenile fishes in a by-passed floodplain of the upper River Rhone, France. Folia Zoologica 41, 281-288.

Phillips, B. (Ed.) (2006). ‘Native fish and wetlands in the Murray-Darling Basin: Action plan, knowledge gaps and supporting papers. Proceedings of a workshop held in Canberra ACT, 7-8 June 2005.’ (Murray-Darling Basin Commission: Canberra).

Poff, L. N., J. D. Allan, M. A. Palmer, D. A. Hart, B. D. Richter, A. H. Arthington, K. H. Rogers, J. L. Meyer and J. A. Stanford (2003). River flows and water wars: emerging science for environmental decision making. Frontiers in Ecology and Environment 1, 298-306.

Poff, N. L., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D. Richter, R. E. Sparks and J. C. Stromberg (1997). The natural flow regime: a paradigm for river conservation and restoration. BioScience 47, 769-784.

Power, M. E., A. Sun, G. Parker, W. E. Dietrich and J. T. Wootton (1995). Hydraulic food-chain models. BioScience 45, 159-167.

Puckridge, J. T., F. Sheldon, K. F. Walker and A. J. Boulton (1998). Flow variability and the ecology of large rivers. Marine and Freshwater Research 49, 55-72.

Reichard, M., P. Juradja and M. Ondrackova (2002). The effect of light intensity on the drift of young-of-the-year cyprinid fishes. Journal of Fish Biology 61, 1063-1066.

Reid, D. D., J. H. Harris and D. J. Chapman (1997). ‘NSW Inland Commercial Fishery Data Analysis.’ NSW Fisheries, Sydney, Australia.

Reynolds, L. F. (1983). Migration patterns of five fish species in the Murray-Darling river system. Australian Journal of Marine and Freshwater Research 34, 857-871.

Robertson, A. I., M. R. Healey and A. J. King (1997). Experimental manipulations of the biomass of introduced carp (Cyprinus carpio) in billabongs. II. Impacts of benthic properties and processes. Marine and Freshwater Research 48, 445-454.

Rood, S. B., C. R. Gourley, E. M. Ammon, L. G. Heki, J. R. Klotz, M. L. Morrison, D. Mosley, G. G. Scoppettone, S. Swanson and P. L. Wagner (2003). Flows for floodplain forests: a successful riparian restoration. BioScience 53, 647-656.

Rowland, S. J. (1989). Aspects of the history and fishery of the Murray cod, Maccullochella peeli (Mitchell) (Percichthyidae). Proceedings of the Linnean Society of New South Wales 111, 201-203.

79

References

Rowland, S. J. (1998). Aspects of the reproductive biology of Murray cod, Maccullochella peelii peelii. Proceedings of the Linnean Society of New South Wales 120, 147-162.

Schiller, C. B. and J. H. Harris (2001). Chapter 6: Native and alien fish. In ‘Rivers as Ecological Systems: The Murray-Darling Basin’. (Ed. W. J. Young) pp. 229-258. (Murray Darling Basin Commission: Canberra, Australia)

Secor, D. H., J. M. Dean and J. Hansbarger (1992). Modification of the quadrefoil light trap for use in hatchery ponds. The Progressive Fish Culturist 54, 202-205.

Serafini, L. and P. Humphries (2004). ‘Preliminary guide to the identification of larvae of fish, with a bibliography of their studies, from the Murray-Darling Basin. Identification and Ecology Guide No. 48.’ Cooperative Research Centre for Freshwater Ecology, Albury, Australia.

Shirley, M. J. (2002). The ecology of billabong fish communities of the River Murray, with a focus on the interactions of European perch (Perca fluviatilis). PhD thesis. Monash University, Melbourne.

Sparks, R. E. (1995). Need for ecosystem management of large rivers and their floodplains. BioScience 45, 168-182.

Stanford, J. A., J. V. Ward, W. J. Liss, C. A. Frissell, R. N. Williams, J. A. Lichatowich and C. C. Coutant (1996). A general protocol for restoration of regulated rivers. Regulated Rivers: Research and Management 12, 391-413.

Stoffels, R. J. and P. Humphries (2003). Ontogenetic variation in the diurnal food and habitat associations of an endemic and an exotic fish in floodplain ponds: consequences for niche partitioning. Environmental Biology of Fishes 66, 293-305.

Stuart, I. G. and M. Jones (2002). ‘Ecology and management of common carp in the Barmah-Millewa forest.’ Arthur Rylah Institute for Environmental Research, Heidelberg. 214 pp.

Stuart, I. G. and M. Jones (2006a). Large, regulated forest floodplain is an ideal recruitment zone for non-native common carp (Cyprinus carpio L.). Marine and Freshwater Research 57, 333-347.

Stuart, I. G. and M. Jones (2006b). Movement of common carp, Cyprinus carpio, in a regulated lowland Australian river: implications for management. Fisheries Management and Ecology 2006, 231-219.

Sturt, C. (1838). Hand-written journal for 1838. State Library of Victoria.

Swee, U. B. and H. R. McCrimmon (1966). Reproductive biology of the carp, Cyprinus carpio L., in Lake St. Lawrence, Ontario. Transactions of the American Fisheries Society 95, 372-380.

Thoms, M. C., P. Suter, J. Roberts, J. D. Koehn, G. Jones, T. J. Hillman and A. Close. (2000). Report of the River Murray scientific panel on environmental flows. River Murray - Dartmouth to Wellington and the Lower Darling River. Murray-Darling Basin Commission, Canberra, Australia. 168 pp.

Tonkin, Z., A. J. King, J. Mahoney and J. Morrongiello (2007). Diel and spatial drifting patterns of silver perch Bidyanus bidyanus eggs in an Australian lowland river. Journal of Fish Biology 70, 313-317.

Travinchek, V. H., M. B. Bain and M. J. Maceina (1995). Recovery of a warmwater fish assemblage after initiation of a minimum-flow release downstream from a hydroelectric dam. Transactions of the American Fisheries Society 124, 836-844.

Trippel, E. A. and R. C. Chambers (1997). The early life history of fishes and its role in recruitment processes. In ‘Early Life History and Recruitment in Fish Populations’. (Ed. R. C. a. T. Chambers, E.A) pp. 21-32. (Chapman and Hall: London, UK)

Turner, T. F., J. C. Trexler, G. L. Miller and K. E. Toyer (1994). Temporal and spatial dynamics of larval and juvenile fish abundance in a temperate floodplain river. Copeia 1994, 174-183.

Valdez, R. A., T. L. Hoffinagle, C. C. McIvor, T. McKinney and W. C. Leibfried (2001). Effects of a test flood on fishes of the Colorado River in Grand Canyon, Arizona. Ecological Applications 11, 686-700.

Venables, W. N. and C. M. Dichmont (2004). GLMs, GAMs and GLMMs: an overview of theory for applications in fisheries research. Fisheries Research 70, 319-337.

80


Walker, K. F. and M. C. Thoms (1993). Environmental effects of flow regulation on the lower River Murray, Australia. Regulated Rivers: Research and Management 8, 103-119.

Ward, J. V., K. Tockner and F. Schiemer (1999). Biodiversity of floodplain ecosystems: ecotones and connectivity. Regulated Rivers: Research and Management 15, 125-139.

Ward, K. A. (2005). Water management in the changing Barmah-Millewa wetlands. Proceedings of the Royal Society of Victoria 117, 77-84.

Waters, J. M., M. J. Shirley and G. P. Closs (2002). Hydroelectric development and translocation of Galaxias brevipinnis: a cloud at the end of the tunnel? Canadian Journal of Fisheries and Aquatic Sciences 59, 49-56.

Welcomme, R. L. (1985). River Fisheries. Food and Agriculture Organisation of the United Nations, FAO Fisheries Technical Paper 262, Rome, Italy. 303 pp.

Ye, Q., K. Jones and B. E. Pierce (2000). ‘Murray cod (Maccullochella peelii peelii), fishery assessment report to PIRSA for Inland Waters Fishery Management Committee, South Australian Fisheries Assessment Series 2000/17.’ SARDI, Adelaide, Australia.

Zitek, A., S. Schmutz and A. Ploner (2004). Fish drift in a Danube sidearm-system: II. Seasonal and diurnal patterns. Journal of Fish Biology 65, 1339-1357.

81

APPENDIX 1: SCIENTIFIC PUBLICATIONS AND PRESENTATIONS

Peer-reviewed Journal Articles

King, A.J., Tonkin, Z. and Mahoney, J. (subm). Environmental flows enhance native fish spawning and recruitment in the Murray River, Australia.

Tonkin, Z ., King, A.J., and Mahoney, J. (in press). Testing a modification to a standard passive drift net to capture drifting ichthyofauna. Fisheries Management and Ecology.

Tonkin, Z., A. King, J. Mahoney & J. Morrongiello (2007). Diel and spatial drifting patterns of silver perch Bidyanus bidyanus eggs in an Australian lowland river. Journal of Fish Biology, 70, 313-317.

King, A.J. (2005). Fish in the Barmah-Millewa Forest – history, status and management challenges. Proceedings of the Royal Society of Victoria, 11, (1) 117-126.

King, A.J., Crook, D.A., Koster, W.M, Mahoney, J. and Tonkin, Z. (2005). Comparison of larval fish drift in the Lower Goulburn and mid-Murray Rivers. Ecological Management and Restoration, 6, (2), 136-138.

Other Publications

McCarthy, B., Nielsen, D., Baldwin, D., Meredith, S., Roberts, J., King, A., and Reid, J. (2006). Barmah Wetland System Environmental Monitoring Program, Part B: Monitoring Program. Report to the Goulburn Broken Catchment Management Authority. Murray-Darling Freshwater Research Centre.

King, A.J., Tonkin, Z. and Mahoney, J. (2006). Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest – 2005/06 Progress report. Report to Murray-Darling Basin Commission. Freshwater Ecology, Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment.

King, A.J., Tonkin, Z. and Mahoney, J. (2005). Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest – 2004/05 Progress report. Report to Murray-Darling Basin Commission. Freshwater Ecology, Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment.

King, A.J., Mahoney, J. and Tonkin, Z. (2004). Assessing the effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest – 2003/04 Progress report. Report to Murray-Darling Basin Commission. Freshwater Ecology, Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment.

Conference/workshop presentations

King A.J. Tonkin, Z. and Mahoney, J. (2006). Environmental flows and fish recruitment in Barmah-Millewa Forest. Spoken presentation, invited seminar to Goulburn Valley Angling and Associated Clubs. 26th October 2006.

King, A.J. (2006). The role of floods and floodplains for recruitment of native fish. Spoken presentation, invited plenary. Australian Society for Limnology, 26th – 29th September 2006.

King, A.J. Tonkin, Z. and Mahoney, J. (2006). Native fish flourish in Barmah-Millewa Forest environmental flows. Spoken presentation. Australian Society for Fish Biology 2006 Conference, Hobart, 31 August – 1st September 2006.

King, A.J. Tonkin, Z. and Mahoney, J. (2006) Assessing effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest. Spoken presentation. Invited seminar MDBC, 24th August 2006.

King, A.J. Tonkin, Z. and Mahoney, J. (2006) Assessing effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest. Spoken presentation. MDBC Native Fish Strategy Forum. 14-15th June 2006.

82


King, A.J. (2006) Want fish – just add water … … Spoken presentation to BM Coordinating Committee. 2nd June 2006

King, A.J. (2006) Fish in the Barmah-Millewa Forest – history, status and management challenges. Spoken presentation to NFS community liaison group and Barmah indigenous representatives. 10th May 2006.

Crook, D.A., and King, A.J. (2005). Fish assemblages in the Lower Goulburn River – status and future challenges. Spoken presentation. Goulburn-Broken CMA, 9th November 2005.

Tonkin, Z.D. (2005). Assessing effectiveness of environmental flows on fish recruitment in Barmah-Millewa Forest. Spoken presentation to Jim Barrett, MDBC. 24th October 2005.

Crook, D.A., and King, A.J. (2005). Fish assemblages in the Lower Goulburn River – status and future challenges. Spoken presentation. DSE’s River Health Branch, September 23rd 2005.

King, A.J., Tonkin, Z. and Mahoney, J. (2005). Can’t see the Forest for the fish! – Fish recruitment in the Barmah-Millewa Forest, Murray River. Spoken presentation. Australian Society for Fish Biology 2005 Conference, Darwin NT, July 14-15, 2005.

King, A.J. 2005. Fish in the Barmah-Millewa Forest – history, status and management challenges. Spoken presentation, Royal Society of Victoria conference on Barmah-Millewa Forest, Melbourne, 18-19 June 2005.

King, A.J. and Humphries, P. (2004) The highs and lows of fish recruitment in floodplain rivers of the Murray-Darling Basin. Spoken presentation. Australian Society for Fish Biology 2004 Conference and Fisheries Ecosystem Symposium, Adelaide SA, September 19-24 2004.

83

APPENDIX 2: MEDIA COMMUNICATIONS

February 2006

• Interviewed for article on Murray cod breeding for Herald Sun Newspaper. “The cod is back”, Sunday Herald Sun, 5th February 2006

• Publication of article in Victorian Fishing Monthly. “Natives flourish in Barmah-Millewa floods”, by Alison King, Victorian Fishing Monthly, February 2006.

January 2006

• DSE Inform article January 2006 “Native fish flourish in Barmah-Millewa Forest floods”

• “Invigorating the Barmah-Millewa Forest”, Victorian Fishing Monthly, January 2006.

December 2005

• Interviewed for ABC TV documentary “Two men in a Tinnie” with Tim Flannery and John Doyle. Screened in October 2006.

• Interviewed for newspaper article, Sunraysia daily, 21/12/06, “Flooding livens Barmah”

• Interviewed for information for Environment Victoria’s website news article “Booming fish numbers prove need for more flows”

• “Flushing flows for Forest and fish” Victorian fishing monthly, December 2005.

November 2005

• Interviewed for various newspaper articles on fish response to Barmah-Millewa Environmental Flow.

• “Flood spawns breeding of Murray’s native perch” The Australian, 23/11/05

• Breeding frenzy in Barmah, The Age article, 26/11/05

• “Native fish species spawning”, Sunraysia Daily, 9/11/05

• “Fish spawn in Barmah wetlands”, Numurkah Leader,16/11/05

• Filmed and interviewed on ABC TV Stateline, screened 2/12/05.

84


“Lower Goulburn and Mid-Murray Native fish”

Victorian Fishing Monthly, March 2005.

“Native fish species spawning”

Sunraysia Daily, 9th November 2005.

85

APPENDIX 2: Media communications

“Fish spawn in Barmah wetlands”

Numurkah Leader, 16th November 2005.

“Flood spawns breeding of Murray’s native perch”

The Australian, 23rd November 2005

86


“A little water fuels a frenzy”

The Age, 26th November 2005

A little water fuels a frenzy

By Melissa Fyfe, Environment Reporter November 26, 2005

DEEP in the Barmah forest there’s a breeding frenzy going on, the likes of which have not been seen for five years.

Birds are at it, frogs are doing it, fish are going for it. And it’s all because of a little water. Well, perhaps more than a little.

Like an irrigator, the Barmah — which straddles the Murray River between Echuca and Tocumwal — has its own water allocation. The two state governments have been saving up the 100 billion litres each year for five years in dams, waiting for the right time to release the biggest environmental flow in Australia’s history.

The timing became crucial this year because it was a long time between drinks for the forest, one of the Murray’s much-loved sites and an internationally renowned wetland. If water managers had left it much longer, waterbirds, which rely on floods to breed, would have suffered a massive population crash.

“We had our hearts in our mouths when planning this flood,” said Kevin Ritchie, the Department of Sustainability and Environment’s north-east regional director. “It was five years since we had a flood in the forest and we really needed to get bird breeding going.”

Many of Barmah’s old red gums are now soaking their thirsty roots in a metre of water. The two state governments plan to release 500 billion litres into the forest, sustaining a five-month-long flood.

The flooding began in October and the results already are spectacular — a relief to governments and freshwater scientists who are keen to demonstrate the benefits of giving the environment more water.

In Kakadu-style, the Barmah wetland is alive and teeming with lush growth, birds, frogs, huge dragonflies, black snakes, turtles and fish.

The great egret is breeding on the Victorian side of the Barmah for the first time in 30 years. Pelicans swirl around in groups. In the red gums, ibis are building nesting platforms over the water. The massive roots of old red gums are surrounded by flood water, while up in the canopy, koalas are feasting on the new green growth.

“It really is looking like an amazing place at the moment,” says Keith Ward, the environmental water reserves manager with the Goulburn Broken Catchment Management Authority. And excited government scientists are out documenting it all: counting birds, checking eggs, monitoring fish.

Scientists are carefully monitoring the fish breeding. Already the results show an excellent response from golden perch, a favourite of recreational anglers, and silver perch, a threatened species that needs conservation.This will make for excellent fishing in five years.

Water managers such as Mr Ward are getting better at timing the floods to get the best outcome. Floods need to be the right length, depth and frequency. If the waters subside too quickly, waterbirds abandon their nests. All three factors have been hit and miss as the Murray’s dams and water extraction reduce natural flows.

This water is not lost — after flooding the forest, about 80 per cent returns to the Murray’s main channel, cleaner and richer in nutrients.

The Yorta Yorta people have signed a joint management agreement with the Victorian Government and are giving their advice about the water regime. “Ideally we would like to see flooding more regularly,” said Neville Atkinson, convener of the joint body. “But we have to keep in mind we are just coming out of a drought period.”

87


“Flooding livens Barmah”

Sunraysia Daily, 21st December 2006.

88


“Booming fish numbers prove need for more flows”

Environment Victoria’s website, 15th December 2005.

89


90


“Flushing flows for Forest and fish”

Victorian Fishing Monthly, December 2005.

“Invigorating the Barmah-Millewa Forest”

Victorian Fishing Monthly, January 2006.

91


“The cod is back”

Sunday Herald Sun, 5th February 2006.

92


“Natives flourish in Barmah-Millewa floods”

Victorian Fishing Monthly, February 2006.

93

APPENDIX 3: SPECIES LIST AND DEVELOPMENTAL STAGE COLLECTED AT ALL SITES IN BARMAH-MILLEWA FOREST DURING NORMAL SAMPLING

* indicates species/stage collected, ^ indicates introduced species.

Barmah Lake

2003–04 2004–05 2005–06

Larvae Juveniles Adults Larvae Juveniles Adults Larvae Juveniles Adults

Australian smelt * * * * * * * * *

carp gudgeon * * * * * *

unspecked hardyhead * * * *

carp^ * * * * * *

goldfish^ * * * *

redfin perch^ * * * *

gambusia^ * * * * * *

oriental weatherloach^ * * *

Budgee Creek

2003–04 2004–05 2005–06



carp gudgeon * * * * * * * * *

Murray cod *

unspecked hardyhead * * *

southern pigmy perch *

Flat-headed gudgeon *

carp^ * * * * * *

goldfish^ * * * *

redfin perch^ * * * * * *

gambusia^ * * * * * * * *

oriental weatherloach^ * * * * *

Black Engine Billabong

2003–04 2004–05 2005–06


Australian smelt * * * * * * *

carp gudgeon * * * * * * * * *


carp^ * *

goldfish^ *

gambusia^ * * * * * * * *

Bunyip Billabong

2003–04 2004–05 2005–06


Australian smelt * * * * * *

carp gudgeon * * * * * * * * *

southern pygmy perch *

carp^ * * * * *

goldfish^ * * * *

gambusia^ * * * * * * * *

94


Choke – Murray River

2003–04 2004–05 2005–06



carp gudgeon * * * * * * * *

Murray cod * * *

silver perch * *

golden perch * *

unspecked hardyhead * * * * * * * * *


* * *

flat-headed gudgeon * * * *

carp^ * * * *

goldfish^ * * * *

Redfin perch^ *

gambusia^ * * * *

oriental weatherloach^ * *

Flat Swamp

2003/04 2004/05 2005/06


Australian smelt * * * * * * * *

carp gudgeon * * * * * * * * *

unspecked hardyhead * * * * * * *

southern pygmy perch * *

flat-headed gudgeon * *

carp^ * * * *

goldfish^ * *

redfin perch^ * * *

gambusia^ * * * * * * * *

oriental weatherloach^ * *

Gulpa Creek

2003/04 2004/05 2005/06



carp gudgeon * * * * * * *

Murray cod * * *



* * * * *

flat-headed gudgeon *

carp^ * * * *

goldfish^ * *

gambusia^ * * * * * * * *


95

APPENDIX 3

Hut Lake

2003/04 2004/05 2005/06



carp gudgeon * * * * * * * * *

unspecked hardyhead * * * * *


carp^ * * * * *

goldfish^ * * * * *

redfin perch^ *

gambusia^ * * * * * * * *


Ladgroves Beach

2003–04 2004–05 2005–06



carp gudgeon * * * * * * * * *

Murray cod * * * *

silver perch * * *

golden perch * *

unspecked hardyhead * * * * * * * *

flat-headed gudgeon * * * * * * * * *

carp^ * * * * *

goldfish^ * * *

redfin perch^ *

gambusia^ * *

Morning Glory – Murray River

2003–04 2004–05 2005–06



carp gudgeon * * * * * * * * *

Murray cod * *

silver perch * *

golden perch *

unspecked hardyhead * * * * * * *

flathead gudgeon * * * * * *

carp^ * * * * * *

goldfish^ * * * *

redfin perch^ * *

Gambusia^ * * * *

oriental weatherloach^ * * * *

96


Moira Lake2003–04 2004–05 2005–06


Australian smelt * * * * * * * * *carp gudgeon * * * * * * * * *flat-headed gudgeon * * *

carp^ * * * * * *

goldfish^ * * * * * *

gambusia^ * * * * * * * * *oriental weatherloach^ * * *

Tarmah Swamp2003/04 2004/05 2005/06


Australian smelt * * * * * * * * *carp gudgeon * * * * * * * * *southern pygmy perch * * * * *flat-headed gudgeon *

carp^ * * * * *

goldfish^ * * * * *redfin perch^ * * * *

gambusia^ * * * * * * * * *oriental weatherloach^ * * * * *

Tongalong Creek2003/04 2004/05 2005/06


Australian smelt * * * * * * * * *carp gudgeon * * * * * * * * *Murray cod * * * *

trout cod *

unspecked hardyhead * * * * * * * *

southern pygmy perch * *flat-headed gudgeon * * * * * * * *carp^ * * * * * *

goldfish^ * * * *

redfin perch^ * *

gambusia^ * * * * * * * * *oriental weatherloach^ * *

Tullah Creek2003/04 2004/05 2005/06


Australian smelt * * * * * * * * *carp gudgeon * * * * * * * * *unspecked hardyhead *


flat-headed gudgeon * * *

carp^ * * * * * *

goldfish^ * * * * * * * *redfin perch^ * * * * * *

gambusia^ * * * * * * * * *oriental weatherloach^ * * * *

97

APPENDIX 4: SPECIES LIST OF FISH COLLECTED IN THE MURRAY RIVER BOAT ELECTROFISHING SAMPLING TARGETED AT JUVENILES.

Note: carp and goldfish were not recorded in 2005 sampling event.

Year

Common name Stage 2005 2006 Total

Murray cod Adult 28 46 74

YOY 5 25 30

trout cod Adult 30 36 66

YOY 4 67 71

golden perch Adult 18 33 51

YOY 3 2 5

silver perch Adult 13 27 40

YOY 0 5 5

Murray-Darling rainbowfish Adult 1 0 1

bony herring Adult 0 1 1

carp Adult 0 188 188

YOY 0 158 158

goldfish Adult 0 22 22

YOY 0 224 224

redfin perch YOY 1 4 5

oriental weatherloach Adult 0 4 4

98


APPENDIX 5: RESULTS OF REML POST-HOC ANALYSES FROM CHAPTER 3

Australian smelt SNE

Total larvae 2003–04 2004–05 2005–06

2003–04

2004–05 ***

2005–06 * *

CREEK LAKE RIVER WETLAND

CREEK

LAKE *

RIVER ** ***

WETLAND * NS **

Comments: 2003/4>2004–05 season, other comparisons very similar.

Greater in River than other habitats, next most in creeks

lower numbers in lakes and wetlands

Australian smelt Light trap

Juveniles 200304 200405 200506

200304

200405 ***

200506 ** ***

Comments: 2003–04>2005–06>2004–05

carp gudgeons Hand trawl

Total larvae CREEK LAKE WETLAND

CREEK

LAKE ***

WETLAND *** ***

Comments: wetland>lake>creek

carp gudgeons SNE

Juveniles CREEK LAKE RIVER WETLAND

CREEK

LAKE ns

RIVER ns ns

WETLAND *** *** ***

99

APPENDIX 5

carp gudgeons Light trap

Juveniles 200304 200405 200506

200304

200405 *

200506 ns *


CREEK

LAKE **

RIVER ns ns

WETLAND *** ns *

Comments: 2003–04>2004–05>2005–06

Wetland>other habitats

carp gudgeons Light trap

Adults CREEK LAKE RIVER WETLAND

CREEK

LAKE ***

RIVER ns **

WETLAND *** *** ***

Comments: Wetland>other habitats

flat-headed gudgeon Drift

Larvae 200304 200405 200506

200304

200405 ns

200506 * **

Comments: 2003–04=2004–05>2005–06

flat-headed gudgeon SNE


CREEK

LAKE ns

RIVER ** ***

WETLAND ns ns ***

Comments: River> other habitats

unspecked hardyhead SNE

Total larvae CREEK LAKE RIVER WETLAND

CREEK

LAKE ns

RIVER *** ***

WETLAND ns ns ***

100


unspecked hardyhead Light trap


CREEK

LAKE ns

RIVER *** **

WETLAND ns ns ***




CREEK

LAKE ns

RIVER ** **

WETLAND ns ns *




CREEK

LAKE ***

RIVER *** ***

WETLAND ns ns ***


southern pygmy perch Light trap

Juveniles 200304 200405 200506

200304

200405 ns

200506 * ns


CREEK

LAKE ns

RIVER ns ns

WETLAND ns * *

Comments: 2005–06>2003–04

Creek = wetland >lake and river

carp SNE

Juveniles 200304 200405 200506

200304

200405 ***

200506 *** ns

101

APPENDIX 5

carp Light trap


CREEK

LAKE ns

RIVER * **

WETLAND ns ns **

Comments: 2005–06>2003–04

Lake=wetland>river, creek

goldfish SNE


CREEK

LAKE ns

RIVER ns ns

WETLAND * ** Ns

Comments: wetland>creek, lake

goldfish SNE

Juveniles 200304 200405 200506

200304

200405 *

200506 ** ns


CREEK

LAKE ns

RIVER * **

WETLAND ns ns *

Comments: 2003–04 sig lower than other years

river sig lower than other habitats

gambusia SNE


CREEK

LAKE ns

RIVER ns ns

WETLAND ** * **


gambusia Light traps


CREEK

LAKE ns

RIVER * **

WETLAND *** *** ***


102


gambusia Light traps


CREEK

LAKE ns

RIVER ** *

WETLAND *** *** ***


oriental weatherloach SNE

Juveniles 200304 200405 200506

200304

200405 ns

200506 * *

Comments: 2005–06>03/04 and 04/05

Light trap

Adults 200304 200405 200506

200304

200405 ns

200506 * ns

Comments: 2003–04>2004–05

Total larvae SNE

200304 200405 200506

200304

200405 *

200506 ns ns

Comments: 2003–04>2004–05

Total larvae Light trap

200304 200405 200506

200304

200405 *

200506 ** ns

Comments: 2003–04>2004–05=2005–06

Total juveniles SNE


CREEK

LAKE ns

RIVER ns ns

WETLAND ns ** **

Comments: wetland>lake and river

no diff between creek and other habitats

Documents

MURRAY-DARLING BASIN AUTHORITY Assessing the …Much speculation has surrounded the role of flooding in the spawning and recruitment of native fish in the Murray-Darling Basin. This